From 4b9d2090a444d44f16788b9ad60180011d133f97 Mon Sep 17 00:00:00 2001 From: Matt Ranostay Date: Tue, 24 May 2016 21:29:19 -0700 Subject: iio: electricalconductivity: add IIO_ELECTRICALCONDUCTIVITY type Signed-off-by: Matt Ranostay Signed-off-by: Jonathan Cameron --- include/uapi/linux/iio/types.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/iio/types.h b/include/uapi/linux/iio/types.h index b0916fc72cce..22e5e589a274 100644 --- a/include/uapi/linux/iio/types.h +++ b/include/uapi/linux/iio/types.h @@ -39,6 +39,7 @@ enum iio_chan_type { IIO_RESISTANCE, IIO_PH, IIO_UVINDEX, + IIO_ELECTRICALCONDUCTIVITY, }; enum iio_modifier { -- cgit v1.2.3 From 97c79a38cd454602645f0470ffb444b3b75ce574 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 28 Apr 2016 13:16:33 -0300 Subject: perf core: Per event callchain limit Additionally to being able to control the system wide maximum depth via /proc/sys/kernel/perf_event_max_stack, now we are able to ask for different depths per event, using perf_event_attr.sample_max_stack for that. This uses an u16 hole at the end of perf_event_attr, that, when perf_event_attr.sample_type has the PERF_SAMPLE_CALLCHAIN, if sample_max_stack is zero, means use perf_event_max_stack, otherwise it'll be bounds checked under callchain_mutex. Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Alexei Starovoitov Cc: Brendan Gregg Cc: David Ahern Cc: Frederic Weisbecker Cc: He Kuang Cc: Jiri Olsa Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Milian Wolff Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: Wang Nan Cc: Zefan Li Link: http://lkml.kernel.org/n/tip-kolmn1yo40p7jhswxwrc7rrd@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/perf_event.h | 2 +- include/uapi/linux/perf_event.h | 6 +++++- kernel/bpf/stackmap.c | 2 +- kernel/events/callchain.c | 14 ++++++++++++-- kernel/events/core.c | 5 ++++- 5 files changed, 23 insertions(+), 6 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 6b87be908790..0e43355c7aad 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1076,7 +1076,7 @@ extern void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct extern struct perf_callchain_entry * get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user, u32 max_stack, bool crosstask, bool add_mark); -extern int get_callchain_buffers(void); +extern int get_callchain_buffers(int max_stack); extern void put_callchain_buffers(void); extern int sysctl_perf_event_max_stack; diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index 36ce552cf6a9..c66a485a24ac 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -276,6 +276,9 @@ enum perf_event_read_format { /* * Hardware event_id to monitor via a performance monitoring event: + * + * @sample_max_stack: Max number of frame pointers in a callchain, + * should be < /proc/sys/kernel/perf_event_max_stack */ struct perf_event_attr { @@ -385,7 +388,8 @@ struct perf_event_attr { * Wakeup watermark for AUX area */ __u32 aux_watermark; - __u32 __reserved_2; /* align to __u64 */ + __u16 sample_max_stack; + __u16 __reserved_2; /* align to __u64 */ }; #define perf_flags(attr) (*(&(attr)->read_format + 1)) diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c index a82d7605db3f..f1de5c1a2af6 100644 --- a/kernel/bpf/stackmap.c +++ b/kernel/bpf/stackmap.c @@ -99,7 +99,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr) if (err) goto free_smap; - err = get_callchain_buffers(); + err = get_callchain_buffers(sysctl_perf_event_max_stack); if (err) goto free_smap; diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c index 179ef4640964..e9fdb5203de5 100644 --- a/kernel/events/callchain.c +++ b/kernel/events/callchain.c @@ -104,7 +104,7 @@ fail: return -ENOMEM; } -int get_callchain_buffers(void) +int get_callchain_buffers(int event_max_stack) { int err = 0; int count; @@ -121,6 +121,15 @@ int get_callchain_buffers(void) /* If the allocation failed, give up */ if (!callchain_cpus_entries) err = -ENOMEM; + /* + * If requesting per event more than the global cap, + * return a different error to help userspace figure + * this out. + * + * And also do it here so that we have &callchain_mutex held. + */ + if (event_max_stack > sysctl_perf_event_max_stack) + err = -EOVERFLOW; goto exit; } @@ -174,11 +183,12 @@ perf_callchain(struct perf_event *event, struct pt_regs *regs) bool user = !event->attr.exclude_callchain_user; /* Disallow cross-task user callchains. */ bool crosstask = event->ctx->task && event->ctx->task != current; + const u32 max_stack = event->attr.sample_max_stack; if (!kernel && !user) return NULL; - return get_perf_callchain(regs, 0, kernel, user, sysctl_perf_event_max_stack, crosstask, true); + return get_perf_callchain(regs, 0, kernel, user, max_stack, crosstask, true); } struct perf_callchain_entry * diff --git a/kernel/events/core.c b/kernel/events/core.c index 050a290c72c7..79363f298445 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -8843,7 +8843,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, if (!event->parent) { if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) { - err = get_callchain_buffers(); + err = get_callchain_buffers(attr->sample_max_stack); if (err) goto err_addr_filters; } @@ -9165,6 +9165,9 @@ SYSCALL_DEFINE5(perf_event_open, return -EINVAL; } + if (!attr.sample_max_stack) + attr.sample_max_stack = sysctl_perf_event_max_stack; + /* * In cgroup mode, the pid argument is used to pass the fd * opened to the cgroup directory in cgroupfs. The cpu argument -- cgit v1.2.3 From bf1ecd210541ef5f3a110e88e8ca5d33b4aa5c23 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Tue, 31 May 2016 00:16:50 +0300 Subject: cfg80211: Allow cfg80211_connect_result() errors to be distinguished Previously, the status parameter to cfg80211_connect_result() was documented as using WLAN_STATUS_UNSPECIFIED_FAILURE (1) when the real status code for the failure is not known. This value can be used by an AP (and often is) and as such, user space cannot distinguish between explicitly rejected authentication/association and not being able to even try to associate or not receiving a response from the AP. Add a new inline function, cfg80211_connect_timeout(), to be used when the driver knows that the connection attempt failed due to a reason where connection could not be attempt or no response was received from the AP. The internal functions now allow a negative status value (-1) to be used as an indication of this special case. This results in the NL80211_ATTR_TIMED_OUT to be added to the NL80211_CMD_CONNECT event to allow user space to determine this case was hit. For backwards compatibility, NL80211_STATUS_CODE with the value WLAN_STATUS_UNSPECIFIED_FAILURE is still indicated in the event in such a case. Signed-off-by: Jouni Malinen [johannes: fix cfg80211_connect_bss() prototype to use int for status, add cfg80211_connect_timeout() to docbook, fix docbook] Signed-off-by: Johannes Berg --- Documentation/DocBook/80211.tmpl | 1 + include/net/cfg80211.h | 53 ++++++++++++++++++++++++++++++---------- include/uapi/linux/nl80211.h | 7 +++++- net/wireless/core.h | 4 +-- net/wireless/nl80211.c | 7 ++++-- net/wireless/nl80211.h | 2 +- net/wireless/sme.c | 8 +++--- 7 files changed, 58 insertions(+), 24 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/DocBook/80211.tmpl b/Documentation/DocBook/80211.tmpl index 5f7c55999c77..800fe7a9024c 100644 --- a/Documentation/DocBook/80211.tmpl +++ b/Documentation/DocBook/80211.tmpl @@ -136,6 +136,7 @@ !Finclude/net/cfg80211.h cfg80211_ibss_joined !Finclude/net/cfg80211.h cfg80211_connect_result !Finclude/net/cfg80211.h cfg80211_connect_bss +!Finclude/net/cfg80211.h cfg80211_connect_timeout !Finclude/net/cfg80211.h cfg80211_roamed !Finclude/net/cfg80211.h cfg80211_disconnected !Finclude/net/cfg80211.h cfg80211_ready_on_channel diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 63921672bed0..537f010cf5e1 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2367,19 +2367,23 @@ struct cfg80211_qos_map { * (invoked with the wireless_dev mutex held) * * @connect: Connect to the ESS with the specified parameters. When connected, - * call cfg80211_connect_result() with status code %WLAN_STATUS_SUCCESS. - * If the connection fails for some reason, call cfg80211_connect_result() - * with the status from the AP. The driver is allowed to roam to other - * BSSes within the ESS when the other BSS matches the connect parameters. - * When such roaming is initiated by the driver, the driver is expected to - * verify that the target matches the configured security parameters and - * to use Reassociation Request frame instead of Association Request frame. - * The connect function can also be used to request the driver to perform - * a specific roam when connected to an ESS. In that case, the prev_bssid + * call cfg80211_connect_result()/cfg80211_connect_bss() with status code + * %WLAN_STATUS_SUCCESS. If the connection fails for some reason, call + * cfg80211_connect_result()/cfg80211_connect_bss() with the status code + * from the AP or cfg80211_connect_timeout() if no frame with status code + * was received. + * The driver is allowed to roam to other BSSes within the ESS when the + * other BSS matches the connect parameters. When such roaming is initiated + * by the driver, the driver is expected to verify that the target matches + * the configured security parameters and to use Reassociation Request + * frame instead of Association Request frame. + * The connect function can also be used to request the driver to perform a + * specific roam when connected to an ESS. In that case, the prev_bssid * parameter is set to the BSSID of the currently associated BSS as an - * indication of requesting reassociation. In both the driver-initiated and - * new connect() call initiated roaming cases, the result of roaming is - * indicated with a call to cfg80211_roamed() or cfg80211_roamed_bss(). + * indication of requesting reassociation. + * In both the driver-initiated and new connect() call initiated roaming + * cases, the result of roaming is indicated with a call to + * cfg80211_roamed() or cfg80211_roamed_bss(). * (invoked with the wireless_dev mutex held) * @disconnect: Disconnect from the BSS/ESS. * (invoked with the wireless_dev mutex held) @@ -4680,7 +4684,7 @@ static inline void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp) void cfg80211_connect_bss(struct net_device *dev, const u8 *bssid, struct cfg80211_bss *bss, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, - size_t resp_ie_len, u16 status, gfp_t gfp); + size_t resp_ie_len, int status, gfp_t gfp); /** * cfg80211_connect_result - notify cfg80211 of connection result @@ -4709,6 +4713,29 @@ cfg80211_connect_result(struct net_device *dev, const u8 *bssid, resp_ie_len, status, gfp); } +/** + * cfg80211_connect_timeout - notify cfg80211 of connection timeout + * + * @dev: network device + * @bssid: the BSSID of the AP + * @req_ie: association request IEs (maybe be %NULL) + * @req_ie_len: association request IEs length + * @gfp: allocation flags + * + * It should be called by the underlying driver whenever connect() has failed + * in a sequence where no explicit authentication/association rejection was + * received from the AP. This could happen, e.g., due to not being able to send + * out the Authentication or Association Request frame or timing out while + * waiting for the response. + */ +static inline void +cfg80211_connect_timeout(struct net_device *dev, const u8 *bssid, + const u8 *req_ie, size_t req_ie_len, gfp_t gfp) +{ + cfg80211_connect_bss(dev, bssid, NULL, req_ie, req_ie_len, NULL, 0, -1, + gfp); +} + /** * cfg80211_roamed - notify cfg80211 of roaming * diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index e23d78685a01..8d995316aadb 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -493,7 +493,12 @@ * This attribute is ignored if driver does not support roam scan. * It is also sent as an event, with the BSSID and response IEs when the * connection is established or failed to be established. This can be - * determined by the STATUS_CODE attribute. + * determined by the %NL80211_ATTR_STATUS_CODE attribute (0 = success, + * non-zero = failure). If %NL80211_ATTR_TIMED_OUT is included in the + * event, the connection attempt failed due to not being able to initiate + * authentication/association or not receiving a response from the AP. + * Non-zero %NL80211_ATTR_STATUS_CODE value is indicated in that case as + * well to remain backwards compatible. * @NL80211_CMD_ROAM: request that the card roam (currently not implemented), * sent as an event when the card/driver roamed by itself. * @NL80211_CMD_DISCONNECT: drop a given connection; also used to notify diff --git a/net/wireless/core.h b/net/wireless/core.h index 025b7a5d508b..a4d547f99f8d 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -214,7 +214,7 @@ struct cfg80211_event { size_t req_ie_len; size_t resp_ie_len; struct cfg80211_bss *bss; - u16 status; + int status; /* -1 = failed; 0..65535 = status code */ } cr; struct { const u8 *req_ie; @@ -374,7 +374,7 @@ int cfg80211_connect(struct cfg80211_registered_device *rdev, void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, bool wextev, + int status, bool wextev, struct cfg80211_bss *bss); void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, size_t ie_len, u16 reason, bool from_ap); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index bf75afa18699..03ac2ba8b174 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -12092,7 +12092,7 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, gfp_t gfp) + int status, gfp_t gfp) { struct sk_buff *msg; void *hdr; @@ -12110,7 +12110,10 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) || nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) || (bssid && nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, bssid)) || - nla_put_u16(msg, NL80211_ATTR_STATUS_CODE, status) || + nla_put_u16(msg, NL80211_ATTR_STATUS_CODE, + status < 0 ? WLAN_STATUS_UNSPECIFIED_FAILURE : + status) || + (status < 0 && nla_put_flag(msg, NL80211_ATTR_TIMED_OUT)) || (req_ie && nla_put(msg, NL80211_ATTR_REQ_IE, req_ie_len, req_ie)) || (resp_ie && diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index 84d4edf1d545..a63f402b10b7 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -55,7 +55,7 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, gfp_t gfp); + int status, gfp_t gfp); void nl80211_send_roamed(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, diff --git a/net/wireless/sme.c b/net/wireless/sme.c index 584fdc347221..add6824c44fd 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -244,9 +244,7 @@ void cfg80211_conn_work(struct work_struct *work) if (cfg80211_conn_do_work(wdev)) { __cfg80211_connect_result( wdev->netdev, bssid, - NULL, 0, NULL, 0, - WLAN_STATUS_UNSPECIFIED_FAILURE, - false, NULL); + NULL, 0, NULL, 0, -1, false, NULL); } wdev_unlock(wdev); } @@ -648,7 +646,7 @@ static DECLARE_WORK(cfg80211_disconnect_work, disconnect_work); void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, bool wextev, + int status, bool wextev, struct cfg80211_bss *bss) { struct wireless_dev *wdev = dev->ieee80211_ptr; @@ -757,7 +755,7 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, void cfg80211_connect_bss(struct net_device *dev, const u8 *bssid, struct cfg80211_bss *bss, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, - size_t resp_ie_len, u16 status, gfp_t gfp) + size_t resp_ie_len, int status, gfp_t gfp) { struct wireless_dev *wdev = dev->ieee80211_ptr; struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); -- cgit v1.2.3 From 019ae3a918811715192b22c400ac78d54acc26a9 Mon Sep 17 00:00:00 2001 From: "Kanchanapally, Vidyullatha" Date: Mon, 16 May 2016 10:41:04 +0530 Subject: cfg80211: Advertise extended capabilities per interface type to userspace The driver extended capabilities may differ for different interface types which the userspace needs to know (for example the fine timing measurement initiator and responder bits might differ for a station and AP). Add a new nl80211 attribute to provide extended capabilities per interface type to userspace. Signed-off-by: Vidyullatha Kanchanapally Reviewed-by: Jouni Malinen Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 28 +++++++++++++++++++++++++++- include/uapi/linux/nl80211.h | 7 +++++++ net/wireless/core.c | 30 ++++++++++++++++++++++++++++++ net/wireless/nl80211.c | 43 ++++++++++++++++++++++++++++++++++++++++++- 4 files changed, 106 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 537f010cf5e1..7bbb00d8b2cd 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -3083,6 +3083,24 @@ struct wiphy_vendor_command { unsigned long *storage); }; +/** + * struct wiphy_iftype_ext_capab - extended capabilities per interface type + * @iftype: interface type + * @extended_capabilities: extended capabilities supported by the driver, + * additional capabilities might be supported by userspace; these are the + * 802.11 extended capabilities ("Extended Capabilities element") and are + * in the same format as in the information element. See IEEE Std + * 802.11-2012 8.4.2.29 for the defined fields. + * @extended_capabilities_mask: mask of the valid values + * @extended_capabilities_len: length of the extended capabilities + */ +struct wiphy_iftype_ext_capab { + enum nl80211_iftype iftype; + const u8 *extended_capabilities; + const u8 *extended_capabilities_mask; + u8 extended_capabilities_len; +}; + /** * struct wiphy - wireless hardware description * @reg_notifier: the driver's regulatory notification callback, @@ -3203,9 +3221,14 @@ struct wiphy_vendor_command { * additional capabilities might be supported by userspace; these are * the 802.11 extended capabilities ("Extended Capabilities element") * and are in the same format as in the information element. See - * 802.11-2012 8.4.2.29 for the defined fields. + * 802.11-2012 8.4.2.29 for the defined fields. These are the default + * extended capabilities to be used if the capabilities are not specified + * for a specific interface type in iftype_ext_capab. * @extended_capabilities_mask: mask of the valid values * @extended_capabilities_len: length of the extended capabilities + * @iftype_ext_capab: array of extended capabilities per interface type + * @num_iftype_ext_capab: number of interface types for which extended + * capabilities are specified separately. * @coalesce: packet coalescing support information * * @vendor_commands: array of vendor commands supported by the hardware @@ -3305,6 +3328,9 @@ struct wiphy { const u8 *extended_capabilities, *extended_capabilities_mask; u8 extended_capabilities_len; + const struct wiphy_iftype_ext_capab *iftype_ext_capab; + unsigned int num_iftype_ext_capab; + /* If multiple wiphys are registered and you're handed e.g. * a regular netdev with assigned ieee80211_ptr, you won't * know whether it points to a wiphy your driver has registered diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 8d995316aadb..53c8278827a0 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1824,6 +1824,11 @@ enum nl80211_commands { * * @NL80211_ATTR_PAD: attribute used for padding for 64-bit alignment * + * @NL80211_ATTR_IFTYPE_EXT_CAPA: Nested attribute of the following attributes: + * %NL80211_ATTR_IFTYPE, %NL80211_ATTR_EXT_CAPA, + * %NL80211_ATTR_EXT_CAPA_MASK, to specify the extended capabilities per + * interface type. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2206,6 +2211,8 @@ enum nl80211_attrs { NL80211_ATTR_PAD, + NL80211_ATTR_IFTYPE_EXT_CAPA, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/wireless/core.c b/net/wireless/core.c index d25c82bc1bbe..b8e10a952111 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -750,6 +750,36 @@ int wiphy_register(struct wiphy *wiphy) nl80211_send_reg_change_event(&request); } + /* Check that nobody globally advertises any capabilities they do not + * advertise on all possible interface types. + */ + if (wiphy->extended_capabilities_len && + wiphy->num_iftype_ext_capab && + wiphy->iftype_ext_capab) { + u8 supported_on_all, j; + const struct wiphy_iftype_ext_capab *capab; + + capab = wiphy->iftype_ext_capab; + for (j = 0; j < wiphy->extended_capabilities_len; j++) { + if (capab[0].extended_capabilities_len > j) + supported_on_all = + capab[0].extended_capabilities[j]; + else + supported_on_all = 0x00; + for (i = 1; i < wiphy->num_iftype_ext_capab; i++) { + if (j >= capab[i].extended_capabilities_len) { + supported_on_all = 0x00; + break; + } + supported_on_all &= + capab[i].extended_capabilities[j]; + } + if (WARN_ON(wiphy->extended_capabilities[j] & + ~supported_on_all)) + break; + } + } + rdev->wiphy.registered = true; rtnl_unlock(); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 03ac2ba8b174..d12044996a0e 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1264,7 +1264,7 @@ nl80211_send_mgmt_stypes(struct sk_buff *msg, struct nl80211_dump_wiphy_state { s64 filter_wiphy; long start; - long split_start, band_start, chan_start; + long split_start, band_start, chan_start, capa_start; bool split; }; @@ -1761,6 +1761,47 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev, nla_nest_end(msg, nested); } + state->split_start++; + break; + case 13: + if (rdev->wiphy.num_iftype_ext_capab && + rdev->wiphy.iftype_ext_capab) { + struct nlattr *nested_ext_capab, *nested; + + nested = nla_nest_start(msg, + NL80211_ATTR_IFTYPE_EXT_CAPA); + if (!nested) + goto nla_put_failure; + + for (i = state->capa_start; + i < rdev->wiphy.num_iftype_ext_capab; i++) { + const struct wiphy_iftype_ext_capab *capab; + + capab = &rdev->wiphy.iftype_ext_capab[i]; + + nested_ext_capab = nla_nest_start(msg, i); + if (!nested_ext_capab || + nla_put_u32(msg, NL80211_ATTR_IFTYPE, + capab->iftype) || + nla_put(msg, NL80211_ATTR_EXT_CAPA, + capab->extended_capabilities_len, + capab->extended_capabilities) || + nla_put(msg, NL80211_ATTR_EXT_CAPA_MASK, + capab->extended_capabilities_len, + capab->extended_capabilities_mask)) + goto nla_put_failure; + + nla_nest_end(msg, nested_ext_capab); + if (state->split) + break; + } + nla_nest_end(msg, nested); + if (i < rdev->wiphy.num_iftype_ext_capab) { + state->capa_start = i + 1; + break; + } + } + /* done */ state->split_start = 0; break; -- cgit v1.2.3 From 14de9d114a82a564b94388c95af79a701dc93134 Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Fri, 3 Jun 2016 16:57:12 -0400 Subject: virtio-net: Add initial MTU advice feature This commit adds the feature bit and associated mtu device entry for the virtio network device. When a virtio device comes up, it checks the feature bit for the VIRTIO_NET_F_MTU feature. If such feature bit is enabled, the driver will read the advised MTU and use it as the initial value. Signed-off-by: Aaron Conole Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 10 ++++++++++ include/uapi/linux/virtio_net.h | 3 +++ 2 files changed, 13 insertions(+) (limited to 'include/uapi') diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index e0638e556fe7..192f321247b9 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1780,6 +1780,7 @@ static int virtnet_probe(struct virtio_device *vdev) struct net_device *dev; struct virtnet_info *vi; u16 max_queue_pairs; + int mtu; if (!vdev->config->get) { dev_err(&vdev->dev, "%s failure: config access disabled\n", @@ -1896,6 +1897,14 @@ static int virtnet_probe(struct virtio_device *vdev) if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ)) vi->has_cvq = true; + if (virtio_has_feature(vdev, VIRTIO_NET_F_MTU)) { + mtu = virtio_cread16(vdev, + offsetof(struct virtio_net_config, + mtu)); + if (virtnet_change_mtu(dev, mtu)) + __virtio_clear_bit(vdev, VIRTIO_NET_F_MTU); + } + if (vi->any_header_sg) dev->needed_headroom = vi->hdr_len; @@ -2067,6 +2076,7 @@ static unsigned int features[] = { VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ, VIRTIO_NET_F_CTRL_MAC_ADDR, VIRTIO_F_ANY_LAYOUT, + VIRTIO_NET_F_MTU, }; static struct virtio_driver virtio_net_driver = { diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index ec32293a00db..1ab4ea6ec847 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -55,6 +55,7 @@ #define VIRTIO_NET_F_MQ 22 /* Device supports Receive Flow * Steering */ #define VIRTIO_NET_F_CTRL_MAC_ADDR 23 /* Set MAC address */ +#define VIRTIO_NET_F_MTU 25 /* Initial MTU advice */ #ifndef VIRTIO_NET_NO_LEGACY #define VIRTIO_NET_F_GSO 6 /* Host handles pkts w/ any GSO type */ @@ -73,6 +74,8 @@ struct virtio_net_config { * Legal values are between 1 and 0x8000 */ __u16 max_virtqueue_pairs; + /* Default maximum transmit unit advice */ + __u16 mtu; } __attribute__((packed)); /* -- cgit v1.2.3 From 53eb440f4ada034ea43b295891feec3df0fa7a29 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 6 Jun 2016 06:32:54 -0400 Subject: net sched actions: introduce timestamp for firsttime use Useful to know when the action was first used for accounting (and debugging) Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/net/act_api.h | 2 ++ include/uapi/linux/pkt_cls.h | 1 + net/sched/act_api.c | 1 + net/sched/act_bpf.c | 1 + net/sched/act_connmark.c | 1 + net/sched/act_csum.c | 1 + net/sched/act_gact.c | 1 + net/sched/act_ipt.c | 1 + net/sched/act_mirred.c | 1 + net/sched/act_nat.c | 1 + net/sched/act_pedit.c | 1 + net/sched/act_police.c | 2 ++ net/sched/act_simple.c | 1 + net/sched/act_skbedit.c | 1 + net/sched/act_vlan.c | 1 + 15 files changed, 17 insertions(+) (limited to 'include/uapi') diff --git a/include/net/act_api.h b/include/net/act_api.h index 9a9a8edc138f..8389c007076f 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -76,6 +76,8 @@ static inline void tcf_lastuse_update(struct tcf_t *tm) if (tm->lastuse != now) tm->lastuse = now; + if (unlikely(!tm->firstuse)) + tm->firstuse = now; } struct tc_action { diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index f4297c8a42fe..9ba1410bd21d 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -124,6 +124,7 @@ struct tcf_t { __u64 install; __u64 lastuse; __u64 expires; + __u64 firstuse; }; struct tc_cnt { diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 336774a535c3..5ebf6d6f85f6 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -283,6 +283,7 @@ err2: p->tcfc_index = index ? index : tcf_hash_new_index(tn); p->tcfc_tm.install = jiffies; p->tcfc_tm.lastuse = jiffies; + p->tcfc_tm.firstuse = 0; if (est) { err = gen_new_estimator(&p->tcfc_bstats, p->cpu_bstats, &p->tcfc_rate_est, diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index c7123e01c2ca..e4b877f9d322 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -156,6 +156,7 @@ static int tcf_bpf_dump(struct sk_buff *skb, struct tc_action *act, tm.install = jiffies_to_clock_t(jiffies - prog->tcf_tm.install); tm.lastuse = jiffies_to_clock_t(jiffies - prog->tcf_tm.lastuse); + tm.firstuse = jiffies_to_clock_t(jiffies - prog->tcf_tm.firstuse); tm.expires = jiffies_to_clock_t(prog->tcf_tm.expires); if (nla_put_64bit(skb, TCA_ACT_BPF_TM, sizeof(tm), &tm, diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index e0e6c6876bc7..e3f64f2d6206 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -163,6 +163,7 @@ static inline int tcf_connmark_dump(struct sk_buff *skb, struct tc_action *a, t.install = jiffies_to_clock_t(jiffies - ci->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - ci->tcf_tm.lastuse); t.expires = jiffies_to_clock_t(ci->tcf_tm.expires); + t.firstuse = jiffies_to_clock_t(jiffies - ci->tcf_tm.firstuse); if (nla_put_64bit(skb, TCA_CONNMARK_TM, sizeof(t), &t, TCA_CONNMARK_PAD)) goto nla_put_failure; diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 065f71618276..7725eafbe581 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -548,6 +548,7 @@ static int tcf_csum_dump(struct sk_buff *skb, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(p->tcf_tm.expires); if (nla_put_64bit(skb, TCA_CSUM_TM, sizeof(t), &t, TCA_CSUM_PAD)) goto nla_put_failure; diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index ec5cc8435238..c9d59f38a3f8 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -190,6 +190,7 @@ static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int #endif t.install = jiffies_to_clock_t(jiffies - gact->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(gact->tcf_tm.expires); if (nla_put_64bit(skb, TCA_GACT_TM, sizeof(t), &t, TCA_GACT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 30e9087b3536..47525ee201a9 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -279,6 +279,7 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int goto nla_put_failure; tm.install = jiffies_to_clock_t(jiffies - ipt->tcf_tm.install); tm.lastuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.lastuse); + tm.firstuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.firstuse); tm.expires = jiffies_to_clock_t(ipt->tcf_tm.expires); if (nla_put_64bit(skb, TCA_IPT_TM, sizeof(tm), &tm, TCA_IPT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index d3ac73e90f00..1b06093f8ef8 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -220,6 +220,7 @@ static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, i goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - m->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - m->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - m->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(m->tcf_tm.expires); if (nla_put_64bit(skb, TCA_MIRRED_TM, sizeof(t), &t, TCA_MIRRED_PAD)) goto nla_put_failure; diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 9135aa8f2970..9fbf780a04c7 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -266,6 +266,7 @@ static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(p->tcf_tm.expires); if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 67a17265c967..fb89275bc595 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -202,6 +202,7 @@ static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(p->tcf_tm.expires); if (nla_put_64bit(skb, TCA_PEDIT_TM, sizeof(t), &t, TCA_PEDIT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_police.c b/net/sched/act_police.c index b884dae692a1..820b11686f85 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -241,6 +241,7 @@ override: tcf_hash_new_index(tn); police->tcf_tm.install = jiffies; police->tcf_tm.lastuse = jiffies; + police->tcf_tm.firstuse = 0; h = tcf_hash(police->tcf_index, POL_TAB_MASK); spin_lock_bh(&hinfo->lock); hlist_add_head(&police->tcf_head, &hinfo->htab[h]); @@ -347,6 +348,7 @@ tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) t.install = jiffies_to_clock_t(jiffies - police->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - police->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - police->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(police->tcf_tm.expires); if (nla_put_64bit(skb, TCA_POLICE_TM, sizeof(t), &t, TCA_POLICE_PAD)) goto nla_put_failure; diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index f95d1c596986..81040f1cc3bb 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -160,6 +160,7 @@ static int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - d->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(d->tcf_tm.expires); if (nla_put_64bit(skb, TCA_DEF_TM, sizeof(t), &t, TCA_DEF_PAD)) goto nla_put_failure; diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 82105c8cb60f..cf34f31f4106 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -170,6 +170,7 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - d->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(d->tcf_tm.expires); if (nla_put_64bit(skb, TCA_SKBEDIT_TM, sizeof(t), &t, TCA_SKBEDIT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index da5120f5513f..978ec4c89684 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -184,6 +184,7 @@ static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, t.install = jiffies_to_clock_t(jiffies - v->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - v->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - v->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(v->tcf_tm.expires); if (nla_put_64bit(skb, TCA_VLAN_TM, sizeof(t), &t, TCA_VLAN_PAD)) goto nla_put_failure; -- cgit v1.2.3 From 0b0f43fe2e7291aa97b1febeaa5a0de453d007ca Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 5 Jun 2016 10:41:32 -0400 Subject: net sched: indentation and other OCD stylistic fixes Signed-off-by: Jamal Hadi Salim Acked-by: Cong Wang --- include/net/act_api.h | 14 ++++++++------ include/net/tc_act/tc_defact.h | 4 ++-- include/uapi/linux/pkt_cls.h | 6 +++--- net/sched/act_api.c | 19 +++++++++++-------- net/sched/act_bpf.c | 3 ++- net/sched/act_gact.c | 3 ++- net/sched/act_ipt.c | 6 ++++-- net/sched/act_vlan.c | 3 ++- net/sched/cls_api.c | 11 +++++++---- 9 files changed, 41 insertions(+), 28 deletions(-) (limited to 'include/uapi') diff --git a/include/net/act_api.h b/include/net/act_api.h index a891978310e9..db218a12efb5 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -2,8 +2,8 @@ #define __NET_ACT_API_H /* - * Public police action API for classifiers/qdiscs - */ + * Public action API for classifiers/qdiscs +*/ #include #include @@ -107,7 +107,8 @@ struct tc_action_ops { char kind[IFNAMSIZ]; __u32 type; /* TBD to match kind */ struct module *owner; - int (*act)(struct sk_buff *, const struct tc_action *, struct tcf_result *); + int (*act)(struct sk_buff *, const struct tc_action *, + struct tcf_result *); int (*dump)(struct sk_buff *, struct tc_action *, int, int); void (*cleanup)(struct tc_action *, int bind); int (*lookup)(struct net *, struct tc_action *, u32); @@ -125,8 +126,8 @@ struct tc_action_net { }; static inline -int tc_action_net_init(struct tc_action_net *tn, const struct tc_action_ops *ops, - unsigned int mask) +int tc_action_net_init(struct tc_action_net *tn, + const struct tc_action_ops *ops, unsigned int mask) { int err = 0; @@ -169,7 +170,8 @@ static inline int tcf_hash_release(struct tc_action *a, bool bind) } int tcf_register_action(struct tc_action_ops *a, struct pernet_operations *ops); -int tcf_unregister_action(struct tc_action_ops *a, struct pernet_operations *ops); +int tcf_unregister_action(struct tc_action_ops *a, + struct pernet_operations *ops); int tcf_action_destroy(struct list_head *actions, int bind); int tcf_action_exec(struct sk_buff *skb, const struct list_head *actions, struct tcf_result *res); diff --git a/include/net/tc_act/tc_defact.h b/include/net/tc_act/tc_defact.h index 9763dcbb9bc3..ab9b5d6be67b 100644 --- a/include/net/tc_act/tc_defact.h +++ b/include/net/tc_act/tc_defact.h @@ -5,8 +5,8 @@ struct tcf_defact { struct tcf_common common; - u32 tcfd_datalen; - void *tcfd_defdata; + u32 tcfd_datalen; + void *tcfd_defdata; }; #define to_defact(a) \ container_of(a->priv, struct tcf_defact, common) diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index 9ba1410bd21d..5702e933dc07 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -115,8 +115,8 @@ struct tc_police { __u32 mtu; struct tc_ratespec rate; struct tc_ratespec peakrate; - int refcnt; - int bindcnt; + int refcnt; + int bindcnt; __u32 capab; }; @@ -128,7 +128,7 @@ struct tcf_t { }; struct tc_cnt { - int refcnt; + int refcnt; int bindcnt; }; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 5ebf6d6f85f6..719bc2e85852 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -504,8 +504,8 @@ nla_put_failure: } EXPORT_SYMBOL(tcf_action_dump_1); -int -tcf_action_dump(struct sk_buff *skb, struct list_head *actions, int bind, int ref) +int tcf_action_dump(struct sk_buff *skb, struct list_head *actions, + int bind, int ref) { struct tc_action *a; int err = -EINVAL; @@ -688,9 +688,9 @@ errout: return -1; } -static int -tca_get_fill(struct sk_buff *skb, struct list_head *actions, u32 portid, u32 seq, - u16 flags, int event, int bind, int ref) +static int tca_get_fill(struct sk_buff *skb, struct list_head *actions, + u32 portid, u32 seq, u16 flags, int event, int bind, + int ref) { struct tcamsg *t; struct nlmsghdr *nlh; @@ -731,7 +731,8 @@ act_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; - if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, event, 0, 0) <= 0) { + if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, event, + 0, 0) <= 0) { kfree_skb(skb); return -EINVAL; } @@ -839,7 +840,8 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, if (a.ops == NULL) /*some idjot trying to flush unknown action */ goto err_out; - nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, sizeof(*t), 0); + nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, + sizeof(*t), 0); if (!nlh) goto out_module_put; t = nlmsg_data(nlh); @@ -1002,7 +1004,8 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n) u32 portid = skb ? NETLINK_CB(skb).portid : 0; int ret = 0, ovr = 0; - if ((n->nlmsg_type != RTM_GETACTION) && !netlink_capable(skb, CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETACTION) && + !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL); diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index ae0e7cbe488c..f7b6cf49ea6f 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -169,7 +169,8 @@ nla_put_failure: static const struct nla_policy act_bpf_policy[TCA_ACT_BPF_MAX + 1] = { [TCA_ACT_BPF_PARMS] = { .len = sizeof(struct tc_act_bpf) }, [TCA_ACT_BPF_FD] = { .type = NLA_U32 }, - [TCA_ACT_BPF_NAME] = { .type = NLA_NUL_STRING, .len = ACT_BPF_NAME_LEN }, + [TCA_ACT_BPF_NAME] = { .type = NLA_NUL_STRING, + .len = ACT_BPF_NAME_LEN }, [TCA_ACT_BPF_OPS_LEN] = { .type = NLA_U16 }, [TCA_ACT_BPF_OPS] = { .type = NLA_BINARY, .len = sizeof(struct sock_filter) * BPF_MAXINSNS }, diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index 4c6e0085054a..19058a7f3e5c 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -162,7 +162,8 @@ static void tcf_gact_stats_update(struct tc_action *a, u64 bytes, u32 packets, tm->lastuse = lastuse; } -static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_gact *gact = a->priv; diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 3fcde44b8f4d..e7c0f4d944a2 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -34,7 +34,8 @@ static int ipt_net_id; static int xt_net_id; -static int ipt_init_target(struct xt_entry_target *t, char *table, unsigned int hook) +static int ipt_init_target(struct xt_entry_target *t, char *table, + unsigned int hook) { struct xt_tgchk_param par; struct xt_target *target; @@ -250,7 +251,8 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, } -static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_ipt *ipt = a->priv; diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index f0a08a11f54f..b075d50e0fc3 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -179,7 +179,8 @@ static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, if (v->tcfv_action == TCA_VLAN_ACT_PUSH && (nla_put_u16(skb, TCA_VLAN_PUSH_VLAN_ID, v->tcfv_push_vid) || - nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, v->tcfv_push_proto))) + nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, + v->tcfv_push_proto))) goto nla_put_failure; tcf_tm_dump(&t, &v->tcf_tm); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index a75864d93142..aafa6bce173e 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -351,8 +351,9 @@ errout: return err; } -static int tcf_fill_node(struct net *net, struct sk_buff *skb, struct tcf_proto *tp, - unsigned long fh, u32 portid, u32 seq, u16 flags, int event) +static int tcf_fill_node(struct net *net, struct sk_buff *skb, + struct tcf_proto *tp, unsigned long fh, u32 portid, + u32 seq, u16 flags, int event) { struct tcmsg *tcm; struct nlmsghdr *nlh; @@ -474,9 +475,11 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) TC_H_MIN(tcm->tcm_info) != tp->protocol) continue; if (t > s_t) - memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0])); + memset(&cb->args[1], 0, + sizeof(cb->args)-sizeof(cb->args[0])); if (cb->args[1] == 0) { - if (tcf_fill_node(net, skb, tp, 0, NETLINK_CB(cb->skb).portid, + if (tcf_fill_node(net, skb, tp, 0, + NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTFILTER) <= 0) break; -- cgit v1.2.3 From 96c63fa7393d0a346acfe5a91e0c7d4c7782641b Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 8 Jun 2016 10:55:39 -0700 Subject: net: Add l3mdev rule Currently, VRFs require 1 oif and 1 iif rule per address family per VRF. As the number of VRF devices increases it brings scalability issues with the increasing rule list. All of the VRF rules have the same format with the exception of the specific table id to direct the lookup. Since the table id is available from the oif or iif in the loopup, the VRF rules can be consolidated to a single rule that pulls the table from the VRF device. This patch introduces a new rule attribute l3mdev. The l3mdev rule means the table id used for the lookup is pulled from the L3 master device (e.g., VRF) rather than being statically defined. With the l3mdev rule all of the basic VRF FIB rules are reduced to 1 l3mdev rule per address family (IPv4 and IPv6). If an admin wishes to insert higher priority rules for specific VRFs those rules will co-exist with the l3mdev rule. This capability means current VRF scripts will co-exist with this new simpler implementation. Currently, the rules list for both ipv4 and ipv6 look like this: $ ip ru ls 1000: from all oif vrf1 lookup 1001 1000: from all iif vrf1 lookup 1001 1000: from all oif vrf2 lookup 1002 1000: from all iif vrf2 lookup 1002 1000: from all oif vrf3 lookup 1003 1000: from all iif vrf3 lookup 1003 1000: from all oif vrf4 lookup 1004 1000: from all iif vrf4 lookup 1004 1000: from all oif vrf5 lookup 1005 1000: from all iif vrf5 lookup 1005 1000: from all oif vrf6 lookup 1006 1000: from all iif vrf6 lookup 1006 1000: from all oif vrf7 lookup 1007 1000: from all iif vrf7 lookup 1007 1000: from all oif vrf8 lookup 1008 1000: from all iif vrf8 lookup 1008 ... 32765: from all lookup local 32766: from all lookup main 32767: from all lookup default With the l3mdev rule the list is just the following regardless of the number of VRFs: $ ip ru ls 1000: from all lookup [l3mdev table] 32765: from all lookup local 32766: from all lookup main 32767: from all lookup default (Note: the above pretty print of the rule is based on an iproute2 prototype. Actual verbage may change) Signed-off-by: David Ahern Signed-off-by: David S. Miller --- include/net/fib_rules.h | 24 ++++++++++++++++++++++-- include/net/l3mdev.h | 12 ++++++++++++ include/uapi/linux/fib_rules.h | 1 + net/core/fib_rules.c | 33 ++++++++++++++++++++++++++++----- net/ipv4/fib_rules.c | 6 ++++-- net/ipv6/fib6_rules.c | 6 ++++-- net/l3mdev/l3mdev.c | 38 ++++++++++++++++++++++++++++++++++++++ 7 files changed, 109 insertions(+), 11 deletions(-) (limited to 'include/uapi') diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 59160de702b6..456e4a6006ab 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -17,7 +17,8 @@ struct fib_rule { u32 flags; u32 table; u8 action; - /* 3 bytes hole, try to use */ + u8 l3mdev; + /* 2 bytes hole, try to use */ u32 target; __be64 tun_id; struct fib_rule __rcu *ctarget; @@ -36,6 +37,7 @@ struct fib_lookup_arg { void *lookup_ptr; void *result; struct fib_rule *rule; + u32 table; int flags; #define FIB_LOOKUP_NOREF 1 #define FIB_LOOKUP_IGNORE_LINKSTATE 2 @@ -89,7 +91,8 @@ struct fib_rules_ops { [FRA_TABLE] = { .type = NLA_U32 }, \ [FRA_SUPPRESS_PREFIXLEN] = { .type = NLA_U32 }, \ [FRA_SUPPRESS_IFGROUP] = { .type = NLA_U32 }, \ - [FRA_GOTO] = { .type = NLA_U32 } + [FRA_GOTO] = { .type = NLA_U32 }, \ + [FRA_L3MDEV] = { .type = NLA_U8 } static inline void fib_rule_get(struct fib_rule *rule) { @@ -102,6 +105,20 @@ static inline void fib_rule_put(struct fib_rule *rule) kfree_rcu(rule, rcu); } +#ifdef CONFIG_NET_L3_MASTER_DEV +static inline u32 fib_rule_get_table(struct fib_rule *rule, + struct fib_lookup_arg *arg) +{ + return rule->l3mdev ? arg->table : rule->table; +} +#else +static inline u32 fib_rule_get_table(struct fib_rule *rule, + struct fib_lookup_arg *arg) +{ + return rule->table; +} +#endif + static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla) { if (nla[FRA_TABLE]) @@ -117,4 +134,7 @@ int fib_rules_lookup(struct fib_rules_ops *, struct flowi *, int flags, struct fib_lookup_arg *); int fib_default_rule_add(struct fib_rules_ops *, u32 pref, u32 table, u32 flags); + +int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh); +int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr *nlh); #endif diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index 374388dc01c8..34f33eb96a5e 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -11,6 +11,8 @@ #ifndef _NET_L3MDEV_H_ #define _NET_L3MDEV_H_ +#include + /** * struct l3mdev_ops - l3mdev operations * @@ -41,6 +43,9 @@ struct l3mdev_ops { #ifdef CONFIG_NET_L3_MASTER_DEV +int l3mdev_fib_rule_match(struct net *net, struct flowi *fl, + struct fib_lookup_arg *arg); + int l3mdev_master_ifindex_rcu(const struct net_device *dev); static inline int l3mdev_master_ifindex(struct net_device *dev) { @@ -236,6 +241,13 @@ struct sk_buff *l3mdev_ip6_rcv(struct sk_buff *skb) { return skb; } + +static inline +int l3mdev_fib_rule_match(struct net *net, struct flowi *fl, + struct fib_lookup_arg *arg) +{ + return 1; +} #endif #endif /* _NET_L3MDEV_H_ */ diff --git a/include/uapi/linux/fib_rules.h b/include/uapi/linux/fib_rules.h index 620c8a5ddc00..14404b3ebb89 100644 --- a/include/uapi/linux/fib_rules.h +++ b/include/uapi/linux/fib_rules.h @@ -50,6 +50,7 @@ enum { FRA_FWMASK, /* mask for netfilter mark */ FRA_OIFNAME, FRA_PAD, + FRA_L3MDEV, /* iif or oif is l3mdev goto its table */ __FRA_MAX }; diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 840acebbb80c..98298b11f534 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -173,7 +173,8 @@ void fib_rules_unregister(struct fib_rules_ops *ops) EXPORT_SYMBOL_GPL(fib_rules_unregister); static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, - struct flowi *fl, int flags) + struct flowi *fl, int flags, + struct fib_lookup_arg *arg) { int ret = 0; @@ -189,6 +190,9 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, if (rule->tun_id && (rule->tun_id != fl->flowi_tun_key.tun_id)) goto out; + if (rule->l3mdev && !l3mdev_fib_rule_match(rule->fr_net, fl, arg)) + goto out; + ret = ops->match(rule, fl, flags); out: return (rule->flags & FIB_RULE_INVERT) ? !ret : ret; @@ -204,7 +208,7 @@ int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, list_for_each_entry_rcu(rule, &ops->rules_list, list) { jumped: - if (!fib_rule_match(rule, ops, fl, flags)) + if (!fib_rule_match(rule, ops, fl, flags, arg)) continue; if (rule->action == FR_ACT_GOTO) { @@ -265,7 +269,7 @@ errout: return err; } -static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) +int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh) { struct net *net = sock_net(skb->sk); struct fib_rule_hdr *frh = nlmsg_data(nlh); @@ -336,6 +340,14 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) if (tb[FRA_TUN_ID]) rule->tun_id = nla_get_be64(tb[FRA_TUN_ID]); + if (tb[FRA_L3MDEV]) { +#ifdef CONFIG_NET_L3_MASTER_DEV + rule->l3mdev = nla_get_u8(tb[FRA_L3MDEV]); + if (rule->l3mdev != 1) +#endif + goto errout_free; + } + rule->action = frh->action; rule->flags = frh->flags; rule->table = frh_get_table(frh, tb); @@ -371,6 +383,9 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) } else if (rule->action == FR_ACT_GOTO) goto errout_free; + if (rule->l3mdev && rule->table) + goto errout_free; + err = ops->configure(rule, skb, frh, tb); if (err < 0) goto errout_free; @@ -424,8 +439,9 @@ errout: rules_ops_put(ops); return err; } +EXPORT_SYMBOL_GPL(fib_nl_newrule); -static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh) +int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr *nlh) { struct net *net = sock_net(skb->sk); struct fib_rule_hdr *frh = nlmsg_data(nlh); @@ -483,6 +499,10 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh) (rule->tun_id != nla_get_be64(tb[FRA_TUN_ID]))) continue; + if (tb[FRA_L3MDEV] && + (rule->l3mdev != nla_get_u8(tb[FRA_L3MDEV]))) + continue; + if (!ops->compare(rule, frh, tb)) continue; @@ -536,6 +556,7 @@ errout: rules_ops_put(ops); return err; } +EXPORT_SYMBOL_GPL(fib_nl_delrule); static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops, struct fib_rule *rule) @@ -607,7 +628,9 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, (rule->target && nla_put_u32(skb, FRA_GOTO, rule->target)) || (rule->tun_id && - nla_put_be64(skb, FRA_TUN_ID, rule->tun_id, FRA_PAD))) + nla_put_be64(skb, FRA_TUN_ID, rule->tun_id, FRA_PAD)) || + (rule->l3mdev && + nla_put_u8(skb, FRA_L3MDEV, rule->l3mdev))) goto nla_put_failure; if (rule->suppress_ifgroup != -1) { diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index f2bda9e89c61..6e9ea69e5f75 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -76,6 +76,7 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, { int err = -EAGAIN; struct fib_table *tbl; + u32 tb_id; switch (rule->action) { case FR_ACT_TO_TBL: @@ -94,7 +95,8 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, rcu_read_lock(); - tbl = fib_get_table(rule->fr_net, rule->table); + tb_id = fib_rule_get_table(rule, arg); + tbl = fib_get_table(rule->fr_net, tb_id); if (tbl) err = fib_table_lookup(tbl, &flp->u.ip4, (struct fib_result *)arg->result, @@ -180,7 +182,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (err) goto errout; - if (rule->table == RT_TABLE_UNSPEC) { + if (rule->table == RT_TABLE_UNSPEC && !rule->l3mdev) { if (rule->action == FR_ACT_TO_TBL) { struct fib_table *table; diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index ed33abf57abd..5857c1fc8b67 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -67,6 +67,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, struct net *net = rule->fr_net; pol_lookup_t lookup = arg->lookup_ptr; int err = 0; + u32 tb_id; switch (rule->action) { case FR_ACT_TO_TBL: @@ -86,7 +87,8 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, goto discard_pkt; } - table = fib6_get_table(net, rule->table); + tb_id = fib_rule_get_table(rule, arg); + table = fib6_get_table(net, tb_id); if (!table) { err = -EAGAIN; goto out; @@ -199,7 +201,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct net *net = sock_net(skb->sk); struct fib6_rule *rule6 = (struct fib6_rule *) rule; - if (rule->action == FR_ACT_TO_TBL) { + if (rule->action == FR_ACT_TO_TBL && !rule->l3mdev) { if (rule->table == RT6_TABLE_UNSPEC) goto errout; diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c index 6651a78e100c..7da97809a7e8 100644 --- a/net/l3mdev/l3mdev.c +++ b/net/l3mdev/l3mdev.c @@ -10,6 +10,7 @@ */ #include +#include #include /** @@ -160,3 +161,40 @@ int l3mdev_get_saddr(struct net *net, int ifindex, struct flowi4 *fl4) return rc; } EXPORT_SYMBOL_GPL(l3mdev_get_saddr); + +/** + * l3mdev_fib_rule_match - Determine if flowi references an + * L3 master device + * @net: network namespace for device index lookup + * @fl: flow struct + */ + +int l3mdev_fib_rule_match(struct net *net, struct flowi *fl, + struct fib_lookup_arg *arg) +{ + struct net_device *dev; + int rc = 0; + + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, fl->flowi_oif); + if (dev && netif_is_l3_master(dev) && + dev->l3mdev_ops->l3mdev_fib_table) { + arg->table = dev->l3mdev_ops->l3mdev_fib_table(dev); + rc = 1; + goto out; + } + + dev = dev_get_by_index_rcu(net, fl->flowi_iif); + if (dev && netif_is_l3_master(dev) && + dev->l3mdev_ops->l3mdev_fib_table) { + arg->table = dev->l3mdev_ops->l3mdev_fib_table(dev); + rc = 1; + goto out; + } + +out: + rcu_read_unlock(); + + return rc; +} -- cgit v1.2.3 From 1dad640b9ef320e8de92c418bcc08448d67590a4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 9 Jun 2016 10:14:39 +0200 Subject: wext: reformat struct/union declarations Everytime I need to look for these, my usual strategy fails because it assumes the right formatting. Fix the formatting here to make it consistent with the rest of the kernel. Signed-off-by: Johannes Berg --- include/uapi/linux/wireless.h | 63 +++++++++++++++---------------------------- 1 file changed, 22 insertions(+), 41 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/wireless.h b/include/uapi/linux/wireless.h index c1592e3e4036..d9ecd7c6d691 100644 --- a/include/uapi/linux/wireless.h +++ b/include/uapi/linux/wireless.h @@ -670,8 +670,7 @@ /* * Generic format for most parameters that fit in an int */ -struct iw_param -{ +struct iw_param { __s32 value; /* The value of the parameter itself */ __u8 fixed; /* Hardware should not use auto select */ __u8 disabled; /* Disable the feature */ @@ -682,8 +681,7 @@ struct iw_param * For all data larger than 16 octets, we need to use a * pointer to memory allocated in user space. */ -struct iw_point -{ +struct iw_point { void __user *pointer; /* Pointer to the data (in user space) */ __u16 length; /* number of fields or size in bytes */ __u16 flags; /* Optional params */ @@ -698,8 +696,7 @@ struct iw_point * of 10 to get 'm' lower than 10^9, with 'm'= f / (10^'e')... * The power of 10 is in 'e', the result of the division is in 'm'. */ -struct iw_freq -{ +struct iw_freq { __s32 m; /* Mantissa */ __s16 e; /* Exponent */ __u8 i; /* List index (when in range struct) */ @@ -709,8 +706,7 @@ struct iw_freq /* * Quality of the link */ -struct iw_quality -{ +struct iw_quality { __u8 qual; /* link quality (%retries, SNR, %missed beacons or better...) */ __u8 level; /* signal level (dBm) */ @@ -725,8 +721,7 @@ struct iw_quality * is already pretty exhaustive, and you should use that first. * This is only additional stats... */ -struct iw_discarded -{ +struct iw_discarded { __u32 nwid; /* Rx : Wrong nwid/essid */ __u32 code; /* Rx : Unable to code/decode (WEP) */ __u32 fragment; /* Rx : Can't perform MAC reassembly */ @@ -738,16 +733,14 @@ struct iw_discarded * Packet/Time period missed in the wireless adapter due to * "wireless" specific problems... */ -struct iw_missed -{ +struct iw_missed { __u32 beacon; /* Missed beacons/superframe */ }; /* * Quality range (for spy threshold) */ -struct iw_thrspy -{ +struct iw_thrspy { struct sockaddr addr; /* Source address (hw/mac) */ struct iw_quality qual; /* Quality of the link */ struct iw_quality low; /* Low threshold */ @@ -765,8 +758,7 @@ struct iw_thrspy * Especially, scan results are required to include an entry for the * current BSS if the driver is in Managed mode and associated with an AP. */ -struct iw_scan_req -{ +struct iw_scan_req { __u8 scan_type; /* IW_SCAN_TYPE_{ACTIVE,PASSIVE} */ __u8 essid_len; __u8 num_channels; /* num entries in channel_list; @@ -827,8 +819,7 @@ struct iw_scan_req * RX_SEQ_VALID for SIOCGIWENCODEEXT are optional, but can be useful for * debugging/testing. */ -struct iw_encode_ext -{ +struct iw_encode_ext { __u32 ext_flags; /* IW_ENCODE_EXT_* */ __u8 tx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ __u8 rx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ @@ -841,8 +832,7 @@ struct iw_encode_ext }; /* SIOCSIWMLME data */ -struct iw_mlme -{ +struct iw_mlme { __u16 cmd; /* IW_MLME_* */ __u16 reason_code; struct sockaddr addr; @@ -855,16 +845,14 @@ struct iw_mlme #define IW_PMKID_LEN 16 -struct iw_pmksa -{ +struct iw_pmksa { __u32 cmd; /* IW_PMKSA_* */ struct sockaddr bssid; __u8 pmkid[IW_PMKID_LEN]; }; /* IWEVMICHAELMICFAILURE data */ -struct iw_michaelmicfailure -{ +struct iw_michaelmicfailure { __u32 flags; struct sockaddr src_addr; __u8 tsc[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ @@ -872,8 +860,7 @@ struct iw_michaelmicfailure /* IWEVPMKIDCAND data */ #define IW_PMKID_CAND_PREAUTH 0x00000001 /* RNS pre-authentication enabled */ -struct iw_pmkid_cand -{ +struct iw_pmkid_cand { __u32 flags; /* IW_PMKID_CAND_* */ __u32 index; /* the smaller the index, the higher the * priority */ @@ -884,8 +871,7 @@ struct iw_pmkid_cand /* * Wireless statistics (used for /proc/net/wireless) */ -struct iw_statistics -{ +struct iw_statistics { __u16 status; /* Status * - device dependent for now */ @@ -897,7 +883,7 @@ struct iw_statistics /* ------------------------ IOCTL REQUEST ------------------------ */ /* - * This structure defines the payload of an ioctl, and is used + * This structure defines the payload of an ioctl, and is used * below. * * Note that this structure should fit on the memory footprint @@ -906,8 +892,7 @@ struct iw_statistics * You should check this when increasing the structures defined * above in this file... */ -union iwreq_data -{ +union iwreq_data { /* Config - generic */ char name[IFNAMSIZ]; /* Name : used to verify the presence of wireless extensions. @@ -944,15 +929,14 @@ union iwreq_data * convenience... * Do I need to remind you about structure size (32 octets) ? */ -struct iwreq -{ +struct iwreq { union { char ifrn_name[IFNAMSIZ]; /* if name, e.g. "eth0" */ } ifr_ifrn; /* Data part (defined just above) */ - union iwreq_data u; + union iwreq_data u; }; /* -------------------------- IOCTL DATA -------------------------- */ @@ -965,8 +949,7 @@ struct iwreq * Range of parameters */ -struct iw_range -{ +struct iw_range { /* Informative stuff (to choose between different interface) */ __u32 throughput; /* To give an idea... */ /* In theory this value should be the maximum benchmarked @@ -1069,9 +1052,8 @@ struct iw_range /* * Private ioctl interface information */ - -struct iw_priv_args -{ + +struct iw_priv_args { __u32 cmd; /* Number of the ioctl to issue */ __u16 set_args; /* Type and number of args */ __u16 get_args; /* Type and number of args */ @@ -1088,8 +1070,7 @@ struct iw_priv_args /* * A Wireless Event. Contains basically the same data as the ioctl... */ -struct iw_event -{ +struct iw_event { __u16 len; /* Real length of this stuff */ __u16 cmd; /* Wireless IOCTL */ union iwreq_data u; /* IOCTL fixed payload */ -- cgit v1.2.3 From 7d84e37e114215be0a0c80095891c8268a99352b Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Thu, 9 Jun 2016 13:41:15 -0400 Subject: virtio_net: Update the feature bit to comply with spec A draft version of the MTU Advice feature bit was specified as 25. This bit is not within the allowed range for network device feature bits, and should be changed to be feature bit 3 to fully comply with the spec. Fixes 14de9d114a82 ('virtio-net: Add initial MTU advice feature') Signed-off-by: Aaron Conole Suggested-by: "Michael S. Tsirkin" Signed-off-by: David S. Miller --- include/uapi/linux/virtio_net.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index 1ab4ea6ec847..0da0e3a98f17 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -35,6 +35,7 @@ #define VIRTIO_NET_F_CSUM 0 /* Host handles pkts w/ partial csum */ #define VIRTIO_NET_F_GUEST_CSUM 1 /* Guest handles pkts w/ partial csum */ #define VIRTIO_NET_F_CTRL_GUEST_OFFLOADS 2 /* Dynamic offload configuration. */ +#define VIRTIO_NET_F_MTU 3 /* Initial MTU advice */ #define VIRTIO_NET_F_MAC 5 /* Host has given MAC address. */ #define VIRTIO_NET_F_GUEST_TSO4 7 /* Guest can handle TSOv4 in. */ #define VIRTIO_NET_F_GUEST_TSO6 8 /* Guest can handle TSOv6 in. */ @@ -55,7 +56,6 @@ #define VIRTIO_NET_F_MQ 22 /* Device supports Receive Flow * Steering */ #define VIRTIO_NET_F_CTRL_MAC_ADDR 23 /* Set MAC address */ -#define VIRTIO_NET_F_MTU 25 /* Initial MTU advice */ #ifndef VIRTIO_NET_NO_LEGACY #define VIRTIO_NET_F_GSO 6 /* Host handles pkts w/ any GSO type */ -- cgit v1.2.3 From f2a4d086ed4c588d32fe9b7aa67fead7280e7bf1 Mon Sep 17 00:00:00 2001 From: William Tu Date: Fri, 10 Jun 2016 11:49:33 -0700 Subject: openvswitch: Add packet truncation support. The patch adds a new OVS action, OVS_ACTION_ATTR_TRUNC, in order to truncate packets. A 'max_len' is added for setting up the maximum packet size, and a 'cutlen' field is to record the number of bytes to trim the packet when the packet is outputting to a port, or when the packet is sent to userspace. Signed-off-by: William Tu Cc: Pravin Shelar Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- include/uapi/linux/openvswitch.h | 6 ++++++ net/openvswitch/actions.c | 40 ++++++++++++++++++++++++++++++++++++---- net/openvswitch/datapath.c | 29 +++++++++++++++++------------ net/openvswitch/datapath.h | 5 ++++- net/openvswitch/flow_netlink.c | 9 +++++++++ net/openvswitch/vport.c | 1 + 6 files changed, 73 insertions(+), 17 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index bb0d515b7654..8274675ba9a3 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -580,6 +580,10 @@ enum ovs_userspace_attr { #define OVS_USERSPACE_ATTR_MAX (__OVS_USERSPACE_ATTR_MAX - 1) +struct ovs_action_trunc { + uint32_t max_len; /* Max packet size in bytes. */ +}; + /** * struct ovs_action_push_mpls - %OVS_ACTION_ATTR_PUSH_MPLS action argument. * @mpls_lse: MPLS label stack entry to push. @@ -703,6 +707,7 @@ enum ovs_nat_attr { * enum ovs_action_attr - Action types. * * @OVS_ACTION_ATTR_OUTPUT: Output packet to port. + * @OVS_ACTION_ATTR_TRUNC: Output packet to port with truncated packet size. * @OVS_ACTION_ATTR_USERSPACE: Send packet to userspace according to nested * %OVS_USERSPACE_ATTR_* attributes. * @OVS_ACTION_ATTR_SET: Replaces the contents of an existing header. The @@ -756,6 +761,7 @@ enum ovs_action_attr { * The data must be zero for the unmasked * bits. */ OVS_ACTION_ATTR_CT, /* Nested OVS_CT_ATTR_* . */ + OVS_ACTION_ATTR_TRUNC, /* u32 struct ovs_action_trunc. */ __OVS_ACTION_ATTR_MAX, /* Nothing past this will be accepted * from userspace. */ diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 9a3eb7a0ebf4..1ecbd7715f6d 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -750,6 +750,14 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, if (likely(vport)) { u16 mru = OVS_CB(skb)->mru; + u32 cutlen = OVS_CB(skb)->cutlen; + + if (unlikely(cutlen > 0)) { + if (skb->len - cutlen > ETH_HLEN) + pskb_trim(skb, skb->len - cutlen); + else + pskb_trim(skb, ETH_HLEN); + } if (likely(!mru || (skb->len <= mru + ETH_HLEN))) { ovs_vport_send(vport, skb); @@ -775,7 +783,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, static int output_userspace(struct datapath *dp, struct sk_buff *skb, struct sw_flow_key *key, const struct nlattr *attr, - const struct nlattr *actions, int actions_len) + const struct nlattr *actions, int actions_len, + uint32_t cutlen) { struct dp_upcall_info upcall; const struct nlattr *a; @@ -822,7 +831,7 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb, } /* End of switch. */ } - return ovs_dp_upcall(dp, skb, key, &upcall); + return ovs_dp_upcall(dp, skb, key, &upcall, cutlen); } static int sample(struct datapath *dp, struct sk_buff *skb, @@ -832,6 +841,7 @@ static int sample(struct datapath *dp, struct sk_buff *skb, const struct nlattr *acts_list = NULL; const struct nlattr *a; int rem; + u32 cutlen = 0; for (a = nla_data(attr), rem = nla_len(attr); rem > 0; a = nla_next(a, &rem)) { @@ -858,13 +868,24 @@ static int sample(struct datapath *dp, struct sk_buff *skb, return 0; /* The only known usage of sample action is having a single user-space + * action, or having a truncate action followed by a single user-space * action. Treat this usage as a special case. * The output_userspace() should clone the skb to be sent to the * user space. This skb will be consumed by its caller. */ + if (unlikely(nla_type(a) == OVS_ACTION_ATTR_TRUNC)) { + struct ovs_action_trunc *trunc = nla_data(a); + + if (skb->len > trunc->max_len) + cutlen = skb->len - trunc->max_len; + + a = nla_next(a, &rem); + } + if (likely(nla_type(a) == OVS_ACTION_ATTR_USERSPACE && nla_is_last(a, rem))) - return output_userspace(dp, skb, key, a, actions, actions_len); + return output_userspace(dp, skb, key, a, actions, + actions_len, cutlen); skb = skb_clone(skb, GFP_ATOMIC); if (!skb) @@ -1051,6 +1072,7 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, if (out_skb) do_output(dp, out_skb, prev_port, key); + OVS_CB(skb)->cutlen = 0; prev_port = -1; } @@ -1059,8 +1081,18 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, prev_port = nla_get_u32(a); break; + case OVS_ACTION_ATTR_TRUNC: { + struct ovs_action_trunc *trunc = nla_data(a); + + if (skb->len > trunc->max_len) + OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + break; + } + case OVS_ACTION_ATTR_USERSPACE: - output_userspace(dp, skb, key, a, attr, len); + output_userspace(dp, skb, key, a, attr, + len, OVS_CB(skb)->cutlen); + OVS_CB(skb)->cutlen = 0; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 856bd8dba676..673934295333 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -137,10 +137,12 @@ EXPORT_SYMBOL_GPL(lockdep_ovsl_is_held); static struct vport *new_vport(const struct vport_parms *); static int queue_gso_packets(struct datapath *dp, struct sk_buff *, const struct sw_flow_key *, - const struct dp_upcall_info *); + const struct dp_upcall_info *, + uint32_t cutlen); static int queue_userspace_packet(struct datapath *dp, struct sk_buff *, const struct sw_flow_key *, - const struct dp_upcall_info *); + const struct dp_upcall_info *, + uint32_t cutlen); /* Must be called with rcu_read_lock. */ static struct datapath *get_dp_rcu(struct net *net, int dp_ifindex) @@ -275,7 +277,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall); + error = ovs_dp_upcall(dp, skb, key, &upcall, 0); if (unlikely(error)) kfree_skb(skb); else @@ -300,7 +302,8 @@ out: int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb, const struct sw_flow_key *key, - const struct dp_upcall_info *upcall_info) + const struct dp_upcall_info *upcall_info, + uint32_t cutlen) { struct dp_stats_percpu *stats; int err; @@ -311,9 +314,9 @@ int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb, } if (!skb_is_gso(skb)) - err = queue_userspace_packet(dp, skb, key, upcall_info); + err = queue_userspace_packet(dp, skb, key, upcall_info, cutlen); else - err = queue_gso_packets(dp, skb, key, upcall_info); + err = queue_gso_packets(dp, skb, key, upcall_info, cutlen); if (err) goto err; @@ -331,7 +334,8 @@ err: static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, const struct sw_flow_key *key, - const struct dp_upcall_info *upcall_info) + const struct dp_upcall_info *upcall_info, + uint32_t cutlen) { unsigned short gso_type = skb_shinfo(skb)->gso_type; struct sw_flow_key later_key; @@ -360,7 +364,7 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, if (gso_type & SKB_GSO_UDP && skb != segs) key = &later_key; - err = queue_userspace_packet(dp, skb, key, upcall_info); + err = queue_userspace_packet(dp, skb, key, upcall_info, cutlen); if (err) break; @@ -416,7 +420,8 @@ static void pad_packet(struct datapath *dp, struct sk_buff *skb) static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, const struct sw_flow_key *key, - const struct dp_upcall_info *upcall_info) + const struct dp_upcall_info *upcall_info, + uint32_t cutlen) { struct ovs_header *upcall; struct sk_buff *nskb = NULL; @@ -461,7 +466,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, else hlen = skb->len; - len = upcall_msg_size(upcall_info, hlen); + len = upcall_msg_size(upcall_info, hlen - cutlen); user_skb = genlmsg_new(len, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; @@ -515,9 +520,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len); + nla->nla_len = nla_attr_size(skb->len - cutlen); - err = skb_zerocopy(user_skb, skb, skb->len, hlen); + err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); if (err) goto out; diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 427e39a045cf..ab85c1cae255 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -100,11 +100,13 @@ struct datapath { * @input_vport: The original vport packet came in on. This value is cached * when a packet is received by OVS. * @mru: The maximum received fragement size; 0 if the packet is not + * @cutlen: The number of bytes from the packet end to be removed. * fragmented. */ struct ovs_skb_cb { struct vport *input_vport; u16 mru; + u32 cutlen; }; #define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb) @@ -194,7 +196,8 @@ extern struct genl_family dp_vport_genl_family; void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key); void ovs_dp_detach_port(struct vport *); int ovs_dp_upcall(struct datapath *, struct sk_buff *, - const struct sw_flow_key *, const struct dp_upcall_info *); + const struct sw_flow_key *, const struct dp_upcall_info *, + uint32_t cutlen); const char *ovs_dp_name(const struct datapath *dp); struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 pid, u32 seq, diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index 0bb650f4f219..c78a6a1476fb 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -2229,6 +2229,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, [OVS_ACTION_ATTR_SAMPLE] = (u32)-1, [OVS_ACTION_ATTR_HASH] = sizeof(struct ovs_action_hash), [OVS_ACTION_ATTR_CT] = (u32)-1, + [OVS_ACTION_ATTR_TRUNC] = sizeof(struct ovs_action_trunc), }; const struct ovs_action_push_vlan *vlan; int type = nla_type(a); @@ -2255,6 +2256,14 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, return -EINVAL; break; + case OVS_ACTION_ATTR_TRUNC: { + const struct ovs_action_trunc *trunc = nla_data(a); + + if (trunc->max_len < ETH_HLEN) + return -EINVAL; + break; + } + case OVS_ACTION_ATTR_HASH: { const struct ovs_action_hash *act_hash = nla_data(a); diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index 31cbc8c5c7db..6b21fd068d87 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -444,6 +444,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; + OVS_CB(skb)->cutlen = 0; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- cgit v1.2.3 From 678ece30226ac05e95768d5f70db53a475569d37 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:17 +0300 Subject: virtio_net: add _UAPI prefix to virtio_net header guards This gives better namespacing and prevents conflicts with no-uapi version of virtio_net header that will be introduced in the following patch. Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller --- include/uapi/linux/virtio_net.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index 0da0e3a98f17..fc353b518288 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -1,5 +1,5 @@ -#ifndef _LINUX_VIRTIO_NET_H -#define _LINUX_VIRTIO_NET_H +#ifndef _UAPI_LINUX_VIRTIO_NET_H +#define _UAPI_LINUX_VIRTIO_NET_H /* This header is BSD licensed so anyone can use the definitions to implement * compatible drivers/servers. * @@ -245,4 +245,4 @@ struct virtio_net_ctrl_mq { #define VIRTIO_NET_CTRL_GUEST_OFFLOADS 5 #define VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET 0 -#endif /* _LINUX_VIRTIO_NET_H */ +#endif /* _UAPI_LINUX_VIRTIO_NET_H */ -- cgit v1.2.3 From d7c51b47ac11e66f547b55640405c1c474642d72 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 26 Apr 2016 10:35:29 +0200 Subject: gpio: userspace ABI for reading/writing GPIO lines This adds a userspace ABI for reading and writing GPIO lines. The mechanism returns an anonymous file handle to a request to read/write n offsets from a gpiochip. This file handle in turn accepts two ioctl()s: one that reads and one that writes values to the selected lines. - Handles can be requested as input/output, active low, open drain, open source, however when you issue a request for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all have the same flags, i.e. all inputs or all outputs, all open drain etc. If a granular control of the flags for each line is desired, they need to be requested individually, not in a batch. - The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be issued also to output lines to verify that the hardware is in the expected state. - It reads and writes up to GPIOHANDLES_MAX lines at once, utilizing the .set_multiple() call in the driver if possible, making the call efficient if several lines can be written with a single register update. The limitation of GPIOHANDLES_MAX to 64 lines is done under the assumption that we may expect hardware that can issue a transaction updating 64 bits at an instant but unlikely anything larger than that. ChangeLog v2->v3: - Use gpiod_get_value_cansleep() so we support also slowpath GPIO drivers. - Fix up the UAPI docs kerneldoc. - Allocate the anonymous fd last, so that the release function don't get called until that point of something fails. After this point, skip the errorpath. ChangeLog v1->v2: - Handle ioctl_compat() properly based on a similar patch to the other ioctl() handling code. - Use _IOWR() as we pass pointers both in and out of the ioctl() - Use kmalloc() and kfree() for the linehandled, do not try to be fancy with devm_* it doesn't work the way I thought. - Fix const-correctness on the linehandle name field. Acked-by: Michael Welling Signed-off-by: Linus Walleij --- drivers/gpio/gpiolib.c | 193 ++++++++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/gpio.h | 61 ++++++++++++++- 2 files changed, 251 insertions(+), 3 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index 24f60d28f0c0..5f2e73e99fa1 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include "gpiolib.h" @@ -310,6 +311,196 @@ static int gpiochip_set_desc_names(struct gpio_chip *gc) return 0; } +/* + * GPIO line handle management + */ + +/** + * struct linehandle_state - contains the state of a userspace handle + * @gdev: the GPIO device the handle pertains to + * @label: consumer label used to tag descriptors + * @descs: the GPIO descriptors held by this handle + * @numdescs: the number of descriptors held in the descs array + */ +struct linehandle_state { + struct gpio_device *gdev; + const char *label; + struct gpio_desc *descs[GPIOHANDLES_MAX]; + u32 numdescs; +}; + +static long linehandle_ioctl(struct file *filep, unsigned int cmd, + unsigned long arg) +{ + struct linehandle_state *lh = filep->private_data; + void __user *ip = (void __user *)arg; + struct gpiohandle_data ghd; + int i; + + if (cmd == GPIOHANDLE_GET_LINE_VALUES_IOCTL) { + int val; + + /* TODO: check if descriptors are really input */ + for (i = 0; i < lh->numdescs; i++) { + val = gpiod_get_value_cansleep(lh->descs[i]); + if (val < 0) + return val; + ghd.values[i] = val; + } + + if (copy_to_user(ip, &ghd, sizeof(ghd))) + return -EFAULT; + + return 0; + } else if (cmd == GPIOHANDLE_SET_LINE_VALUES_IOCTL) { + int vals[GPIOHANDLES_MAX]; + + /* TODO: check if descriptors are really output */ + if (copy_from_user(&ghd, ip, sizeof(ghd))) + return -EFAULT; + + /* Clamp all values to [0,1] */ + for (i = 0; i < lh->numdescs; i++) + vals[i] = !!ghd.values[i]; + + /* Reuse the array setting function */ + gpiod_set_array_value_complex(false, + true, + lh->numdescs, + lh->descs, + vals); + return 0; + } + return -EINVAL; +} + +#ifdef CONFIG_COMPAT +static long linehandle_ioctl_compat(struct file *filep, unsigned int cmd, + unsigned long arg) +{ + return linehandle_ioctl(filep, cmd, (unsigned long)compat_ptr(arg)); +} +#endif + +static int linehandle_release(struct inode *inode, struct file *filep) +{ + struct linehandle_state *lh = filep->private_data; + struct gpio_device *gdev = lh->gdev; + int i; + + for (i = 0; i < lh->numdescs; i++) + gpiod_free(lh->descs[i]); + kfree(lh->label); + kfree(lh); + put_device(&gdev->dev); + return 0; +} + +static const struct file_operations linehandle_fileops = { + .release = linehandle_release, + .owner = THIS_MODULE, + .llseek = noop_llseek, + .unlocked_ioctl = linehandle_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = linehandle_ioctl_compat, +#endif +}; + +static int linehandle_create(struct gpio_device *gdev, void __user *ip) +{ + struct gpiohandle_request handlereq; + struct linehandle_state *lh; + int fd, i, ret; + + if (copy_from_user(&handlereq, ip, sizeof(handlereq))) + return -EFAULT; + if ((handlereq.lines == 0) || (handlereq.lines > GPIOHANDLES_MAX)) + return -EINVAL; + + lh = kzalloc(sizeof(*lh), GFP_KERNEL); + if (!lh) + return -ENOMEM; + lh->gdev = gdev; + get_device(&gdev->dev); + + /* Make sure this is terminated */ + handlereq.consumer_label[sizeof(handlereq.consumer_label)-1] = '\0'; + if (strlen(handlereq.consumer_label)) { + lh->label = kstrdup(handlereq.consumer_label, + GFP_KERNEL); + if (!lh->label) { + ret = -ENOMEM; + goto out_free_lh; + } + } + + /* Request each GPIO */ + for (i = 0; i < handlereq.lines; i++) { + u32 offset = handlereq.lineoffsets[i]; + u32 lflags = handlereq.flags; + struct gpio_desc *desc; + + desc = &gdev->descs[offset]; + ret = gpiod_request(desc, lh->label); + if (ret) + goto out_free_descs; + lh->descs[i] = desc; + + if (lflags & GPIOHANDLE_REQUEST_ACTIVE_LOW) + set_bit(FLAG_ACTIVE_LOW, &desc->flags); + if (lflags & GPIOHANDLE_REQUEST_OPEN_DRAIN) + set_bit(FLAG_OPEN_DRAIN, &desc->flags); + if (lflags & GPIOHANDLE_REQUEST_OPEN_SOURCE) + set_bit(FLAG_OPEN_SOURCE, &desc->flags); + + /* + * Lines have to be requested explicitly for input + * or output, else the line will be treated "as is". + */ + if (lflags & GPIOHANDLE_REQUEST_OUTPUT) { + int val = !!handlereq.default_values[i]; + + ret = gpiod_direction_output(desc, val); + if (ret) + goto out_free_descs; + } else if (lflags & GPIOHANDLE_REQUEST_INPUT) { + ret = gpiod_direction_input(desc); + if (ret) + goto out_free_descs; + } + dev_dbg(&gdev->dev, "registered chardev handle for line %d\n", + offset); + } + lh->numdescs = handlereq.lines; + + fd = anon_inode_getfd("gpio-linehandle", + &linehandle_fileops, + lh, + O_RDONLY | O_CLOEXEC); + if (fd < 0) { + ret = fd; + goto out_free_descs; + } + + handlereq.fd = fd; + if (copy_to_user(ip, &handlereq, sizeof(handlereq))) + return -EFAULT; + + dev_dbg(&gdev->dev, "registered chardev handle for %d lines\n", + lh->numdescs); + + return 0; + +out_free_descs: + for (; i >= 0; i--) + gpiod_free(lh->descs[i]); + kfree(lh->label); +out_free_lh: + kfree(lh); + put_device(&gdev->dev); + return ret; +} + /** * gpio_ioctl() - ioctl handler for the GPIO chardev */ @@ -385,6 +576,8 @@ static long gpio_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (copy_to_user(ip, &lineinfo, sizeof(lineinfo))) return -EFAULT; return 0; + } else if (cmd == GPIO_GET_LINEHANDLE_IOCTL) { + return linehandle_create(gdev, ip); } return -EINVAL; } diff --git a/include/uapi/linux/gpio.h b/include/uapi/linux/gpio.h index d0a3cac72250..21905531dcf4 100644 --- a/include/uapi/linux/gpio.h +++ b/include/uapi/linux/gpio.h @@ -1,7 +1,7 @@ /* * - userspace ABI for the GPIO character devices * - * Copyright (C) 2015 Linus Walleij + * Copyright (C) 2016 Linus Walleij * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License version 2 as published by @@ -26,8 +26,8 @@ struct gpiochip_info { __u32 lines; }; -/* Line is in use by the kernel */ -#define GPIOLINE_FLAG_KERNEL (1UL << 0) +/* Informational flags */ +#define GPIOLINE_FLAG_KERNEL (1UL << 0) /* Line used by the kernel */ #define GPIOLINE_FLAG_IS_OUT (1UL << 1) #define GPIOLINE_FLAG_ACTIVE_LOW (1UL << 2) #define GPIOLINE_FLAG_OPEN_DRAIN (1UL << 3) @@ -52,7 +52,62 @@ struct gpioline_info { char consumer[32]; }; +/* Maximum number of requested handles */ +#define GPIOHANDLES_MAX 64 + +/* Request flags */ +#define GPIOHANDLE_REQUEST_INPUT (1UL << 0) +#define GPIOHANDLE_REQUEST_OUTPUT (1UL << 1) +#define GPIOHANDLE_REQUEST_ACTIVE_LOW (1UL << 2) +#define GPIOHANDLE_REQUEST_OPEN_DRAIN (1UL << 3) +#define GPIOHANDLE_REQUEST_OPEN_SOURCE (1UL << 4) + +/** + * struct gpiohandle_request - Information about a GPIO handle request + * @lineoffsets: an array desired lines, specified by offset index for the + * associated GPIO device + * @flags: desired flags for the desired GPIO lines, such as + * GPIOHANDLE_REQUEST_OUTPUT, GPIOHANDLE_REQUEST_ACTIVE_LOW etc, OR:ed + * together. Note that even if multiple lines are requested, the same flags + * must be applicable to all of them, if you want lines with individual + * flags set, request them one by one. It is possible to select + * a batch of input or output lines, but they must all have the same + * characteristics, i.e. all inputs or all outputs, all active low etc + * @default_values: if the GPIOHANDLE_REQUEST_OUTPUT is set for a requested + * line, this specifies the default output value, should be 0 (low) or + * 1 (high), anything else than 0 or 1 will be interpreted as 1 (high) + * @consumer_label: a desired consumer label for the selected GPIO line(s) + * such as "my-bitbanged-relay" + * @lines: number of lines requested in this request, i.e. the number of + * valid fields in the above arrays, set to 1 to request a single line + * @fd: if successful this field will contain a valid anonymous file handle + * after a GPIO_GET_LINEHANDLE_IOCTL operation, zero or negative value + * means error + */ +struct gpiohandle_request { + __u32 lineoffsets[GPIOHANDLES_MAX]; + __u32 flags; + __u8 default_values[GPIOHANDLES_MAX]; + char consumer_label[32]; + __u32 lines; + int fd; +}; + #define GPIO_GET_CHIPINFO_IOCTL _IOR(0xB4, 0x01, struct gpiochip_info) #define GPIO_GET_LINEINFO_IOCTL _IOWR(0xB4, 0x02, struct gpioline_info) +#define GPIO_GET_LINEHANDLE_IOCTL _IOWR(0xB4, 0x03, struct gpiohandle_request) + +/** + * struct gpiohandle_data - Information of values on a GPIO handle + * @values: when getting the state of lines this contains the current + * state of a line, when setting the state of lines these should contain + * the desired target state + */ +struct gpiohandle_data { + __u8 values[GPIOHANDLES_MAX]; +}; + +#define GPIOHANDLE_GET_LINE_VALUES_IOCTL _IOWR(0xB4, 0x08, struct gpiohandle_data) +#define GPIOHANDLE_SET_LINE_VALUES_IOCTL _IOWR(0xB4, 0x09, struct gpiohandle_data) #endif /* _UAPI_GPIO_H_ */ -- cgit v1.2.3 From 61f922db72216b00386581c851db9c9095961522 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 2 Jun 2016 11:30:15 +0200 Subject: gpio: userspace ABI for reading GPIO line events This adds an ABI for listening to events on GPIO lines. The mechanism returns an anonymous file handle to a request to listen to a specific offset on a specific gpiochip. To fetch the stream of events from the file handle, userspace simply reads an event. - Events can be requested with the same flags as ordinary handles, i.e. open drain or open source. An ioctl() call GPIO_GET_LINEEVENT_IOCTL is issued indicating the desired line. - Events can be requested for falling edge events, rising edge events, or both. - All events are timestamped using the kernel real time nanosecond timestamp (the same as is used by IIO). - The supplied consumer label will appear in "lsgpio" listings of the lines, and in /proc/interrupts as the mechanism will request an interrupt from the gpio chip. - Events are not supported on gpiochips that do not serve interrupts (no legal .to_irq() call). The event interrupt is threaded to avoid any realtime problems. - It is possible to also directly read the current value of the registered GPIO line by issuing the same GPIOHANDLE_GET_LINE_VALUES_IOCTL as used by the line handles. Setting the value is not supported: we do not listen to events on output lines. This ABI is strongly influenced by Industrial I/O and surpasses the old sysfs ABI by providing proper precision timestamps, making it possible to set flags like open drain, and put consumer names on the GPIO lines. Signed-off-by: Linus Walleij --- drivers/gpio/gpiolib.c | 298 ++++++++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/gpio.h | 54 ++++++++- 2 files changed, 347 insertions(+), 5 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index 5f2e73e99fa1..5239230ad075 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -22,6 +22,9 @@ #include #include #include +#include +#include +#include #include #include "gpiolib.h" @@ -501,6 +504,299 @@ out_free_lh: return ret; } +/* + * GPIO line event management + */ + +/** + * struct lineevent_state - contains the state of a userspace event + * @gdev: the GPIO device the event pertains to + * @label: consumer label used to tag descriptors + * @desc: the GPIO descriptor held by this event + * @eflags: the event flags this line was requested with + * @irq: the interrupt that trigger in response to events on this GPIO + * @wait: wait queue that handles blocking reads of events + * @events: KFIFO for the GPIO events + * @read_lock: mutex lock to protect reads from colliding with adding + * new events to the FIFO + */ +struct lineevent_state { + struct gpio_device *gdev; + const char *label; + struct gpio_desc *desc; + u32 eflags; + int irq; + wait_queue_head_t wait; + DECLARE_KFIFO(events, struct gpioevent_data, 16); + struct mutex read_lock; +}; + +static unsigned int lineevent_poll(struct file *filep, + struct poll_table_struct *wait) +{ + struct lineevent_state *le = filep->private_data; + unsigned int events = 0; + + poll_wait(filep, &le->wait, wait); + + if (!kfifo_is_empty(&le->events)) + events = POLLIN | POLLRDNORM; + + return events; +} + + +static ssize_t lineevent_read(struct file *filep, + char __user *buf, + size_t count, + loff_t *f_ps) +{ + struct lineevent_state *le = filep->private_data; + unsigned int copied; + int ret; + + if (count < sizeof(struct gpioevent_data)) + return -EINVAL; + + do { + if (kfifo_is_empty(&le->events)) { + if (filep->f_flags & O_NONBLOCK) + return -EAGAIN; + + ret = wait_event_interruptible(le->wait, + !kfifo_is_empty(&le->events)); + if (ret) + return ret; + } + + if (mutex_lock_interruptible(&le->read_lock)) + return -ERESTARTSYS; + ret = kfifo_to_user(&le->events, buf, count, &copied); + mutex_unlock(&le->read_lock); + + if (ret) + return ret; + + /* + * If we couldn't read anything from the fifo (a different + * thread might have been faster) we either return -EAGAIN if + * the file descriptor is non-blocking, otherwise we go back to + * sleep and wait for more data to arrive. + */ + if (copied == 0 && (filep->f_flags & O_NONBLOCK)) + return -EAGAIN; + + } while (copied == 0); + + return copied; +} + +static int lineevent_release(struct inode *inode, struct file *filep) +{ + struct lineevent_state *le = filep->private_data; + struct gpio_device *gdev = le->gdev; + + free_irq(le->irq, le); + gpiod_free(le->desc); + kfree(le->label); + kfree(le); + put_device(&gdev->dev); + return 0; +} + +static long lineevent_ioctl(struct file *filep, unsigned int cmd, + unsigned long arg) +{ + struct lineevent_state *le = filep->private_data; + void __user *ip = (void __user *)arg; + struct gpiohandle_data ghd; + + /* + * We can get the value for an event line but not set it, + * because it is input by definition. + */ + if (cmd == GPIOHANDLE_GET_LINE_VALUES_IOCTL) { + int val; + + val = gpiod_get_value_cansleep(le->desc); + if (val < 0) + return val; + ghd.values[0] = val; + + if (copy_to_user(ip, &ghd, sizeof(ghd))) + return -EFAULT; + + return 0; + } + return -EINVAL; +} + +#ifdef CONFIG_COMPAT +static long lineevent_ioctl_compat(struct file *filep, unsigned int cmd, + unsigned long arg) +{ + return lineevent_ioctl(filep, cmd, (unsigned long)compat_ptr(arg)); +} +#endif + +static const struct file_operations lineevent_fileops = { + .release = lineevent_release, + .read = lineevent_read, + .poll = lineevent_poll, + .owner = THIS_MODULE, + .llseek = noop_llseek, + .unlocked_ioctl = lineevent_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = lineevent_ioctl_compat, +#endif +}; + +irqreturn_t lineevent_irq_thread(int irq, void *p) +{ + struct lineevent_state *le = p; + struct gpioevent_data ge; + int ret; + + ge.timestamp = ktime_get_real_ns(); + + if (le->eflags & GPIOEVENT_REQUEST_BOTH_EDGES) { + int level = gpiod_get_value_cansleep(le->desc); + + if (level) + /* Emit low-to-high event */ + ge.id = GPIOEVENT_EVENT_RISING_EDGE; + else + /* Emit high-to-low event */ + ge.id = GPIOEVENT_EVENT_FALLING_EDGE; + } else if (le->eflags & GPIOEVENT_REQUEST_RISING_EDGE) { + /* Emit low-to-high event */ + ge.id = GPIOEVENT_EVENT_RISING_EDGE; + } else if (le->eflags & GPIOEVENT_REQUEST_FALLING_EDGE) { + /* Emit high-to-low event */ + ge.id = GPIOEVENT_EVENT_FALLING_EDGE; + } + + ret = kfifo_put(&le->events, ge); + if (ret != 0) + wake_up_poll(&le->wait, POLLIN); + + return IRQ_HANDLED; +} + +static int lineevent_create(struct gpio_device *gdev, void __user *ip) +{ + struct gpioevent_request eventreq; + struct lineevent_state *le; + struct gpio_desc *desc; + u32 offset; + u32 lflags; + u32 eflags; + int fd; + int ret; + int irqflags = 0; + + if (copy_from_user(&eventreq, ip, sizeof(eventreq))) + return -EFAULT; + + le = kzalloc(sizeof(*le), GFP_KERNEL); + if (!le) + return -ENOMEM; + le->gdev = gdev; + get_device(&gdev->dev); + + /* Make sure this is terminated */ + eventreq.consumer_label[sizeof(eventreq.consumer_label)-1] = '\0'; + if (strlen(eventreq.consumer_label)) { + le->label = kstrdup(eventreq.consumer_label, + GFP_KERNEL); + if (!le->label) { + ret = -ENOMEM; + goto out_free_le; + } + } + + offset = eventreq.lineoffset; + lflags = eventreq.handleflags; + eflags = eventreq.eventflags; + + /* This is just wrong: we don't look for events on output lines */ + if (lflags & GPIOHANDLE_REQUEST_OUTPUT) { + ret = -EINVAL; + goto out_free_label; + } + + desc = &gdev->descs[offset]; + ret = gpiod_request(desc, le->label); + if (ret) + goto out_free_desc; + le->desc = desc; + le->eflags = eflags; + + if (lflags & GPIOHANDLE_REQUEST_ACTIVE_LOW) + set_bit(FLAG_ACTIVE_LOW, &desc->flags); + if (lflags & GPIOHANDLE_REQUEST_OPEN_DRAIN) + set_bit(FLAG_OPEN_DRAIN, &desc->flags); + if (lflags & GPIOHANDLE_REQUEST_OPEN_SOURCE) + set_bit(FLAG_OPEN_SOURCE, &desc->flags); + + ret = gpiod_direction_input(desc); + if (ret) + goto out_free_desc; + + le->irq = gpiod_to_irq(desc); + if (le->irq <= 0) { + ret = -ENODEV; + goto out_free_desc; + } + + if (eflags & GPIOEVENT_REQUEST_RISING_EDGE) + irqflags |= IRQF_TRIGGER_RISING; + if (eflags & GPIOEVENT_REQUEST_FALLING_EDGE) + irqflags |= IRQF_TRIGGER_FALLING; + irqflags |= IRQF_ONESHOT; + irqflags |= IRQF_SHARED; + + INIT_KFIFO(le->events); + init_waitqueue_head(&le->wait); + mutex_init(&le->read_lock); + + /* Request a thread to read the events */ + ret = request_threaded_irq(le->irq, + NULL, + lineevent_irq_thread, + irqflags, + le->label, + le); + if (ret) + goto out_free_desc; + + fd = anon_inode_getfd("gpio-event", + &lineevent_fileops, + le, + O_RDONLY | O_CLOEXEC); + if (fd < 0) { + ret = fd; + goto out_free_irq; + } + + eventreq.fd = fd; + if (copy_to_user(ip, &eventreq, sizeof(eventreq))) + return -EFAULT; + + return 0; + +out_free_irq: + free_irq(le->irq, le); +out_free_desc: + gpiod_free(le->desc); +out_free_label: + kfree(le->label); +out_free_le: + kfree(le); + put_device(&gdev->dev); + return ret; +} + /** * gpio_ioctl() - ioctl handler for the GPIO chardev */ @@ -578,6 +874,8 @@ static long gpio_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return 0; } else if (cmd == GPIO_GET_LINEHANDLE_IOCTL) { return linehandle_create(gdev, ip); + } else if (cmd == GPIO_GET_LINEEVENT_IOCTL) { + return lineevent_create(gdev, ip); } return -EINVAL; } diff --git a/include/uapi/linux/gpio.h b/include/uapi/linux/gpio.h index 21905531dcf4..333d3544c964 100644 --- a/include/uapi/linux/gpio.h +++ b/include/uapi/linux/gpio.h @@ -55,7 +55,7 @@ struct gpioline_info { /* Maximum number of requested handles */ #define GPIOHANDLES_MAX 64 -/* Request flags */ +/* Linerequest flags */ #define GPIOHANDLE_REQUEST_INPUT (1UL << 0) #define GPIOHANDLE_REQUEST_OUTPUT (1UL << 1) #define GPIOHANDLE_REQUEST_ACTIVE_LOW (1UL << 2) @@ -93,10 +93,6 @@ struct gpiohandle_request { int fd; }; -#define GPIO_GET_CHIPINFO_IOCTL _IOR(0xB4, 0x01, struct gpiochip_info) -#define GPIO_GET_LINEINFO_IOCTL _IOWR(0xB4, 0x02, struct gpioline_info) -#define GPIO_GET_LINEHANDLE_IOCTL _IOWR(0xB4, 0x03, struct gpiohandle_request) - /** * struct gpiohandle_data - Information of values on a GPIO handle * @values: when getting the state of lines this contains the current @@ -110,4 +106,52 @@ struct gpiohandle_data { #define GPIOHANDLE_GET_LINE_VALUES_IOCTL _IOWR(0xB4, 0x08, struct gpiohandle_data) #define GPIOHANDLE_SET_LINE_VALUES_IOCTL _IOWR(0xB4, 0x09, struct gpiohandle_data) +/* Eventrequest flags */ +#define GPIOEVENT_REQUEST_RISING_EDGE (1UL << 0) +#define GPIOEVENT_REQUEST_FALLING_EDGE (1UL << 1) +#define GPIOEVENT_REQUEST_BOTH_EDGES ((1UL << 0) | (1UL << 1)) + +/** + * struct gpioevent_request - Information about a GPIO event request + * @lineoffset: the desired line to subscribe to events from, specified by + * offset index for the associated GPIO device + * @handleflags: desired handle flags for the desired GPIO line, such as + * GPIOHANDLE_REQUEST_ACTIVE_LOW or GPIOHANDLE_REQUEST_OPEN_DRAIN + * @eventflags: desired flags for the desired GPIO event line, such as + * GPIOEVENT_REQUEST_RISING_EDGE or GPIOEVENT_REQUEST_FALLING_EDGE + * @consumer_label: a desired consumer label for the selected GPIO line(s) + * such as "my-listener" + * @fd: if successful this field will contain a valid anonymous file handle + * after a GPIO_GET_LINEEVENT_IOCTL operation, zero or negative value + * means error + */ +struct gpioevent_request { + __u32 lineoffset; + __u32 handleflags; + __u32 eventflags; + char consumer_label[32]; + int fd; +}; + +/** + * GPIO event types + */ +#define GPIOEVENT_EVENT_RISING_EDGE 0x01 +#define GPIOEVENT_EVENT_FALLING_EDGE 0x02 + +/** + * struct gpioevent_data - The actual event being pushed to userspace + * @timestamp: best estimate of time of event occurrence, in nanoseconds + * @id: event identifier + */ +struct gpioevent_data { + __u64 timestamp; + __u32 id; +}; + +#define GPIO_GET_CHIPINFO_IOCTL _IOR(0xB4, 0x01, struct gpiochip_info) +#define GPIO_GET_LINEINFO_IOCTL _IOWR(0xB4, 0x02, struct gpioline_info) +#define GPIO_GET_LINEHANDLE_IOCTL _IOWR(0xB4, 0x03, struct gpiohandle_request) +#define GPIO_GET_LINEEVENT_IOCTL _IOWR(0xB4, 0x04, struct gpioevent_request) + #endif /* _UAPI_GPIO_H_ */ -- cgit v1.2.3 From 22a59be8b7693eb2d0897a9638f5991f2f8e4ddd Mon Sep 17 00:00:00 2001 From: Philip Prindeville Date: Tue, 14 Jun 2016 15:53:02 -0600 Subject: net: ipv4: Add ability to have GRE ignore DF bit in IPv4 payloads In the presence of firewalls which improperly block ICMP Unreachable (including Fragmentation Required) messages, Path MTU Discovery is prevented from working. A workaround is to handle IPv4 payloads opaquely, ignoring the DF bit--as is done for other payloads like AppleTalk--and doing transparent fragmentation and reassembly. Redux includes the enforcement of mutual exclusion between this feature and Path MTU Discovery as suggested by Alexander Duyck. Cc: Alexander Duyck Reviewed-by: Stephen Hemminger Signed-off-by: Philip Prindeville Signed-off-by: David S. Miller --- include/net/ip_tunnels.h | 1 + include/uapi/linux/if_tunnel.h | 1 + net/ipv4/ip_gre.c | 42 +++++++++++++++++++++++++++++++++--------- net/ipv4/ip_tunnel.c | 2 +- 4 files changed, 36 insertions(+), 10 deletions(-) (limited to 'include/uapi') diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index dbf444428437..9222678426a1 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -132,6 +132,7 @@ struct ip_tunnel { int ip_tnl_net_id; struct gro_cells gro_cells; bool collect_md; + bool ignore_df; }; #define TUNNEL_CSUM __cpu_to_be16(0x01) diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index af4de90ba27d..1046f5515174 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -113,6 +113,7 @@ enum { IFLA_GRE_ENCAP_SPORT, IFLA_GRE_ENCAP_DPORT, IFLA_GRE_COLLECT_METADATA, + IFLA_GRE_IGNORE_DF, __IFLA_GRE_MAX, }; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 4d2025f7ec57..0f8ca3fca00a 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -841,17 +841,19 @@ out: return ipgre_tunnel_validate(tb, data); } -static void ipgre_netlink_parms(struct net_device *dev, +static int ipgre_netlink_parms(struct net_device *dev, struct nlattr *data[], struct nlattr *tb[], struct ip_tunnel_parm *parms) { + struct ip_tunnel *t = netdev_priv(dev); + memset(parms, 0, sizeof(*parms)); parms->iph.protocol = IPPROTO_GRE; if (!data) - return; + return 0; if (data[IFLA_GRE_LINK]) parms->link = nla_get_u32(data[IFLA_GRE_LINK]); @@ -880,16 +882,26 @@ static void ipgre_netlink_parms(struct net_device *dev, if (data[IFLA_GRE_TOS]) parms->iph.tos = nla_get_u8(data[IFLA_GRE_TOS]); - if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC])) + if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC])) { + if (t->ignore_df) + return -EINVAL; parms->iph.frag_off = htons(IP_DF); + } if (data[IFLA_GRE_COLLECT_METADATA]) { - struct ip_tunnel *t = netdev_priv(dev); - t->collect_md = true; if (dev->type == ARPHRD_IPGRE) dev->type = ARPHRD_NONE; } + + if (data[IFLA_GRE_IGNORE_DF]) { + if (nla_get_u8(data[IFLA_GRE_IGNORE_DF]) + && (parms->iph.frag_off & htons(IP_DF))) + return -EINVAL; + t->ignore_df = !!nla_get_u8(data[IFLA_GRE_IGNORE_DF]); + } + + return 0; } /* This function returns true when ENCAP attributes are present in the nl msg */ @@ -960,16 +972,19 @@ static int ipgre_newlink(struct net *src_net, struct net_device *dev, { struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + int err; if (ipgre_netlink_encap_parms(data, &ipencap)) { struct ip_tunnel *t = netdev_priv(dev); - int err = ip_tunnel_encap_setup(t, &ipencap); + err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) return err; } - ipgre_netlink_parms(dev, data, tb, &p); + err = ipgre_netlink_parms(dev, data, tb, &p); + if (err < 0) + return err; return ip_tunnel_newlink(dev, tb, &p); } @@ -978,16 +993,19 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], { struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + int err; if (ipgre_netlink_encap_parms(data, &ipencap)) { struct ip_tunnel *t = netdev_priv(dev); - int err = ip_tunnel_encap_setup(t, &ipencap); + err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) return err; } - ipgre_netlink_parms(dev, data, tb, &p); + err = ipgre_netlink_parms(dev, data, tb, &p); + if (err < 0) + return err; return ip_tunnel_changelink(dev, tb, &p); } @@ -1024,6 +1042,8 @@ static size_t ipgre_get_size(const struct net_device *dev) nla_total_size(2) + /* IFLA_GRE_COLLECT_METADATA */ nla_total_size(0) + + /* IFLA_GRE_IGNORE_DF */ + nla_total_size(1) + 0; } @@ -1057,6 +1077,9 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) t->encap.flags)) goto nla_put_failure; + if (nla_put_u8(skb, IFLA_GRE_IGNORE_DF, t->ignore_df)) + goto nla_put_failure; + if (t->collect_md) { if (nla_put_flag(skb, IFLA_GRE_COLLECT_METADATA)) goto nla_put_failure; @@ -1084,6 +1107,7 @@ static const struct nla_policy ipgre_policy[IFLA_GRE_MAX + 1] = { [IFLA_GRE_ENCAP_SPORT] = { .type = NLA_U16 }, [IFLA_GRE_ENCAP_DPORT] = { .type = NLA_U16 }, [IFLA_GRE_COLLECT_METADATA] = { .type = NLA_FLAG }, + [IFLA_GRE_IGNORE_DF] = { .type = NLA_U8 }, }; static struct rtnl_link_ops ipgre_link_ops __read_mostly = { diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index d8f5e0a269f5..95649ebd2874 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -682,7 +682,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, } df = tnl_params->frag_off; - if (skb->protocol == htons(ETH_P_IP)) + if (skb->protocol == htons(ETH_P_IP) && !tunnel->ignore_df) df |= (inner_iph->frag_off&htons(IP_DF)); max_headroom = LL_RESERVED_SPACE(rt->dst.dev) + sizeof(struct iphdr) -- cgit v1.2.3 From 5fb384b066d7c320832c4541658e5c655c590ac5 Mon Sep 17 00:00:00 2001 From: Fabien Siron Date: Wed, 15 Jun 2016 15:37:49 +0000 Subject: netlink: Add comment to warn about deprecated netlink rings attribute request Signed-off-by: Fabien Siron Signed-off-by: David S. Miller --- include/uapi/linux/netlink_diag.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/netlink_diag.h b/include/uapi/linux/netlink_diag.h index d79399394b46..76b4d87c83a8 100644 --- a/include/uapi/linux/netlink_diag.h +++ b/include/uapi/linux/netlink_diag.h @@ -49,6 +49,7 @@ enum { #define NDIAG_SHOW_MEMINFO 0x00000001 /* show memory info of a socket */ #define NDIAG_SHOW_GROUPS 0x00000002 /* show groups of a netlink socket */ #ifndef __KERNEL__ +/* deprecated since 4.6 */ #define NDIAG_SHOW_RING_CFG 0x00000004 /* show ring configuration */ #endif -- cgit v1.2.3 From e456cd37bc28abe47dc65189df916ac0510ac1d4 Mon Sep 17 00:00:00 2001 From: Benjamin Tissoires Date: Thu, 9 Jun 2016 16:53:48 +0200 Subject: i2c: smbus: add SMBus Host Notify support SMBus Host Notify allows a slave device to act as a master on a bus to notify the host of an interrupt. On Intel chipsets, the functionality is directly implemented in the firmware. We just need to export a function to call .alert() on the proper device driver. i2c_handle_smbus_host_notify() behaves like i2c_handle_smbus_alert(). When called, it schedules a task that will be able to sleep to go through the list of devices attached to the adapter. The current implementation allows one Host Notification to be scheduled while an other is running. Tested-by: Andrew Duggan Signed-off-by: Benjamin Tissoires Signed-off-by: Wolfram Sang --- Documentation/i2c/smbus-protocol | 6 +++ drivers/i2c/i2c-smbus.c | 113 +++++++++++++++++++++++++++++++++++++-- include/linux/i2c-smbus.h | 44 +++++++++++++++ include/linux/i2c.h | 3 ++ include/uapi/linux/i2c.h | 1 + 5 files changed, 162 insertions(+), 5 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/i2c/smbus-protocol b/Documentation/i2c/smbus-protocol index 6012b12b3510..14d4ec1be245 100644 --- a/Documentation/i2c/smbus-protocol +++ b/Documentation/i2c/smbus-protocol @@ -199,6 +199,12 @@ alerting device's address. [S] [HostAddr] [Wr] A [DevAddr] A [DataLow] A [DataHigh] A [P] +This is implemented in the following way in the Linux kernel: +* I2C bus drivers which support SMBus Host Notify should call + i2c_setup_smbus_host_notify() to setup SMBus Host Notify support. +* I2C drivers for devices which can trigger SMBus Host Notify should implement + the optional alert() callback. + Packet Error Checking (PEC) =========================== diff --git a/drivers/i2c/i2c-smbus.c b/drivers/i2c/i2c-smbus.c index 3b6765a4ebe9..f574995b41c1 100644 --- a/drivers/i2c/i2c-smbus.c +++ b/drivers/i2c/i2c-smbus.c @@ -33,7 +33,8 @@ struct i2c_smbus_alert { struct alert_data { unsigned short addr; - u8 flag:1; + enum i2c_alert_protocol type; + unsigned int data; }; /* If this is the alerting device, notify its driver */ @@ -56,8 +57,7 @@ static int smbus_do_alert(struct device *dev, void *addrp) if (client->dev.driver) { driver = to_i2c_driver(client->dev.driver); if (driver->alert) - driver->alert(client, I2C_PROTOCOL_SMBUS_ALERT, - data->flag); + driver->alert(client, data->type, data->data); else dev_warn(&client->dev, "no driver alert()!\n"); } else @@ -97,8 +97,9 @@ static void smbus_alert(struct work_struct *work) if (status < 0) break; - data.flag = status & 1; + data.data = status & 1; data.addr = status >> 1; + data.type = I2C_PROTOCOL_SMBUS_ALERT; if (data.addr == prev_addr) { dev_warn(&ara->dev, "Duplicate SMBALERT# from dev " @@ -106,7 +107,7 @@ static void smbus_alert(struct work_struct *work) break; } dev_dbg(&ara->dev, "SMBALERT# from dev 0x%02x, flag %d\n", - data.addr, data.flag); + data.addr, data.data); /* Notify driver for the device which issued the alert */ device_for_each_child(&ara->adapter->dev, &data, @@ -240,6 +241,108 @@ int i2c_handle_smbus_alert(struct i2c_client *ara) } EXPORT_SYMBOL_GPL(i2c_handle_smbus_alert); +static void smbus_host_notify_work(struct work_struct *work) +{ + struct alert_data alert; + struct i2c_adapter *adapter; + unsigned long flags; + u16 payload; + u8 addr; + struct smbus_host_notify *data; + + data = container_of(work, struct smbus_host_notify, work); + + spin_lock_irqsave(&data->lock, flags); + payload = data->payload; + addr = data->addr; + adapter = data->adapter; + + /* clear the pending bit and release the spinlock */ + data->pending = false; + spin_unlock_irqrestore(&data->lock, flags); + + if (!adapter || !addr) + return; + + alert.type = I2C_PROTOCOL_SMBUS_HOST_NOTIFY; + alert.addr = addr; + alert.data = payload; + + device_for_each_child(&adapter->dev, &alert, smbus_do_alert); +} + +/** + * i2c_setup_smbus_host_notify - Allocate a new smbus_host_notify for the given + * I2C adapter. + * @adapter: the adapter we want to associate a Host Notify function + * + * Returns a struct smbus_host_notify pointer on success, and NULL on failure. + * The resulting smbus_host_notify must not be freed afterwards, it is a + * managed resource already. + */ +struct smbus_host_notify *i2c_setup_smbus_host_notify(struct i2c_adapter *adap) +{ + struct smbus_host_notify *host_notify; + + host_notify = devm_kzalloc(&adap->dev, sizeof(struct smbus_host_notify), + GFP_KERNEL); + if (!host_notify) + return NULL; + + host_notify->adapter = adap; + + spin_lock_init(&host_notify->lock); + INIT_WORK(&host_notify->work, smbus_host_notify_work); + + return host_notify; +} +EXPORT_SYMBOL_GPL(i2c_setup_smbus_host_notify); + +/** + * i2c_handle_smbus_host_notify - Forward a Host Notify event to the correct + * I2C client. + * @host_notify: the struct host_notify attached to the relevant adapter + * @data: the Host Notify data which contains the payload and address of the + * client + * Context: can't sleep + * + * Helper function to be called from an I2C bus driver's interrupt + * handler. It will schedule the Host Notify work, in turn calling the + * corresponding I2C device driver's alert function. + * + * host_notify should be a valid pointer previously returned by + * i2c_setup_smbus_host_notify(). + */ +int i2c_handle_smbus_host_notify(struct smbus_host_notify *host_notify, + unsigned short addr, unsigned int data) +{ + unsigned long flags; + struct i2c_adapter *adapter; + + if (!host_notify || !host_notify->adapter) + return -EINVAL; + + adapter = host_notify->adapter; + + spin_lock_irqsave(&host_notify->lock, flags); + + if (host_notify->pending) { + spin_unlock_irqrestore(&host_notify->lock, flags); + dev_warn(&adapter->dev, "Host Notify already scheduled.\n"); + return -EBUSY; + } + + host_notify->payload = data; + host_notify->addr = addr; + + /* Mark that there is a pending notification and release the lock */ + host_notify->pending = true; + spin_unlock_irqrestore(&host_notify->lock, flags); + + return schedule_work(&host_notify->work); +} +EXPORT_SYMBOL_GPL(i2c_handle_smbus_host_notify); + module_i2c_driver(smbalert_driver); MODULE_AUTHOR("Jean Delvare "); diff --git a/include/linux/i2c-smbus.h b/include/linux/i2c-smbus.h index 8f1b086ca5bc..4ac95bbe53ef 100644 --- a/include/linux/i2c-smbus.h +++ b/include/linux/i2c-smbus.h @@ -23,6 +23,8 @@ #define _LINUX_I2C_SMBUS_H #include +#include +#include /** @@ -48,4 +50,46 @@ struct i2c_client *i2c_setup_smbus_alert(struct i2c_adapter *adapter, struct i2c_smbus_alert_setup *setup); int i2c_handle_smbus_alert(struct i2c_client *ara); +/** + * smbus_host_notify - internal structure used by the Host Notify mechanism. + * @adapter: the I2C adapter associated with this struct + * @work: worker used to schedule the IRQ in the slave device + * @lock: spinlock to check if a notification is already pending + * @pending: flag set when a notification is pending (any new notification will + * be rejected if pending is true) + * @payload: the actual payload of the Host Notify event + * @addr: the address of the slave device which raised the notification + * + * This struct needs to be allocated by i2c_setup_smbus_host_notify() and does + * not need to be freed. Internally, i2c_setup_smbus_host_notify() uses a + * managed resource to clean this up when the adapter get released. + */ +struct smbus_host_notify { + struct i2c_adapter *adapter; + struct work_struct work; + spinlock_t lock; + bool pending; + u16 payload; + u8 addr; +}; + +#if IS_ENABLED(CONFIG_I2C_SMBUS) +struct smbus_host_notify *i2c_setup_smbus_host_notify(struct i2c_adapter *adap); +int i2c_handle_smbus_host_notify(struct smbus_host_notify *host_notify, + unsigned short addr, unsigned int data); +#else +static inline struct smbus_host_notify * +i2c_setup_smbus_host_notify(struct i2c_adapter *adap) +{ + return NULL; +} + +static inline int +i2c_handle_smbus_host_notify(struct smbus_host_notify *host_notify, + unsigned short addr, unsigned int data) +{ + return 0; +} +#endif /* I2C_SMBUS */ + #endif /* _LINUX_I2C_SMBUS_H */ diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 37a45dcd6592..fffdc270ca18 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -128,6 +128,7 @@ i2c_smbus_read_i2c_block_data_or_emulated(const struct i2c_client *client, enum i2c_alert_protocol { I2C_PROTOCOL_SMBUS_ALERT, + I2C_PROTOCOL_SMBUS_HOST_NOTIFY, }; /** @@ -184,6 +185,8 @@ struct i2c_driver { * The format and meaning of the data value depends on the protocol. * For the SMBus alert protocol, there is a single bit of data passed * as the alert response's low bit ("event flag"). + * For the SMBus Host Notify protocol, the data corresponds to the + * 16-bit payload data reported by the slave device acting as master. */ void (*alert)(struct i2c_client *, enum i2c_alert_protocol protocol, unsigned int data); diff --git a/include/uapi/linux/i2c.h b/include/uapi/linux/i2c.h index adcbef4bff61..009e27bb9abe 100644 --- a/include/uapi/linux/i2c.h +++ b/include/uapi/linux/i2c.h @@ -102,6 +102,7 @@ struct i2c_msg { #define I2C_FUNC_SMBUS_WRITE_BLOCK_DATA 0x02000000 #define I2C_FUNC_SMBUS_READ_I2C_BLOCK 0x04000000 /* I2C-like block xfer */ #define I2C_FUNC_SMBUS_WRITE_I2C_BLOCK 0x08000000 /* w/ 1-byte reg. addr. */ +#define I2C_FUNC_SMBUS_HOST_NOTIFY 0x10000000 #define I2C_FUNC_SMBUS_BYTE (I2C_FUNC_SMBUS_READ_BYTE | \ I2C_FUNC_SMBUS_WRITE_BYTE) -- cgit v1.2.3 From 6f3b911d5f29b98752e5da86a295210c0c4f4e14 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 17 Jun 2016 15:35:27 +0200 Subject: can: bcm: add support for CAN FD frames The programming API of the CAN_BCM depends on struct can_frame which is given as array directly behind the bcm_msg_head structure. To follow this schema for the CAN FD frames a new flag 'CAN_FD_FRAME' in the bcm_msg_head flags indicates that the concatenated CAN frame structures behind the bcm_msg_head are defined as struct canfd_frame. This patch adds the support to handle CAN and CAN FD frames on a per BCM-op base. Main changes: - generally use struct canfd_frames instead if struct can_frames - use canfd_frame.flags instead of can_frame.can_dlc for private BCM flags - make all CAN frame sizes depending on the new CAN_FD_FRAME flags - separate between CAN and CAN FD when sending/receiving frames Due to the dependence of the CAN_FD_FRAME flag the former binary interface for classic CAN frames remains stable. Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde --- include/uapi/linux/can/bcm.h | 1 + net/can/bcm.c | 223 ++++++++++++++++++++++++++----------------- 2 files changed, 136 insertions(+), 88 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/can/bcm.h b/include/uapi/linux/can/bcm.h index 7a291dc1ff15..cefb304414ba 100644 --- a/include/uapi/linux/can/bcm.h +++ b/include/uapi/linux/can/bcm.h @@ -99,5 +99,6 @@ enum { #define RX_ANNOUNCE_RESUME 0x0100 #define TX_RESET_MULTI_IDX 0x0200 #define RX_RTR_FRAME 0x0400 +#define CAN_FD_FRAME 0x0800 #endif /* !_UAPI_CAN_BCM_H */ diff --git a/net/can/bcm.c b/net/can/bcm.c index f3bf3875313a..8e999ffdf28b 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -1,7 +1,7 @@ /* * bcm.c - Broadcast Manager to filter/send (cyclic) CAN content * - * Copyright (c) 2002-2007 Volkswagen Group Electronic Research + * Copyright (c) 2002-2016 Volkswagen Group Electronic Research * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -67,27 +67,31 @@ */ #define MAX_NFRAMES 256 -/* use of last_frames[index].can_dlc */ +/* use of last_frames[index].flags */ #define RX_RECV 0x40 /* received data for this element */ #define RX_THR 0x80 /* element not been sent due to throttle feature */ -#define BCM_CAN_DLC_MASK 0x0F /* clean private flags in can_dlc by masking */ +#define BCM_CAN_FLAGS_MASK 0x3F /* to clean private flags after usage */ /* get best masking value for can_rx_register() for a given single can_id */ #define REGMASK(id) ((id & CAN_EFF_FLAG) ? \ (CAN_EFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG) : \ (CAN_SFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG)) -#define CAN_BCM_VERSION CAN_VERSION +#define CAN_BCM_VERSION "20160617" MODULE_DESCRIPTION("PF_CAN broadcast manager protocol"); MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Oliver Hartkopp "); MODULE_ALIAS("can-proto-2"); -/* easy access to CAN frame payload */ -static inline u64 GET_U64(const struct can_frame *cp) +/* + * easy access to the first 64 bit of can(fd)_frame payload. cp->data is + * 64 bit aligned so the offset has to be multiples of 8 which is ensured + * by the only callers in bcm_rx_cmp_to_index() bcm_rx_handler(). + */ +static inline u64 get_u64(const struct canfd_frame *cp, int offset) { - return *(u64 *)cp->data; + return *(u64 *)(cp->data + offset); } struct bcm_op { @@ -101,13 +105,14 @@ struct bcm_op { struct tasklet_struct tsklet, thrtsklet; ktime_t rx_stamp, kt_ival1, kt_ival2, kt_lastmsg; int rx_ifindex; + int cfsiz; u32 count; u32 nframes; u32 currframe; - struct can_frame *frames; - struct can_frame *last_frames; - struct can_frame sframe; - struct can_frame last_sframe; + struct canfd_frame *frames; + struct canfd_frame *last_frames; + struct canfd_frame sframe; + struct canfd_frame last_sframe; struct sock *sk; struct net_device *rx_reg_dev; }; @@ -136,7 +141,7 @@ static inline ktime_t bcm_timeval_to_ktime(struct bcm_timeval tv) return ktime_set(tv.tv_sec, tv.tv_usec * NSEC_PER_USEC); } -#define CFSIZ sizeof(struct can_frame) +#define CFSIZ(flags) ((flags & CAN_FD_FRAME) ? CANFD_MTU : CAN_MTU) #define OPSIZ sizeof(struct bcm_op) #define MHSIZ sizeof(struct bcm_msg_head) @@ -183,10 +188,16 @@ static int bcm_proc_show(struct seq_file *m, void *v) if (!op->frames_abs) continue; - seq_printf(m, "rx_op: %03X %-5s ", - op->can_id, bcm_proc_getifname(ifname, op->ifindex)); - seq_printf(m, "[%u]%c ", op->nframes, - (op->flags & RX_CHECK_DLC) ? 'd' : ' '); + seq_printf(m, "rx_op: %03X %-5s ", op->can_id, + bcm_proc_getifname(ifname, op->ifindex)); + + if (op->flags & CAN_FD_FRAME) + seq_printf(m, "(%u)", op->nframes); + else + seq_printf(m, "[%u]", op->nframes); + + seq_printf(m, "%c ", (op->flags & RX_CHECK_DLC) ? 'd' : ' '); + if (op->kt_ival1.tv64) seq_printf(m, "timeo=%lld ", (long long)ktime_to_us(op->kt_ival1)); @@ -206,10 +217,13 @@ static int bcm_proc_show(struct seq_file *m, void *v) list_for_each_entry(op, &bo->tx_ops, list) { - seq_printf(m, "tx_op: %03X %s [%u] ", - op->can_id, - bcm_proc_getifname(ifname, op->ifindex), - op->nframes); + seq_printf(m, "tx_op: %03X %s ", op->can_id, + bcm_proc_getifname(ifname, op->ifindex)); + + if (op->flags & CAN_FD_FRAME) + seq_printf(m, "(%u) ", op->nframes); + else + seq_printf(m, "[%u] ", op->nframes); if (op->kt_ival1.tv64) seq_printf(m, "t1=%lld ", @@ -246,7 +260,7 @@ static void bcm_can_tx(struct bcm_op *op) { struct sk_buff *skb; struct net_device *dev; - struct can_frame *cf = &op->frames[op->currframe]; + struct canfd_frame *cf = op->frames + op->cfsiz * op->currframe; /* no target device? => exit */ if (!op->ifindex) @@ -258,7 +272,7 @@ static void bcm_can_tx(struct bcm_op *op) return; } - skb = alloc_skb(CFSIZ + sizeof(struct can_skb_priv), gfp_any()); + skb = alloc_skb(op->cfsiz + sizeof(struct can_skb_priv), gfp_any()); if (!skb) goto out; @@ -266,7 +280,7 @@ static void bcm_can_tx(struct bcm_op *op) can_skb_prv(skb)->ifindex = dev->ifindex; can_skb_prv(skb)->skbcnt = 0; - memcpy(skb_put(skb, CFSIZ), cf, CFSIZ); + memcpy(skb_put(skb, op->cfsiz), cf, op->cfsiz); /* send with loopback */ skb->dev = dev; @@ -289,13 +303,13 @@ out: * (consisting of bcm_msg_head + x CAN frames) */ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, - struct can_frame *frames, int has_timestamp) + struct canfd_frame *frames, int has_timestamp) { struct sk_buff *skb; - struct can_frame *firstframe; + struct canfd_frame *firstframe; struct sockaddr_can *addr; struct sock *sk = op->sk; - unsigned int datalen = head->nframes * CFSIZ; + unsigned int datalen = head->nframes * op->cfsiz; int err; skb = alloc_skb(sizeof(*head) + datalen, gfp_any()); @@ -306,18 +320,18 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, if (head->nframes) { /* CAN frames starting here */ - firstframe = (struct can_frame *)skb_tail_pointer(skb); + firstframe = (struct canfd_frame *)skb_tail_pointer(skb); memcpy(skb_put(skb, datalen), frames, datalen); /* - * the BCM uses the can_dlc-element of the CAN frame + * the BCM uses the flags-element of the canfd_frame * structure for internal purposes. This is only * relevant for updates that are generated by the * BCM, where nframes is 1 */ if (head->nframes == 1) - firstframe->can_dlc &= BCM_CAN_DLC_MASK; + firstframe->flags &= BCM_CAN_FLAGS_MASK; } if (has_timestamp) { @@ -404,7 +418,7 @@ static enum hrtimer_restart bcm_tx_timeout_handler(struct hrtimer *hrtimer) /* * bcm_rx_changed - create a RX_CHANGED notification due to changed content */ -static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) +static void bcm_rx_changed(struct bcm_op *op, struct canfd_frame *data) { struct bcm_msg_head head; @@ -416,7 +430,7 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) op->frames_filtered = op->frames_abs = 0; /* this element is not throttled anymore */ - data->can_dlc &= (BCM_CAN_DLC_MASK|RX_RECV); + data->flags &= (BCM_CAN_FLAGS_MASK|RX_RECV); head.opcode = RX_CHANGED; head.flags = op->flags; @@ -435,13 +449,13 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) * 2. send a notification to the user (if possible) */ static void bcm_rx_update_and_send(struct bcm_op *op, - struct can_frame *lastdata, - const struct can_frame *rxdata) + struct canfd_frame *lastdata, + const struct canfd_frame *rxdata) { - memcpy(lastdata, rxdata, CFSIZ); + memcpy(lastdata, rxdata, op->cfsiz); /* mark as used and throttled by default */ - lastdata->can_dlc |= (RX_RECV|RX_THR); + lastdata->flags |= (RX_RECV|RX_THR); /* throttling mode inactive ? */ if (!op->kt_ival2.tv64) { @@ -479,33 +493,36 @@ rx_changed_settime: * received data stored in op->last_frames[] */ static void bcm_rx_cmp_to_index(struct bcm_op *op, unsigned int index, - const struct can_frame *rxdata) + const struct canfd_frame *rxdata) { + struct canfd_frame *cf = op->frames + op->cfsiz * index; + struct canfd_frame *lcf = op->last_frames + op->cfsiz * index; + int i; + /* - * no one uses the MSBs of can_dlc for comparison, + * no one uses the MSBs of flags for comparison, * so we use it here to detect the first time of reception */ - if (!(op->last_frames[index].can_dlc & RX_RECV)) { + if (!(lcf->flags & RX_RECV)) { /* received data for the first time => send update to user */ - bcm_rx_update_and_send(op, &op->last_frames[index], rxdata); + bcm_rx_update_and_send(op, lcf, rxdata); return; } /* do a real check in CAN frame data section */ - - if ((GET_U64(&op->frames[index]) & GET_U64(rxdata)) != - (GET_U64(&op->frames[index]) & GET_U64(&op->last_frames[index]))) { - bcm_rx_update_and_send(op, &op->last_frames[index], rxdata); - return; + for (i = 0; i < rxdata->len; i += 8) { + if ((get_u64(cf, i) & get_u64(rxdata, i)) != + (get_u64(cf, i) & get_u64(lcf, i))) { + bcm_rx_update_and_send(op, lcf, rxdata); + return; + } } if (op->flags & RX_CHECK_DLC) { - /* do a real check in CAN frame dlc */ - if (rxdata->can_dlc != (op->last_frames[index].can_dlc & - BCM_CAN_DLC_MASK)) { - bcm_rx_update_and_send(op, &op->last_frames[index], - rxdata); + /* do a real check in CAN frame length */ + if (rxdata->len != lcf->len) { + bcm_rx_update_and_send(op, lcf, rxdata); return; } } @@ -555,7 +572,7 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) /* if user wants to be informed, when cyclic CAN-Messages come back */ if ((op->flags & RX_ANNOUNCE_RESUME) && op->last_frames) { /* clear received CAN frames to indicate 'nothing received' */ - memset(op->last_frames, 0, op->nframes * CFSIZ); + memset(op->last_frames, 0, op->nframes * op->cfsiz); } return HRTIMER_NORESTART; @@ -567,9 +584,11 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) static inline int bcm_rx_do_flush(struct bcm_op *op, int update, unsigned int index) { - if ((op->last_frames) && (op->last_frames[index].can_dlc & RX_THR)) { + struct canfd_frame *lcf = op->last_frames + op->cfsiz * index; + + if ((op->last_frames) && (lcf->flags & RX_THR)) { if (update) - bcm_rx_changed(op, &op->last_frames[index]); + bcm_rx_changed(op, lcf); return 1; } return 0; @@ -634,15 +653,19 @@ static enum hrtimer_restart bcm_rx_thr_handler(struct hrtimer *hrtimer) static void bcm_rx_handler(struct sk_buff *skb, void *data) { struct bcm_op *op = (struct bcm_op *)data; - const struct can_frame *rxframe = (struct can_frame *)skb->data; + const struct canfd_frame *rxframe = (struct canfd_frame *)skb->data; unsigned int i; - /* disable timeout */ - hrtimer_cancel(&op->timer); - if (op->can_id != rxframe->can_id) return; + /* make sure to handle the correct frame type (CAN / CAN FD) */ + if (skb->len != op->cfsiz) + return; + + /* disable timeout */ + hrtimer_cancel(&op->timer); + /* save rx timestamp */ op->rx_stamp = skb->tstamp; /* save originator for recvfrom() */ @@ -673,13 +696,14 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) * multiplex compare * * find the first multiplex mask that fits. - * Remark: The MUX-mask is stored in index 0 + * Remark: The MUX-mask is stored in index 0 - but only the + * first 64 bits of the frame data[] are relevant (CAN FD) */ for (i = 1; i < op->nframes; i++) { - if ((GET_U64(&op->frames[0]) & GET_U64(rxframe)) == - (GET_U64(&op->frames[0]) & - GET_U64(&op->frames[i]))) { + if ((get_u64(op->frames, 0) & get_u64(rxframe, 0)) == + (get_u64(op->frames, 0) & + get_u64(op->frames + op->cfsiz * i, 0))) { bcm_rx_cmp_to_index(op, i, rxframe); break; } @@ -699,7 +723,8 @@ static struct bcm_op *bcm_find_op(struct list_head *ops, struct bcm_op *op; list_for_each_entry(op, ops, list) { - if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex) && + (op->flags & CAN_FD_FRAME) == (mh->flags & CAN_FD_FRAME)) return op; } @@ -748,7 +773,8 @@ static int bcm_delete_rx_op(struct list_head *ops, struct bcm_msg_head *mh, struct bcm_op *op, *n; list_for_each_entry_safe(op, n, ops, list) { - if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) { + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex) && + (op->flags & CAN_FD_FRAME) == (mh->flags & CAN_FD_FRAME)) { /* * Don't care if we're bound or not (due to netdev @@ -794,7 +820,8 @@ static int bcm_delete_tx_op(struct list_head *ops, struct bcm_msg_head *mh, struct bcm_op *op, *n; list_for_each_entry_safe(op, n, ops, list) { - if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) { + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex) && + (op->flags & CAN_FD_FRAME) == (mh->flags & CAN_FD_FRAME)) { list_del(&op->list); bcm_remove_op(op); return 1; /* done */ @@ -835,6 +862,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, { struct bcm_sock *bo = bcm_sk(sk); struct bcm_op *op; + struct canfd_frame *cf; unsigned int i; int err; @@ -861,19 +889,27 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* update CAN frames content */ for (i = 0; i < msg_head->nframes; i++) { - err = memcpy_from_msg((u8 *)&op->frames[i], msg, CFSIZ); - if (op->frames[i].can_dlc > 8) - err = -EINVAL; + cf = op->frames + op->cfsiz * i; + err = memcpy_from_msg((u8 *)cf, msg, op->cfsiz); + + if (op->flags & CAN_FD_FRAME) { + if (cf->len > 64) + err = -EINVAL; + } else { + if (cf->len > 8) + err = -EINVAL; + } if (err < 0) return err; if (msg_head->flags & TX_CP_CAN_ID) { /* copy can_id into frame */ - op->frames[i].can_id = msg_head->can_id; + cf->can_id = msg_head->can_id; } } + op->flags = msg_head->flags; } else { /* insert new BCM operation for the given can_id */ @@ -882,11 +918,13 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (!op) return -ENOMEM; - op->can_id = msg_head->can_id; + op->can_id = msg_head->can_id; + op->cfsiz = CFSIZ(msg_head->flags); + op->flags = msg_head->flags; /* create array for CAN frames and copy the data */ if (msg_head->nframes > 1) { - op->frames = kmalloc(msg_head->nframes * CFSIZ, + op->frames = kmalloc(msg_head->nframes * op->cfsiz, GFP_KERNEL); if (!op->frames) { kfree(op); @@ -896,10 +934,17 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, op->frames = &op->sframe; for (i = 0; i < msg_head->nframes; i++) { - err = memcpy_from_msg((u8 *)&op->frames[i], msg, CFSIZ); - if (op->frames[i].can_dlc > 8) - err = -EINVAL; + cf = op->frames + op->cfsiz * i; + err = memcpy_from_msg((u8 *)cf, msg, op->cfsiz); + + if (op->flags & CAN_FD_FRAME) { + if (cf->len > 64) + err = -EINVAL; + } else { + if (cf->len > 8) + err = -EINVAL; + } if (err < 0) { if (op->frames != &op->sframe) @@ -910,7 +955,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (msg_head->flags & TX_CP_CAN_ID) { /* copy can_id into frame */ - op->frames[i].can_id = msg_head->can_id; + cf->can_id = msg_head->can_id; } } @@ -945,8 +990,6 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* check flags */ - op->flags = msg_head->flags; - if (op->flags & TX_RESET_MULTI_IDX) { /* start multiple frame transmission with index 0 */ op->currframe = 0; @@ -980,7 +1023,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (op->flags & STARTTIMER) bcm_tx_start_timer(op); - return msg_head->nframes * CFSIZ + MHSIZ; + return msg_head->nframes * op->cfsiz + MHSIZ; } /* @@ -1026,15 +1069,16 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (msg_head->nframes) { /* update CAN frames content */ err = memcpy_from_msg((u8 *)op->frames, msg, - msg_head->nframes * CFSIZ); + msg_head->nframes * op->cfsiz); if (err < 0) return err; /* clear last_frames to indicate 'nothing received' */ - memset(op->last_frames, 0, msg_head->nframes * CFSIZ); + memset(op->last_frames, 0, msg_head->nframes * op->cfsiz); } op->nframes = msg_head->nframes; + op->flags = msg_head->flags; /* Only an update -> do not call can_rx_register() */ do_rx_register = 0; @@ -1045,12 +1089,14 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (!op) return -ENOMEM; - op->can_id = msg_head->can_id; - op->nframes = msg_head->nframes; + op->can_id = msg_head->can_id; + op->nframes = msg_head->nframes; + op->cfsiz = CFSIZ(msg_head->flags); + op->flags = msg_head->flags; if (msg_head->nframes > 1) { /* create array for CAN frames and copy the data */ - op->frames = kmalloc(msg_head->nframes * CFSIZ, + op->frames = kmalloc(msg_head->nframes * op->cfsiz, GFP_KERNEL); if (!op->frames) { kfree(op); @@ -1058,7 +1104,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } /* create and init array for received CAN frames */ - op->last_frames = kzalloc(msg_head->nframes * CFSIZ, + op->last_frames = kzalloc(msg_head->nframes * op->cfsiz, GFP_KERNEL); if (!op->last_frames) { kfree(op->frames); @@ -1073,7 +1119,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (msg_head->nframes) { err = memcpy_from_msg((u8 *)op->frames, msg, - msg_head->nframes * CFSIZ); + msg_head->nframes * op->cfsiz); if (err < 0) { if (op->frames != &op->sframe) kfree(op->frames); @@ -1115,7 +1161,6 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } /* if ((op = bcm_find_op(&bo->rx_ops, msg_head->can_id, ifindex))) */ /* check flags */ - op->flags = msg_head->flags; if (op->flags & RX_RTR_FRAME) { @@ -1187,7 +1232,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } } - return msg_head->nframes * CFSIZ + MHSIZ; + return msg_head->nframes * op->cfsiz + MHSIZ; } /* @@ -1244,6 +1289,7 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) struct bcm_sock *bo = bcm_sk(sk); int ifindex = bo->ifindex; /* default ifindex for this bcm_op */ struct bcm_msg_head msg_head; + int cfsiz; int ret; /* read bytes or error codes as return value */ if (!bo->bound) @@ -1258,7 +1304,8 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) if (ret < 0) return ret; - if ((size - MHSIZ) % CFSIZ) + cfsiz = CFSIZ(msg_head.flags); + if ((size - MHSIZ) % cfsiz) return -EINVAL; /* check for alternative ifindex for this bcm_op */ @@ -1332,10 +1379,10 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) case TX_SEND: /* we need exactly one CAN frame behind the msg head */ - if ((msg_head.nframes != 1) || (size != CFSIZ + MHSIZ)) + if ((msg_head.nframes != 1) || (size != cfsiz + MHSIZ)) ret = -EINVAL; else - ret = bcm_tx_send(msg, ifindex, sk, CFSIZ); + ret = bcm_tx_send(msg, ifindex, sk, cfsiz); break; default: -- cgit v1.2.3 From 20e1954fe238dbe5f8d3a979e593fe352bd703cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:06 -0700 Subject: ipv6: RFC 4884 partial support for SIT/GRE tunnels When receiving an ICMPv4 message containing extensions as defined in RFC 4884, and translating it to ICMPv6 at SIT or GRE tunnel, we need some extra manipulation in order to properly forward the extensions. This patch only takes care of Time Exceeded messages as they are the ones that typically carry information from various routers in a fabric during a traceroute session. It also avoids complex skb logic if the data_len is not a multiple of 8. RFC states : The "original datagram" field MUST contain at least 128 octets. If the original datagram did not contain 128 octets, the "original datagram" field MUST be zero padded to 128 octets. In practice routers use 128 bytes of original datagram, not more. Initial translation was added in commit ca15a078bd90 ("sit: generate icmpv6 error when receiving icmpv4 error") Signed-off-by: Eric Dumazet Cc: Oussama Ghorbel Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 3 ++- include/uapi/linux/icmp.h | 1 + net/ipv4/ip_gre.c | 5 ++++- net/ipv6/icmp.c | 28 ++++++++++++++++++++++++---- net/ipv6/sit.c | 4 +++- 5 files changed, 34 insertions(+), 7 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 97ae98071a03..57086e9fc64c 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,7 +18,8 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, + unsigned int data_len); #else diff --git a/include/uapi/linux/icmp.h b/include/uapi/linux/icmp.h index 16fff055f734..fddd9d736284 100644 --- a/include/uapi/linux/icmp.h +++ b/include/uapi/linux/icmp.h @@ -79,6 +79,7 @@ struct icmphdr { __be16 __unused; __be16 mtu; } frag; + __u8 reserved[4]; } un; }; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index ab4cff8e563d..8eec78f53f9e 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -144,6 +144,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info, const struct iphdr *iph; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; + unsigned int data_len = 0; struct ip_tunnel *t; switch (type) { @@ -169,6 +170,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info, case ICMP_TIME_EXCEEDED: if (code != ICMP_EXC_TTL) return; + data_len = icmp_hdr(skb)->un.reserved[1] * 4; /* RFC 4884 4.1 */ break; case ICMP_REDIRECT: @@ -189,7 +191,8 @@ static void ipgre_err(struct sk_buff *skb, u32 info, #if IS_ENABLED(CONFIG_IPV6) if (tpi->proto == htons(ETH_P_IPV6) && - !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, type)) + !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, + type, data_len)) return; #endif diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 867aebc34248..fd11f5856ce8 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -564,16 +564,22 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) * Either an IPv4 header for SIT encap * an IPv4 header + GRE header for GRE encap */ -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, + unsigned int data_len) { struct in6_addr temp_saddr; struct rt6_info *rt; struct sk_buff *skb2; + u32 info = 0; if (!pskb_may_pull(skb, nhs + sizeof(struct ipv6hdr) + 8)) return 1; - skb2 = skb_clone(skb, GFP_ATOMIC); + /* RFC 4884 (partial) support for ICMP extensions */ + if (data_len < 128 || (data_len & 7) || skb->len < data_len) + data_len = 0; + + skb2 = data_len ? skb_copy(skb, GFP_ATOMIC) : skb_clone(skb, GFP_ATOMIC); if (!skb2) return 1; @@ -588,12 +594,26 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) skb2->dev = rt->dst.dev; ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr, &temp_saddr); + + if (data_len) { + /* RFC 4884 (partial) support : + * insert 0 padding at the end, before the extensions + */ + __skb_push(skb2, nhs); + skb_reset_network_header(skb2); + memmove(skb2->data, skb2->data + nhs, data_len - nhs); + memset(skb2->data + data_len - nhs, 0, nhs); + /* RFC 4884 4.5 : Length is measured in 64-bit words, + * and stored in reserved[0] + */ + info = (data_len/8) << 24; + } if (type == ICMP_TIME_EXCEEDED) icmp6_send(skb2, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, - 0, &temp_saddr); + info, &temp_saddr); else icmp6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, - 0, &temp_saddr); + info, &temp_saddr); if (rt) ip6_rt_put(rt); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index d7a36114eb50..cdd714690f95 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -484,6 +484,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) const struct iphdr *iph = (const struct iphdr *)skb->data; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; + unsigned int data_len = 0; struct ip_tunnel *t; int err; @@ -508,6 +509,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) case ICMP_TIME_EXCEEDED: if (code != ICMP_EXC_TTL) return 0; + data_len = icmp_hdr(skb)->un.reserved[1] * 4; /* RFC 4884 4.1 */ break; case ICMP_REDIRECT: break; @@ -536,7 +538,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) } err = 0; - if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type)) + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type, data_len)) goto out; if (t->parms.iph.daddr == 0) -- cgit v1.2.3 From e02fb7264d8a31dddb9a80fbde603feb502d6478 Mon Sep 17 00:00:00 2001 From: stuart hayes Date: Thu, 26 May 2016 11:38:41 -0500 Subject: nfit: add Microsoft NVDIMM DSM command set to white list Add the Microsoft _DSM command set to the white list of NVDIMM command sets. This command set is documented at: https://msdn.microsoft.com/library/windows/hardware/mt604741 Cc: Pavel Machek [pavel: fix up braces] Signed-off-by: Stuart Hayes Signed-off-by: Dan Williams --- drivers/acpi/nfit.c | 11 +++++++---- drivers/acpi/nfit.h | 4 ++++ include/uapi/linux/ndctl.h | 1 + 3 files changed, 12 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/drivers/acpi/nfit.c b/drivers/acpi/nfit.c index 2215fc847fa9..da14c89f4667 100644 --- a/drivers/acpi/nfit.c +++ b/drivers/acpi/nfit.c @@ -1130,11 +1130,11 @@ static int acpi_nfit_add_dimm(struct acpi_nfit_desc *acpi_desc, } /* - * Until standardization materializes we need to consider up to 3 + * Until standardization materializes we need to consider 4 * different command sets. Note, that checking for function0 (bit0) * tells us if any commands are reachable through this uuid. */ - for (i = NVDIMM_FAMILY_INTEL; i <= NVDIMM_FAMILY_HPE2; i++) + for (i = NVDIMM_FAMILY_INTEL; i <= NVDIMM_FAMILY_MSFT; i++) if (acpi_check_dsm(adev_dimm->handle, to_nfit_uuid(i), 1, 1)) break; @@ -1144,12 +1144,14 @@ static int acpi_nfit_add_dimm(struct acpi_nfit_desc *acpi_desc, dsm_mask = 0x3fe; if (disable_vendor_specific) dsm_mask &= ~(1 << ND_CMD_VENDOR); - } else if (nfit_mem->family == NVDIMM_FAMILY_HPE1) + } else if (nfit_mem->family == NVDIMM_FAMILY_HPE1) { dsm_mask = 0x1c3c76; - else if (nfit_mem->family == NVDIMM_FAMILY_HPE2) { + } else if (nfit_mem->family == NVDIMM_FAMILY_HPE2) { dsm_mask = 0x1fe; if (disable_vendor_specific) dsm_mask &= ~(1 << 8); + } else if (nfit_mem->family == NVDIMM_FAMILY_MSFT) { + dsm_mask = 0xffffffff; } else { dev_err(dev, "unknown dimm command family\n"); nfit_mem->family = -1; @@ -2692,6 +2694,7 @@ static __init int nfit_init(void) acpi_str_to_uuid(UUID_NFIT_DIMM, nfit_uuid[NFIT_DEV_DIMM]); acpi_str_to_uuid(UUID_NFIT_DIMM_N_HPE1, nfit_uuid[NFIT_DEV_DIMM_N_HPE1]); acpi_str_to_uuid(UUID_NFIT_DIMM_N_HPE2, nfit_uuid[NFIT_DEV_DIMM_N_HPE2]); + acpi_str_to_uuid(UUID_NFIT_DIMM_N_MSFT, nfit_uuid[NFIT_DEV_DIMM_N_MSFT]); nfit_wq = create_singlethread_workqueue("nfit"); if (!nfit_wq) diff --git a/drivers/acpi/nfit.h b/drivers/acpi/nfit.h index 11cb38348aef..f06fa91c5abf 100644 --- a/drivers/acpi/nfit.h +++ b/drivers/acpi/nfit.h @@ -31,6 +31,9 @@ #define UUID_NFIT_DIMM_N_HPE1 "9002c334-acf3-4c0e-9642-a235f0d53bc6" #define UUID_NFIT_DIMM_N_HPE2 "5008664b-b758-41a0-a03c-27c2f2d04f7e" +/* https://msdn.microsoft.com/library/windows/hardware/mt604741 */ +#define UUID_NFIT_DIMM_N_MSFT "1ee68b36-d4bd-4a1a-9a16-4f8e53d46e05" + #define ACPI_NFIT_MEM_FAILED_MASK (ACPI_NFIT_MEM_SAVE_FAILED \ | ACPI_NFIT_MEM_RESTORE_FAILED | ACPI_NFIT_MEM_FLUSH_FAILED \ | ACPI_NFIT_MEM_NOT_ARMED) @@ -40,6 +43,7 @@ enum nfit_uuids { NFIT_DEV_DIMM = NVDIMM_FAMILY_INTEL, NFIT_DEV_DIMM_N_HPE1 = NVDIMM_FAMILY_HPE1, NFIT_DEV_DIMM_N_HPE2 = NVDIMM_FAMILY_HPE2, + NFIT_DEV_DIMM_N_MSFT = NVDIMM_FAMILY_MSFT, NFIT_SPA_VOLATILE, NFIT_SPA_PM, NFIT_SPA_DCR, diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h index 309915f74492..ba5a8c79652a 100644 --- a/include/uapi/linux/ndctl.h +++ b/include/uapi/linux/ndctl.h @@ -298,6 +298,7 @@ struct nd_cmd_pkg { #define NVDIMM_FAMILY_INTEL 0 #define NVDIMM_FAMILY_HPE1 1 #define NVDIMM_FAMILY_HPE2 2 +#define NVDIMM_FAMILY_MSFT 3 #define ND_IOCTL_CALL _IOWR(ND_IOCTL, ND_CMD_CALL,\ struct nd_cmd_pkg) -- cgit v1.2.3 From b95e5928fcc76d156352570858abdea7b2628efd Mon Sep 17 00:00:00 2001 From: William Tu Date: Mon, 20 Jun 2016 07:26:17 -0700 Subject: openvswitch: Add packet len info to upcall. The commit f2a4d086ed4c ("openvswitch: Add packet truncation support.") introduces packet truncation before sending to userspace upcall receiver. This patch passes up the skb->len before truncation so that the upcall receiver knows the original packet size. Potentially this will be used by sFlow, where OVS translates sFlow config header=N to a sample action, truncating packet to N byte in kernel datapath. Thus, only N bytes instead of full-packet size is copied from kernel to userspace, saving the kernel-to-userspace bandwidth. Signed-off-by: William Tu Cc: Pravin Shelar Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- include/uapi/linux/openvswitch.h | 2 ++ net/openvswitch/datapath.c | 13 ++++++++++++- 2 files changed, 14 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index 8274675ba9a3..d95a3018f6a1 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -166,6 +166,7 @@ enum ovs_packet_cmd { * output port is actually a tunnel port. Contains the output tunnel key * extracted from the packet as nested %OVS_TUNNEL_KEY_ATTR_* attributes. * @OVS_PACKET_ATTR_MRU: Present for an %OVS_PACKET_CMD_ACTION and + * @OVS_PACKET_ATTR_LEN: Packet size before truncation. * %OVS_PACKET_ATTR_USERSPACE action specify the Maximum received fragment * size. * @@ -185,6 +186,7 @@ enum ovs_packet_attr { OVS_PACKET_ATTR_PROBE, /* Packet operation is a feature probe, error logging should be suppressed. */ OVS_PACKET_ATTR_MRU, /* Maximum received IP fragment size. */ + OVS_PACKET_ATTR_LEN, /* Packet size before truncation. */ __OVS_PACKET_ATTR_MAX }; diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 673934295333..524c0fd3078e 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -387,7 +387,8 @@ static size_t upcall_msg_size(const struct dp_upcall_info *upcall_info, { size_t size = NLMSG_ALIGN(sizeof(struct ovs_header)) + nla_total_size(hdrlen) /* OVS_PACKET_ATTR_PACKET */ - + nla_total_size(ovs_key_attr_size()); /* OVS_PACKET_ATTR_KEY */ + + nla_total_size(ovs_key_attr_size()) /* OVS_PACKET_ATTR_KEY */ + + nla_total_size(sizeof(unsigned int)); /* OVS_PACKET_ATTR_LEN */ /* OVS_PACKET_ATTR_USERDATA */ if (upcall_info->userdata) @@ -514,6 +515,16 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, pad_packet(dp, user_skb); } + /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ + if (cutlen > 0) { + if (nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, + skb->len)) { + err = -ENOBUFS; + goto out; + } + pad_packet(dp, user_skb); + } + /* Only reserve room for attribute header, packet data is added * in skb_zerocopy() */ if (!(nla = nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, 0))) { -- cgit v1.2.3 From 4e5f2c400765e3a3ce512dc1ae890bac53401798 Mon Sep 17 00:00:00 2001 From: Salvatore Benedetto Date: Wed, 22 Jun 2016 17:49:13 +0100 Subject: crypto: kpp - Key-agreement Protocol Primitives API (KPP) Add key-agreement protocol primitives (kpp) API which allows to implement primitives required by protocols such as DH and ECDH. The API is composed mainly by the following functions * set_secret() - It allows the user to set his secret, also referred to as his private key, along with the parameters known to both parties involved in the key-agreement session. * generate_public_key() - It generates the public key to be sent to the other counterpart involved in the key-agreement session. The function has to be called after set_params() and set_secret() * generate_secret() - It generates the shared secret for the session Other functions such as init() and exit() are provided for allowing cryptographic hardware to be inizialized properly before use Signed-off-by: Salvatore Benedetto Signed-off-by: Herbert Xu --- crypto/Kconfig | 10 ++ crypto/Makefile | 1 + crypto/crypto_user.c | 20 +++ crypto/kpp.c | 123 +++++++++++++++ include/crypto/internal/kpp.h | 64 ++++++++ include/crypto/kpp.h | 328 ++++++++++++++++++++++++++++++++++++++++ include/linux/crypto.h | 1 + include/uapi/linux/cryptouser.h | 5 + 8 files changed, 552 insertions(+) create mode 100644 crypto/kpp.c create mode 100644 include/crypto/internal/kpp.h create mode 100644 include/crypto/kpp.h (limited to 'include/uapi') diff --git a/crypto/Kconfig b/crypto/Kconfig index 6881d1a5f859..e72c4270173d 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -93,6 +93,15 @@ config CRYPTO_AKCIPHER select CRYPTO_AKCIPHER2 select CRYPTO_ALGAPI +config CRYPTO_KPP2 + tristate + select CRYPTO_ALGAPI2 + +config CRYPTO_KPP + tristate + select CRYPTO_ALGAPI + select CRYPTO_KPP2 + config CRYPTO_RSA tristate "RSA algorithm" select CRYPTO_AKCIPHER @@ -115,6 +124,7 @@ config CRYPTO_MANAGER2 select CRYPTO_HASH2 select CRYPTO_BLKCIPHER2 select CRYPTO_AKCIPHER2 + select CRYPTO_KPP2 config CRYPTO_USER tristate "Userspace cryptographic algorithm configuration" diff --git a/crypto/Makefile b/crypto/Makefile index 0b82c4753743..07b0f51bd645 100644 --- a/crypto/Makefile +++ b/crypto/Makefile @@ -30,6 +30,7 @@ crypto_hash-y += shash.o obj-$(CONFIG_CRYPTO_HASH2) += crypto_hash.o obj-$(CONFIG_CRYPTO_AKCIPHER2) += akcipher.o +obj-$(CONFIG_CRYPTO_KPP2) += kpp.o $(obj)/rsapubkey-asn1.o: $(obj)/rsapubkey-asn1.c $(obj)/rsapubkey-asn1.h $(obj)/rsaprivkey-asn1.o: $(obj)/rsaprivkey-asn1.c $(obj)/rsaprivkey-asn1.h diff --git a/crypto/crypto_user.c b/crypto/crypto_user.c index 43fe85f20d57..d28513fb5a90 100644 --- a/crypto/crypto_user.c +++ b/crypto/crypto_user.c @@ -28,6 +28,7 @@ #include #include #include +#include #include "internal.h" @@ -126,6 +127,21 @@ nla_put_failure: return -EMSGSIZE; } +static int crypto_report_kpp(struct sk_buff *skb, struct crypto_alg *alg) +{ + struct crypto_report_kpp rkpp; + + strncpy(rkpp.type, "kpp", sizeof(rkpp.type)); + + if (nla_put(skb, CRYPTOCFGA_REPORT_KPP, + sizeof(struct crypto_report_kpp), &rkpp)) + goto nla_put_failure; + return 0; + +nla_put_failure: + return -EMSGSIZE; +} + static int crypto_report_one(struct crypto_alg *alg, struct crypto_user_alg *ualg, struct sk_buff *skb) { @@ -176,6 +192,10 @@ static int crypto_report_one(struct crypto_alg *alg, goto nla_put_failure; break; + case CRYPTO_ALG_TYPE_KPP: + if (crypto_report_kpp(skb, alg)) + goto nla_put_failure; + break; } out: diff --git a/crypto/kpp.c b/crypto/kpp.c new file mode 100644 index 000000000000..d36ce05eee43 --- /dev/null +++ b/crypto/kpp.c @@ -0,0 +1,123 @@ +/* + * Key-agreement Protocol Primitives (KPP) + * + * Copyright (c) 2016, Intel Corporation + * Authors: Salvatore Benedetto + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +#ifdef CONFIG_NET +static int crypto_kpp_report(struct sk_buff *skb, struct crypto_alg *alg) +{ + struct crypto_report_kpp rkpp; + + strncpy(rkpp.type, "kpp", sizeof(rkpp.type)); + + if (nla_put(skb, CRYPTOCFGA_REPORT_KPP, + sizeof(struct crypto_report_kpp), &rkpp)) + goto nla_put_failure; + return 0; + +nla_put_failure: + return -EMSGSIZE; +} +#else +static int crypto_kpp_report(struct sk_buff *skb, struct crypto_alg *alg) +{ + return -ENOSYS; +} +#endif + +static void crypto_kpp_show(struct seq_file *m, struct crypto_alg *alg) + __attribute__ ((unused)); + +static void crypto_kpp_show(struct seq_file *m, struct crypto_alg *alg) +{ + seq_puts(m, "type : kpp\n"); +} + +static void crypto_kpp_exit_tfm(struct crypto_tfm *tfm) +{ + struct crypto_kpp *kpp = __crypto_kpp_tfm(tfm); + struct kpp_alg *alg = crypto_kpp_alg(kpp); + + alg->exit(kpp); +} + +static int crypto_kpp_init_tfm(struct crypto_tfm *tfm) +{ + struct crypto_kpp *kpp = __crypto_kpp_tfm(tfm); + struct kpp_alg *alg = crypto_kpp_alg(kpp); + + if (alg->exit) + kpp->base.exit = crypto_kpp_exit_tfm; + + if (alg->init) + return alg->init(kpp); + + return 0; +} + +static const struct crypto_type crypto_kpp_type = { + .extsize = crypto_alg_extsize, + .init_tfm = crypto_kpp_init_tfm, +#ifdef CONFIG_PROC_FS + .show = crypto_kpp_show, +#endif + .report = crypto_kpp_report, + .maskclear = ~CRYPTO_ALG_TYPE_MASK, + .maskset = CRYPTO_ALG_TYPE_MASK, + .type = CRYPTO_ALG_TYPE_KPP, + .tfmsize = offsetof(struct crypto_kpp, base), +}; + +struct crypto_kpp *crypto_alloc_kpp(const char *alg_name, u32 type, u32 mask) +{ + return crypto_alloc_tfm(alg_name, &crypto_kpp_type, type, mask); +} +EXPORT_SYMBOL_GPL(crypto_alloc_kpp); + +static void kpp_prepare_alg(struct kpp_alg *alg) +{ + struct crypto_alg *base = &alg->base; + + base->cra_type = &crypto_kpp_type; + base->cra_flags &= ~CRYPTO_ALG_TYPE_MASK; + base->cra_flags |= CRYPTO_ALG_TYPE_KPP; +} + +int crypto_register_kpp(struct kpp_alg *alg) +{ + struct crypto_alg *base = &alg->base; + + kpp_prepare_alg(alg); + return crypto_register_alg(base); +} +EXPORT_SYMBOL_GPL(crypto_register_kpp); + +void crypto_unregister_kpp(struct kpp_alg *alg) +{ + crypto_unregister_alg(&alg->base); +} +EXPORT_SYMBOL_GPL(crypto_unregister_kpp); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Key-agreement Protocol Primitives"); diff --git a/include/crypto/internal/kpp.h b/include/crypto/internal/kpp.h new file mode 100644 index 000000000000..ad3acf3649be --- /dev/null +++ b/include/crypto/internal/kpp.h @@ -0,0 +1,64 @@ +/* + * Key-agreement Protocol Primitives (KPP) + * + * Copyright (c) 2016, Intel Corporation + * Authors: Salvatore Benedetto + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ +#ifndef _CRYPTO_KPP_INT_H +#define _CRYPTO_KPP_INT_H +#include +#include + +/* + * Transform internal helpers. + */ +static inline void *kpp_request_ctx(struct kpp_request *req) +{ + return req->__ctx; +} + +static inline void *kpp_tfm_ctx(struct crypto_kpp *tfm) +{ + return tfm->base.__crt_ctx; +} + +static inline void kpp_request_complete(struct kpp_request *req, int err) +{ + req->base.complete(&req->base, err); +} + +static inline const char *kpp_alg_name(struct crypto_kpp *tfm) +{ + return crypto_kpp_tfm(tfm)->__crt_alg->cra_name; +} + +/** + * crypto_register_kpp() -- Register key-agreement protocol primitives algorithm + * + * Function registers an implementation of a key-agreement protocol primitive + * algorithm + * + * @alg: algorithm definition + * + * Return: zero on success; error code in case of error + */ +int crypto_register_kpp(struct kpp_alg *alg); + +/** + * crypto_unregister_kpp() -- Unregister key-agreement protocol primitive + * algorithm + * + * Function unregisters an implementation of a key-agreement protocol primitive + * algorithm + * + * @alg: algorithm definition + */ +void crypto_unregister_kpp(struct kpp_alg *alg); + +#endif diff --git a/include/crypto/kpp.h b/include/crypto/kpp.h new file mode 100644 index 000000000000..4fa897f3366b --- /dev/null +++ b/include/crypto/kpp.h @@ -0,0 +1,328 @@ +/* + * Key-agreement Protocol Primitives (KPP) + * + * Copyright (c) 2016, Intel Corporation + * Authors: Salvatore Benedetto + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + */ + +#ifndef _CRYPTO_KPP_ +#define _CRYPTO_KPP_ +#include + +/** + * struct kpp_request + * + * @base: Common attributes for async crypto requests + * @src: Source data + * @dst: Destination data + * @src_len: Size of the input buffer + * @dst_len: Size of the output buffer. It needs to be at least + * as big as the expected result depending on the operation + * After operation it will be updated with the actual size of the + * result. In case of error where the dst sgl size was insufficient, + * it will be updated to the size required for the operation. + * @__ctx: Start of private context data + */ +struct kpp_request { + struct crypto_async_request base; + struct scatterlist *src; + struct scatterlist *dst; + unsigned int src_len; + unsigned int dst_len; + void *__ctx[] CRYPTO_MINALIGN_ATTR; +}; + +/** + * struct crypto_kpp - user-instantiated object which encapsulate + * algorithms and core processing logic + * + * @base: Common crypto API algorithm data structure + */ +struct crypto_kpp { + struct crypto_tfm base; +}; + +/** + * struct kpp_alg - generic key-agreement protocol primitives + * + * @set_secret: Function invokes the protocol specific function to + * store the secret private key along with parameters. + * The implementation knows how to decode thie buffer + * @generate_public_key: Function generate the public key to be sent to the + * counterpart. In case of error, where output is not big + * enough req->dst_len will be updated to the size + * required + * @compute_shared_secret: Function compute the shared secret as defined by + * the algorithm. The result is given back to the user. + * In case of error, where output is not big enough, + * req->dst_len will be updated to the size required + * @max_size: Function returns the size of the output buffer + * @init: Initialize the object. This is called only once at + * instantiation time. In case the cryptographic hardware + * needs to be initialized. Software fallback should be + * put in place here. + * @exit: Undo everything @init did. + * + * @reqsize: Request context size required by algorithm + * implementation + * @base Common crypto API algorithm data structure + */ +struct kpp_alg { + int (*set_secret)(struct crypto_kpp *tfm, void *buffer, + unsigned int len); + int (*generate_public_key)(struct kpp_request *req); + int (*compute_shared_secret)(struct kpp_request *req); + + int (*max_size)(struct crypto_kpp *tfm); + + int (*init)(struct crypto_kpp *tfm); + void (*exit)(struct crypto_kpp *tfm); + + unsigned int reqsize; + struct crypto_alg base; +}; + +/** + * DOC: Generic Key-agreement Protocol Primitevs API + * + * The KPP API is used with the algorithm type + * CRYPTO_ALG_TYPE_KPP (listed as type "kpp" in /proc/crypto) + */ + +/** + * crypto_alloc_kpp() - allocate KPP tfm handle + * @alg_name: is the name of the kpp algorithm (e.g. "dh", "ecdh") + * @type: specifies the type of the algorithm + * @mask: specifies the mask for the algorithm + * + * Allocate a handle for kpp algorithm. The returned struct crypto_kpp + * is requeried for any following API invocation + * + * Return: allocated handle in case of success; IS_ERR() is true in case of + * an error, PTR_ERR() returns the error code. + */ +struct crypto_kpp *crypto_alloc_kpp(const char *alg_name, u32 type, u32 mask); + +static inline struct crypto_tfm *crypto_kpp_tfm(struct crypto_kpp *tfm) +{ + return &tfm->base; +} + +static inline struct kpp_alg *__crypto_kpp_alg(struct crypto_alg *alg) +{ + return container_of(alg, struct kpp_alg, base); +} + +static inline struct crypto_kpp *__crypto_kpp_tfm(struct crypto_tfm *tfm) +{ + return container_of(tfm, struct crypto_kpp, base); +} + +static inline struct kpp_alg *crypto_kpp_alg(struct crypto_kpp *tfm) +{ + return __crypto_kpp_alg(crypto_kpp_tfm(tfm)->__crt_alg); +} + +static inline unsigned int crypto_kpp_reqsize(struct crypto_kpp *tfm) +{ + return crypto_kpp_alg(tfm)->reqsize; +} + +static inline void kpp_request_set_tfm(struct kpp_request *req, + struct crypto_kpp *tfm) +{ + req->base.tfm = crypto_kpp_tfm(tfm); +} + +static inline struct crypto_kpp *crypto_kpp_reqtfm(struct kpp_request *req) +{ + return __crypto_kpp_tfm(req->base.tfm); +} + +/** + * crypto_free_kpp() - free KPP tfm handle + * + * @tfm: KPP tfm handle allocated with crypto_alloc_kpp() + */ +static inline void crypto_free_kpp(struct crypto_kpp *tfm) +{ + crypto_destroy_tfm(tfm, crypto_kpp_tfm(tfm)); +} + +/** + * kpp_request_alloc() - allocates kpp request + * + * @tfm: KPP tfm handle allocated with crypto_alloc_kpp() + * @gfp: allocation flags + * + * Return: allocated handle in case of success or NULL in case of an error. + */ +static inline struct kpp_request *kpp_request_alloc(struct crypto_kpp *tfm, + gfp_t gfp) +{ + struct kpp_request *req; + + req = kmalloc(sizeof(*req) + crypto_kpp_reqsize(tfm), gfp); + if (likely(req)) + kpp_request_set_tfm(req, tfm); + + return req; +} + +/** + * kpp_request_free() - zeroize and free kpp request + * + * @req: request to free + */ +static inline void kpp_request_free(struct kpp_request *req) +{ + kzfree(req); +} + +/** + * kpp_request_set_callback() - Sets an asynchronous callback. + * + * Callback will be called when an asynchronous operation on a given + * request is finished. + * + * @req: request that the callback will be set for + * @flgs: specify for instance if the operation may backlog + * @cmpl: callback which will be called + * @data: private data used by the caller + */ +static inline void kpp_request_set_callback(struct kpp_request *req, + u32 flgs, + crypto_completion_t cmpl, + void *data) +{ + req->base.complete = cmpl; + req->base.data = data; + req->base.flags = flgs; +} + +/** + * kpp_request_set_input() - Sets input buffer + * + * Sets parameters required by generate_public_key + * + * @req: kpp request + * @input: ptr to input scatter list + * @input_len: size of the input scatter list + */ +static inline void kpp_request_set_input(struct kpp_request *req, + struct scatterlist *input, + unsigned int input_len) +{ + req->src = input; + req->src_len = input_len; +} + +/** + * kpp_request_set_output() - Sets output buffer + * + * Sets parameters required by kpp operation + * + * @req: kpp request + * @output: ptr to output scatter list + * @output_len: size of the output scatter list + */ +static inline void kpp_request_set_output(struct kpp_request *req, + struct scatterlist *output, + unsigned int output_len) +{ + req->dst = output; + req->dst_len = output_len; +} + +enum { + CRYPTO_KPP_SECRET_TYPE_UNKNOWN, +}; + +/** + * struct kpp_secret - small header for packing secret buffer + * + * @type: define type of secret. Each kpp type will define its own + * @len: specify the len of the secret, include the header, that + * follows the struct + */ +struct kpp_secret { + unsigned short type; + unsigned short len; +}; + +/** + * crypto_kpp_set_secret() - Invoke kpp operation + * + * Function invokes the specific kpp operation for a given alg. + * + * @tfm: tfm handle + * + * Return: zero on success; error code in case of error + */ +static inline int crypto_kpp_set_secret(struct crypto_kpp *tfm, void *buffer, + unsigned int len) +{ + struct kpp_alg *alg = crypto_kpp_alg(tfm); + + return alg->set_secret(tfm, buffer, len); +} + +/** + * crypto_kpp_generate_public_key() - Invoke kpp operation + * + * Function invokes the specific kpp operation for generating the public part + * for a given kpp algorithm + * + * @req: kpp key request + * + * Return: zero on success; error code in case of error + */ +static inline int crypto_kpp_generate_public_key(struct kpp_request *req) +{ + struct crypto_kpp *tfm = crypto_kpp_reqtfm(req); + struct kpp_alg *alg = crypto_kpp_alg(tfm); + + return alg->generate_public_key(req); +} + +/** + * crypto_kpp_compute_shared_secret() - Invoke kpp operation + * + * Function invokes the specific kpp operation for computing the shared secret + * for a given kpp algorithm. + * + * @req: kpp key request + * + * Return: zero on success; error code in case of error + */ +static inline int crypto_kpp_compute_shared_secret(struct kpp_request *req) +{ + struct crypto_kpp *tfm = crypto_kpp_reqtfm(req); + struct kpp_alg *alg = crypto_kpp_alg(tfm); + + return alg->compute_shared_secret(req); +} + +/** + * crypto_kpp_maxsize() - Get len for output buffer + * + * Function returns the output buffer size required + * + * @tfm: KPP tfm handle allocated with crypto_alloc_kpp() + * + * Return: minimum len for output buffer or error code if key hasn't been set + */ +static inline int crypto_kpp_maxsize(struct crypto_kpp *tfm) +{ + struct kpp_alg *alg = crypto_kpp_alg(tfm); + + return alg->max_size(tfm); +} + +#endif diff --git a/include/linux/crypto.h b/include/linux/crypto.h index d844cbc365f7..992cfc2e5df1 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -48,6 +48,7 @@ #define CRYPTO_ALG_TYPE_BLKCIPHER 0x00000004 #define CRYPTO_ALG_TYPE_ABLKCIPHER 0x00000005 #define CRYPTO_ALG_TYPE_GIVCIPHER 0x00000006 +#define CRYPTO_ALG_TYPE_KPP 0x00000008 #define CRYPTO_ALG_TYPE_RNG 0x0000000c #define CRYPTO_ALG_TYPE_AKCIPHER 0x0000000d #define CRYPTO_ALG_TYPE_DIGEST 0x0000000e diff --git a/include/uapi/linux/cryptouser.h b/include/uapi/linux/cryptouser.h index 2e67bb64c1da..79b5ded2001a 100644 --- a/include/uapi/linux/cryptouser.h +++ b/include/uapi/linux/cryptouser.h @@ -45,6 +45,7 @@ enum crypto_attr_type_t { CRYPTOCFGA_REPORT_RNG, /* struct crypto_report_rng */ CRYPTOCFGA_REPORT_CIPHER, /* struct crypto_report_cipher */ CRYPTOCFGA_REPORT_AKCIPHER, /* struct crypto_report_akcipher */ + CRYPTOCFGA_REPORT_KPP, /* struct crypto_report_kpp */ __CRYPTOCFGA_MAX #define CRYPTOCFGA_MAX (__CRYPTOCFGA_MAX - 1) @@ -107,5 +108,9 @@ struct crypto_report_akcipher { char type[CRYPTO_MAX_NAME]; }; +struct crypto_report_kpp { + char type[CRYPTO_MAX_NAME]; +}; + #define CRYPTO_REPORT_MAXSIZE (sizeof(struct crypto_user_alg) + \ sizeof(struct crypto_report_blkcipher)) -- cgit v1.2.3 From f213c05272100f385912372fff678d0af4d7f8ad Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 23 May 2016 15:20:49 +0300 Subject: IB/uverbs: Add WQ support User space applications which use RSS functionality need to create a work queue object (WQ). The lifetime of such an object is: * Create a WQ * Modify the WQ from reset to init state. * Use the WQ (by downstream patches). * Destroy the WQ. These commands are added to the uverbs API. Signed-off-by: Yishai Hadas Signed-off-by: Matan Barak Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/core/uverbs.h | 9 ++ drivers/infiniband/core/uverbs_cmd.c | 243 ++++++++++++++++++++++++++++++++++ drivers/infiniband/core/uverbs_main.c | 25 ++++ include/rdma/ib_verbs.h | 3 + include/uapi/rdma/ib_user_verbs.h | 41 ++++++ 5 files changed, 321 insertions(+) (limited to 'include/uapi') diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index 612ccfd39bf9..74776c6531f4 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -162,6 +162,10 @@ struct ib_uqp_object { struct ib_uxrcd_object *uxrcd; }; +struct ib_uwq_object { + struct ib_uevent_object uevent; +}; + struct ib_ucq_object { struct ib_uobject uobject; struct ib_uverbs_file *uverbs_file; @@ -181,6 +185,7 @@ extern struct idr ib_uverbs_qp_idr; extern struct idr ib_uverbs_srq_idr; extern struct idr ib_uverbs_xrcd_idr; extern struct idr ib_uverbs_rule_idr; +extern struct idr ib_uverbs_wq_idr; void idr_remove_uobj(struct idr *idp, struct ib_uobject *uobj); @@ -199,6 +204,7 @@ void ib_uverbs_release_uevent(struct ib_uverbs_file *file, void ib_uverbs_comp_handler(struct ib_cq *cq, void *cq_context); void ib_uverbs_cq_event_handler(struct ib_event *event, void *context_ptr); void ib_uverbs_qp_event_handler(struct ib_event *event, void *context_ptr); +void ib_uverbs_wq_event_handler(struct ib_event *event, void *context_ptr); void ib_uverbs_srq_event_handler(struct ib_event *event, void *context_ptr); void ib_uverbs_event_handler(struct ib_event_handler *handler, struct ib_event *event); @@ -275,5 +281,8 @@ IB_UVERBS_DECLARE_EX_CMD(destroy_flow); IB_UVERBS_DECLARE_EX_CMD(query_device); IB_UVERBS_DECLARE_EX_CMD(create_cq); IB_UVERBS_DECLARE_EX_CMD(create_qp); +IB_UVERBS_DECLARE_EX_CMD(create_wq); +IB_UVERBS_DECLARE_EX_CMD(modify_wq); +IB_UVERBS_DECLARE_EX_CMD(destroy_wq); #endif /* UVERBS_H */ diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 1a8babb8ee3c..22e617391657 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -57,6 +57,7 @@ static struct uverbs_lock_class ah_lock_class = { .name = "AH-uobj" }; static struct uverbs_lock_class srq_lock_class = { .name = "SRQ-uobj" }; static struct uverbs_lock_class xrcd_lock_class = { .name = "XRCD-uobj" }; static struct uverbs_lock_class rule_lock_class = { .name = "RULE-uobj" }; +static struct uverbs_lock_class wq_lock_class = { .name = "WQ-uobj" }; /* * The ib_uobject locking scheme is as follows: @@ -243,6 +244,16 @@ static struct ib_qp *idr_read_qp(int qp_handle, struct ib_ucontext *context) return idr_read_obj(&ib_uverbs_qp_idr, qp_handle, context, 0); } +static struct ib_wq *idr_read_wq(int wq_handle, struct ib_ucontext *context) +{ + return idr_read_obj(&ib_uverbs_wq_idr, wq_handle, context, 0); +} + +static void put_wq_read(struct ib_wq *wq) +{ + put_uobj_read(wq->uobject); +} + static struct ib_qp *idr_write_qp(int qp_handle, struct ib_ucontext *context) { struct ib_uobject *uobj; @@ -326,6 +337,7 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, INIT_LIST_HEAD(&ucontext->qp_list); INIT_LIST_HEAD(&ucontext->srq_list); INIT_LIST_HEAD(&ucontext->ah_list); + INIT_LIST_HEAD(&ucontext->wq_list); INIT_LIST_HEAD(&ucontext->xrcd_list); INIT_LIST_HEAD(&ucontext->rule_list); rcu_read_lock(); @@ -3056,6 +3068,237 @@ static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec, return 0; } +int ib_uverbs_ex_create_wq(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + struct ib_udata *ucore, + struct ib_udata *uhw) +{ + struct ib_uverbs_ex_create_wq cmd = {}; + struct ib_uverbs_ex_create_wq_resp resp = {}; + struct ib_uwq_object *obj; + int err = 0; + struct ib_cq *cq; + struct ib_pd *pd; + struct ib_wq *wq; + struct ib_wq_init_attr wq_init_attr = {}; + size_t required_cmd_sz; + size_t required_resp_len; + + required_cmd_sz = offsetof(typeof(cmd), max_sge) + sizeof(cmd.max_sge); + required_resp_len = offsetof(typeof(resp), wqn) + sizeof(resp.wqn); + + if (ucore->inlen < required_cmd_sz) + return -EINVAL; + + if (ucore->outlen < required_resp_len) + return -ENOSPC; + + if (ucore->inlen > sizeof(cmd) && + !ib_is_udata_cleared(ucore, sizeof(cmd), + ucore->inlen - sizeof(cmd))) + return -EOPNOTSUPP; + + err = ib_copy_from_udata(&cmd, ucore, min(sizeof(cmd), ucore->inlen)); + if (err) + return err; + + if (cmd.comp_mask) + return -EOPNOTSUPP; + + obj = kmalloc(sizeof(*obj), GFP_KERNEL); + if (!obj) + return -ENOMEM; + + init_uobj(&obj->uevent.uobject, cmd.user_handle, file->ucontext, + &wq_lock_class); + down_write(&obj->uevent.uobject.mutex); + pd = idr_read_pd(cmd.pd_handle, file->ucontext); + if (!pd) { + err = -EINVAL; + goto err_uobj; + } + + cq = idr_read_cq(cmd.cq_handle, file->ucontext, 0); + if (!cq) { + err = -EINVAL; + goto err_put_pd; + } + + wq_init_attr.cq = cq; + wq_init_attr.max_sge = cmd.max_sge; + wq_init_attr.max_wr = cmd.max_wr; + wq_init_attr.wq_context = file; + wq_init_attr.wq_type = cmd.wq_type; + wq_init_attr.event_handler = ib_uverbs_wq_event_handler; + obj->uevent.events_reported = 0; + INIT_LIST_HEAD(&obj->uevent.event_list); + wq = pd->device->create_wq(pd, &wq_init_attr, uhw); + if (IS_ERR(wq)) { + err = PTR_ERR(wq); + goto err_put_cq; + } + + wq->uobject = &obj->uevent.uobject; + obj->uevent.uobject.object = wq; + wq->wq_type = wq_init_attr.wq_type; + wq->cq = cq; + wq->pd = pd; + wq->device = pd->device; + wq->wq_context = wq_init_attr.wq_context; + atomic_set(&wq->usecnt, 0); + atomic_inc(&pd->usecnt); + atomic_inc(&cq->usecnt); + wq->uobject = &obj->uevent.uobject; + obj->uevent.uobject.object = wq; + err = idr_add_uobj(&ib_uverbs_wq_idr, &obj->uevent.uobject); + if (err) + goto destroy_wq; + + memset(&resp, 0, sizeof(resp)); + resp.wq_handle = obj->uevent.uobject.id; + resp.max_sge = wq_init_attr.max_sge; + resp.max_wr = wq_init_attr.max_wr; + resp.wqn = wq->wq_num; + resp.response_length = required_resp_len; + err = ib_copy_to_udata(ucore, + &resp, resp.response_length); + if (err) + goto err_copy; + + put_pd_read(pd); + put_cq_read(cq); + + mutex_lock(&file->mutex); + list_add_tail(&obj->uevent.uobject.list, &file->ucontext->wq_list); + mutex_unlock(&file->mutex); + + obj->uevent.uobject.live = 1; + up_write(&obj->uevent.uobject.mutex); + return 0; + +err_copy: + idr_remove_uobj(&ib_uverbs_wq_idr, &obj->uevent.uobject); +destroy_wq: + ib_destroy_wq(wq); +err_put_cq: + put_cq_read(cq); +err_put_pd: + put_pd_read(pd); +err_uobj: + put_uobj_write(&obj->uevent.uobject); + + return err; +} + +int ib_uverbs_ex_destroy_wq(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + struct ib_udata *ucore, + struct ib_udata *uhw) +{ + struct ib_uverbs_ex_destroy_wq cmd = {}; + struct ib_uverbs_ex_destroy_wq_resp resp = {}; + struct ib_wq *wq; + struct ib_uobject *uobj; + struct ib_uwq_object *obj; + size_t required_cmd_sz; + size_t required_resp_len; + int ret; + + required_cmd_sz = offsetof(typeof(cmd), wq_handle) + sizeof(cmd.wq_handle); + required_resp_len = offsetof(typeof(resp), reserved) + sizeof(resp.reserved); + + if (ucore->inlen < required_cmd_sz) + return -EINVAL; + + if (ucore->outlen < required_resp_len) + return -ENOSPC; + + if (ucore->inlen > sizeof(cmd) && + !ib_is_udata_cleared(ucore, sizeof(cmd), + ucore->inlen - sizeof(cmd))) + return -EOPNOTSUPP; + + ret = ib_copy_from_udata(&cmd, ucore, min(sizeof(cmd), ucore->inlen)); + if (ret) + return ret; + + if (cmd.comp_mask) + return -EOPNOTSUPP; + + resp.response_length = required_resp_len; + uobj = idr_write_uobj(&ib_uverbs_wq_idr, cmd.wq_handle, + file->ucontext); + if (!uobj) + return -EINVAL; + + wq = uobj->object; + obj = container_of(uobj, struct ib_uwq_object, uevent.uobject); + ret = ib_destroy_wq(wq); + if (!ret) + uobj->live = 0; + + put_uobj_write(uobj); + if (ret) + return ret; + + idr_remove_uobj(&ib_uverbs_wq_idr, uobj); + + mutex_lock(&file->mutex); + list_del(&uobj->list); + mutex_unlock(&file->mutex); + + ib_uverbs_release_uevent(file, &obj->uevent); + resp.events_reported = obj->uevent.events_reported; + put_uobj(uobj); + + ret = ib_copy_to_udata(ucore, &resp, resp.response_length); + if (ret) + return ret; + + return 0; +} + +int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + struct ib_udata *ucore, + struct ib_udata *uhw) +{ + struct ib_uverbs_ex_modify_wq cmd = {}; + struct ib_wq *wq; + struct ib_wq_attr wq_attr = {}; + size_t required_cmd_sz; + int ret; + + required_cmd_sz = offsetof(typeof(cmd), curr_wq_state) + sizeof(cmd.curr_wq_state); + if (ucore->inlen < required_cmd_sz) + return -EINVAL; + + if (ucore->inlen > sizeof(cmd) && + !ib_is_udata_cleared(ucore, sizeof(cmd), + ucore->inlen - sizeof(cmd))) + return -EOPNOTSUPP; + + ret = ib_copy_from_udata(&cmd, ucore, min(sizeof(cmd), ucore->inlen)); + if (ret) + return ret; + + if (!cmd.attr_mask) + return -EINVAL; + + if (cmd.attr_mask > (IB_WQ_STATE | IB_WQ_CUR_STATE)) + return -EINVAL; + + wq = idr_read_wq(cmd.wq_handle, file->ucontext); + if (!wq) + return -EINVAL; + + wq_attr.curr_wq_state = cmd.curr_wq_state; + wq_attr.wq_state = cmd.wq_state; + ret = wq->device->modify_wq(wq, &wq_attr, cmd.attr_mask, uhw); + put_wq_read(wq); + return ret; +} + int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, struct ib_device *ib_dev, struct ib_udata *ucore, diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 31f422a70623..91cb36f66ea7 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -76,6 +76,7 @@ DEFINE_IDR(ib_uverbs_qp_idr); DEFINE_IDR(ib_uverbs_srq_idr); DEFINE_IDR(ib_uverbs_xrcd_idr); DEFINE_IDR(ib_uverbs_rule_idr); +DEFINE_IDR(ib_uverbs_wq_idr); static DEFINE_SPINLOCK(map_lock); static DECLARE_BITMAP(dev_map, IB_UVERBS_MAX_DEVICES); @@ -130,6 +131,9 @@ static int (*uverbs_ex_cmd_table[])(struct ib_uverbs_file *file, [IB_USER_VERBS_EX_CMD_QUERY_DEVICE] = ib_uverbs_ex_query_device, [IB_USER_VERBS_EX_CMD_CREATE_CQ] = ib_uverbs_ex_create_cq, [IB_USER_VERBS_EX_CMD_CREATE_QP] = ib_uverbs_ex_create_qp, + [IB_USER_VERBS_EX_CMD_CREATE_WQ] = ib_uverbs_ex_create_wq, + [IB_USER_VERBS_EX_CMD_MODIFY_WQ] = ib_uverbs_ex_modify_wq, + [IB_USER_VERBS_EX_CMD_DESTROY_WQ] = ib_uverbs_ex_destroy_wq, }; static void ib_uverbs_add_one(struct ib_device *device); @@ -265,6 +269,17 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, kfree(uqp); } + list_for_each_entry_safe(uobj, tmp, &context->wq_list, list) { + struct ib_wq *wq = uobj->object; + struct ib_uwq_object *uwq = + container_of(uobj, struct ib_uwq_object, uevent.uobject); + + idr_remove_uobj(&ib_uverbs_wq_idr, uobj); + ib_destroy_wq(wq); + ib_uverbs_release_uevent(file, &uwq->uevent); + kfree(uwq); + } + list_for_each_entry_safe(uobj, tmp, &context->srq_list, list) { struct ib_srq *srq = uobj->object; struct ib_uevent_object *uevent = @@ -568,6 +583,16 @@ void ib_uverbs_qp_event_handler(struct ib_event *event, void *context_ptr) &uobj->events_reported); } +void ib_uverbs_wq_event_handler(struct ib_event *event, void *context_ptr) +{ + struct ib_uevent_object *uobj = container_of(event->element.wq->uobject, + struct ib_uevent_object, uobject); + + ib_uverbs_async_handler(context_ptr, uobj->uobject.user_handle, + event->event, &uobj->event_list, + &uobj->events_reported); +} + void ib_uverbs_srq_event_handler(struct ib_event *event, void *context_ptr) { struct ib_uevent_object *uobj; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index f2d954ac42f6..0c1956a98573 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -562,6 +562,7 @@ enum ib_event_type { IB_EVENT_QP_LAST_WQE_REACHED, IB_EVENT_CLIENT_REREGISTER, IB_EVENT_GID_CHANGE, + IB_EVENT_WQ_FATAL, }; const char *__attribute_const__ ib_event_msg(enum ib_event_type event); @@ -572,6 +573,7 @@ struct ib_event { struct ib_cq *cq; struct ib_qp *qp; struct ib_srq *srq; + struct ib_wq *wq; u8 port_num; } element; enum ib_event_type event; @@ -1323,6 +1325,7 @@ struct ib_ucontext { struct list_head ah_list; struct list_head xrcd_list; struct list_head rule_list; + struct list_head wq_list; int closing; struct pid *tgid; diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index b6543d73d20a..c9470e511e03 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -95,6 +95,9 @@ enum { IB_USER_VERBS_EX_CMD_CREATE_QP = IB_USER_VERBS_CMD_CREATE_QP, IB_USER_VERBS_EX_CMD_CREATE_FLOW = IB_USER_VERBS_CMD_THRESHOLD, IB_USER_VERBS_EX_CMD_DESTROY_FLOW, + IB_USER_VERBS_EX_CMD_CREATE_WQ, + IB_USER_VERBS_EX_CMD_MODIFY_WQ, + IB_USER_VERBS_EX_CMD_DESTROY_WQ, }; /* @@ -946,4 +949,42 @@ struct ib_uverbs_destroy_srq_resp { __u32 events_reported; }; +struct ib_uverbs_ex_create_wq { + __u32 comp_mask; + __u32 wq_type; + __u64 user_handle; + __u32 pd_handle; + __u32 cq_handle; + __u32 max_wr; + __u32 max_sge; +}; + +struct ib_uverbs_ex_create_wq_resp { + __u32 comp_mask; + __u32 response_length; + __u32 wq_handle; + __u32 max_wr; + __u32 max_sge; + __u32 wqn; +}; + +struct ib_uverbs_ex_destroy_wq { + __u32 comp_mask; + __u32 wq_handle; +}; + +struct ib_uverbs_ex_destroy_wq_resp { + __u32 comp_mask; + __u32 response_length; + __u32 events_reported; + __u32 reserved; +}; + +struct ib_uverbs_ex_modify_wq { + __u32 attr_mask; + __u32 wq_handle; + __u32 wq_state; + __u32 curr_wq_state; +}; + #endif /* IB_USER_VERBS_H */ -- cgit v1.2.3 From de019a94049d579608a5511f8c50652faf125182 Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 23 May 2016 15:20:52 +0300 Subject: IB/uverbs: Introduce RWQ Indirection table User applications that want to spread traffic on several WQs, need to create an indirection table, by using already created WQs. Adding uverbs API in order to create and destroy this table. Signed-off-by: Yishai Hadas Signed-off-by: Matan Barak Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/core/uverbs.h | 3 + drivers/infiniband/core/uverbs_cmd.c | 210 ++++++++++++++++++++++++++++++++++ drivers/infiniband/core/uverbs_main.c | 13 +++ include/rdma/ib_verbs.h | 1 + include/uapi/rdma/ib_user_verbs.h | 26 +++++ 5 files changed, 253 insertions(+) (limited to 'include/uapi') diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index 74776c6531f4..6c2292397cd6 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -186,6 +186,7 @@ extern struct idr ib_uverbs_srq_idr; extern struct idr ib_uverbs_xrcd_idr; extern struct idr ib_uverbs_rule_idr; extern struct idr ib_uverbs_wq_idr; +extern struct idr ib_uverbs_rwq_ind_tbl_idr; void idr_remove_uobj(struct idr *idp, struct ib_uobject *uobj); @@ -284,5 +285,7 @@ IB_UVERBS_DECLARE_EX_CMD(create_qp); IB_UVERBS_DECLARE_EX_CMD(create_wq); IB_UVERBS_DECLARE_EX_CMD(modify_wq); IB_UVERBS_DECLARE_EX_CMD(destroy_wq); +IB_UVERBS_DECLARE_EX_CMD(create_rwq_ind_table); +IB_UVERBS_DECLARE_EX_CMD(destroy_rwq_ind_table); #endif /* UVERBS_H */ diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 22e617391657..327a56cccc27 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -58,6 +58,7 @@ static struct uverbs_lock_class srq_lock_class = { .name = "SRQ-uobj" }; static struct uverbs_lock_class xrcd_lock_class = { .name = "XRCD-uobj" }; static struct uverbs_lock_class rule_lock_class = { .name = "RULE-uobj" }; static struct uverbs_lock_class wq_lock_class = { .name = "WQ-uobj" }; +static struct uverbs_lock_class rwq_ind_table_lock_class = { .name = "IND_TBL-uobj" }; /* * The ib_uobject locking scheme is as follows: @@ -338,6 +339,7 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, INIT_LIST_HEAD(&ucontext->srq_list); INIT_LIST_HEAD(&ucontext->ah_list); INIT_LIST_HEAD(&ucontext->wq_list); + INIT_LIST_HEAD(&ucontext->rwq_ind_tbl_list); INIT_LIST_HEAD(&ucontext->xrcd_list); INIT_LIST_HEAD(&ucontext->rule_list); rcu_read_lock(); @@ -3299,6 +3301,214 @@ int ib_uverbs_ex_modify_wq(struct ib_uverbs_file *file, return ret; } +int ib_uverbs_ex_create_rwq_ind_table(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + struct ib_udata *ucore, + struct ib_udata *uhw) +{ + struct ib_uverbs_ex_create_rwq_ind_table cmd = {}; + struct ib_uverbs_ex_create_rwq_ind_table_resp resp = {}; + struct ib_uobject *uobj; + int err = 0; + struct ib_rwq_ind_table_init_attr init_attr = {}; + struct ib_rwq_ind_table *rwq_ind_tbl; + struct ib_wq **wqs = NULL; + u32 *wqs_handles = NULL; + struct ib_wq *wq = NULL; + int i, j, num_read_wqs; + u32 num_wq_handles; + u32 expected_in_size; + size_t required_cmd_sz_header; + size_t required_resp_len; + + required_cmd_sz_header = offsetof(typeof(cmd), log_ind_tbl_size) + sizeof(cmd.log_ind_tbl_size); + required_resp_len = offsetof(typeof(resp), ind_tbl_num) + sizeof(resp.ind_tbl_num); + + if (ucore->inlen < required_cmd_sz_header) + return -EINVAL; + + if (ucore->outlen < required_resp_len) + return -ENOSPC; + + err = ib_copy_from_udata(&cmd, ucore, required_cmd_sz_header); + if (err) + return err; + + ucore->inbuf += required_cmd_sz_header; + ucore->inlen -= required_cmd_sz_header; + + if (cmd.comp_mask) + return -EOPNOTSUPP; + + if (cmd.log_ind_tbl_size > IB_USER_VERBS_MAX_LOG_IND_TBL_SIZE) + return -EINVAL; + + num_wq_handles = 1 << cmd.log_ind_tbl_size; + expected_in_size = num_wq_handles * sizeof(__u32); + if (num_wq_handles == 1) + /* input size for wq handles is u64 aligned */ + expected_in_size += sizeof(__u32); + + if (ucore->inlen < expected_in_size) + return -EINVAL; + + if (ucore->inlen > expected_in_size && + !ib_is_udata_cleared(ucore, expected_in_size, + ucore->inlen - expected_in_size)) + return -EOPNOTSUPP; + + wqs_handles = kcalloc(num_wq_handles, sizeof(*wqs_handles), + GFP_KERNEL); + if (!wqs_handles) + return -ENOMEM; + + err = ib_copy_from_udata(wqs_handles, ucore, + num_wq_handles * sizeof(__u32)); + if (err) + goto err_free; + + wqs = kcalloc(num_wq_handles, sizeof(*wqs), GFP_KERNEL); + if (!wqs) { + err = -ENOMEM; + goto err_free; + } + + for (num_read_wqs = 0; num_read_wqs < num_wq_handles; + num_read_wqs++) { + wq = idr_read_wq(wqs_handles[num_read_wqs], file->ucontext); + if (!wq) { + err = -EINVAL; + goto put_wqs; + } + + wqs[num_read_wqs] = wq; + } + + uobj = kmalloc(sizeof(*uobj), GFP_KERNEL); + if (!uobj) { + err = -ENOMEM; + goto put_wqs; + } + + init_uobj(uobj, 0, file->ucontext, &rwq_ind_table_lock_class); + down_write(&uobj->mutex); + init_attr.log_ind_tbl_size = cmd.log_ind_tbl_size; + init_attr.ind_tbl = wqs; + rwq_ind_tbl = ib_dev->create_rwq_ind_table(ib_dev, &init_attr, uhw); + + if (IS_ERR(rwq_ind_tbl)) { + err = PTR_ERR(rwq_ind_tbl); + goto err_uobj; + } + + rwq_ind_tbl->ind_tbl = wqs; + rwq_ind_tbl->log_ind_tbl_size = init_attr.log_ind_tbl_size; + rwq_ind_tbl->uobject = uobj; + uobj->object = rwq_ind_tbl; + rwq_ind_tbl->device = ib_dev; + atomic_set(&rwq_ind_tbl->usecnt, 0); + + for (i = 0; i < num_wq_handles; i++) + atomic_inc(&wqs[i]->usecnt); + + err = idr_add_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); + if (err) + goto destroy_ind_tbl; + + resp.ind_tbl_handle = uobj->id; + resp.ind_tbl_num = rwq_ind_tbl->ind_tbl_num; + resp.response_length = required_resp_len; + + err = ib_copy_to_udata(ucore, + &resp, resp.response_length); + if (err) + goto err_copy; + + kfree(wqs_handles); + + for (j = 0; j < num_read_wqs; j++) + put_wq_read(wqs[j]); + + mutex_lock(&file->mutex); + list_add_tail(&uobj->list, &file->ucontext->rwq_ind_tbl_list); + mutex_unlock(&file->mutex); + + uobj->live = 1; + + up_write(&uobj->mutex); + return 0; + +err_copy: + idr_remove_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); +destroy_ind_tbl: + ib_destroy_rwq_ind_table(rwq_ind_tbl); +err_uobj: + put_uobj_write(uobj); +put_wqs: + for (j = 0; j < num_read_wqs; j++) + put_wq_read(wqs[j]); +err_free: + kfree(wqs_handles); + kfree(wqs); + return err; +} + +int ib_uverbs_ex_destroy_rwq_ind_table(struct ib_uverbs_file *file, + struct ib_device *ib_dev, + struct ib_udata *ucore, + struct ib_udata *uhw) +{ + struct ib_uverbs_ex_destroy_rwq_ind_table cmd = {}; + struct ib_rwq_ind_table *rwq_ind_tbl; + struct ib_uobject *uobj; + int ret; + struct ib_wq **ind_tbl; + size_t required_cmd_sz; + + required_cmd_sz = offsetof(typeof(cmd), ind_tbl_handle) + sizeof(cmd.ind_tbl_handle); + + if (ucore->inlen < required_cmd_sz) + return -EINVAL; + + if (ucore->inlen > sizeof(cmd) && + !ib_is_udata_cleared(ucore, sizeof(cmd), + ucore->inlen - sizeof(cmd))) + return -EOPNOTSUPP; + + ret = ib_copy_from_udata(&cmd, ucore, min(sizeof(cmd), ucore->inlen)); + if (ret) + return ret; + + if (cmd.comp_mask) + return -EOPNOTSUPP; + + uobj = idr_write_uobj(&ib_uverbs_rwq_ind_tbl_idr, cmd.ind_tbl_handle, + file->ucontext); + if (!uobj) + return -EINVAL; + rwq_ind_tbl = uobj->object; + ind_tbl = rwq_ind_tbl->ind_tbl; + + ret = ib_destroy_rwq_ind_table(rwq_ind_tbl); + if (!ret) + uobj->live = 0; + + put_uobj_write(uobj); + + if (ret) + return ret; + + idr_remove_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); + + mutex_lock(&file->mutex); + list_del(&uobj->list); + mutex_unlock(&file->mutex); + + put_uobj(uobj); + kfree(ind_tbl); + return ret; +} + int ib_uverbs_ex_create_flow(struct ib_uverbs_file *file, struct ib_device *ib_dev, struct ib_udata *ucore, diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 91cb36f66ea7..426e0ac203fc 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -77,6 +77,7 @@ DEFINE_IDR(ib_uverbs_srq_idr); DEFINE_IDR(ib_uverbs_xrcd_idr); DEFINE_IDR(ib_uverbs_rule_idr); DEFINE_IDR(ib_uverbs_wq_idr); +DEFINE_IDR(ib_uverbs_rwq_ind_tbl_idr); static DEFINE_SPINLOCK(map_lock); static DECLARE_BITMAP(dev_map, IB_UVERBS_MAX_DEVICES); @@ -134,6 +135,8 @@ static int (*uverbs_ex_cmd_table[])(struct ib_uverbs_file *file, [IB_USER_VERBS_EX_CMD_CREATE_WQ] = ib_uverbs_ex_create_wq, [IB_USER_VERBS_EX_CMD_MODIFY_WQ] = ib_uverbs_ex_modify_wq, [IB_USER_VERBS_EX_CMD_DESTROY_WQ] = ib_uverbs_ex_destroy_wq, + [IB_USER_VERBS_EX_CMD_CREATE_RWQ_IND_TBL] = ib_uverbs_ex_create_rwq_ind_table, + [IB_USER_VERBS_EX_CMD_DESTROY_RWQ_IND_TBL] = ib_uverbs_ex_destroy_rwq_ind_table, }; static void ib_uverbs_add_one(struct ib_device *device); @@ -269,6 +272,16 @@ static int ib_uverbs_cleanup_ucontext(struct ib_uverbs_file *file, kfree(uqp); } + list_for_each_entry_safe(uobj, tmp, &context->rwq_ind_tbl_list, list) { + struct ib_rwq_ind_table *rwq_ind_tbl = uobj->object; + struct ib_wq **ind_tbl = rwq_ind_tbl->ind_tbl; + + idr_remove_uobj(&ib_uverbs_rwq_ind_tbl_idr, uobj); + ib_destroy_rwq_ind_table(rwq_ind_tbl); + kfree(ind_tbl); + kfree(uobj); + } + list_for_each_entry_safe(uobj, tmp, &context->wq_list, list) { struct ib_wq *wq = uobj->object; struct ib_uwq_object *uwq = diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index fa2e0184dcc5..e305c9a36663 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1326,6 +1326,7 @@ struct ib_ucontext { struct list_head xrcd_list; struct list_head rule_list; struct list_head wq_list; + struct list_head rwq_ind_tbl_list; int closing; struct pid *tgid; diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index c9470e511e03..2cf7c95860d8 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -98,6 +98,8 @@ enum { IB_USER_VERBS_EX_CMD_CREATE_WQ, IB_USER_VERBS_EX_CMD_MODIFY_WQ, IB_USER_VERBS_EX_CMD_DESTROY_WQ, + IB_USER_VERBS_EX_CMD_CREATE_RWQ_IND_TBL, + IB_USER_VERBS_EX_CMD_DESTROY_RWQ_IND_TBL }; /* @@ -987,4 +989,28 @@ struct ib_uverbs_ex_modify_wq { __u32 curr_wq_state; }; +/* Prevent memory allocation rather than max expected size */ +#define IB_USER_VERBS_MAX_LOG_IND_TBL_SIZE 0x0d +struct ib_uverbs_ex_create_rwq_ind_table { + __u32 comp_mask; + __u32 log_ind_tbl_size; + /* Following are the wq handles according to log_ind_tbl_size + * wq_handle1 + * wq_handle2 + */ + __u32 wq_handles[0]; +}; + +struct ib_uverbs_ex_create_rwq_ind_table_resp { + __u32 comp_mask; + __u32 response_length; + __u32 ind_tbl_handle; + __u32 ind_tbl_num; +}; + +struct ib_uverbs_ex_destroy_rwq_ind_table { + __u32 comp_mask; + __u32 ind_tbl_handle; +}; + #endif /* IB_USER_VERBS_H */ -- cgit v1.2.3 From c70285f880e88cb4f73effb722065a182ba5936f Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 23 May 2016 15:20:55 +0300 Subject: IB/uverbs: Extend create QP to get RWQ indirection table User applications that want to spread incoming traffic between several WQs should create a QP which contains an indirection table. When such a QP is created other receive side parameters are not valid and should not be given. Its send side is optional and assumed active based on max_send_wr capability value. Extend create QP to work accordingly. Signed-off-by: Yishai Hadas Signed-off-by: Matan Barak Reviewed-by: Sagi Grimberg Signed-off-by: Doug Ledford --- drivers/infiniband/core/uverbs_cmd.c | 75 ++++++++++++++++++++++++++++++------ include/uapi/rdma/ib_user_verbs.h | 10 +++++ 2 files changed, 73 insertions(+), 12 deletions(-) (limited to 'include/uapi') diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 327a56cccc27..65ab2097cd81 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -255,6 +255,17 @@ static void put_wq_read(struct ib_wq *wq) put_uobj_read(wq->uobject); } +static struct ib_rwq_ind_table *idr_read_rwq_indirection_table(int ind_table_handle, + struct ib_ucontext *context) +{ + return idr_read_obj(&ib_uverbs_rwq_ind_tbl_idr, ind_table_handle, context, 0); +} + +static void put_rwq_indirection_table_read(struct ib_rwq_ind_table *ind_table) +{ + put_uobj_read(ind_table->uobject); +} + static struct ib_qp *idr_write_qp(int qp_handle, struct ib_ucontext *context) { struct ib_uobject *uobj; @@ -1761,9 +1772,11 @@ static int create_qp(struct ib_uverbs_file *file, struct ib_srq *srq = NULL; struct ib_qp *qp; char *buf; - struct ib_qp_init_attr attr; + struct ib_qp_init_attr attr = {}; struct ib_uverbs_ex_create_qp_resp resp; int ret; + struct ib_rwq_ind_table *ind_tbl = NULL; + bool has_sq = true; if (cmd->qp_type == IB_QPT_RAW_PACKET && !capable(CAP_NET_RAW)) return -EPERM; @@ -1775,6 +1788,32 @@ static int create_qp(struct ib_uverbs_file *file, init_uobj(&obj->uevent.uobject, cmd->user_handle, file->ucontext, &qp_lock_class); down_write(&obj->uevent.uobject.mutex); + if (cmd_sz >= offsetof(typeof(*cmd), rwq_ind_tbl_handle) + + sizeof(cmd->rwq_ind_tbl_handle) && + (cmd->comp_mask & IB_UVERBS_CREATE_QP_MASK_IND_TABLE)) { + ind_tbl = idr_read_rwq_indirection_table(cmd->rwq_ind_tbl_handle, + file->ucontext); + if (!ind_tbl) { + ret = -EINVAL; + goto err_put; + } + + attr.rwq_ind_tbl = ind_tbl; + } + + if ((cmd_sz >= offsetof(typeof(*cmd), reserved1) + + sizeof(cmd->reserved1)) && cmd->reserved1) { + ret = -EOPNOTSUPP; + goto err_put; + } + + if (ind_tbl && (cmd->max_recv_wr || cmd->max_recv_sge || cmd->is_srq)) { + ret = -EINVAL; + goto err_put; + } + + if (ind_tbl && !cmd->max_send_wr) + has_sq = false; if (cmd->qp_type == IB_QPT_XRC_TGT) { xrcd = idr_read_xrcd(cmd->pd_handle, file->ucontext, @@ -1798,20 +1837,24 @@ static int create_qp(struct ib_uverbs_file *file, } } - if (cmd->recv_cq_handle != cmd->send_cq_handle) { - rcq = idr_read_cq(cmd->recv_cq_handle, - file->ucontext, 0); - if (!rcq) { - ret = -EINVAL; - goto err_put; + if (!ind_tbl) { + if (cmd->recv_cq_handle != cmd->send_cq_handle) { + rcq = idr_read_cq(cmd->recv_cq_handle, + file->ucontext, 0); + if (!rcq) { + ret = -EINVAL; + goto err_put; + } } } } - scq = idr_read_cq(cmd->send_cq_handle, file->ucontext, !!rcq); - rcq = rcq ?: scq; + if (has_sq) + scq = idr_read_cq(cmd->send_cq_handle, file->ucontext, !!rcq); + if (!ind_tbl) + rcq = rcq ?: scq; pd = idr_read_pd(cmd->pd_handle, file->ucontext); - if (!pd || !scq) { + if (!pd || (!scq && has_sq)) { ret = -EINVAL; goto err_put; } @@ -1878,16 +1921,20 @@ static int create_qp(struct ib_uverbs_file *file, qp->send_cq = attr.send_cq; qp->recv_cq = attr.recv_cq; qp->srq = attr.srq; + qp->rwq_ind_tbl = ind_tbl; qp->event_handler = attr.event_handler; qp->qp_context = attr.qp_context; qp->qp_type = attr.qp_type; atomic_set(&qp->usecnt, 0); atomic_inc(&pd->usecnt); - atomic_inc(&attr.send_cq->usecnt); + if (attr.send_cq) + atomic_inc(&attr.send_cq->usecnt); if (attr.recv_cq) atomic_inc(&attr.recv_cq->usecnt); if (attr.srq) atomic_inc(&attr.srq->usecnt); + if (ind_tbl) + atomic_inc(&ind_tbl->usecnt); } qp->uobject = &obj->uevent.uobject; @@ -1927,6 +1974,8 @@ static int create_qp(struct ib_uverbs_file *file, put_cq_read(rcq); if (srq) put_srq_read(srq); + if (ind_tbl) + put_rwq_indirection_table_read(ind_tbl); mutex_lock(&file->mutex); list_add_tail(&obj->uevent.uobject.list, &file->ucontext->qp_list); @@ -1954,6 +2003,8 @@ err_put: put_cq_read(rcq); if (srq) put_srq_read(srq); + if (ind_tbl) + put_rwq_indirection_table_read(ind_tbl); put_uobj_write(&obj->uevent.uobject); return ret; @@ -2047,7 +2098,7 @@ int ib_uverbs_ex_create_qp(struct ib_uverbs_file *file, if (err) return err; - if (cmd.comp_mask) + if (cmd.comp_mask & ~IB_UVERBS_CREATE_QP_SUP_COMP_MASK) return -EINVAL; if (cmd.reserved) diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index 2cf7c95860d8..2c8bca8cd2c2 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -523,6 +523,14 @@ struct ib_uverbs_create_qp { __u64 driver_data[0]; }; +enum ib_uverbs_create_qp_mask { + IB_UVERBS_CREATE_QP_MASK_IND_TABLE = 1UL << 0, +}; + +enum { + IB_UVERBS_CREATE_QP_SUP_COMP_MASK = IB_UVERBS_CREATE_QP_MASK_IND_TABLE, +}; + struct ib_uverbs_ex_create_qp { __u64 user_handle; __u32 pd_handle; @@ -540,6 +548,8 @@ struct ib_uverbs_ex_create_qp { __u8 reserved; __u32 comp_mask; __u32 create_flags; + __u32 rwq_ind_tbl_handle; + __u32 reserved1; }; struct ib_uverbs_open_qp { -- cgit v1.2.3 From 4c2aae712cb024f9d30a1fa62e3ba2ff785c6a3e Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Fri, 17 Jun 2016 15:14:50 +0300 Subject: IB/core: Add IPv6 support to flow steering Add IPv6 flow specification support. Signed-off-by: Maor Gottlieb Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/core/uverbs.h | 1 + drivers/infiniband/core/uverbs_cmd.c | 9 +++++++++ include/rdma/ib_verbs.h | 14 ++++++++++++++ include/uapi/rdma/ib_user_verbs.h | 18 ++++++++++++++++++ 4 files changed, 42 insertions(+) (limited to 'include/uapi') diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index 6c2292397cd6..b7f3b8d359b9 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -226,6 +226,7 @@ struct ib_uverbs_flow_spec { struct ib_uverbs_flow_spec_eth eth; struct ib_uverbs_flow_spec_ipv4 ipv4; struct ib_uverbs_flow_spec_tcp_udp tcp_udp; + struct ib_uverbs_flow_spec_ipv6 ipv6; }; }; diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 65ab2097cd81..f6647318138d 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -3105,6 +3105,15 @@ static int kern_spec_to_ib_spec(struct ib_uverbs_flow_spec *kern_spec, memcpy(&ib_spec->ipv4.mask, &kern_spec->ipv4.mask, sizeof(struct ib_flow_ipv4_filter)); break; + case IB_FLOW_SPEC_IPV6: + ib_spec->ipv6.size = sizeof(struct ib_flow_spec_ipv6); + if (ib_spec->ipv6.size != kern_spec->ipv6.size) + return -EINVAL; + memcpy(&ib_spec->ipv6.val, &kern_spec->ipv6.val, + sizeof(struct ib_flow_ipv6_filter)); + memcpy(&ib_spec->ipv6.mask, &kern_spec->ipv6.mask, + sizeof(struct ib_flow_ipv6_filter)); + break; case IB_FLOW_SPEC_TCP: case IB_FLOW_SPEC_UDP: ib_spec->tcp_udp.size = sizeof(struct ib_flow_spec_tcp_udp); diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 9b2fafe5eb38..9bbca6887f7f 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1569,6 +1569,7 @@ enum ib_flow_spec_type { IB_FLOW_SPEC_IB = 0x22, /* L3 header*/ IB_FLOW_SPEC_IPV4 = 0x30, + IB_FLOW_SPEC_IPV6 = 0x31, /* L4 headers*/ IB_FLOW_SPEC_TCP = 0x40, IB_FLOW_SPEC_UDP = 0x41 @@ -1630,6 +1631,18 @@ struct ib_flow_spec_ipv4 { struct ib_flow_ipv4_filter mask; }; +struct ib_flow_ipv6_filter { + u8 src_ip[16]; + u8 dst_ip[16]; +}; + +struct ib_flow_spec_ipv6 { + enum ib_flow_spec_type type; + u16 size; + struct ib_flow_ipv6_filter val; + struct ib_flow_ipv6_filter mask; +}; + struct ib_flow_tcp_udp_filter { __be16 dst_port; __be16 src_port; @@ -1651,6 +1664,7 @@ union ib_flow_spec { struct ib_flow_spec_ib ib; struct ib_flow_spec_ipv4 ipv4; struct ib_flow_spec_tcp_udp tcp_udp; + struct ib_flow_spec_ipv6 ipv6; }; struct ib_flow_attr { diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h index 2c8bca8cd2c2..7f035f4b53b0 100644 --- a/include/uapi/rdma/ib_user_verbs.h +++ b/include/uapi/rdma/ib_user_verbs.h @@ -867,6 +867,24 @@ struct ib_uverbs_flow_spec_tcp_udp { struct ib_uverbs_flow_tcp_udp_filter mask; }; +struct ib_uverbs_flow_ipv6_filter { + __u8 src_ip[16]; + __u8 dst_ip[16]; +}; + +struct ib_uverbs_flow_spec_ipv6 { + union { + struct ib_uverbs_flow_spec_hdr hdr; + struct { + __u32 type; + __u16 size; + __u16 reserved; + }; + }; + struct ib_uverbs_flow_ipv6_filter val; + struct ib_uverbs_flow_ipv6_filter mask; +}; + struct ib_uverbs_flow_attr { __u32 type; __u16 size; -- cgit v1.2.3 From 7643507fe8b5bd8ab7522f6a81058cc1209d2585 Mon Sep 17 00:00:00 2001 From: Vishwanath Pai Date: Tue, 21 Jun 2016 14:58:46 -0400 Subject: netfilter: xt_NFLOG: nflog-range does not truncate packets li->u.ulog.copy_len is currently ignored by the kernel, we should truncate the packet to either li->u.ulog.copy_len (if set) or copy_range before sending it to userspace. 0 is a valid input for copy_len, so add a new flag to indicate whether this was option was specified by the user or not. Add two flags to indicate whether nflog-size/copy_len was set or not. XT_NFLOG_F_COPY_LEN is for XT_NFLOG and NFLOG_F_COPY_LEN for nfnetlink_log On the userspace side, this was initially represented by the option nflog-range, this will be replaced by --nflog-size now. --nflog-range would still exist but does not do anything. Reported-by: Joe Dollard Reviewed-by: Josh Hunt Signed-off-by: Vishwanath Pai Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_log.h | 7 +++++++ include/uapi/linux/netfilter/xt_NFLOG.h | 6 +++++- net/netfilter/nfnetlink_log.c | 9 ++++++--- net/netfilter/xt_NFLOG.c | 3 +++ 4 files changed, 21 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/include/net/netfilter/nf_log.h b/include/net/netfilter/nf_log.h index 57639fca223a..83d855ba6af1 100644 --- a/include/net/netfilter/nf_log.h +++ b/include/net/netfilter/nf_log.h @@ -12,6 +12,9 @@ #define NF_LOG_UID 0x08 /* Log UID owning local socket */ #define NF_LOG_MASK 0x0f +/* This flag indicates that copy_len field in nf_loginfo is set */ +#define NF_LOG_F_COPY_LEN 0x1 + enum nf_log_type { NF_LOG_TYPE_LOG = 0, NF_LOG_TYPE_ULOG, @@ -22,9 +25,13 @@ struct nf_loginfo { u_int8_t type; union { struct { + /* copy_len will be used iff you set + * NF_LOG_F_COPY_LEN in flags + */ u_int32_t copy_len; u_int16_t group; u_int16_t qthreshold; + u_int16_t flags; } ulog; struct { u_int8_t level; diff --git a/include/uapi/linux/netfilter/xt_NFLOG.h b/include/uapi/linux/netfilter/xt_NFLOG.h index 87b58311ce6b..f33070730fc8 100644 --- a/include/uapi/linux/netfilter/xt_NFLOG.h +++ b/include/uapi/linux/netfilter/xt_NFLOG.h @@ -6,9 +6,13 @@ #define XT_NFLOG_DEFAULT_GROUP 0x1 #define XT_NFLOG_DEFAULT_THRESHOLD 0 -#define XT_NFLOG_MASK 0x0 +#define XT_NFLOG_MASK 0x1 + +/* This flag indicates that 'len' field in xt_nflog_info is set*/ +#define XT_NFLOG_F_COPY_LEN 0x1 struct xt_nflog_info { + /* 'len' will be used iff you set XT_NFLOG_F_COPY_LEN in flags */ __u32 len; __u16 group; __u16 threshold; diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 11f81c8385fc..cbcfdfb586a6 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -700,10 +700,13 @@ nfulnl_log_packet(struct net *net, break; case NFULNL_COPY_PACKET: - if (inst->copy_range > skb->len) + data_len = inst->copy_range; + if ((li->u.ulog.flags & NF_LOG_F_COPY_LEN) && + (li->u.ulog.copy_len < data_len)) + data_len = li->u.ulog.copy_len; + + if (data_len > skb->len) data_len = skb->len; - else - data_len = inst->copy_range; size += nla_total_size(data_len); break; diff --git a/net/netfilter/xt_NFLOG.c b/net/netfilter/xt_NFLOG.c index a1fa2c800cb9..018eed7e1ff1 100644 --- a/net/netfilter/xt_NFLOG.c +++ b/net/netfilter/xt_NFLOG.c @@ -33,6 +33,9 @@ nflog_tg(struct sk_buff *skb, const struct xt_action_param *par) li.u.ulog.group = info->group; li.u.ulog.qthreshold = info->threshold; + if (info->flags & XT_NFLOG_F_COPY_LEN) + li.u.ulog.flags |= NF_LOG_F_COPY_LEN; + nfulnl_log_packet(net, par->family, par->hooknum, skb, par->in, par->out, &li, info->prefix); return XT_CONTINUE; -- cgit v1.2.3 From 0071e184a535e40ce487528cb04f4690cb0da881 Mon Sep 17 00:00:00 2001 From: Arturo Borrero Date: Thu, 23 Jun 2016 12:24:08 +0200 Subject: netfilter: nf_tables: add support for inverted logic in nft_lookup Introduce a new configuration option for this expression, which allows users to invert the logic of set lookups. In _init() we will now return EINVAL if NFT_LOOKUP_F_INV is in anyway related to a map lookup. The code in the _eval() function has been untangled and updated to sopport the XOR of options, as we should consider 4 cases: * lookup false, invert false -> NFT_BREAK * lookup false, invert true -> return w/o NFT_BREAK * lookup true, invert false -> return w/o NFT_BREAK * lookup true, invert true -> NFT_BREAK Signed-off-by: Arturo Borrero Gonzalez Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 6 ++++++ net/netfilter/nft_lookup.c | 37 +++++++++++++++++++++++++++----- 2 files changed, 38 insertions(+), 5 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 6a4dbe04f09e..01751faccaf8 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -546,6 +546,10 @@ enum nft_cmp_attributes { }; #define NFTA_CMP_MAX (__NFTA_CMP_MAX - 1) +enum nft_lookup_flags { + NFT_LOOKUP_F_INV = (1 << 0), +}; + /** * enum nft_lookup_attributes - nf_tables set lookup expression netlink attributes * @@ -553,6 +557,7 @@ enum nft_cmp_attributes { * @NFTA_LOOKUP_SREG: source register of the data to look for (NLA_U32: nft_registers) * @NFTA_LOOKUP_DREG: destination register (NLA_U32: nft_registers) * @NFTA_LOOKUP_SET_ID: uniquely identifies a set in a transaction (NLA_U32) + * @NFTA_LOOKUP_FLAGS: flags (NLA_U32: enum nft_lookup_flags) */ enum nft_lookup_attributes { NFTA_LOOKUP_UNSPEC, @@ -560,6 +565,7 @@ enum nft_lookup_attributes { NFTA_LOOKUP_SREG, NFTA_LOOKUP_DREG, NFTA_LOOKUP_SET_ID, + NFTA_LOOKUP_FLAGS, __NFTA_LOOKUP_MAX }; #define NFTA_LOOKUP_MAX (__NFTA_LOOKUP_MAX - 1) diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 8a102cf855d0..b8d18f598569 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -22,6 +22,7 @@ struct nft_lookup { struct nft_set *set; enum nft_registers sreg:8; enum nft_registers dreg:8; + bool invert; struct nft_set_binding binding; }; @@ -32,14 +33,20 @@ static void nft_lookup_eval(const struct nft_expr *expr, const struct nft_lookup *priv = nft_expr_priv(expr); const struct nft_set *set = priv->set; const struct nft_set_ext *ext; + bool found; - if (set->ops->lookup(set, ®s->data[priv->sreg], &ext)) { - if (set->flags & NFT_SET_MAP) - nft_data_copy(®s->data[priv->dreg], - nft_set_ext_data(ext), set->dlen); + found = set->ops->lookup(set, ®s->data[priv->sreg], &ext) ^ + priv->invert; + + if (!found) { + regs->verdict.code = NFT_BREAK; return; } - regs->verdict.code = NFT_BREAK; + + if (found && set->flags & NFT_SET_MAP) + nft_data_copy(®s->data[priv->dreg], + nft_set_ext_data(ext), set->dlen); + } static const struct nla_policy nft_lookup_policy[NFTA_LOOKUP_MAX + 1] = { @@ -47,6 +54,7 @@ static const struct nla_policy nft_lookup_policy[NFTA_LOOKUP_MAX + 1] = { [NFTA_LOOKUP_SET_ID] = { .type = NLA_U32 }, [NFTA_LOOKUP_SREG] = { .type = NLA_U32 }, [NFTA_LOOKUP_DREG] = { .type = NLA_U32 }, + [NFTA_LOOKUP_FLAGS] = { .type = NLA_U32 }, }; static int nft_lookup_init(const struct nft_ctx *ctx, @@ -56,6 +64,7 @@ static int nft_lookup_init(const struct nft_ctx *ctx, struct nft_lookup *priv = nft_expr_priv(expr); u8 genmask = nft_genmask_next(ctx->net); struct nft_set *set; + u32 flags; int err; if (tb[NFTA_LOOKUP_SET] == NULL || @@ -81,7 +90,22 @@ static int nft_lookup_init(const struct nft_ctx *ctx, if (err < 0) return err; + if (tb[NFTA_LOOKUP_FLAGS]) { + flags = ntohl(nla_get_be32(tb[NFTA_LOOKUP_FLAGS])); + + if (flags & ~NFT_LOOKUP_F_INV) + return -EINVAL; + + if (flags & NFT_LOOKUP_F_INV) { + if (set->flags & NFT_SET_MAP) + return -EINVAL; + priv->invert = true; + } + } + if (tb[NFTA_LOOKUP_DREG] != NULL) { + if (priv->invert) + return -EINVAL; if (!(set->flags & NFT_SET_MAP)) return -EINVAL; @@ -114,6 +138,7 @@ static void nft_lookup_destroy(const struct nft_ctx *ctx, static int nft_lookup_dump(struct sk_buff *skb, const struct nft_expr *expr) { const struct nft_lookup *priv = nft_expr_priv(expr); + u32 flags = priv->invert ? NFT_LOOKUP_F_INV : 0; if (nla_put_string(skb, NFTA_LOOKUP_SET, priv->set->name)) goto nla_put_failure; @@ -122,6 +147,8 @@ static int nft_lookup_dump(struct sk_buff *skb, const struct nft_expr *expr) if (priv->set->flags & NFT_SET_MAP) if (nft_dump_register(skb, NFTA_LOOKUP_DREG, priv->dreg)) goto nla_put_failure; + if (nla_put_be32(skb, NFTA_LOOKUP_FLAGS, htonl(flags))) + goto nla_put_failure; return 0; nla_put_failure: -- cgit v1.2.3 From 6f99612e250041a2402d3b1694bccb149cd424a4 Mon Sep 17 00:00:00 2001 From: Stefan Berger Date: Mon, 18 Apr 2016 13:26:15 -0400 Subject: tpm: Proxy driver for supporting multiple emulated TPMs This patch implements a proxy driver for supporting multiple emulated TPMs in a system. The driver implements a device /dev/vtpmx that is used to created a client device pair /dev/tpmX (e.g., /dev/tpm10) and a server side that is accessed using a file descriptor returned by an ioctl. The device /dev/tpmX is the usual TPM device created by the core TPM driver. Applications or kernel subsystems can send TPM commands to it and the corresponding server-side file descriptor receives these commands and delivers them to an emulated TPM. The driver retrievs the TPM 1.2 durations and timeouts. Since this requires the startup of the TPM, we send a startup for TPM 1.2 as well as TPM 2. Signed-off-by: Stefan Berger Reviewed-by: Jason Gunthorpe CC: linux-kernel@vger.kernel.org CC: linux-doc@vger.kernel.org CC: linux-api@vger.kernel.org Reviewed-by: Jarkko Sakkinen Tested-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen --- drivers/char/tpm/Kconfig | 10 + drivers/char/tpm/Makefile | 1 + drivers/char/tpm/tpm_vtpm_proxy.c | 644 ++++++++++++++++++++++++++++++++++++++ include/uapi/linux/Kbuild | 1 + include/uapi/linux/vtpm_proxy.h | 36 +++ 5 files changed, 692 insertions(+) create mode 100644 drivers/char/tpm/tpm_vtpm_proxy.c create mode 100644 include/uapi/linux/vtpm_proxy.h (limited to 'include/uapi') diff --git a/drivers/char/tpm/Kconfig b/drivers/char/tpm/Kconfig index 3b84a8b1bfbe..0eac596e33d1 100644 --- a/drivers/char/tpm/Kconfig +++ b/drivers/char/tpm/Kconfig @@ -122,5 +122,15 @@ config TCG_CRB from within Linux. To compile this driver as a module, choose M here; the module will be called tpm_crb. +config TCG_VTPM_PROXY + tristate "VTPM Proxy Interface" + depends on TCG_TPM + ---help--- + This driver proxies for an emulated TPM (vTPM) running in userspace. + A device /dev/vtpmx is provided that creates a device pair + /dev/vtpmX and a server-side file descriptor on which the vTPM + can receive commands. + + source "drivers/char/tpm/st33zp24/Kconfig" endif # TCG_TPM diff --git a/drivers/char/tpm/Makefile b/drivers/char/tpm/Makefile index 56e8f1f3dc7e..98de5e676f3c 100644 --- a/drivers/char/tpm/Makefile +++ b/drivers/char/tpm/Makefile @@ -23,3 +23,4 @@ obj-$(CONFIG_TCG_IBMVTPM) += tpm_ibmvtpm.o obj-$(CONFIG_TCG_TIS_ST33ZP24) += st33zp24/ obj-$(CONFIG_TCG_XEN) += xen-tpmfront.o obj-$(CONFIG_TCG_CRB) += tpm_crb.o +obj-$(CONFIG_TCG_VTPM_PROXY) += tpm_vtpm_proxy.o diff --git a/drivers/char/tpm/tpm_vtpm_proxy.c b/drivers/char/tpm/tpm_vtpm_proxy.c new file mode 100644 index 000000000000..4384042b3b50 --- /dev/null +++ b/drivers/char/tpm/tpm_vtpm_proxy.c @@ -0,0 +1,644 @@ +/* + * Copyright (C) 2015, 2016 IBM Corporation + * + * Author: Stefan Berger + * + * Maintained by: + * + * Device driver for vTPM (vTPM proxy driver) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation, version 2 of the + * License. + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "tpm.h" + +#define VTPM_PROXY_REQ_COMPLETE_FLAG BIT(0) + +struct proxy_dev { + struct tpm_chip *chip; + + u32 flags; /* public API flags */ + + wait_queue_head_t wq; + + struct mutex buf_lock; /* protect buffer and flags */ + + long state; /* internal state */ +#define STATE_OPENED_FLAG BIT(0) +#define STATE_WAIT_RESPONSE_FLAG BIT(1) /* waiting for emulator response */ + + size_t req_len; /* length of queued TPM request */ + size_t resp_len; /* length of queued TPM response */ + u8 buffer[TPM_BUFSIZE]; /* request/response buffer */ + + struct work_struct work; /* task that retrieves TPM timeouts */ +}; + +/* all supported flags */ +#define VTPM_PROXY_FLAGS_ALL (VTPM_PROXY_FLAG_TPM2) + +static struct workqueue_struct *workqueue; + +static void vtpm_proxy_delete_device(struct proxy_dev *proxy_dev); + +/* + * Functions related to 'server side' + */ + +/** + * vtpm_proxy_fops_read - Read TPM commands on 'server side' + * + * Return value: + * Number of bytes read or negative error code + */ +static ssize_t vtpm_proxy_fops_read(struct file *filp, char __user *buf, + size_t count, loff_t *off) +{ + struct proxy_dev *proxy_dev = filp->private_data; + size_t len; + int sig, rc; + + sig = wait_event_interruptible(proxy_dev->wq, + proxy_dev->req_len != 0 || + !(proxy_dev->state & STATE_OPENED_FLAG)); + if (sig) + return -EINTR; + + mutex_lock(&proxy_dev->buf_lock); + + if (!(proxy_dev->state & STATE_OPENED_FLAG)) { + mutex_unlock(&proxy_dev->buf_lock); + return -EPIPE; + } + + len = proxy_dev->req_len; + + if (count < len) { + mutex_unlock(&proxy_dev->buf_lock); + pr_debug("Invalid size in recv: count=%zd, req_len=%zd\n", + count, len); + return -EIO; + } + + rc = copy_to_user(buf, proxy_dev->buffer, len); + memset(proxy_dev->buffer, 0, len); + proxy_dev->req_len = 0; + + if (!rc) + proxy_dev->state |= STATE_WAIT_RESPONSE_FLAG; + + mutex_unlock(&proxy_dev->buf_lock); + + if (rc) + return -EFAULT; + + return len; +} + +/** + * vtpm_proxy_fops_write - Write TPM responses on 'server side' + * + * Return value: + * Number of bytes read or negative error value + */ +static ssize_t vtpm_proxy_fops_write(struct file *filp, const char __user *buf, + size_t count, loff_t *off) +{ + struct proxy_dev *proxy_dev = filp->private_data; + + mutex_lock(&proxy_dev->buf_lock); + + if (!(proxy_dev->state & STATE_OPENED_FLAG)) { + mutex_unlock(&proxy_dev->buf_lock); + return -EPIPE; + } + + if (count > sizeof(proxy_dev->buffer) || + !(proxy_dev->state & STATE_WAIT_RESPONSE_FLAG)) { + mutex_unlock(&proxy_dev->buf_lock); + return -EIO; + } + + proxy_dev->state &= ~STATE_WAIT_RESPONSE_FLAG; + + proxy_dev->req_len = 0; + + if (copy_from_user(proxy_dev->buffer, buf, count)) { + mutex_unlock(&proxy_dev->buf_lock); + return -EFAULT; + } + + proxy_dev->resp_len = count; + + mutex_unlock(&proxy_dev->buf_lock); + + wake_up_interruptible(&proxy_dev->wq); + + return count; +} + +/* + * vtpm_proxy_fops_poll: Poll status on 'server side' + * + * Return value: + * Poll flags + */ +static unsigned int vtpm_proxy_fops_poll(struct file *filp, poll_table *wait) +{ + struct proxy_dev *proxy_dev = filp->private_data; + unsigned ret; + + poll_wait(filp, &proxy_dev->wq, wait); + + ret = POLLOUT; + + mutex_lock(&proxy_dev->buf_lock); + + if (proxy_dev->req_len) + ret |= POLLIN | POLLRDNORM; + + if (!(proxy_dev->state & STATE_OPENED_FLAG)) + ret |= POLLHUP; + + mutex_unlock(&proxy_dev->buf_lock); + + return ret; +} + +/* + * vtpm_proxy_fops_open - Open vTPM device on 'server side' + * + * Called when setting up the anonymous file descriptor + */ +static void vtpm_proxy_fops_open(struct file *filp) +{ + struct proxy_dev *proxy_dev = filp->private_data; + + proxy_dev->state |= STATE_OPENED_FLAG; +} + +/** + * vtpm_proxy_fops_undo_open - counter-part to vtpm_fops_open + * + * Call to undo vtpm_proxy_fops_open + */ +static void vtpm_proxy_fops_undo_open(struct proxy_dev *proxy_dev) +{ + mutex_lock(&proxy_dev->buf_lock); + + proxy_dev->state &= ~STATE_OPENED_FLAG; + + mutex_unlock(&proxy_dev->buf_lock); + + /* no more TPM responses -- wake up anyone waiting for them */ + wake_up_interruptible(&proxy_dev->wq); +} + +/* + * vtpm_proxy_fops_release: Close 'server side' + * + * Return value: + * Always returns 0. + */ +static int vtpm_proxy_fops_release(struct inode *inode, struct file *filp) +{ + struct proxy_dev *proxy_dev = filp->private_data; + + filp->private_data = NULL; + + vtpm_proxy_delete_device(proxy_dev); + + return 0; +} + +static const struct file_operations vtpm_proxy_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .read = vtpm_proxy_fops_read, + .write = vtpm_proxy_fops_write, + .poll = vtpm_proxy_fops_poll, + .release = vtpm_proxy_fops_release, +}; + +/* + * Functions invoked by the core TPM driver to send TPM commands to + * 'server side' and receive responses from there. + */ + +/* + * Called when core TPM driver reads TPM responses from 'server side' + * + * Return value: + * Number of TPM response bytes read, negative error value otherwise + */ +static int vtpm_proxy_tpm_op_recv(struct tpm_chip *chip, u8 *buf, size_t count) +{ + struct proxy_dev *proxy_dev = dev_get_drvdata(&chip->dev); + size_t len; + + /* process gone ? */ + mutex_lock(&proxy_dev->buf_lock); + + if (!(proxy_dev->state & STATE_OPENED_FLAG)) { + mutex_unlock(&proxy_dev->buf_lock); + return -EPIPE; + } + + len = proxy_dev->resp_len; + if (count < len) { + dev_err(&chip->dev, + "Invalid size in recv: count=%zd, resp_len=%zd\n", + count, len); + len = -EIO; + goto out; + } + + memcpy(buf, proxy_dev->buffer, len); + proxy_dev->resp_len = 0; + +out: + mutex_unlock(&proxy_dev->buf_lock); + + return len; +} + +/* + * Called when core TPM driver forwards TPM requests to 'server side'. + * + * Return value: + * 0 in case of success, negative error value otherwise. + */ +static int vtpm_proxy_tpm_op_send(struct tpm_chip *chip, u8 *buf, size_t count) +{ + struct proxy_dev *proxy_dev = dev_get_drvdata(&chip->dev); + int rc = 0; + + if (count > sizeof(proxy_dev->buffer)) { + dev_err(&chip->dev, + "Invalid size in send: count=%zd, buffer size=%zd\n", + count, sizeof(proxy_dev->buffer)); + return -EIO; + } + + mutex_lock(&proxy_dev->buf_lock); + + if (!(proxy_dev->state & STATE_OPENED_FLAG)) { + mutex_unlock(&proxy_dev->buf_lock); + return -EPIPE; + } + + proxy_dev->resp_len = 0; + + proxy_dev->req_len = count; + memcpy(proxy_dev->buffer, buf, count); + + proxy_dev->state &= ~STATE_WAIT_RESPONSE_FLAG; + + mutex_unlock(&proxy_dev->buf_lock); + + wake_up_interruptible(&proxy_dev->wq); + + return rc; +} + +static void vtpm_proxy_tpm_op_cancel(struct tpm_chip *chip) +{ + /* not supported */ +} + +static u8 vtpm_proxy_tpm_op_status(struct tpm_chip *chip) +{ + struct proxy_dev *proxy_dev = dev_get_drvdata(&chip->dev); + + if (proxy_dev->resp_len) + return VTPM_PROXY_REQ_COMPLETE_FLAG; + + return 0; +} + +static bool vtpm_proxy_tpm_req_canceled(struct tpm_chip *chip, u8 status) +{ + struct proxy_dev *proxy_dev = dev_get_drvdata(&chip->dev); + bool ret; + + mutex_lock(&proxy_dev->buf_lock); + + ret = !(proxy_dev->state & STATE_OPENED_FLAG); + + mutex_unlock(&proxy_dev->buf_lock); + + return ret; +} + +static const struct tpm_class_ops vtpm_proxy_tpm_ops = { + .recv = vtpm_proxy_tpm_op_recv, + .send = vtpm_proxy_tpm_op_send, + .cancel = vtpm_proxy_tpm_op_cancel, + .status = vtpm_proxy_tpm_op_status, + .req_complete_mask = VTPM_PROXY_REQ_COMPLETE_FLAG, + .req_complete_val = VTPM_PROXY_REQ_COMPLETE_FLAG, + .req_canceled = vtpm_proxy_tpm_req_canceled, +}; + +/* + * Code related to the startup of the TPM 2 and startup of TPM 1.2 + + * retrieval of timeouts and durations. + */ + +static void vtpm_proxy_work(struct work_struct *work) +{ + struct proxy_dev *proxy_dev = container_of(work, struct proxy_dev, + work); + int rc; + + if (proxy_dev->flags & VTPM_PROXY_FLAG_TPM2) + rc = tpm2_startup(proxy_dev->chip, TPM2_SU_CLEAR); + else + rc = tpm_get_timeouts(proxy_dev->chip); + + if (rc) + goto err; + + rc = tpm_chip_register(proxy_dev->chip); + if (rc) + goto err; + + return; + +err: + vtpm_proxy_fops_undo_open(proxy_dev); +} + +/* + * vtpm_proxy_work_stop: make sure the work has finished + * + * This function is useful when user space closed the fd + * while the driver still determines timeouts. + */ +static void vtpm_proxy_work_stop(struct proxy_dev *proxy_dev) +{ + vtpm_proxy_fops_undo_open(proxy_dev); + flush_work(&proxy_dev->work); +} + +/* + * vtpm_proxy_work_start: Schedule the work for TPM 1.2 & 2 initialization + */ +static inline void vtpm_proxy_work_start(struct proxy_dev *proxy_dev) +{ + queue_work(workqueue, &proxy_dev->work); +} + +/* + * Code related to creation and deletion of device pairs + */ +static struct proxy_dev *vtpm_proxy_create_proxy_dev(void) +{ + struct proxy_dev *proxy_dev; + struct tpm_chip *chip; + int err; + + proxy_dev = kzalloc(sizeof(*proxy_dev), GFP_KERNEL); + if (proxy_dev == NULL) + return ERR_PTR(-ENOMEM); + + init_waitqueue_head(&proxy_dev->wq); + mutex_init(&proxy_dev->buf_lock); + INIT_WORK(&proxy_dev->work, vtpm_proxy_work); + + chip = tpm_chip_alloc(NULL, &vtpm_proxy_tpm_ops); + if (IS_ERR(chip)) { + err = PTR_ERR(chip); + goto err_proxy_dev_free; + } + dev_set_drvdata(&chip->dev, proxy_dev); + + proxy_dev->chip = chip; + + return proxy_dev; + +err_proxy_dev_free: + kfree(proxy_dev); + + return ERR_PTR(err); +} + +/* + * Undo what has been done in vtpm_create_proxy_dev + */ +static inline void vtpm_proxy_delete_proxy_dev(struct proxy_dev *proxy_dev) +{ + put_device(&proxy_dev->chip->dev); /* frees chip */ + kfree(proxy_dev); +} + +/* + * Create a /dev/tpm%d and 'server side' file descriptor pair + * + * Return value: + * Returns file pointer on success, an error value otherwise + */ +static struct file *vtpm_proxy_create_device( + struct vtpm_proxy_new_dev *vtpm_new_dev) +{ + struct proxy_dev *proxy_dev; + int rc, fd; + struct file *file; + + if (vtpm_new_dev->flags & ~VTPM_PROXY_FLAGS_ALL) + return ERR_PTR(-EOPNOTSUPP); + + proxy_dev = vtpm_proxy_create_proxy_dev(); + if (IS_ERR(proxy_dev)) + return ERR_CAST(proxy_dev); + + proxy_dev->flags = vtpm_new_dev->flags; + + /* setup an anonymous file for the server-side */ + fd = get_unused_fd_flags(O_RDWR); + if (fd < 0) { + rc = fd; + goto err_delete_proxy_dev; + } + + file = anon_inode_getfile("[vtpms]", &vtpm_proxy_fops, proxy_dev, + O_RDWR); + if (IS_ERR(file)) { + rc = PTR_ERR(file); + goto err_put_unused_fd; + } + + /* from now on we can unwind with put_unused_fd() + fput() */ + /* simulate an open() on the server side */ + vtpm_proxy_fops_open(file); + + if (proxy_dev->flags & VTPM_PROXY_FLAG_TPM2) + proxy_dev->chip->flags |= TPM_CHIP_FLAG_TPM2; + + vtpm_proxy_work_start(proxy_dev); + + vtpm_new_dev->fd = fd; + vtpm_new_dev->major = MAJOR(proxy_dev->chip->dev.devt); + vtpm_new_dev->minor = MINOR(proxy_dev->chip->dev.devt); + vtpm_new_dev->tpm_num = proxy_dev->chip->dev_num; + + return file; + +err_put_unused_fd: + put_unused_fd(fd); + +err_delete_proxy_dev: + vtpm_proxy_delete_proxy_dev(proxy_dev); + + return ERR_PTR(rc); +} + +/* + * Counter part to vtpm_create_device. + */ +static void vtpm_proxy_delete_device(struct proxy_dev *proxy_dev) +{ + vtpm_proxy_work_stop(proxy_dev); + + /* + * A client may hold the 'ops' lock, so let it know that the server + * side shuts down before we try to grab the 'ops' lock when + * unregistering the chip. + */ + vtpm_proxy_fops_undo_open(proxy_dev); + + tpm_chip_unregister(proxy_dev->chip); + + vtpm_proxy_delete_proxy_dev(proxy_dev); +} + +/* + * Code related to the control device /dev/vtpmx + */ + +/* + * vtpmx_fops_ioctl: ioctl on /dev/vtpmx + * + * Return value: + * Returns 0 on success, a negative error code otherwise. + */ +static long vtpmx_fops_ioctl(struct file *f, unsigned int ioctl, + unsigned long arg) +{ + void __user *argp = (void __user *)arg; + struct vtpm_proxy_new_dev *vtpm_new_dev_p; + struct vtpm_proxy_new_dev vtpm_new_dev; + struct file *file; + + switch (ioctl) { + case VTPM_PROXY_IOC_NEW_DEV: + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + vtpm_new_dev_p = argp; + if (copy_from_user(&vtpm_new_dev, vtpm_new_dev_p, + sizeof(vtpm_new_dev))) + return -EFAULT; + file = vtpm_proxy_create_device(&vtpm_new_dev); + if (IS_ERR(file)) + return PTR_ERR(file); + if (copy_to_user(vtpm_new_dev_p, &vtpm_new_dev, + sizeof(vtpm_new_dev))) { + put_unused_fd(vtpm_new_dev.fd); + fput(file); + return -EFAULT; + } + + fd_install(vtpm_new_dev.fd, file); + return 0; + + default: + return -ENOIOCTLCMD; + } +} + +#ifdef CONFIG_COMPAT +static long vtpmx_fops_compat_ioctl(struct file *f, unsigned int ioctl, + unsigned long arg) +{ + return vtpmx_fops_ioctl(f, ioctl, (unsigned long)compat_ptr(arg)); +} +#endif + +static const struct file_operations vtpmx_fops = { + .owner = THIS_MODULE, + .unlocked_ioctl = vtpmx_fops_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = vtpmx_fops_compat_ioctl, +#endif + .llseek = noop_llseek, +}; + +static struct miscdevice vtpmx_miscdev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "vtpmx", + .fops = &vtpmx_fops, +}; + +static int vtpmx_init(void) +{ + return misc_register(&vtpmx_miscdev); +} + +static void vtpmx_cleanup(void) +{ + misc_deregister(&vtpmx_miscdev); +} + +static int __init vtpm_module_init(void) +{ + int rc; + + rc = vtpmx_init(); + if (rc) { + pr_err("couldn't create vtpmx device\n"); + return rc; + } + + workqueue = create_workqueue("tpm-vtpm"); + if (!workqueue) { + pr_err("couldn't create workqueue\n"); + rc = -ENOMEM; + goto err_vtpmx_cleanup; + } + + return 0; + +err_vtpmx_cleanup: + vtpmx_cleanup(); + + return rc; +} + +static void __exit vtpm_module_exit(void) +{ + destroy_workqueue(workqueue); + vtpmx_cleanup(); +} + +module_init(vtpm_module_init); +module_exit(vtpm_module_exit); + +MODULE_AUTHOR("Stefan Berger (stefanb@us.ibm.com)"); +MODULE_DESCRIPTION("vTPM Driver"); +MODULE_VERSION("0.1"); +MODULE_LICENSE("GPL"); diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild index 8bdae34d1f9a..ad821a3a99f5 100644 --- a/include/uapi/linux/Kbuild +++ b/include/uapi/linux/Kbuild @@ -454,6 +454,7 @@ header-y += virtio_scsi.h header-y += virtio_types.h header-y += vm_sockets.h header-y += vt.h +header-y += vtpm_proxy.h header-y += wait.h header-y += wanrouter.h header-y += watchdog.h diff --git a/include/uapi/linux/vtpm_proxy.h b/include/uapi/linux/vtpm_proxy.h new file mode 100644 index 000000000000..41e8e2252a30 --- /dev/null +++ b/include/uapi/linux/vtpm_proxy.h @@ -0,0 +1,36 @@ +/* + * Definitions for the VTPM proxy driver + * Copyright (c) 2015, 2016, IBM Corporation + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef _UAPI_LINUX_VTPM_PROXY_H +#define _UAPI_LINUX_VTPM_PROXY_H + +#include +#include + +/* ioctls */ + +struct vtpm_proxy_new_dev { + __u32 flags; /* input */ + __u32 tpm_num; /* output */ + __u32 fd; /* output */ + __u32 major; /* output */ + __u32 minor; /* output */ +}; + +/* above flags */ +#define VTPM_PROXY_FLAG_TPM2 1 /* emulator is TPM 2 */ + +#define VTPM_PROXY_IOC_NEW_DEV _IOWR(0xa1, 0x00, struct vtpm_proxy_new_dev) + +#endif /* _UAPI_LINUX_VTPM_PROXY_H */ -- cgit v1.2.3 From 89da45b8b5b2187734a11038b8593714f964ffd1 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:43 +0300 Subject: ethtool: Add 50G baseSR2 link mode Add ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT bit. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Cc: Ben Hutchings Cc: David Decotigny Acked-By: David Decotigny Signed-off-by: David S. Miller --- include/uapi/linux/ethtool.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index 5f030b46cff4..b8f38e84d93a 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -1362,6 +1362,7 @@ enum ethtool_link_mode_bit_indices { ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT = 37, ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT = 38, ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT = 39, + ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT = 40, /* Last allowed bit for __ETHTOOL_LINK_MODE_LEGACY_MASK is bit * 31. Please do NOT define any SUPPORTED_* or ADVERTISED_* @@ -1370,7 +1371,7 @@ enum ethtool_link_mode_bit_indices { */ __ETHTOOL_LINK_MODE_LAST - = ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT, + = ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT, }; #define __ETHTOOL_LINK_MODE_LEGACY_MASK(base_name) \ -- cgit v1.2.3 From d28f6df1305a86715e4e7ea0f043ba01c0a0e8d9 Mon Sep 17 00:00:00 2001 From: Geoff Levand Date: Thu, 23 Jun 2016 17:54:48 +0000 Subject: arm64/kexec: Add core kexec support Add three new files, kexec.h, machine_kexec.c and relocate_kernel.S to the arm64 architecture that add support for the kexec re-boot mechanism (CONFIG_KEXEC) on arm64 platforms. Signed-off-by: Geoff Levand Reviewed-by: James Morse [catalin.marinas@arm.com: removed dead code following James Morse's comments] Signed-off-by: Catalin Marinas --- arch/arm64/Kconfig | 10 +++ arch/arm64/include/asm/kexec.h | 48 ++++++++++ arch/arm64/kernel/Makefile | 2 + arch/arm64/kernel/machine_kexec.c | 170 ++++++++++++++++++++++++++++++++++++ arch/arm64/kernel/relocate_kernel.S | 130 +++++++++++++++++++++++++++ include/uapi/linux/kexec.h | 1 + 6 files changed, 361 insertions(+) create mode 100644 arch/arm64/include/asm/kexec.h create mode 100644 arch/arm64/kernel/machine_kexec.c create mode 100644 arch/arm64/kernel/relocate_kernel.S (limited to 'include/uapi') diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index eb0b0a05751e..1b196bf99320 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -665,6 +665,16 @@ config PARAVIRT_TIME_ACCOUNTING If in doubt, say N here. +config KEXEC + depends on PM_SLEEP_SMP + select KEXEC_CORE + bool "kexec system call" + ---help--- + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but it is independent of the system firmware. And like a reboot + you can start any kernel with it, not just Linux. + config XEN_DOM0 def_bool y depends on XEN diff --git a/arch/arm64/include/asm/kexec.h b/arch/arm64/include/asm/kexec.h new file mode 100644 index 000000000000..04744dc5fb61 --- /dev/null +++ b/arch/arm64/include/asm/kexec.h @@ -0,0 +1,48 @@ +/* + * kexec for arm64 + * + * Copyright (C) Linaro. + * Copyright (C) Huawei Futurewei Technologies. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef _ARM64_KEXEC_H +#define _ARM64_KEXEC_H + +/* Maximum physical address we can use pages from */ + +#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL) + +/* Maximum address we can reach in physical address mode */ + +#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL) + +/* Maximum address we can use for the control code buffer */ + +#define KEXEC_CONTROL_MEMORY_LIMIT (-1UL) + +#define KEXEC_CONTROL_PAGE_SIZE 4096 + +#define KEXEC_ARCH KEXEC_ARCH_AARCH64 + +#ifndef __ASSEMBLY__ + +/** + * crash_setup_regs() - save registers for the panic kernel + * + * @newregs: registers are saved here + * @oldregs: registers to be saved (may be %NULL) + */ + +static inline void crash_setup_regs(struct pt_regs *newregs, + struct pt_regs *oldregs) +{ + /* Empty routine needed to avoid build errors. */ +} + +#endif /* __ASSEMBLY__ */ + +#endif diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 2173149d8954..7700c0c23962 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -46,6 +46,8 @@ arm64-obj-$(CONFIG_ARM64_ACPI_PARKING_PROTOCOL) += acpi_parking_protocol.o arm64-obj-$(CONFIG_PARAVIRT) += paravirt.o arm64-obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o arm64-obj-$(CONFIG_HIBERNATION) += hibernate.o hibernate-asm.o +arm64-obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o \ + cpu-reset.o obj-y += $(arm64-obj-y) vdso/ obj-m += $(arm64-obj-m) diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c new file mode 100644 index 000000000000..c40e64607545 --- /dev/null +++ b/arch/arm64/kernel/machine_kexec.c @@ -0,0 +1,170 @@ +/* + * kexec for arm64 + * + * Copyright (C) Linaro. + * Copyright (C) Huawei Futurewei Technologies. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include + +#include +#include +#include + +#include "cpu-reset.h" + +/* Global variables for the arm64_relocate_new_kernel routine. */ +extern const unsigned char arm64_relocate_new_kernel[]; +extern const unsigned long arm64_relocate_new_kernel_size; + +static unsigned long kimage_start; + +void machine_kexec_cleanup(struct kimage *kimage) +{ + /* Empty routine needed to avoid build errors. */ +} + +/** + * machine_kexec_prepare - Prepare for a kexec reboot. + * + * Called from the core kexec code when a kernel image is loaded. + * Forbid loading a kexec kernel if we have no way of hotplugging cpus or cpus + * are stuck in the kernel. This avoids a panic once we hit machine_kexec(). + */ +int machine_kexec_prepare(struct kimage *kimage) +{ + kimage_start = kimage->start; + + if (kimage->type != KEXEC_TYPE_CRASH && cpus_are_stuck_in_kernel()) { + pr_err("Can't kexec: CPUs are stuck in the kernel.\n"); + return -EBUSY; + } + + return 0; +} + +/** + * kexec_list_flush - Helper to flush the kimage list and source pages to PoC. + */ +static void kexec_list_flush(struct kimage *kimage) +{ + kimage_entry_t *entry; + + for (entry = &kimage->head; ; entry++) { + unsigned int flag; + void *addr; + + /* flush the list entries. */ + __flush_dcache_area(entry, sizeof(kimage_entry_t)); + + flag = *entry & IND_FLAGS; + if (flag == IND_DONE) + break; + + addr = phys_to_virt(*entry & PAGE_MASK); + + switch (flag) { + case IND_INDIRECTION: + /* Set entry point just before the new list page. */ + entry = (kimage_entry_t *)addr - 1; + break; + case IND_SOURCE: + /* flush the source pages. */ + __flush_dcache_area(addr, PAGE_SIZE); + break; + case IND_DESTINATION: + break; + default: + BUG(); + } + } +} + +/** + * kexec_segment_flush - Helper to flush the kimage segments to PoC. + */ +static void kexec_segment_flush(const struct kimage *kimage) +{ + unsigned long i; + + pr_debug("%s:\n", __func__); + + for (i = 0; i < kimage->nr_segments; i++) { + pr_debug(" segment[%lu]: %016lx - %016lx, 0x%lx bytes, %lu pages\n", + i, + kimage->segment[i].mem, + kimage->segment[i].mem + kimage->segment[i].memsz, + kimage->segment[i].memsz, + kimage->segment[i].memsz / PAGE_SIZE); + + __flush_dcache_area(phys_to_virt(kimage->segment[i].mem), + kimage->segment[i].memsz); + } +} + +/** + * machine_kexec - Do the kexec reboot. + * + * Called from the core kexec code for a sys_reboot with LINUX_REBOOT_CMD_KEXEC. + */ +void machine_kexec(struct kimage *kimage) +{ + phys_addr_t reboot_code_buffer_phys; + void *reboot_code_buffer; + + /* + * New cpus may have become stuck_in_kernel after we loaded the image. + */ + BUG_ON(cpus_are_stuck_in_kernel() || (num_online_cpus() > 1)); + + reboot_code_buffer_phys = page_to_phys(kimage->control_code_page); + reboot_code_buffer = phys_to_virt(reboot_code_buffer_phys); + + /* + * Copy arm64_relocate_new_kernel to the reboot_code_buffer for use + * after the kernel is shut down. + */ + memcpy(reboot_code_buffer, arm64_relocate_new_kernel, + arm64_relocate_new_kernel_size); + + /* Flush the reboot_code_buffer in preparation for its execution. */ + __flush_dcache_area(reboot_code_buffer, arm64_relocate_new_kernel_size); + flush_icache_range((uintptr_t)reboot_code_buffer, + arm64_relocate_new_kernel_size); + + /* Flush the kimage list and its buffers. */ + kexec_list_flush(kimage); + + /* Flush the new image if already in place. */ + if (kimage->head & IND_DONE) + kexec_segment_flush(kimage); + + pr_info("Bye!\n"); + + /* Disable all DAIF exceptions. */ + asm volatile ("msr daifset, #0xf" : : : "memory"); + + /* + * cpu_soft_restart will shutdown the MMU, disable data caches, then + * transfer control to the reboot_code_buffer which contains a copy of + * the arm64_relocate_new_kernel routine. arm64_relocate_new_kernel + * uses physical addressing to relocate the new image to its final + * position and transfers control to the image entry point when the + * relocation is complete. + */ + + cpu_soft_restart(1, reboot_code_buffer_phys, kimage->head, + kimage_start, 0); + + BUG(); /* Should never get here. */ +} + +void machine_crash_shutdown(struct pt_regs *regs) +{ + /* Empty routine needed to avoid build errors. */ +} diff --git a/arch/arm64/kernel/relocate_kernel.S b/arch/arm64/kernel/relocate_kernel.S new file mode 100644 index 000000000000..51b73cdde287 --- /dev/null +++ b/arch/arm64/kernel/relocate_kernel.S @@ -0,0 +1,130 @@ +/* + * kexec for arm64 + * + * Copyright (C) Linaro. + * Copyright (C) Huawei Futurewei Technologies. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include + +#include +#include +#include +#include + +/* + * arm64_relocate_new_kernel - Put a 2nd stage image in place and boot it. + * + * The memory that the old kernel occupies may be overwritten when coping the + * new image to its final location. To assure that the + * arm64_relocate_new_kernel routine which does that copy is not overwritten, + * all code and data needed by arm64_relocate_new_kernel must be between the + * symbols arm64_relocate_new_kernel and arm64_relocate_new_kernel_end. The + * machine_kexec() routine will copy arm64_relocate_new_kernel to the kexec + * control_code_page, a special page which has been set up to be preserved + * during the copy operation. + */ +ENTRY(arm64_relocate_new_kernel) + + /* Setup the list loop variables. */ + mov x17, x1 /* x17 = kimage_start */ + mov x16, x0 /* x16 = kimage_head */ + dcache_line_size x15, x0 /* x15 = dcache line size */ + mov x14, xzr /* x14 = entry ptr */ + mov x13, xzr /* x13 = copy dest */ + + /* Clear the sctlr_el2 flags. */ + mrs x0, CurrentEL + cmp x0, #CurrentEL_EL2 + b.ne 1f + mrs x0, sctlr_el2 + ldr x1, =SCTLR_ELx_FLAGS + bic x0, x0, x1 + msr sctlr_el2, x0 + isb +1: + + /* Check if the new image needs relocation. */ + tbnz x16, IND_DONE_BIT, .Ldone + +.Lloop: + and x12, x16, PAGE_MASK /* x12 = addr */ + + /* Test the entry flags. */ +.Ltest_source: + tbz x16, IND_SOURCE_BIT, .Ltest_indirection + + /* Invalidate dest page to PoC. */ + mov x0, x13 + add x20, x0, #PAGE_SIZE + sub x1, x15, #1 + bic x0, x0, x1 +2: dc ivac, x0 + add x0, x0, x15 + cmp x0, x20 + b.lo 2b + dsb sy + + mov x20, x13 + mov x21, x12 + copy_page x20, x21, x0, x1, x2, x3, x4, x5, x6, x7 + + /* dest += PAGE_SIZE */ + add x13, x13, PAGE_SIZE + b .Lnext + +.Ltest_indirection: + tbz x16, IND_INDIRECTION_BIT, .Ltest_destination + + /* ptr = addr */ + mov x14, x12 + b .Lnext + +.Ltest_destination: + tbz x16, IND_DESTINATION_BIT, .Lnext + + /* dest = addr */ + mov x13, x12 + +.Lnext: + /* entry = *ptr++ */ + ldr x16, [x14], #8 + + /* while (!(entry & DONE)) */ + tbz x16, IND_DONE_BIT, .Lloop + +.Ldone: + /* wait for writes from copy_page to finish */ + dsb nsh + ic iallu + dsb nsh + isb + + /* Start new image. */ + mov x0, xzr + mov x1, xzr + mov x2, xzr + mov x3, xzr + br x17 + +ENDPROC(arm64_relocate_new_kernel) + +.ltorg + +.align 3 /* To keep the 64-bit values below naturally aligned. */ + +.Lcopy_end: +.org KEXEC_CONTROL_PAGE_SIZE + +/* + * arm64_relocate_new_kernel_size - Number of bytes to copy to the + * control_code_page. + */ +.globl arm64_relocate_new_kernel_size +arm64_relocate_new_kernel_size: + .quad .Lcopy_end - arm64_relocate_new_kernel diff --git a/include/uapi/linux/kexec.h b/include/uapi/linux/kexec.h index 99048e501b88..aae5ebf2022b 100644 --- a/include/uapi/linux/kexec.h +++ b/include/uapi/linux/kexec.h @@ -39,6 +39,7 @@ #define KEXEC_ARCH_SH (42 << 16) #define KEXEC_ARCH_MIPS_LE (10 << 16) #define KEXEC_ARCH_MIPS ( 8 << 16) +#define KEXEC_ARCH_AARCH64 (183 << 16) /* The artificial cap on the number of segments passed to kexec_load. */ #define KEXEC_SEGMENT_MAX 16 -- cgit v1.2.3 From cb72d38211eacda2dd90b09540542b6582da614e Mon Sep 17 00:00:00 2001 From: Huw Davies Date: Mon, 27 Jun 2016 15:02:46 -0400 Subject: netlabel: Initial support for the CALIPSO netlink protocol. CALIPSO is a packet labelling protocol for IPv6 which is very similar to CIPSO. It is specified in RFC 5570. Much of the code is based on the current CIPSO code. This adds support for adding passthrough-type CALIPSO DOIs through the NLBL_CALIPSO_C_ADD command. It requires attributes: NLBL_CALIPSO_A_TYPE which must be CALIPSO_MAP_PASS. NLBL_CALIPSO_A_DOI. In passthrough mode the CALIPSO engine will map MLS secattr levels and categories directly to the packet label. At this stage, the major difference between this and the CIPSO code is that IPv6 may be compiled as a module. To allow for this the CALIPSO functions are registered at module init time. Signed-off-by: Huw Davies Signed-off-by: Paul Moore --- include/net/calipso.h | 79 +++++++++++++++ include/net/netlabel.h | 23 +++++ include/uapi/linux/audit.h | 2 + net/ipv6/Makefile | 1 + net/ipv6/af_inet6.c | 9 +- net/ipv6/calipso.c | 169 +++++++++++++++++++++++++++++++ net/netlabel/Makefile | 2 +- net/netlabel/netlabel_calipso.c | 216 ++++++++++++++++++++++++++++++++++++++++ net/netlabel/netlabel_calipso.h | 90 +++++++++++++++++ net/netlabel/netlabel_kapi.c | 1 + net/netlabel/netlabel_mgmt.c | 9 ++ net/netlabel/netlabel_user.c | 5 + 12 files changed, 604 insertions(+), 2 deletions(-) create mode 100644 include/net/calipso.h create mode 100644 net/ipv6/calipso.c create mode 100644 net/netlabel/netlabel_calipso.c create mode 100644 net/netlabel/netlabel_calipso.h (limited to 'include/uapi') diff --git a/include/net/calipso.h b/include/net/calipso.h new file mode 100644 index 000000000000..38dbb4707150 --- /dev/null +++ b/include/net/calipso.h @@ -0,0 +1,79 @@ +/* + * CALIPSO - Common Architecture Label IPv6 Security Option + * + * This is an implementation of the CALIPSO protocol as specified in + * RFC 5570. + * + * Authors: Paul Moore + * Huw Davies + * + */ + +/* + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Huw Davies , 2015 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + * + */ + +#ifndef _CALIPSO_H +#define _CALIPSO_H + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* known doi values */ +#define CALIPSO_DOI_UNKNOWN 0x00000000 + +/* doi mapping types */ +#define CALIPSO_MAP_UNKNOWN 0 +#define CALIPSO_MAP_PASS 2 + +/* + * CALIPSO DOI definitions + */ + +/* DOI definition struct */ +struct calipso_doi { + u32 doi; + u32 type; + + atomic_t refcount; + struct list_head list; + struct rcu_head rcu; +}; + +#ifdef CONFIG_NETLABEL +int __init calipso_init(void); +void calipso_exit(void); +#else +static inline int __init calipso_init(void) +{ + return 0; +} + +static inline void calipso_exit(void) +{ +} +#endif /* CONFIG_NETLABEL */ + +#endif /* _CALIPSO_H */ diff --git a/include/net/netlabel.h b/include/net/netlabel.h index 7b5a300de7f5..6af1bb6df4ab 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -40,6 +40,7 @@ #include struct cipso_v4_doi; +struct calipso_doi; /* * NetLabel - A management interface for maintaining network packet label @@ -94,6 +95,8 @@ struct cipso_v4_doi; #define NETLBL_NLTYPE_UNLABELED_NAME "NLBL_UNLBL" #define NETLBL_NLTYPE_ADDRSELECT 6 #define NETLBL_NLTYPE_ADDRSELECT_NAME "NLBL_ADRSEL" +#define NETLBL_NLTYPE_CALIPSO 7 +#define NETLBL_NLTYPE_CALIPSO_NAME "NLBL_CALIPSO" /* * NetLabel - Kernel API for accessing the network packet label mappings. @@ -216,6 +219,23 @@ struct netlbl_lsm_secattr { } attr; }; +/** + * struct netlbl_calipso_ops - NetLabel CALIPSO operations + * @doi_add: add a CALIPSO DOI + * @doi_free: free a CALIPSO DOI + * + * Description: + * This structure is filled out by the CALIPSO engine and passed + * to the NetLabel core via a call to netlbl_calipso_ops_register(). + * It enables the CALIPSO engine (and hence IPv6) to be compiled + * as a module. + */ +struct netlbl_calipso_ops { + int (*doi_add)(struct calipso_doi *doi_def, + struct netlbl_audit *audit_info); + void (*doi_free)(struct calipso_doi *doi_def); +}; + /* * LSM security attribute operations (inline) */ @@ -598,4 +618,7 @@ static inline struct audit_buffer *netlbl_audit_start(int type, } #endif /* CONFIG_NETLABEL */ +const struct netlbl_calipso_ops * +netlbl_calipso_ops_register(const struct netlbl_calipso_ops *ops); + #endif /* _NETLABEL_H */ diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h index d820aa979620..82e8aa59446b 100644 --- a/include/uapi/linux/audit.h +++ b/include/uapi/linux/audit.h @@ -130,6 +130,8 @@ #define AUDIT_MAC_IPSEC_EVENT 1415 /* Audit an IPSec event */ #define AUDIT_MAC_UNLBL_STCADD 1416 /* NetLabel: add a static label */ #define AUDIT_MAC_UNLBL_STCDEL 1417 /* NetLabel: del a static label */ +#define AUDIT_MAC_CALIPSO_ADD 1418 /* NetLabel: add CALIPSO DOI entry */ +#define AUDIT_MAC_CALIPSO_DEL 1419 /* NetLabel: del CALIPSO DOI entry */ #define AUDIT_FIRST_KERN_ANOM_MSG 1700 #define AUDIT_LAST_KERN_ANOM_MSG 1799 diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 2fbd90bf8d33..88ad4477705c 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -21,6 +21,7 @@ ipv6-$(CONFIG_NETFILTER) += netfilter.o ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o ipv6-$(CONFIG_PROC_FS) += proc.o ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o +ipv6-$(CONFIG_NETLABEL) += calipso.o ipv6-objs += $(ipv6-y) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index b11c37cfd67c..c241c1805728 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -60,6 +60,7 @@ #ifdef CONFIG_IPV6_TUNNEL #include #endif +#include #include #include @@ -970,6 +971,10 @@ static int __init inet6_init(void) if (err) goto pingv6_fail; + err = calipso_init(); + if (err) + goto calipso_fail; + #ifdef CONFIG_SYSCTL err = ipv6_sysctl_register(); if (err) @@ -980,8 +985,10 @@ out: #ifdef CONFIG_SYSCTL sysctl_fail: - pingv6_exit(); + calipso_exit(); #endif +calipso_fail: + pingv6_exit(); pingv6_fail: ipv6_packet_cleanup(); ipv6_packet_fail: diff --git a/net/ipv6/calipso.c b/net/ipv6/calipso.c new file mode 100644 index 000000000000..aa44310120be --- /dev/null +++ b/net/ipv6/calipso.c @@ -0,0 +1,169 @@ +/* + * CALIPSO - Common Architecture Label IPv6 Security Option + * + * This is an implementation of the CALIPSO protocol as specified in + * RFC 5570. + * + * Authors: Paul Moore + * Huw Davies + * + */ + +/* (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008 + * (c) Copyright Huw Davies , 2015 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* List of available DOI definitions */ +static DEFINE_SPINLOCK(calipso_doi_list_lock); +static LIST_HEAD(calipso_doi_list); + +/* DOI List Functions + */ + +/** + * calipso_doi_search - Searches for a DOI definition + * @doi: the DOI to search for + * + * Description: + * Search the DOI definition list for a DOI definition with a DOI value that + * matches @doi. The caller is responsible for calling rcu_read_[un]lock(). + * Returns a pointer to the DOI definition on success and NULL on failure. + */ +static struct calipso_doi *calipso_doi_search(u32 doi) +{ + struct calipso_doi *iter; + + list_for_each_entry_rcu(iter, &calipso_doi_list, list) + if (iter->doi == doi && atomic_read(&iter->refcount)) + return iter; + return NULL; +} + +/** + * calipso_doi_add - Add a new DOI to the CALIPSO protocol engine + * @doi_def: the DOI structure + * @audit_info: NetLabel audit information + * + * Description: + * The caller defines a new DOI for use by the CALIPSO engine and calls this + * function to add it to the list of acceptable domains. The caller must + * ensure that the mapping table specified in @doi_def->map meets all of the + * requirements of the mapping type (see calipso.h for details). Returns + * zero on success and non-zero on failure. + * + */ +static int calipso_doi_add(struct calipso_doi *doi_def, + struct netlbl_audit *audit_info) +{ + int ret_val = -EINVAL; + u32 doi; + u32 doi_type; + struct audit_buffer *audit_buf; + + doi = doi_def->doi; + doi_type = doi_def->type; + + if (doi_def->doi == CALIPSO_DOI_UNKNOWN) + goto doi_add_return; + + atomic_set(&doi_def->refcount, 1); + + spin_lock(&calipso_doi_list_lock); + if (calipso_doi_search(doi_def->doi)) { + spin_unlock(&calipso_doi_list_lock); + ret_val = -EEXIST; + goto doi_add_return; + } + list_add_tail_rcu(&doi_def->list, &calipso_doi_list); + spin_unlock(&calipso_doi_list_lock); + ret_val = 0; + +doi_add_return: + audit_buf = netlbl_audit_start(AUDIT_MAC_CALIPSO_ADD, audit_info); + if (audit_buf) { + const char *type_str; + + switch (doi_type) { + case CALIPSO_MAP_PASS: + type_str = "pass"; + break; + default: + type_str = "(unknown)"; + } + audit_log_format(audit_buf, + " calipso_doi=%u calipso_type=%s res=%u", + doi, type_str, ret_val == 0 ? 1 : 0); + audit_log_end(audit_buf); + } + + return ret_val; +} + +/** + * calipso_doi_free - Frees a DOI definition + * @doi_def: the DOI definition + * + * Description: + * This function frees all of the memory associated with a DOI definition. + * + */ +static void calipso_doi_free(struct calipso_doi *doi_def) +{ + kfree(doi_def); +} + +static const struct netlbl_calipso_ops ops = { + .doi_add = calipso_doi_add, + .doi_free = calipso_doi_free, +}; + +/** + * calipso_init - Initialize the CALIPSO module + * + * Description: + * Initialize the CALIPSO module and prepare it for use. Returns zero on + * success and negative values on failure. + * + */ +int __init calipso_init(void) +{ + netlbl_calipso_ops_register(&ops); + return 0; +} + +void calipso_exit(void) +{ + netlbl_calipso_ops_register(NULL); +} diff --git a/net/netlabel/Makefile b/net/netlabel/Makefile index d2732fc952e2..d341ede0dca5 100644 --- a/net/netlabel/Makefile +++ b/net/netlabel/Makefile @@ -12,4 +12,4 @@ obj-y += netlabel_mgmt.o # protocol modules obj-y += netlabel_unlabeled.o obj-y += netlabel_cipso_v4.o - +obj-$(subst m,y,$(CONFIG_IPV6)) += netlabel_calipso.o diff --git a/net/netlabel/netlabel_calipso.c b/net/netlabel/netlabel_calipso.c new file mode 100644 index 000000000000..8a113f91b7a1 --- /dev/null +++ b/net/netlabel/netlabel_calipso.c @@ -0,0 +1,216 @@ +/* + * NetLabel CALIPSO/IPv6 Support + * + * This file defines the CALIPSO/IPv6 functions for the NetLabel system. The + * NetLabel system manages static and dynamic label mappings for network + * protocols such as CIPSO and CALIPSO. + * + * Authors: Paul Moore + * Huw Davies + * + */ + +/* (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Huw Davies , 2015 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "netlabel_user.h" +#include "netlabel_calipso.h" +#include "netlabel_mgmt.h" +#include "netlabel_domainhash.h" + +/* NetLabel Generic NETLINK CALIPSO family */ +static struct genl_family netlbl_calipso_gnl_family = { + .id = GENL_ID_GENERATE, + .hdrsize = 0, + .name = NETLBL_NLTYPE_CALIPSO_NAME, + .version = NETLBL_PROTO_VERSION, + .maxattr = NLBL_CALIPSO_A_MAX, +}; + +/* NetLabel Netlink attribute policy */ +static const struct nla_policy calipso_genl_policy[NLBL_CALIPSO_A_MAX + 1] = { + [NLBL_CALIPSO_A_DOI] = { .type = NLA_U32 }, + [NLBL_CALIPSO_A_MTYPE] = { .type = NLA_U32 }, +}; + +/* NetLabel Command Handlers + */ +/** + * netlbl_calipso_add_pass - Adds a CALIPSO pass DOI definition + * @info: the Generic NETLINK info block + * @audit_info: NetLabel audit information + * + * Description: + * Create a new CALIPSO_MAP_PASS DOI definition based on the given ADD message + * and add it to the CALIPSO engine. Return zero on success and non-zero on + * error. + * + */ +static int netlbl_calipso_add_pass(struct genl_info *info, + struct netlbl_audit *audit_info) +{ + int ret_val; + struct calipso_doi *doi_def = NULL; + + doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL); + if (!doi_def) + return -ENOMEM; + doi_def->type = CALIPSO_MAP_PASS; + doi_def->doi = nla_get_u32(info->attrs[NLBL_CALIPSO_A_DOI]); + ret_val = calipso_doi_add(doi_def, audit_info); + if (ret_val != 0) + calipso_doi_free(doi_def); + + return ret_val; +} + +/** + * netlbl_calipso_add - Handle an ADD message + * @skb: the NETLINK buffer + * @info: the Generic NETLINK info block + * + * Description: + * Create a new DOI definition based on the given ADD message and add it to the + * CALIPSO engine. Returns zero on success, negative values on failure. + * + */ +static int netlbl_calipso_add(struct sk_buff *skb, struct genl_info *info) + +{ + int ret_val = -EINVAL; + struct netlbl_audit audit_info; + + if (!info->attrs[NLBL_CALIPSO_A_DOI] || + !info->attrs[NLBL_CALIPSO_A_MTYPE]) + return -EINVAL; + + netlbl_netlink_auditinfo(skb, &audit_info); + switch (nla_get_u32(info->attrs[NLBL_CALIPSO_A_MTYPE])) { + case CALIPSO_MAP_PASS: + ret_val = netlbl_calipso_add_pass(info, &audit_info); + break; + } + if (ret_val == 0) + atomic_inc(&netlabel_mgmt_protocount); + + return ret_val; +} + +/* NetLabel Generic NETLINK Command Definitions + */ + +static const struct genl_ops netlbl_calipso_ops[] = { + { + .cmd = NLBL_CALIPSO_C_ADD, + .flags = GENL_ADMIN_PERM, + .policy = calipso_genl_policy, + .doit = netlbl_calipso_add, + .dumpit = NULL, + }, +}; + +/* NetLabel Generic NETLINK Protocol Functions + */ + +/** + * netlbl_calipso_genl_init - Register the CALIPSO NetLabel component + * + * Description: + * Register the CALIPSO packet NetLabel component with the Generic NETLINK + * mechanism. Returns zero on success, negative values on failure. + * + */ +int __init netlbl_calipso_genl_init(void) +{ + return genl_register_family_with_ops(&netlbl_calipso_gnl_family, + netlbl_calipso_ops); +} + +static const struct netlbl_calipso_ops *calipso_ops; + +/** + * netlbl_calipso_ops_register - Register the CALIPSO operations + * + * Description: + * Register the CALIPSO packet engine operations. + * + */ +const struct netlbl_calipso_ops * +netlbl_calipso_ops_register(const struct netlbl_calipso_ops *ops) +{ + return xchg(&calipso_ops, ops); +} +EXPORT_SYMBOL(netlbl_calipso_ops_register); + +static const struct netlbl_calipso_ops *netlbl_calipso_ops_get(void) +{ + return ACCESS_ONCE(calipso_ops); +} + +/** + * calipso_doi_add - Add a new DOI to the CALIPSO protocol engine + * @doi_def: the DOI structure + * @audit_info: NetLabel audit information + * + * Description: + * The caller defines a new DOI for use by the CALIPSO engine and calls this + * function to add it to the list of acceptable domains. The caller must + * ensure that the mapping table specified in @doi_def->map meets all of the + * requirements of the mapping type (see calipso.h for details). Returns + * zero on success and non-zero on failure. + * + */ +int calipso_doi_add(struct calipso_doi *doi_def, + struct netlbl_audit *audit_info) +{ + int ret_val = -ENOMSG; + const struct netlbl_calipso_ops *ops = netlbl_calipso_ops_get(); + + if (ops) + ret_val = ops->doi_add(doi_def, audit_info); + return ret_val; +} + +/** + * calipso_doi_free - Frees a DOI definition + * @doi_def: the DOI definition + * + * Description: + * This function frees all of the memory associated with a DOI definition. + * + */ +void calipso_doi_free(struct calipso_doi *doi_def) +{ + const struct netlbl_calipso_ops *ops = netlbl_calipso_ops_get(); + + if (ops) + ops->doi_free(doi_def); +} diff --git a/net/netlabel/netlabel_calipso.h b/net/netlabel/netlabel_calipso.h new file mode 100644 index 000000000000..f78790a6ce4f --- /dev/null +++ b/net/netlabel/netlabel_calipso.h @@ -0,0 +1,90 @@ +/* + * NetLabel CALIPSO Support + * + * This file defines the CALIPSO functions for the NetLabel system. The + * NetLabel system manages static and dynamic label mappings for network + * protocols such as CIPSO and RIPSO. + * + * Authors: Paul Moore + * Huw Davies + * + */ + +/* (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Huw Davies , 2015 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + * + */ + +#ifndef _NETLABEL_CALIPSO +#define _NETLABEL_CALIPSO + +#include +#include + +/* The following NetLabel payloads are supported by the CALIPSO subsystem. + * + * o ADD: + * Sent by an application to add a new DOI mapping table. + * + * Required attributes: + * + * NLBL_CALIPSO_A_DOI + * NLBL_CALIPSO_A_MTYPE + * + * If using CALIPSO_MAP_PASS no additional attributes are required. + * + */ + +/* NetLabel CALIPSO commands */ +enum { + NLBL_CALIPSO_C_UNSPEC, + NLBL_CALIPSO_C_ADD, + NLBL_CALIPSO_C_REMOVE, + NLBL_CALIPSO_C_LIST, + NLBL_CALIPSO_C_LISTALL, + __NLBL_CALIPSO_C_MAX, +}; + +/* NetLabel CALIPSO attributes */ +enum { + NLBL_CALIPSO_A_UNSPEC, + NLBL_CALIPSO_A_DOI, + /* (NLA_U32) + * the DOI value */ + NLBL_CALIPSO_A_MTYPE, + /* (NLA_U32) + * the mapping table type (defined in the calipso.h header as + * CALIPSO_MAP_*) */ + __NLBL_CALIPSO_A_MAX, +}; + +#define NLBL_CALIPSO_A_MAX (__NLBL_CALIPSO_A_MAX - 1) + +/* NetLabel protocol functions */ +#if IS_ENABLED(CONFIG_IPV6) +int netlbl_calipso_genl_init(void); +#else +static inline int netlbl_calipso_genl_init(void) +{ + return 0; +} +#endif + +int calipso_doi_add(struct calipso_doi *doi_def, + struct netlbl_audit *audit_info); +void calipso_doi_free(struct calipso_doi *doi_def); + +#endif diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index 7e2a68f9d165..8d2e483167a8 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -1170,6 +1170,7 @@ struct audit_buffer *netlbl_audit_start(int type, { return netlbl_audit_start_common(type, audit_info); } +EXPORT_SYMBOL(netlbl_audit_start); /* * Setup Functions diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index b2aeb5d44601..975b1e93271e 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -674,6 +674,15 @@ static int netlbl_mgmt_protocols(struct sk_buff *skb, goto protocols_return; protos_sent++; } +#if IS_ENABLED(CONFIG_IPV6) + if (protos_sent == 2) { + if (netlbl_mgmt_protocols_cb(skb, + cb, + NETLBL_NLTYPE_CALIPSO) < 0) + goto protocols_return; + protos_sent++; + } +#endif protocols_return: cb->args[0] = protos_sent; diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c index adf8b7900da2..58495f44c62a 100644 --- a/net/netlabel/netlabel_user.c +++ b/net/netlabel/netlabel_user.c @@ -44,6 +44,7 @@ #include "netlabel_mgmt.h" #include "netlabel_unlabeled.h" #include "netlabel_cipso_v4.h" +#include "netlabel_calipso.h" #include "netlabel_user.h" /* @@ -71,6 +72,10 @@ int __init netlbl_netlink_init(void) if (ret_val != 0) return ret_val; + ret_val = netlbl_calipso_genl_init(); + if (ret_val != 0) + return ret_val; + return netlbl_unlabel_genl_init(); } -- cgit v1.2.3 From ceba1832b1b2da0149c51de62a847c00bca1677a Mon Sep 17 00:00:00 2001 From: Huw Davies Date: Mon, 27 Jun 2016 15:02:51 -0400 Subject: calipso: Set the calipso socket label to match the secattr. CALIPSO is a hop-by-hop IPv6 option. A lot of this patch is based on the equivalent CISPO code. The main difference is due to manipulating the options in the hop-by-hop header. Signed-off-by: Huw Davies Signed-off-by: Paul Moore --- include/net/ipv6.h | 2 + include/net/netlabel.h | 9 + include/uapi/linux/in6.h | 1 + net/ipv6/calipso.c | 595 ++++++++++++++++++++++++++++++++++++++++ net/ipv6/ipv6_sockglue.c | 1 - net/netlabel/Kconfig | 1 + net/netlabel/netlabel_calipso.c | 64 +++++ net/netlabel/netlabel_calipso.h | 5 + net/netlabel/netlabel_kapi.c | 58 +++- security/selinux/netlabel.c | 2 +- 10 files changed, 728 insertions(+), 10 deletions(-) (limited to 'include/uapi') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 887313d978d0..4e279a83cdd0 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -319,6 +319,8 @@ struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space, bool ipv6_opt_accepted(const struct sock *sk, const struct sk_buff *skb, const struct inet6_skb_parm *opt); +struct ipv6_txoptions *ipv6_update_options(struct sock *sk, + struct ipv6_txoptions *opt); static inline bool ipv6_accept_ra(struct inet6_dev *idev) { diff --git a/include/net/netlabel.h b/include/net/netlabel.h index 9fc2cab9be98..918a6044c89c 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -226,6 +226,9 @@ struct netlbl_lsm_secattr { * @doi_getdef: returns a reference to a DOI * @doi_putdef: releases a reference of a DOI * @doi_walk: enumerate the DOI list + * @sock_getattr: retrieve the socket's attr + * @sock_setattr: set the socket's attr + * @sock_delattr: remove the socket's attr * * Description: * This structure is filled out by the CALIPSO engine and passed @@ -243,6 +246,12 @@ struct netlbl_calipso_ops { int (*doi_walk)(u32 *skip_cnt, int (*callback)(struct calipso_doi *doi_def, void *arg), void *cb_arg); + int (*sock_getattr)(struct sock *sk, + struct netlbl_lsm_secattr *secattr); + int (*sock_setattr)(struct sock *sk, + const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr); + void (*sock_delattr)(struct sock *sk); }; /* diff --git a/include/uapi/linux/in6.h b/include/uapi/linux/in6.h index 318a4828bf98..b39ea4f2e701 100644 --- a/include/uapi/linux/in6.h +++ b/include/uapi/linux/in6.h @@ -143,6 +143,7 @@ struct in6_flowlabel_req { #define IPV6_TLV_PAD1 0 #define IPV6_TLV_PADN 1 #define IPV6_TLV_ROUTERALERT 5 +#define IPV6_TLV_CALIPSO 7 /* RFC 5570 */ #define IPV6_TLV_JUMBO 194 #define IPV6_TLV_HAO 201 /* home address option */ diff --git a/net/ipv6/calipso.c b/net/ipv6/calipso.c index d7df7a4bd32e..959db5268b38 100644 --- a/net/ipv6/calipso.c +++ b/net/ipv6/calipso.c @@ -44,6 +44,24 @@ #include #include #include +#include + +/* Maximium size of the calipso option including + * the two-byte TLV header. + */ +#define CALIPSO_OPT_LEN_MAX (2 + 252) + +/* Size of the minimum calipso option including + * the two-byte TLV header. + */ +#define CALIPSO_HDR_LEN (2 + 8) + +/* Maximium size of the calipso option including + * the two-byte TLV header and upto 3 bytes of + * leading pad and 7 bytes of trailing pad. + */ +#define CALIPSO_OPT_LEN_MAX_WITH_PAD (3 + CALIPSO_OPT_LEN_MAX + 7) + /* List of available DOI definitions */ static DEFINE_SPINLOCK(calipso_doi_list_lock); @@ -297,6 +315,580 @@ doi_walk_return: return ret_val; } +/** + * calipso_map_cat_hton - Perform a category mapping from host to network + * @doi_def: the DOI definition + * @secattr: the security attributes + * @net_cat: the zero'd out category bitmap in network/CALIPSO format + * @net_cat_len: the length of the CALIPSO bitmap in bytes + * + * Description: + * Perform a label mapping to translate a local MLS category bitmap to the + * correct CALIPSO bitmap using the given DOI definition. Returns the minimum + * size in bytes of the network bitmap on success, negative values otherwise. + * + */ +static int calipso_map_cat_hton(const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr, + unsigned char *net_cat, + u32 net_cat_len) +{ + int spot = -1; + u32 net_spot_max = 0; + u32 net_clen_bits = net_cat_len * 8; + + for (;;) { + spot = netlbl_catmap_walk(secattr->attr.mls.cat, + spot + 1); + if (spot < 0) + break; + if (spot >= net_clen_bits) + return -ENOSPC; + netlbl_bitmap_setbit(net_cat, spot, 1); + + if (spot > net_spot_max) + net_spot_max = spot; + } + + return (net_spot_max / 32 + 1) * 4; +} + +/** + * calipso_map_cat_ntoh - Perform a category mapping from network to host + * @doi_def: the DOI definition + * @net_cat: the category bitmap in network/CALIPSO format + * @net_cat_len: the length of the CALIPSO bitmap in bytes + * @secattr: the security attributes + * + * Description: + * Perform a label mapping to translate a CALIPSO bitmap to the correct local + * MLS category bitmap using the given DOI definition. Returns zero on + * success, negative values on failure. + * + */ +static int calipso_map_cat_ntoh(const struct calipso_doi *doi_def, + const unsigned char *net_cat, + u32 net_cat_len, + struct netlbl_lsm_secattr *secattr) +{ + int ret_val; + int spot = -1; + u32 net_clen_bits = net_cat_len * 8; + + for (;;) { + spot = netlbl_bitmap_walk(net_cat, + net_clen_bits, + spot + 1, + 1); + if (spot < 0) { + if (spot == -2) + return -EFAULT; + return 0; + } + + ret_val = netlbl_catmap_setbit(&secattr->attr.mls.cat, + spot, + GFP_ATOMIC); + if (ret_val != 0) + return ret_val; + } + + return -EINVAL; +} + +/** + * calipso_pad_write - Writes pad bytes in TLV format + * @buf: the buffer + * @offset: offset from start of buffer to write padding + * @count: number of pad bytes to write + * + * Description: + * Write @count bytes of TLV padding into @buffer starting at offset @offset. + * @count should be less than 8 - see RFC 4942. + * + */ +static int calipso_pad_write(unsigned char *buf, unsigned int offset, + unsigned int count) +{ + if (WARN_ON_ONCE(count >= 8)) + return -EINVAL; + + switch (count) { + case 0: + break; + case 1: + buf[offset] = IPV6_TLV_PAD1; + break; + default: + buf[offset] = IPV6_TLV_PADN; + buf[offset + 1] = count - 2; + if (count > 2) + memset(buf + offset + 2, 0, count - 2); + break; + } + return 0; +} + +/** + * calipso_genopt - Generate a CALIPSO option + * @buf: the option buffer + * @start: offset from which to write + * @buf_len: the size of opt_buf + * @doi_def: the CALIPSO DOI to use + * @secattr: the security attributes + * + * Description: + * Generate a CALIPSO option using the DOI definition and security attributes + * passed to the function. This also generates upto three bytes of leading + * padding that ensures that the option is 4n + 2 aligned. It returns the + * number of bytes written (including any initial padding). + */ +static int calipso_genopt(unsigned char *buf, u32 start, u32 buf_len, + const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val; + u32 len, pad; + u16 crc; + static const unsigned char padding[4] = {2, 1, 0, 3}; + unsigned char *calipso; + + /* CALIPSO has 4n + 2 alignment */ + pad = padding[start & 3]; + if (buf_len <= start + pad + CALIPSO_HDR_LEN) + return -ENOSPC; + + if ((secattr->flags & NETLBL_SECATTR_MLS_LVL) == 0) + return -EPERM; + + len = CALIPSO_HDR_LEN; + + if (secattr->flags & NETLBL_SECATTR_MLS_CAT) { + ret_val = calipso_map_cat_hton(doi_def, + secattr, + buf + start + pad + len, + buf_len - start - pad - len); + if (ret_val < 0) + return ret_val; + len += ret_val; + } + + calipso_pad_write(buf, start, pad); + calipso = buf + start + pad; + + calipso[0] = IPV6_TLV_CALIPSO; + calipso[1] = len - 2; + *(__be32 *)(calipso + 2) = htonl(doi_def->doi); + calipso[6] = (len - CALIPSO_HDR_LEN) / 4; + calipso[7] = secattr->attr.mls.lvl, + crc = ~crc_ccitt(0xffff, calipso, len); + calipso[8] = crc & 0xff; + calipso[9] = (crc >> 8) & 0xff; + return pad + len; +} + +/* Hop-by-hop hdr helper functions + */ + +/** + * calipso_opt_update - Replaces socket's hop options with a new set + * @sk: the socket + * @hop: new hop options + * + * Description: + * Replaces @sk's hop options with @hop. @hop may be NULL to leave + * the socket with no hop options. + * + */ +static int calipso_opt_update(struct sock *sk, struct ipv6_opt_hdr *hop) +{ + struct ipv6_txoptions *old = txopt_get(inet6_sk(sk)), *txopts; + + txopts = ipv6_renew_options_kern(sk, old, IPV6_HOPOPTS, + hop, hop ? ipv6_optlen(hop) : 0); + txopt_put(old); + if (IS_ERR(txopts)) + return PTR_ERR(txopts); + + txopts = ipv6_update_options(sk, txopts); + if (txopts) { + atomic_sub(txopts->tot_len, &sk->sk_omem_alloc); + txopt_put(txopts); + } + + return 0; +} + +/** + * calipso_tlv_len - Returns the length of the TLV + * @opt: the option header + * @offset: offset of the TLV within the header + * + * Description: + * Returns the length of the TLV option at offset @offset within + * the option header @opt. Checks that the entire TLV fits inside + * the option header, returns a negative value if this is not the case. + */ +static int calipso_tlv_len(struct ipv6_opt_hdr *opt, unsigned int offset) +{ + unsigned char *tlv = (unsigned char *)opt; + unsigned int opt_len = ipv6_optlen(opt), tlv_len; + + if (offset < sizeof(*opt) || offset >= opt_len) + return -EINVAL; + if (tlv[offset] == IPV6_TLV_PAD1) + return 1; + if (offset + 1 >= opt_len) + return -EINVAL; + tlv_len = tlv[offset + 1] + 2; + if (offset + tlv_len > opt_len) + return -EINVAL; + return tlv_len; +} + +/** + * calipso_opt_find - Finds the CALIPSO option in an IPv6 hop options header + * @hop: the hop options header + * @start: on return holds the offset of any leading padding + * @end: on return holds the offset of the first non-pad TLV after CALIPSO + * + * Description: + * Finds the space occupied by a CALIPSO option (including any leading and + * trailing padding). + * + * If a CALIPSO option exists set @start and @end to the + * offsets within @hop of the start of padding before the first + * CALIPSO option and the end of padding after the first CALIPSO + * option. In this case the function returns 0. + * + * In the absence of a CALIPSO option, @start and @end will be + * set to the start and end of any trailing padding in the header. + * This is useful when appending a new option, as the caller may want + * to overwrite some of this padding. In this case the function will + * return -ENOENT. + */ +static int calipso_opt_find(struct ipv6_opt_hdr *hop, unsigned int *start, + unsigned int *end) +{ + int ret_val = -ENOENT, tlv_len; + unsigned int opt_len, offset, offset_s = 0, offset_e = 0; + unsigned char *opt = (unsigned char *)hop; + + opt_len = ipv6_optlen(hop); + offset = sizeof(*hop); + + while (offset < opt_len) { + tlv_len = calipso_tlv_len(hop, offset); + if (tlv_len < 0) + return tlv_len; + + switch (opt[offset]) { + case IPV6_TLV_PAD1: + case IPV6_TLV_PADN: + if (offset_e) + offset_e = offset; + break; + case IPV6_TLV_CALIPSO: + ret_val = 0; + offset_e = offset; + break; + default: + if (offset_e == 0) + offset_s = offset; + else + goto out; + } + offset += tlv_len; + } + +out: + if (offset_s) + *start = offset_s + calipso_tlv_len(hop, offset_s); + else + *start = sizeof(*hop); + if (offset_e) + *end = offset_e + calipso_tlv_len(hop, offset_e); + else + *end = opt_len; + + return ret_val; +} + +/** + * calipso_opt_insert - Inserts a CALIPSO option into an IPv6 hop opt hdr + * @hop: the original hop options header + * @doi_def: the CALIPSO DOI to use + * @secattr: the specific security attributes of the socket + * + * Description: + * Creates a new hop options header based on @hop with a + * CALIPSO option added to it. If @hop already contains a CALIPSO + * option this is overwritten, otherwise the new option is appended + * after any existing options. If @hop is NULL then the new header + * will contain just the CALIPSO option and any needed padding. + * + */ +static struct ipv6_opt_hdr * +calipso_opt_insert(struct ipv6_opt_hdr *hop, + const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) +{ + unsigned int start, end, buf_len, pad, hop_len; + struct ipv6_opt_hdr *new; + int ret_val; + + if (hop) { + hop_len = ipv6_optlen(hop); + ret_val = calipso_opt_find(hop, &start, &end); + if (ret_val && ret_val != -ENOENT) + return ERR_PTR(ret_val); + } else { + hop_len = 0; + start = sizeof(*hop); + end = 0; + } + + buf_len = hop_len + start - end + CALIPSO_OPT_LEN_MAX_WITH_PAD; + new = kzalloc(buf_len, GFP_ATOMIC); + if (!new) + return ERR_PTR(-ENOMEM); + + if (start > sizeof(*hop)) + memcpy(new, hop, start); + ret_val = calipso_genopt((unsigned char *)new, start, buf_len, doi_def, + secattr); + if (ret_val < 0) + return ERR_PTR(ret_val); + + buf_len = start + ret_val; + /* At this point buf_len aligns to 4n, so (buf_len & 4) pads to 8n */ + pad = ((buf_len & 4) + (end & 7)) & 7; + calipso_pad_write((unsigned char *)new, buf_len, pad); + buf_len += pad; + + if (end != hop_len) { + memcpy((char *)new + buf_len, (char *)hop + end, hop_len - end); + buf_len += hop_len - end; + } + new->nexthdr = 0; + new->hdrlen = buf_len / 8 - 1; + + return new; +} + +/** + * calipso_opt_del - Removes the CALIPSO option from an option header + * @hop: the original header + * @new: the new header + * + * Description: + * Creates a new header based on @hop without any CALIPSO option. If @hop + * doesn't contain a CALIPSO option it returns -ENOENT. If @hop contains + * no other non-padding options, it returns zero with @new set to NULL. + * Otherwise it returns zero, creates a new header without the CALIPSO + * option (and removing as much padding as possible) and returns with + * @new set to that header. + * + */ +static int calipso_opt_del(struct ipv6_opt_hdr *hop, + struct ipv6_opt_hdr **new) +{ + int ret_val; + unsigned int start, end, delta, pad, hop_len; + + ret_val = calipso_opt_find(hop, &start, &end); + if (ret_val) + return ret_val; + + hop_len = ipv6_optlen(hop); + if (start == sizeof(*hop) && end == hop_len) { + /* There's no other option in the header so return NULL */ + *new = NULL; + return 0; + } + + delta = (end - start) & ~7; + *new = kzalloc(hop_len - delta, GFP_ATOMIC); + if (!*new) + return -ENOMEM; + + memcpy(*new, hop, start); + (*new)->hdrlen -= delta / 8; + pad = (end - start) & 7; + calipso_pad_write((unsigned char *)*new, start, pad); + if (end != hop_len) + memcpy((char *)*new + start + pad, (char *)hop + end, + hop_len - end); + + return 0; +} + +/** + * calipso_opt_getattr - Get the security attributes from a memory block + * @calipso: the CALIPSO option + * @secattr: the security attributes + * + * Description: + * Inspect @calipso and return the security attributes in @secattr. + * Returns zero on success and negative values on failure. + * + */ +static int calipso_opt_getattr(const unsigned char *calipso, + struct netlbl_lsm_secattr *secattr) +{ + int ret_val = -ENOMSG; + u32 doi, len = calipso[1], cat_len = calipso[6] * 4; + struct calipso_doi *doi_def; + + if (cat_len + 8 > len) + return -EINVAL; + + doi = get_unaligned_be32(calipso + 2); + rcu_read_lock(); + doi_def = calipso_doi_search(doi); + if (!doi_def) + goto getattr_return; + + secattr->attr.mls.lvl = calipso[7]; + secattr->flags |= NETLBL_SECATTR_MLS_LVL; + + if (cat_len) { + ret_val = calipso_map_cat_ntoh(doi_def, + calipso + 10, + cat_len, + secattr); + if (ret_val != 0) { + netlbl_catmap_free(secattr->attr.mls.cat); + goto getattr_return; + } + + secattr->flags |= NETLBL_SECATTR_MLS_CAT; + } + + secattr->type = NETLBL_NLTYPE_CALIPSO; + +getattr_return: + rcu_read_unlock(); + return ret_val; +} + +/* sock functions. + */ + +/** + * calipso_sock_getattr - Get the security attributes from a sock + * @sk: the sock + * @secattr: the security attributes + * + * Description: + * Query @sk to see if there is a CALIPSO option attached to the sock and if + * there is return the CALIPSO security attributes in @secattr. This function + * requires that @sk be locked, or privately held, but it does not do any + * locking itself. Returns zero on success and negative values on failure. + * + */ +static int calipso_sock_getattr(struct sock *sk, + struct netlbl_lsm_secattr *secattr) +{ + struct ipv6_opt_hdr *hop; + int opt_len, len, ret_val = -ENOMSG, offset; + unsigned char *opt; + struct ipv6_txoptions *txopts = txopt_get(inet6_sk(sk)); + + if (!txopts || !txopts->hopopt) + goto done; + + hop = txopts->hopopt; + opt = (unsigned char *)hop; + opt_len = ipv6_optlen(hop); + offset = sizeof(*hop); + while (offset < opt_len) { + len = calipso_tlv_len(hop, offset); + if (len < 0) { + ret_val = len; + goto done; + } + switch (opt[offset]) { + case IPV6_TLV_CALIPSO: + if (len < CALIPSO_HDR_LEN) + ret_val = -EINVAL; + else + ret_val = calipso_opt_getattr(&opt[offset], + secattr); + goto done; + default: + offset += len; + break; + } + } +done: + txopt_put(txopts); + return ret_val; +} + +/** + * calipso_sock_setattr - Add a CALIPSO option to a socket + * @sk: the socket + * @doi_def: the CALIPSO DOI to use + * @secattr: the specific security attributes of the socket + * + * Description: + * Set the CALIPSO option on the given socket using the DOI definition and + * security attributes passed to the function. This function requires + * exclusive access to @sk, which means it either needs to be in the + * process of being created or locked. Returns zero on success and negative + * values on failure. + * + */ +static int calipso_sock_setattr(struct sock *sk, + const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val; + struct ipv6_opt_hdr *old, *new; + struct ipv6_txoptions *txopts = txopt_get(inet6_sk(sk)); + + old = NULL; + if (txopts) + old = txopts->hopopt; + + new = calipso_opt_insert(old, doi_def, secattr); + txopt_put(txopts); + if (IS_ERR(new)) + return PTR_ERR(new); + + ret_val = calipso_opt_update(sk, new); + + kfree(new); + return ret_val; +} + +/** + * calipso_sock_delattr - Delete the CALIPSO option from a socket + * @sk: the socket + * + * Description: + * Removes the CALIPSO option from a socket, if present. + * + */ +static void calipso_sock_delattr(struct sock *sk) +{ + struct ipv6_opt_hdr *new_hop; + struct ipv6_txoptions *txopts = txopt_get(inet6_sk(sk)); + + if (!txopts || !txopts->hopopt) + goto done; + + if (calipso_opt_del(txopts->hopopt, &new_hop)) + goto done; + + calipso_opt_update(sk, new_hop); + kfree(new_hop); + +done: + txopt_put(txopts); +} + static const struct netlbl_calipso_ops ops = { .doi_add = calipso_doi_add, .doi_free = calipso_doi_free, @@ -304,6 +896,9 @@ static const struct netlbl_calipso_ops ops = { .doi_getdef = calipso_doi_getdef, .doi_putdef = calipso_doi_putdef, .doi_walk = calipso_doi_walk, + .sock_getattr = calipso_sock_getattr, + .sock_setattr = calipso_sock_setattr, + .sock_delattr = calipso_sock_delattr, }; /** diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 4449ad1f8114..8a80d59bed34 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -98,7 +98,6 @@ int ip6_ra_control(struct sock *sk, int sel) return 0; } -static struct ipv6_txoptions *ipv6_update_options(struct sock *sk, struct ipv6_txoptions *opt) { diff --git a/net/netlabel/Kconfig b/net/netlabel/Kconfig index 56958c85f2b4..d9eaa30ffe3f 100644 --- a/net/netlabel/Kconfig +++ b/net/netlabel/Kconfig @@ -5,6 +5,7 @@ config NETLABEL bool "NetLabel subsystem support" depends on SECURITY + select CRC_CCITT if IPV6 default n ---help--- NetLabel provides support for explicit network packet labeling diff --git a/net/netlabel/netlabel_calipso.c b/net/netlabel/netlabel_calipso.c index 2857673e8802..6f9c6589a022 100644 --- a/net/netlabel/netlabel_calipso.c +++ b/net/netlabel/netlabel_calipso.c @@ -514,3 +514,67 @@ int calipso_doi_walk(u32 *skip_cnt, ret_val = ops->doi_walk(skip_cnt, callback, cb_arg); return ret_val; } + +/** + * calipso_sock_getattr - Get the security attributes from a sock + * @sk: the sock + * @secattr: the security attributes + * + * Description: + * Query @sk to see if there is a CALIPSO option attached to the sock and if + * there is return the CALIPSO security attributes in @secattr. This function + * requires that @sk be locked, or privately held, but it does not do any + * locking itself. Returns zero on success and negative values on failure. + * + */ +int calipso_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr) +{ + int ret_val = -ENOMSG; + const struct netlbl_calipso_ops *ops = netlbl_calipso_ops_get(); + + if (ops) + ret_val = ops->sock_getattr(sk, secattr); + return ret_val; +} + +/** + * calipso_sock_setattr - Add a CALIPSO option to a socket + * @sk: the socket + * @doi_def: the CALIPSO DOI to use + * @secattr: the specific security attributes of the socket + * + * Description: + * Set the CALIPSO option on the given socket using the DOI definition and + * security attributes passed to the function. This function requires + * exclusive access to @sk, which means it either needs to be in the + * process of being created or locked. Returns zero on success and negative + * values on failure. + * + */ +int calipso_sock_setattr(struct sock *sk, + const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val = -ENOMSG; + const struct netlbl_calipso_ops *ops = netlbl_calipso_ops_get(); + + if (ops) + ret_val = ops->sock_setattr(sk, doi_def, secattr); + return ret_val; +} + +/** + * calipso_sock_delattr - Delete the CALIPSO option from a socket + * @sk: the socket + * + * Description: + * Removes the CALIPSO option from a socket, if present. + * + */ +void calipso_sock_delattr(struct sock *sk) +{ + const struct netlbl_calipso_ops *ops = netlbl_calipso_ops_get(); + + if (ops) + ops->sock_delattr(sk); +} diff --git a/net/netlabel/netlabel_calipso.h b/net/netlabel/netlabel_calipso.h index ed78554ba472..49bc116705c4 100644 --- a/net/netlabel/netlabel_calipso.h +++ b/net/netlabel/netlabel_calipso.h @@ -128,5 +128,10 @@ void calipso_doi_putdef(struct calipso_doi *doi_def); int calipso_doi_walk(u32 *skip_cnt, int (*callback)(struct calipso_doi *doi_def, void *arg), void *cb_arg); +int calipso_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr); +int calipso_sock_setattr(struct sock *sk, + const struct calipso_doi *doi_def, + const struct netlbl_lsm_secattr *secattr); +void calipso_sock_delattr(struct sock *sk); #endif diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index 54f13a33b52c..00bab51c291e 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -37,12 +37,14 @@ #include #include #include +#include #include #include #include "netlabel_domainhash.h" #include "netlabel_unlabeled.h" #include "netlabel_cipso_v4.h" +#include "netlabel_calipso.h" #include "netlabel_user.h" #include "netlabel_mgmt.h" #include "netlabel_addrlist.h" @@ -521,6 +523,7 @@ int netlbl_catmap_walk(struct netlbl_lsm_catmap *catmap, u32 offset) return -ENOENT; } +EXPORT_SYMBOL(netlbl_catmap_walk); /** * netlbl_catmap_walkrng - Find the end of a string of set bits @@ -656,6 +659,7 @@ int netlbl_catmap_setbit(struct netlbl_lsm_catmap **catmap, return 0; } +EXPORT_SYMBOL(netlbl_catmap_setbit); /** * netlbl_catmap_setrng - Set a range of bits in a LSM secattr catmap @@ -870,9 +874,21 @@ int netlbl_sock_setattr(struct sock *sk, break; #if IS_ENABLED(CONFIG_IPV6) case AF_INET6: - /* since we don't support any IPv6 labeling protocols right - * now we can optimize everything away until we do */ - ret_val = 0; + switch (dom_entry->def.type) { + case NETLBL_NLTYPE_ADDRSELECT: + ret_val = -EDESTADDRREQ; + break; + case NETLBL_NLTYPE_CALIPSO: + ret_val = calipso_sock_setattr(sk, + dom_entry->def.calipso, + secattr); + break; + case NETLBL_NLTYPE_UNLABELED: + ret_val = 0; + break; + default: + ret_val = -ENOENT; + } break; #endif /* IPv6 */ default: @@ -899,6 +915,11 @@ void netlbl_sock_delattr(struct sock *sk) case AF_INET: cipso_v4_sock_delattr(sk); break; +#if IS_ENABLED(CONFIG_IPV6) + case AF_INET6: + calipso_sock_delattr(sk); + break; +#endif /* IPv6 */ } } @@ -925,7 +946,7 @@ int netlbl_sock_getattr(struct sock *sk, break; #if IS_ENABLED(CONFIG_IPV6) case AF_INET6: - ret_val = -ENOMSG; + ret_val = calipso_sock_getattr(sk, secattr); break; #endif /* IPv6 */ default: @@ -953,6 +974,9 @@ int netlbl_conn_setattr(struct sock *sk, { int ret_val; struct sockaddr_in *addr4; +#if IS_ENABLED(CONFIG_IPV6) + struct sockaddr_in6 *addr6; +#endif struct netlbl_dommap_def *entry; rcu_read_lock(); @@ -973,7 +997,7 @@ int netlbl_conn_setattr(struct sock *sk, case NETLBL_NLTYPE_UNLABELED: /* just delete the protocols we support for right now * but we could remove other protocols if needed */ - cipso_v4_sock_delattr(sk); + netlbl_sock_delattr(sk); ret_val = 0; break; default: @@ -982,9 +1006,27 @@ int netlbl_conn_setattr(struct sock *sk, break; #if IS_ENABLED(CONFIG_IPV6) case AF_INET6: - /* since we don't support any IPv6 labeling protocols right - * now we can optimize everything away until we do */ - ret_val = 0; + addr6 = (struct sockaddr_in6 *)addr; + entry = netlbl_domhsh_getentry_af6(secattr->domain, + &addr6->sin6_addr); + if (entry == NULL) { + ret_val = -ENOENT; + goto conn_setattr_return; + } + switch (entry->type) { + case NETLBL_NLTYPE_CALIPSO: + ret_val = calipso_sock_setattr(sk, + entry->calipso, secattr); + break; + case NETLBL_NLTYPE_UNLABELED: + /* just delete the protocols we support for right now + * but we could remove other protocols if needed */ + netlbl_sock_delattr(sk); + ret_val = 0; + break; + default: + ret_val = -ENOENT; + } break; #endif /* IPv6 */ default: diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c index 1f989a539fd4..5470f32eca54 100644 --- a/security/selinux/netlabel.c +++ b/security/selinux/netlabel.c @@ -333,7 +333,7 @@ int selinux_netlbl_socket_post_create(struct sock *sk, u16 family) struct sk_security_struct *sksec = sk->sk_security; struct netlbl_lsm_secattr *secattr; - if (family != PF_INET) + if (family != PF_INET && family != PF_INET6) return 0; secattr = selinux_netlbl_sock_genattr(sk); -- cgit v1.2.3 From b810253bd9342f863a86ec7dfff4a5a7a0394d2f Mon Sep 17 00:00:00 2001 From: Philippe Bergheaud Date: Thu, 23 Jun 2016 15:03:53 +0200 Subject: cxl: Add mechanism for delivering AFU driver specific events This adds an afu_driver_ops structure with fetch_event() and event_delivered() callbacks. An AFU driver such as cxlflash can fill this out and associate it with a context to enable passing custom AFU specific events to userspace. This also adds a new kernel API function cxl_context_pending_events(), that the AFU driver can use to notify the cxl driver that new specific events are ready to be delivered, and wake up anyone waiting on the context wait queue. The current count of AFU driver specific events is stored in the field afu_driver_events of the context structure. The cxl driver checks the afu_driver_events count during poll, select, read, etc. calls to check if an AFU driver specific event is pending, and calls fetch_event() to obtain and deliver that event. This way, the cxl driver takes care of all the usual locking semantics around these calls and handles all the generic cxl events, so that the AFU driver only needs to worry about it's own events. fetch_event() return a struct cxl_event_afu_driver_reserved, allocated by the AFU driver, and filled in with the specific event information and size. Total event size (header + data) should not be greater than CXL_READ_MIN_SIZE (4K). Th cxl driver prepends an appropriate cxl event header, copies the event to userspace, and finally calls event_delivered() to return the status of the operation to the AFU driver. The event is identified by the context and cxl_event_afu_driver_reserved pointers. Since AFU drivers provide their own means for userspace to obtain the AFU file descriptor (i.e. cxlflash uses an ioctl on their scsi file descriptor to obtain the AFU file descriptor) and the generic cxl driver will never use this event, the ABI of the event is up to each individual AFU driver. Signed-off-by: Philippe Bergheaud Signed-off-by: Michael Ellerman --- drivers/misc/cxl/Kconfig | 5 ++++ drivers/misc/cxl/api.c | 17 +++++++++++++ drivers/misc/cxl/cxl.h | 7 +++++- drivers/misc/cxl/file.c | 64 ++++++++++++++++++++++++++++++++++++++++++------ include/misc/cxl.h | 48 ++++++++++++++++++++++++++++++++++++ include/uapi/misc/cxl.h | 17 +++++++++++++ 6 files changed, 149 insertions(+), 9 deletions(-) (limited to 'include/uapi') diff --git a/drivers/misc/cxl/Kconfig b/drivers/misc/cxl/Kconfig index 8756d06e2bb8..560412cd4c98 100644 --- a/drivers/misc/cxl/Kconfig +++ b/drivers/misc/cxl/Kconfig @@ -15,12 +15,17 @@ config CXL_EEH bool default n +config CXL_AFU_DRIVER_OPS + bool + default n + config CXL tristate "Support for IBM Coherent Accelerators (CXL)" depends on PPC_POWERNV && PCI_MSI && EEH select CXL_BASE select CXL_KERNEL_API select CXL_EEH + select CXL_AFU_DRIVER_OPS default m help Select this option to enable driver support for IBM Coherent diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c index 99081b821f12..f11dc0e5fdd6 100644 --- a/drivers/misc/cxl/api.c +++ b/drivers/misc/cxl/api.c @@ -333,6 +333,23 @@ struct cxl_context *cxl_fops_get_context(struct file *file) } EXPORT_SYMBOL_GPL(cxl_fops_get_context); +void cxl_set_driver_ops(struct cxl_context *ctx, + struct cxl_afu_driver_ops *ops) +{ + WARN_ON(!ops->fetch_event || !ops->event_delivered); + atomic_set(&ctx->afu_driver_events, 0); + ctx->afu_driver_ops = ops; +} +EXPORT_SYMBOL_GPL(cxl_set_driver_ops); + +void cxl_context_events_pending(struct cxl_context *ctx, + unsigned int new_events) +{ + atomic_add(new_events, &ctx->afu_driver_events); + wake_up_all(&ctx->wq); +} +EXPORT_SYMBOL_GPL(cxl_context_events_pending); + int cxl_start_work(struct cxl_context *ctx, struct cxl_ioctl_start_work *work) { diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h index ce2b9d513069..422ee53868a8 100644 --- a/drivers/misc/cxl/cxl.h +++ b/drivers/misc/cxl/cxl.h @@ -24,6 +24,7 @@ #include #include +#include #include extern uint cxl_verbose; @@ -34,7 +35,7 @@ extern uint cxl_verbose; * Bump version each time a user API change is made, whether it is * backwards compatible ot not. */ -#define CXL_API_VERSION 2 +#define CXL_API_VERSION 3 #define CXL_API_VERSION_COMPATIBLE 1 /* @@ -528,6 +529,10 @@ struct cxl_context { bool pending_fault; bool pending_afu_err; + /* Used by AFU drivers for driver specific event delivery */ + struct cxl_afu_driver_ops *afu_driver_ops; + atomic_t afu_driver_events; + struct rcu_head rcu; }; diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c index eec468f1612f..5fb9894b157f 100644 --- a/drivers/misc/cxl/file.c +++ b/drivers/misc/cxl/file.c @@ -293,6 +293,17 @@ int afu_mmap(struct file *file, struct vm_area_struct *vm) return cxl_context_iomap(ctx, vm); } +static inline bool ctx_event_pending(struct cxl_context *ctx) +{ + if (ctx->pending_irq || ctx->pending_fault || ctx->pending_afu_err) + return true; + + if (ctx->afu_driver_ops && atomic_read(&ctx->afu_driver_events)) + return true; + + return false; +} + unsigned int afu_poll(struct file *file, struct poll_table_struct *poll) { struct cxl_context *ctx = file->private_data; @@ -305,8 +316,7 @@ unsigned int afu_poll(struct file *file, struct poll_table_struct *poll) pr_devel("afu_poll wait done pe: %i\n", ctx->pe); spin_lock_irqsave(&ctx->lock, flags); - if (ctx->pending_irq || ctx->pending_fault || - ctx->pending_afu_err) + if (ctx_event_pending(ctx)) mask |= POLLIN | POLLRDNORM; else if (ctx->status == CLOSED) /* Only error on closed when there are no futher events pending @@ -319,16 +329,46 @@ unsigned int afu_poll(struct file *file, struct poll_table_struct *poll) return mask; } -static inline int ctx_event_pending(struct cxl_context *ctx) +static ssize_t afu_driver_event_copy(struct cxl_context *ctx, + char __user *buf, + struct cxl_event *event, + struct cxl_event_afu_driver_reserved *pl) { - return (ctx->pending_irq || ctx->pending_fault || - ctx->pending_afu_err || (ctx->status == CLOSED)); + /* Check event */ + if (!pl) { + ctx->afu_driver_ops->event_delivered(ctx, pl, -EINVAL); + return -EFAULT; + } + + /* Check event size */ + event->header.size += pl->data_size; + if (event->header.size > CXL_READ_MIN_SIZE) { + ctx->afu_driver_ops->event_delivered(ctx, pl, -EINVAL); + return -EFAULT; + } + + /* Copy event header */ + if (copy_to_user(buf, event, sizeof(struct cxl_event_header))) { + ctx->afu_driver_ops->event_delivered(ctx, pl, -EFAULT); + return -EFAULT; + } + + /* Copy event data */ + buf += sizeof(struct cxl_event_header); + if (copy_to_user(buf, &pl->data, pl->data_size)) { + ctx->afu_driver_ops->event_delivered(ctx, pl, -EFAULT); + return -EFAULT; + } + + ctx->afu_driver_ops->event_delivered(ctx, pl, 0); /* Success */ + return event->header.size; } ssize_t afu_read(struct file *file, char __user *buf, size_t count, loff_t *off) { struct cxl_context *ctx = file->private_data; + struct cxl_event_afu_driver_reserved *pl = NULL; struct cxl_event event; unsigned long flags; int rc; @@ -344,7 +384,7 @@ ssize_t afu_read(struct file *file, char __user *buf, size_t count, for (;;) { prepare_to_wait(&ctx->wq, &wait, TASK_INTERRUPTIBLE); - if (ctx_event_pending(ctx)) + if (ctx_event_pending(ctx) || (ctx->status == CLOSED)) break; if (!cxl_ops->link_ok(ctx->afu->adapter, ctx->afu)) { @@ -374,7 +414,12 @@ ssize_t afu_read(struct file *file, char __user *buf, size_t count, memset(&event, 0, sizeof(event)); event.header.process_element = ctx->pe; event.header.size = sizeof(struct cxl_event_header); - if (ctx->pending_irq) { + if (ctx->afu_driver_ops && atomic_read(&ctx->afu_driver_events)) { + pr_devel("afu_read delivering AFU driver specific event\n"); + pl = ctx->afu_driver_ops->fetch_event(ctx); + atomic_dec(&ctx->afu_driver_events); + event.header.type = CXL_EVENT_AFU_DRIVER; + } else if (ctx->pending_irq) { pr_devel("afu_read delivering AFU interrupt\n"); event.header.size += sizeof(struct cxl_event_afu_interrupt); event.header.type = CXL_EVENT_AFU_INTERRUPT; @@ -404,6 +449,9 @@ ssize_t afu_read(struct file *file, char __user *buf, size_t count, spin_unlock_irqrestore(&ctx->lock, flags); + if (event.header.type == CXL_EVENT_AFU_DRIVER) + return afu_driver_event_copy(ctx, buf, &event, pl); + if (copy_to_user(buf, &event, event.header.size)) return -EFAULT; return event.header.size; @@ -558,7 +606,7 @@ int __init cxl_file_init(void) * If these change we really need to update API. Either change some * flags or update API version number CXL_API_VERSION. */ - BUILD_BUG_ON(CXL_API_VERSION != 2); + BUILD_BUG_ON(CXL_API_VERSION != 3); BUILD_BUG_ON(sizeof(struct cxl_ioctl_start_work) != 64); BUILD_BUG_ON(sizeof(struct cxl_event_header) != 8); BUILD_BUG_ON(sizeof(struct cxl_event_afu_interrupt) != 8); diff --git a/include/misc/cxl.h b/include/misc/cxl.h index 56560c5781b4..17419f61e611 100644 --- a/include/misc/cxl.h +++ b/include/misc/cxl.h @@ -220,4 +220,52 @@ void cxl_perst_reloads_same_image(struct cxl_afu *afu, */ ssize_t cxl_read_adapter_vpd(struct pci_dev *dev, void *buf, size_t count); +/* + * AFU driver ops allow an AFU driver to create their own events to pass to + * userspace through the file descriptor as a simpler alternative to overriding + * the read() and poll() calls that works with the generic cxl events. These + * events are given priority over the generic cxl events, so they will be + * delivered first if multiple types of events are pending. + * + * The AFU driver must call cxl_context_events_pending() to notify the cxl + * driver that new events are ready to be delivered for a specific context. + * cxl_context_events_pending() will adjust the current count of AFU driver + * events for this context, and wake up anyone waiting on the context wait + * queue. + * + * The cxl driver will then call fetch_event() to get a structure defining + * the size and address of the driver specific event data. The cxl driver + * will build a cxl header with type and process_element fields filled in, + * and header.size set to sizeof(struct cxl_event_header) + data_size. + * The total size of the event is limited to CXL_READ_MIN_SIZE (4K). + * + * fetch_event() is called with a spin lock held, so it must not sleep. + * + * The cxl driver will then deliver the event to userspace, and finally + * call event_delivered() to return the status of the operation, identified + * by cxl context and AFU driver event data pointers. + * 0 Success + * -EFAULT copy_to_user() has failed + * -EINVAL Event data pointer is NULL, or event size is greater than + * CXL_READ_MIN_SIZE. + */ +struct cxl_afu_driver_ops { + struct cxl_event_afu_driver_reserved *(*fetch_event) ( + struct cxl_context *ctx); + void (*event_delivered) (struct cxl_context *ctx, + struct cxl_event_afu_driver_reserved *event, + int rc); +}; + +/* + * Associate the above driver ops with a specific context. + * Reset the current count of AFU driver events. + */ +void cxl_set_driver_ops(struct cxl_context *ctx, + struct cxl_afu_driver_ops *ops); + +/* Notify cxl driver that new events are ready to be delivered for context */ +void cxl_context_events_pending(struct cxl_context *ctx, + unsigned int new_events); + #endif /* _MISC_CXL_H */ diff --git a/include/uapi/misc/cxl.h b/include/uapi/misc/cxl.h index 8cd334f99ddc..cbae529b7ce0 100644 --- a/include/uapi/misc/cxl.h +++ b/include/uapi/misc/cxl.h @@ -93,6 +93,7 @@ enum cxl_event_type { CXL_EVENT_AFU_INTERRUPT = 1, CXL_EVENT_DATA_STORAGE = 2, CXL_EVENT_AFU_ERROR = 3, + CXL_EVENT_AFU_DRIVER = 4, }; struct cxl_event_header { @@ -124,12 +125,28 @@ struct cxl_event_afu_error { __u64 error; }; +struct cxl_event_afu_driver_reserved { + /* + * Defines the buffer passed to the cxl driver by the AFU driver. + * + * This is not ABI since the event header.size passed to the user for + * existing events is set in the read call to sizeof(cxl_event_header) + * + sizeof(whatever event is being dispatched) and the user is already + * required to use a 4K buffer on the read call. + * + * Of course the contents will be ABI, but that's up the AFU driver. + */ + size_t data_size; + u8 data[]; +}; + struct cxl_event { struct cxl_event_header header; union { struct cxl_event_afu_interrupt irq; struct cxl_event_data_storage fault; struct cxl_event_afu_error afu_error; + struct cxl_event_afu_driver_reserved afu_driver_event; }; }; -- cgit v1.2.3 From 637c841dd7a5f9bd97b75cbe90b526fa1a52e530 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 23 Jun 2016 18:42:51 -0700 Subject: net: diag: Add support to filter on device index Add support to inet_diag facility to filter sockets based on device index. If an interface index is in the filter only sockets bound to that index (sk_bound_dev_if) are returned. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- include/uapi/linux/inet_diag.h | 1 + net/ipv4/inet_diag.c | 25 +++++++++++++++++++++++++ 2 files changed, 26 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/inet_diag.h b/include/uapi/linux/inet_diag.h index a16643705669..abbd1dc5d683 100644 --- a/include/uapi/linux/inet_diag.h +++ b/include/uapi/linux/inet_diag.h @@ -72,6 +72,7 @@ enum { INET_DIAG_BC_AUTO, INET_DIAG_BC_S_COND, INET_DIAG_BC_D_COND, + INET_DIAG_BC_DEV_COND, /* u32 ifindex */ }; struct inet_diag_hostcond { diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 25af1243649b..38c2c47fe0e8 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -44,6 +44,7 @@ struct inet_diag_entry { u16 dport; u16 family; u16 userlocks; + u32 ifindex; }; static DEFINE_MUTEX(inet_diag_table_mutex); @@ -571,6 +572,14 @@ static int inet_diag_bc_run(const struct nlattr *_bc, yes = 0; break; } + case INET_DIAG_BC_DEV_COND: { + u32 ifindex; + + ifindex = *((const u32 *)(op + 1)); + if (ifindex != entry->ifindex) + yes = 0; + break; + } } if (yes) { @@ -613,6 +622,7 @@ int inet_diag_bc_sk(const struct nlattr *bc, struct sock *sk) entry_fill_addrs(&entry, sk); entry.sport = inet->inet_num; entry.dport = ntohs(inet->inet_dport); + entry.ifindex = sk->sk_bound_dev_if; entry.userlocks = sk_fullsock(sk) ? sk->sk_userlocks : 0; return inet_diag_bc_run(bc, &entry); @@ -636,6 +646,17 @@ static int valid_cc(const void *bc, int len, int cc) return 0; } +/* data is u32 ifindex */ +static bool valid_devcond(const struct inet_diag_bc_op *op, int len, + int *min_len) +{ + /* Check ifindex space. */ + *min_len += sizeof(u32); + if (len < *min_len) + return false; + + return true; +} /* Validate an inet_diag_hostcond. */ static bool valid_hostcond(const struct inet_diag_bc_op *op, int len, int *min_len) @@ -700,6 +721,10 @@ static int inet_diag_bc_audit(const void *bytecode, int bytecode_len) if (!valid_hostcond(bc, len, &min_len)) return -EINVAL; break; + case INET_DIAG_BC_DEV_COND: + if (!valid_devcond(bc, len, &min_len)) + return -EINVAL; + break; case INET_DIAG_BC_S_GE: case INET_DIAG_BC_S_LE: case INET_DIAG_BC_D_GE: -- cgit v1.2.3 From 79c8d083e226fbfc20d9e801352fccfa7b08a8b3 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Wed, 15 Jun 2016 20:27:58 -0300 Subject: [media] videodev2.h: Group YUV 3 planes formats together The formats are interleaved with the YUV packed and miscellaneous formats, making the result confusing especially with the YUV444 format being packed and not planar like YUV410 or YUV420. Move them to their own group as the 2 planes or 3 non-contiguous planes formats to clarify the header. Signed-off-by: Laurent Pinchart Acked-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- include/uapi/linux/videodev2.h | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/videodev2.h b/include/uapi/linux/videodev2.h index 8f951917be74..50ff346c4118 100644 --- a/include/uapi/linux/videodev2.h +++ b/include/uapi/linux/videodev2.h @@ -504,22 +504,16 @@ struct v4l2_pix_format { #define V4L2_PIX_FMT_UV8 v4l2_fourcc('U', 'V', '8', ' ') /* 8 UV 4:4 */ /* Luminance+Chrominance formats */ -#define V4L2_PIX_FMT_YVU410 v4l2_fourcc('Y', 'V', 'U', '9') /* 9 YVU 4:1:0 */ -#define V4L2_PIX_FMT_YVU420 v4l2_fourcc('Y', 'V', '1', '2') /* 12 YVU 4:2:0 */ #define V4L2_PIX_FMT_YUYV v4l2_fourcc('Y', 'U', 'Y', 'V') /* 16 YUV 4:2:2 */ #define V4L2_PIX_FMT_YYUV v4l2_fourcc('Y', 'Y', 'U', 'V') /* 16 YUV 4:2:2 */ #define V4L2_PIX_FMT_YVYU v4l2_fourcc('Y', 'V', 'Y', 'U') /* 16 YVU 4:2:2 */ #define V4L2_PIX_FMT_UYVY v4l2_fourcc('U', 'Y', 'V', 'Y') /* 16 YUV 4:2:2 */ #define V4L2_PIX_FMT_VYUY v4l2_fourcc('V', 'Y', 'U', 'Y') /* 16 YUV 4:2:2 */ -#define V4L2_PIX_FMT_YUV422P v4l2_fourcc('4', '2', '2', 'P') /* 16 YVU422 planar */ -#define V4L2_PIX_FMT_YUV411P v4l2_fourcc('4', '1', '1', 'P') /* 16 YVU411 planar */ #define V4L2_PIX_FMT_Y41P v4l2_fourcc('Y', '4', '1', 'P') /* 12 YUV 4:1:1 */ #define V4L2_PIX_FMT_YUV444 v4l2_fourcc('Y', '4', '4', '4') /* 16 xxxxyyyy uuuuvvvv */ #define V4L2_PIX_FMT_YUV555 v4l2_fourcc('Y', 'U', 'V', 'O') /* 16 YUV-5-5-5 */ #define V4L2_PIX_FMT_YUV565 v4l2_fourcc('Y', 'U', 'V', 'P') /* 16 YUV-5-6-5 */ #define V4L2_PIX_FMT_YUV32 v4l2_fourcc('Y', 'U', 'V', '4') /* 32 YUV-8-8-8-8 */ -#define V4L2_PIX_FMT_YUV410 v4l2_fourcc('Y', 'U', 'V', '9') /* 9 YUV 4:1:0 */ -#define V4L2_PIX_FMT_YUV420 v4l2_fourcc('Y', 'U', '1', '2') /* 12 YUV 4:2:0 */ #define V4L2_PIX_FMT_HI240 v4l2_fourcc('H', 'I', '2', '4') /* 8 8-bit color */ #define V4L2_PIX_FMT_HM12 v4l2_fourcc('H', 'M', '1', '2') /* 8 YUV 4:2:0 16x16 macroblocks */ #define V4L2_PIX_FMT_M420 v4l2_fourcc('M', '4', '2', '0') /* 12 YUV 4:2:0 2 lines y, 1 line uv interleaved */ @@ -540,6 +534,14 @@ struct v4l2_pix_format { #define V4L2_PIX_FMT_NV12MT v4l2_fourcc('T', 'M', '1', '2') /* 12 Y/CbCr 4:2:0 64x32 macroblocks */ #define V4L2_PIX_FMT_NV12MT_16X16 v4l2_fourcc('V', 'M', '1', '2') /* 12 Y/CbCr 4:2:0 16x16 macroblocks */ +/* three planes - Y Cb, Cr */ +#define V4L2_PIX_FMT_YUV410 v4l2_fourcc('Y', 'U', 'V', '9') /* 9 YUV 4:1:0 */ +#define V4L2_PIX_FMT_YVU410 v4l2_fourcc('Y', 'V', 'U', '9') /* 9 YVU 4:1:0 */ +#define V4L2_PIX_FMT_YUV411P v4l2_fourcc('4', '1', '1', 'P') /* 16 YVU411 planar */ +#define V4L2_PIX_FMT_YUV420 v4l2_fourcc('Y', 'U', '1', '2') /* 12 YUV 4:2:0 */ +#define V4L2_PIX_FMT_YVU420 v4l2_fourcc('Y', 'V', '1', '2') /* 12 YVU 4:2:0 */ +#define V4L2_PIX_FMT_YUV422P v4l2_fourcc('4', '2', '2', 'P') /* 16 YVU422 planar */ + /* three non contiguous planes - Y, Cb, Cr */ #define V4L2_PIX_FMT_YUV420M v4l2_fourcc('Y', 'M', '1', '2') /* 12 YUV420 planar */ #define V4L2_PIX_FMT_YVU420M v4l2_fourcc('Y', 'M', '2', '1') /* 12 YVU420 planar */ -- cgit v1.2.3 From ab46f6d24bf57ddac0f5abe2f546a78af57b476c Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Wed, 15 Jun 2016 20:35:36 -0300 Subject: [media] videodev2.h: Fix V4L2_PIX_FMT_YUV411P description YUV 4:1:1 uses 12 bits per pixel on average, not 16. Signed-off-by: Laurent Pinchart Acked-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- include/uapi/linux/videodev2.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/videodev2.h b/include/uapi/linux/videodev2.h index 50ff346c4118..724f43e69d03 100644 --- a/include/uapi/linux/videodev2.h +++ b/include/uapi/linux/videodev2.h @@ -537,7 +537,7 @@ struct v4l2_pix_format { /* three planes - Y Cb, Cr */ #define V4L2_PIX_FMT_YUV410 v4l2_fourcc('Y', 'U', 'V', '9') /* 9 YUV 4:1:0 */ #define V4L2_PIX_FMT_YVU410 v4l2_fourcc('Y', 'V', 'U', '9') /* 9 YVU 4:1:0 */ -#define V4L2_PIX_FMT_YUV411P v4l2_fourcc('4', '1', '1', 'P') /* 16 YVU411 planar */ +#define V4L2_PIX_FMT_YUV411P v4l2_fourcc('4', '1', '1', 'P') /* 12 YVU411 planar */ #define V4L2_PIX_FMT_YUV420 v4l2_fourcc('Y', 'U', '1', '2') /* 12 YUV 4:2:0 */ #define V4L2_PIX_FMT_YVU420 v4l2_fourcc('Y', 'V', '1', '2') /* 12 YVU 4:2:0 */ #define V4L2_PIX_FMT_YUV422P v4l2_fourcc('4', '2', '2', 'P') /* 16 YVU422 planar */ -- cgit v1.2.3 From 1179aab13db3c6251484afe492c8dbd869ca8b05 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Mon, 15 Feb 2016 22:00:30 -0200 Subject: [media] media: Add video processing entity functions Add composer, pixel formatter, pixel encoding converter and scaler functions. Signed-off-by: Laurent Pinchart Acked-by: Hans Verkuil Acked-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- Documentation/DocBook/media/v4l/media-types.xml | 55 +++++++++++++++++++++++++ include/uapi/linux/media.h | 9 ++++ 2 files changed, 64 insertions(+) (limited to 'include/uapi') diff --git a/Documentation/DocBook/media/v4l/media-types.xml b/Documentation/DocBook/media/v4l/media-types.xml index 5e3f20fdcf17..60fe841f8846 100644 --- a/Documentation/DocBook/media/v4l/media-types.xml +++ b/Documentation/DocBook/media/v4l/media-types.xml @@ -121,6 +121,61 @@ MEDIA_ENT_F_AUDIO_MIXER Audio Mixer Function Entity. + + MEDIA_ENT_F_PROC_VIDEO_COMPOSER + Video composer (blender). An entity capable of video + composing must have at least two sink pads and one source + pad, and composes input video frames onto output video + frames. Composition can be performed using alpha blending, + color keying, raster operations (ROP), stitching or any other + means. + + + + MEDIA_ENT_F_PROC_VIDEO_PIXEL_FORMATTER + Video pixel formatter. An entity capable of pixel formatting + must have at least one sink pad and one source pad. Read + pixel formatters read pixels from memory and perform a subset + of unpacking, cropping, color keying, alpha multiplication + and pixel encoding conversion. Write pixel formatters perform + a subset of dithering, pixel encoding conversion and packing + and write pixels to memory. + + + + MEDIA_ENT_F_PROC_VIDEO_PIXEL_ENC_CONV + Video pixel encoding converter. An entity capable of pixel + enconding conversion must have at least one sink pad and one + source pad, and convert the encoding of pixels received on + its sink pad(s) to a different encoding output on its source + pad(s). Pixel encoding conversion includes but isn't limited + to RGB to/from HSV, RGB to/from YUV and CFA (Bayer) to RGB + conversions. + + + + MEDIA_ENT_F_PROC_VIDEO_LUT + Video look-up table. An entity capable of video lookup table + processing must have one sink pad and one source pad. It uses + the values of the pixels received on its sink pad to look up + entries in internal tables and output them on its source pad. + The lookup processing can be performed on all components + separately or combine them for multi-dimensional table + lookups. + + + + MEDIA_ENT_F_PROC_VIDEO_SCALER + Video scaler. An entity capable of video scaling must have + at least one sink pad and one source pad, and scale the + video frame(s) received on its sink pad(s) to a different + resolution output on its source pad(s). The range of + supported scaling ratios is entity-specific and can differ + between the horizontal and vertical directions (in particular + scaling can be supported in one direction only). Binning and + skipping are considered as scaling. + + diff --git a/include/uapi/linux/media.h b/include/uapi/linux/media.h index df59edee25d1..3136686c4bd0 100644 --- a/include/uapi/linux/media.h +++ b/include/uapi/linux/media.h @@ -94,6 +94,15 @@ struct media_device_info { #define MEDIA_ENT_F_AUDIO_PLAYBACK (MEDIA_ENT_F_BASE + 0x03002) #define MEDIA_ENT_F_AUDIO_MIXER (MEDIA_ENT_F_BASE + 0x03003) +/* + * Processing entities + */ +#define MEDIA_ENT_F_PROC_VIDEO_COMPOSER (MEDIA_ENT_F_BASE + 0x4001) +#define MEDIA_ENT_F_PROC_VIDEO_PIXEL_FORMATTER (MEDIA_ENT_F_BASE + 0x4002) +#define MEDIA_ENT_F_PROC_VIDEO_PIXEL_ENC_CONV (MEDIA_ENT_F_BASE + 0x4003) +#define MEDIA_ENT_F_PROC_VIDEO_LUT (MEDIA_ENT_F_BASE + 0x4004) +#define MEDIA_ENT_F_PROC_VIDEO_SCALER (MEDIA_ENT_F_BASE + 0x4005) + /* * Connectors */ -- cgit v1.2.3 From eaa0b96bbb6586a95e66d3dedd974313d7b0d94f Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Tue, 1 Mar 2016 21:12:27 -0300 Subject: [media] media: Add video statistics computation functions The video statistics function describes entities such as video histogram engines or 3A statistics engines. Signed-off-by: Laurent Pinchart Acked-by: Hans Verkuil Acked-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- Documentation/DocBook/media/v4l/media-types.xml | 9 +++++++++ include/uapi/linux/media.h | 1 + 2 files changed, 10 insertions(+) (limited to 'include/uapi') diff --git a/Documentation/DocBook/media/v4l/media-types.xml b/Documentation/DocBook/media/v4l/media-types.xml index 60fe841f8846..95aa1f9c836a 100644 --- a/Documentation/DocBook/media/v4l/media-types.xml +++ b/Documentation/DocBook/media/v4l/media-types.xml @@ -176,6 +176,15 @@ skipping are considered as scaling. + + MEDIA_ENT_F_PROC_VIDEO_STATISTICS + Video statistics computation (histogram, 3A, ...). An entity + capable of statistics computation must have one sink pad and + one source pad. It computes statistics over the frames + received on its sink pad and outputs the statistics data on + its source pad. + + diff --git a/include/uapi/linux/media.h b/include/uapi/linux/media.h index 3136686c4bd0..7acf0f634f70 100644 --- a/include/uapi/linux/media.h +++ b/include/uapi/linux/media.h @@ -102,6 +102,7 @@ struct media_device_info { #define MEDIA_ENT_F_PROC_VIDEO_PIXEL_ENC_CONV (MEDIA_ENT_F_BASE + 0x4003) #define MEDIA_ENT_F_PROC_VIDEO_LUT (MEDIA_ENT_F_BASE + 0x4004) #define MEDIA_ENT_F_PROC_VIDEO_SCALER (MEDIA_ENT_F_BASE + 0x4005) +#define MEDIA_ENT_F_PROC_VIDEO_STATISTICS (MEDIA_ENT_F_BASE + 0x4006) /* * Connectors -- cgit v1.2.3 From 42e89bed23000c0bf985a741422ef943df0ee1e9 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Fri, 13 May 2016 13:52:11 -0300 Subject: [media] v4l: vsp1: lut: Expose configuration through a control Replace the custom ioctl with a V4L2 control in order to standardize the API. Signed-off-by: Laurent Pinchart Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/vsp1/vsp1_lut.c | 76 +++++++++++++++++++++++----------- drivers/media/platform/vsp1/vsp1_lut.h | 6 +-- include/uapi/linux/vsp1.h | 34 --------------- 3 files changed, 54 insertions(+), 62 deletions(-) delete mode 100644 include/uapi/linux/vsp1.h (limited to 'include/uapi') diff --git a/drivers/media/platform/vsp1/vsp1_lut.c b/drivers/media/platform/vsp1/vsp1_lut.c index 4a4724965de1..db8f01dbab84 100644 --- a/drivers/media/platform/vsp1/vsp1_lut.c +++ b/drivers/media/platform/vsp1/vsp1_lut.c @@ -13,7 +13,6 @@ #include #include -#include #include @@ -35,43 +34,60 @@ static inline void vsp1_lut_write(struct vsp1_lut *lut, struct vsp1_dl_list *dl, } /* ----------------------------------------------------------------------------- - * V4L2 Subdevice Core Operations + * Controls */ -static int lut_set_table(struct vsp1_lut *lut, struct vsp1_lut_config *config) +#define V4L2_CID_VSP1_LUT_TABLE (V4L2_CID_USER_BASE | 0x1001) + +static int lut_set_table(struct vsp1_lut *lut, struct v4l2_ctrl *ctrl) { struct vsp1_dl_body *dlb; unsigned int i; - dlb = vsp1_dl_fragment_alloc(lut->entity.vsp1, ARRAY_SIZE(config->lut)); + dlb = vsp1_dl_fragment_alloc(lut->entity.vsp1, 256); if (!dlb) return -ENOMEM; - for (i = 0; i < ARRAY_SIZE(config->lut); ++i) + for (i = 0; i < 256; ++i) vsp1_dl_fragment_write(dlb, VI6_LUT_TABLE + 4 * i, - config->lut[i]); + ctrl->p_new.p_u32[i]); - mutex_lock(&lut->lock); swap(lut->lut, dlb); - mutex_unlock(&lut->lock); vsp1_dl_fragment_free(dlb); return 0; } -static long lut_ioctl(struct v4l2_subdev *subdev, unsigned int cmd, void *arg) +static int lut_s_ctrl(struct v4l2_ctrl *ctrl) { - struct vsp1_lut *lut = to_lut(subdev); - - switch (cmd) { - case VIDIOC_VSP1_LUT_CONFIG: - return lut_set_table(lut, arg); + struct vsp1_lut *lut = + container_of(ctrl->handler, struct vsp1_lut, ctrls); - default: - return -ENOIOCTLCMD; + switch (ctrl->id) { + case V4L2_CID_VSP1_LUT_TABLE: + lut_set_table(lut, ctrl); + break; } + + return 0; } +static const struct v4l2_ctrl_ops lut_ctrl_ops = { + .s_ctrl = lut_s_ctrl, +}; + +static const struct v4l2_ctrl_config lut_table_control = { + .ops = &lut_ctrl_ops, + .id = V4L2_CID_VSP1_LUT_TABLE, + .name = "Look-Up Table", + .type = V4L2_CTRL_TYPE_U32, + .min = 0x00000000, + .max = 0x00ffffff, + .step = 1, + .def = 0, + .dims = { 256}, +}; + /* ----------------------------------------------------------------------------- * V4L2 Subdevice Pad Operations */ @@ -147,10 +163,6 @@ static int lut_set_format(struct v4l2_subdev *subdev, * V4L2 Subdevice Operations */ -static const struct v4l2_subdev_core_ops lut_core_ops = { - .ioctl = lut_ioctl, -}; - static const struct v4l2_subdev_pad_ops lut_pad_ops = { .init_cfg = vsp1_entity_init_cfg, .enum_mbus_code = lut_enum_mbus_code, @@ -160,7 +172,6 @@ static const struct v4l2_subdev_pad_ops lut_pad_ops = { }; static const struct v4l2_subdev_ops lut_ops = { - .core = &lut_core_ops, .pad = &lut_pad_ops, }; @@ -176,12 +187,14 @@ static void lut_configure(struct vsp1_entity *entity, vsp1_lut_write(lut, dl, VI6_LUT_CTRL, VI6_LUT_CTRL_EN); - mutex_lock(&lut->lock); + mutex_lock(lut->ctrls.lock); + if (lut->lut) { vsp1_dl_list_add_fragment(dl, lut->lut); lut->lut = NULL; } - mutex_unlock(&lut->lock); + + mutex_unlock(lut->ctrls.lock); } static const struct vsp1_entity_operations lut_entity_ops = { @@ -201,8 +214,6 @@ struct vsp1_lut *vsp1_lut_create(struct vsp1_device *vsp1) if (lut == NULL) return ERR_PTR(-ENOMEM); - mutex_init(&lut->lock); - lut->entity.ops = &lut_entity_ops; lut->entity.type = VSP1_ENTITY_LUT; @@ -211,5 +222,20 @@ struct vsp1_lut *vsp1_lut_create(struct vsp1_device *vsp1) if (ret < 0) return ERR_PTR(ret); + /* Initialize the control handler. */ + v4l2_ctrl_handler_init(&lut->ctrls, 1); + v4l2_ctrl_new_custom(&lut->ctrls, &lut_table_control, NULL); + + lut->entity.subdev.ctrl_handler = &lut->ctrls; + + if (lut->ctrls.error) { + dev_err(vsp1->dev, "lut: failed to initialize controls\n"); + ret = lut->ctrls.error; + vsp1_entity_destroy(&lut->entity); + return ERR_PTR(ret); + } + + v4l2_ctrl_handler_setup(&lut->ctrls); + return lut; } diff --git a/drivers/media/platform/vsp1/vsp1_lut.h b/drivers/media/platform/vsp1/vsp1_lut.h index cef874f22b6a..021898fc0ce5 100644 --- a/drivers/media/platform/vsp1/vsp1_lut.h +++ b/drivers/media/platform/vsp1/vsp1_lut.h @@ -13,9 +13,8 @@ #ifndef __VSP1_LUT_H__ #define __VSP1_LUT_H__ -#include - #include +#include #include #include "vsp1_entity.h" @@ -28,7 +27,8 @@ struct vsp1_device; struct vsp1_lut { struct vsp1_entity entity; - struct mutex lock; + struct v4l2_ctrl_handler ctrls; + struct vsp1_dl_body *lut; }; diff --git a/include/uapi/linux/vsp1.h b/include/uapi/linux/vsp1.h deleted file mode 100644 index 9a823696d816..000000000000 --- a/include/uapi/linux/vsp1.h +++ /dev/null @@ -1,34 +0,0 @@ -/* - * vsp1.h - * - * Renesas R-Car VSP1 - User-space API - * - * Copyright (C) 2013 Renesas Corporation - * - * Contacts: Laurent Pinchart - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#ifndef __VSP1_USER_H__ -#define __VSP1_USER_H__ - -#include -#include - -/* - * Private IOCTLs - * - * VIDIOC_VSP1_LUT_CONFIG - Configure the lookup table - */ - -#define VIDIOC_VSP1_LUT_CONFIG \ - _IOWR('V', BASE_VIDIOC_PRIVATE + 1, struct vsp1_lut_config) - -struct vsp1_lut_config { - __u32 lut[256]; -}; - -#endif /* __VSP1_USER_H__ */ -- cgit v1.2.3 From 6816a7ffce32e999601825ddfd887f36d3052932 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:25 +0200 Subject: bpf, trace: add BPF_F_CURRENT_CPU flag for bpf_perf_event_read Follow-up commit to 1e33759c788c ("bpf, trace: add BPF_F_CURRENT_CPU flag for bpf_perf_event_output") to add the same functionality into bpf_perf_event_read() helper. The split of index into flags and index component is also safe here, since such large maps are rejected during map allocation time. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 2 +- kernel/trace/bpf_trace.c | 11 ++++++++--- 2 files changed, 9 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 406459b935a2..58df2da3e9bf 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -347,7 +347,7 @@ enum bpf_func_id { #define BPF_F_ZERO_CSUM_TX (1ULL << 1) #define BPF_F_DONT_FRAGMENT (1ULL << 2) -/* BPF_FUNC_perf_event_output flags. */ +/* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */ #define BPF_F_INDEX_MASK 0xffffffffULL #define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 505f9e9cdb3b..19c5b4a5c3eb 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -188,13 +188,19 @@ const struct bpf_func_proto *bpf_get_trace_printk_proto(void) return &bpf_trace_printk_proto; } -static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) +static u64 bpf_perf_event_read(u64 r1, u64 flags, u64 r3, u64 r4, u64 r5) { struct bpf_map *map = (struct bpf_map *) (unsigned long) r1; struct bpf_array *array = container_of(map, struct bpf_array, map); + unsigned int cpu = smp_processor_id(); + u64 index = flags & BPF_F_INDEX_MASK; struct bpf_event_entry *ee; struct perf_event *event; + if (unlikely(flags & ~(BPF_F_INDEX_MASK))) + return -EINVAL; + if (index == BPF_F_CURRENT_CPU) + index = cpu; if (unlikely(index >= array->map.max_entries)) return -E2BIG; @@ -208,8 +214,7 @@ static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) return -EINVAL; /* make sure event is local and doesn't have pmu::count */ - if (event->oncpu != smp_processor_id() || - event->pmu->count) + if (unlikely(event->oncpu != cpu || event->pmu->count)) return -EINVAL; /* -- cgit v1.2.3 From 6578171a7ff0c31dc73258f93da7407510abf085 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:27 +0200 Subject: bpf: add bpf_skb_change_proto helper This patch adds a minimal helper for doing the groundwork of changing the skb->protocol in a controlled way. Currently supported is v4 to v6 and vice versa transitions, which allows f.e. for a minimal, static nat64 implementation where applications in containers that still require IPv4 can be transparently operated in an IPv6-only environment. For example, host facing veth of the container can transparently do the transitions in a programmatic way with the help of clsact qdisc and cls_bpf. Idea is to separate concerns for keeping complexity of the helper lower, which means that the programs utilize bpf_skb_change_proto(), bpf_skb_store_bytes() and bpf_lX_csum_replace() to get the job done, instead of doing everything in a single helper (and thus partially duplicating helper functionality). Also, bpf_skb_change_proto() shouldn't need to deal with raw packet data as this is done by other helpers. bpf_skb_proto_6_to_4() and bpf_skb_proto_4_to_6() unclone the skb to operate on a private one, push or pop additionally required header space and migrate the gso/gro meta data from the shared info. We do mark the gso type as dodgy so that headers are checked and segs recalculated by the gso/gro engine. The gso_size target is adapted as well. The flags argument added is currently reserved and can be used for future extensions. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 14 ++++ net/core/filter.c | 200 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 214 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 58df2da3e9bf..66cd738a937a 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -313,6 +313,20 @@ enum bpf_func_id { */ BPF_FUNC_skb_get_tunnel_opt, BPF_FUNC_skb_set_tunnel_opt, + + /** + * bpf_skb_change_proto(skb, proto, flags) + * Change protocol of the skb. Currently supported is + * v4 -> v6, v6 -> v4 transitions. The helper will also + * resize the skb. eBPF program is expected to fill the + * new headers via skb_store_bytes and lX_csum_replace. + * @skb: pointer to skb + * @proto: new skb->protocol type + * @flags: reserved + * Return: 0 on success or negative error + */ + BPF_FUNC_skb_change_proto, + __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index 46c88d9cec5c..d983e765787a 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1783,6 +1783,202 @@ const struct bpf_func_proto bpf_skb_vlan_pop_proto = { }; EXPORT_SYMBOL_GPL(bpf_skb_vlan_pop_proto); +static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len) +{ + /* Caller already did skb_cow() with len as headroom, + * so no need to do it here. + */ + skb_push(skb, len); + memmove(skb->data, skb->data + len, off); + memset(skb->data + off, 0, len); + + /* No skb_postpush_rcsum(skb, skb->data + off, len) + * needed here as it does not change the skb->csum + * result for checksum complete when summing over + * zeroed blocks. + */ + return 0; +} + +static int bpf_skb_generic_pop(struct sk_buff *skb, u32 off, u32 len) +{ + /* skb_ensure_writable() is not needed here, as we're + * already working on an uncloned skb. + */ + if (unlikely(!pskb_may_pull(skb, off + len))) + return -ENOMEM; + + skb_postpull_rcsum(skb, skb->data + off, len); + memmove(skb->data + len, skb->data, off); + __skb_pull(skb, len); + + return 0; +} + +static int bpf_skb_net_hdr_push(struct sk_buff *skb, u32 off, u32 len) +{ + bool trans_same = skb->transport_header == skb->network_header; + int ret; + + /* There's no need for __skb_push()/__skb_pull() pair to + * get to the start of the mac header as we're guaranteed + * to always start from here under eBPF. + */ + ret = bpf_skb_generic_push(skb, off, len); + if (likely(!ret)) { + skb->mac_header -= len; + skb->network_header -= len; + if (trans_same) + skb->transport_header = skb->network_header; + } + + return ret; +} + +static int bpf_skb_net_hdr_pop(struct sk_buff *skb, u32 off, u32 len) +{ + bool trans_same = skb->transport_header == skb->network_header; + int ret; + + /* Same here, __skb_push()/__skb_pull() pair not needed. */ + ret = bpf_skb_generic_pop(skb, off, len); + if (likely(!ret)) { + skb->mac_header += len; + skb->network_header += len; + if (trans_same) + skb->transport_header = skb->network_header; + } + + return ret; +} + +static int bpf_skb_proto_4_to_6(struct sk_buff *skb) +{ + const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr); + u32 off = skb->network_header - skb->mac_header; + int ret; + + ret = skb_cow(skb, len_diff); + if (unlikely(ret < 0)) + return ret; + + ret = bpf_skb_net_hdr_push(skb, off, len_diff); + if (unlikely(ret < 0)) + return ret; + + if (skb_is_gso(skb)) { + /* SKB_GSO_UDP stays as is. SKB_GSO_TCPV4 needs to + * be changed into SKB_GSO_TCPV6. + */ + if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4) { + skb_shinfo(skb)->gso_type &= ~SKB_GSO_TCPV4; + skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV6; + } + + /* Due to IPv6 header, MSS needs to be downgraded. */ + skb_shinfo(skb)->gso_size -= len_diff; + /* Header must be checked, and gso_segs recomputed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + } + + skb->protocol = htons(ETH_P_IPV6); + skb_clear_hash(skb); + + return 0; +} + +static int bpf_skb_proto_6_to_4(struct sk_buff *skb) +{ + const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr); + u32 off = skb->network_header - skb->mac_header; + int ret; + + ret = skb_unclone(skb, GFP_ATOMIC); + if (unlikely(ret < 0)) + return ret; + + ret = bpf_skb_net_hdr_pop(skb, off, len_diff); + if (unlikely(ret < 0)) + return ret; + + if (skb_is_gso(skb)) { + /* SKB_GSO_UDP stays as is. SKB_GSO_TCPV6 needs to + * be changed into SKB_GSO_TCPV4. + */ + if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) { + skb_shinfo(skb)->gso_type &= ~SKB_GSO_TCPV6; + skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV4; + } + + /* Due to IPv4 header, MSS can be upgraded. */ + skb_shinfo(skb)->gso_size += len_diff; + /* Header must be checked, and gso_segs recomputed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + } + + skb->protocol = htons(ETH_P_IP); + skb_clear_hash(skb); + + return 0; +} + +static int bpf_skb_proto_xlat(struct sk_buff *skb, __be16 to_proto) +{ + __be16 from_proto = skb->protocol; + + if (from_proto == htons(ETH_P_IP) && + to_proto == htons(ETH_P_IPV6)) + return bpf_skb_proto_4_to_6(skb); + + if (from_proto == htons(ETH_P_IPV6) && + to_proto == htons(ETH_P_IP)) + return bpf_skb_proto_6_to_4(skb); + + return -ENOTSUPP; +} + +static u64 bpf_skb_change_proto(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5) +{ + struct sk_buff *skb = (struct sk_buff *) (long) r1; + __be16 proto = (__force __be16) r2; + int ret; + + if (unlikely(flags)) + return -EINVAL; + + /* General idea is that this helper does the basic groundwork + * needed for changing the protocol, and eBPF program fills the + * rest through bpf_skb_store_bytes(), bpf_lX_csum_replace() + * and other helpers, rather than passing a raw buffer here. + * + * The rationale is to keep this minimal and without a need to + * deal with raw packet data. F.e. even if we would pass buffers + * here, the program still needs to call the bpf_lX_csum_replace() + * helpers anyway. Plus, this way we keep also separation of + * concerns, since f.e. bpf_skb_store_bytes() should only take + * care of stores. + * + * Currently, additional options and extension header space are + * not supported, but flags register is reserved so we can adapt + * that. For offloads, we mark packet as dodgy, so that headers + * need to be verified first. + */ + ret = bpf_skb_proto_xlat(skb, proto); + bpf_compute_data_end(skb); + return ret; +} + +static const struct bpf_func_proto bpf_skb_change_proto_proto = { + .func = bpf_skb_change_proto, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, +}; + bool bpf_helper_changes_skb_data(void *func) { if (func == bpf_skb_vlan_push) @@ -1791,6 +1987,8 @@ bool bpf_helper_changes_skb_data(void *func) return true; if (func == bpf_skb_store_bytes) return true; + if (func == bpf_skb_change_proto) + return true; if (func == bpf_l3_csum_replace) return true; if (func == bpf_l4_csum_replace) @@ -2078,6 +2276,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_skb_vlan_push_proto; case BPF_FUNC_skb_vlan_pop: return &bpf_skb_vlan_pop_proto; + case BPF_FUNC_skb_change_proto: + return &bpf_skb_change_proto_proto; case BPF_FUNC_skb_get_tunnel_key: return &bpf_skb_get_tunnel_key_proto; case BPF_FUNC_skb_set_tunnel_key: -- cgit v1.2.3 From d2485c4242a826fdf493fd3a27b8b792965b9b9e Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:28 +0200 Subject: bpf: add bpf_skb_change_type helper This work adds a helper for changing skb->pkt_type in a controlled way. We only allow a subset of possible values and can extend that in future should other use cases come up. Doing this as a helper has the advantage that errors can be handeled gracefully and thus helper kept extensible. It's a write counterpart to pkt_type member we can already read from struct __sk_buff context. Major use case is to change incoming skbs to PACKET_HOST in a programmatic way instead of having to recirculate via redirect(..., BPF_F_INGRESS), for example. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 9 +++++++++ net/core/filter.c | 24 ++++++++++++++++++++++++ 2 files changed, 33 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 66cd738a937a..be6ac1291680 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -327,6 +327,15 @@ enum bpf_func_id { */ BPF_FUNC_skb_change_proto, + /** + * bpf_skb_change_type(skb, type) + * Change packet type of skb. + * @skb: pointer to skb + * @type: new skb->pkt_type type + * Return: 0 on success or negative error + */ + BPF_FUNC_skb_change_type, + __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index d983e765787a..76f9a4938be4 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1979,6 +1979,28 @@ static const struct bpf_func_proto bpf_skb_change_proto_proto = { .arg3_type = ARG_ANYTHING, }; +static u64 bpf_skb_change_type(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + struct sk_buff *skb = (struct sk_buff *) (long) r1; + u32 pkt_type = r2; + + /* We only allow a restricted subset to be changed for now. */ + if (unlikely(skb->pkt_type > PACKET_OTHERHOST || + pkt_type > PACKET_OTHERHOST)) + return -EINVAL; + + skb->pkt_type = pkt_type; + return 0; +} + +static const struct bpf_func_proto bpf_skb_change_type_proto = { + .func = bpf_skb_change_type, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, +}; + bool bpf_helper_changes_skb_data(void *func) { if (func == bpf_skb_vlan_push) @@ -2278,6 +2300,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_skb_vlan_pop_proto; case BPF_FUNC_skb_change_proto: return &bpf_skb_change_proto_proto; + case BPF_FUNC_skb_change_type: + return &bpf_skb_change_type_proto; case BPF_FUNC_skb_get_tunnel_key: return &bpf_skb_get_tunnel_key_proto; case BPF_FUNC_skb_set_tunnel_key: -- cgit v1.2.3 From 80e73cc563c4359be809a03bcb8e7e28141a813a Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jun 2016 16:57:05 +0200 Subject: net: rtnetlink: add support for the IFLA_STATS_LINK_XSTATS_SLAVE attribute This patch adds support for the IFLA_STATS_LINK_XSTATS_SLAVE attribute which allows to export per-slave statistics if the master device supports the linkxstats callback. The attribute is passed down to the linkxstats callback and it is up to the callback user to use it (an example has been added to the only current user - the bridge). This allows us to query only specific slaves of master devices like bridge ports and export only what we're interested in instead of having to dump all ports and searching only for a single one. This will be used to export per-port IGMP/MLD stats and also per-port vlan stats in the future, possibly other statistics as well. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 5 ++-- include/uapi/linux/if_link.h | 1 + net/bridge/br_netlink.c | 59 +++++++++++++++++++++++++++++++++++++++++--- net/core/rtnetlink.c | 50 +++++++++++++++++++++++++++++++++++-- 4 files changed, 108 insertions(+), 7 deletions(-) (limited to 'include/uapi') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 006a7b81d758..4113916cc1bb 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -98,10 +98,11 @@ struct rtnl_link_ops { const struct net_device *dev, const struct net_device *slave_dev); struct net *(*get_link_net)(const struct net_device *dev); - size_t (*get_linkxstats_size)(const struct net_device *dev); + size_t (*get_linkxstats_size)(const struct net_device *dev, + int attr); int (*fill_linkxstats)(struct sk_buff *skb, const struct net_device *dev, - int *prividx); + int *prividx, int attr); }; int __rtnl_link_register(struct rtnl_link_ops *ops); diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index bb36bd5675a7..db2458ade81c 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -822,6 +822,7 @@ enum { IFLA_STATS_UNSPEC, /* also used as 64bit pad attribute */ IFLA_STATS_LINK_64, IFLA_STATS_LINK_XSTATS, + IFLA_STATS_LINK_XSTATS_SLAVE, __IFLA_STATS_MAX, }; diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 85e89f693589..ed75ff9ff9e6 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1234,7 +1234,7 @@ static int br_fill_info(struct sk_buff *skb, const struct net_device *brdev) return 0; } -static size_t br_get_linkxstats_size(const struct net_device *dev) +static size_t bridge_get_linkxstats_size(const struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); struct net_bridge_vlan_group *vg; @@ -1254,8 +1254,30 @@ static size_t br_get_linkxstats_size(const struct net_device *dev) nla_total_size(0); } -static int br_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, - int *prividx) +static size_t brport_get_linkxstats_size(const struct net_device *dev) +{ + return nla_total_size(0); +} + +static size_t br_get_linkxstats_size(const struct net_device *dev, int attr) +{ + size_t retsize = 0; + + switch (attr) { + case IFLA_STATS_LINK_XSTATS: + retsize = bridge_get_linkxstats_size(dev); + break; + case IFLA_STATS_LINK_XSTATS_SLAVE: + retsize = brport_get_linkxstats_size(dev); + break; + } + + return retsize; +} + +static int bridge_fill_linkxstats(struct sk_buff *skb, + const struct net_device *dev, + int *prividx) { struct net_bridge *br = netdev_priv(dev); struct net_bridge_vlan_group *vg; @@ -1298,6 +1320,37 @@ nla_put_failure: return -EMSGSIZE; } +static int brport_fill_linkxstats(struct sk_buff *skb, + const struct net_device *dev, + int *prividx) +{ + struct nlattr *nest; + + nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); + if (!nest) + return -EMSGSIZE; + nla_nest_end(skb, nest); + + return 0; +} + +static int br_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, + int *prividx, int attr) +{ + int ret = -EINVAL; + + switch (attr) { + case IFLA_STATS_LINK_XSTATS: + ret = bridge_fill_linkxstats(skb, dev, prividx); + break; + case IFLA_STATS_LINK_XSTATS_SLAVE: + ret = brport_fill_linkxstats(skb, dev, prividx); + break; + } + + return ret; +} + static struct rtnl_af_ops br_af_ops __read_mostly = { .family = AF_BRIDGE, .get_link_af_size = br_get_link_af_size_filtered, diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index eb49ca24274a..cfed7bc14ee6 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -3519,7 +3519,32 @@ static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev, if (!attr) goto nla_put_failure; - err = ops->fill_linkxstats(skb, dev, prividx); + err = ops->fill_linkxstats(skb, dev, prividx, *idxattr); + nla_nest_end(skb, attr); + if (err) + goto nla_put_failure; + *idxattr = 0; + } + } + + if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS_SLAVE, + *idxattr)) { + const struct rtnl_link_ops *ops = NULL; + const struct net_device *master; + + master = netdev_master_upper_dev_get(dev); + if (master) + ops = master->rtnl_link_ops; + if (ops && ops->fill_linkxstats) { + int err; + + *idxattr = IFLA_STATS_LINK_XSTATS_SLAVE; + attr = nla_nest_start(skb, + IFLA_STATS_LINK_XSTATS_SLAVE); + if (!attr) + goto nla_put_failure; + + err = ops->fill_linkxstats(skb, dev, prividx, *idxattr); nla_nest_end(skb, attr); if (err) goto nla_put_failure; @@ -3555,14 +3580,35 @@ static size_t if_nlmsg_stats_size(const struct net_device *dev, if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, 0)) { const struct rtnl_link_ops *ops = dev->rtnl_link_ops; + int attr = IFLA_STATS_LINK_XSTATS; if (ops && ops->get_linkxstats_size) { - size += nla_total_size(ops->get_linkxstats_size(dev)); + size += nla_total_size(ops->get_linkxstats_size(dev, + attr)); /* for IFLA_STATS_LINK_XSTATS */ size += nla_total_size(0); } } + if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS_SLAVE, 0)) { + struct net_device *_dev = (struct net_device *)dev; + const struct rtnl_link_ops *ops = NULL; + const struct net_device *master; + + /* netdev_master_upper_dev_get can't take const */ + master = netdev_master_upper_dev_get(_dev); + if (master) + ops = master->rtnl_link_ops; + if (ops && ops->get_linkxstats_size) { + int attr = IFLA_STATS_LINK_XSTATS_SLAVE; + + size += nla_total_size(ops->get_linkxstats_size(dev, + attr)); + /* for IFLA_STATS_LINK_XSTATS_SLAVE */ + size += nla_total_size(0); + } + } + return size; } -- cgit v1.2.3 From 1080ab95e3c7bdd77870e209aff83c763fdcf439 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jun 2016 16:57:06 +0200 Subject: net: bridge: add support for IGMP/MLD stats and export them via netlink This patch adds stats support for the currently used IGMP/MLD types by the bridge. The stats are per-port (plus one stat per-bridge) and per-direction (RX/TX). The stats are exported via netlink via the new linkxstats API (RTM_GETSTATS). In order to minimize the performance impact, a new option is used to enable/disable the stats - multicast_stats_enabled, similar to the recent vlan stats. Also in order to avoid multiple IGMP/MLD type lookups and checks, we make use of the current "igmp" member of the bridge private skb->cb region to record the type on Rx (both host-generated and external packets pass by multicast_rcv()). We can do that since the igmp member was used as a boolean and all the valid IGMP/MLD types are positive values. The normal bridge fast-path is not affected at all, the only affected paths are the flooding ones and since we make use of the IGMP/MLD type, we can quickly determine if the packet should be counted using cache-hot data (cb's igmp member). We add counters for: * IGMP Queries * IGMP Leaves * IGMP v1/v2/v3 reports * MLD Queries * MLD Leaves * MLD v1/v2 reports These are invaluable when monitoring or debugging complex multicast setups with bridges. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/uapi/linux/if_bridge.h | 26 +++++ include/uapi/linux/if_link.h | 1 + net/bridge/br_device.c | 10 +- net/bridge/br_forward.c | 13 ++- net/bridge/br_if.c | 9 +- net/bridge/br_input.c | 3 + net/bridge/br_multicast.c | 217 ++++++++++++++++++++++++++++++++++++++--- net/bridge/br_netlink.c | 91 ++++++++++++----- net/bridge/br_private.h | 41 +++++++- net/bridge/br_sysfs_br.c | 25 +++++ 10 files changed, 390 insertions(+), 46 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h index 397d503fdedb..8304fe6f0561 100644 --- a/include/uapi/linux/if_bridge.h +++ b/include/uapi/linux/if_bridge.h @@ -247,8 +247,34 @@ enum { enum { BRIDGE_XSTATS_UNSPEC, BRIDGE_XSTATS_VLAN, + BRIDGE_XSTATS_MCAST, + BRIDGE_XSTATS_PAD, __BRIDGE_XSTATS_MAX }; #define BRIDGE_XSTATS_MAX (__BRIDGE_XSTATS_MAX - 1) +enum { + BR_MCAST_DIR_RX, + BR_MCAST_DIR_TX, + BR_MCAST_DIR_SIZE +}; + +/* IGMP/MLD statistics */ +struct br_mcast_stats { + __u64 igmp_queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_leaves[BR_MCAST_DIR_SIZE]; + __u64 igmp_v1reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_v2reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_v3reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_parse_errors; + + __u64 mld_queries[BR_MCAST_DIR_SIZE]; + __u64 mld_leaves[BR_MCAST_DIR_SIZE]; + __u64 mld_v1reports[BR_MCAST_DIR_SIZE]; + __u64 mld_v2reports[BR_MCAST_DIR_SIZE]; + __u64 mld_parse_errors; + + __u64 mcast_bytes[BR_MCAST_DIR_SIZE]; + __u64 mcast_packets[BR_MCAST_DIR_SIZE]; +}; #endif /* _UAPI_LINUX_IF_BRIDGE_H */ diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index db2458ade81c..4285ac31e865 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -273,6 +273,7 @@ enum { IFLA_BR_VLAN_DEFAULT_PVID, IFLA_BR_PAD, IFLA_BR_VLAN_STATS_ENABLED, + IFLA_BR_MCAST_STATS_ENABLED, __IFLA_BR_MAX, }; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 2c8095a5d824..0c39e0f6da09 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -104,8 +104,16 @@ static int br_dev_init(struct net_device *dev) return -ENOMEM; err = br_vlan_init(br); - if (err) + if (err) { free_percpu(br->stats); + return err; + } + + err = br_multicast_init_stats(br); + if (err) { + free_percpu(br->stats); + br_vlan_flush(br); + } br_set_lockdep_class(dev); return err; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index f47759f05b6d..6c196037d818 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -198,8 +198,10 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, struct sk_buff *skb), bool unicast) { - struct net_bridge_port *p; + u8 igmp_type = br_multicast_igmp_type(skb); + __be16 proto = skb->protocol; struct net_bridge_port *prev; + struct net_bridge_port *p; prev = NULL; @@ -218,6 +220,9 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, prev = maybe_deliver(prev, p, skb, __packet_hook); if (IS_ERR(prev)) goto out; + if (prev == p) + br_multicast_count(p->br, p, proto, igmp_type, + BR_MCAST_DIR_TX); } if (!prev) @@ -257,9 +262,12 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, struct sk_buff *skb)) { struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; + u8 igmp_type = br_multicast_igmp_type(skb); struct net_bridge *br = netdev_priv(dev); struct net_bridge_port *prev = NULL; struct net_bridge_port_group *p; + __be16 proto = skb->protocol; + struct hlist_node *rp; rp = rcu_dereference(hlist_first_rcu(&br->router_list)); @@ -277,6 +285,9 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, prev = maybe_deliver(prev, port, skb, __packet_hook); if (IS_ERR(prev)) goto out; + if (prev == port) + br_multicast_count(port->br, port, proto, igmp_type, + BR_MCAST_DIR_TX); if ((unsigned long)lport >= (unsigned long)port) p = rcu_dereference(p->next); diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 8217aecf025b..f2fede05d32c 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -345,8 +345,8 @@ static int find_portno(struct net_bridge *br) static struct net_bridge_port *new_nbp(struct net_bridge *br, struct net_device *dev) { - int index; struct net_bridge_port *p; + int index, err; index = find_portno(br); if (index < 0) @@ -366,7 +366,12 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br, br_init_port(p); br_set_state(p, BR_STATE_DISABLED); br_stp_port_timer_init(p); - br_multicast_add_port(p); + err = br_multicast_add_port(p); + if (err) { + dev_put(dev); + kfree(p); + p = ERR_PTR(err); + } return p; } diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 43d2cd862bc2..786602bc0567 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -60,6 +60,9 @@ static int br_pass_frame_up(struct sk_buff *skb) skb = br_handle_vlan(br, vg, skb); if (!skb) return NET_RX_DROP; + /* update the multicast stats if the packet is IGMP/MLD */ + br_multicast_count(br, NULL, skb->protocol, br_multicast_igmp_type(skb), + BR_MCAST_DIR_TX); return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, dev_net(indev), NULL, skb, indev, NULL, diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 43844144c9c4..e405eef0ae2e 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -361,7 +361,8 @@ out: } static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br, - __be32 group) + __be32 group, + u8 *igmp_type) { struct sk_buff *skb; struct igmphdr *ih; @@ -411,6 +412,7 @@ static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br, skb_set_transport_header(skb, skb->len); ih = igmp_hdr(skb); + *igmp_type = IGMP_HOST_MEMBERSHIP_QUERY; ih->type = IGMP_HOST_MEMBERSHIP_QUERY; ih->code = (group ? br->multicast_last_member_interval : br->multicast_query_response_interval) / @@ -428,7 +430,8 @@ out: #if IS_ENABLED(CONFIG_IPV6) static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br, - const struct in6_addr *group) + const struct in6_addr *grp, + u8 *igmp_type) { struct sk_buff *skb; struct ipv6hdr *ip6h; @@ -487,16 +490,17 @@ static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br, skb_set_transport_header(skb, skb->len); mldq = (struct mld_msg *) icmp6_hdr(skb); - interval = ipv6_addr_any(group) ? + interval = ipv6_addr_any(grp) ? br->multicast_query_response_interval : br->multicast_last_member_interval; + *igmp_type = ICMPV6_MGM_QUERY; mldq->mld_type = ICMPV6_MGM_QUERY; mldq->mld_code = 0; mldq->mld_cksum = 0; mldq->mld_maxdelay = htons((u16)jiffies_to_msecs(interval)); mldq->mld_reserved = 0; - mldq->mld_mca = *group; + mldq->mld_mca = *grp; /* checksum */ mldq->mld_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, @@ -513,14 +517,16 @@ out: #endif static struct sk_buff *br_multicast_alloc_query(struct net_bridge *br, - struct br_ip *addr) + struct br_ip *addr, + u8 *igmp_type) { switch (addr->proto) { case htons(ETH_P_IP): - return br_ip4_multicast_alloc_query(br, addr->u.ip4); + return br_ip4_multicast_alloc_query(br, addr->u.ip4, igmp_type); #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): - return br_ip6_multicast_alloc_query(br, &addr->u.ip6); + return br_ip6_multicast_alloc_query(br, &addr->u.ip6, + igmp_type); #endif } return NULL; @@ -829,18 +835,23 @@ static void __br_multicast_send_query(struct net_bridge *br, struct br_ip *ip) { struct sk_buff *skb; + u8 igmp_type; - skb = br_multicast_alloc_query(br, ip); + skb = br_multicast_alloc_query(br, ip, &igmp_type); if (!skb) return; if (port) { skb->dev = port->dev; + br_multicast_count(br, port, skb->protocol, igmp_type, + BR_MCAST_DIR_TX); NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, dev_net(port->dev), NULL, skb, NULL, skb->dev, br_dev_queue_push_xmit); } else { br_multicast_select_own_querier(br, ip, skb); + br_multicast_count(br, port, skb->protocol, igmp_type, + BR_MCAST_DIR_RX); netif_rx(skb); } } @@ -918,7 +929,7 @@ static void br_ip6_multicast_port_query_expired(unsigned long data) } #endif -void br_multicast_add_port(struct net_bridge_port *port) +int br_multicast_add_port(struct net_bridge_port *port) { port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY; @@ -930,6 +941,11 @@ void br_multicast_add_port(struct net_bridge_port *port) setup_timer(&port->ip6_own_query.timer, br_ip6_multicast_port_query_expired, (unsigned long)port); #endif + port->mcast_stats = netdev_alloc_pcpu_stats(struct bridge_mcast_stats); + if (!port->mcast_stats) + return -ENOMEM; + + return 0; } void br_multicast_del_port(struct net_bridge_port *port) @@ -944,6 +960,7 @@ void br_multicast_del_port(struct net_bridge_port *port) br_multicast_del_pg(br, pg); spin_unlock_bh(&br->multicast_lock); del_timer_sync(&port->multicast_router_timer); + free_percpu(port->mcast_stats); } static void br_multicast_enable(struct bridge_mcast_own_query *query) @@ -1583,6 +1600,39 @@ static void br_ip6_multicast_leave_group(struct net_bridge *br, } #endif +static void br_multicast_err_count(const struct net_bridge *br, + const struct net_bridge_port *p, + __be16 proto) +{ + struct bridge_mcast_stats __percpu *stats; + struct bridge_mcast_stats *pstats; + + if (!br->multicast_stats_enabled) + return; + + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + pstats = this_cpu_ptr(stats); + + u64_stats_update_begin(&pstats->syncp); + switch (proto) { + case htons(ETH_P_IP): + pstats->mstats.igmp_parse_errors++; + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + pstats->mstats.mld_parse_errors++; + break; +#endif + } + u64_stats_update_end(&pstats->syncp); +} + static int br_multicast_ipv4_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, @@ -1599,11 +1649,12 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, BR_INPUT_SKB_CB(skb)->mrouters_only = 1; return 0; } else if (err < 0) { + br_multicast_err_count(br, port, skb->protocol); return err; } - BR_INPUT_SKB_CB(skb)->igmp = 1; ih = igmp_hdr(skb); + BR_INPUT_SKB_CB(skb)->igmp = ih->type; switch (ih->type) { case IGMP_HOST_MEMBERSHIP_REPORT: @@ -1625,6 +1676,9 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); + br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + BR_MCAST_DIR_RX); + return err; } @@ -1645,11 +1699,12 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, BR_INPUT_SKB_CB(skb)->mrouters_only = 1; return 0; } else if (err < 0) { + br_multicast_err_count(br, port, skb->protocol); return err; } - BR_INPUT_SKB_CB(skb)->igmp = 1; mld = (struct mld_msg *)skb_transport_header(skb); + BR_INPUT_SKB_CB(skb)->igmp = mld->mld_type; switch (mld->mld_type) { case ICMPV6_MGM_REPORT: @@ -1670,6 +1725,9 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); + br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + BR_MCAST_DIR_RX); + return err; } #endif @@ -1677,6 +1735,8 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, u16 vid) { + int ret = 0; + BR_INPUT_SKB_CB(skb)->igmp = 0; BR_INPUT_SKB_CB(skb)->mrouters_only = 0; @@ -1685,14 +1745,16 @@ int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, switch (skb->protocol) { case htons(ETH_P_IP): - return br_multicast_ipv4_rcv(br, port, skb, vid); + ret = br_multicast_ipv4_rcv(br, port, skb, vid); + break; #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): - return br_multicast_ipv6_rcv(br, port, skb, vid); + ret = br_multicast_ipv6_rcv(br, port, skb, vid); + break; #endif } - return 0; + return ret; } static void br_multicast_query_expired(struct net_bridge *br, @@ -1831,6 +1893,8 @@ void br_multicast_dev_del(struct net_bridge *br) out: spin_unlock_bh(&br->multicast_lock); + + free_percpu(br->mcast_stats); } int br_multicast_set_router(struct net_bridge *br, unsigned long val) @@ -2185,3 +2249,128 @@ unlock: return ret; } EXPORT_SYMBOL_GPL(br_multicast_has_querier_adjacent); + +static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, + __be16 proto, u8 type, u8 dir) +{ + struct bridge_mcast_stats *pstats = this_cpu_ptr(stats); + + u64_stats_update_begin(&pstats->syncp); + switch (proto) { + case htons(ETH_P_IP): + switch (type) { + case IGMP_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v1reports[dir]++; + break; + case IGMPV2_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v2reports[dir]++; + break; + case IGMPV3_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v3reports[dir]++; + break; + case IGMP_HOST_MEMBERSHIP_QUERY: + pstats->mstats.igmp_queries[dir]++; + break; + case IGMP_HOST_LEAVE_MESSAGE: + pstats->mstats.igmp_leaves[dir]++; + break; + } + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + switch (type) { + case ICMPV6_MGM_REPORT: + pstats->mstats.mld_v1reports[dir]++; + break; + case ICMPV6_MLD2_REPORT: + pstats->mstats.mld_v2reports[dir]++; + break; + case ICMPV6_MGM_QUERY: + pstats->mstats.mld_queries[dir]++; + break; + case ICMPV6_MGM_REDUCTION: + pstats->mstats.mld_leaves[dir]++; + break; + } + break; +#endif /* CONFIG_IPV6 */ + } + u64_stats_update_end(&pstats->syncp); +} + +void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir) +{ + struct bridge_mcast_stats __percpu *stats; + + /* if multicast_disabled is true then igmp type can't be set */ + if (!type || !br->multicast_stats_enabled) + return; + + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + br_mcast_stats_add(stats, proto, type, dir); +} + +int br_multicast_init_stats(struct net_bridge *br) +{ + br->mcast_stats = netdev_alloc_pcpu_stats(struct bridge_mcast_stats); + if (!br->mcast_stats) + return -ENOMEM; + + return 0; +} + +static void mcast_stats_add_dir(u64 *dst, u64 *src) +{ + dst[BR_MCAST_DIR_RX] += src[BR_MCAST_DIR_RX]; + dst[BR_MCAST_DIR_TX] += src[BR_MCAST_DIR_TX]; +} + +void br_multicast_get_stats(const struct net_bridge *br, + const struct net_bridge_port *p, + struct br_mcast_stats *dest) +{ + struct bridge_mcast_stats __percpu *stats; + struct br_mcast_stats tdst; + int i; + + memset(dest, 0, sizeof(*dest)); + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + memset(&tdst, 0, sizeof(tdst)); + for_each_possible_cpu(i) { + struct bridge_mcast_stats *cpu_stats = per_cpu_ptr(stats, i); + struct br_mcast_stats temp; + unsigned int start; + + do { + start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); + memcpy(&temp, &cpu_stats->mstats, sizeof(temp)); + } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); + + mcast_stats_add_dir(tdst.igmp_queries, temp.igmp_queries); + mcast_stats_add_dir(tdst.igmp_leaves, temp.igmp_leaves); + mcast_stats_add_dir(tdst.igmp_v1reports, temp.igmp_v1reports); + mcast_stats_add_dir(tdst.igmp_v2reports, temp.igmp_v2reports); + mcast_stats_add_dir(tdst.igmp_v3reports, temp.igmp_v3reports); + tdst.igmp_parse_errors += temp.igmp_parse_errors; + + mcast_stats_add_dir(tdst.mld_queries, temp.mld_queries); + mcast_stats_add_dir(tdst.mld_leaves, temp.mld_leaves); + mcast_stats_add_dir(tdst.mld_v1reports, temp.mld_v1reports); + mcast_stats_add_dir(tdst.mld_v2reports, temp.mld_v2reports); + tdst.mld_parse_errors += temp.mld_parse_errors; + } + memcpy(dest, &tdst, sizeof(*dest)); +} diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index ed75ff9ff9e6..f2a29e467e78 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -851,6 +851,7 @@ static const struct nla_policy br_policy[IFLA_BR_MAX + 1] = { [IFLA_BR_NF_CALL_ARPTABLES] = { .type = NLA_U8 }, [IFLA_BR_VLAN_DEFAULT_PVID] = { .type = NLA_U16 }, [IFLA_BR_VLAN_STATS_ENABLED] = { .type = NLA_U8 }, + [IFLA_BR_MCAST_STATS_ENABLED] = { .type = NLA_U8 }, }; static int br_changelink(struct net_device *brdev, struct nlattr *tb[], @@ -1055,6 +1056,13 @@ static int br_changelink(struct net_device *brdev, struct nlattr *tb[], br->multicast_startup_query_interval = clock_t_to_jiffies(val); } + + if (data[IFLA_BR_MCAST_STATS_ENABLED]) { + __u8 mcast_stats; + + mcast_stats = nla_get_u8(data[IFLA_BR_MCAST_STATS_ENABLED]); + br->multicast_stats_enabled = !!mcast_stats; + } #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) if (data[IFLA_BR_NF_CALL_IPTABLES]) { @@ -1110,6 +1118,7 @@ static size_t br_get_size(const struct net_device *brdev) nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_SNOOPING */ nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_QUERY_USE_IFADDR */ nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_QUERIER */ + nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_STATS_ENABLED */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_HASH_ELASTICITY */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_HASH_MAX */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_LAST_MEMBER_CNT */ @@ -1187,6 +1196,8 @@ static int br_fill_info(struct sk_buff *skb, const struct net_device *brdev) nla_put_u8(skb, IFLA_BR_MCAST_QUERY_USE_IFADDR, br->multicast_query_use_ifaddr) || nla_put_u8(skb, IFLA_BR_MCAST_QUERIER, br->multicast_querier) || + nla_put_u8(skb, IFLA_BR_MCAST_STATS_ENABLED, + br->multicast_stats_enabled) || nla_put_u32(skb, IFLA_BR_MCAST_HASH_ELASTICITY, br->hash_elasticity) || nla_put_u32(skb, IFLA_BR_MCAST_HASH_MAX, br->hash_max) || @@ -1242,21 +1253,21 @@ static size_t bridge_get_linkxstats_size(const struct net_device *dev) int numvls = 0; vg = br_vlan_group(br); - if (!vg) - return 0; - - /* we need to count all, even placeholder entries */ - list_for_each_entry(v, &vg->vlan_list, vlist) - numvls++; + if (vg) { + /* we need to count all, even placeholder entries */ + list_for_each_entry(v, &vg->vlan_list, vlist) + numvls++; + } - /* account for the vlans and the link xstats type nest attribute */ return numvls * nla_total_size(sizeof(struct bridge_vlan_xstats)) + + nla_total_size(sizeof(struct br_mcast_stats)) + nla_total_size(0); } static size_t brport_get_linkxstats_size(const struct net_device *dev) { - return nla_total_size(0); + return nla_total_size(sizeof(struct br_mcast_stats)) + + nla_total_size(0); } static size_t br_get_linkxstats_size(const struct net_device *dev, int attr) @@ -1280,37 +1291,50 @@ static int bridge_fill_linkxstats(struct sk_buff *skb, int *prividx) { struct net_bridge *br = netdev_priv(dev); + struct nlattr *nla __maybe_unused; struct net_bridge_vlan_group *vg; struct net_bridge_vlan *v; struct nlattr *nest; int vl_idx = 0; - vg = br_vlan_group(br); - if (!vg) - goto out; nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); if (!nest) return -EMSGSIZE; - list_for_each_entry(v, &vg->vlan_list, vlist) { - struct bridge_vlan_xstats vxi; - struct br_vlan_stats stats; - if (++vl_idx < *prividx) - continue; - memset(&vxi, 0, sizeof(vxi)); - vxi.vid = v->vid; - br_vlan_get_stats(v, &stats); - vxi.rx_bytes = stats.rx_bytes; - vxi.rx_packets = stats.rx_packets; - vxi.tx_bytes = stats.tx_bytes; - vxi.tx_packets = stats.tx_packets; - - if (nla_put(skb, BRIDGE_XSTATS_VLAN, sizeof(vxi), &vxi)) + vg = br_vlan_group(br); + if (vg) { + list_for_each_entry(v, &vg->vlan_list, vlist) { + struct bridge_vlan_xstats vxi; + struct br_vlan_stats stats; + + if (++vl_idx < *prividx) + continue; + memset(&vxi, 0, sizeof(vxi)); + vxi.vid = v->vid; + br_vlan_get_stats(v, &stats); + vxi.rx_bytes = stats.rx_bytes; + vxi.rx_packets = stats.rx_packets; + vxi.tx_bytes = stats.tx_bytes; + vxi.tx_packets = stats.tx_packets; + + if (nla_put(skb, BRIDGE_XSTATS_VLAN, sizeof(vxi), &vxi)) + goto nla_put_failure; + } + } + +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + if (++vl_idx >= *prividx) { + nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_MCAST, + sizeof(struct br_mcast_stats), + BRIDGE_XSTATS_PAD); + if (!nla) goto nla_put_failure; + br_multicast_get_stats(br, NULL, nla_data(nla)); } +#endif nla_nest_end(skb, nest); *prividx = 0; -out: + return 0; nla_put_failure: @@ -1324,11 +1348,26 @@ static int brport_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, int *prividx) { + struct net_bridge_port *p = br_port_get_rtnl(dev); + struct nlattr *nla __maybe_unused; struct nlattr *nest; + if (!p) + return 0; + nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); if (!nest) return -EMSGSIZE; +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_MCAST, + sizeof(struct br_mcast_stats), + BRIDGE_XSTATS_PAD); + if (!nla) { + nla_nest_end(skb, nest); + return -EMSGSIZE; + } + br_multicast_get_stats(p->br, p, nla_data(nla)); +#endif nla_nest_end(skb, nest); return 0; diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 52edecf3c294..4dc851166ad1 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -75,6 +75,12 @@ struct bridge_mcast_querier { struct br_ip addr; struct net_bridge_port __rcu *port; }; + +/* IGMP/MLD statistics */ +struct bridge_mcast_stats { + struct br_mcast_stats mstats; + struct u64_stats_sync syncp; +}; #endif struct br_vlan_stats { @@ -229,6 +235,7 @@ struct net_bridge_port struct bridge_mcast_own_query ip6_own_query; #endif /* IS_ENABLED(CONFIG_IPV6) */ unsigned char multicast_router; + struct bridge_mcast_stats __percpu *mcast_stats; struct timer_list multicast_router_timer; struct hlist_head mglist; struct hlist_node rlist; @@ -315,6 +322,7 @@ struct net_bridge u8 multicast_querier:1; u8 multicast_query_use_ifaddr:1; u8 has_ipv6_addr:1; + u8 multicast_stats_enabled:1; u32 hash_elasticity; u32 hash_max; @@ -337,6 +345,7 @@ struct net_bridge struct bridge_mcast_other_query ip4_other_query; struct bridge_mcast_own_query ip4_own_query; struct bridge_mcast_querier ip4_querier; + struct bridge_mcast_stats __percpu *mcast_stats; #if IS_ENABLED(CONFIG_IPV6) struct bridge_mcast_other_query ip6_other_query; struct bridge_mcast_own_query ip6_own_query; @@ -543,7 +552,7 @@ int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, u16 vid); struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br, struct sk_buff *skb, u16 vid); -void br_multicast_add_port(struct net_bridge_port *port); +int br_multicast_add_port(struct net_bridge_port *port); void br_multicast_del_port(struct net_bridge_port *port); void br_multicast_enable_port(struct net_bridge_port *port); void br_multicast_disable_port(struct net_bridge_port *port); @@ -576,6 +585,12 @@ void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port, struct br_ip *group, int type, u8 flags); void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port, int type); +void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir); +int br_multicast_init_stats(struct net_bridge *br); +void br_multicast_get_stats(const struct net_bridge *br, + const struct net_bridge_port *p, + struct br_mcast_stats *dest); #define mlock_dereference(X, br) \ rcu_dereference_protected(X, lockdep_is_held(&br->multicast_lock)) @@ -623,6 +638,11 @@ static inline bool br_multicast_querier_exists(struct net_bridge *br, return false; } } + +static inline int br_multicast_igmp_type(const struct sk_buff *skb) +{ + return BR_INPUT_SKB_CB(skb)->igmp; +} #else static inline int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, @@ -638,8 +658,9 @@ static inline struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br, return NULL; } -static inline void br_multicast_add_port(struct net_bridge_port *port) +static inline int br_multicast_add_port(struct net_bridge_port *port) { + return 0; } static inline void br_multicast_del_port(struct net_bridge_port *port) @@ -695,6 +716,22 @@ static inline void br_mdb_init(void) static inline void br_mdb_uninit(void) { } + +static inline void br_multicast_count(struct net_bridge *br, + const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir) +{ +} + +static inline int br_multicast_init_stats(struct net_bridge *br) +{ + return 0; +} + +static inline int br_multicast_igmp_type(const struct sk_buff *skb) +{ + return 0; +} #endif /* br_vlan.c */ diff --git a/net/bridge/br_sysfs_br.c b/net/bridge/br_sysfs_br.c index beb47071e38d..e120307c6e36 100644 --- a/net/bridge/br_sysfs_br.c +++ b/net/bridge/br_sysfs_br.c @@ -618,6 +618,30 @@ static ssize_t multicast_startup_query_interval_store( return store_bridge_parm(d, buf, len, set_startup_query_interval); } static DEVICE_ATTR_RW(multicast_startup_query_interval); + +static ssize_t multicast_stats_enabled_show(struct device *d, + struct device_attribute *attr, + char *buf) +{ + struct net_bridge *br = to_bridge(d); + + return sprintf(buf, "%u\n", br->multicast_stats_enabled); +} + +static int set_stats_enabled(struct net_bridge *br, unsigned long val) +{ + br->multicast_stats_enabled = !!val; + return 0; +} + +static ssize_t multicast_stats_enabled_store(struct device *d, + struct device_attribute *attr, + const char *buf, + size_t len) +{ + return store_bridge_parm(d, buf, len, set_stats_enabled); +} +static DEVICE_ATTR_RW(multicast_stats_enabled); #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) static ssize_t nf_call_iptables_show( @@ -784,6 +808,7 @@ static struct attribute *bridge_attrs[] = { &dev_attr_multicast_query_interval.attr, &dev_attr_multicast_query_response_interval.attr, &dev_attr_multicast_startup_query_interval.attr, + &dev_attr_multicast_stats_enabled.attr, #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) &dev_attr_nf_call_iptables.attr, -- cgit v1.2.3 From b1ed4c4fa9a5ccf325184fd90edc50978ef6e33a Mon Sep 17 00:00:00 2001 From: Andrey Vagin Date: Mon, 27 Jun 2016 15:33:56 -0700 Subject: tcp: add an ability to dump and restore window parameters We found that sometimes a restored tcp socket doesn't work. A reason of this bug is incorrect window parameters and in this case tcp_acceptable_seq() returns tcp_wnd_end(tp) instead of tp->snd_nxt. The other side drops packets with this seq, because seq is less than tp->rcv_nxt ( tcp_sequence() ). Data from a send queue is sent only if there is enough space in a window, so when we restore unacked data, we need to expand a window to fit this data. This was in a first version of this patch: "tcp: extend window to fit all restored unacked data in a send queue" Then Alexey recommended me to restore window parameters instead of adjusted them according with data in a sent queue. This sounds resonable. rcv_wnd has to be restored, because it was reported to another side and the offered window is never shrunk. One of reasons why we need to restore snd_wnd was described above. Cc: Pavel Emelyanov Cc: "David S. Miller" Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Signed-off-by: Andrey Vagin Signed-off-by: David S. Miller --- include/uapi/linux/tcp.h | 10 +++++++++ net/ipv4/tcp.c | 57 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 67 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index 53e8e3fe6b1b..482898fc433a 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -115,12 +115,22 @@ enum { #define TCP_CC_INFO 26 /* Get Congestion Control (optional) info */ #define TCP_SAVE_SYN 27 /* Record SYN headers for new connections */ #define TCP_SAVED_SYN 28 /* Get SYN headers recorded for connection */ +#define TCP_REPAIR_WINDOW 29 /* Get/set window parameters */ struct tcp_repair_opt { __u32 opt_code; __u32 opt_val; }; +struct tcp_repair_window { + __u32 snd_wl1; + __u32 snd_wnd; + __u32 max_window; + + __u32 rcv_wnd; + __u32 rcv_wup; +}; + enum { TCP_NO_QUEUE, TCP_RECV_QUEUE, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 5c7ed147449c..108ef2a6665c 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2277,6 +2277,38 @@ static inline bool tcp_can_repair_sock(const struct sock *sk) ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_ESTABLISHED)); } +static int tcp_repair_set_window(struct tcp_sock *tp, char __user *optbuf, int len) +{ + struct tcp_repair_window opt; + + if (!tp->repair) + return -EPERM; + + if (len != sizeof(opt)) + return -EINVAL; + + if (copy_from_user(&opt, optbuf, sizeof(opt))) + return -EFAULT; + + if (opt.max_window < opt.snd_wnd) + return -EINVAL; + + if (after(opt.snd_wl1, tp->rcv_nxt + opt.rcv_wnd)) + return -EINVAL; + + if (after(opt.rcv_wup, tp->rcv_nxt)) + return -EINVAL; + + tp->snd_wl1 = opt.snd_wl1; + tp->snd_wnd = opt.snd_wnd; + tp->max_window = opt.max_window; + + tp->rcv_wnd = opt.rcv_wnd; + tp->rcv_wup = opt.rcv_wup; + + return 0; +} + static int tcp_repair_options_est(struct tcp_sock *tp, struct tcp_repair_opt __user *optbuf, unsigned int len) { @@ -2604,6 +2636,9 @@ static int do_tcp_setsockopt(struct sock *sk, int level, else tp->tsoffset = val - tcp_time_stamp; break; + case TCP_REPAIR_WINDOW: + err = tcp_repair_set_window(tp, optval, optlen); + break; case TCP_NOTSENT_LOWAT: tp->notsent_lowat = val; sk->sk_write_space(sk); @@ -2860,6 +2895,28 @@ static int do_tcp_getsockopt(struct sock *sk, int level, return -EINVAL; break; + case TCP_REPAIR_WINDOW: { + struct tcp_repair_window opt; + + if (get_user(len, optlen)) + return -EFAULT; + + if (len != sizeof(opt)) + return -EINVAL; + + if (!tp->repair) + return -EPERM; + + opt.snd_wl1 = tp->snd_wl1; + opt.snd_wnd = tp->snd_wnd; + opt.max_window = tp->max_window; + opt.rcv_wnd = tp->rcv_wnd; + opt.rcv_wup = tp->rcv_wup; + + if (copy_to_user(optval, &opt, len)) + return -EFAULT; + return 0; + } case TCP_QUEUE_SEQ: if (tp->repair_queue == TCP_SEND_QUEUE) val = tp->write_seq; -- cgit v1.2.3 From 37f501afed23fa1126017255495d5be5e97c9d6d Mon Sep 17 00:00:00 2001 From: "arun.siluvery@linux.intel.com" Date: Fri, 1 Jul 2016 11:43:02 +0100 Subject: drm/i915/bxt: Export pooled eu info to userspace Pooled EU is a bxt only feature and kernel changes are already merged. This feature is not yet exposed to userspace as the support was not yet available. Beignet team expressed interest and added patches to use this. Since we now have a user and patches to use them, expose them from the kernel side as well. v2: fix compile error [1] https://lists.freedesktop.org/archives/beignet/2016-June/007698.html [2] https://lists.freedesktop.org/archives/beignet/2016-June/007699.html Cc: Winiarski, Michal Cc: Zou, Nanhai Cc: Yang, Rong R Cc: Tim Gore Cc: Jeff McGee Signed-off-by: Arun Siluvery Acked-by: Chris Wilson Signed-off-by: Tvrtko Ursulin Link: http://patchwork.freedesktop.org/patch/msgid/1467369782-25992-1-git-send-email-arun.siluvery@linux.intel.com Acked-by: Jani Nikula --- drivers/gpu/drm/i915/i915_drv.c | 6 ++++++ include/uapi/drm/i915_drm.h | 2 ++ 2 files changed, 8 insertions(+) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index c580e24095b0..8a2674013aef 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -365,6 +365,12 @@ static int i915_getparam(struct drm_device *dev, void *data, case I915_PARAM_HAS_EXEC_SOFTPIN: value = 1; break; + case I915_PARAM_HAS_POOLED_EU: + value = HAS_POOLED_EU(dev); + break; + case I915_PARAM_MIN_EU_IN_POOL: + value = INTEL_INFO(dev)->min_eu_in_pool; + break; default: DRM_DEBUG("Unknown parameter %d\n", param->param); return -EINVAL; diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index c17d63d8b543..a642bbc7777d 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -361,6 +361,8 @@ typedef struct drm_i915_irq_wait { #define I915_PARAM_HAS_GPU_RESET 35 #define I915_PARAM_HAS_RESOURCE_STREAMER 36 #define I915_PARAM_HAS_EXEC_SOFTPIN 37 +#define I915_PARAM_HAS_POOLED_EU 38 +#define I915_PARAM_MIN_EU_IN_POOL 39 typedef struct drm_i915_getparam { __s32 param; -- cgit v1.2.3 From 4ed8ec521ed57c4e207ad464ca0388776de74d4b Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:43 -0700 Subject: cgroup: bpf: Add BPF_MAP_TYPE_CGROUP_ARRAY Add a BPF_MAP_TYPE_CGROUP_ARRAY and its bpf_map_ops's implementations. To update an element, the caller is expected to obtain a cgroup2 backed fd by open(cgroup2_dir) and then update the array with that fd. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 1 + kernel/bpf/arraymap.c | 43 +++++++++++++++++++++++++++++++++++++++++++ kernel/bpf/syscall.c | 3 ++- kernel/bpf/verifier.c | 2 ++ 4 files changed, 48 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index be6ac1291680..26c04be32003 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -84,6 +84,7 @@ enum bpf_map_type { BPF_MAP_TYPE_PERCPU_HASH, BPF_MAP_TYPE_PERCPU_ARRAY, BPF_MAP_TYPE_STACK_TRACE, + BPF_MAP_TYPE_CGROUP_ARRAY, }; enum bpf_prog_type { diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 4ec57a649b1f..db1a743e3db2 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -537,3 +537,46 @@ static int __init register_perf_event_array_map(void) return 0; } late_initcall(register_perf_event_array_map); + +#ifdef CONFIG_SOCK_CGROUP_DATA +static void *cgroup_fd_array_get_ptr(struct bpf_map *map, + struct file *map_file /* not used */, + int fd) +{ + return cgroup_get_from_fd(fd); +} + +static void cgroup_fd_array_put_ptr(void *ptr) +{ + /* cgroup_put free cgrp after a rcu grace period */ + cgroup_put(ptr); +} + +static void cgroup_fd_array_free(struct bpf_map *map) +{ + bpf_fd_array_map_clear(map); + fd_array_map_free(map); +} + +static const struct bpf_map_ops cgroup_array_ops = { + .map_alloc = fd_array_map_alloc, + .map_free = cgroup_fd_array_free, + .map_get_next_key = array_map_get_next_key, + .map_lookup_elem = fd_array_map_lookup_elem, + .map_delete_elem = fd_array_map_delete_elem, + .map_fd_get_ptr = cgroup_fd_array_get_ptr, + .map_fd_put_ptr = cgroup_fd_array_put_ptr, +}; + +static struct bpf_map_type_list cgroup_array_type __read_mostly = { + .ops = &cgroup_array_ops, + .type = BPF_MAP_TYPE_CGROUP_ARRAY, +}; + +static int __init register_cgroup_array_map(void) +{ + bpf_register_map_type(&cgroup_array_type); + return 0; +} +late_initcall(register_cgroup_array_map); +#endif diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 22863d9872b1..96d938a22050 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -393,7 +393,8 @@ static int map_update_elem(union bpf_attr *attr) } else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { err = bpf_percpu_array_update(map, key, value, attr->flags); } else if (map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || - map->map_type == BPF_MAP_TYPE_PROG_ARRAY) { + map->map_type == BPF_MAP_TYPE_PROG_ARRAY || + map->map_type == BPF_MAP_TYPE_CGROUP_ARRAY) { rcu_read_lock(); err = bpf_fd_array_map_update_elem(map, f.file, key, value, attr->flags); diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index eec9f90ba030..69ba2251a22b 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1035,6 +1035,8 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) if (func_id != BPF_FUNC_get_stackid) goto error; break; + case BPF_MAP_TYPE_CGROUP_ARRAY: + goto error; default: break; } -- cgit v1.2.3 From 4a482f34afcc162d8456f449b137ec2a95be60d8 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:44 -0700 Subject: cgroup: bpf: Add bpf_skb_in_cgroup_proto Adds a bpf helper, bpf_skb_in_cgroup, to decide if a skb->sk belongs to a descendant of a cgroup2. It is similar to the feature added in netfilter: commit c38c4597e4bf ("netfilter: implement xt_cgroup cgroup2 path match") The user is expected to populate a BPF_MAP_TYPE_CGROUP_ARRAY which will be used by the bpf_skb_in_cgroup. Modifications to the bpf verifier is to ensure BPF_MAP_TYPE_CGROUP_ARRAY and bpf_skb_in_cgroup() are always used together. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 11 +++++++++++ kernel/bpf/verifier.c | 8 +++++++- net/core/filter.c | 38 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 56 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 26c04be32003..f44504d875e2 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -337,6 +337,17 @@ enum bpf_func_id { */ BPF_FUNC_skb_change_type, + /** + * bpf_skb_in_cgroup(skb, map, index) - Check cgroup2 membership of skb + * @skb: pointer to skb + * @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type + * @index: index of the cgroup in the bpf_map + * Return: + * == 0 skb failed the cgroup2 descendant test + * == 1 skb succeeded the cgroup2 descendant test + * < 0 error + */ + BPF_FUNC_skb_in_cgroup, __BPF_FUNC_MAX_ID, }; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 69ba2251a22b..e206c2181412 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1036,7 +1036,9 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) goto error; break; case BPF_MAP_TYPE_CGROUP_ARRAY: - goto error; + if (func_id != BPF_FUNC_skb_in_cgroup) + goto error; + break; default: break; } @@ -1056,6 +1058,10 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) if (map->map_type != BPF_MAP_TYPE_STACK_TRACE) goto error; break; + case BPF_FUNC_skb_in_cgroup: + if (map->map_type != BPF_MAP_TYPE_CGROUP_ARRAY) + goto error; + break; default: break; } diff --git a/net/core/filter.c b/net/core/filter.c index 76fee35da244..54071cf70fb5 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2239,6 +2239,40 @@ bpf_get_skb_set_tunnel_proto(enum bpf_func_id which) } } +#ifdef CONFIG_SOCK_CGROUP_DATA +static u64 bpf_skb_in_cgroup(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + struct sk_buff *skb = (struct sk_buff *)(long)r1; + struct bpf_map *map = (struct bpf_map *)(long)r2; + struct bpf_array *array = container_of(map, struct bpf_array, map); + struct cgroup *cgrp; + struct sock *sk; + u32 i = (u32)r3; + + sk = skb->sk; + if (!sk || !sk_fullsock(sk)) + return -ENOENT; + + if (unlikely(i >= array->map.max_entries)) + return -E2BIG; + + cgrp = READ_ONCE(array->ptrs[i]); + if (unlikely(!cgrp)) + return -EAGAIN; + + return cgroup_is_descendant(sock_cgroup_ptr(&sk->sk_cgrp_data), cgrp); +} + +static const struct bpf_func_proto bpf_skb_in_cgroup_proto = { + .func = bpf_skb_in_cgroup, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_CONST_MAP_PTR, + .arg3_type = ARG_ANYTHING, +}; +#endif + static const struct bpf_func_proto * sk_filter_func_proto(enum bpf_func_id func_id) { @@ -2307,6 +2341,10 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return bpf_get_event_output_proto(); case BPF_FUNC_get_smp_processor_id: return &bpf_get_smp_processor_id_proto; +#ifdef CONFIG_SOCK_CGROUP_DATA + case BPF_FUNC_skb_in_cgroup: + return &bpf_skb_in_cgroup_proto; +#endif default: return sk_filter_func_proto(func_id); } -- cgit v1.2.3 From 08f4b5918b2d6b491f0403cc1886f5cdccef89bb Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:51:01 +0300 Subject: net/devlink: Add E-Switch mode control Add the commands to set and show the mode of SRIOV E-Switch, two modes are supported: * legacy: operating in the "old" L2 based mode (DMAC --> VF vport) * switchdev: the E-Switch is referred to as whitebox switch configured using standard tools such as tc, bridge, openvswitch etc. To allow working with the tools, for each VF, a VF representor netdevice is created by the E-Switch manager vendor device driver instance (e.g PF). Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- include/net/devlink.h | 3 ++ include/uapi/linux/devlink.h | 8 ++++ net/core/devlink.c | 87 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 98 insertions(+) (limited to 'include/uapi') diff --git a/include/net/devlink.h b/include/net/devlink.h index 1d45b61cb320..c99ffe8cef3c 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -90,6 +90,9 @@ struct devlink_ops { u16 tc_index, enum devlink_sb_pool_type pool_type, u32 *p_cur, u32 *p_max); + + int (*eswitch_mode_get)(struct devlink *devlink, u16 *p_mode); + int (*eswitch_mode_set)(struct devlink *devlink, u16 mode); }; static inline void *devlink_priv(struct devlink *devlink) diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h index ba0073b26fa6..915bfa74458c 100644 --- a/include/uapi/linux/devlink.h +++ b/include/uapi/linux/devlink.h @@ -57,6 +57,8 @@ enum devlink_command { DEVLINK_CMD_SB_OCC_SNAPSHOT, DEVLINK_CMD_SB_OCC_MAX_CLEAR, + DEVLINK_CMD_ESWITCH_MODE_GET, + DEVLINK_CMD_ESWITCH_MODE_SET, /* add new commands above here */ __DEVLINK_CMD_MAX, @@ -95,6 +97,11 @@ enum devlink_sb_threshold_type { #define DEVLINK_SB_THRESHOLD_TO_ALPHA_MAX 20 +enum devlink_eswitch_mode { + DEVLINK_ESWITCH_MODE_LEGACY, + DEVLINK_ESWITCH_MODE_SWITCHDEV, +}; + enum devlink_attr { /* don't change the order or add anything between, this is ABI! */ DEVLINK_ATTR_UNSPEC, @@ -125,6 +132,7 @@ enum devlink_attr { DEVLINK_ATTR_SB_TC_INDEX, /* u16 */ DEVLINK_ATTR_SB_OCC_CUR, /* u32 */ DEVLINK_ATTR_SB_OCC_MAX, /* u32 */ + DEVLINK_ATTR_ESWITCH_MODE, /* u16 */ /* add new attributes above here, update the policy in devlink.c */ diff --git a/net/core/devlink.c b/net/core/devlink.c index 933e8d4d3968..b2e592a198c0 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -1394,6 +1394,78 @@ static int devlink_nl_cmd_sb_occ_max_clear_doit(struct sk_buff *skb, return -EOPNOTSUPP; } +static int devlink_eswitch_fill(struct sk_buff *msg, struct devlink *devlink, + enum devlink_command cmd, u32 portid, + u32 seq, int flags, u16 mode) +{ + void *hdr; + + hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd); + if (!hdr) + return -EMSGSIZE; + + if (devlink_nl_put_handle(msg, devlink)) + goto nla_put_failure; + + if (nla_put_u16(msg, DEVLINK_ATTR_ESWITCH_MODE, mode)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + return -EMSGSIZE; +} + +static int devlink_nl_cmd_eswitch_mode_get_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct devlink *devlink = info->user_ptr[0]; + const struct devlink_ops *ops = devlink->ops; + struct sk_buff *msg; + u16 mode; + int err; + + if (!ops || !ops->eswitch_mode_get) + return -EOPNOTSUPP; + + err = ops->eswitch_mode_get(devlink, &mode); + if (err) + return err; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + err = devlink_eswitch_fill(msg, devlink, DEVLINK_CMD_ESWITCH_MODE_GET, + info->snd_portid, info->snd_seq, 0, mode); + + if (err) { + nlmsg_free(msg); + return err; + } + + return genlmsg_reply(msg, info); +} + +static int devlink_nl_cmd_eswitch_mode_set_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct devlink *devlink = info->user_ptr[0]; + const struct devlink_ops *ops = devlink->ops; + u16 mode; + + if (!info->attrs[DEVLINK_ATTR_ESWITCH_MODE]) + return -EINVAL; + + mode = nla_get_u16(info->attrs[DEVLINK_ATTR_ESWITCH_MODE]); + + if (ops && ops->eswitch_mode_set) + return ops->eswitch_mode_set(devlink, mode); + return -EOPNOTSUPP; +} + static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING }, [DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING }, @@ -1407,6 +1479,7 @@ static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_SB_POOL_THRESHOLD_TYPE] = { .type = NLA_U8 }, [DEVLINK_ATTR_SB_THRESHOLD] = { .type = NLA_U32 }, [DEVLINK_ATTR_SB_TC_INDEX] = { .type = NLA_U16 }, + [DEVLINK_ATTR_ESWITCH_MODE] = { .type = NLA_U16 }, }; static const struct genl_ops devlink_nl_ops[] = { @@ -1525,6 +1598,20 @@ static const struct genl_ops devlink_nl_ops[] = { DEVLINK_NL_FLAG_NEED_SB | DEVLINK_NL_FLAG_LOCK_PORTS, }, + { + .cmd = DEVLINK_CMD_ESWITCH_MODE_GET, + .doit = devlink_nl_cmd_eswitch_mode_get_doit, + .policy = devlink_nl_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + }, + { + .cmd = DEVLINK_CMD_ESWITCH_MODE_SET, + .doit = devlink_nl_cmd_eswitch_mode_set_doit, + .policy = devlink_nl_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + }, }; /** -- cgit v1.2.3 From bc3d674462e5df5f2b33adbfcaad9edff8b827f4 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Mon, 4 Jul 2016 08:08:39 +0100 Subject: drm/i915: Allow userspace to request no-error-capture upon GPU hangs igt likes to inject GPU hangs into its command streams. However, as we expect these hangs, we don't actually want them recorded in the dmesg output or stored in the i915_error_state (usually). To accommodate this allow userspace to set a flag on the context that any hang emanating from that context will not be recorded. We still do the error capture (otherwise how do we find the guilty context and know its intent?) as part of the reason for random GPU hang injection is to exercise the race conditions between the error capture and normal execution. v2: Split out the request->ringbuf error capture changes. v3: Move the flag defines next to the intel_context->flags definition Signed-off-by: Chris Wilson Acked-by: Daniel Vetter Reviewed-by: Dave Gordon Link: http://patchwork.freedesktop.org/patch/msgid/1467616119-4093-9-git-send-email-chris@chris-wilson.co.uk --- drivers/gpu/drm/i915/i915_drv.h | 4 +++- drivers/gpu/drm/i915/i915_gem_context.c | 13 +++++++++++++ drivers/gpu/drm/i915/i915_gpu_error.c | 20 ++++++++++++-------- include/uapi/drm/i915_drm.h | 1 + 4 files changed, 29 insertions(+), 9 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 488891853cb5..251a08d8808d 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -475,6 +475,7 @@ struct drm_i915_error_state { struct timeval time; char error_msg[128]; + bool simulated; int iommu; u32 reset_count; u32 suspend_count; @@ -875,9 +876,10 @@ struct i915_gem_context { /* Unique identifier for this context, used by the hw for tracking */ unsigned long flags; +#define CONTEXT_NO_ZEROMAP BIT(0) +#define CONTEXT_NO_ERROR_CAPTURE BIT(1) unsigned hw_id; u32 user_handle; -#define CONTEXT_NO_ZEROMAP (1<<0) u32 ggtt_alignment; diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index 3a6594b70900..8e952b1a31b3 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -1026,6 +1026,9 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data, else args->value = to_i915(dev)->ggtt.base.total; break; + case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE: + args->value = !!(ctx->flags & CONTEXT_NO_ERROR_CAPTURE); + break; default: ret = -EINVAL; break; @@ -1071,6 +1074,16 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data, ctx->flags |= args->value ? CONTEXT_NO_ZEROMAP : 0; } break; + case I915_CONTEXT_PARAM_NO_ERROR_CAPTURE: + if (args->size) { + ret = -EINVAL; + } else { + if (args->value) + ctx->flags |= CONTEXT_NO_ERROR_CAPTURE; + else + ctx->flags &= ~CONTEXT_NO_ERROR_CAPTURE; + } + break; default: ret = -EINVAL; break; diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 1be63590a7fe..c6e05cccbedf 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -1093,9 +1093,8 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv, struct i915_address_space *vm; struct intel_ringbuffer *rb; - vm = request->ctx && request->ctx->ppgtt ? - &request->ctx->ppgtt->base : - &ggtt->base; + vm = request->ctx->ppgtt ? + &request->ctx->ppgtt->base : &ggtt->base; /* We need to copy these to an anonymous buffer * as the simplest method to avoid being overwritten @@ -1123,6 +1122,9 @@ static void i915_gem_record_rings(struct drm_i915_private *dev_priv, rcu_read_unlock(); } + error->simulated |= + request->ctx->flags & CONTEXT_NO_ERROR_CAPTURE; + rb = request->ringbuf; error->ring[i].cpu_ring_head = rb->head; error->ring[i].cpu_ring_tail = rb->tail; @@ -1422,12 +1424,14 @@ void i915_capture_error_state(struct drm_i915_private *dev_priv, i915_error_capture_msg(dev_priv, error, engine_mask, error_msg); DRM_INFO("%s\n", error->error_msg); - spin_lock_irqsave(&dev_priv->gpu_error.lock, flags); - if (dev_priv->gpu_error.first_error == NULL) { - dev_priv->gpu_error.first_error = error; - error = NULL; + if (!error->simulated) { + spin_lock_irqsave(&dev_priv->gpu_error.lock, flags); + if (!dev_priv->gpu_error.first_error) { + dev_priv->gpu_error.first_error = error; + error = NULL; + } + spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags); } - spin_unlock_irqrestore(&dev_priv->gpu_error.lock, flags); if (error) { i915_error_state_free(&error->ref); diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index a642bbc7777d..d7e81a3886fd 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -1173,6 +1173,7 @@ struct drm_i915_gem_context_param { #define I915_CONTEXT_PARAM_BAN_PERIOD 0x1 #define I915_CONTEXT_PARAM_NO_ZEROMAP 0x2 #define I915_CONTEXT_PARAM_GTT_SIZE 0x3 +#define I915_CONTEXT_PARAM_NO_ERROR_CAPTURE 0x4 __u64 value; }; -- cgit v1.2.3 From 09748a22f4ab7b0ab5a83c432f6e18f65f18e09b Mon Sep 17 00:00:00 2001 From: Matthias Schiffer Date: Mon, 9 May 2016 18:41:08 +0200 Subject: batman-adv: add generic netlink family for batman-adv debugfs is currently severely broken virtually everywhere in the kernel where files are dynamically added and removed (see http://lkml.iu.edu/hypermail/linux/kernel/1506.1/02196.html for some details). In addition to that, debugfs is not namespace-aware. Instead of adding new debugfs entries, the whole infrastructure should be moved to netlink. This will fix the long standing problem of large buffers for debug tables and hard to parse text files. Signed-off-by: Matthias Schiffer Signed-off-by: Andrew Lunn [sven.eckelmann@open-mesh.com: Strip down patch to only add genl family, add missing kerneldoc] Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich --- MAINTAINERS | 1 + include/uapi/linux/batman_adv.h | 53 ++++++++++++++++++++++++++++++++++++++ net/batman-adv/Makefile | 1 + net/batman-adv/main.c | 3 +++ net/batman-adv/netlink.c | 57 +++++++++++++++++++++++++++++++++++++++++ net/batman-adv/netlink.h | 26 +++++++++++++++++++ 6 files changed, 141 insertions(+) create mode 100644 include/uapi/linux/batman_adv.h create mode 100644 net/batman-adv/netlink.c create mode 100644 net/batman-adv/netlink.h (limited to 'include/uapi') diff --git a/MAINTAINERS b/MAINTAINERS index 0e26025718c2..e25091f4e583 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2291,6 +2291,7 @@ S: Maintained F: Documentation/ABI/testing/sysfs-class-net-batman-adv F: Documentation/ABI/testing/sysfs-class-net-mesh F: Documentation/networking/batman-adv.txt +F: include/uapi/linux/batman_adv.h F: net/batman-adv/ BAYCOM/HDLCDRV DRIVERS FOR AX.25 diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h new file mode 100644 index 000000000000..79f797281b87 --- /dev/null +++ b/include/uapi/linux/batman_adv.h @@ -0,0 +1,53 @@ +/* Copyright (C) 2016 B.A.T.M.A.N. contributors: + * + * Matthias Schiffer + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef _UAPI_LINUX_BATMAN_ADV_H_ +#define _UAPI_LINUX_BATMAN_ADV_H_ + +#define BATADV_NL_NAME "batadv" + +/** + * enum batadv_nl_attrs - batman-adv netlink attributes + * + * @BATADV_ATTR_UNSPEC: unspecified attribute to catch errors + * @__BATADV_ATTR_AFTER_LAST: internal use + * @NUM_BATADV_ATTR: total number of batadv_nl_attrs available + * @BATADV_ATTR_MAX: highest attribute number currently defined + */ +enum batadv_nl_attrs { + BATADV_ATTR_UNSPEC, + /* add attributes above here, update the policy in netlink.c */ + __BATADV_ATTR_AFTER_LAST, + NUM_BATADV_ATTR = __BATADV_ATTR_AFTER_LAST, + BATADV_ATTR_MAX = __BATADV_ATTR_AFTER_LAST - 1 +}; + +/** + * enum batadv_nl_commands - supported batman-adv netlink commands + * + * @BATADV_CMD_UNSPEC: unspecified command to catch errors + * @__BATADV_CMD_AFTER_LAST: internal use + * @BATADV_CMD_MAX: highest used command number + */ +enum batadv_nl_commands { + BATADV_CMD_UNSPEC, + /* add new commands above here */ + __BATADV_CMD_AFTER_LAST, + BATADV_CMD_MAX = __BATADV_CMD_AFTER_LAST - 1 +}; + +#endif /* _UAPI_LINUX_BATMAN_ADV_H_ */ diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index a55f4ec97068..7da59014e134 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -35,6 +35,7 @@ batman-adv-y += icmp_socket.o batman-adv-$(CONFIG_BATMAN_ADV_DEBUG) += log.o batman-adv-y += main.o batman-adv-$(CONFIG_BATMAN_ADV_MCAST) += multicast.o +batman-adv-y += netlink.o batman-adv-$(CONFIG_BATMAN_ADV_NC) += network-coding.o batman-adv-y += originator.o batman-adv-y += routing.o diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index eab9d1b8a6eb..275604b7c64e 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -57,6 +57,7 @@ #include "icmp_socket.h" #include "log.h" #include "multicast.h" +#include "netlink.h" #include "network-coding.h" #include "originator.h" #include "packet.h" @@ -99,6 +100,7 @@ static int __init batadv_init(void) register_netdevice_notifier(&batadv_hard_if_notifier); rtnl_link_register(&batadv_link_ops); + batadv_netlink_register(); pr_info("B.A.T.M.A.N. advanced %s (compatibility version %i) loaded\n", BATADV_SOURCE_VERSION, BATADV_COMPAT_VERSION); @@ -109,6 +111,7 @@ static int __init batadv_init(void) static void __exit batadv_exit(void) { batadv_debugfs_destroy(); + batadv_netlink_unregister(); rtnl_link_unregister(&batadv_link_ops); unregister_netdevice_notifier(&batadv_hard_if_notifier); batadv_hardif_remove_interfaces(); diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c new file mode 100644 index 000000000000..9e4c865a9cc0 --- /dev/null +++ b/net/batman-adv/netlink.c @@ -0,0 +1,57 @@ +/* Copyright (C) 2016 B.A.T.M.A.N. contributors: + * + * Matthias Schiffer + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "netlink.h" +#include "main.h" + +#include +#include +#include +#include +#include + +static struct genl_family batadv_netlink_family = { + .id = GENL_ID_GENERATE, + .hdrsize = 0, + .name = BATADV_NL_NAME, + .version = 1, + .maxattr = BATADV_ATTR_MAX, +}; + +static struct genl_ops batadv_netlink_ops[] = { +}; + +/** + * batadv_netlink_register - register batadv genl netlink family + */ +void __init batadv_netlink_register(void) +{ + int ret; + + ret = genl_register_family_with_ops(&batadv_netlink_family, + batadv_netlink_ops); + if (ret) + pr_warn("unable to register netlink family"); +} + +/** + * batadv_netlink_unregister - unregister batadv genl netlink family + */ +void batadv_netlink_unregister(void) +{ + genl_unregister_family(&batadv_netlink_family); +} diff --git a/net/batman-adv/netlink.h b/net/batman-adv/netlink.h new file mode 100644 index 000000000000..39044ccff662 --- /dev/null +++ b/net/batman-adv/netlink.h @@ -0,0 +1,26 @@ +/* Copyright (C) 2016 B.A.T.M.A.N. contributors: + * + * Matthias Schiffer + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_NETLINK_H_ +#define _NET_BATMAN_ADV_NETLINK_H_ + +#include "main.h" + +void batadv_netlink_register(void); +void batadv_netlink_unregister(void); + +#endif /* _NET_BATMAN_ADV_NETLINK_H_ */ -- cgit v1.2.3 From 5da0aef5e93591b373010c10f374c4161b37728c Mon Sep 17 00:00:00 2001 From: Matthias Schiffer Date: Mon, 9 May 2016 18:41:09 +0200 Subject: batman-adv: add netlink command to query generic mesh information files BATADV_CMD_GET_MESH_INFO is used to query basic information about a batman-adv softif (name, index and MAC address for both the softif and the primary hardif; routing algorithm; batman-adv version). Signed-off-by: Matthias Schiffer Signed-off-by: Andrew Lunn [sven.eckelmann@open-mesh.com: Reduce the number of changes to BATADV_CMD_GET_MESH_INFO, add missing kerneldoc, add policy for attributes] Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich --- include/uapi/linux/batman_adv.h | 18 ++++++ net/batman-adv/netlink.c | 137 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 155 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h index 79f797281b87..c39623c7109e 100644 --- a/include/uapi/linux/batman_adv.h +++ b/include/uapi/linux/batman_adv.h @@ -24,12 +24,28 @@ * enum batadv_nl_attrs - batman-adv netlink attributes * * @BATADV_ATTR_UNSPEC: unspecified attribute to catch errors + * @BATADV_ATTR_VERSION: batman-adv version string + * @BATADV_ATTR_ALGO_NAME: name of routing algorithm + * @BATADV_ATTR_MESH_IFINDEX: index of the batman-adv interface + * @BATADV_ATTR_MESH_IFNAME: name of the batman-adv interface + * @BATADV_ATTR_MESH_ADDRESS: mac address of the batman-adv interface + * @BATADV_ATTR_HARD_IFINDEX: index of the non-batman-adv interface + * @BATADV_ATTR_HARD_IFNAME: name of the non-batman-adv interface + * @BATADV_ATTR_HARD_ADDRESS: mac address of the non-batman-adv interface * @__BATADV_ATTR_AFTER_LAST: internal use * @NUM_BATADV_ATTR: total number of batadv_nl_attrs available * @BATADV_ATTR_MAX: highest attribute number currently defined */ enum batadv_nl_attrs { BATADV_ATTR_UNSPEC, + BATADV_ATTR_VERSION, + BATADV_ATTR_ALGO_NAME, + BATADV_ATTR_MESH_IFINDEX, + BATADV_ATTR_MESH_IFNAME, + BATADV_ATTR_MESH_ADDRESS, + BATADV_ATTR_HARD_IFINDEX, + BATADV_ATTR_HARD_IFNAME, + BATADV_ATTR_HARD_ADDRESS, /* add attributes above here, update the policy in netlink.c */ __BATADV_ATTR_AFTER_LAST, NUM_BATADV_ATTR = __BATADV_ATTR_AFTER_LAST, @@ -40,11 +56,13 @@ enum batadv_nl_attrs { * enum batadv_nl_commands - supported batman-adv netlink commands * * @BATADV_CMD_UNSPEC: unspecified command to catch errors + * @BATADV_CMD_GET_MESH_INFO: Query basic information about batman-adv device * @__BATADV_CMD_AFTER_LAST: internal use * @BATADV_CMD_MAX: highest used command number */ enum batadv_nl_commands { BATADV_CMD_UNSPEC, + BATADV_CMD_GET_MESH_INFO, /* add new commands above here */ __BATADV_CMD_AFTER_LAST, BATADV_CMD_MAX = __BATADV_CMD_AFTER_LAST - 1 diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index 9e4c865a9cc0..68152aa9bb26 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -18,12 +18,24 @@ #include "netlink.h" #include "main.h" +#include +#include #include +#include #include +#include +#include #include +#include #include +#include #include +#include "hard-interface.h" +#include "soft-interface.h" + +struct sk_buff; + static struct genl_family batadv_netlink_family = { .id = GENL_ID_GENERATE, .hdrsize = 0, @@ -32,7 +44,132 @@ static struct genl_family batadv_netlink_family = { .maxattr = BATADV_ATTR_MAX, }; +static struct nla_policy batadv_netlink_policy[NUM_BATADV_ATTR] = { + [BATADV_ATTR_VERSION] = { .type = NLA_STRING }, + [BATADV_ATTR_ALGO_NAME] = { .type = NLA_STRING }, + [BATADV_ATTR_MESH_IFINDEX] = { .type = NLA_U32 }, + [BATADV_ATTR_MESH_IFNAME] = { .type = NLA_STRING }, + [BATADV_ATTR_MESH_ADDRESS] = { .len = ETH_ALEN }, + [BATADV_ATTR_HARD_IFINDEX] = { .type = NLA_U32 }, + [BATADV_ATTR_HARD_IFNAME] = { .type = NLA_STRING }, + [BATADV_ATTR_HARD_ADDRESS] = { .len = ETH_ALEN }, +}; + +/** + * batadv_netlink_mesh_info_put - fill in generic information about mesh + * interface + * @msg: netlink message to be sent back + * @soft_iface: interface for which the data should be taken + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_mesh_info_put(struct sk_buff *msg, struct net_device *soft_iface) +{ + struct batadv_priv *bat_priv = netdev_priv(soft_iface); + struct batadv_hard_iface *primary_if = NULL; + struct net_device *hard_iface; + int ret = -ENOBUFS; + + if (nla_put_string(msg, BATADV_ATTR_VERSION, BATADV_SOURCE_VERSION) || + nla_put_string(msg, BATADV_ATTR_ALGO_NAME, + bat_priv->bat_algo_ops->name) || + nla_put_u32(msg, BATADV_ATTR_MESH_IFINDEX, soft_iface->ifindex) || + nla_put_string(msg, BATADV_ATTR_MESH_IFNAME, soft_iface->name) || + nla_put(msg, BATADV_ATTR_MESH_ADDRESS, ETH_ALEN, + soft_iface->dev_addr)) + goto out; + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (primary_if && primary_if->if_status == BATADV_IF_ACTIVE) { + hard_iface = primary_if->net_dev; + + if (nla_put_u32(msg, BATADV_ATTR_HARD_IFINDEX, + hard_iface->ifindex) || + nla_put_string(msg, BATADV_ATTR_HARD_IFNAME, + hard_iface->name) || + nla_put(msg, BATADV_ATTR_HARD_ADDRESS, ETH_ALEN, + hard_iface->dev_addr)) + goto out; + } + + ret = 0; + + out: + if (primary_if) + batadv_hardif_put(primary_if); + + return ret; +} + +/** + * batadv_netlink_get_mesh_info - handle incoming BATADV_CMD_GET_MESH_INFO + * netlink request + * @skb: received netlink message + * @info: receiver information + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_get_mesh_info(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct net_device *soft_iface; + struct sk_buff *msg = NULL; + void *msg_head; + int ifindex; + int ret; + + if (!info->attrs[BATADV_ATTR_MESH_IFINDEX]) + return -EINVAL; + + ifindex = nla_get_u32(info->attrs[BATADV_ATTR_MESH_IFINDEX]); + if (!ifindex) + return -EINVAL; + + soft_iface = dev_get_by_index(net, ifindex); + if (!soft_iface || !batadv_softif_is_valid(soft_iface)) { + ret = -ENODEV; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + ret = -ENOMEM; + goto out; + } + + msg_head = genlmsg_put(msg, info->snd_portid, info->snd_seq, + &batadv_netlink_family, 0, + BATADV_CMD_GET_MESH_INFO); + if (!msg_head) { + ret = -ENOBUFS; + goto out; + } + + ret = batadv_netlink_mesh_info_put(msg, soft_iface); + + out: + if (soft_iface) + dev_put(soft_iface); + + if (ret) { + if (msg) + nlmsg_free(msg); + return ret; + } + + genlmsg_end(msg, msg_head); + return genlmsg_reply(msg, info); +} + static struct genl_ops batadv_netlink_ops[] = { + { + .cmd = BATADV_CMD_GET_MESH_INFO, + .flags = GENL_ADMIN_PERM, + .policy = batadv_netlink_policy, + .doit = batadv_netlink_get_mesh_info, + }, }; /** -- cgit v1.2.3 From 33a3bb4a3345bb511f9c69c913da95d4693e2a4e Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Thu, 5 May 2016 13:09:43 +0200 Subject: batman-adv: throughput meter implementation The throughput meter module is a simple, kernel-space replacement for throughtput measurements tool like iperf and netperf. It is intended to approximate TCP behaviour. It is invoked through batctl: the protocol is connection oriented, with cumulative acknowledgment and a dynamic-size sliding window. The test *can* be interrupted by batctl. A receiver side timeout avoids unlimited waitings for sender packets: after one second of inactivity, the receiver abort the ongoing test. Based on a prototype from Edo Monticelli Signed-off-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich --- include/uapi/linux/batman_adv.h | 43 ++ net/batman-adv/Makefile | 1 + net/batman-adv/log.h | 18 +- net/batman-adv/main.c | 4 + net/batman-adv/main.h | 8 + net/batman-adv/netlink.c | 234 +++++- net/batman-adv/netlink.h | 6 + net/batman-adv/packet.h | 54 ++ net/batman-adv/routing.c | 8 + net/batman-adv/soft-interface.c | 2 + net/batman-adv/tp_meter.c | 1507 +++++++++++++++++++++++++++++++++++++++ net/batman-adv/tp_meter.h | 34 + net/batman-adv/types.h | 112 +++ 13 files changed, 2021 insertions(+), 10 deletions(-) create mode 100644 net/batman-adv/tp_meter.c create mode 100644 net/batman-adv/tp_meter.h (limited to 'include/uapi') diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h index c39623c7109e..0fbf6fd4711b 100644 --- a/include/uapi/linux/batman_adv.h +++ b/include/uapi/linux/batman_adv.h @@ -20,6 +20,8 @@ #define BATADV_NL_NAME "batadv" +#define BATADV_NL_MCAST_GROUP_TPMETER "tpmeter" + /** * enum batadv_nl_attrs - batman-adv netlink attributes * @@ -32,6 +34,12 @@ * @BATADV_ATTR_HARD_IFINDEX: index of the non-batman-adv interface * @BATADV_ATTR_HARD_IFNAME: name of the non-batman-adv interface * @BATADV_ATTR_HARD_ADDRESS: mac address of the non-batman-adv interface + * @BATADV_ATTR_ORIG_ADDRESS: originator mac address + * @BATADV_ATTR_TPMETER_RESULT: result of run (see batadv_tp_meter_status) + * @BATADV_ATTR_TPMETER_TEST_TIME: time (msec) the run took + * @BATADV_ATTR_TPMETER_BYTES: amount of acked bytes during run + * @BATADV_ATTR_TPMETER_COOKIE: session cookie to match tp_meter session + * @BATADV_ATTR_PAD: attribute used for padding for 64-bit alignment * @__BATADV_ATTR_AFTER_LAST: internal use * @NUM_BATADV_ATTR: total number of batadv_nl_attrs available * @BATADV_ATTR_MAX: highest attribute number currently defined @@ -46,6 +54,12 @@ enum batadv_nl_attrs { BATADV_ATTR_HARD_IFINDEX, BATADV_ATTR_HARD_IFNAME, BATADV_ATTR_HARD_ADDRESS, + BATADV_ATTR_ORIG_ADDRESS, + BATADV_ATTR_TPMETER_RESULT, + BATADV_ATTR_TPMETER_TEST_TIME, + BATADV_ATTR_TPMETER_BYTES, + BATADV_ATTR_TPMETER_COOKIE, + BATADV_ATTR_PAD, /* add attributes above here, update the policy in netlink.c */ __BATADV_ATTR_AFTER_LAST, NUM_BATADV_ATTR = __BATADV_ATTR_AFTER_LAST, @@ -57,15 +71,44 @@ enum batadv_nl_attrs { * * @BATADV_CMD_UNSPEC: unspecified command to catch errors * @BATADV_CMD_GET_MESH_INFO: Query basic information about batman-adv device + * @BATADV_CMD_TP_METER: Start a tp meter session + * @BATADV_CMD_TP_METER_CANCEL: Cancel a tp meter session * @__BATADV_CMD_AFTER_LAST: internal use * @BATADV_CMD_MAX: highest used command number */ enum batadv_nl_commands { BATADV_CMD_UNSPEC, BATADV_CMD_GET_MESH_INFO, + BATADV_CMD_TP_METER, + BATADV_CMD_TP_METER_CANCEL, /* add new commands above here */ __BATADV_CMD_AFTER_LAST, BATADV_CMD_MAX = __BATADV_CMD_AFTER_LAST - 1 }; +/** + * enum batadv_tp_meter_reason - reason of a tp meter test run stop + * @BATADV_TP_REASON_COMPLETE: sender finished tp run + * @BATADV_TP_REASON_CANCEL: sender was stopped during run + * @BATADV_TP_REASON_DST_UNREACHABLE: receiver could not be reached or didn't + * answer + * @BATADV_TP_REASON_RESEND_LIMIT: (unused) sender retry reached limit + * @BATADV_TP_REASON_ALREADY_ONGOING: test to or from the same node already + * ongoing + * @BATADV_TP_REASON_MEMORY_ERROR: test was stopped due to low memory + * @BATADV_TP_REASON_CANT_SEND: failed to send via outgoing interface + * @BATADV_TP_REASON_TOO_MANY: too many ongoing sessions + */ +enum batadv_tp_meter_reason { + BATADV_TP_REASON_COMPLETE = 3, + BATADV_TP_REASON_CANCEL = 4, + /* error status >= 128 */ + BATADV_TP_REASON_DST_UNREACHABLE = 128, + BATADV_TP_REASON_RESEND_LIMIT = 129, + BATADV_TP_REASON_ALREADY_ONGOING = 130, + BATADV_TP_REASON_MEMORY_ERROR = 131, + BATADV_TP_REASON_CANT_SEND = 132, + BATADV_TP_REASON_TOO_MANY = 133, +}; + #endif /* _UAPI_LINUX_BATMAN_ADV_H_ */ diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index 7da59014e134..a83fc6c58d19 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -42,5 +42,6 @@ batman-adv-y += routing.o batman-adv-y += send.o batman-adv-y += soft-interface.o batman-adv-y += sysfs.o +batman-adv-y += tp_meter.o batman-adv-y += translation-table.o batman-adv-y += tvlv.o diff --git a/net/batman-adv/log.h b/net/batman-adv/log.h index 9948e56eabaa..e0e1a88c3e58 100644 --- a/net/batman-adv/log.h +++ b/net/batman-adv/log.h @@ -51,17 +51,19 @@ static inline void batadv_debug_log_cleanup(struct batadv_priv *bat_priv) * @BATADV_DBG_DAT: ARP snooping and DAT related messages * @BATADV_DBG_NC: network coding related messages * @BATADV_DBG_MCAST: multicast related messages + * @BATADV_DBG_TP_METER: throughput meter messages * @BATADV_DBG_ALL: the union of all the above log levels */ enum batadv_dbg_level { - BATADV_DBG_BATMAN = BIT(0), - BATADV_DBG_ROUTES = BIT(1), - BATADV_DBG_TT = BIT(2), - BATADV_DBG_BLA = BIT(3), - BATADV_DBG_DAT = BIT(4), - BATADV_DBG_NC = BIT(5), - BATADV_DBG_MCAST = BIT(6), - BATADV_DBG_ALL = 127, + BATADV_DBG_BATMAN = BIT(0), + BATADV_DBG_ROUTES = BIT(1), + BATADV_DBG_TT = BIT(2), + BATADV_DBG_BLA = BIT(3), + BATADV_DBG_DAT = BIT(4), + BATADV_DBG_NC = BIT(5), + BATADV_DBG_MCAST = BIT(6), + BATADV_DBG_TP_METER = BIT(7), + BATADV_DBG_ALL = 127, }; #ifdef CONFIG_BATMAN_ADV_DEBUG diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 275604b7c64e..fe4c5e29f96b 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -64,6 +64,7 @@ #include "routing.h" #include "send.h" #include "soft-interface.h" +#include "tp_meter.h" #include "translation-table.h" /* List manipulations on hardif_list have to be rtnl_lock()'ed, @@ -89,6 +90,7 @@ static int __init batadv_init(void) batadv_v_init(); batadv_iv_init(); batadv_nc_init(); + batadv_tp_meter_init(); batadv_event_workqueue = create_singlethread_workqueue("bat_events"); @@ -142,6 +144,7 @@ int batadv_mesh_init(struct net_device *soft_iface) spin_lock_init(&bat_priv->tvlv.container_list_lock); spin_lock_init(&bat_priv->tvlv.handler_list_lock); spin_lock_init(&bat_priv->softif_vlan_list_lock); + spin_lock_init(&bat_priv->tp_list_lock); INIT_HLIST_HEAD(&bat_priv->forw_bat_list); INIT_HLIST_HEAD(&bat_priv->forw_bcast_list); @@ -160,6 +163,7 @@ int batadv_mesh_init(struct net_device *soft_iface) INIT_HLIST_HEAD(&bat_priv->tvlv.container_list); INIT_HLIST_HEAD(&bat_priv->tvlv.handler_list); INIT_HLIST_HEAD(&bat_priv->softif_vlan_list); + INIT_HLIST_HEAD(&bat_priv->tp_list); ret = batadv_v_mesh_init(bat_priv); if (ret < 0) diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 857fb5a4e37a..06a860845434 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -100,6 +100,9 @@ #define BATADV_NUM_BCASTS_WIRELESS 3 #define BATADV_NUM_BCASTS_MAX 3 +/* length of the single packet used by the TP meter */ +#define BATADV_TP_PACKET_LEN ETH_DATA_LEN + /* msecs after which an ARP_REQUEST is sent in broadcast as fallback */ #define ARP_REQ_DELAY 250 /* numbers of originator to contact for any PUT/GET DHT operation */ @@ -131,6 +134,11 @@ #define BATADV_NC_NODE_TIMEOUT 10000 /* Milliseconds */ +/** + * BATADV_TP_MAX_NUM - maximum number of simultaneously active tp sessions + */ +#define BATADV_TP_MAX_NUM 5 + enum batadv_mesh_state { BATADV_MESH_INACTIVE, BATADV_MESH_ACTIVE, diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index 68152aa9bb26..c25bbb8ab06c 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -27,12 +27,14 @@ #include #include #include +#include #include #include #include #include "hard-interface.h" #include "soft-interface.h" +#include "tp_meter.h" struct sk_buff; @@ -44,6 +46,15 @@ static struct genl_family batadv_netlink_family = { .maxattr = BATADV_ATTR_MAX, }; +/* multicast groups */ +enum batadv_netlink_multicast_groups { + BATADV_NL_MCGRP_TPMETER, +}; + +static struct genl_multicast_group batadv_netlink_mcgrps[] = { + [BATADV_NL_MCGRP_TPMETER] = { .name = BATADV_NL_MCAST_GROUP_TPMETER }, +}; + static struct nla_policy batadv_netlink_policy[NUM_BATADV_ATTR] = { [BATADV_ATTR_VERSION] = { .type = NLA_STRING }, [BATADV_ATTR_ALGO_NAME] = { .type = NLA_STRING }, @@ -53,6 +64,11 @@ static struct nla_policy batadv_netlink_policy[NUM_BATADV_ATTR] = { [BATADV_ATTR_HARD_IFINDEX] = { .type = NLA_U32 }, [BATADV_ATTR_HARD_IFNAME] = { .type = NLA_STRING }, [BATADV_ATTR_HARD_ADDRESS] = { .len = ETH_ALEN }, + [BATADV_ATTR_ORIG_ADDRESS] = { .len = ETH_ALEN }, + [BATADV_ATTR_TPMETER_RESULT] = { .type = NLA_U8 }, + [BATADV_ATTR_TPMETER_TEST_TIME] = { .type = NLA_U32 }, + [BATADV_ATTR_TPMETER_BYTES] = { .type = NLA_U64 }, + [BATADV_ATTR_TPMETER_COOKIE] = { .type = NLA_U32 }, }; /** @@ -163,6 +179,207 @@ batadv_netlink_get_mesh_info(struct sk_buff *skb, struct genl_info *info) return genlmsg_reply(msg, info); } +/** + * batadv_netlink_tp_meter_put - Fill information of started tp_meter session + * @msg: netlink message to be sent back + * @cookie: tp meter session cookie + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_tp_meter_put(struct sk_buff *msg, u32 cookie) +{ + if (nla_put_u32(msg, BATADV_ATTR_TPMETER_COOKIE, cookie)) + return -ENOBUFS; + + return 0; +} + +/** + * batadv_netlink_tpmeter_notify - send tp_meter result via netlink to client + * @bat_priv: the bat priv with all the soft interface information + * @dst: destination of tp_meter session + * @result: reason for tp meter session stop + * @test_time: total time ot the tp_meter session + * @total_bytes: bytes acked to the receiver + * @cookie: cookie of tp_meter session + * + * Return: 0 on success, < 0 on error + */ +int batadv_netlink_tpmeter_notify(struct batadv_priv *bat_priv, const u8 *dst, + u8 result, u32 test_time, u64 total_bytes, + u32 cookie) +{ + struct sk_buff *msg; + void *hdr; + int ret; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + hdr = genlmsg_put(msg, 0, 0, &batadv_netlink_family, 0, + BATADV_CMD_TP_METER); + if (!hdr) { + ret = -ENOBUFS; + goto err_genlmsg; + } + + if (nla_put_u32(msg, BATADV_ATTR_TPMETER_COOKIE, cookie)) + goto nla_put_failure; + + if (nla_put_u32(msg, BATADV_ATTR_TPMETER_TEST_TIME, test_time)) + goto nla_put_failure; + + if (nla_put_u64_64bit(msg, BATADV_ATTR_TPMETER_BYTES, total_bytes, + BATADV_ATTR_PAD)) + goto nla_put_failure; + + if (nla_put_u8(msg, BATADV_ATTR_TPMETER_RESULT, result)) + goto nla_put_failure; + + if (nla_put(msg, BATADV_ATTR_ORIG_ADDRESS, ETH_ALEN, dst)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + genlmsg_multicast_netns(&batadv_netlink_family, + dev_net(bat_priv->soft_iface), msg, 0, + BATADV_NL_MCGRP_TPMETER, GFP_KERNEL); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + ret = -EMSGSIZE; + +err_genlmsg: + nlmsg_free(msg); + return ret; +} + +/** + * batadv_netlink_tp_meter_start - Start a new tp_meter session + * @skb: received netlink message + * @info: receiver information + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_tp_meter_start(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct net_device *soft_iface; + struct batadv_priv *bat_priv; + struct sk_buff *msg = NULL; + u32 test_length; + void *msg_head; + int ifindex; + u32 cookie; + u8 *dst; + int ret; + + if (!info->attrs[BATADV_ATTR_MESH_IFINDEX]) + return -EINVAL; + + if (!info->attrs[BATADV_ATTR_ORIG_ADDRESS]) + return -EINVAL; + + if (!info->attrs[BATADV_ATTR_TPMETER_TEST_TIME]) + return -EINVAL; + + ifindex = nla_get_u32(info->attrs[BATADV_ATTR_MESH_IFINDEX]); + if (!ifindex) + return -EINVAL; + + dst = nla_data(info->attrs[BATADV_ATTR_ORIG_ADDRESS]); + + test_length = nla_get_u32(info->attrs[BATADV_ATTR_TPMETER_TEST_TIME]); + + soft_iface = dev_get_by_index(net, ifindex); + if (!soft_iface || !batadv_softif_is_valid(soft_iface)) { + ret = -ENODEV; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + ret = -ENOMEM; + goto out; + } + + msg_head = genlmsg_put(msg, info->snd_portid, info->snd_seq, + &batadv_netlink_family, 0, + BATADV_CMD_TP_METER); + if (!msg_head) { + ret = -ENOBUFS; + goto out; + } + + bat_priv = netdev_priv(soft_iface); + batadv_tp_start(bat_priv, dst, test_length, &cookie); + + ret = batadv_netlink_tp_meter_put(msg, cookie); + + out: + if (soft_iface) + dev_put(soft_iface); + + if (ret) { + if (msg) + nlmsg_free(msg); + return ret; + } + + genlmsg_end(msg, msg_head); + return genlmsg_reply(msg, info); +} + +/** + * batadv_netlink_tp_meter_start - Cancel a running tp_meter session + * @skb: received netlink message + * @info: receiver information + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_tp_meter_cancel(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct net_device *soft_iface; + struct batadv_priv *bat_priv; + int ifindex; + u8 *dst; + int ret = 0; + + if (!info->attrs[BATADV_ATTR_MESH_IFINDEX]) + return -EINVAL; + + if (!info->attrs[BATADV_ATTR_ORIG_ADDRESS]) + return -EINVAL; + + ifindex = nla_get_u32(info->attrs[BATADV_ATTR_MESH_IFINDEX]); + if (!ifindex) + return -EINVAL; + + dst = nla_data(info->attrs[BATADV_ATTR_ORIG_ADDRESS]); + + soft_iface = dev_get_by_index(net, ifindex); + if (!soft_iface || !batadv_softif_is_valid(soft_iface)) { + ret = -ENODEV; + goto out; + } + + bat_priv = netdev_priv(soft_iface); + batadv_tp_stop(bat_priv, dst, BATADV_TP_REASON_CANCEL); + +out: + if (soft_iface) + dev_put(soft_iface); + + return ret; +} + static struct genl_ops batadv_netlink_ops[] = { { .cmd = BATADV_CMD_GET_MESH_INFO, @@ -170,6 +387,18 @@ static struct genl_ops batadv_netlink_ops[] = { .policy = batadv_netlink_policy, .doit = batadv_netlink_get_mesh_info, }, + { + .cmd = BATADV_CMD_TP_METER, + .flags = GENL_ADMIN_PERM, + .policy = batadv_netlink_policy, + .doit = batadv_netlink_tp_meter_start, + }, + { + .cmd = BATADV_CMD_TP_METER_CANCEL, + .flags = GENL_ADMIN_PERM, + .policy = batadv_netlink_policy, + .doit = batadv_netlink_tp_meter_cancel, + }, }; /** @@ -179,8 +408,9 @@ void __init batadv_netlink_register(void) { int ret; - ret = genl_register_family_with_ops(&batadv_netlink_family, - batadv_netlink_ops); + ret = genl_register_family_with_ops_groups(&batadv_netlink_family, + batadv_netlink_ops, + batadv_netlink_mcgrps); if (ret) pr_warn("unable to register netlink family"); } diff --git a/net/batman-adv/netlink.h b/net/batman-adv/netlink.h index 39044ccff662..945653ab58c6 100644 --- a/net/batman-adv/netlink.h +++ b/net/batman-adv/netlink.h @@ -20,7 +20,13 @@ #include "main.h" +#include + void batadv_netlink_register(void); void batadv_netlink_unregister(void); +int batadv_netlink_tpmeter_notify(struct batadv_priv *bat_priv, const u8 *dst, + u8 result, u32 test_time, u64 total_bytes, + u32 cookie); + #endif /* _NET_BATMAN_ADV_NETLINK_H_ */ diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h index 71567794df17..6b011ff64dd8 100644 --- a/net/batman-adv/packet.h +++ b/net/batman-adv/packet.h @@ -21,6 +21,8 @@ #include #include +#define batadv_tp_is_error(n) ((u8)n > 127 ? 1 : 0) + /** * enum batadv_packettype - types for batman-adv encapsulated packets * @BATADV_IV_OGM: originator messages for B.A.T.M.A.N. IV @@ -93,6 +95,7 @@ enum batadv_icmp_packettype { BATADV_ECHO_REQUEST = 8, BATADV_TTL_EXCEEDED = 11, BATADV_PARAMETER_PROBLEM = 12, + BATADV_TP = 15, }; /** @@ -284,6 +287,16 @@ struct batadv_elp_packet { #define BATADV_ELP_HLEN sizeof(struct batadv_elp_packet) +/** + * enum batadv_icmp_user_cmd_type - types for batman-adv icmp cmd modes + * @BATADV_TP_START: start a throughput meter run + * @BATADV_TP_STOP: stop a throughput meter run + */ +enum batadv_icmp_user_cmd_type { + BATADV_TP_START = 0, + BATADV_TP_STOP = 2, +}; + /** * struct batadv_icmp_header - common members among all the ICMP packets * @packet_type: batman-adv packet type, part of the general header @@ -334,6 +347,47 @@ struct batadv_icmp_packet { __be16 seqno; }; +/** + * struct batadv_icmp_tp_packet - ICMP TP Meter packet + * @packet_type: batman-adv packet type, part of the general header + * @version: batman-adv protocol version, part of the genereal header + * @ttl: time to live for this packet, part of the genereal header + * @msg_type: ICMP packet type + * @dst: address of the destination node + * @orig: address of the source node + * @uid: local ICMP socket identifier + * @subtype: TP packet subtype (see batadv_icmp_tp_subtype) + * @session: TP session identifier + * @seqno: the TP sequence number + * @timestamp: time when the packet has been sent. This value is filled in a + * TP_MSG and echoed back in the next TP_ACK so that the sender can compute the + * RTT. Since it is read only by the host which wrote it, there is no need to + * store it using network order + */ +struct batadv_icmp_tp_packet { + u8 packet_type; + u8 version; + u8 ttl; + u8 msg_type; /* see ICMP message types above */ + u8 dst[ETH_ALEN]; + u8 orig[ETH_ALEN]; + u8 uid; + u8 subtype; + u8 session[2]; + __be32 seqno; + __be32 timestamp; +}; + +/** + * enum batadv_icmp_tp_subtype - ICMP TP Meter packet subtypes + * @BATADV_TP_MSG: Msg from sender to receiver + * @BATADV_TP_ACK: acknowledgment from receiver to sender + */ +enum batadv_icmp_tp_subtype { + BATADV_TP_MSG = 0, + BATADV_TP_ACK, +}; + #define BATADV_RR_LEN 16 /** diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 76de583fe866..7b5de402ee0d 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -46,6 +46,7 @@ #include "packet.h" #include "send.h" #include "soft-interface.h" +#include "tp_meter.h" #include "translation-table.h" #include "tvlv.h" @@ -276,6 +277,13 @@ static int batadv_recv_my_icmp_packet(struct batadv_priv *bat_priv, ret = NET_RX_SUCCESS; break; + case BATADV_TP: + if (!pskb_may_pull(skb, sizeof(struct batadv_icmp_tp_packet))) + goto out; + + batadv_tp_meter_recv(bat_priv, skb); + ret = NET_RX_SUCCESS; + goto out; default: /* drop unknown type */ goto out; diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index f75631e21e48..18b6d07c3233 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -842,6 +842,8 @@ static int batadv_softif_init_late(struct net_device *dev) #ifdef CONFIG_BATMAN_ADV_BLA atomic_set(&bat_priv->bla.num_requests, 0); #endif + atomic_set(&bat_priv->tp_num, 0); + bat_priv->tt.last_changeset = NULL; bat_priv->tt.last_changeset_len = 0; bat_priv->isolation_mark = 0; diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c new file mode 100644 index 000000000000..2333777f919d --- /dev/null +++ b/net/batman-adv/tp_meter.c @@ -0,0 +1,1507 @@ +/* Copyright (C) 2012-2016 B.A.T.M.A.N. contributors: + * + * Edo Monticelli, Antonio Quartulli + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "tp_meter.h" +#include "main.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "hard-interface.h" +#include "log.h" +#include "netlink.h" +#include "originator.h" +#include "packet.h" +#include "send.h" + +/** + * BATADV_TP_DEF_TEST_LENGTH - Default test length if not specified by the user + * in milliseconds + */ +#define BATADV_TP_DEF_TEST_LENGTH 10000 + +/** + * BATADV_TP_AWND - Advertised window by the receiver (in bytes) + */ +#define BATADV_TP_AWND 0x20000000 + +/** + * BATADV_TP_RECV_TIMEOUT - Receiver activity timeout. If the receiver does not + * get anything for such amount of milliseconds, the connection is killed + */ +#define BATADV_TP_RECV_TIMEOUT 1000 + +/** + * BATADV_TP_MAX_RTO - Maximum sender timeout. If the sender RTO gets beyond + * such amound of milliseconds, the receiver is considered unreachable and the + * connection is killed + */ +#define BATADV_TP_MAX_RTO 30000 + +/** + * BATADV_TP_FIRST_SEQ - First seqno of each session. The number is rather high + * in order to immediately trigger a wrap around (test purposes) + */ +#define BATADV_TP_FIRST_SEQ ((u32)-1 - 2000) + +/** + * BATADV_TP_PLEN - length of the payload (data after the batadv_unicast header) + * to simulate + */ +#define BATADV_TP_PLEN (BATADV_TP_PACKET_LEN - ETH_HLEN - \ + sizeof(struct batadv_unicast_packet)) + +static u8 batadv_tp_prerandom[4096] __read_mostly; + +/** + * batadv_tp_session_cookie - generate session cookie based on session ids + * @session: TP session identifier + * @icmp_uid: icmp pseudo uid of the tp session + * + * Return: 32 bit tp_meter session cookie + */ +static u32 batadv_tp_session_cookie(const u8 session[2], u8 icmp_uid) +{ + u32 cookie; + + cookie = icmp_uid << 16; + cookie |= session[0] << 8; + cookie |= session[1]; + + return cookie; +} + +/** + * batadv_tp_cwnd - compute the new cwnd size + * @base: base cwnd size value + * @increment: the value to add to base to get the new size + * @min: minumim cwnd value (usually MSS) + * + * Return the new cwnd size and ensures it does not exceed the Advertised + * Receiver Window size. It is wrap around safe. + * For details refer to Section 3.1 of RFC5681 + * + * Return: new congestion window size in bytes + */ +static u32 batadv_tp_cwnd(u32 base, u32 increment, u32 min) +{ + u32 new_size = base + increment; + + /* check for wrap-around */ + if (new_size < base) + new_size = (u32)ULONG_MAX; + + new_size = min_t(u32, new_size, BATADV_TP_AWND); + + return max_t(u32, new_size, min); +} + +/** + * batadv_tp_updated_cwnd - update the Congestion Windows + * @tp_vars: the private data of the current TP meter session + * @mss: maximum segment size of transmission + * + * 1) if the session is in Slow Start, the CWND has to be increased by 1 + * MSS every unique received ACK + * 2) if the session is in Congestion Avoidance, the CWND has to be + * increased by MSS * MSS / CWND for every unique received ACK + */ +static void batadv_tp_update_cwnd(struct batadv_tp_vars *tp_vars, u32 mss) +{ + spin_lock_bh(&tp_vars->cwnd_lock); + + /* slow start... */ + if (tp_vars->cwnd <= tp_vars->ss_threshold) { + tp_vars->dec_cwnd = 0; + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->cwnd, mss, mss); + spin_unlock_bh(&tp_vars->cwnd_lock); + return; + } + + /* increment CWND at least of 1 (section 3.1 of RFC5681) */ + tp_vars->dec_cwnd += max_t(u32, 1U << 3, + ((mss * mss) << 6) / (tp_vars->cwnd << 3)); + if (tp_vars->dec_cwnd < (mss << 3)) { + spin_unlock_bh(&tp_vars->cwnd_lock); + return; + } + + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->cwnd, mss, mss); + tp_vars->dec_cwnd = 0; + + spin_unlock_bh(&tp_vars->cwnd_lock); +} + +/** + * batadv_tp_update_rto - calculate new retransmission timeout + * @tp_vars: the private data of the current TP meter session + * @new_rtt: new roundtrip time in msec + */ +static void batadv_tp_update_rto(struct batadv_tp_vars *tp_vars, + u32 new_rtt) +{ + long m = new_rtt; + + /* RTT update + * Details in Section 2.2 and 2.3 of RFC6298 + * + * It's tricky to understand. Don't lose hair please. + * Inspired by tcp_rtt_estimator() tcp_input.c + */ + if (tp_vars->srtt != 0) { + m -= (tp_vars->srtt >> 3); /* m is now error in rtt est */ + tp_vars->srtt += m; /* rtt = 7/8 srtt + 1/8 new */ + if (m < 0) + m = -m; + + m -= (tp_vars->rttvar >> 2); + tp_vars->rttvar += m; /* mdev ~= 3/4 rttvar + 1/4 new */ + } else { + /* first measure getting in */ + tp_vars->srtt = m << 3; /* take the measured time to be srtt */ + tp_vars->rttvar = m << 1; /* new_rtt / 2 */ + } + + /* rto = srtt + 4 * rttvar. + * rttvar is scaled by 4, therefore doesn't need to be multiplied + */ + tp_vars->rto = (tp_vars->srtt >> 3) + tp_vars->rttvar; +} + +/** + * batadv_tp_batctl_notify - send client status result to client + * @reason: reason for tp meter session stop + * @dst: destination of tp_meter session + * @bat_priv: the bat priv with all the soft interface information + * @start_time: start of transmission in jiffies + * @total_sent: bytes acked to the receiver + * @cookie: cookie of tp_meter session + */ +static void batadv_tp_batctl_notify(enum batadv_tp_meter_reason reason, + const u8 *dst, struct batadv_priv *bat_priv, + unsigned long start_time, u64 total_sent, + u32 cookie) +{ + u32 test_time; + u8 result; + u32 total_bytes; + + if (!batadv_tp_is_error(reason)) { + result = BATADV_TP_REASON_COMPLETE; + test_time = jiffies_to_msecs(jiffies - start_time); + total_bytes = total_sent; + } else { + result = reason; + test_time = 0; + total_bytes = 0; + } + + batadv_netlink_tpmeter_notify(bat_priv, dst, result, test_time, + total_bytes, cookie); +} + +/** + * batadv_tp_batctl_error_notify - send client error result to client + * @reason: reason for tp meter session stop + * @dst: destination of tp_meter session + * @bat_priv: the bat priv with all the soft interface information + * @cookie: cookie of tp_meter session + */ +static void batadv_tp_batctl_error_notify(enum batadv_tp_meter_reason reason, + const u8 *dst, + struct batadv_priv *bat_priv, + u32 cookie) +{ + batadv_tp_batctl_notify(reason, dst, bat_priv, 0, 0, cookie); +} + +/** + * batadv_tp_list_find - find a tp_vars object in the global list + * @bat_priv: the bat priv with all the soft interface information + * @dst: the other endpoint MAC address to look for + * + * Look for a tp_vars object matching dst as end_point and return it after + * having incremented the refcounter. Return NULL is not found + * + * Return: matching tp_vars or NULL when no tp_vars with @dst was found + */ +static struct batadv_tp_vars *batadv_tp_list_find(struct batadv_priv *bat_priv, + const u8 *dst) +{ + struct batadv_tp_vars *pos, *tp_vars = NULL; + + rcu_read_lock(); + hlist_for_each_entry_rcu(pos, &bat_priv->tp_list, list) { + if (!batadv_compare_eth(pos->other_end, dst)) + continue; + + /* most of the time this function is invoked during the normal + * process..it makes sens to pay more when the session is + * finished and to speed the process up during the measurement + */ + if (unlikely(!kref_get_unless_zero(&pos->refcount))) + continue; + + tp_vars = pos; + break; + } + rcu_read_unlock(); + + return tp_vars; +} + +/** + * batadv_tp_list_find_session - find tp_vars session object in the global list + * @bat_priv: the bat priv with all the soft interface information + * @dst: the other endpoint MAC address to look for + * @session: session identifier + * + * Look for a tp_vars object matching dst as end_point, session as tp meter + * session and return it after having incremented the refcounter. Return NULL + * is not found + * + * Return: matching tp_vars or NULL when no tp_vars was found + */ +static struct batadv_tp_vars * +batadv_tp_list_find_session(struct batadv_priv *bat_priv, const u8 *dst, + const u8 *session) +{ + struct batadv_tp_vars *pos, *tp_vars = NULL; + + rcu_read_lock(); + hlist_for_each_entry_rcu(pos, &bat_priv->tp_list, list) { + if (!batadv_compare_eth(pos->other_end, dst)) + continue; + + if (memcmp(pos->session, session, sizeof(pos->session)) != 0) + continue; + + /* most of the time this function is invoked during the normal + * process..it makes sense to pay more when the session is + * finished and to speed the process up during the measurement + */ + if (unlikely(!kref_get_unless_zero(&pos->refcount))) + continue; + + tp_vars = pos; + break; + } + rcu_read_unlock(); + + return tp_vars; +} + +/** + * batadv_tp_vars_release - release batadv_tp_vars from lists and queue for + * free after rcu grace period + * @ref: kref pointer of the batadv_tp_vars + */ +static void batadv_tp_vars_release(struct kref *ref) +{ + struct batadv_tp_vars *tp_vars; + struct batadv_tp_unacked *un, *safe; + + tp_vars = container_of(ref, struct batadv_tp_vars, refcount); + + /* lock should not be needed because this object is now out of any + * context! + */ + spin_lock_bh(&tp_vars->unacked_lock); + list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { + list_del(&un->list); + kfree(un); + } + spin_unlock_bh(&tp_vars->unacked_lock); + + kfree_rcu(tp_vars, rcu); +} + +/** + * batadv_tp_vars_put - decrement the batadv_tp_vars refcounter and possibly + * release it + * @tp_vars: the private data of the current TP meter session to be free'd + */ +static void batadv_tp_vars_put(struct batadv_tp_vars *tp_vars) +{ + kref_put(&tp_vars->refcount, batadv_tp_vars_release); +} + +/** + * batadv_tp_sender_cleanup - cleanup sender data and drop and timer + * @bat_priv: the bat priv with all the soft interface information + * @tp_vars: the private data of the current TP meter session to cleanup + */ +static void batadv_tp_sender_cleanup(struct batadv_priv *bat_priv, + struct batadv_tp_vars *tp_vars) +{ + cancel_delayed_work(&tp_vars->finish_work); + + spin_lock_bh(&tp_vars->bat_priv->tp_list_lock); + hlist_del_rcu(&tp_vars->list); + spin_unlock_bh(&tp_vars->bat_priv->tp_list_lock); + + /* drop list reference */ + batadv_tp_vars_put(tp_vars); + + atomic_dec(&tp_vars->bat_priv->tp_num); + + /* kill the timer and remove its reference */ + del_timer_sync(&tp_vars->timer); + /* the worker might have rearmed itself therefore we kill it again. Note + * that if the worker should run again before invoking the following + * del_timer(), it would not re-arm itself once again because the status + * is OFF now + */ + del_timer(&tp_vars->timer); + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_sender_end - print info about ended session and inform client + * @bat_priv: the bat priv with all the soft interface information + * @tp_vars: the private data of the current TP meter session + */ +static void batadv_tp_sender_end(struct batadv_priv *bat_priv, + struct batadv_tp_vars *tp_vars) +{ + u32 session_cookie; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Test towards %pM finished..shutting down (reason=%d)\n", + tp_vars->other_end, tp_vars->reason); + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Last timing stats: SRTT=%ums RTTVAR=%ums RTO=%ums\n", + tp_vars->srtt >> 3, tp_vars->rttvar >> 2, tp_vars->rto); + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Final values: cwnd=%u ss_threshold=%u\n", + tp_vars->cwnd, tp_vars->ss_threshold); + + session_cookie = batadv_tp_session_cookie(tp_vars->session, + tp_vars->icmp_uid); + + batadv_tp_batctl_notify(tp_vars->reason, + tp_vars->other_end, + bat_priv, + tp_vars->start_time, + atomic64_read(&tp_vars->tot_sent), + session_cookie); +} + +/** + * batadv_tp_sender_shutdown - let sender thread/timer stop gracefully + * @tp_vars: the private data of the current TP meter session + * @reason: reason for tp meter session stop + */ +static void batadv_tp_sender_shutdown(struct batadv_tp_vars *tp_vars, + enum batadv_tp_meter_reason reason) +{ + if (!atomic_dec_and_test(&tp_vars->sending)) + return; + + tp_vars->reason = reason; +} + +/** + * batadv_tp_sender_finish - stop sender session after test_length was reached + * @work: delayed work reference of the related tp_vars + */ +static void batadv_tp_sender_finish(struct work_struct *work) +{ + struct delayed_work *delayed_work; + struct batadv_tp_vars *tp_vars; + + delayed_work = to_delayed_work(work); + tp_vars = container_of(delayed_work, struct batadv_tp_vars, + finish_work); + + batadv_tp_sender_shutdown(tp_vars, BATADV_TP_REASON_COMPLETE); +} + +/** + * batadv_tp_reset_sender_timer - reschedule the sender timer + * @tp_vars: the private TP meter data for this session + * + * Reschedule the timer using tp_vars->rto as delay + */ +static void batadv_tp_reset_sender_timer(struct batadv_tp_vars *tp_vars) +{ + /* most of the time this function is invoked while normal packet + * reception... + */ + if (unlikely(atomic_read(&tp_vars->sending) == 0)) + /* timer ref will be dropped in batadv_tp_sender_cleanup */ + return; + + mod_timer(&tp_vars->timer, jiffies + msecs_to_jiffies(tp_vars->rto)); +} + +/** + * batadv_tp_sender_timeout - timer that fires in case of packet loss + * @arg: address of the related tp_vars + * + * If fired it means that there was packet loss. + * Switch to Slow Start, set the ss_threshold to half of the current cwnd and + * reset the cwnd to 3*MSS + */ +static void batadv_tp_sender_timeout(unsigned long arg) +{ + struct batadv_tp_vars *tp_vars = (struct batadv_tp_vars *)arg; + struct batadv_priv *bat_priv = tp_vars->bat_priv; + + if (atomic_read(&tp_vars->sending) == 0) + return; + + /* if the user waited long enough...shutdown the test */ + if (unlikely(tp_vars->rto >= BATADV_TP_MAX_RTO)) { + batadv_tp_sender_shutdown(tp_vars, + BATADV_TP_REASON_DST_UNREACHABLE); + return; + } + + /* RTO exponential backoff + * Details in Section 5.5 of RFC6298 + */ + tp_vars->rto <<= 1; + + spin_lock_bh(&tp_vars->cwnd_lock); + + tp_vars->ss_threshold = tp_vars->cwnd >> 1; + if (tp_vars->ss_threshold < BATADV_TP_PLEN * 2) + tp_vars->ss_threshold = BATADV_TP_PLEN * 2; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: RTO fired during test towards %pM! cwnd=%u new ss_thr=%u, resetting last_sent to %u\n", + tp_vars->other_end, tp_vars->cwnd, tp_vars->ss_threshold, + atomic_read(&tp_vars->last_acked)); + + tp_vars->cwnd = BATADV_TP_PLEN * 3; + + spin_unlock_bh(&tp_vars->cwnd_lock); + + /* resend the non-ACKed packets.. */ + tp_vars->last_sent = atomic_read(&tp_vars->last_acked); + wake_up(&tp_vars->more_bytes); + + batadv_tp_reset_sender_timer(tp_vars); +} + +/** + * batadv_tp_fill_prerandom - Fill buffer with prefetched random bytes + * @tp_vars: the private TP meter data for this session + * @buf: Buffer to fill with bytes + * @nbytes: amount of pseudorandom bytes + */ +static void batadv_tp_fill_prerandom(struct batadv_tp_vars *tp_vars, + u8 *buf, size_t nbytes) +{ + u32 local_offset; + size_t bytes_inbuf; + size_t to_copy; + size_t pos = 0; + + spin_lock_bh(&tp_vars->prerandom_lock); + local_offset = tp_vars->prerandom_offset; + tp_vars->prerandom_offset += nbytes; + tp_vars->prerandom_offset %= sizeof(batadv_tp_prerandom); + spin_unlock_bh(&tp_vars->prerandom_lock); + + while (nbytes) { + local_offset %= sizeof(batadv_tp_prerandom); + bytes_inbuf = sizeof(batadv_tp_prerandom) - local_offset; + to_copy = min(nbytes, bytes_inbuf); + + memcpy(&buf[pos], &batadv_tp_prerandom[local_offset], to_copy); + pos += to_copy; + nbytes -= to_copy; + local_offset = 0; + } +} + +/** + * batadv_tp_send_msg - send a single message + * @tp_vars: the private TP meter data for this session + * @src: source mac address + * @orig_node: the originator of the destination + * @seqno: sequence number of this packet + * @len: length of the entire packet + * @session: session identifier + * @uid: local ICMP "socket" index + * @timestamp: timestamp in jiffies which is replied in ack + * + * Create and send a single TP Meter message. + * + * Return: 0 on success, BATADV_TP_REASON_DST_UNREACHABLE if the destination is + * not reachable, BATADV_TP_REASON_MEMORY_ERROR if the packet couldn't be + * allocated + */ +static int batadv_tp_send_msg(struct batadv_tp_vars *tp_vars, const u8 *src, + struct batadv_orig_node *orig_node, + u32 seqno, size_t len, const u8 *session, + int uid, u32 timestamp) +{ + struct batadv_icmp_tp_packet *icmp; + struct sk_buff *skb; + int r; + u8 *data; + size_t data_len; + + skb = netdev_alloc_skb_ip_align(NULL, len + ETH_HLEN); + if (unlikely(!skb)) + return BATADV_TP_REASON_MEMORY_ERROR; + + skb_reserve(skb, ETH_HLEN); + icmp = (struct batadv_icmp_tp_packet *)skb_put(skb, sizeof(*icmp)); + + /* fill the icmp header */ + ether_addr_copy(icmp->dst, orig_node->orig); + ether_addr_copy(icmp->orig, src); + icmp->version = BATADV_COMPAT_VERSION; + icmp->packet_type = BATADV_ICMP; + icmp->ttl = BATADV_TTL; + icmp->msg_type = BATADV_TP; + icmp->uid = uid; + + icmp->subtype = BATADV_TP_MSG; + memcpy(icmp->session, session, sizeof(icmp->session)); + icmp->seqno = htonl(seqno); + icmp->timestamp = htonl(timestamp); + + data_len = len - sizeof(*icmp); + data = (u8 *)skb_put(skb, data_len); + batadv_tp_fill_prerandom(tp_vars, data, data_len); + + r = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (r == -1) + kfree_skb(skb); + + if (r == NET_XMIT_SUCCESS) + return 0; + + return BATADV_TP_REASON_CANT_SEND; +} + +/** + * batadv_tp_recv_ack - ACK receiving function + * @bat_priv: the bat priv with all the soft interface information + * @skb: the buffer containing the received packet + * + * Process a received TP ACK packet + */ +static void batadv_tp_recv_ack(struct batadv_priv *bat_priv, + const struct sk_buff *skb) +{ + struct batadv_hard_iface *primary_if = NULL; + struct batadv_orig_node *orig_node = NULL; + const struct batadv_icmp_tp_packet *icmp; + struct batadv_tp_vars *tp_vars; + size_t packet_len, mss; + u32 rtt, recv_ack, cwnd; + unsigned char *dev_addr; + + packet_len = BATADV_TP_PLEN; + mss = BATADV_TP_PLEN; + packet_len += sizeof(struct batadv_unicast_packet); + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + /* find the tp_vars */ + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, + icmp->session); + if (unlikely(!tp_vars)) + return; + + if (unlikely(atomic_read(&tp_vars->sending) == 0)) + goto out; + + /* old ACK? silently drop it.. */ + if (batadv_seq_before(ntohl(icmp->seqno), + (u32)atomic_read(&tp_vars->last_acked))) + goto out; + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (unlikely(!primary_if)) + goto out; + + orig_node = batadv_orig_hash_find(bat_priv, icmp->orig); + if (unlikely(!orig_node)) + goto out; + + /* update RTO with the new sampled RTT, if any */ + rtt = jiffies_to_msecs(jiffies) - ntohl(icmp->timestamp); + if (icmp->timestamp && rtt) + batadv_tp_update_rto(tp_vars, rtt); + + /* ACK for new data... reset the timer */ + batadv_tp_reset_sender_timer(tp_vars); + + recv_ack = ntohl(icmp->seqno); + + /* check if this ACK is a duplicate */ + if (atomic_read(&tp_vars->last_acked) == recv_ack) { + atomic_inc(&tp_vars->dup_acks); + if (atomic_read(&tp_vars->dup_acks) != 3) + goto out; + + if (recv_ack >= tp_vars->recover) + goto out; + + /* if this is the third duplicate ACK do Fast Retransmit */ + batadv_tp_send_msg(tp_vars, primary_if->net_dev->dev_addr, + orig_node, recv_ack, packet_len, + icmp->session, icmp->uid, + jiffies_to_msecs(jiffies)); + + spin_lock_bh(&tp_vars->cwnd_lock); + + /* Fast Recovery */ + tp_vars->fast_recovery = true; + /* Set recover to the last outstanding seqno when Fast Recovery + * is entered. RFC6582, Section 3.2, step 1 + */ + tp_vars->recover = tp_vars->last_sent; + tp_vars->ss_threshold = tp_vars->cwnd >> 1; + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: Fast Recovery, (cur cwnd=%u) ss_thr=%u last_sent=%u recv_ack=%u\n", + tp_vars->cwnd, tp_vars->ss_threshold, + tp_vars->last_sent, recv_ack); + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->ss_threshold, 3 * mss, + mss); + tp_vars->dec_cwnd = 0; + tp_vars->last_sent = recv_ack; + + spin_unlock_bh(&tp_vars->cwnd_lock); + } else { + /* count the acked data */ + atomic64_add(recv_ack - atomic_read(&tp_vars->last_acked), + &tp_vars->tot_sent); + /* reset the duplicate ACKs counter */ + atomic_set(&tp_vars->dup_acks, 0); + + if (tp_vars->fast_recovery) { + /* partial ACK */ + if (batadv_seq_before(recv_ack, tp_vars->recover)) { + /* this is another hole in the window. React + * immediately as specified by NewReno (see + * Section 3.2 of RFC6582 for details) + */ + dev_addr = primary_if->net_dev->dev_addr; + batadv_tp_send_msg(tp_vars, dev_addr, + orig_node, recv_ack, + packet_len, icmp->session, + icmp->uid, + jiffies_to_msecs(jiffies)); + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->cwnd, + mss, mss); + } else { + tp_vars->fast_recovery = false; + /* set cwnd to the value of ss_threshold at the + * moment that Fast Recovery was entered. + * RFC6582, Section 3.2, step 3 + */ + cwnd = batadv_tp_cwnd(tp_vars->ss_threshold, 0, + mss); + tp_vars->cwnd = cwnd; + } + goto move_twnd; + } + + if (recv_ack - atomic_read(&tp_vars->last_acked) >= mss) + batadv_tp_update_cwnd(tp_vars, mss); +move_twnd: + /* move the Transmit Window */ + atomic_set(&tp_vars->last_acked, recv_ack); + } + + wake_up(&tp_vars->more_bytes); +out: + if (likely(primary_if)) + batadv_hardif_put(primary_if); + if (likely(orig_node)) + batadv_orig_node_put(orig_node); + if (likely(tp_vars)) + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_avail - check if congestion window is not full + * @tp_vars: the private data of the current TP meter session + * @payload_len: size of the payload of a single message + * + * Return: true when congestion window is not full, false otherwise + */ +static bool batadv_tp_avail(struct batadv_tp_vars *tp_vars, + size_t payload_len) +{ + u32 win_left, win_limit; + + win_limit = atomic_read(&tp_vars->last_acked) + tp_vars->cwnd; + win_left = win_limit - tp_vars->last_sent; + + return win_left >= payload_len; +} + +/** + * batadv_tp_wait_available - wait until congestion window becomes free or + * timeout is reached + * @tp_vars: the private data of the current TP meter session + * @plen: size of the payload of a single message + * + * Return: 0 if the condition evaluated to false after the timeout elapsed, + * 1 if the condition evaluated to true after the timeout elapsed, the + * remaining jiffies (at least 1) if the condition evaluated to true before + * the timeout elapsed, or -ERESTARTSYS if it was interrupted by a signal. + */ +static int batadv_tp_wait_available(struct batadv_tp_vars *tp_vars, size_t plen) +{ + int ret; + + ret = wait_event_interruptible_timeout(tp_vars->more_bytes, + batadv_tp_avail(tp_vars, plen), + HZ / 10); + + return ret; +} + +/** + * batadv_tp_send - main sending thread of a tp meter session + * @arg: address of the related tp_vars + * + * Return: nothing, this function never returns + */ +static int batadv_tp_send(void *arg) +{ + struct batadv_tp_vars *tp_vars = arg; + struct batadv_priv *bat_priv = tp_vars->bat_priv; + struct batadv_hard_iface *primary_if = NULL; + struct batadv_orig_node *orig_node = NULL; + size_t payload_len, packet_len; + int err = 0; + + if (unlikely(tp_vars->role != BATADV_TP_SENDER)) { + err = BATADV_TP_REASON_DST_UNREACHABLE; + tp_vars->reason = err; + goto out; + } + + orig_node = batadv_orig_hash_find(bat_priv, tp_vars->other_end); + if (unlikely(!orig_node)) { + err = BATADV_TP_REASON_DST_UNREACHABLE; + tp_vars->reason = err; + goto out; + } + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (unlikely(!primary_if)) { + err = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + + /* assume that all the hard_interfaces have a correctly + * configured MTU, so use the soft_iface MTU as MSS. + * This might not be true and in that case the fragmentation + * should be used. + * Now, try to send the packet as it is + */ + payload_len = BATADV_TP_PLEN; + BUILD_BUG_ON(sizeof(struct batadv_icmp_tp_packet) > BATADV_TP_PLEN); + + batadv_tp_reset_sender_timer(tp_vars); + + /* queue the worker in charge of terminating the test */ + queue_delayed_work(batadv_event_workqueue, &tp_vars->finish_work, + msecs_to_jiffies(tp_vars->test_length)); + + while (atomic_read(&tp_vars->sending) != 0) { + if (unlikely(!batadv_tp_avail(tp_vars, payload_len))) { + batadv_tp_wait_available(tp_vars, payload_len); + continue; + } + + /* to emulate normal unicast traffic, add to the payload len + * the size of the unicast header + */ + packet_len = payload_len + sizeof(struct batadv_unicast_packet); + + err = batadv_tp_send_msg(tp_vars, primary_if->net_dev->dev_addr, + orig_node, tp_vars->last_sent, + packet_len, + tp_vars->session, tp_vars->icmp_uid, + jiffies_to_msecs(jiffies)); + + /* something went wrong during the preparation/transmission */ + if (unlikely(err && err != BATADV_TP_REASON_CANT_SEND)) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: batadv_tp_send() cannot send packets (%d)\n", + err); + /* ensure nobody else tries to stop the thread now */ + if (atomic_dec_and_test(&tp_vars->sending)) + tp_vars->reason = err; + break; + } + + /* right-shift the TWND */ + if (!err) + tp_vars->last_sent += payload_len; + + cond_resched(); + } + +out: + if (likely(primary_if)) + batadv_hardif_put(primary_if); + if (likely(orig_node)) + batadv_orig_node_put(orig_node); + + batadv_tp_sender_end(bat_priv, tp_vars); + batadv_tp_sender_cleanup(bat_priv, tp_vars); + + batadv_tp_vars_put(tp_vars); + + do_exit(0); +} + +/** + * batadv_tp_start_kthread - start new thread which manages the tp meter sender + * @tp_vars: the private data of the current TP meter session + */ +static void batadv_tp_start_kthread(struct batadv_tp_vars *tp_vars) +{ + struct task_struct *kthread; + struct batadv_priv *bat_priv = tp_vars->bat_priv; + u32 session_cookie; + + kref_get(&tp_vars->refcount); + kthread = kthread_create(batadv_tp_send, tp_vars, "kbatadv_tp_meter"); + if (IS_ERR(kthread)) { + session_cookie = batadv_tp_session_cookie(tp_vars->session, + tp_vars->icmp_uid); + pr_err("batadv: cannot create tp meter kthread\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_MEMORY_ERROR, + tp_vars->other_end, + bat_priv, session_cookie); + + /* drop reserved reference for kthread */ + batadv_tp_vars_put(tp_vars); + + /* cleanup of failed tp meter variables */ + batadv_tp_sender_cleanup(bat_priv, tp_vars); + return; + } + + wake_up_process(kthread); +} + +/** + * batadv_tp_start - start a new tp meter session + * @bat_priv: the bat priv with all the soft interface information + * @dst: the receiver MAC address + * @test_length: test length in milliseconds + * @cookie: session cookie + */ +void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, + u32 test_length, u32 *cookie) +{ + struct batadv_tp_vars *tp_vars; + u8 session_id[2]; + u8 icmp_uid; + u32 session_cookie; + + get_random_bytes(session_id, sizeof(session_id)); + get_random_bytes(&icmp_uid, 1); + session_cookie = batadv_tp_session_cookie(session_id, icmp_uid); + *cookie = session_cookie; + + /* look for an already existing test towards this node */ + spin_lock_bh(&bat_priv->tp_list_lock); + tp_vars = batadv_tp_list_find(bat_priv, dst); + if (tp_vars) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_tp_vars_put(tp_vars); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: test to or from the same node already ongoing, aborting\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_ALREADY_ONGOING, + dst, bat_priv, session_cookie); + return; + } + + if (!atomic_add_unless(&bat_priv->tp_num, 1, BATADV_TP_MAX_NUM)) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: too many ongoing sessions, aborting (SEND)\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_TOO_MANY, dst, + bat_priv, session_cookie); + return; + } + + tp_vars = kmalloc(sizeof(*tp_vars), GFP_ATOMIC); + if (!tp_vars) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: batadv_tp_start cannot allocate list elements\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_MEMORY_ERROR, + dst, bat_priv, session_cookie); + return; + } + + /* initialize tp_vars */ + ether_addr_copy(tp_vars->other_end, dst); + kref_init(&tp_vars->refcount); + tp_vars->role = BATADV_TP_SENDER; + atomic_set(&tp_vars->sending, 1); + memcpy(tp_vars->session, session_id, sizeof(session_id)); + tp_vars->icmp_uid = icmp_uid; + + tp_vars->last_sent = BATADV_TP_FIRST_SEQ; + atomic_set(&tp_vars->last_acked, BATADV_TP_FIRST_SEQ); + tp_vars->fast_recovery = false; + tp_vars->recover = BATADV_TP_FIRST_SEQ; + + /* initialise the CWND to 3*MSS (Section 3.1 in RFC5681). + * For batman-adv the MSS is the size of the payload received by the + * soft_interface, hence its MTU + */ + tp_vars->cwnd = BATADV_TP_PLEN * 3; + /* at the beginning initialise the SS threshold to the biggest possible + * window size, hence the AWND size + */ + tp_vars->ss_threshold = BATADV_TP_AWND; + + /* RTO initial value is 3 seconds. + * Details in Section 2.1 of RFC6298 + */ + tp_vars->rto = 1000; + tp_vars->srtt = 0; + tp_vars->rttvar = 0; + + atomic64_set(&tp_vars->tot_sent, 0); + + kref_get(&tp_vars->refcount); + setup_timer(&tp_vars->timer, batadv_tp_sender_timeout, + (unsigned long)tp_vars); + + tp_vars->bat_priv = bat_priv; + tp_vars->start_time = jiffies; + + init_waitqueue_head(&tp_vars->more_bytes); + + spin_lock_init(&tp_vars->unacked_lock); + INIT_LIST_HEAD(&tp_vars->unacked_list); + + spin_lock_init(&tp_vars->cwnd_lock); + + tp_vars->prerandom_offset = 0; + spin_lock_init(&tp_vars->prerandom_lock); + + kref_get(&tp_vars->refcount); + hlist_add_head_rcu(&tp_vars->list, &bat_priv->tp_list); + spin_unlock_bh(&bat_priv->tp_list_lock); + + tp_vars->test_length = test_length; + if (!tp_vars->test_length) + tp_vars->test_length = BATADV_TP_DEF_TEST_LENGTH; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: starting throughput meter towards %pM (length=%ums)\n", + dst, test_length); + + /* init work item for finished tp tests */ + INIT_DELAYED_WORK(&tp_vars->finish_work, batadv_tp_sender_finish); + + /* start tp kthread. This way the write() call issued from userspace can + * happily return and avoid to block + */ + batadv_tp_start_kthread(tp_vars); + + /* don't return reference to new tp_vars */ + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_stop - stop currently running tp meter session + * @bat_priv: the bat priv with all the soft interface information + * @dst: the receiver MAC address + * @return_value: reason for tp meter session stop + */ +void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, + u8 return_value) +{ + struct batadv_orig_node *orig_node; + struct batadv_tp_vars *tp_vars; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: stopping test towards %pM\n", dst); + + orig_node = batadv_orig_hash_find(bat_priv, dst); + if (!orig_node) + return; + + tp_vars = batadv_tp_list_find(bat_priv, orig_node->orig); + if (!tp_vars) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: trying to interrupt an already over connection\n"); + goto out; + } + + batadv_tp_sender_shutdown(tp_vars, return_value); + batadv_tp_vars_put(tp_vars); +out: + batadv_orig_node_put(orig_node); +} + +/** + * batadv_tp_reset_receiver_timer - reset the receiver shutdown timer + * @tp_vars: the private data of the current TP meter session + * + * start the receiver shutdown timer or reset it if already started + */ +static void batadv_tp_reset_receiver_timer(struct batadv_tp_vars *tp_vars) +{ + mod_timer(&tp_vars->timer, + jiffies + msecs_to_jiffies(BATADV_TP_RECV_TIMEOUT)); +} + +/** + * batadv_tp_receiver_shutdown - stop a tp meter receiver when timeout is + * reached without received ack + * @arg: address of the related tp_vars + */ +static void batadv_tp_receiver_shutdown(unsigned long arg) +{ + struct batadv_tp_vars *tp_vars = (struct batadv_tp_vars *)arg; + struct batadv_tp_unacked *un, *safe; + struct batadv_priv *bat_priv; + + bat_priv = tp_vars->bat_priv; + + /* if there is recent activity rearm the timer */ + if (!batadv_has_timed_out(tp_vars->last_recv_time, + BATADV_TP_RECV_TIMEOUT)) { + /* reset the receiver shutdown timer */ + batadv_tp_reset_receiver_timer(tp_vars); + return; + } + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Shutting down for inactivity (more than %dms) from %pM\n", + BATADV_TP_RECV_TIMEOUT, tp_vars->other_end); + + spin_lock_bh(&tp_vars->bat_priv->tp_list_lock); + hlist_del_rcu(&tp_vars->list); + spin_unlock_bh(&tp_vars->bat_priv->tp_list_lock); + + /* drop list reference */ + batadv_tp_vars_put(tp_vars); + + atomic_dec(&bat_priv->tp_num); + + spin_lock_bh(&tp_vars->unacked_lock); + list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { + list_del(&un->list); + kfree(un); + } + spin_unlock_bh(&tp_vars->unacked_lock); + + /* drop reference of timer */ + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_send_ack - send an ACK packet + * @bat_priv: the bat priv with all the soft interface information + * @dst: the mac address of the destination originator + * @seq: the sequence number to ACK + * @timestamp: the timestamp to echo back in the ACK + * @session: session identifier + * @socket_index: local ICMP socket identifier + * + * Return: 0 on success, a positive integer representing the reason of the + * failure otherwise + */ +static int batadv_tp_send_ack(struct batadv_priv *bat_priv, const u8 *dst, + u32 seq, __be32 timestamp, const u8 *session, + int socket_index) +{ + struct batadv_hard_iface *primary_if = NULL; + struct batadv_orig_node *orig_node; + struct batadv_icmp_tp_packet *icmp; + struct sk_buff *skb; + int r, ret; + + orig_node = batadv_orig_hash_find(bat_priv, dst); + if (unlikely(!orig_node)) { + ret = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (unlikely(!primary_if)) { + ret = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + + skb = netdev_alloc_skb_ip_align(NULL, sizeof(*icmp) + ETH_HLEN); + if (unlikely(!skb)) { + ret = BATADV_TP_REASON_MEMORY_ERROR; + goto out; + } + + skb_reserve(skb, ETH_HLEN); + icmp = (struct batadv_icmp_tp_packet *)skb_put(skb, sizeof(*icmp)); + icmp->packet_type = BATADV_ICMP; + icmp->version = BATADV_COMPAT_VERSION; + icmp->ttl = BATADV_TTL; + icmp->msg_type = BATADV_TP; + ether_addr_copy(icmp->dst, orig_node->orig); + ether_addr_copy(icmp->orig, primary_if->net_dev->dev_addr); + icmp->uid = socket_index; + + icmp->subtype = BATADV_TP_ACK; + memcpy(icmp->session, session, sizeof(icmp->session)); + icmp->seqno = htonl(seq); + icmp->timestamp = timestamp; + + /* send the ack */ + r = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (r == -1) + kfree_skb(skb); + + if (unlikely(r < 0) || (r == NET_XMIT_DROP)) { + ret = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + ret = 0; + +out: + if (likely(orig_node)) + batadv_orig_node_put(orig_node); + if (likely(primary_if)) + batadv_hardif_put(primary_if); + + return ret; +} + +/** + * batadv_tp_handle_out_of_order - store an out of order packet + * @tp_vars: the private data of the current TP meter session + * @skb: the buffer containing the received packet + * + * Store the out of order packet in the unacked list for late processing. This + * packets are kept in this list so that they can be ACKed at once as soon as + * all the previous packets have been received + * + * Return: true if the packed has been successfully processed, false otherwise + */ +static bool batadv_tp_handle_out_of_order(struct batadv_tp_vars *tp_vars, + const struct sk_buff *skb) +{ + const struct batadv_icmp_tp_packet *icmp; + struct batadv_tp_unacked *un, *new; + u32 payload_len; + bool added = false; + + new = kmalloc(sizeof(*new), GFP_ATOMIC); + if (unlikely(!new)) + return false; + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + new->seqno = ntohl(icmp->seqno); + payload_len = skb->len - sizeof(struct batadv_unicast_packet); + new->len = payload_len; + + spin_lock_bh(&tp_vars->unacked_lock); + /* if the list is empty immediately attach this new object */ + if (list_empty(&tp_vars->unacked_list)) { + list_add(&new->list, &tp_vars->unacked_list); + goto out; + } + + /* otherwise loop over the list and either drop the packet because this + * is a duplicate or store it at the right position. + * + * The iteration is done in the reverse way because it is likely that + * the last received packet (the one being processed now) has a bigger + * seqno than all the others already stored. + */ + list_for_each_entry_reverse(un, &tp_vars->unacked_list, list) { + /* check for duplicates */ + if (new->seqno == un->seqno) { + if (new->len > un->len) + un->len = new->len; + kfree(new); + added = true; + break; + } + + /* look for the right position */ + if (batadv_seq_before(new->seqno, un->seqno)) + continue; + + /* as soon as an entry having a bigger seqno is found, the new + * one is attached _after_ it. In this way the list is kept in + * ascending order + */ + list_add_tail(&new->list, &un->list); + added = true; + break; + } + + /* received packet with smallest seqno out of order; add it to front */ + if (!added) + list_add(&new->list, &tp_vars->unacked_list); + +out: + spin_unlock_bh(&tp_vars->unacked_lock); + + return true; +} + +/** + * batadv_tp_ack_unordered - update number received bytes in current stream + * without gaps + * @tp_vars: the private data of the current TP meter session + */ +static void batadv_tp_ack_unordered(struct batadv_tp_vars *tp_vars) +{ + struct batadv_tp_unacked *un, *safe; + u32 to_ack; + + /* go through the unacked packet list and possibly ACK them as + * well + */ + spin_lock_bh(&tp_vars->unacked_lock); + list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { + /* the list is ordered, therefore it is possible to stop as soon + * there is a gap between the last acked seqno and the seqno of + * the packet under inspection + */ + if (batadv_seq_before(tp_vars->last_recv, un->seqno)) + break; + + to_ack = un->seqno + un->len - tp_vars->last_recv; + + if (batadv_seq_before(tp_vars->last_recv, un->seqno + un->len)) + tp_vars->last_recv += to_ack; + + list_del(&un->list); + kfree(un); + } + spin_unlock_bh(&tp_vars->unacked_lock); +} + +/** + * batadv_tp_init_recv - return matching or create new receiver tp_vars + * @bat_priv: the bat priv with all the soft interface information + * @icmp: received icmp tp msg + * + * Return: corresponding tp_vars or NULL on errors + */ +static struct batadv_tp_vars * +batadv_tp_init_recv(struct batadv_priv *bat_priv, + const struct batadv_icmp_tp_packet *icmp) +{ + struct batadv_tp_vars *tp_vars; + + spin_lock_bh(&bat_priv->tp_list_lock); + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, + icmp->session); + if (tp_vars) + goto out_unlock; + + if (!atomic_add_unless(&bat_priv->tp_num, 1, BATADV_TP_MAX_NUM)) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: too many ongoing sessions, aborting (RECV)\n"); + goto out_unlock; + } + + tp_vars = kmalloc(sizeof(*tp_vars), GFP_ATOMIC); + if (!tp_vars) + goto out_unlock; + + ether_addr_copy(tp_vars->other_end, icmp->orig); + tp_vars->role = BATADV_TP_RECEIVER; + memcpy(tp_vars->session, icmp->session, sizeof(tp_vars->session)); + tp_vars->last_recv = BATADV_TP_FIRST_SEQ; + tp_vars->bat_priv = bat_priv; + kref_init(&tp_vars->refcount); + + spin_lock_init(&tp_vars->unacked_lock); + INIT_LIST_HEAD(&tp_vars->unacked_list); + + kref_get(&tp_vars->refcount); + hlist_add_head_rcu(&tp_vars->list, &bat_priv->tp_list); + + kref_get(&tp_vars->refcount); + setup_timer(&tp_vars->timer, batadv_tp_receiver_shutdown, + (unsigned long)tp_vars); + + batadv_tp_reset_receiver_timer(tp_vars); + +out_unlock: + spin_unlock_bh(&bat_priv->tp_list_lock); + + return tp_vars; +} + +/** + * batadv_tp_recv_msg - process a single data message + * @bat_priv: the bat priv with all the soft interface information + * @skb: the buffer containing the received packet + * + * Process a received TP MSG packet + */ +static void batadv_tp_recv_msg(struct batadv_priv *bat_priv, + const struct sk_buff *skb) +{ + const struct batadv_icmp_tp_packet *icmp; + struct batadv_tp_vars *tp_vars; + size_t packet_size; + u32 seqno; + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + seqno = ntohl(icmp->seqno); + /* check if this is the first seqno. This means that if the + * first packet is lost, the tp meter does not work anymore! + */ + if (seqno == BATADV_TP_FIRST_SEQ) { + tp_vars = batadv_tp_init_recv(bat_priv, icmp); + if (!tp_vars) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: seqno != BATADV_TP_FIRST_SEQ cannot initiate connection\n"); + goto out; + } + } else { + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, + icmp->session); + if (!tp_vars) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Unexpected packet from %pM!\n", + icmp->orig); + goto out; + } + } + + if (unlikely(tp_vars->role != BATADV_TP_RECEIVER)) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: dropping packet: not expected (role=%u)\n", + tp_vars->role); + goto out; + } + + tp_vars->last_recv_time = jiffies; + + /* if the packet is a duplicate, it may be the case that an ACK has been + * lost. Resend the ACK + */ + if (batadv_seq_before(seqno, tp_vars->last_recv)) + goto send_ack; + + /* if the packet is out of order enqueue it */ + if (ntohl(icmp->seqno) != tp_vars->last_recv) { + /* exit immediately (and do not send any ACK) if the packet has + * not been enqueued correctly + */ + if (!batadv_tp_handle_out_of_order(tp_vars, skb)) + goto out; + + /* send a duplicate ACK */ + goto send_ack; + } + + /* if everything was fine count the ACKed bytes */ + packet_size = skb->len - sizeof(struct batadv_unicast_packet); + tp_vars->last_recv += packet_size; + + /* check if this ordered message filled a gap.... */ + batadv_tp_ack_unordered(tp_vars); + +send_ack: + /* send the ACK. If the received packet was out of order, the ACK that + * is going to be sent is a duplicate (the sender will count them and + * possibly enter Fast Retransmit as soon as it has reached 3) + */ + batadv_tp_send_ack(bat_priv, icmp->orig, tp_vars->last_recv, + icmp->timestamp, icmp->session, icmp->uid); +out: + if (likely(tp_vars)) + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_meter_recv - main TP Meter receiving function + * @bat_priv: the bat priv with all the soft interface information + * @skb: the buffer containing the received packet + */ +void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb) +{ + struct batadv_icmp_tp_packet *icmp; + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + switch (icmp->subtype) { + case BATADV_TP_MSG: + batadv_tp_recv_msg(bat_priv, skb); + break; + case BATADV_TP_ACK: + batadv_tp_recv_ack(bat_priv, skb); + break; + default: + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Received unknown TP Metric packet type %u\n", + icmp->subtype); + } + consume_skb(skb); +} + +/** + * batadv_tp_meter_init - initialize global tp_meter structures + */ +void batadv_tp_meter_init(void) +{ + get_random_bytes(batadv_tp_prerandom, sizeof(batadv_tp_prerandom)); +} diff --git a/net/batman-adv/tp_meter.h b/net/batman-adv/tp_meter.h new file mode 100644 index 000000000000..ba922c425e56 --- /dev/null +++ b/net/batman-adv/tp_meter.h @@ -0,0 +1,34 @@ +/* Copyright (C) 2012-2016 B.A.T.M.A.N. contributors: + * + * Edo Monticelli, Antonio Quartulli + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_TP_METER_H_ +#define _NET_BATMAN_ADV_TP_METER_H_ + +#include "main.h" + +#include + +struct sk_buff; + +void batadv_tp_meter_init(void); +void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, + u32 test_length, u32 *cookie); +void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, + u8 return_value); +void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb); + +#endif /* _NET_BATMAN_ADV_TP_METER_H_ */ diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index ab863a5ab2b8..a331e3ab93d1 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -33,6 +33,7 @@ #include #include #include +#include #include "packet.h" @@ -832,6 +833,111 @@ struct batadv_priv_nc { struct batadv_hashtable *decoding_hash; }; +/** + * struct batadv_tp_unacked - unacked packet meta-information + * @seqno: seqno of the unacked packet + * @len: length of the packet + * @list: list node for batadv_tp_vars::unacked_list + * + * This struct is supposed to represent a buffer unacked packet. However, since + * the purpose of the TP meter is to count the traffic only, there is no need to + * store the entire sk_buff, the starting offset and the length are enough + */ +struct batadv_tp_unacked { + u32 seqno; + u16 len; + struct list_head list; +}; + +/** + * enum batadv_tp_meter_role - Modus in tp meter session + * @BATADV_TP_RECEIVER: Initialized as receiver + * @BATADV_TP_SENDER: Initialized as sender + */ +enum batadv_tp_meter_role { + BATADV_TP_RECEIVER, + BATADV_TP_SENDER +}; + +/** + * struct batadv_tp_vars - tp meter private variables per session + * @list: list node for bat_priv::tp_list + * @timer: timer for ack (receiver) and retry (sender) + * @bat_priv: pointer to the mesh object + * @start_time: start time in jiffies + * @other_end: mac address of remote + * @role: receiver/sender modi + * @sending: sending binary semaphore: 1 if sending, 0 is not + * @reason: reason for a stopped session + * @finish_work: work item for the finishing procedure + * @test_length: test length in milliseconds + * @session: TP session identifier + * @icmp_uid: local ICMP "socket" index + * @dec_cwnd: decimal part of the cwnd used during linear growth + * @cwnd: current size of the congestion window + * @cwnd_lock: lock do protect @cwnd & @dec_cwnd + * @ss_threshold: Slow Start threshold. Once cwnd exceeds this value the + * connection switches to the Congestion Avoidance state + * @last_acked: last acked byte + * @last_sent: last sent byte, not yet acked + * @tot_sent: amount of data sent/ACKed so far + * @dup_acks: duplicate ACKs counter + * @fast_recovery: true if in Fast Recovery mode + * @recover: last sent seqno when entering Fast Recovery + * @rto: sender timeout + * @srtt: smoothed RTT scaled by 2^3 + * @rttvar: RTT variation scaled by 2^2 + * @more_bytes: waiting queue anchor when waiting for more ack/retry timeout + * @prerandom_offset: offset inside the prerandom buffer + * @prerandom_lock: spinlock protecting access to prerandom_offset + * @last_recv: last in-order received packet + * @unacked_list: list of unacked packets (meta-info only) + * @unacked_lock: protect unacked_list + * @last_recv_time: time time (jiffies) a msg was received + * @refcount: number of context where the object is used + * @rcu: struct used for freeing in an RCU-safe manner + */ +struct batadv_tp_vars { + struct hlist_node list; + struct timer_list timer; + struct batadv_priv *bat_priv; + unsigned long start_time; + u8 other_end[ETH_ALEN]; + enum batadv_tp_meter_role role; + atomic_t sending; + enum batadv_tp_meter_reason reason; + struct delayed_work finish_work; + u32 test_length; + u8 session[2]; + u8 icmp_uid; + + /* sender variables */ + u16 dec_cwnd; + u32 cwnd; + spinlock_t cwnd_lock; /* Protects cwnd & dec_cwnd */ + u32 ss_threshold; + atomic_t last_acked; + u32 last_sent; + atomic64_t tot_sent; + atomic_t dup_acks; + bool fast_recovery; + u32 recover; + u32 rto; + u32 srtt; + u32 rttvar; + wait_queue_head_t more_bytes; + u32 prerandom_offset; + spinlock_t prerandom_lock; /* Protects prerandom_offset */ + + /* receiver variables */ + u32 last_recv; + struct list_head unacked_list; + spinlock_t unacked_lock; /* Protects unacked_list */ + unsigned long last_recv_time; + struct kref refcount; + struct rcu_head rcu; +}; + /** * struct batadv_softif_vlan - per VLAN attributes set * @bat_priv: pointer to the mesh object @@ -900,9 +1006,12 @@ struct batadv_priv_bat_v { * @debug_dir: dentry for debugfs batman-adv subdirectory * @forw_bat_list: list of aggregated OGMs that will be forwarded * @forw_bcast_list: list of broadcast packets that will be rebroadcasted + * @tp_list: list of tp sessions + * @tp_num: number of currently active tp sessions * @orig_hash: hash table containing mesh participants (orig nodes) * @forw_bat_list_lock: lock protecting forw_bat_list * @forw_bcast_list_lock: lock protecting forw_bcast_list + * @tp_list_lock: spinlock protecting @tp_list * @orig_work: work queue callback item for orig node purging * @cleanup_work: work queue callback item for soft-interface deinit * @primary_if: one of the hard-interfaces assigned to this mesh interface @@ -956,9 +1065,12 @@ struct batadv_priv { struct dentry *debug_dir; struct hlist_head forw_bat_list; struct hlist_head forw_bcast_list; + struct hlist_head tp_list; struct batadv_hashtable *orig_hash; spinlock_t forw_bat_list_lock; /* protects forw_bat_list */ spinlock_t forw_bcast_list_lock; /* protects forw_bcast_list */ + spinlock_t tp_list_lock; /* protects tp_list */ + atomic_t tp_num; struct delayed_work orig_work; struct work_struct cleanup_work; struct batadv_hard_iface __rcu *primary_if; /* rcu protected pointer */ -- cgit v1.2.3 From ff202ee1ed8f032f05b80b541664cf02e75d7080 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Jul 2016 06:43:15 -0400 Subject: net sched actions: skbedit add support for mod-ing skb pkt_type Extremely useful for setting packet type to host so i dont have to modify the dst mac address using pedit (which requires that i know the mac address) Example usage: tc filter add dev eth0 parent ffff: protocol ip pref 9 u32 \ match ip src 5.5.5.5/32 \ flowid 1:5 action skbedit ptype host This will tag all packets incoming from 5.5.5.5 with type PACKET_HOST Signed-off-by: Jamal Hadi Salim Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/net/tc_act/tc_skbedit.h | 10 +++++----- include/uapi/linux/tc_act/tc_skbedit.h | 2 ++ net/sched/act_skbedit.c | 18 +++++++++++++++++- 3 files changed, 24 insertions(+), 6 deletions(-) (limited to 'include/uapi') diff --git a/include/net/tc_act/tc_skbedit.h b/include/net/tc_act/tc_skbedit.h index b496d5ad7d42..d01a5d40cfb5 100644 --- a/include/net/tc_act/tc_skbedit.h +++ b/include/net/tc_act/tc_skbedit.h @@ -24,11 +24,11 @@ struct tcf_skbedit { struct tcf_common common; - u32 flags; - u32 priority; - u32 mark; - u16 queue_mapping; - /* XXX: 16-bit pad here? */ + u32 flags; + u32 priority; + u32 mark; + u16 queue_mapping; + u16 ptype; }; #define to_skbedit(a) \ container_of(a->priv, struct tcf_skbedit, common) diff --git a/include/uapi/linux/tc_act/tc_skbedit.h b/include/uapi/linux/tc_act/tc_skbedit.h index fecb5cc48c40..a4d00c608d8f 100644 --- a/include/uapi/linux/tc_act/tc_skbedit.h +++ b/include/uapi/linux/tc_act/tc_skbedit.h @@ -27,6 +27,7 @@ #define SKBEDIT_F_PRIORITY 0x1 #define SKBEDIT_F_QUEUE_MAPPING 0x2 #define SKBEDIT_F_MARK 0x4 +#define SKBEDIT_F_PTYPE 0x8 struct tc_skbedit { tc_gen; @@ -40,6 +41,7 @@ enum { TCA_SKBEDIT_QUEUE_MAPPING, TCA_SKBEDIT_MARK, TCA_SKBEDIT_PAD, + TCA_SKBEDIT_PTYPE, __TCA_SKBEDIT_MAX }; #define TCA_SKBEDIT_MAX (__TCA_SKBEDIT_MAX - 1) diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 53d1486cddf7..1c4c9240a3f8 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -47,6 +47,8 @@ static int tcf_skbedit(struct sk_buff *skb, const struct tc_action *a, skb_set_queue_mapping(skb, d->queue_mapping); if (d->flags & SKBEDIT_F_MARK) skb->mark = d->mark; + if (d->flags & SKBEDIT_F_PTYPE) + skb->pkt_type = d->ptype; spin_unlock(&d->tcf_lock); return d->tcf_action; @@ -57,6 +59,7 @@ static const struct nla_policy skbedit_policy[TCA_SKBEDIT_MAX + 1] = { [TCA_SKBEDIT_PRIORITY] = { .len = sizeof(u32) }, [TCA_SKBEDIT_QUEUE_MAPPING] = { .len = sizeof(u16) }, [TCA_SKBEDIT_MARK] = { .len = sizeof(u32) }, + [TCA_SKBEDIT_PTYPE] = { .len = sizeof(u16) }, }; static int tcf_skbedit_init(struct net *net, struct nlattr *nla, @@ -68,7 +71,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, struct tc_skbedit *parm; struct tcf_skbedit *d; u32 flags = 0, *priority = NULL, *mark = NULL; - u16 *queue_mapping = NULL; + u16 *queue_mapping = NULL, *ptype = NULL; bool exists = false; int ret = 0, err; @@ -92,6 +95,13 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, queue_mapping = nla_data(tb[TCA_SKBEDIT_QUEUE_MAPPING]); } + if (tb[TCA_SKBEDIT_PTYPE] != NULL) { + ptype = nla_data(tb[TCA_SKBEDIT_PTYPE]); + if (!skb_pkt_type_ok(*ptype)) + return -EINVAL; + flags |= SKBEDIT_F_PTYPE; + } + if (tb[TCA_SKBEDIT_MARK] != NULL) { flags |= SKBEDIT_F_MARK; mark = nla_data(tb[TCA_SKBEDIT_MARK]); @@ -132,6 +142,8 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, d->queue_mapping = *queue_mapping; if (flags & SKBEDIT_F_MARK) d->mark = *mark; + if (flags & SKBEDIT_F_PTYPE) + d->ptype = *ptype; d->tcf_action = parm->action; @@ -169,6 +181,10 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, nla_put(skb, TCA_SKBEDIT_MARK, sizeof(d->mark), &d->mark)) goto nla_put_failure; + if ((d->flags & SKBEDIT_F_PTYPE) && + nla_put(skb, TCA_SKBEDIT_PTYPE, sizeof(d->ptype), + &d->ptype)) + goto nla_put_failure; tcf_tm_dump(&t, &d->tcf_tm); if (nla_put_64bit(skb, TCA_SKBEDIT_TM, sizeof(t), &t, TCA_SKBEDIT_PAD)) -- cgit v1.2.3 From 13c5c240f789bbd2bcacb14a23771491485ae61f Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Sun, 3 Jul 2016 01:28:47 +0200 Subject: bpf: add bpf_get_hash_recalc helper If skb_clear_hash() was invoked due to mangling of relevant headers and BPF program needs skb->hash later on, we can add a helper to trigger hash recalculation via bpf_get_hash_recalc(). The helper will return the newly retrieved hash directly, but later access can also be done via skb context again through skb->hash directly (inline) without needing to call the helper once more. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 9 +++++++++ net/core/filter.c | 19 +++++++++++++++++++ 2 files changed, 28 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index f44504d875e2..c14ca1cd6297 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -348,6 +348,15 @@ enum bpf_func_id { * < 0 error */ BPF_FUNC_skb_in_cgroup, + + /** + * bpf_get_hash_recalc(skb) + * Retrieve and possibly recalculate skb->hash. + * @skb: pointer to skb + * Return: hash + */ + BPF_FUNC_get_hash_recalc, + __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index 54071cf70fb5..10c4a2f9e8bb 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1729,6 +1729,23 @@ static const struct bpf_func_proto bpf_get_route_realm_proto = { .arg1_type = ARG_PTR_TO_CTX, }; +static u64 bpf_get_hash_recalc(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + /* If skb_clear_hash() was called due to mangling, we can + * trigger SW recalculation here. Later access to hash + * can then use the inline skb->hash via context directly + * instead of calling this helper again. + */ + return skb_get_hash((struct sk_buff *) (unsigned long) r1); +} + +static const struct bpf_func_proto bpf_get_hash_recalc_proto = { + .func = bpf_get_hash_recalc, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, +}; + static u64 bpf_skb_vlan_push(u64 r1, u64 r2, u64 vlan_tci, u64 r4, u64 r5) { struct sk_buff *skb = (struct sk_buff *) (long) r1; @@ -2337,6 +2354,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_redirect_proto; case BPF_FUNC_get_route_realm: return &bpf_get_route_realm_proto; + case BPF_FUNC_get_hash_recalc: + return &bpf_get_hash_recalc_proto; case BPF_FUNC_perf_event_output: return bpf_get_event_output_proto(); case BPF_FUNC_get_smp_processor_id: -- cgit v1.2.3 From c6e6a0c8be575c830a97b1942dabeab70f423fe0 Mon Sep 17 00:00:00 2001 From: Aviya Erenfeld Date: Tue, 5 Jul 2016 15:23:08 +0300 Subject: nl80211: Add API to support VHT MU-MIMO air sniffer add API to support VHT MU-MIMO air sniffer. in MU-MIMO there are parallel frames on the air while the HW has only one RX. add the capability to sniff one of the MU-MIMO parallel frames by giving the sniffer additional information so it'll know which of the parallel frames it shall follow. Add attribute - NL80211_ATTR_MU_MIMO_GROUP_DATA - for getting a MU-MIMO groupID in order to monitor packets from that group using VHT MU-MIMO. And add attribute -NL80211_ATTR_MU_MIMO_FOLLOW_ADDR - for passing MAC address to monitor mode. that option will be used by VHT MU-MIMO air sniffer to follow a station according to it's MAC address using VHT MU-MIMO. Signed-off-by: Aviya Erenfeld Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 10 ++++++++-- include/uapi/linux/nl80211.h | 29 +++++++++++++++++++++++++++++ net/wireless/nl80211.c | 36 ++++++++++++++++++++++++++++++++++++ 3 files changed, 73 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 7bbb00d8b2cd..fa4f0f793817 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -330,6 +330,9 @@ struct ieee80211_supported_band { * in a separate chapter. */ +#define VHT_MUMIMO_GROUPS_DATA_LEN (WLAN_MEMBERSHIP_LEN +\ + WLAN_USER_POSITION_LEN) + /** * struct vif_params - describes virtual interface parameters * @use_4addr: use 4-address frames @@ -339,10 +342,13 @@ struct ieee80211_supported_band { * This feature is only fully supported by drivers that enable the * %NL80211_FEATURE_MAC_ON_CREATE flag. Others may support creating ** only p2p devices with specified MAC. + * @vht_mumimo_groups: MU-MIMO groupID. used for monitoring only + * packets belonging to that MU-MIMO groupID. */ struct vif_params { - int use_4addr; - u8 macaddr[ETH_ALEN]; + int use_4addr; + u8 macaddr[ETH_ALEN]; + u8 vht_mumimo_groups[VHT_MUMIMO_GROUPS_DATA_LEN]; }; /** diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 53c8278827a0..1d7da7888dcf 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1829,6 +1829,25 @@ enum nl80211_commands { * %NL80211_ATTR_EXT_CAPA_MASK, to specify the extended capabilities per * interface type. * + * @NL80211_ATTR_MU_MIMO_GROUP_DATA: array of 24 bytes that defines a MU-MIMO + * groupID for monitor mode. + * The first 8 bytes are a mask that defines the membership in each + * group (there are 64 groups, group 0 and 63 are reserved), + * each bit represents a group and set to 1 for being a member in + * that group and 0 for not being a member. + * The remaining 16 bytes define the position in each group: 2 bits for + * each group. + * (smaller group numbers represented on most significant bits and bigger + * group numbers on least significant bits.) + * This attribute is used only if all interfaces are in monitor mode. + * Set this attribute in order to monitor packets using the given MU-MIMO + * groupID data. + * to turn off that feature set all the bits of the groupID to zero. + * @NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR: mac address for the sniffer to follow + * when using MU-MIMO air sniffer. + * to turn that feature off set an invalid mac address + * (e.g. FF:FF:FF:FF:FF:FF) + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2213,6 +2232,9 @@ enum nl80211_attrs { NL80211_ATTR_IFTYPE_EXT_CAPA, + NL80211_ATTR_MU_MIMO_GROUP_DATA, + NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -4479,6 +4501,12 @@ enum nl80211_feature_flags { * %NL80211_CMD_ASSOCIATE and %NL80211_CMD_CONNECT requests, which will set * the ASSOC_REQ_USE_RRM flag in the association request even if * NL80211_FEATURE_QUIET is not advertized. + * @NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER: This device supports MU-MIMO air + * sniffer which means that it can be configured to hear packets from + * certain groups which can be configured by the + * %NL80211_ATTR_MU_MIMO_GROUP_DATA attribute, + * or can be configured to follow a station by configuring the + * %NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR attribute. * * @NUM_NL80211_EXT_FEATURES: number of extended features. * @MAX_NL80211_EXT_FEATURES: highest extended feature index. @@ -4486,6 +4514,7 @@ enum nl80211_feature_flags { enum nl80211_ext_feature_index { NL80211_EXT_FEATURE_VHT_IBSS, NL80211_EXT_FEATURE_RRM, + NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER, /* add new features before the definition below */ NUM_NL80211_EXT_FEATURES, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 244d552d5647..447026f8cc76 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -405,6 +405,10 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_PBSS] = { .type = NLA_FLAG }, [NL80211_ATTR_BSS_SELECT] = { .type = NLA_NESTED }, [NL80211_ATTR_STA_SUPPORT_P2P_PS] = { .type = NLA_U8 }, + [NL80211_ATTR_MU_MIMO_GROUP_DATA] = { + .len = VHT_MUMIMO_GROUPS_DATA_LEN + }, + [NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR] = { .len = ETH_ALEN }, }; /* policy for the key attributes */ @@ -2695,6 +2699,38 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) change = true; } + if (info->attrs[NL80211_ATTR_MU_MIMO_GROUP_DATA]) { + const u8 *mumimo_groups; + u32 cap_flag = NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER; + + if (!wiphy_ext_feature_isset(&rdev->wiphy, cap_flag)) + return -EOPNOTSUPP; + + mumimo_groups = + nla_data(info->attrs[NL80211_ATTR_MU_MIMO_GROUP_DATA]); + + /* bits 0 and 63 are reserved and must be zero */ + if ((mumimo_groups[0] & BIT(7)) || + (mumimo_groups[VHT_MUMIMO_GROUPS_DATA_LEN - 1] & BIT(0))) + return -EINVAL; + + memcpy(params.vht_mumimo_groups, mumimo_groups, + VHT_MUMIMO_GROUPS_DATA_LEN); + change = true; + } + + if (info->attrs[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR]) { + u32 cap_flag = NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER; + + if (!wiphy_ext_feature_isset(&rdev->wiphy, cap_flag)) + return -EOPNOTSUPP; + + nla_memcpy(params.macaddr, + info->attrs[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR], + ETH_ALEN); + change = true; + } + if (flags && (*flags & MONITOR_FLAG_ACTIVE) && !(rdev->wiphy.features & NL80211_FEATURE_ACTIVE_MONITOR)) return -EOPNOTSUPP; -- cgit v1.2.3 From 1d76250bd34af86c6498fc51e50cab3bfbbeceaa Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Tue, 5 Jul 2016 17:10:13 +0300 Subject: nl80211: support beacon report scanning Beacon report radio measurement requires reporting observed BSSs on the channels specified in the beacon request. If the measurement mode is set to passive or active, it requires actually performing a scan (passive or active, accordingly), and reporting the time that the scan was started and the time each beacon/probe was received (both in terms of TSF of the BSS of the requesting AP). If the request mode is table, this information is optional. In addition, the radio measurement request specifies the channel dwell time for the measurement. In order to use scan for beacon report when the mode is active or passive, add a parameter to scan request that specifies the channel dwell time, and add scan start time and beacon received time to scan results information. Supporting beacon report is required for Multi Band Operation (MBO). Signed-off-by: Assaf Krauss Signed-off-by: David Spinadel Signed-off-by: Avraham Stern Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- drivers/net/wireless/ath/ath6kl/cfg80211.c | 17 +++++++-- drivers/net/wireless/ath/wil6210/cfg80211.c | 6 +++- drivers/net/wireless/ath/wil6210/main.c | 12 +++++-- drivers/net/wireless/ath/wil6210/p2p.c | 6 +++- drivers/net/wireless/ath/wil6210/wmi.c | 8 +++-- .../broadcom/brcm80211/brcmfmac/cfg80211.c | 6 +++- drivers/net/wireless/intersil/orinoco/scan.c | 12 +++++-- drivers/net/wireless/marvell/libertas/cfg.c | 11 ++++-- drivers/net/wireless/marvell/mwifiex/cmdevt.c | 12 +++++-- drivers/net/wireless/marvell/mwifiex/main.c | 6 +++- drivers/net/wireless/marvell/mwifiex/scan.c | 12 +++++-- drivers/net/wireless/rndis_wlan.c | 10 ++++-- drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c | 11 ++++-- drivers/staging/wilc1000/wilc_wfi_cfgoperations.c | 12 +++++-- drivers/staging/wlan-ng/cfg80211.c | 5 ++- include/net/cfg80211.h | 40 ++++++++++++++++++--- include/uapi/linux/nl80211.h | 42 ++++++++++++++++++++++ net/mac80211/scan.c | 9 +++-- net/wireless/core.c | 4 +-- net/wireless/core.h | 12 +++++++ net/wireless/nl80211.c | 27 ++++++++++++++ net/wireless/scan.c | 18 ++++++---- net/wireless/trace.h | 33 ++++++++++++----- 23 files changed, 279 insertions(+), 52 deletions(-) (limited to 'include/uapi') diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c index 4e11ba06f089..ef5b40ef6d67 100644 --- a/drivers/net/wireless/ath/ath6kl/cfg80211.c +++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c @@ -859,7 +859,11 @@ void ath6kl_cfg80211_disconnect_event(struct ath6kl_vif *vif, u8 reason, struct ath6kl *ar = vif->ar; if (vif->scan_req) { - cfg80211_scan_done(vif->scan_req, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(vif->scan_req, &info); vif->scan_req = NULL; } @@ -1069,6 +1073,9 @@ static int ath6kl_cfg80211_scan(struct wiphy *wiphy, void ath6kl_cfg80211_scan_complete_event(struct ath6kl_vif *vif, bool aborted) { struct ath6kl *ar = vif->ar; + struct cfg80211_scan_info info = { + .aborted = aborted, + }; int i; ath6kl_dbg(ATH6KL_DBG_WLAN_CFG, "%s: status%s\n", __func__, @@ -1089,7 +1096,7 @@ void ath6kl_cfg80211_scan_complete_event(struct ath6kl_vif *vif, bool aborted) } out: - cfg80211_scan_done(vif->scan_req, aborted); + cfg80211_scan_done(vif->scan_req, &info); vif->scan_req = NULL; } @@ -3614,7 +3621,11 @@ void ath6kl_cfg80211_vif_stop(struct ath6kl_vif *vif, bool wmi_ready) } if (vif->scan_req) { - cfg80211_scan_done(vif->scan_req, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(vif->scan_req, &info); vif->scan_req = NULL; } diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c index 62bf9331bd7f..f0e1175fb76a 100644 --- a/drivers/net/wireless/ath/wil6210/cfg80211.c +++ b/drivers/net/wireless/ath/wil6210/cfg80211.c @@ -1369,7 +1369,11 @@ static void wil_cfg80211_stop_p2p_device(struct wiphy *wiphy, mutex_lock(&wil->mutex); started = wil_p2p_stop_discovery(wil); if (started && wil->scan_request) { - cfg80211_scan_done(wil->scan_request, 1); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; wil->radio_wdev = wil->wdev; } diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c index 8e31d755bbee..4bc92e54984a 100644 --- a/drivers/net/wireless/ath/wil6210/main.c +++ b/drivers/net/wireless/ath/wil6210/main.c @@ -850,10 +850,14 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw) mutex_unlock(&wil->wmi_mutex); if (wil->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + wil_dbg_misc(wil, "Abort scan_request 0x%p\n", wil->scan_request); del_timer_sync(&wil->scan_timer); - cfg80211_scan_done(wil->scan_request, true); + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; } @@ -1049,10 +1053,14 @@ int __wil_down(struct wil6210_priv *wil) (void)wil_p2p_stop_discovery(wil); if (wil->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + wil_dbg_misc(wil, "Abort scan_request 0x%p\n", wil->scan_request); del_timer_sync(&wil->scan_timer); - cfg80211_scan_done(wil->scan_request, true); + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; } diff --git a/drivers/net/wireless/ath/wil6210/p2p.c b/drivers/net/wireless/ath/wil6210/p2p.c index 213b8259638c..e0f8aa0ebfac 100644 --- a/drivers/net/wireless/ath/wil6210/p2p.c +++ b/drivers/net/wireless/ath/wil6210/p2p.c @@ -252,8 +252,12 @@ void wil_p2p_search_expired(struct work_struct *work) mutex_unlock(&wil->mutex); if (started) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + mutex_lock(&wil->p2p_wdev_mutex); - cfg80211_scan_done(wil->scan_request, 0); + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; wil->radio_wdev = wil->wdev; mutex_unlock(&wil->p2p_wdev_mutex); diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c index b80c5d850e1e..4d92541913c0 100644 --- a/drivers/net/wireless/ath/wil6210/wmi.c +++ b/drivers/net/wireless/ath/wil6210/wmi.c @@ -426,15 +426,17 @@ static void wmi_evt_scan_complete(struct wil6210_priv *wil, int id, { if (wil->scan_request) { struct wmi_scan_complete_event *data = d; - bool aborted = (data->status != WMI_SCAN_SUCCESS); + struct cfg80211_scan_info info = { + .aborted = (data->status != WMI_SCAN_SUCCESS), + }; wil_dbg_wmi(wil, "SCAN_COMPLETE(0x%08x)\n", data->status); wil_dbg_misc(wil, "Complete scan_request 0x%p aborted %d\n", - wil->scan_request, aborted); + wil->scan_request, info.aborted); del_timer_sync(&wil->scan_timer); mutex_lock(&wil->p2p_wdev_mutex); - cfg80211_scan_done(wil->scan_request, aborted); + cfg80211_scan_done(wil->scan_request, &info); wil->radio_wdev = wil->wdev; mutex_unlock(&wil->p2p_wdev_mutex); wil->scan_request = NULL; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 264bd638a3d9..afe2b202040a 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -775,9 +775,13 @@ s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg, if (!aborted) cfg80211_sched_scan_results(cfg_to_wiphy(cfg)); } else if (scan_request) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + brcmf_dbg(SCAN, "ESCAN Completed scan: %s\n", aborted ? "Aborted" : "Done"); - cfg80211_scan_done(scan_request, aborted); + cfg80211_scan_done(scan_request, &info); } if (!test_and_clear_bit(BRCMF_SCAN_STATUS_BUSY, &cfg->scan_status)) brcmf_dbg(SCAN, "Scan complete, probably P2P scan\n"); diff --git a/drivers/net/wireless/intersil/orinoco/scan.c b/drivers/net/wireless/intersil/orinoco/scan.c index d0ceb06c72d0..6d1d084854fb 100644 --- a/drivers/net/wireless/intersil/orinoco/scan.c +++ b/drivers/net/wireless/intersil/orinoco/scan.c @@ -237,7 +237,11 @@ void orinoco_add_hostscan_results(struct orinoco_private *priv, scan_abort: if (priv->scan_request) { - cfg80211_scan_done(priv->scan_request, abort); + struct cfg80211_scan_info info = { + .aborted = abort, + }; + + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } @@ -245,7 +249,11 @@ void orinoco_add_hostscan_results(struct orinoco_private *priv, void orinoco_scan_done(struct orinoco_private *priv, bool abort) { if (priv->scan_request) { - cfg80211_scan_done(priv->scan_request, abort); + struct cfg80211_scan_info info = { + .aborted = abort, + }; + + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } diff --git a/drivers/net/wireless/marvell/libertas/cfg.c b/drivers/net/wireless/marvell/libertas/cfg.c index 776b44bfd93a..ea4802446618 100644 --- a/drivers/net/wireless/marvell/libertas/cfg.c +++ b/drivers/net/wireless/marvell/libertas/cfg.c @@ -796,10 +796,15 @@ void lbs_scan_done(struct lbs_private *priv) { WARN_ON(!priv->scan_req); - if (priv->internal_scan) + if (priv->internal_scan) { kfree(priv->scan_req); - else - cfg80211_scan_done(priv->scan_req, false); + } else { + struct cfg80211_scan_info info = { + .aborted = false, + }; + + cfg80211_scan_done(priv->scan_req, &info); + } priv->scan_req = NULL; } diff --git a/drivers/net/wireless/marvell/mwifiex/cmdevt.c b/drivers/net/wireless/marvell/mwifiex/cmdevt.c index 6bc2011d8609..e7a21443647e 100644 --- a/drivers/net/wireless/marvell/mwifiex/cmdevt.c +++ b/drivers/net/wireless/marvell/mwifiex/cmdevt.c @@ -1057,8 +1057,12 @@ mwifiex_cancel_all_pending_cmd(struct mwifiex_adapter *adapter) if (!priv) continue; if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(adapter, WARN, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } @@ -1112,8 +1116,12 @@ mwifiex_cancel_pending_ioctl(struct mwifiex_adapter *adapter) if (!priv) continue; if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(adapter, WARN, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c index 0e280f879b58..db4925db39aa 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.c +++ b/drivers/net/wireless/marvell/mwifiex/main.c @@ -697,9 +697,13 @@ mwifiex_close(struct net_device *dev) struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev); if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(priv->adapter, INFO, "aborting scan on ndo_stop\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; priv->scan_aborting = true; } diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c index bc5e52cebce1..fdd749110fcb 100644 --- a/drivers/net/wireless/marvell/mwifiex/scan.c +++ b/drivers/net/wireless/marvell/mwifiex/scan.c @@ -1956,9 +1956,13 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) mwifiex_complete_scan(priv); if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + mwifiex_dbg(adapter, INFO, "info: notifying scan done\n"); - cfg80211_scan_done(priv->scan_request, 0); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } else { priv->scan_aborting = false; @@ -1977,9 +1981,13 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) if (!adapter->active_scan_triggered) { if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(adapter, INFO, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } else { priv->scan_aborting = false; diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index 569918c485b4..603c90470225 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -2134,6 +2134,7 @@ static void rndis_get_scan_results(struct work_struct *work) struct rndis_wlan_private *priv = container_of(work, struct rndis_wlan_private, scan_work.work); struct usbnet *usbdev = priv->usbdev; + struct cfg80211_scan_info info = {}; int ret; netdev_dbg(usbdev->net, "get_scan_results\n"); @@ -2143,7 +2144,8 @@ static void rndis_get_scan_results(struct work_struct *work) ret = rndis_check_bssid_list(usbdev, NULL, NULL); - cfg80211_scan_done(priv->scan_request, ret < 0); + info.aborted = ret < 0; + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } @@ -3574,7 +3576,11 @@ static int rndis_wlan_stop(struct usbnet *usbdev) flush_workqueue(priv->workqueue); if (priv->scan_request) { - cfg80211_scan_done(priv->scan_request, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } diff --git a/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c b/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c index 0da559d929bc..d0ba3778990e 100644 --- a/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c +++ b/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c @@ -1256,10 +1256,15 @@ void rtw_cfg80211_indicate_scan_done(struct rtw_wdev_priv *pwdev_priv, DBG_8723A("%s with scan req\n", __func__); if (pwdev_priv->scan_request->wiphy != - pwdev_priv->rtw_wdev->wiphy) + pwdev_priv->rtw_wdev->wiphy) { DBG_8723A("error wiphy compare\n"); - else - cfg80211_scan_done(pwdev_priv->scan_request, aborted); + } else { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + + cfg80211_scan_done(pwdev_priv->scan_request, &info); + } pwdev_priv->scan_request = NULL; } else { diff --git a/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c b/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c index 51aff4ff7d7c..a0d8e22e575b 100644 --- a/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c +++ b/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c @@ -454,7 +454,11 @@ static void CfgScanResult(enum scan_event scan_event, mutex_lock(&priv->scan_req_lock); if (priv->pstrScanReq) { - cfg80211_scan_done(priv->pstrScanReq, false); + struct cfg80211_scan_info info = { + .aborted = false, + }; + + cfg80211_scan_done(priv->pstrScanReq, &info); priv->u32RcvdChCount = 0; priv->bCfgScanning = false; priv->pstrScanReq = NULL; @@ -464,10 +468,14 @@ static void CfgScanResult(enum scan_event scan_event, mutex_lock(&priv->scan_req_lock); if (priv->pstrScanReq) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + update_scan_time(); refresh_scan(priv, 1, false); - cfg80211_scan_done(priv->pstrScanReq, false); + cfg80211_scan_done(priv->pstrScanReq, &info); priv->bCfgScanning = false; priv->pstrScanReq = NULL; } diff --git a/drivers/staging/wlan-ng/cfg80211.c b/drivers/staging/wlan-ng/cfg80211.c index a6e6fb9f42e1..f46dfe6b24e8 100644 --- a/drivers/staging/wlan-ng/cfg80211.c +++ b/drivers/staging/wlan-ng/cfg80211.c @@ -338,6 +338,8 @@ static int prism2_scan(struct wiphy *wiphy, struct p80211msg_dot11req_scan msg1; struct p80211msg_dot11req_scan_results msg2; struct cfg80211_bss *bss; + struct cfg80211_scan_info info = {}; + int result; int err = 0; int numbss = 0; @@ -440,7 +442,8 @@ static int prism2_scan(struct wiphy *wiphy, err = prism2_result2err(msg2.resultcode.data); exit: - cfg80211_scan_done(request, err ? 1 : 0); + info.aborted = !!(err); + cfg80211_scan_done(request, &info); priv->scan_request = NULL; return err; } diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index fa4f0f793817..e2658e392a1f 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1423,6 +1423,21 @@ struct cfg80211_ssid { u8 ssid_len; }; +/** + * struct cfg80211_scan_info - information about completed scan + * @scan_start_tsf: scan start time in terms of the TSF of the BSS that the + * wireless device that requested the scan is connected to. If this + * information is not available, this field is left zero. + * @tsf_bssid: the BSSID according to which %scan_start_tsf is set. + * @aborted: set to true if the scan was aborted for any reason, + * userspace will be notified of that + */ +struct cfg80211_scan_info { + u64 scan_start_tsf; + u8 tsf_bssid[ETH_ALEN] __aligned(2); + bool aborted; +}; + /** * struct cfg80211_scan_request - scan request description * @@ -1433,12 +1448,17 @@ struct cfg80211_ssid { * @scan_width: channel width for scanning * @ie: optional information element(s) to add into Probe Request or %NULL * @ie_len: length of ie in octets + * @duration: how long to listen on each channel, in TUs. If + * %duration_mandatory is not set, this is the maximum dwell time and + * the actual dwell time may be shorter. + * @duration_mandatory: if set, the scan duration must be as specified by the + * %duration field. * @flags: bit field of flags controlling operation * @rates: bitmap of rates to advertise for each band * @wiphy: the wiphy this was for * @scan_start: time (in jiffies) when the scan started * @wdev: the wireless device to scan for - * @aborted: (internal) scan request was notified as aborted + * @info: (internal) information about completed scan * @notified: (internal) scan request was notified as done or aborted * @no_cck: used to send probe requests at non CCK rate in 2GHz band * @mac_addr: MAC address used with randomisation @@ -1454,6 +1474,8 @@ struct cfg80211_scan_request { enum nl80211_bss_scan_width scan_width; const u8 *ie; size_t ie_len; + u16 duration; + bool duration_mandatory; u32 flags; u32 rates[NUM_NL80211_BANDS]; @@ -1467,7 +1489,8 @@ struct cfg80211_scan_request { /* internal */ struct wiphy *wiphy; unsigned long scan_start; - bool aborted, notified; + struct cfg80211_scan_info info; + bool notified; bool no_cck; /* keep last */ @@ -1600,12 +1623,19 @@ enum cfg80211_signal_type { * buffered on the device) and be accurate to about 10ms. * If the frame isn't buffered, just passing the return value of * ktime_get_boot_ns() is likely appropriate. + * @parent_tsf: the time at the start of reception of the first octet of the + * timestamp field of the frame. The time is the TSF of the BSS specified + * by %parent_bssid. + * @parent_bssid: the BSS according to which %parent_tsf is set. This is set to + * the BSS that requested the scan in which the beacon/probe was received. */ struct cfg80211_inform_bss { struct ieee80211_channel *chan; enum nl80211_bss_scan_width scan_width; s32 signal; u64 boottime_ns; + u64 parent_tsf; + u8 parent_bssid[ETH_ALEN] __aligned(2); }; /** @@ -4067,10 +4097,10 @@ const char *reg_initiator_name(enum nl80211_reg_initiator initiator); * cfg80211_scan_done - notify that scan finished * * @request: the corresponding scan request - * @aborted: set to true if the scan was aborted for any reason, - * userspace will be notified of that + * @info: information about the completed scan */ -void cfg80211_scan_done(struct cfg80211_scan_request *request, bool aborted); +void cfg80211_scan_done(struct cfg80211_scan_request *request, + struct cfg80211_scan_info *info); /** * cfg80211_sched_scan_results - notify that new scan results are available diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 1d7da7888dcf..b39ccab45333 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1848,6 +1848,22 @@ enum nl80211_commands { * to turn that feature off set an invalid mac address * (e.g. FF:FF:FF:FF:FF:FF) * + * @NL80211_ATTR_SCAN_START_TIME_TSF: The time at which the scan was actually + * started (u64). The time is the TSF of the BSS the interface that + * requested the scan is connected to (if available, otherwise this + * attribute must not be included). + * @NL80211_ATTR_SCAN_START_TIME_TSF_BSSID: The BSS according to which + * %NL80211_ATTR_SCAN_START_TIME_TSF is set. + * @NL80211_ATTR_MEASUREMENT_DURATION: measurement duration in TUs (u16). If + * %NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY is not set, this is the + * maximum measurement duration allowed. This attribute is used with + * measurement requests. It can also be used with %NL80211_CMD_TRIGGER_SCAN + * if the scan is used for beacon report radio measurement. + * @NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY: flag attribute that indicates + * that the duration specified with %NL80211_ATTR_MEASUREMENT_DURATION is + * mandatory. If this flag is not set, the duration is the maximum duration + * and the actual measurement duration may be shorter. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2235,6 +2251,11 @@ enum nl80211_attrs { NL80211_ATTR_MU_MIMO_GROUP_DATA, NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR, + NL80211_ATTR_SCAN_START_TIME_TSF, + NL80211_ATTR_SCAN_START_TIME_TSF_BSSID, + NL80211_ATTR_MEASUREMENT_DURATION, + NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -3496,6 +3517,12 @@ enum nl80211_bss_scan_width { * was last updated by a received frame. The value is expected to be * accurate to about 10ms. (u64, nanoseconds) * @NL80211_BSS_PAD: attribute used for padding for 64-bit alignment + * @NL80211_BSS_PARENT_TSF: the time at the start of reception of the first + * octet of the timestamp field of the last beacon/probe received for + * this BSS. The time is the TSF of the BSS specified by + * @NL80211_BSS_PARENT_BSSID. (u64). + * @NL80211_BSS_PARENT_BSSID: the BSS according to which @NL80211_BSS_PARENT_TSF + * is set. * @__NL80211_BSS_AFTER_LAST: internal * @NL80211_BSS_MAX: highest BSS attribute */ @@ -3517,6 +3544,8 @@ enum nl80211_bss { NL80211_BSS_PRESP_DATA, NL80211_BSS_LAST_SEEN_BOOTTIME, NL80211_BSS_PAD, + NL80211_BSS_PARENT_TSF, + NL80211_BSS_PARENT_BSSID, /* keep last */ __NL80211_BSS_AFTER_LAST, @@ -4507,6 +4536,16 @@ enum nl80211_feature_flags { * %NL80211_ATTR_MU_MIMO_GROUP_DATA attribute, * or can be configured to follow a station by configuring the * %NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR attribute. + * @NL80211_EXT_FEATURE_SCAN_START_TIME: This driver includes the actual + * time the scan started in scan results event. The time is the TSF of + * the BSS that the interface that requested the scan is connected to + * (if available). + * @NL80211_EXT_FEATURE_BSS_PARENT_TSF: Per BSS, this driver reports the + * time the last beacon/probe was received. The time is the TSF of the + * BSS that the interface that requested the scan is connected to + * (if available). + * @NL80211_EXT_FEATURE_SET_SCAN_DWELL: This driver supports configuration of + * channel dwell time. * * @NUM_NL80211_EXT_FEATURES: number of extended features. * @MAX_NL80211_EXT_FEATURES: highest extended feature index. @@ -4515,6 +4554,9 @@ enum nl80211_ext_feature_index { NL80211_EXT_FEATURE_VHT_IBSS, NL80211_EXT_FEATURE_RRM, NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER, + NL80211_EXT_FEATURE_SCAN_START_TIME, + NL80211_EXT_FEATURE_BSS_PARENT_TSF, + NL80211_EXT_FEATURE_SET_SCAN_DWELL, /* add new features before the definition below */ NUM_NL80211_EXT_FEATURES, diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index f9648ef9e31f..4ec1c52a1549 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -353,8 +353,13 @@ static void __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) scan_req = rcu_dereference_protected(local->scan_req, lockdep_is_held(&local->mtx)); - if (scan_req != local->int_scan_req) - cfg80211_scan_done(scan_req, aborted); + if (scan_req != local->int_scan_req) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + + cfg80211_scan_done(scan_req, &info); + } RCU_INIT_POINTER(local->scan_req, NULL); scan_sdata = rcu_dereference_protected(local->scan_sdata, diff --git a/net/wireless/core.c b/net/wireless/core.c index 39d9abd309ea..7645e97362c0 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -220,7 +220,7 @@ void cfg80211_stop_p2p_device(struct cfg80211_registered_device *rdev, if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified)) - rdev->scan_req->aborted = true; + rdev->scan_req->info.aborted = true; ___cfg80211_scan_done(rdev, false); } } @@ -1087,7 +1087,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, cfg80211_update_iface_num(rdev, wdev->iftype, -1); if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified)) - rdev->scan_req->aborted = true; + rdev->scan_req->info.aborted = true; ___cfg80211_scan_done(rdev, false); } diff --git a/net/wireless/core.h b/net/wireless/core.h index a4d547f99f8d..eee91443924d 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -141,6 +141,18 @@ struct cfg80211_internal_bss { unsigned long refcount; atomic_t hold; + /* time at the start of the reception of the first octet of the + * timestamp field of the last beacon/probe received for this BSS. + * The time is the TSF of the BSS specified by %parent_bssid. + */ + u64 parent_tsf; + + /* the BSS according to which %parent_tsf is set. This is set to + * the BSS that the interface that requested the scan was connected to + * when the beacon/probe was received. + */ + u8 parent_bssid[ETH_ALEN] __aligned(2); + /* must be last because of priv member */ struct cfg80211_bss pub; }; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 447026f8cc76..c53b5462ed00 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -6223,6 +6223,19 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) } } + if (info->attrs[NL80211_ATTR_MEASUREMENT_DURATION]) { + if (!wiphy_ext_feature_isset(wiphy, + NL80211_EXT_FEATURE_SET_SCAN_DWELL)) { + err = -EOPNOTSUPP; + goto out_free; + } + + request->duration = + nla_get_u16(info->attrs[NL80211_ATTR_MEASUREMENT_DURATION]); + request->duration_mandatory = + nla_get_flag(info->attrs[NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY]); + } + if (info->attrs[NL80211_ATTR_SCAN_FLAGS]) { request->flags = nla_get_u32( info->attrs[NL80211_ATTR_SCAN_FLAGS]); @@ -7056,6 +7069,13 @@ static int nl80211_send_bss(struct sk_buff *msg, struct netlink_callback *cb, jiffies_to_msecs(jiffies - intbss->ts))) goto nla_put_failure; + if (intbss->parent_tsf && + (nla_put_u64_64bit(msg, NL80211_BSS_PARENT_TSF, + intbss->parent_tsf, NL80211_BSS_PAD) || + nla_put(msg, NL80211_BSS_PARENT_BSSID, ETH_ALEN, + intbss->parent_bssid))) + goto nla_put_failure; + if (intbss->ts_boottime && nla_put_u64_64bit(msg, NL80211_BSS_LAST_SEEN_BOOTTIME, intbss->ts_boottime, NL80211_BSS_PAD)) @@ -11829,6 +11849,13 @@ static int nl80211_add_scan_req(struct sk_buff *msg, nla_put_u32(msg, NL80211_ATTR_SCAN_FLAGS, req->flags)) goto nla_put_failure; + if (req->info.scan_start_tsf && + (nla_put_u64_64bit(msg, NL80211_ATTR_SCAN_START_TIME_TSF, + req->info.scan_start_tsf, NL80211_BSS_PAD) || + nla_put(msg, NL80211_ATTR_SCAN_START_TIME_TSF_BSSID, ETH_ALEN, + req->info.tsf_bssid))) + goto nla_put_failure; + return 0; nla_put_failure: return -ENOBUFS; diff --git a/net/wireless/scan.c b/net/wireless/scan.c index ef2955c89a00..0358e12be54b 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -3,6 +3,7 @@ * * Copyright 2008 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH + * Copyright 2016 Intel Deutschland GmbH */ #include #include @@ -194,7 +195,7 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, if (wdev->netdev) cfg80211_sme_scan_done(wdev->netdev); - if (!request->aborted && + if (!request->info.aborted && request->flags & NL80211_SCAN_FLAG_FLUSH) { /* flush entries from previous scans */ spin_lock_bh(&rdev->bss_lock); @@ -202,10 +203,10 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, spin_unlock_bh(&rdev->bss_lock); } - msg = nl80211_build_scan_msg(rdev, wdev, request->aborted); + msg = nl80211_build_scan_msg(rdev, wdev, request->info.aborted); #ifdef CONFIG_CFG80211_WEXT - if (wdev->netdev && !request->aborted) { + if (wdev->netdev && !request->info.aborted) { memset(&wrqu, 0, sizeof(wrqu)); wireless_send_event(wdev->netdev, SIOCGIWSCAN, &wrqu, NULL); @@ -236,12 +237,13 @@ void __cfg80211_scan_done(struct work_struct *wk) rtnl_unlock(); } -void cfg80211_scan_done(struct cfg80211_scan_request *request, bool aborted) +void cfg80211_scan_done(struct cfg80211_scan_request *request, + struct cfg80211_scan_info *info) { - trace_cfg80211_scan_done(request, aborted); + trace_cfg80211_scan_done(request, info); WARN_ON(request != wiphy_to_rdev(request->wiphy)->scan_req); - request->aborted = aborted; + request->info = *info; request->notified = true; queue_work(cfg80211_wq, &wiphy_to_rdev(request->wiphy)->scan_done_wk); } @@ -843,6 +845,8 @@ cfg80211_bss_update(struct cfg80211_registered_device *rdev, found->pub.capability = tmp->pub.capability; found->ts = tmp->ts; found->ts_boottime = tmp->ts_boottime; + found->parent_tsf = tmp->parent_tsf; + ether_addr_copy(found->parent_bssid, tmp->parent_bssid); } else { struct cfg80211_internal_bss *new; struct cfg80211_internal_bss *hidden; @@ -1086,6 +1090,8 @@ cfg80211_inform_bss_frame_data(struct wiphy *wiphy, tmp.pub.beacon_interval = le16_to_cpu(mgmt->u.probe_resp.beacon_int); tmp.pub.capability = le16_to_cpu(mgmt->u.probe_resp.capab_info); tmp.ts_boottime = data->boottime_ns; + tmp.parent_tsf = data->parent_tsf; + ether_addr_copy(tmp.parent_bssid, data->parent_bssid); signal_valid = abs(data->chan->center_freq - channel->center_freq) <= wiphy->max_adj_channel_rssi_comp; diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 3c1091ae6c36..72b5255cefe2 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -2642,8 +2642,9 @@ TRACE_EVENT(cfg80211_tdls_oper_request, ); TRACE_EVENT(cfg80211_scan_done, - TP_PROTO(struct cfg80211_scan_request *request, bool aborted), - TP_ARGS(request, aborted), + TP_PROTO(struct cfg80211_scan_request *request, + struct cfg80211_scan_info *info), + TP_ARGS(request, info), TP_STRUCT__entry( __field(u32, n_channels) __dynamic_array(u8, ie, request ? request->ie_len : 0) @@ -2652,6 +2653,8 @@ TRACE_EVENT(cfg80211_scan_done, MAC_ENTRY(wiphy_mac) __field(bool, no_cck) __field(bool, aborted) + __field(u64, scan_start_tsf) + MAC_ENTRY(tsf_bssid) ), TP_fast_assign( if (request) { @@ -2666,9 +2669,16 @@ TRACE_EVENT(cfg80211_scan_done, request->wiphy->perm_addr); __entry->no_cck = request->no_cck; } - __entry->aborted = aborted; + if (info) { + __entry->aborted = info->aborted; + __entry->scan_start_tsf = info->scan_start_tsf; + MAC_ASSIGN(tsf_bssid, info->tsf_bssid); + } ), - TP_printk("aborted: %s", BOOL_TO_STR(__entry->aborted)) + TP_printk("aborted: %s, scan start (TSF): %llu, tsf_bssid: " MAC_PR_FMT, + BOOL_TO_STR(__entry->aborted), + (unsigned long long)__entry->scan_start_tsf, + MAC_PR_ARG(tsf_bssid)) ); DEFINE_EVENT(wiphy_only_evt, cfg80211_sched_scan_results, @@ -2721,6 +2731,8 @@ TRACE_EVENT(cfg80211_inform_bss_frame, __dynamic_array(u8, mgmt, len) __field(s32, signal) __field(u64, ts_boottime) + __field(u64, parent_tsf) + MAC_ENTRY(parent_bssid) ), TP_fast_assign( WIPHY_ASSIGN; @@ -2730,10 +2742,15 @@ TRACE_EVENT(cfg80211_inform_bss_frame, memcpy(__get_dynamic_array(mgmt), mgmt, len); __entry->signal = data->signal; __entry->ts_boottime = data->boottime_ns; - ), - TP_printk(WIPHY_PR_FMT ", " CHAN_PR_FMT "(scan_width: %d) signal: %d, tsb:%llu", - WIPHY_PR_ARG, CHAN_PR_ARG, __entry->scan_width, - __entry->signal, (unsigned long long)__entry->ts_boottime) + __entry->parent_tsf = data->parent_tsf; + MAC_ASSIGN(parent_bssid, data->parent_bssid); + ), + TP_printk(WIPHY_PR_FMT ", " CHAN_PR_FMT + "(scan_width: %d) signal: %d, tsb:%llu, detect_tsf:%llu, tsf_bssid: " + MAC_PR_FMT, WIPHY_PR_ARG, CHAN_PR_ARG, __entry->scan_width, + __entry->signal, (unsigned long long)__entry->ts_boottime, + (unsigned long long)__entry->parent_tsf, + MAC_PR_ARG(parent_bssid)) ); DECLARE_EVENT_CLASS(cfg80211_bss_evt, -- cgit v1.2.3 From 7d27a0ba7adc8ef30c2aae7592fce4c162aee4df Mon Sep 17 00:00:00 2001 From: Masashi Honma Date: Fri, 1 Jul 2016 10:19:34 +0900 Subject: cfg80211: Add mesh peer AID setting API Previously, mesh power management functionality works only with kernel MPM. Because user space MPM did not report mesh peer AID to kernel, the kernel could not identify the bit in TIM element. So this patch adds mesh peer AID setting API. Signed-off-by: Masashi Honma Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 2 ++ include/uapi/linux/nl80211.h | 5 +++++ net/mac80211/cfg.c | 1 + net/wireless/nl80211.c | 6 ++++++ 4 files changed, 14 insertions(+) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index e2658e392a1f..9c23f4d33e06 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -780,6 +780,7 @@ enum station_parameters_apply_mask { * (bitmask of BIT(NL80211_STA_FLAG_...)) * @listen_interval: listen interval or -1 for no change * @aid: AID or zero for no change + * @peer_aid: mesh peer AID or zero for no change * @plink_action: plink action to take * @plink_state: set the peer link state for a station * @ht_capa: HT capabilities of station @@ -811,6 +812,7 @@ struct station_parameters { u32 sta_modify_mask; int listen_interval; u16 aid; + u16 peer_aid; u8 supported_rates_len; u8 plink_action; u8 plink_state; diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index b39ccab45333..220694151434 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1864,6 +1864,9 @@ enum nl80211_commands { * mandatory. If this flag is not set, the duration is the maximum duration * and the actual measurement duration may be shorter. * + * @NL80211_ATTR_MESH_PEER_AID: Association ID for the mesh peer (u16). This is + * used to pull the stored data for mesh peer in power save state. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2256,6 +2259,8 @@ enum nl80211_attrs { NL80211_ATTR_MEASUREMENT_DURATION, NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY, + NL80211_ATTR_MESH_PEER_AID, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 0c12e4001f19..47e99ab8d97a 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -997,6 +997,7 @@ static void sta_apply_mesh_params(struct ieee80211_local *local, if (sta->mesh->plink_state != NL80211_PLINK_ESTAB) changed = mesh_plink_inc_estab_count(sdata); sta->mesh->plink_state = params->plink_state; + sta->mesh->aid = params->peer_aid; ieee80211_mps_sta_status_update(sta); changed |= ieee80211_mps_set_sta_local_pm(sta, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index c53b5462ed00..5782f718d567 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -4446,6 +4446,12 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_STATE]); if (params.plink_state >= NUM_NL80211_PLINK_STATES) return -EINVAL; + if (info->attrs[NL80211_ATTR_MESH_PEER_AID]) { + params.peer_aid = nla_get_u16( + info->attrs[NL80211_ATTR_MESH_PEER_AID]); + if (params.peer_aid > IEEE80211_MAX_AID) + return -EINVAL; + } params.sta_modify_mask |= STATION_PARAM_APPLY_PLINK_STATE; } -- cgit v1.2.3 From 000cab9a61ea9e8dc42144e39a6eb8333a402b86 Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Sun, 12 Jun 2016 15:44:44 +0800 Subject: drm/amdgpu: factor out the AMDGPU_INFO_FW_VERSION case branch into amdgpu_firmware_info MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The new amdgpu_firmware_info function will be used on amdgpu firmware version debugfs. Suggested-by: Christian König Signed-off-by: Huang Rui Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 116 ++++++++++++++++++-------------- include/uapi/drm/amdgpu_drm.h | 32 ++++----- 2 files changed, 81 insertions(+), 67 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index d851ea15059f..56c857f6e7ca 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -142,6 +142,65 @@ out: return r; } +static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info, + struct drm_amdgpu_query_fw *query_fw, + struct amdgpu_device *adev) +{ + switch (query_fw->fw_type) { + case AMDGPU_INFO_FW_VCE: + fw_info->ver = adev->vce.fw_version; + fw_info->feature = adev->vce.fb_version; + break; + case AMDGPU_INFO_FW_UVD: + fw_info->ver = adev->uvd.fw_version; + fw_info->feature = 0; + break; + case AMDGPU_INFO_FW_GMC: + fw_info->ver = adev->mc.fw_version; + fw_info->feature = 0; + break; + case AMDGPU_INFO_FW_GFX_ME: + fw_info->ver = adev->gfx.me_fw_version; + fw_info->feature = adev->gfx.me_feature_version; + break; + case AMDGPU_INFO_FW_GFX_PFP: + fw_info->ver = adev->gfx.pfp_fw_version; + fw_info->feature = adev->gfx.pfp_feature_version; + break; + case AMDGPU_INFO_FW_GFX_CE: + fw_info->ver = adev->gfx.ce_fw_version; + fw_info->feature = adev->gfx.ce_feature_version; + break; + case AMDGPU_INFO_FW_GFX_RLC: + fw_info->ver = adev->gfx.rlc_fw_version; + fw_info->feature = adev->gfx.rlc_feature_version; + break; + case AMDGPU_INFO_FW_GFX_MEC: + if (query_fw->index == 0) { + fw_info->ver = adev->gfx.mec_fw_version; + fw_info->feature = adev->gfx.mec_feature_version; + } else if (query_fw->index == 1) { + fw_info->ver = adev->gfx.mec2_fw_version; + fw_info->feature = adev->gfx.mec2_feature_version; + } else + return -EINVAL; + break; + case AMDGPU_INFO_FW_SMC: + fw_info->ver = adev->pm.fw_version; + fw_info->feature = 0; + break; + case AMDGPU_INFO_FW_SDMA: + if (query_fw->index >= adev->sdma.num_instances) + return -EINVAL; + fw_info->ver = adev->sdma.instance[query_fw->index].fw_version; + fw_info->feature = adev->sdma.instance[query_fw->index].feature_version; + break; + default: + return -EINVAL; + } + return 0; +} + /* * Userspace get information ioctl */ @@ -292,63 +351,16 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0; case AMDGPU_INFO_FW_VERSION: { struct drm_amdgpu_info_firmware fw_info; + int ret; /* We only support one instance of each IP block right now. */ if (info->query_fw.ip_instance != 0) return -EINVAL; - switch (info->query_fw.fw_type) { - case AMDGPU_INFO_FW_VCE: - fw_info.ver = adev->vce.fw_version; - fw_info.feature = adev->vce.fb_version; - break; - case AMDGPU_INFO_FW_UVD: - fw_info.ver = adev->uvd.fw_version; - fw_info.feature = 0; - break; - case AMDGPU_INFO_FW_GMC: - fw_info.ver = adev->mc.fw_version; - fw_info.feature = 0; - break; - case AMDGPU_INFO_FW_GFX_ME: - fw_info.ver = adev->gfx.me_fw_version; - fw_info.feature = adev->gfx.me_feature_version; - break; - case AMDGPU_INFO_FW_GFX_PFP: - fw_info.ver = adev->gfx.pfp_fw_version; - fw_info.feature = adev->gfx.pfp_feature_version; - break; - case AMDGPU_INFO_FW_GFX_CE: - fw_info.ver = adev->gfx.ce_fw_version; - fw_info.feature = adev->gfx.ce_feature_version; - break; - case AMDGPU_INFO_FW_GFX_RLC: - fw_info.ver = adev->gfx.rlc_fw_version; - fw_info.feature = adev->gfx.rlc_feature_version; - break; - case AMDGPU_INFO_FW_GFX_MEC: - if (info->query_fw.index == 0) { - fw_info.ver = adev->gfx.mec_fw_version; - fw_info.feature = adev->gfx.mec_feature_version; - } else if (info->query_fw.index == 1) { - fw_info.ver = adev->gfx.mec2_fw_version; - fw_info.feature = adev->gfx.mec2_feature_version; - } else - return -EINVAL; - break; - case AMDGPU_INFO_FW_SMC: - fw_info.ver = adev->pm.fw_version; - fw_info.feature = 0; - break; - case AMDGPU_INFO_FW_SDMA: - if (info->query_fw.index >= adev->sdma.num_instances) - return -EINVAL; - fw_info.ver = adev->sdma.instance[info->query_fw.index].fw_version; - fw_info.feature = adev->sdma.instance[info->query_fw.index].feature_version; - break; - default: - return -EINVAL; - } + ret = amdgpu_firmware_info(&fw_info, &info->query_fw, adev); + if (ret) + return ret; + return copy_to_user(out, &fw_info, min((size_t)size, sizeof(fw_info))) ? -EFAULT : 0; } diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index cdecf87576e8..462246aa200e 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -487,6 +487,22 @@ struct drm_amdgpu_cs_chunk_data { #define AMDGPU_INFO_MMR_SH_INDEX_SHIFT 8 #define AMDGPU_INFO_MMR_SH_INDEX_MASK 0xff +struct drm_amdgpu_query_fw { + /** AMDGPU_INFO_FW_* */ + __u32 fw_type; + /** + * Index of the IP if there are more IPs of + * the same type. + */ + __u32 ip_instance; + /** + * Index of the engine. Whether this is used depends + * on the firmware type. (e.g. MEC, SDMA) + */ + __u32 index; + __u32 _pad; +}; + /* Input structure for the INFO ioctl */ struct drm_amdgpu_info { /* Where the return value will be stored */ @@ -522,21 +538,7 @@ struct drm_amdgpu_info { __u32 flags; } read_mmr_reg; - struct { - /** AMDGPU_INFO_FW_* */ - __u32 fw_type; - /** - * Index of the IP if there are more IPs of - * the same type. - */ - __u32 ip_instance; - /** - * Index of the engine. Whether this is used depends - * on the firmware type. (e.g. MEC, SDMA) - */ - __u32 index; - __u32 _pad; - } query_fw; + struct drm_amdgpu_query_fw query_fw; }; }; -- cgit v1.2.3 From 606274c5abd8e245add01bc7145a8cbb92b69ba8 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Wed, 6 Jul 2016 22:38:36 -0700 Subject: bpf: introduce bpf_get_current_task() helper over time there were multiple requests to access different data structures and fields of task_struct current, so finally add the helper to access 'current' as-is. Tracing bpf programs will do the rest of walking the pointers via bpf_probe_read(). Note that current can be null and bpf program has to deal it with, but even dumb passing null into bpf_probe_read() is still safe. Suggested-by: Brendan Gregg Signed-off-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 7 +++++++ kernel/trace/bpf_trace.c | 13 +++++++++++++ 2 files changed, 20 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index c14ca1cd6297..262a7e883b19 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -357,6 +357,13 @@ enum bpf_func_id { */ BPF_FUNC_get_hash_recalc, + /** + * u64 bpf_get_current_task(void) + * Returns current task_struct + * Return: current + */ + BPF_FUNC_get_current_task, + __BPF_FUNC_MAX_ID, }; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 19c5b4a5c3eb..094c716154ed 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -312,6 +312,17 @@ const struct bpf_func_proto *bpf_get_event_output_proto(void) return &bpf_event_output_proto; } +static u64 bpf_get_current_task(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + return (long) current; +} + +static const struct bpf_func_proto bpf_get_current_task_proto = { + .func = bpf_get_current_task, + .gpl_only = true, + .ret_type = RET_INTEGER, +}; + static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) { switch (func_id) { @@ -329,6 +340,8 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) return &bpf_tail_call_proto; case BPF_FUNC_get_current_pid_tgid: return &bpf_get_current_pid_tgid_proto; + case BPF_FUNC_get_current_task: + return &bpf_get_current_task_proto; case BPF_FUNC_get_current_uid_gid: return &bpf_get_current_uid_gid_proto; case BPF_FUNC_get_current_comm: -- cgit v1.2.3 From a65056ecf4b48be0d0284a7b6a57b6dace10b843 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 6 Jul 2016 12:12:21 -0700 Subject: net: bridge: extend MLD/IGMP query stats MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As was suggested this patch adds support for the different versions of MLD and IGMP query types. Since the user visible structure is still in net-next we can augment it instead of adding netlink attributes. The distinction between the different IGMP/MLD query types is done as suggested in Section 7.1, RFC 3376 [1] and Section 8.1, RFC 3810 [2] based on query payload size and code for IGMP. Since all IGMP packets go through multicast_rcv() and it uses ip_mc_check_igmp/ipv6_mc_check_mld we can be sure that at least the ip/ipv6 header can be directly used. [1] https://tools.ietf.org/html/rfc3376#section-7 [2] https://tools.ietf.org/html/rfc3810#section-8.1 Suggested-by: Linus Lüssing Signed-off-by: Nikolay Aleksandrov Acked-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/uapi/linux/if_bridge.h | 7 ++++-- net/bridge/br_forward.c | 7 ++---- net/bridge/br_input.c | 2 +- net/bridge/br_multicast.c | 48 ++++++++++++++++++++++++++++++++---------- net/bridge/br_private.h | 5 +++-- 5 files changed, 48 insertions(+), 21 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h index 8304fe6f0561..c186f64fffca 100644 --- a/include/uapi/linux/if_bridge.h +++ b/include/uapi/linux/if_bridge.h @@ -261,14 +261,17 @@ enum { /* IGMP/MLD statistics */ struct br_mcast_stats { - __u64 igmp_queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_v1queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_v2queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_v3queries[BR_MCAST_DIR_SIZE]; __u64 igmp_leaves[BR_MCAST_DIR_SIZE]; __u64 igmp_v1reports[BR_MCAST_DIR_SIZE]; __u64 igmp_v2reports[BR_MCAST_DIR_SIZE]; __u64 igmp_v3reports[BR_MCAST_DIR_SIZE]; __u64 igmp_parse_errors; - __u64 mld_queries[BR_MCAST_DIR_SIZE]; + __u64 mld_v1queries[BR_MCAST_DIR_SIZE]; + __u64 mld_v2queries[BR_MCAST_DIR_SIZE]; __u64 mld_leaves[BR_MCAST_DIR_SIZE]; __u64 mld_v1reports[BR_MCAST_DIR_SIZE]; __u64 mld_v2reports[BR_MCAST_DIR_SIZE]; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 6c196037d818..d610644368b9 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -199,7 +199,6 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, bool unicast) { u8 igmp_type = br_multicast_igmp_type(skb); - __be16 proto = skb->protocol; struct net_bridge_port *prev; struct net_bridge_port *p; @@ -221,7 +220,7 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, if (IS_ERR(prev)) goto out; if (prev == p) - br_multicast_count(p->br, p, proto, igmp_type, + br_multicast_count(p->br, p, skb, igmp_type, BR_MCAST_DIR_TX); } @@ -266,8 +265,6 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, struct net_bridge *br = netdev_priv(dev); struct net_bridge_port *prev = NULL; struct net_bridge_port_group *p; - __be16 proto = skb->protocol; - struct hlist_node *rp; rp = rcu_dereference(hlist_first_rcu(&br->router_list)); @@ -286,7 +283,7 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, if (IS_ERR(prev)) goto out; if (prev == port) - br_multicast_count(port->br, port, proto, igmp_type, + br_multicast_count(port->br, port, skb, igmp_type, BR_MCAST_DIR_TX); if ((unsigned long)lport >= (unsigned long)port) diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 786602bc0567..a7817e6f306f 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -61,7 +61,7 @@ static int br_pass_frame_up(struct sk_buff *skb) if (!skb) return NET_RX_DROP; /* update the multicast stats if the packet is IGMP/MLD */ - br_multicast_count(br, NULL, skb->protocol, br_multicast_igmp_type(skb), + br_multicast_count(br, NULL, skb, br_multicast_igmp_type(skb), BR_MCAST_DIR_TX); return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index e405eef0ae2e..a5423a1eec05 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -843,14 +843,14 @@ static void __br_multicast_send_query(struct net_bridge *br, if (port) { skb->dev = port->dev; - br_multicast_count(br, port, skb->protocol, igmp_type, + br_multicast_count(br, port, skb, igmp_type, BR_MCAST_DIR_TX); NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, dev_net(port->dev), NULL, skb, NULL, skb->dev, br_dev_queue_push_xmit); } else { br_multicast_select_own_querier(br, ip, skb); - br_multicast_count(br, port, skb->protocol, igmp_type, + br_multicast_count(br, port, skb, igmp_type, BR_MCAST_DIR_RX); netif_rx(skb); } @@ -1676,7 +1676,7 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); - br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + br_multicast_count(br, port, skb, BR_INPUT_SKB_CB(skb)->igmp, BR_MCAST_DIR_RX); return err; @@ -1725,7 +1725,7 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); - br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + br_multicast_count(br, port, skb, BR_INPUT_SKB_CB(skb)->igmp, BR_MCAST_DIR_RX); return err; @@ -2251,13 +2251,16 @@ unlock: EXPORT_SYMBOL_GPL(br_multicast_has_querier_adjacent); static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, - __be16 proto, u8 type, u8 dir) + const struct sk_buff *skb, u8 type, u8 dir) { struct bridge_mcast_stats *pstats = this_cpu_ptr(stats); + __be16 proto = skb->protocol; + unsigned int t_len; u64_stats_update_begin(&pstats->syncp); switch (proto) { case htons(ETH_P_IP): + t_len = ntohs(ip_hdr(skb)->tot_len) - ip_hdrlen(skb); switch (type) { case IGMP_HOST_MEMBERSHIP_REPORT: pstats->mstats.igmp_v1reports[dir]++; @@ -2269,7 +2272,21 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, pstats->mstats.igmp_v3reports[dir]++; break; case IGMP_HOST_MEMBERSHIP_QUERY: - pstats->mstats.igmp_queries[dir]++; + if (t_len != sizeof(struct igmphdr)) { + pstats->mstats.igmp_v3queries[dir]++; + } else { + unsigned int offset = skb_transport_offset(skb); + struct igmphdr *ih, _ihdr; + + ih = skb_header_pointer(skb, offset, + sizeof(_ihdr), &_ihdr); + if (!ih) + break; + if (!ih->code) + pstats->mstats.igmp_v1queries[dir]++; + else + pstats->mstats.igmp_v2queries[dir]++; + } break; case IGMP_HOST_LEAVE_MESSAGE: pstats->mstats.igmp_leaves[dir]++; @@ -2278,6 +2295,9 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, break; #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): + t_len = ntohs(ipv6_hdr(skb)->payload_len) + + sizeof(struct ipv6hdr); + t_len -= skb_network_header_len(skb); switch (type) { case ICMPV6_MGM_REPORT: pstats->mstats.mld_v1reports[dir]++; @@ -2286,7 +2306,10 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, pstats->mstats.mld_v2reports[dir]++; break; case ICMPV6_MGM_QUERY: - pstats->mstats.mld_queries[dir]++; + if (t_len != sizeof(struct mld_msg)) + pstats->mstats.mld_v2queries[dir]++; + else + pstats->mstats.mld_v1queries[dir]++; break; case ICMPV6_MGM_REDUCTION: pstats->mstats.mld_leaves[dir]++; @@ -2299,7 +2322,7 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, } void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, - __be16 proto, u8 type, u8 dir) + const struct sk_buff *skb, u8 type, u8 dir) { struct bridge_mcast_stats __percpu *stats; @@ -2314,7 +2337,7 @@ void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, if (WARN_ON(!stats)) return; - br_mcast_stats_add(stats, proto, type, dir); + br_mcast_stats_add(stats, skb, type, dir); } int br_multicast_init_stats(struct net_bridge *br) @@ -2359,14 +2382,17 @@ void br_multicast_get_stats(const struct net_bridge *br, memcpy(&temp, &cpu_stats->mstats, sizeof(temp)); } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); - mcast_stats_add_dir(tdst.igmp_queries, temp.igmp_queries); + mcast_stats_add_dir(tdst.igmp_v1queries, temp.igmp_v1queries); + mcast_stats_add_dir(tdst.igmp_v2queries, temp.igmp_v2queries); + mcast_stats_add_dir(tdst.igmp_v3queries, temp.igmp_v3queries); mcast_stats_add_dir(tdst.igmp_leaves, temp.igmp_leaves); mcast_stats_add_dir(tdst.igmp_v1reports, temp.igmp_v1reports); mcast_stats_add_dir(tdst.igmp_v2reports, temp.igmp_v2reports); mcast_stats_add_dir(tdst.igmp_v3reports, temp.igmp_v3reports); tdst.igmp_parse_errors += temp.igmp_parse_errors; - mcast_stats_add_dir(tdst.mld_queries, temp.mld_queries); + mcast_stats_add_dir(tdst.mld_v1queries, temp.mld_v1queries); + mcast_stats_add_dir(tdst.mld_v2queries, temp.mld_v2queries); mcast_stats_add_dir(tdst.mld_leaves, temp.mld_leaves); mcast_stats_add_dir(tdst.mld_v1reports, temp.mld_v1reports); mcast_stats_add_dir(tdst.mld_v2reports, temp.mld_v2reports); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 4dc851166ad1..40f200947ddc 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -586,7 +586,7 @@ void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port, void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port, int type); void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, - __be16 proto, u8 type, u8 dir); + const struct sk_buff *skb, u8 type, u8 dir); int br_multicast_init_stats(struct net_bridge *br); void br_multicast_get_stats(const struct net_bridge *br, const struct net_bridge_port *p, @@ -719,7 +719,8 @@ static inline void br_mdb_uninit(void) static inline void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, - __be16 proto, u8 type, u8 dir) + const struct sk_buff *skb, + u8 type, u8 dir) { } -- cgit v1.2.3 From d55f09abe24b4dfadab246b6f217da547361cdb6 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Mon, 11 Jul 2016 09:45:32 -0300 Subject: [media] lirc.h: remove several unused ioctls While reviewing the documentation gaps on LIRC, it was noticed that several ioctls aren't used by any LIRC drivers (nor at staging or mainstream). It doesn't make sense to document them, as they're not used anywhere. So, let's remove those from the lirc header. Signed-off-by: Mauro Carvalho Chehab --- include/uapi/linux/lirc.h | 39 ++------------------------------------- 1 file changed, 2 insertions(+), 37 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/lirc.h b/include/uapi/linux/lirc.h index 4b3ab2966b5a..991ab4570b8e 100644 --- a/include/uapi/linux/lirc.h +++ b/include/uapi/linux/lirc.h @@ -90,20 +90,11 @@ #define LIRC_GET_SEND_MODE _IOR('i', 0x00000001, __u32) #define LIRC_GET_REC_MODE _IOR('i', 0x00000002, __u32) -#define LIRC_GET_SEND_CARRIER _IOR('i', 0x00000003, __u32) -#define LIRC_GET_REC_CARRIER _IOR('i', 0x00000004, __u32) -#define LIRC_GET_SEND_DUTY_CYCLE _IOR('i', 0x00000005, __u32) -#define LIRC_GET_REC_DUTY_CYCLE _IOR('i', 0x00000006, __u32) #define LIRC_GET_REC_RESOLUTION _IOR('i', 0x00000007, __u32) #define LIRC_GET_MIN_TIMEOUT _IOR('i', 0x00000008, __u32) #define LIRC_GET_MAX_TIMEOUT _IOR('i', 0x00000009, __u32) -#define LIRC_GET_MIN_FILTER_PULSE _IOR('i', 0x0000000a, __u32) -#define LIRC_GET_MAX_FILTER_PULSE _IOR('i', 0x0000000b, __u32) -#define LIRC_GET_MIN_FILTER_SPACE _IOR('i', 0x0000000c, __u32) -#define LIRC_GET_MAX_FILTER_SPACE _IOR('i', 0x0000000d, __u32) - /* code length in bits, currently only for LIRC_MODE_LIRCCODE */ #define LIRC_GET_LENGTH _IOR('i', 0x0000000f, __u32) @@ -113,7 +104,6 @@ #define LIRC_SET_SEND_CARRIER _IOW('i', 0x00000013, __u32) #define LIRC_SET_REC_CARRIER _IOW('i', 0x00000014, __u32) #define LIRC_SET_SEND_DUTY_CYCLE _IOW('i', 0x00000015, __u32) -#define LIRC_SET_REC_DUTY_CYCLE _IOW('i', 0x00000016, __u32) #define LIRC_SET_TRANSMITTER_MASK _IOW('i', 0x00000017, __u32) /* @@ -126,22 +116,6 @@ /* 1 enables, 0 disables timeout reports in MODE2 */ #define LIRC_SET_REC_TIMEOUT_REPORTS _IOW('i', 0x00000019, __u32) -/* - * pulses shorter than this are filtered out by hardware (software - * emulation in lirc_dev?) - */ -#define LIRC_SET_REC_FILTER_PULSE _IOW('i', 0x0000001a, __u32) -/* - * spaces shorter than this are filtered out by hardware (software - * emulation in lirc_dev?) - */ -#define LIRC_SET_REC_FILTER_SPACE _IOW('i', 0x0000001b, __u32) -/* - * if filter cannot be set independently for pulse/space, this should - * be used - */ -#define LIRC_SET_REC_FILTER _IOW('i', 0x0000001c, __u32) - /* * if enabled from the next key press on the driver will send * LIRC_MODE2_FREQUENCY packets @@ -149,20 +123,11 @@ #define LIRC_SET_MEASURE_CARRIER_MODE _IOW('i', 0x0000001d, __u32) /* - * to set a range use - * LIRC_SET_REC_DUTY_CYCLE_RANGE/LIRC_SET_REC_CARRIER_RANGE with the - * lower bound first and later - * LIRC_SET_REC_DUTY_CYCLE/LIRC_SET_REC_CARRIER with the upper bound + * to set a range use LIRC_SET_REC_CARRIER_RANGE with the + * lower bound first and later LIRC_SET_REC_CARRIER with the upper bound */ - -#define LIRC_SET_REC_DUTY_CYCLE_RANGE _IOW('i', 0x0000001e, __u32) #define LIRC_SET_REC_CARRIER_RANGE _IOW('i', 0x0000001f, __u32) -#define LIRC_NOTIFY_DECODE _IO('i', 0x00000020) - -#define LIRC_SETUP_START _IO('i', 0x00000021) -#define LIRC_SETUP_END _IO('i', 0x00000022) - #define LIRC_SET_WIDEBAND_RECEIVER _IOW('i', 0x00000023, __u32) #endif -- cgit v1.2.3 From 28aa4c26fce2202db8d42ae76b639ca1d9a23d25 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:40 +0800 Subject: sctp: add SCTP_PR_SUPPORTED on sctp sockopt According to section 4.5 of rfc7496, prsctp_enable should be per asoc. We will add prsctp_enable to both asoc and ep, and replace the places where it used net.sctp->prsctp_enable with asoc->prsctp_enable. ep->prsctp_enable will be initialized with net.sctp->prsctp_enable, and asoc->prsctp_enable will be initialized with ep->prsctp_enable. We can also modify it's value through sockopt SCTP_PR_SUPPORTED. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 6 ++-- include/uapi/linux/sctp.h | 1 + net/sctp/associola.c | 1 + net/sctp/endpointola.c | 1 + net/sctp/sm_make_chunk.c | 12 +++---- net/sctp/socket.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 93 insertions(+), 8 deletions(-) (limited to 'include/uapi') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 83c5ec58b93a..07115ca9de4d 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1256,7 +1256,8 @@ struct sctp_endpoint { /* SCTP-AUTH: endpoint shared keys */ struct list_head endpoint_shared_keys; __u16 active_key_id; - __u8 auth_enable; + __u8 auth_enable:1, + prsctp_enable:1; }; /* Recover the outter endpoint structure. */ @@ -1848,7 +1849,8 @@ struct sctp_association { __u16 active_key_id; __u8 need_ecne:1, /* Need to send an ECNE Chunk? */ - temp:1; /* Is it a temporary association? */ + temp:1, /* Is it a temporary association? */ + prsctp_enable:1; struct sctp_priv_assoc_stats stats; }; diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index ce70fe6b45df..aa08906f292d 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -112,6 +112,7 @@ typedef __s32 sctp_assoc_t; #define SCTP_SOCKOPT_CONNECTX 110 /* CONNECTX requests. */ #define SCTP_SOCKOPT_CONNECTX3 111 /* CONNECTX requests (updated) */ #define SCTP_GET_ASSOC_STATS 112 /* Read only */ +#define SCTP_PR_SUPPORTED 113 /* These are bit fields for msghdr->msg_flags. See section 5.1. */ /* On user space Linux, these live in as an enum. */ diff --git a/net/sctp/associola.c b/net/sctp/associola.c index e1849f3714ad..1c23060c41a6 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -268,6 +268,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a goto fail_init; asoc->active_key_id = ep->active_key_id; + asoc->prsctp_enable = ep->prsctp_enable; /* Save the hmacs and chunks list into this association */ if (ep->auth_hmacs_list) diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 9d494e35e7f9..1f03065686fe 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -163,6 +163,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, */ ep->auth_hmacs_list = auth_hmacs; ep->auth_chunk_list = auth_chunks; + ep->prsctp_enable = net->sctp.prsctp_enable; return ep; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 56f364d8f932..0e3045ef57fa 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -261,7 +261,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, chunksize += WORD_ROUND(SCTP_SAT_LEN(num_types)); chunksize += sizeof(ecap_param); - if (net->sctp.prsctp_enable) + if (asoc->prsctp_enable) chunksize += sizeof(prsctp_param); /* ADDIP: Section 4.2.7: @@ -355,7 +355,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, sctp_addto_param(retval, num_ext, extensions); } - if (net->sctp.prsctp_enable) + if (asoc->prsctp_enable) sctp_addto_chunk(retval, sizeof(prsctp_param), &prsctp_param); if (sp->adaptation_ind) { @@ -2024,8 +2024,8 @@ static void sctp_process_ext_param(struct sctp_association *asoc, for (i = 0; i < num_ext; i++) { switch (param.ext->chunks[i]) { case SCTP_CID_FWD_TSN: - if (net->sctp.prsctp_enable && !asoc->peer.prsctp_capable) - asoc->peer.prsctp_capable = 1; + if (asoc->prsctp_enable && !asoc->peer.prsctp_capable) + asoc->peer.prsctp_capable = 1; break; case SCTP_CID_AUTH: /* if the peer reports AUTH, assume that he @@ -2169,7 +2169,7 @@ static sctp_ierror_t sctp_verify_param(struct net *net, break; case SCTP_PARAM_FWD_TSN_SUPPORT: - if (net->sctp.prsctp_enable) + if (ep->prsctp_enable) break; goto fallthrough; @@ -2653,7 +2653,7 @@ do_addr_param: break; case SCTP_PARAM_FWD_TSN_SUPPORT: - if (net->sctp.prsctp_enable) { + if (asoc->prsctp_enable) { asoc->peer.prsctp_capable = 1; break; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index cdabbd8219b1..7460ddebd9ce 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3661,6 +3661,39 @@ static int sctp_setsockopt_recvnxtinfo(struct sock *sk, return 0; } +static int sctp_setsockopt_pr_supported(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EINVAL; + + if (optlen != sizeof(params)) + goto out; + + if (copy_from_user(¶ms, optval, optlen)) { + retval = -EFAULT; + goto out; + } + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (asoc) { + asoc->prsctp_enable = !!params.assoc_value; + } else if (!params.assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + sp->ep->prsctp_enable = !!params.assoc_value; + } else { + goto out; + } + + retval = 0; + +out: + return retval; +} + /* API 6.2 setsockopt(), getsockopt() * * Applications use setsockopt() and getsockopt() to set or retrieve @@ -3821,6 +3854,9 @@ static int sctp_setsockopt(struct sock *sk, int level, int optname, case SCTP_RECVNXTINFO: retval = sctp_setsockopt_recvnxtinfo(sk, optval, optlen); break; + case SCTP_PR_SUPPORTED: + retval = sctp_setsockopt_pr_supported(sk, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; @@ -6166,6 +6202,47 @@ static int sctp_getsockopt_recvnxtinfo(struct sock *sk, int len, return 0; } +static int sctp_getsockopt_pr_supported(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(params)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (asoc) { + params.assoc_value = asoc->prsctp_enable; + } else if (!params.assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + params.assoc_value = sp->ep->prsctp_enable; + } else { + retval = -EINVAL; + goto out; + } + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, ¶ms, len)) + goto out; + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -6319,6 +6396,9 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, case SCTP_RECVNXTINFO: retval = sctp_getsockopt_recvnxtinfo(sk, len, optval, optlen); break; + case SCTP_PR_SUPPORTED: + retval = sctp_getsockopt_pr_supported(sk, len, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; -- cgit v1.2.3 From f959fb442c35f4b61fea341401b8463dd0a1b959 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:41 +0800 Subject: sctp: add SCTP_DEFAULT_PRINFO into sctp sockopt This patch adds SCTP_DEFAULT_PRINFO to sctp sockopt. It is used to set/get sctp Partially Reliable Policies' default params, which includes 3 policies (ttl, rtx, prio) and their values. Still, if we set policy params in sndinfo, we will use the params of sndinfo against chunks, instead of the default params. In this patch, we will use 5-8bit of sp/asoc->default_flags to store prsctp policies, and reuse asoc->default_timetolive to store their values. It means if we enable and set prsctp policy, prior ttl timeout in sctp will not work any more. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/uapi/linux/sctp.h | 29 +++++++++++++++ net/sctp/socket.c | 91 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 120 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index aa08906f292d..984cf2e9a61d 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -113,6 +113,29 @@ typedef __s32 sctp_assoc_t; #define SCTP_SOCKOPT_CONNECTX3 111 /* CONNECTX requests (updated) */ #define SCTP_GET_ASSOC_STATS 112 /* Read only */ #define SCTP_PR_SUPPORTED 113 +#define SCTP_DEFAULT_PRINFO 114 + +/* PR-SCTP policies */ +#define SCTP_PR_SCTP_NONE 0x0000 +#define SCTP_PR_SCTP_TTL 0x0010 +#define SCTP_PR_SCTP_RTX 0x0020 +#define SCTP_PR_SCTP_PRIO 0x0030 +#define SCTP_PR_SCTP_MAX SCTP_PR_SCTP_PRIO +#define SCTP_PR_SCTP_MASK 0x0030 + +#define __SCTP_PR_INDEX(x) ((x >> 4) - 1) +#define SCTP_PR_INDEX(x) __SCTP_PR_INDEX(SCTP_PR_SCTP_ ## x) + +#define SCTP_PR_POLICY(x) ((x) & SCTP_PR_SCTP_MASK) +#define SCTP_PR_SET_POLICY(flags, x) \ + do { \ + flags &= ~SCTP_PR_SCTP_MASK; \ + flags |= x; \ + } while (0) + +#define SCTP_PR_TTL_ENABLED(x) (SCTP_PR_POLICY(x) == SCTP_PR_SCTP_TTL) +#define SCTP_PR_RTX_ENABLED(x) (SCTP_PR_POLICY(x) == SCTP_PR_SCTP_RTX) +#define SCTP_PR_PRIO_ENABLED(x) (SCTP_PR_POLICY(x) == SCTP_PR_SCTP_PRIO) /* These are bit fields for msghdr->msg_flags. See section 5.1. */ /* On user space Linux, these live in as an enum. */ @@ -903,4 +926,10 @@ struct sctp_paddrthlds { __u16 spt_pathpfthld; }; +struct sctp_default_prinfo { + sctp_assoc_t pr_assoc_id; + __u32 pr_value; + __u16 pr_policy; +}; + #endif /* _UAPI_SCTP_H */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 7460ddebd9ce..c03fe1b76706 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3694,6 +3694,47 @@ out: return retval; } +static int sctp_setsockopt_default_prinfo(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_default_prinfo info; + struct sctp_association *asoc; + int retval = -EINVAL; + + if (optlen != sizeof(info)) + goto out; + + if (copy_from_user(&info, optval, sizeof(info))) { + retval = -EFAULT; + goto out; + } + + if (info.pr_policy & ~SCTP_PR_SCTP_MASK) + goto out; + + if (info.pr_policy == SCTP_PR_SCTP_NONE) + info.pr_value = 0; + + asoc = sctp_id2assoc(sk, info.pr_assoc_id); + if (asoc) { + SCTP_PR_SET_POLICY(asoc->default_flags, info.pr_policy); + asoc->default_timetolive = info.pr_value; + } else if (!info.pr_assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + SCTP_PR_SET_POLICY(sp->default_flags, info.pr_policy); + sp->default_timetolive = info.pr_value; + } else { + goto out; + } + + retval = 0; + +out: + return retval; +} + /* API 6.2 setsockopt(), getsockopt() * * Applications use setsockopt() and getsockopt() to set or retrieve @@ -3857,6 +3898,9 @@ static int sctp_setsockopt(struct sock *sk, int level, int optname, case SCTP_PR_SUPPORTED: retval = sctp_setsockopt_pr_supported(sk, optval, optlen); break; + case SCTP_DEFAULT_PRINFO: + retval = sctp_setsockopt_default_prinfo(sk, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; @@ -6243,6 +6287,49 @@ out: return retval; } +static int sctp_getsockopt_default_prinfo(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_default_prinfo info; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(info)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(info); + if (copy_from_user(&info, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, info.pr_assoc_id); + if (asoc) { + info.pr_policy = SCTP_PR_POLICY(asoc->default_flags); + info.pr_value = asoc->default_timetolive; + } else if (!info.pr_assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + info.pr_policy = SCTP_PR_POLICY(sp->default_flags); + info.pr_value = sp->default_timetolive; + } else { + retval = -EINVAL; + goto out; + } + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, &info, len)) + goto out; + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -6399,6 +6486,10 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, case SCTP_PR_SUPPORTED: retval = sctp_getsockopt_pr_supported(sk, len, optval, optlen); break; + case SCTP_DEFAULT_PRINFO: + retval = sctp_getsockopt_default_prinfo(sk, len, optval, + optlen); + break; default: retval = -ENOPROTOOPT; break; -- cgit v1.2.3 From 826d253d57b11f69add81c8086d2e7f1dce5ec77 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:42 +0800 Subject: sctp: add SCTP_PR_ASSOC_STATUS on sctp sockopt This patch adds SCTP_PR_ASSOC_STATUS to sctp sockopt, which is used to dump the prsctp statistics info from the asoc. The prsctp statistics includes abandoned_sent/unsent from the asoc. abandoned_sent is the count of the packets we drop packets from retransmit/transmited queue, and abandoned_unsent is the count of the packets we drop from out_queue according to the policy. Note: another option for prsctp statistics dump described in rfc is SCTP_PR_STREAM_STATUS, which is used to dump the prsctp statistics info from each stream. But by now, linux doesn't yet have per stream statistics info, it needs rfc6525 to be implemented. As the prsctp statistics for each stream has to be based on per stream statistics, we will delay it until rfc6525 is done in linux. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 3 +++ include/uapi/linux/sctp.h | 12 +++++++++ net/sctp/socket.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 77 insertions(+) (limited to 'include/uapi') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 07115ca9de4d..d8e464aacb20 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1853,6 +1853,9 @@ struct sctp_association { prsctp_enable:1; struct sctp_priv_assoc_stats stats; + + __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; + __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; }; diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index 984cf2e9a61d..d304f4c9792c 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -114,6 +114,7 @@ typedef __s32 sctp_assoc_t; #define SCTP_GET_ASSOC_STATS 112 /* Read only */ #define SCTP_PR_SUPPORTED 113 #define SCTP_DEFAULT_PRINFO 114 +#define SCTP_PR_ASSOC_STATUS 115 /* PR-SCTP policies */ #define SCTP_PR_SCTP_NONE 0x0000 @@ -926,6 +927,17 @@ struct sctp_paddrthlds { __u16 spt_pathpfthld; }; +/* + * Socket Option for Getting the Association/Stream-Specific PR-SCTP Status + */ +struct sctp_prstatus { + sctp_assoc_t sprstat_assoc_id; + __u16 sprstat_sid; + __u16 sprstat_policy; + __u64 sprstat_abandoned_unsent; + __u64 sprstat_abandoned_sent; +}; + struct sctp_default_prinfo { sctp_assoc_t pr_assoc_id; __u32 pr_value; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index c03fe1b76706..c3167c43a9c1 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -6330,6 +6330,64 @@ out: return retval; } +static int sctp_getsockopt_pr_assocstatus(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_prstatus params; + struct sctp_association *asoc; + int policy; + int retval = -EINVAL; + + if (len < sizeof(params)) + goto out; + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) { + retval = -EFAULT; + goto out; + } + + policy = params.sprstat_policy; + if (policy & ~SCTP_PR_SCTP_MASK) + goto out; + + asoc = sctp_id2assoc(sk, params.sprstat_assoc_id); + if (!asoc) + goto out; + + if (policy == SCTP_PR_SCTP_NONE) { + params.sprstat_abandoned_unsent = 0; + params.sprstat_abandoned_sent = 0; + for (policy = 0; policy <= SCTP_PR_INDEX(MAX); policy++) { + params.sprstat_abandoned_unsent += + asoc->abandoned_unsent[policy]; + params.sprstat_abandoned_sent += + asoc->abandoned_sent[policy]; + } + } else { + params.sprstat_abandoned_unsent = + asoc->abandoned_unsent[__SCTP_PR_INDEX(policy)]; + params.sprstat_abandoned_sent = + asoc->abandoned_sent[__SCTP_PR_INDEX(policy)]; + } + + if (put_user(len, optlen)) { + retval = -EFAULT; + goto out; + } + + if (copy_to_user(optval, ¶ms, len)) { + retval = -EFAULT; + goto out; + } + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -6490,6 +6548,10 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, retval = sctp_getsockopt_default_prinfo(sk, len, optval, optlen); break; + case SCTP_PR_ASSOC_STATUS: + retval = sctp_getsockopt_pr_assocstatus(sk, len, optval, + optlen); + break; default: retval = -ENOPROTOOPT; break; -- cgit v1.2.3 From c624b5de5037fab77c75db8272befb2df52a968f Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Sun, 10 Jul 2016 07:08:58 -0300 Subject: [media] input: serio - add new protocol for the Pulse-Eight USB-CEC Adapter This is for the new pulse8-cec staging driver. Signed-off-by: Hans Verkuil Acked-by: Dmitry Torokhov Signed-off-by: Mauro Carvalho Chehab --- include/uapi/linux/serio.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/serio.h b/include/uapi/linux/serio.h index c2ea1698257f..f2447a83ac8d 100644 --- a/include/uapi/linux/serio.h +++ b/include/uapi/linux/serio.h @@ -78,5 +78,6 @@ #define SERIO_TSC40 0x3d #define SERIO_WACOM_IV 0x3e #define SERIO_EGALAX 0x3f +#define SERIO_PULSE8_CEC 0x40 #endif /* _UAPI_SERIO_H */ -- cgit v1.2.3 From 3713131345fbea291cbd859d248e06ed77815962 Mon Sep 17 00:00:00 2001 From: Radim Krčmář Date: Tue, 12 Jul 2016 22:09:27 +0200 Subject: KVM: x86: add KVM_CAP_X2APIC_API MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit KVM_CAP_X2APIC_API is a capability for features related to x2APIC enablement. KVM_X2APIC_API_32BIT_FORMAT feature can be enabled to extend APIC ID in get/set ioctl and MSI addresses to 32 bits. Both are needed to support x2APIC. The feature has to be enableable and disabled by default, because get/set ioctl shifted and truncated APIC ID to 8 bits by using a non-standard protocol inspired by xAPIC and the change is not backward-compatible. Changes to MSI addresses follow the format used by interrupt remapping unit. The upper address word, that used to be 0, contains upper 24 bits of the LAPIC address in its upper 24 bits. Lower 8 bits are reserved as 0. Using the upper address word is not backward-compatible either as we didn't check that userspace zeroed the word. Reserved bits are still not explicitly checked, but non-zero data will affect LAPIC addresses, which will cause a bug. Signed-off-by: Radim Krčmář Signed-off-by: Paolo Bonzini --- Documentation/virtual/kvm/api.txt | 41 +++++++++++++++++++++++++++++++++++++++ arch/x86/include/asm/kvm_host.h | 4 +++- arch/x86/kvm/irq_comm.c | 29 ++++++++++++++++++++++----- arch/x86/kvm/lapic.c | 13 +++++++++---- arch/x86/kvm/vmx.c | 2 +- arch/x86/kvm/x86.c | 15 ++++++++++++++ include/trace/events/kvm.h | 5 +++-- include/uapi/linux/kvm.h | 3 +++ 8 files changed, 99 insertions(+), 13 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 09efa9eb3926..e34e51fa28b0 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -1482,6 +1482,11 @@ struct kvm_irq_routing_msi { __u32 pad; }; +On x86, address_hi is ignored unless the KVM_X2APIC_API_USE_32BIT_IDS +feature of KVM_CAP_X2APIC_API capability is enabled. If it is enabled, +address_hi bits 31-8 provide bits 31-8 of the destination id. Bits 7-0 of +address_hi must be zero. + struct kvm_irq_routing_s390_adapter { __u64 ind_addr; __u64 summary_addr; @@ -1583,6 +1588,17 @@ struct kvm_lapic_state { Reads the Local APIC registers and copies them into the input argument. The data format and layout are the same as documented in the architecture manual. +If KVM_X2APIC_API_USE_32BIT_IDS feature of KVM_CAP_X2APIC_API is +enabled, then the format of APIC_ID register depends on the APIC mode +(reported by MSR_IA32_APICBASE) of its VCPU. x2APIC stores APIC ID in +the APIC_ID register (bytes 32-35). xAPIC only allows an 8-bit APIC ID +which is stored in bits 31-24 of the APIC register, or equivalently in +byte 35 of struct kvm_lapic_state's regs field. KVM_GET_LAPIC must then +be called after MSR_IA32_APICBASE has been set with KVM_SET_MSR. + +If KVM_X2APIC_API_USE_32BIT_IDS feature is disabled, struct kvm_lapic_state +always uses xAPIC format. + 4.58 KVM_SET_LAPIC @@ -1600,6 +1616,10 @@ struct kvm_lapic_state { Copies the input argument into the Local APIC registers. The data format and layout are the same as documented in the architecture manual. +The format of the APIC ID register (bytes 32-35 of struct kvm_lapic_state's +regs field) depends on the state of the KVM_CAP_X2APIC_API capability. +See the note in KVM_GET_LAPIC. + 4.59 KVM_IOEVENTFD @@ -2180,6 +2200,10 @@ struct kvm_msi { No flags are defined so far. The corresponding field must be 0. +On x86, address_hi is ignored unless the KVM_CAP_X2APIC_API capability is +enabled. If it is enabled, address_hi bits 31-8 provide bits 31-8 of the +destination id. Bits 7-0 of address_hi must be zero. + 4.71 KVM_CREATE_PIT2 @@ -3811,6 +3835,23 @@ Allows use of runtime-instrumentation introduced with zEC12 processor. Will return -EINVAL if the machine does not support runtime-instrumentation. Will return -EBUSY if a VCPU has already been created. +7.7 KVM_CAP_X2APIC_API + +Architectures: x86 +Parameters: args[0] - features that should be enabled +Returns: 0 on success, -EINVAL when args[0] contains invalid features + +Valid feature flags in args[0] are + +#define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) + +Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of +KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, +allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their +respective sections. + + + 8. Other capabilities. ---------------------- diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index a2832cc3cb81..7c00ba3242d7 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -782,6 +782,8 @@ struct kvm_arch { u32 ldr_mode; struct page *avic_logical_id_table_page; struct page *avic_physical_id_table_page; + + bool x2apic_format; }; struct kvm_vm_stat { @@ -1364,7 +1366,7 @@ bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu); bool kvm_intr_is_single_vcpu(struct kvm *kvm, struct kvm_lapic_irq *irq, struct kvm_vcpu **dest_vcpu); -void kvm_set_msi_irq(struct kvm_kernel_irq_routing_entry *e, +void kvm_set_msi_irq(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e, struct kvm_lapic_irq *irq); static inline void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu) diff --git a/arch/x86/kvm/irq_comm.c b/arch/x86/kvm/irq_comm.c index 889563d50c55..25810b144b58 100644 --- a/arch/x86/kvm/irq_comm.c +++ b/arch/x86/kvm/irq_comm.c @@ -110,13 +110,17 @@ int kvm_irq_delivery_to_apic(struct kvm *kvm, struct kvm_lapic *src, return r; } -void kvm_set_msi_irq(struct kvm_kernel_irq_routing_entry *e, +void kvm_set_msi_irq(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e, struct kvm_lapic_irq *irq) { - trace_kvm_msi_set_irq(e->msi.address_lo, e->msi.data); + trace_kvm_msi_set_irq(e->msi.address_lo | (kvm->arch.x2apic_format ? + (u64)e->msi.address_hi << 32 : 0), + e->msi.data); irq->dest_id = (e->msi.address_lo & MSI_ADDR_DEST_ID_MASK) >> MSI_ADDR_DEST_ID_SHIFT; + if (kvm->arch.x2apic_format) + irq->dest_id |= MSI_ADDR_EXT_DEST_ID(e->msi.address_hi); irq->vector = (e->msi.data & MSI_DATA_VECTOR_MASK) >> MSI_DATA_VECTOR_SHIFT; irq->dest_mode = (1 << MSI_ADDR_DEST_MODE_SHIFT) & e->msi.address_lo; @@ -129,15 +133,24 @@ void kvm_set_msi_irq(struct kvm_kernel_irq_routing_entry *e, } EXPORT_SYMBOL_GPL(kvm_set_msi_irq); +static inline bool kvm_msi_route_invalid(struct kvm *kvm, + struct kvm_kernel_irq_routing_entry *e) +{ + return kvm->arch.x2apic_format && (e->msi.address_hi & 0xff); +} + int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e, struct kvm *kvm, int irq_source_id, int level, bool line_status) { struct kvm_lapic_irq irq; + if (kvm_msi_route_invalid(kvm, e)) + return -EINVAL; + if (!level) return -1; - kvm_set_msi_irq(e, &irq); + kvm_set_msi_irq(kvm, e, &irq); return kvm_irq_delivery_to_apic(kvm, NULL, &irq, NULL); } @@ -153,7 +166,10 @@ int kvm_arch_set_irq_inatomic(struct kvm_kernel_irq_routing_entry *e, if (unlikely(e->type != KVM_IRQ_ROUTING_MSI)) return -EWOULDBLOCK; - kvm_set_msi_irq(e, &irq); + if (kvm_msi_route_invalid(kvm, e)) + return -EINVAL; + + kvm_set_msi_irq(kvm, e, &irq); if (kvm_irq_delivery_to_apic_fast(kvm, NULL, &irq, &r, NULL)) return r; @@ -286,6 +302,9 @@ int kvm_set_routing_entry(struct kvm *kvm, e->msi.address_lo = ue->u.msi.address_lo; e->msi.address_hi = ue->u.msi.address_hi; e->msi.data = ue->u.msi.data; + + if (kvm_msi_route_invalid(kvm, e)) + goto out; break; case KVM_IRQ_ROUTING_HV_SINT: e->set = kvm_hv_set_sint; @@ -394,7 +413,7 @@ void kvm_scan_ioapic_routes(struct kvm_vcpu *vcpu, if (entry->type != KVM_IRQ_ROUTING_MSI) continue; - kvm_set_msi_irq(entry, &irq); + kvm_set_msi_irq(vcpu->kvm, entry, &irq); if (irq.level && kvm_apic_match_dest(vcpu, NULL, 0, irq.dest_id, irq.dest_mode)) diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 3c2a8c113054..d27a7829a4ce 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1991,10 +1991,15 @@ static int kvm_apic_state_fixup(struct kvm_vcpu *vcpu, if (apic_x2apic_mode(vcpu->arch.apic)) { u32 *id = (u32 *)(s->regs + APIC_ID); - if (set) - *id >>= 24; - else - *id <<= 24; + if (vcpu->kvm->arch.x2apic_format) { + if (*id != vcpu->vcpu_id) + return -EINVAL; + } else { + if (set) + *id >>= 24; + else + *id <<= 24; + } } return 0; diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 7bdd6b1a2373..b61cdadf8623 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -11104,7 +11104,7 @@ static int vmx_update_pi_irte(struct kvm *kvm, unsigned int host_irq, * We will support full lowest-priority interrupt later. */ - kvm_set_msi_irq(e, &irq); + kvm_set_msi_irq(kvm, e, &irq); if (!kvm_intr_is_single_vcpu(kvm, &irq, &vcpu)) { /* * Make sure the IRTE is in remapped mode if diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b6e402d16e0c..d86f563a6896 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -90,6 +90,8 @@ static u64 __read_mostly efer_reserved_bits = ~((u64)EFER_SCE); #define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU +#define KVM_X2APIC_API_VALID_FLAGS (KVM_X2APIC_API_USE_32BIT_IDS) + static void update_cr8_intercept(struct kvm_vcpu *vcpu); static void process_nmi(struct kvm_vcpu *vcpu); static void enter_smm(struct kvm_vcpu *vcpu); @@ -2625,6 +2627,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_TSC_CONTROL: r = kvm_has_tsc_control; break; + case KVM_CAP_X2APIC_API: + r = KVM_X2APIC_API_VALID_FLAGS; + break; default: r = 0; break; @@ -3799,6 +3804,16 @@ split_irqchip_unlock: mutex_unlock(&kvm->lock); break; } + case KVM_CAP_X2APIC_API: + r = -EINVAL; + if (cap->args[0] & ~KVM_X2APIC_API_VALID_FLAGS) + break; + + if (cap->args[0] & KVM_X2APIC_API_USE_32BIT_IDS) + kvm->arch.x2apic_format = true; + + r = 0; + break; default: r = -EINVAL; break; diff --git a/include/trace/events/kvm.h b/include/trace/events/kvm.h index f28292d73ddb..8ade3eb6c640 100644 --- a/include/trace/events/kvm.h +++ b/include/trace/events/kvm.h @@ -151,8 +151,9 @@ TRACE_EVENT(kvm_msi_set_irq, __entry->data = data; ), - TP_printk("dst %u vec %u (%s|%s|%s%s)", - (u8)(__entry->address >> 12), (u8)__entry->data, + TP_printk("dst %llx vec %u (%s|%s|%s%s)", + (u8)(__entry->address >> 12) | ((__entry->address >> 32) & 0xffffff00), + (u8)__entry->data, __print_symbolic((__entry->data >> 8 & 0x7), kvm_deliver_mode), (__entry->address & (1<<2)) ? "logical" : "physical", (__entry->data & (1<<15)) ? "level" : "edge", diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 05ebf475104c..f704403e19a0 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -866,6 +866,7 @@ struct kvm_ppc_smmu_info { #define KVM_CAP_ARM_PMU_V3 126 #define KVM_CAP_VCPU_ATTRIBUTES 127 #define KVM_CAP_MAX_VCPU_ID 128 +#define KVM_CAP_X2APIC_API 129 #ifdef KVM_CAP_IRQ_ROUTING @@ -1313,4 +1314,6 @@ struct kvm_assigned_msix_entry { __u16 padding[3]; }; +#define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) + #endif /* __LINUX_KVM_H */ -- cgit v1.2.3 From c519265f2aa348b2f1b9ecf8fbe20bb7c0fb102e Mon Sep 17 00:00:00 2001 From: Radim Krčmář Date: Tue, 12 Jul 2016 22:09:28 +0200 Subject: KVM: x86: add a flag to disable KVM x2apic broadcast quirk MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK as a feature flag to KVM_CAP_X2APIC_API. The quirk made KVM interpret 0xff as a broadcast even in x2APIC mode. The enableable capability is needed in order to support standard x2APIC and remain backward compatible. Signed-off-by: Radim Krčmář [Expand kvm_apic_mda comment. - Paolo] Signed-off-by: Paolo Bonzini --- Documentation/virtual/kvm/api.txt | 6 +++++ arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/lapic.c | 53 +++++++++++++++++++++++++++++---------- arch/x86/kvm/x86.c | 5 +++- include/uapi/linux/kvm.h | 1 + 5 files changed, 52 insertions(+), 14 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index e34e51fa28b0..c4d2fb0e28de 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -3844,12 +3844,18 @@ Returns: 0 on success, -EINVAL when args[0] contains invalid features Valid feature flags in args[0] are #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) +#define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) Enabling KVM_X2APIC_API_USE_32BIT_IDS changes the behavior of KVM_SET_GSI_ROUTING, KVM_SIGNAL_MSI, KVM_SET_LAPIC, and KVM_GET_LAPIC, allowing the use of 32-bit APIC IDs. See KVM_CAP_X2APIC_API in their respective sections. +KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK must be enabled for x2APIC to work +in logical mode or with more than 255 VCPUs. Otherwise, KVM treats 0xff +as a broadcast even in x2APIC mode in order to support physical x2APIC +without interrupt remapping. This is undesirable in logical mode, +where 0xff represents CPUs 0-7 in cluster 0. 8. Other capabilities. diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 7c00ba3242d7..074b5c760327 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -784,6 +784,7 @@ struct kvm_arch { struct page *avic_physical_id_table_page; bool x2apic_format; + bool x2apic_broadcast_quirk_disabled; }; struct kvm_vm_stat { diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index d27a7829a4ce..a16e0bb95d28 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -616,17 +616,30 @@ static bool kvm_apic_match_logical_addr(struct kvm_lapic *apic, u32 mda) } } -/* KVM APIC implementation has two quirks - * - dest always begins at 0 while xAPIC MDA has offset 24, - * - IOxAPIC messages have to be delivered (directly) to x2APIC. +/* The KVM local APIC implementation has two quirks: + * + * - the xAPIC MDA stores the destination at bits 24-31, while this + * is not true of struct kvm_lapic_irq's dest_id field. This is + * just a quirk in the API and is not problematic. + * + * - in-kernel IOAPIC messages have to be delivered directly to + * x2APIC, because the kernel does not support interrupt remapping. + * In order to support broadcast without interrupt remapping, x2APIC + * rewrites the destination of non-IPI messages from APIC_BROADCAST + * to X2APIC_BROADCAST. + * + * The broadcast quirk can be disabled with KVM_CAP_X2APIC_API. This is + * important when userspace wants to use x2APIC-format MSIs, because + * APIC_BROADCAST (0xff) is a legal route for "cluster 0, CPUs 0-7". */ -static u32 kvm_apic_mda(unsigned int dest_id, struct kvm_lapic *source, - struct kvm_lapic *target) +static u32 kvm_apic_mda(struct kvm_vcpu *vcpu, unsigned int dest_id, + struct kvm_lapic *source, struct kvm_lapic *target) { bool ipi = source != NULL; bool x2apic_mda = apic_x2apic_mode(ipi ? source : target); - if (!ipi && dest_id == APIC_BROADCAST && x2apic_mda) + if (!vcpu->kvm->arch.x2apic_broadcast_quirk_disabled && + !ipi && dest_id == APIC_BROADCAST && x2apic_mda) return X2APIC_BROADCAST; return x2apic_mda ? dest_id : SET_APIC_DEST_FIELD(dest_id); @@ -636,7 +649,7 @@ bool kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source, int short_hand, unsigned int dest, int dest_mode) { struct kvm_lapic *target = vcpu->arch.apic; - u32 mda = kvm_apic_mda(dest, source, target); + u32 mda = kvm_apic_mda(vcpu, dest, source, target); apic_debug("target %p, source %p, dest 0x%x, " "dest_mode 0x%x, short_hand 0x%x\n", @@ -688,6 +701,25 @@ static void kvm_apic_disabled_lapic_found(struct kvm *kvm) } } +static bool kvm_apic_is_broadcast_dest(struct kvm *kvm, struct kvm_lapic **src, + struct kvm_lapic_irq *irq, struct kvm_apic_map *map) +{ + if (kvm->arch.x2apic_broadcast_quirk_disabled) { + if ((irq->dest_id == APIC_BROADCAST && + map->mode != KVM_APIC_MODE_X2APIC)) + return true; + if (irq->dest_id == X2APIC_BROADCAST) + return true; + } else { + bool x2apic_ipi = src && *src && apic_x2apic_mode(*src); + if (irq->dest_id == (x2apic_ipi ? + X2APIC_BROADCAST : APIC_BROADCAST)) + return true; + } + + return false; +} + /* Return true if the interrupt can be handled by using *bitmap as index mask * for valid destinations in *dst array. * Return false if kvm_apic_map_get_dest_lapic did nothing useful. @@ -701,7 +733,6 @@ static inline bool kvm_apic_map_get_dest_lapic(struct kvm *kvm, unsigned long *bitmap) { int i, lowest; - bool x2apic_ipi; if (irq->shorthand == APIC_DEST_SELF && src) { *dst = src; @@ -710,11 +741,7 @@ static inline bool kvm_apic_map_get_dest_lapic(struct kvm *kvm, } else if (irq->shorthand) return false; - x2apic_ipi = src && *src && apic_x2apic_mode(*src); - if (irq->dest_id == (x2apic_ipi ? X2APIC_BROADCAST : APIC_BROADCAST)) - return false; - - if (!map) + if (!map || kvm_apic_is_broadcast_dest(kvm, src, irq, map)) return false; if (irq->dest_mode == APIC_DEST_PHYSICAL) { diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index d86f563a6896..f0d23622bc4e 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -90,7 +90,8 @@ static u64 __read_mostly efer_reserved_bits = ~((u64)EFER_SCE); #define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU -#define KVM_X2APIC_API_VALID_FLAGS (KVM_X2APIC_API_USE_32BIT_IDS) +#define KVM_X2APIC_API_VALID_FLAGS (KVM_X2APIC_API_USE_32BIT_IDS | \ + KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK) static void update_cr8_intercept(struct kvm_vcpu *vcpu); static void process_nmi(struct kvm_vcpu *vcpu); @@ -3811,6 +3812,8 @@ split_irqchip_unlock: if (cap->args[0] & KVM_X2APIC_API_USE_32BIT_IDS) kvm->arch.x2apic_format = true; + if (cap->args[0] & KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK) + kvm->arch.x2apic_broadcast_quirk_disabled = true; r = 0; break; diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index f704403e19a0..4f8030e5b05d 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -1315,5 +1315,6 @@ struct kvm_assigned_msix_entry { }; #define KVM_X2APIC_API_USE_32BIT_IDS (1ULL << 0) +#define KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK (1ULL << 1) #endif /* __LINUX_KVM_H */ -- cgit v1.2.3 From af713795c59fea36161a7debf97dbc10bf652cf7 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Fri, 1 Jul 2016 13:10:38 -0700 Subject: drm/vc4: Add a getparam ioctl for getting the V3D identity regs. As I extend the driver to support different V3D revisions, userspace needs to know what version it's targeting. This is most easily detected using the V3D identity registers. v2: Make sure V3D is runtime PM on when reading the registers. v3: Switch to a 64-bit param value (suggested by Rob Clark in review) Signed-off-by: Eric Anholt Acked-by: Daniel Vetter (v2) Reviewed-by: Rob Clark (v3, over irc) --- drivers/gpu/drm/vc4/vc4_drv.c | 42 ++++++++++++++++++++++++++++++++++++++++++ include/uapi/drm/vc4_drm.h | 12 ++++++++++++ 2 files changed, 54 insertions(+) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index 2f30214ee810..047d7a265ceb 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -14,6 +14,7 @@ #include #include #include +#include #include "drm_fb_cma_helper.h" #include "uapi/drm/vc4_drm.h" @@ -43,6 +44,46 @@ void __iomem *vc4_ioremap_regs(struct platform_device *dev, int index) return map; } +static int vc4_get_param_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct vc4_dev *vc4 = to_vc4_dev(dev); + struct drm_vc4_get_param *args = data; + int ret; + + if (args->pad != 0) + return -EINVAL; + + switch (args->param) { + case DRM_VC4_PARAM_V3D_IDENT0: + ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev); + if (ret) + return ret; + args->value = V3D_READ(V3D_IDENT0); + pm_runtime_put(&vc4->v3d->pdev->dev); + break; + case DRM_VC4_PARAM_V3D_IDENT1: + ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev); + if (ret) + return ret; + args->value = V3D_READ(V3D_IDENT1); + pm_runtime_put(&vc4->v3d->pdev->dev); + break; + case DRM_VC4_PARAM_V3D_IDENT2: + ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev); + if (ret) + return ret; + args->value = V3D_READ(V3D_IDENT2); + pm_runtime_put(&vc4->v3d->pdev->dev); + break; + default: + DRM_DEBUG("Unknown parameter %d\n", args->param); + return -EINVAL; + } + + return 0; +} + static void vc4_lastclose(struct drm_device *dev) { struct vc4_dev *vc4 = to_vc4_dev(dev); @@ -74,6 +115,7 @@ static const struct drm_ioctl_desc vc4_drm_ioctls[] = { DRM_IOCTL_DEF_DRV(VC4_CREATE_SHADER_BO, vc4_create_shader_bo_ioctl, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(VC4_GET_HANG_STATE, vc4_get_hang_state_ioctl, DRM_ROOT_ONLY), + DRM_IOCTL_DEF_DRV(VC4_GET_PARAM, vc4_get_param_ioctl, DRM_RENDER_ALLOW), }; static struct drm_driver vc4_drm_driver = { diff --git a/include/uapi/drm/vc4_drm.h b/include/uapi/drm/vc4_drm.h index af12e8a184c8..1143e954048d 100644 --- a/include/uapi/drm/vc4_drm.h +++ b/include/uapi/drm/vc4_drm.h @@ -37,6 +37,7 @@ extern "C" { #define DRM_VC4_MMAP_BO 0x04 #define DRM_VC4_CREATE_SHADER_BO 0x05 #define DRM_VC4_GET_HANG_STATE 0x06 +#define DRM_VC4_GET_PARAM 0x07 #define DRM_IOCTL_VC4_SUBMIT_CL DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_SUBMIT_CL, struct drm_vc4_submit_cl) #define DRM_IOCTL_VC4_WAIT_SEQNO DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_WAIT_SEQNO, struct drm_vc4_wait_seqno) @@ -45,6 +46,7 @@ extern "C" { #define DRM_IOCTL_VC4_MMAP_BO DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_MMAP_BO, struct drm_vc4_mmap_bo) #define DRM_IOCTL_VC4_CREATE_SHADER_BO DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_CREATE_SHADER_BO, struct drm_vc4_create_shader_bo) #define DRM_IOCTL_VC4_GET_HANG_STATE DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_GET_HANG_STATE, struct drm_vc4_get_hang_state) +#define DRM_IOCTL_VC4_GET_PARAM DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_GET_PARAM, struct drm_vc4_get_param) struct drm_vc4_submit_rcl_surface { __u32 hindex; /* Handle index, or ~0 if not present. */ @@ -280,6 +282,16 @@ struct drm_vc4_get_hang_state { __u32 pad[16]; }; +#define DRM_VC4_PARAM_V3D_IDENT0 0 +#define DRM_VC4_PARAM_V3D_IDENT1 1 +#define DRM_VC4_PARAM_V3D_IDENT2 2 + +struct drm_vc4_get_param { + __u32 param; + __u32 pad; + __u64 value; +}; + #if defined(__cplusplus) } #endif -- cgit v1.2.3 From 555c8a8623a3a87b3c990ba30b7fd2e5914e41d2 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 14 Jul 2016 18:08:05 +0200 Subject: bpf: avoid stack copy and use skb ctx for event output This work addresses a couple of issues bpf_skb_event_output() helper currently has: i) We need two copies instead of just a single one for the skb data when it should be part of a sample. The data can be non-linear and thus needs to be extracted via bpf_skb_load_bytes() helper first, and then copied once again into the ring buffer slot. ii) Since bpf_skb_load_bytes() currently needs to be used first, the helper needs to see a constant size on the passed stack buffer to make sure BPF verifier can do sanity checks on it during verification time. Thus, just passing skb->len (or any other non-constant value) wouldn't work, but changing bpf_skb_load_bytes() is also not the proper solution, since the two copies are generally still needed. iii) bpf_skb_load_bytes() is just for rather small buffers like headers, since they need to sit on the limited BPF stack anyway. Instead of working around in bpf_skb_load_bytes(), this work improves the bpf_skb_event_output() helper to address all 3 at once. We can make use of the passed in skb context that we have in the helper anyway, and use some of the reserved flag bits as a length argument. The helper will use the new __output_custom() facility from perf side with bpf_skb_copy() as callback helper to walk and extract the data. It will pass the data for setup to bpf_event_output(), which generates and pushes the raw record with an additional frag part. The linear data used in the first frag of the record serves as programmatically defined meta data passed along with the appended sample. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 7 ++++++- include/uapi/linux/bpf.h | 2 ++ kernel/bpf/core.c | 6 ++++-- kernel/trace/bpf_trace.c | 33 +++++++++++++++------------------ net/core/filter.c | 43 ++++++++++++++++++++++++++++++++++++++++++- 5 files changed, 69 insertions(+), 22 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index b3336b4f5d04..c13e92b00bf5 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -209,7 +209,12 @@ u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp); const struct bpf_func_proto *bpf_get_trace_printk_proto(void); -const struct bpf_func_proto *bpf_get_event_output_proto(void); + +typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src, + unsigned long len); + +u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy); #ifdef CONFIG_BPF_SYSCALL DECLARE_PER_CPU(int, bpf_prog_active); diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 262a7e883b19..c4d922439d20 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -401,6 +401,8 @@ enum bpf_func_id { /* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */ #define BPF_F_INDEX_MASK 0xffffffffULL #define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK +/* BPF_FUNC_perf_event_output for sk_buff input context. */ +#define BPF_F_CTXLEN_MASK (0xfffffULL << 32) /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index d638062f66d6..03fd23d4d587 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -1054,9 +1054,11 @@ const struct bpf_func_proto * __weak bpf_get_trace_printk_proto(void) return NULL; } -const struct bpf_func_proto * __weak bpf_get_event_output_proto(void) +u64 __weak +bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { - return NULL; + return -ENOTSUPP; } /* Always built-in helper functions. */ diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index c35883a9bc11..ebfbb7dd7033 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -298,29 +298,26 @@ static const struct bpf_func_proto bpf_perf_event_output_proto = { static DEFINE_PER_CPU(struct pt_regs, bpf_pt_regs); -static u64 bpf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) +u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { struct pt_regs *regs = this_cpu_ptr(&bpf_pt_regs); + struct perf_raw_frag frag = { + .copy = ctx_copy, + .size = ctx_size, + .data = ctx, + }; + struct perf_raw_record raw = { + .frag = { + .next = ctx_size ? &frag : NULL, + .size = meta_size, + .data = meta, + }, + }; perf_fetch_caller_regs(regs); - return bpf_perf_event_output((long)regs, r2, flags, r4, size); -} - -static const struct bpf_func_proto bpf_event_output_proto = { - .func = bpf_event_output, - .gpl_only = true, - .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_CTX, - .arg2_type = ARG_CONST_MAP_PTR, - .arg3_type = ARG_ANYTHING, - .arg4_type = ARG_PTR_TO_STACK, - .arg5_type = ARG_CONST_STACK_SIZE, -}; - -const struct bpf_func_proto *bpf_get_event_output_proto(void) -{ - return &bpf_event_output_proto; + return __bpf_perf_event_output(regs, map, flags, &raw); } static u64 bpf_get_current_task(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) diff --git a/net/core/filter.c b/net/core/filter.c index 10c4a2f9e8bb..22e3992c8b48 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2025,6 +2025,47 @@ bool bpf_helper_changes_skb_data(void *func) return false; } +static unsigned long bpf_skb_copy(void *dst_buff, const void *skb, + unsigned long len) +{ + void *ptr = skb_header_pointer(skb, 0, len, dst_buff); + + if (unlikely(!ptr)) + return len; + if (ptr != dst_buff) + memcpy(dst_buff, ptr, len); + + return 0; +} + +static u64 bpf_skb_event_output(u64 r1, u64 r2, u64 flags, u64 r4, + u64 meta_size) +{ + struct sk_buff *skb = (struct sk_buff *)(long) r1; + struct bpf_map *map = (struct bpf_map *)(long) r2; + u64 skb_size = (flags & BPF_F_CTXLEN_MASK) >> 32; + void *meta = (void *)(long) r4; + + if (unlikely(flags & ~(BPF_F_CTXLEN_MASK | BPF_F_INDEX_MASK))) + return -EINVAL; + if (unlikely(skb_size > skb->len)) + return -EFAULT; + + return bpf_event_output(map, flags, meta, meta_size, skb, skb_size, + bpf_skb_copy); +} + +static const struct bpf_func_proto bpf_skb_event_output_proto = { + .func = bpf_skb_event_output, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_CONST_MAP_PTR, + .arg3_type = ARG_ANYTHING, + .arg4_type = ARG_PTR_TO_STACK, + .arg5_type = ARG_CONST_STACK_SIZE, +}; + static unsigned short bpf_tunnel_key_af(u64 flags) { return flags & BPF_F_TUNINFO_IPV6 ? AF_INET6 : AF_INET; @@ -2357,7 +2398,7 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) case BPF_FUNC_get_hash_recalc: return &bpf_get_hash_recalc_proto; case BPF_FUNC_perf_event_output: - return bpf_get_event_output_proto(); + return &bpf_skb_event_output_proto; case BPF_FUNC_get_smp_processor_id: return &bpf_get_smp_processor_id_proto; #ifdef CONFIG_SOCK_CGROUP_DATA -- cgit v1.2.3 From 7363cee5b467c31dc3af2ac98df0634bb8bbc668 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Sat, 2 Jul 2016 14:14:27 -0700 Subject: drm/vc4: Add a getparam to signal support for branches. Userspace needs to know if it can create shaders that do branching. Otherwise, for backwards compatibility with old kernels it needs to lower if statements to conditional assignments. Signed-off-by: Eric Anholt --- drivers/gpu/drm/vc4/vc4_drv.c | 3 +++ include/uapi/drm/vc4_drm.h | 1 + 2 files changed, 4 insertions(+) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index 047d7a265ceb..9435894822d5 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -76,6 +76,9 @@ static int vc4_get_param_ioctl(struct drm_device *dev, void *data, args->value = V3D_READ(V3D_IDENT2); pm_runtime_put(&vc4->v3d->pdev->dev); break; + case DRM_VC4_PARAM_SUPPORTS_BRANCHES: + args->value = true; + break; default: DRM_DEBUG("Unknown parameter %d\n", args->param); return -EINVAL; diff --git a/include/uapi/drm/vc4_drm.h b/include/uapi/drm/vc4_drm.h index 1143e954048d..ad7edc3edf7c 100644 --- a/include/uapi/drm/vc4_drm.h +++ b/include/uapi/drm/vc4_drm.h @@ -285,6 +285,7 @@ struct drm_vc4_get_hang_state { #define DRM_VC4_PARAM_V3D_IDENT0 0 #define DRM_VC4_PARAM_V3D_IDENT1 1 #define DRM_VC4_PARAM_V3D_IDENT2 2 +#define DRM_VC4_PARAM_SUPPORTS_BRANCHES 3 struct drm_vc4_get_param { __u32 param; -- cgit v1.2.3 From 4cd33c48ea25ba17e9d0383fe914c3e58b48f7dd Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Tue, 17 May 2016 15:44:49 -0400 Subject: drm/msm: add madvise ioctl Doesn't do anything too interesting until we wire up shrinker. Pretty much lifted from i915. Signed-off-by: Rob Clark --- drivers/gpu/drm/msm/msm_drv.c | 39 +++++++++++++++++++++++++++++++++++++++ drivers/gpu/drm/msm/msm_drv.h | 1 + drivers/gpu/drm/msm/msm_gem.c | 35 +++++++++++++++++++++++++++++++++-- drivers/gpu/drm/msm/msm_gem.h | 5 +++++ include/uapi/drm/msm_drm.h | 25 ++++++++++++++++++++++++- 5 files changed, 102 insertions(+), 3 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 435a8f90c290..00881f3ed32e 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -690,6 +690,44 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data, return msm_wait_fence(priv->gpu->fctx, args->fence, &timeout, true); } +static int msm_ioctl_gem_madvise(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct drm_msm_gem_madvise *args = data; + struct drm_gem_object *obj; + int ret; + + switch (args->madv) { + case MSM_MADV_DONTNEED: + case MSM_MADV_WILLNEED: + break; + default: + return -EINVAL; + } + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + + obj = drm_gem_object_lookup(file, args->handle); + if (!obj) { + ret = -ENOENT; + goto unlock; + } + + ret = msm_gem_madvise(obj, args->madv); + if (ret >= 0) { + args->retained = ret; + ret = 0; + } + + drm_gem_object_unreference(obj); + +unlock: + mutex_unlock(&dev->struct_mutex); + return ret; +} + static const struct drm_ioctl_desc msm_ioctls[] = { DRM_IOCTL_DEF_DRV(MSM_GET_PARAM, msm_ioctl_get_param, DRM_AUTH|DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_NEW, msm_ioctl_gem_new, DRM_AUTH|DRM_RENDER_ALLOW), @@ -698,6 +736,7 @@ static const struct drm_ioctl_desc msm_ioctls[] = { DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_FINI, msm_ioctl_gem_cpu_fini, DRM_AUTH|DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_GEM_SUBMIT, msm_ioctl_gem_submit, DRM_AUTH|DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_WAIT_FENCE, msm_ioctl_wait_fence, DRM_AUTH|DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(MSM_GEM_MADVISE, msm_ioctl_gem_madvise, DRM_AUTH|DRM_RENDER_ALLOW), }; static const struct vm_operations_struct vm_ops = { diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index be01e3895178..a49d7fd67ec3 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -195,6 +195,7 @@ int msm_gem_prime_pin(struct drm_gem_object *obj); void msm_gem_prime_unpin(struct drm_gem_object *obj); void *msm_gem_vaddr_locked(struct drm_gem_object *obj); void *msm_gem_vaddr(struct drm_gem_object *obj); +int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv); int msm_gem_sync_object(struct drm_gem_object *obj, struct msm_fence_context *fctx, bool exclusive); void msm_gem_move_to_active(struct drm_gem_object *obj, diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 69836f5685b1..c40db08647d1 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -413,6 +413,21 @@ void *msm_gem_vaddr(struct drm_gem_object *obj) return ret; } +/* Update madvise status, returns true if not purged, else + * false or -errno. + */ +int msm_gem_madvise(struct drm_gem_object *obj, unsigned madv) +{ + struct msm_gem_object *msm_obj = to_msm_bo(obj); + + WARN_ON(!mutex_is_locked(&obj->dev->struct_mutex)); + + if (msm_obj->madv != __MSM_MADV_PURGED) + msm_obj->madv = madv; + + return (msm_obj->madv != __MSM_MADV_PURGED); +} + /* must be called before _move_to_active().. */ int msm_gem_sync_object(struct drm_gem_object *obj, struct msm_fence_context *fctx, bool exclusive) @@ -464,6 +479,7 @@ void msm_gem_move_to_active(struct drm_gem_object *obj, struct msm_gpu *gpu, bool exclusive, struct fence *fence) { struct msm_gem_object *msm_obj = to_msm_bo(obj); + WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED); msm_obj->gpu = gpu; if (exclusive) reservation_object_add_excl_fence(msm_obj->resv, fence); @@ -532,13 +548,27 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m) struct reservation_object_list *fobj; struct fence *fence; uint64_t off = drm_vma_node_start(&obj->vma_node); + const char *madv; WARN_ON(!mutex_is_locked(&obj->dev->struct_mutex)); - seq_printf(m, "%08x: %c %2d (%2d) %08llx %p %zu\n", + switch (msm_obj->madv) { + case __MSM_MADV_PURGED: + madv = " purged"; + break; + case MSM_MADV_DONTNEED: + madv = " purgeable"; + break; + case MSM_MADV_WILLNEED: + default: + madv = ""; + break; + } + + seq_printf(m, "%08x: %c %2d (%2d) %08llx %p %zu%s\n", msm_obj->flags, is_active(msm_obj) ? 'A' : 'I', obj->name, obj->refcount.refcount.counter, - off, msm_obj->vaddr, obj->size); + off, msm_obj->vaddr, obj->size, madv); rcu_read_lock(); fobj = rcu_dereference(robj->fence); @@ -688,6 +718,7 @@ static int msm_gem_new_impl(struct drm_device *dev, msm_obj->vram_node = (void *)&msm_obj[1]; msm_obj->flags = flags; + msm_obj->madv = MSM_MADV_WILLNEED; if (resv) { msm_obj->resv = resv; diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index 9facd4b6ffd9..fa8e1f16f18e 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -29,6 +29,11 @@ struct msm_gem_object { uint32_t flags; + /** + * Advice: are the backing pages purgeable? + */ + uint8_t madv; + /* And object is either: * inactive - on priv->inactive_list * active - on one one of the gpu's active_list.. well, at diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index bf19d2cd9078..49f778de8e06 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -201,6 +201,27 @@ struct drm_msm_wait_fence { struct drm_msm_timespec timeout; /* in */ }; +/* madvise provides a way to tell the kernel in case a buffers contents + * can be discarded under memory pressure, which is useful for userspace + * bo cache where we want to optimistically hold on to buffer allocate + * and potential mmap, but allow the pages to be discarded under memory + * pressure. + * + * Typical usage would involve madvise(DONTNEED) when buffer enters BO + * cache, and madvise(WILLNEED) if trying to recycle buffer from BO cache. + * In the WILLNEED case, 'retained' indicates to userspace whether the + * backing pages still exist. + */ +#define MSM_MADV_WILLNEED 0 /* backing pages are needed, status returned in 'retained' */ +#define MSM_MADV_DONTNEED 1 /* backing pages not needed */ +#define __MSM_MADV_PURGED 2 /* internal state */ + +struct drm_msm_gem_madvise { + __u32 handle; /* in, GEM handle */ + __u32 madv; /* in, MSM_MADV_x */ + __u32 retained; /* out, whether backing store still exists */ +}; + #define DRM_MSM_GET_PARAM 0x00 /* placeholder: #define DRM_MSM_SET_PARAM 0x01 @@ -211,7 +232,8 @@ struct drm_msm_wait_fence { #define DRM_MSM_GEM_CPU_FINI 0x05 #define DRM_MSM_GEM_SUBMIT 0x06 #define DRM_MSM_WAIT_FENCE 0x07 -#define DRM_MSM_NUM_IOCTLS 0x08 +#define DRM_MSM_GEM_MADVISE 0x08 +#define DRM_MSM_NUM_IOCTLS 0x09 #define DRM_IOCTL_MSM_GET_PARAM DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GET_PARAM, struct drm_msm_param) #define DRM_IOCTL_MSM_GEM_NEW DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_NEW, struct drm_msm_gem_new) @@ -220,6 +242,7 @@ struct drm_msm_wait_fence { #define DRM_IOCTL_MSM_GEM_CPU_FINI DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_GEM_CPU_FINI, struct drm_msm_gem_cpu_fini) #define DRM_IOCTL_MSM_GEM_SUBMIT DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_SUBMIT, struct drm_msm_gem_submit) #define DRM_IOCTL_MSM_WAIT_FENCE DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_WAIT_FENCE, struct drm_msm_wait_fence) +#define DRM_IOCTL_MSM_GEM_MADVISE DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_MADVISE, struct drm_msm_gem_madvise) #if defined(__cplusplus) } -- cgit v1.2.3 From 4077798484459a2eced2050045099a466ecb618a Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 15 Jul 2016 09:31:11 +0100 Subject: drm/vgem: Attach sw fences to exported vGEM dma-buf (ioctl) vGEM buffers are useful for passing data between software clients and hardware renders. By allowing the user to create and attach fences to the exported vGEM buffers (on the dma-buf), the user can implement a deferred renderer and queue hardware operations like flipping and then signal the buffer readiness (i.e. this allows the user to schedule operations out-of-order, but have them complete in-order). This also makes it much easier to write tightly controlled testcases for dma-buf fencing and signaling between hardware drivers. v2: Don't pretend the fences exist in an ordered timeline, but allocate a separate fence-context for each fence so that the fences are unordered. v3: Make the debug output more interesting, and show the signaled status. v4: Automatically signal the fence to prevent userspace from indefinitely hanging drivers. Testcase: igt/vgem_basic/dmabuf-fence Testcase: igt/vgem_slow/nohang Signed-off-by: Chris Wilson Cc: Sean Paul Cc: Zach Reizner Cc: Gustavo Padovan Cc: Daniel Vetter Acked-by: Zach Reizner Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/1468571471-12610-1-git-send-email-chris@chris-wilson.co.uk --- drivers/gpu/drm/vgem/Makefile | 2 +- drivers/gpu/drm/vgem/vgem_drv.c | 34 +++++ drivers/gpu/drm/vgem/vgem_drv.h | 16 +++ drivers/gpu/drm/vgem/vgem_fence.c | 283 ++++++++++++++++++++++++++++++++++++++ include/uapi/drm/vgem_drm.h | 62 +++++++++ 5 files changed, 396 insertions(+), 1 deletion(-) create mode 100644 drivers/gpu/drm/vgem/vgem_fence.c create mode 100644 include/uapi/drm/vgem_drm.h (limited to 'include/uapi') diff --git a/drivers/gpu/drm/vgem/Makefile b/drivers/gpu/drm/vgem/Makefile index 3f4c7b842028..bfcdea1330e6 100644 --- a/drivers/gpu/drm/vgem/Makefile +++ b/drivers/gpu/drm/vgem/Makefile @@ -1,4 +1,4 @@ ccflags-y := -Iinclude/drm -vgem-y := vgem_drv.o +vgem-y := vgem_drv.o vgem_fence.o obj-$(CONFIG_DRM_VGEM) += vgem.o diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c index 29c2aab3c1a7..c15bafb06665 100644 --- a/drivers/gpu/drm/vgem/vgem_drv.c +++ b/drivers/gpu/drm/vgem/vgem_drv.c @@ -83,6 +83,34 @@ static const struct vm_operations_struct vgem_gem_vm_ops = { .close = drm_gem_vm_close, }; +static int vgem_open(struct drm_device *dev, struct drm_file *file) +{ + struct vgem_file *vfile; + int ret; + + vfile = kzalloc(sizeof(*vfile), GFP_KERNEL); + if (!vfile) + return -ENOMEM; + + file->driver_priv = vfile; + + ret = vgem_fence_open(vfile); + if (ret) { + kfree(vfile); + return ret; + } + + return 0; +} + +static void vgem_preclose(struct drm_device *dev, struct drm_file *file) +{ + struct vgem_file *vfile = file->driver_priv; + + vgem_fence_close(vfile); + kfree(vfile); +} + /* ioctls */ static struct drm_gem_object *vgem_gem_create(struct drm_device *dev, @@ -164,6 +192,8 @@ unref: } static struct drm_ioctl_desc vgem_ioctls[] = { + DRM_IOCTL_DEF_DRV(VGEM_FENCE_ATTACH, vgem_fence_attach_ioctl, DRM_AUTH|DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(VGEM_FENCE_SIGNAL, vgem_fence_signal_ioctl, DRM_AUTH|DRM_RENDER_ALLOW), }; static int vgem_mmap(struct file *filp, struct vm_area_struct *vma) @@ -271,9 +301,12 @@ static int vgem_prime_mmap(struct drm_gem_object *obj, static struct drm_driver vgem_driver = { .driver_features = DRIVER_GEM | DRIVER_PRIME, + .open = vgem_open, + .preclose = vgem_preclose, .gem_free_object_unlocked = vgem_gem_free_object, .gem_vm_ops = &vgem_gem_vm_ops, .ioctls = vgem_ioctls, + .num_ioctls = ARRAY_SIZE(vgem_ioctls), .fops = &vgem_driver_fops, .dumb_create = vgem_gem_dumb_create, @@ -328,5 +361,6 @@ module_init(vgem_init); module_exit(vgem_exit); MODULE_AUTHOR("Red Hat, Inc."); +MODULE_AUTHOR("Intel Corporation"); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_LICENSE("GPL and additional rights"); diff --git a/drivers/gpu/drm/vgem/vgem_drv.h b/drivers/gpu/drm/vgem/vgem_drv.h index 988cbaae7588..1f8798ad329c 100644 --- a/drivers/gpu/drm/vgem/vgem_drv.h +++ b/drivers/gpu/drm/vgem/vgem_drv.h @@ -32,9 +32,25 @@ #include #include +#include + +struct vgem_file { + struct idr fence_idr; + struct mutex fence_mutex; +}; + #define to_vgem_bo(x) container_of(x, struct drm_vgem_gem_object, base) struct drm_vgem_gem_object { struct drm_gem_object base; }; +int vgem_fence_open(struct vgem_file *file); +int vgem_fence_attach_ioctl(struct drm_device *dev, + void *data, + struct drm_file *file); +int vgem_fence_signal_ioctl(struct drm_device *dev, + void *data, + struct drm_file *file); +void vgem_fence_close(struct vgem_file *file); + #endif diff --git a/drivers/gpu/drm/vgem/vgem_fence.c b/drivers/gpu/drm/vgem/vgem_fence.c new file mode 100644 index 000000000000..e77b52208699 --- /dev/null +++ b/drivers/gpu/drm/vgem/vgem_fence.c @@ -0,0 +1,283 @@ +/* + * Copyright 2016 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software") + * to deal in the software without restriction, including without limitation + * on the rights to use, copy, modify, merge, publish, distribute, sub + * license, and/or sell copies of the Software, and to permit persons to whom + * them Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTIBILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES, OR OTHER LIABILITY, WHETHER + * IN AN ACTION OF CONTRACT, TORT, OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ + +#include +#include + +#include "vgem_drv.h" + +#define VGEM_FENCE_TIMEOUT (10*HZ) + +struct vgem_fence { + struct fence base; + struct spinlock lock; + struct timer_list timer; +}; + +static const char *vgem_fence_get_driver_name(struct fence *fence) +{ + return "vgem"; +} + +static const char *vgem_fence_get_timeline_name(struct fence *fence) +{ + return "unbound"; +} + +static bool vgem_fence_signaled(struct fence *fence) +{ + return false; +} + +static bool vgem_fence_enable_signaling(struct fence *fence) +{ + return true; +} + +static void vgem_fence_release(struct fence *base) +{ + struct vgem_fence *fence = container_of(base, typeof(*fence), base); + + del_timer_sync(&fence->timer); + fence_free(&fence->base); +} + +static void vgem_fence_value_str(struct fence *fence, char *str, int size) +{ + snprintf(str, size, "%u", fence->seqno); +} + +static void vgem_fence_timeline_value_str(struct fence *fence, char *str, + int size) +{ + snprintf(str, size, "%u", fence_is_signaled(fence) ? fence->seqno : 0); +} + +const struct fence_ops vgem_fence_ops = { + .get_driver_name = vgem_fence_get_driver_name, + .get_timeline_name = vgem_fence_get_timeline_name, + .enable_signaling = vgem_fence_enable_signaling, + .signaled = vgem_fence_signaled, + .wait = fence_default_wait, + .release = vgem_fence_release, + + .fence_value_str = vgem_fence_value_str, + .timeline_value_str = vgem_fence_timeline_value_str, +}; + +static void vgem_fence_timeout(unsigned long data) +{ + struct vgem_fence *fence = (struct vgem_fence *)data; + + fence_signal(&fence->base); +} + +static struct fence *vgem_fence_create(struct vgem_file *vfile, + unsigned int flags) +{ + struct vgem_fence *fence; + + fence = kzalloc(sizeof(*fence), GFP_KERNEL); + if (!fence) + return NULL; + + spin_lock_init(&fence->lock); + fence_init(&fence->base, &vgem_fence_ops, &fence->lock, + fence_context_alloc(1), 1); + + setup_timer(&fence->timer, vgem_fence_timeout, (unsigned long)fence); + + /* We force the fence to expire within 10s to prevent driver hangs */ + mod_timer(&fence->timer, VGEM_FENCE_TIMEOUT); + + return &fence->base; +} + +static int attach_dmabuf(struct drm_device *dev, + struct drm_gem_object *obj) +{ + struct dma_buf *dmabuf; + + if (obj->dma_buf) + return 0; + + dmabuf = dev->driver->gem_prime_export(dev, obj, 0); + if (IS_ERR(dmabuf)) + return PTR_ERR(dmabuf); + + obj->dma_buf = dmabuf; + drm_gem_object_reference(obj); + return 0; +} + +/* + * vgem_fence_attach_ioctl (DRM_IOCTL_VGEM_FENCE_ATTACH): + * + * Create and attach a fence to the vGEM handle. This fence is then exposed + * via the dma-buf reservation object and visible to consumers of the exported + * dma-buf. If the flags contain VGEM_FENCE_WRITE, the fence indicates the + * vGEM buffer is being written to by the client and is exposed as an exclusive + * fence, otherwise the fence indicates the client is current reading from the + * buffer and all future writes should wait for the client to signal its + * completion. Note that if a conflicting fence is already on the dma-buf (i.e. + * an exclusive fence when adding a read, or any fence when adding a write), + * -EBUSY is reported. Serialisation between operations should be handled + * by waiting upon the dma-buf. + * + * This returns the handle for the new fence that must be signaled within 10 + * seconds (or otherwise it will automatically expire). See + * vgem_fence_signal_ioctl (DRM_IOCTL_VGEM_FENCE_SIGNAL). + * + * If the vGEM handle does not exist, vgem_fence_attach_ioctl returns -ENOENT. + */ +int vgem_fence_attach_ioctl(struct drm_device *dev, + void *data, + struct drm_file *file) +{ + struct drm_vgem_fence_attach *arg = data; + struct vgem_file *vfile = file->driver_priv; + struct reservation_object *resv; + struct drm_gem_object *obj; + struct fence *fence; + int ret; + + if (arg->flags & ~VGEM_FENCE_WRITE) + return -EINVAL; + + if (arg->pad) + return -EINVAL; + + obj = drm_gem_object_lookup(file, arg->handle); + if (!obj) + return -ENOENT; + + ret = attach_dmabuf(dev, obj); + if (ret) + goto err; + + fence = vgem_fence_create(vfile, arg->flags); + if (!fence) { + ret = -ENOMEM; + goto err; + } + + /* Check for a conflicting fence */ + resv = obj->dma_buf->resv; + if (!reservation_object_test_signaled_rcu(resv, + arg->flags & VGEM_FENCE_WRITE)) { + ret = -EBUSY; + goto err_fence; + } + + /* Expose the fence via the dma-buf */ + ret = 0; + mutex_lock(&resv->lock.base); + if (arg->flags & VGEM_FENCE_WRITE) + reservation_object_add_excl_fence(resv, fence); + else if ((ret = reservation_object_reserve_shared(resv)) == 0) + reservation_object_add_shared_fence(resv, fence); + mutex_unlock(&resv->lock.base); + + /* Record the fence in our idr for later signaling */ + if (ret == 0) { + mutex_lock(&vfile->fence_mutex); + ret = idr_alloc(&vfile->fence_idr, fence, 1, 0, GFP_KERNEL); + mutex_unlock(&vfile->fence_mutex); + if (ret > 0) { + arg->out_fence = ret; + ret = 0; + } + } +err_fence: + if (ret) { + fence_signal(fence); + fence_put(fence); + } +err: + drm_gem_object_unreference_unlocked(obj); + return ret; +} + +/* + * vgem_fence_signal_ioctl (DRM_IOCTL_VGEM_FENCE_SIGNAL): + * + * Signal and consume a fence ealier attached to a vGEM handle using + * vgem_fence_attach_ioctl (DRM_IOCTL_VGEM_FENCE_ATTACH). + * + * All fences must be signaled within 10s of attachment or otherwise they + * will automatically expire (and a vgem_fence_signal_ioctl returns -ETIMEDOUT). + * + * Signaling a fence indicates to all consumers of the dma-buf that the + * client has completed the operation associated with the fence, and that the + * buffer is then ready for consumption. + * + * If the fence does not exist (or has already been signaled by the client), + * vgem_fence_signal_ioctl returns -ENOENT. + */ +int vgem_fence_signal_ioctl(struct drm_device *dev, + void *data, + struct drm_file *file) +{ + struct vgem_file *vfile = file->driver_priv; + struct drm_vgem_fence_signal *arg = data; + struct fence *fence; + int ret; + + if (arg->flags) + return -EINVAL; + + mutex_lock(&vfile->fence_mutex); + fence = idr_replace(&vfile->fence_idr, NULL, arg->fence); + mutex_unlock(&vfile->fence_mutex); + if (!fence) + return -ENOENT; + if (IS_ERR(fence)) + return PTR_ERR(fence); + + if (fence_is_signaled(fence)) + ret = -ETIMEDOUT; + + fence_signal(fence); + fence_put(fence); + return ret; +} + +int vgem_fence_open(struct vgem_file *vfile) +{ + mutex_init(&vfile->fence_mutex); + idr_init(&vfile->fence_idr); + + return 0; +} + +static int __vgem_fence_idr_fini(int id, void *p, void *data) +{ + fence_signal(p); + fence_put(p); + return 0; +} + +void vgem_fence_close(struct vgem_file *vfile) +{ + idr_for_each(&vfile->fence_idr, __vgem_fence_idr_fini, vfile); + idr_destroy(&vfile->fence_idr); +} diff --git a/include/uapi/drm/vgem_drm.h b/include/uapi/drm/vgem_drm.h new file mode 100644 index 000000000000..bf66f5db6da8 --- /dev/null +++ b/include/uapi/drm/vgem_drm.h @@ -0,0 +1,62 @@ +/* + * Copyright 2016 Intel Corporation + * All Rights Reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. + * IN NO EVENT SHALL TUNGSTEN GRAPHICS AND/OR ITS SUPPLIERS BE LIABLE FOR + * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, + * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE + * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _UAPI_VGEM_DRM_H_ +#define _UAPI_VGEM_DRM_H_ + +#include "drm.h" + +#if defined(__cplusplus) +extern "C" { +#endif + +/* Please note that modifications to all structs defined here are + * subject to backwards-compatibility constraints. + */ +#define DRM_VGEM_FENCE_ATTACH 0x1 +#define DRM_VGEM_FENCE_SIGNAL 0x2 + +#define DRM_IOCTL_VGEM_FENCE_ATTACH DRM_IOWR( DRM_COMMAND_BASE + DRM_VGEM_FENCE_ATTACH, struct drm_vgem_fence_attach) +#define DRM_IOCTL_VGEM_FENCE_SIGNAL DRM_IOW( DRM_COMMAND_BASE + DRM_VGEM_FENCE_SIGNAL, struct drm_vgem_fence_signal) + +struct drm_vgem_fence_attach { + __u32 handle; + __u32 flags; +#define VGEM_FENCE_WRITE 0x1 + __u32 out_fence; + __u32 pad; +}; + +struct drm_vgem_fence_signal { + __u32 fence; + __u32 flags; +}; + +#if defined(__cplusplus) +} +#endif + +#endif /* _UAPI_VGEM_DRM_H_ */ -- cgit v1.2.3 From 6502a34cfd6695929086187f63fe670cc3050e68 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 21 Jun 2016 14:19:51 +0200 Subject: KVM: s390: allow user space to handle instr 0x0000 We will use illegal instruction 0x0000 for handling 2 byte sw breakpoints from user space. As it can be enabled dynamically via a capability, let's move setting of ICTL_OPEREXC to the post creation step, so we avoid any races when enabling that capability just while adding new cpus. Acked-by: Janosch Frank Reviewed-by: Cornelia Huck Signed-off-by: David Hildenbrand Signed-off-by: Christian Borntraeger --- Documentation/virtual/kvm/api.txt | 13 +++++++++++++ arch/s390/include/asm/kvm_host.h | 2 ++ arch/s390/kvm/intercept.c | 3 +++ arch/s390/kvm/kvm-s390.c | 26 ++++++++++++++++++++++++-- include/uapi/linux/kvm.h | 1 + 5 files changed, 43 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index c4d2fb0e28de..299306db5d84 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -3857,6 +3857,19 @@ as a broadcast even in x2APIC mode in order to support physical x2APIC without interrupt remapping. This is undesirable in logical mode, where 0xff represents CPUs 0-7 in cluster 0. +7.8 KVM_CAP_S390_USER_INSTR0 + +Architectures: s390 +Parameters: none + +With this capability enabled, all illegal instructions 0x0000 (2 bytes) will +be intercepted and forwarded to user space. User space can use this +mechanism e.g. to realize 2-byte software breakpoints. The kernel will +not inject an operating exception for these instructions, user space has +to take care of that. + +This capability can be enabled dynamically even if VCPUs were already +created and are running. 8. Other capabilities. ---------------------- diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 946fc86202fd..183b01727de4 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -43,6 +43,7 @@ /* s390-specific vcpu->requests bit members */ #define KVM_REQ_ENABLE_IBS 8 #define KVM_REQ_DISABLE_IBS 9 +#define KVM_REQ_ICPT_OPEREXC 10 #define SIGP_CTRL_C 0x80 #define SIGP_CTRL_SCN_MASK 0x3f @@ -666,6 +667,7 @@ struct kvm_arch{ int user_cpu_state_ctrl; int user_sigp; int user_stsi; + int user_instr0; struct s390_io_adapter *adapters[MAX_S390_IO_ADAPTERS]; wait_queue_head_t ipte_wq; int ipte_lock_count; diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 850be47c4cc9..7a2f1551bc39 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -359,6 +359,9 @@ static int handle_operexc(struct kvm_vcpu *vcpu) test_kvm_facility(vcpu->kvm, 74)) return handle_sthyi(vcpu); + if (vcpu->arch.sie_block->ipa == 0 && vcpu->kvm->arch.user_instr0) + return -EOPNOTSUPP; + return kvm_s390_inject_program_int(vcpu, PGM_OPERATION); } diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index d42428c11794..63ac7c1641a7 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -364,6 +364,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_S390_USER_STSI: case KVM_CAP_S390_SKEYS: case KVM_CAP_S390_IRQ_STATE: + case KVM_CAP_S390_USER_INSTR0: r = 1; break; case KVM_CAP_S390_MEM_OP: @@ -456,6 +457,16 @@ out: return r; } +static void icpt_operexc_on_all_vcpus(struct kvm *kvm) +{ + unsigned int i; + struct kvm_vcpu *vcpu; + + kvm_for_each_vcpu(i, vcpu, kvm) { + kvm_s390_sync_request(KVM_REQ_ICPT_OPEREXC, vcpu); + } +} + static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) { int r; @@ -507,6 +518,12 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) kvm->arch.user_stsi = 1; r = 0; break; + case KVM_CAP_S390_USER_INSTR0: + VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_INSTR0"); + kvm->arch.user_instr0 = 1; + icpt_operexc_on_all_vcpus(kvm); + r = 0; + break; default: r = -EINVAL; break; @@ -1836,6 +1853,8 @@ void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu) vcpu->arch.gmap = vcpu->kvm->arch.gmap; sca_add_vcpu(vcpu); } + if (test_kvm_facility(vcpu->kvm, 74) || vcpu->kvm->arch.user_instr0) + vcpu->arch.sie_block->ictl |= ICTL_OPEREXC; /* make vcpu_load load the right gmap on the first trigger */ vcpu->arch.enabled_gmap = vcpu->arch.gmap; } @@ -1923,8 +1942,6 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) } vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb; vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE; - if (test_kvm_facility(vcpu->kvm, 74)) - vcpu->arch.sie_block->ictl |= ICTL_OPEREXC; if (vcpu->kvm->arch.use_cmma) { rc = kvm_s390_vcpu_setup_cmma(vcpu); @@ -2369,6 +2386,11 @@ retry: goto retry; } + if (kvm_check_request(KVM_REQ_ICPT_OPEREXC, vcpu)) { + vcpu->arch.sie_block->ictl |= ICTL_OPEREXC; + goto retry; + } + /* nothing to do, just clear the request */ clear_bit(KVM_REQ_UNHALT, &vcpu->requests); diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 4f8030e5b05d..70941f4ab6d8 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -867,6 +867,7 @@ struct kvm_ppc_smmu_info { #define KVM_CAP_VCPU_ATTRIBUTES 127 #define KVM_CAP_MAX_VCPU_ID 128 #define KVM_CAP_X2APIC_API 129 +#define KVM_CAP_S390_USER_INSTR0 130 #ifdef KVM_CAP_IRQ_ROUTING -- cgit v1.2.3 From 2b8ddd9337ee0d001b22507f95596648a1a90992 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Fri, 15 Jul 2016 12:43:24 +0100 Subject: KVM: Extend struct kvm_msi to hold a 32-bit device ID The ARM GICv3 ITS MSI controller requires a device ID to be able to assign the proper interrupt vector. On real hardware, this ID is sampled from the bus. To be able to emulate an ITS controller, extend the KVM MSI interface to let userspace provide such a device ID. For PCI devices, the device ID is simply the 16-bit bus-device-function triplet, which should be easily available to the userland tool. Also there is a new KVM capability which advertises whether the current VM requires a device ID to be set along with the MSI data. This flag is still reported as not available everywhere, later we will enable it when ITS emulation is used. Signed-off-by: Andre Przywara Reviewed-by: Eric Auger Reviewed-by: Marc Zyngier Acked-by: Christoffer Dall Acked-by: Paolo Bonzini Tested-by: Eric Auger Signed-off-by: Marc Zyngier --- Documentation/virtual/kvm/api.txt | 12 ++++++++++-- include/uapi/linux/kvm.h | 5 ++++- 2 files changed, 14 insertions(+), 3 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 09efa9eb3926..65513119fee8 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -2175,10 +2175,18 @@ struct kvm_msi { __u32 address_hi; __u32 data; __u32 flags; - __u8 pad[16]; + __u32 devid; + __u8 pad[12]; }; -No flags are defined so far. The corresponding field must be 0. +flags: KVM_MSI_VALID_DEVID: devid contains a valid value +devid: If KVM_MSI_VALID_DEVID is set, contains a unique device identifier + for the device that wrote the MSI message. + For PCI, this is usually a BFD identifier in the lower 16 bits. + +The per-VM KVM_CAP_MSI_DEVID capability advertises the need to provide +the device ID. If this capability is not set, userland cannot rely on +the kernel to allow the KVM_MSI_VALID_DEVID flag being set. 4.71 KVM_CREATE_PIT2 diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 05ebf475104c..7de96f5bb92c 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -866,6 +866,7 @@ struct kvm_ppc_smmu_info { #define KVM_CAP_ARM_PMU_V3 126 #define KVM_CAP_VCPU_ATTRIBUTES 127 #define KVM_CAP_MAX_VCPU_ID 128 +#define KVM_CAP_MSI_DEVID 129 #ifdef KVM_CAP_IRQ_ROUTING @@ -1024,12 +1025,14 @@ struct kvm_one_reg { __u64 addr; }; +#define KVM_MSI_VALID_DEVID (1U << 0) struct kvm_msi { __u32 address_lo; __u32 address_hi; __u32 data; __u32 flags; - __u8 pad[16]; + __u32 devid; + __u8 pad[12]; }; struct kvm_arm_device_addr { -- cgit v1.2.3 From 1085fdc68c6097244627a02a56bd2d8fe58a1a9c Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Fri, 15 Jul 2016 12:43:31 +0100 Subject: KVM: arm64: vgic-its: Introduce new KVM ITS device Introduce a new KVM device that represents an ARM Interrupt Translation Service (ITS) controller. Since there can be multiple of this per guest, we can't piggy back on the existing GICv3 distributor device, but create a new type of KVM device. On the KVM_CREATE_DEVICE ioctl we allocate and initialize the ITS data structure and store the pointer in the kvm_device data. Upon an explicit init ioctl from userland (after having setup the MMIO address) we register the handlers with the kvm_io_bus framework. Any reference to an ITS thus has to go via this interface. Signed-off-by: Andre Przywara Reviewed-by: Marc Zyngier Tested-by: Eric Auger Signed-off-by: Marc Zyngier --- Documentation/virtual/kvm/devices/arm-vgic.txt | 25 +++-- arch/arm/kvm/arm.c | 1 + arch/arm64/include/uapi/asm/kvm.h | 2 + include/kvm/arm_vgic.h | 3 + include/uapi/linux/kvm.h | 2 + virt/kvm/arm/vgic/vgic-its.c | 135 +++++++++++++++++++++++++ virt/kvm/arm/vgic/vgic-kvm-device.c | 4 +- virt/kvm/arm/vgic/vgic-mmio-v3.c | 2 +- virt/kvm/arm/vgic/vgic.h | 3 + 9 files changed, 168 insertions(+), 9 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/virtual/kvm/devices/arm-vgic.txt b/Documentation/virtual/kvm/devices/arm-vgic.txt index 59541d49e15c..89182f80cc7f 100644 --- a/Documentation/virtual/kvm/devices/arm-vgic.txt +++ b/Documentation/virtual/kvm/devices/arm-vgic.txt @@ -4,16 +4,22 @@ ARM Virtual Generic Interrupt Controller (VGIC) Device types supported: KVM_DEV_TYPE_ARM_VGIC_V2 ARM Generic Interrupt Controller v2.0 KVM_DEV_TYPE_ARM_VGIC_V3 ARM Generic Interrupt Controller v3.0 + KVM_DEV_TYPE_ARM_VGIC_ITS ARM Interrupt Translation Service Controller -Only one VGIC instance may be instantiated through either this API or the -legacy KVM_CREATE_IRQCHIP api. The created VGIC will act as the VM interrupt -controller, requiring emulated user-space devices to inject interrupts to the -VGIC instead of directly to CPUs. +Only one VGIC instance of the V2/V3 types above may be instantiated through +either this API or the legacy KVM_CREATE_IRQCHIP api. The created VGIC will +act as the VM interrupt controller, requiring emulated user-space devices to +inject interrupts to the VGIC instead of directly to CPUs. Creating a guest GICv3 device requires a host GICv3 as well. GICv3 implementations with hardware compatibility support allow a guest GICv2 as well. +Creating a virtual ITS controller requires a host GICv3 (but does not depend +on having physical ITS controllers). +There can be multiple ITS controllers per guest, each of them has to have +a separate, non-overlapping MMIO region. + Groups: KVM_DEV_ARM_VGIC_GRP_ADDR Attributes: @@ -39,6 +45,13 @@ Groups: Only valid for KVM_DEV_TYPE_ARM_VGIC_V3. This address needs to be 64K aligned. + KVM_VGIC_V3_ADDR_TYPE_ITS (rw, 64-bit) + Base address in the guest physical address space of the GICv3 ITS + control register frame. The ITS allows MSI(-X) interrupts to be + injected into guests. This extension is optional. If the kernel + does not support the ITS, the call returns -ENODEV. + Only valid for KVM_DEV_TYPE_ARM_VGIC_ITS. + This address needs to be 64K aligned and the region covers 128K. KVM_DEV_ARM_VGIC_GRP_DIST_REGS Attributes: @@ -109,8 +122,8 @@ Groups: KVM_DEV_ARM_VGIC_GRP_CTRL Attributes: KVM_DEV_ARM_VGIC_CTRL_INIT - request the initialization of the VGIC, no additional parameter in - kvm_device_attr.addr. + request the initialization of the VGIC or ITS, no additional parameter + in kvm_device_attr.addr. Errors: -ENXIO: VGIC not properly configured as required prior to calling this attribute diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 972075cc111c..fb4661cf896e 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/arm64/include/uapi/asm/kvm.h b/arch/arm64/include/uapi/asm/kvm.h index f209ea151dca..3051f86a9b5f 100644 --- a/arch/arm64/include/uapi/asm/kvm.h +++ b/arch/arm64/include/uapi/asm/kvm.h @@ -87,9 +87,11 @@ struct kvm_regs { /* Supported VGICv3 address types */ #define KVM_VGIC_V3_ADDR_TYPE_DIST 2 #define KVM_VGIC_V3_ADDR_TYPE_REDIST 3 +#define KVM_VGIC_ITS_ADDR_TYPE 4 #define KVM_VGIC_V3_DIST_SIZE SZ_64K #define KVM_VGIC_V3_REDIST_SIZE (2 * SZ_64K) +#define KVM_VGIC_V3_ITS_SIZE (2 * SZ_64K) #define KVM_ARM_VCPU_POWER_OFF 0 /* CPU is started in OFF state */ #define KVM_ARM_VCPU_EL1_32BIT 1 /* CPU running a 32bit VM */ diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 685f33975ce4..8609faced83e 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -134,6 +134,7 @@ struct vgic_its { gpa_t vgic_its_base; bool enabled; + bool initialized; struct vgic_io_device iodev; }; @@ -167,6 +168,8 @@ struct vgic_dist { struct vgic_io_device dist_iodev; + bool has_its; + /* * Contains the attributes and gpa of the LPI configuration table. * Since we report GICR_TYPER.CommonLPIAff as 0b00, we can share diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 7de96f5bb92c..d8c4c324cfae 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -1077,6 +1077,8 @@ enum kvm_device_type { #define KVM_DEV_TYPE_FLIC KVM_DEV_TYPE_FLIC KVM_DEV_TYPE_ARM_VGIC_V3, #define KVM_DEV_TYPE_ARM_VGIC_V3 KVM_DEV_TYPE_ARM_VGIC_V3 + KVM_DEV_TYPE_ARM_VGIC_ITS, +#define KVM_DEV_TYPE_ARM_VGIC_ITS KVM_DEV_TYPE_ARM_VGIC_ITS KVM_DEV_TYPE_MAX, }; diff --git a/virt/kvm/arm/vgic/vgic-its.c b/virt/kvm/arm/vgic/vgic-its.c index 4654d6edf6a6..6b47b3674690 100644 --- a/virt/kvm/arm/vgic/vgic-its.c +++ b/virt/kvm/arm/vgic/vgic-its.c @@ -21,6 +21,7 @@ #include #include #include +#include #include @@ -84,6 +85,9 @@ static int vgic_its_init_its(struct kvm *kvm, struct vgic_its *its) struct vgic_io_device *iodev = &its->iodev; int ret; + if (its->initialized) + return 0; + if (IS_VGIC_ADDR_UNDEF(its->vgic_its_base)) return -ENXIO; @@ -99,5 +103,136 @@ static int vgic_its_init_its(struct kvm *kvm, struct vgic_its *its) KVM_VGIC_V3_ITS_SIZE, &iodev->dev); mutex_unlock(&kvm->slots_lock); + if (!ret) + its->initialized = true; + return ret; } + +static int vgic_its_create(struct kvm_device *dev, u32 type) +{ + struct vgic_its *its; + + if (type != KVM_DEV_TYPE_ARM_VGIC_ITS) + return -ENODEV; + + its = kzalloc(sizeof(struct vgic_its), GFP_KERNEL); + if (!its) + return -ENOMEM; + + its->vgic_its_base = VGIC_ADDR_UNDEF; + + dev->kvm->arch.vgic.has_its = true; + its->initialized = false; + its->enabled = false; + + dev->private = its; + + return 0; +} + +static void vgic_its_destroy(struct kvm_device *kvm_dev) +{ + struct vgic_its *its = kvm_dev->private; + + kfree(its); +} + +static int vgic_its_has_attr(struct kvm_device *dev, + struct kvm_device_attr *attr) +{ + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_ADDR: + switch (attr->attr) { + case KVM_VGIC_ITS_ADDR_TYPE: + return 0; + } + break; + case KVM_DEV_ARM_VGIC_GRP_CTRL: + switch (attr->attr) { + case KVM_DEV_ARM_VGIC_CTRL_INIT: + return 0; + } + break; + } + return -ENXIO; +} + +static int vgic_its_set_attr(struct kvm_device *dev, + struct kvm_device_attr *attr) +{ + struct vgic_its *its = dev->private; + int ret; + + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_ADDR: { + u64 __user *uaddr = (u64 __user *)(long)attr->addr; + unsigned long type = (unsigned long)attr->attr; + u64 addr; + + if (type != KVM_VGIC_ITS_ADDR_TYPE) + return -ENODEV; + + if (its->initialized) + return -EBUSY; + + if (copy_from_user(&addr, uaddr, sizeof(addr))) + return -EFAULT; + + ret = vgic_check_ioaddr(dev->kvm, &its->vgic_its_base, + addr, SZ_64K); + if (ret) + return ret; + + its->vgic_its_base = addr; + + return 0; + } + case KVM_DEV_ARM_VGIC_GRP_CTRL: + switch (attr->attr) { + case KVM_DEV_ARM_VGIC_CTRL_INIT: + return vgic_its_init_its(dev->kvm, its); + } + break; + } + return -ENXIO; +} + +static int vgic_its_get_attr(struct kvm_device *dev, + struct kvm_device_attr *attr) +{ + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_ADDR: { + struct vgic_its *its = dev->private; + u64 addr = its->vgic_its_base; + u64 __user *uaddr = (u64 __user *)(long)attr->addr; + unsigned long type = (unsigned long)attr->attr; + + if (type != KVM_VGIC_ITS_ADDR_TYPE) + return -ENODEV; + + if (copy_to_user(uaddr, &addr, sizeof(addr))) + return -EFAULT; + break; + default: + return -ENXIO; + } + } + + return 0; +} + +static struct kvm_device_ops kvm_arm_vgic_its_ops = { + .name = "kvm-arm-vgic-its", + .create = vgic_its_create, + .destroy = vgic_its_destroy, + .set_attr = vgic_its_set_attr, + .get_attr = vgic_its_get_attr, + .has_attr = vgic_its_has_attr, +}; + +int kvm_vgic_register_its_device(void) +{ + return kvm_register_device_ops(&kvm_arm_vgic_its_ops, + KVM_DEV_TYPE_ARM_VGIC_ITS); +} diff --git a/virt/kvm/arm/vgic/vgic-kvm-device.c b/virt/kvm/arm/vgic/vgic-kvm-device.c index 2f24f13c6c90..561d2ba96a4f 100644 --- a/virt/kvm/arm/vgic/vgic-kvm-device.c +++ b/virt/kvm/arm/vgic/vgic-kvm-device.c @@ -21,8 +21,8 @@ /* common helpers */ -static int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr, - phys_addr_t addr, phys_addr_t alignment) +int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr, + phys_addr_t addr, phys_addr_t alignment) { if (addr & ~KVM_PHYS_MASK) return -E2BIG; diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c index b92b7d6cabe6..a5c35050c786 100644 --- a/virt/kvm/arm/vgic/vgic-mmio-v3.c +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c @@ -49,7 +49,7 @@ bool vgic_has_its(struct kvm *kvm) if (dist->vgic_model != KVM_DEV_TYPE_ARM_VGIC_V3) return false; - return false; + return dist->has_its; } static unsigned long vgic_mmio_read_v3_misc(struct kvm_vcpu *vcpu, diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h index 31807c166d2a..8192a293f119 100644 --- a/virt/kvm/arm/vgic/vgic.h +++ b/virt/kvm/arm/vgic/vgic.h @@ -42,6 +42,9 @@ void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq); bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq); void vgic_kick_vcpus(struct kvm *kvm); +int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr, + phys_addr_t addr, phys_addr_t alignment); + void vgic_v2_process_maintenance(struct kvm_vcpu *vcpu); void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu); void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr); -- cgit v1.2.3 From 6389eaa7fa9c3ee6c7d39f6087b86660d17236ac Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:17 +1000 Subject: net/ncsi: NCSI command packet handler The NCSI command packets are sent from MC (Management Controller) to remote end. They are used for multiple purposes: probe existing NCSI package/channel, retrieve NCSI channel's capability, configure NCSI channel etc. This defines struct to represent NCSI command packets and introduces function ncsi_xmit_cmd(), which will be used to transmit NCSI command packet according to the request. The request is represented by struct ncsi_cmd_arg. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller --- include/uapi/linux/if_ether.h | 1 + net/ncsi/Makefile | 2 +- net/ncsi/internal.h | 19 +++ net/ncsi/ncsi-cmd.c | 367 ++++++++++++++++++++++++++++++++++++++++++ net/ncsi/ncsi-pkt.h | 171 ++++++++++++++++++++ 5 files changed, 559 insertions(+), 1 deletion(-) create mode 100644 net/ncsi/ncsi-cmd.c create mode 100644 net/ncsi/ncsi-pkt.h (limited to 'include/uapi') diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h index cec849a239f6..117d02e0fc31 100644 --- a/include/uapi/linux/if_ether.h +++ b/include/uapi/linux/if_ether.h @@ -87,6 +87,7 @@ #define ETH_P_8021AH 0x88E7 /* 802.1ah Backbone Service Tag */ #define ETH_P_MVRP 0x88F5 /* 802.1Q MVRP */ #define ETH_P_1588 0x88F7 /* IEEE 1588 Timesync */ +#define ETH_P_NCSI 0x88F8 /* NCSI protocol */ #define ETH_P_PRP 0x88FB /* IEC 62439-3 PRP/HSRv0 */ #define ETH_P_FCOE 0x8906 /* Fibre Channel over Ethernet */ #define ETH_P_TDLS 0x890D /* TDLS */ diff --git a/net/ncsi/Makefile b/net/ncsi/Makefile index 07b5625155d7..abc404631ca8 100644 --- a/net/ncsi/Makefile +++ b/net/ncsi/Makefile @@ -1,4 +1,4 @@ # # Makefile for NCSI API # -obj-$(CONFIG_NET_NCSI) += ncsi-manage.o +obj-$(CONFIG_NET_NCSI) += ncsi-cmd.o ncsi-manage.o diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index 89028e1f83cd..3d81697a97d0 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -220,6 +220,21 @@ struct ncsi_dev_priv { struct list_head node; /* Form NCSI device list */ }; +struct ncsi_cmd_arg { + struct ncsi_dev_priv *ndp; /* Associated NCSI device */ + unsigned char type; /* Command in the NCSI packet */ + unsigned char id; /* Request ID (sequence number) */ + unsigned char package; /* Destination package ID */ + unsigned char channel; /* Detination channel ID or 0x1f */ + unsigned short payload; /* Command packet payload length */ + bool driven; /* Drive the state machine? */ + union { + unsigned char bytes[16]; /* Command packet specific data */ + unsigned short words[8]; + unsigned int dwords[4]; + }; +}; + extern struct list_head ncsi_dev_list; extern spinlock_t ncsi_dev_lock; @@ -253,4 +268,8 @@ struct ncsi_request *ncsi_alloc_request(struct ncsi_dev_priv *ndp, bool driven); void ncsi_free_request(struct ncsi_request *nr); struct ncsi_dev *ncsi_find_dev(struct net_device *dev); +/* Packet handlers */ +u32 ncsi_calculate_checksum(unsigned char *data, int len); +int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca); + #endif /* __NCSI_INTERNAL_H__ */ diff --git a/net/ncsi/ncsi-cmd.c b/net/ncsi/ncsi-cmd.c new file mode 100644 index 000000000000..21057a8ceeac --- /dev/null +++ b/net/ncsi/ncsi-cmd.c @@ -0,0 +1,367 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "internal.h" +#include "ncsi-pkt.h" + +u32 ncsi_calculate_checksum(unsigned char *data, int len) +{ + u32 checksum = 0; + int i; + + for (i = 0; i < len; i += 2) + checksum += (((u32)data[i] << 8) | data[i + 1]); + + checksum = (~checksum + 1); + return checksum; +} + +/* This function should be called after the data area has been + * populated completely. + */ +static void ncsi_cmd_build_header(struct ncsi_pkt_hdr *h, + struct ncsi_cmd_arg *nca) +{ + u32 checksum; + __be32 *pchecksum; + + h->mc_id = 0; + h->revision = NCSI_PKT_REVISION; + h->reserved = 0; + h->id = nca->id; + h->type = nca->type; + h->channel = NCSI_TO_CHANNEL(nca->package, + nca->channel); + h->length = htons(nca->payload); + h->reserved1[0] = 0; + h->reserved1[1] = 0; + + /* Fill with calculated checksum */ + checksum = ncsi_calculate_checksum((unsigned char *)h, + sizeof(*h) + nca->payload); + pchecksum = (__be32 *)((void *)h + sizeof(struct ncsi_pkt_hdr) + + nca->payload); + *pchecksum = htonl(checksum); +} + +static int ncsi_cmd_handler_default(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_pkt *cmd; + + cmd = (struct ncsi_cmd_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_sp(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_sp_pkt *cmd; + + cmd = (struct ncsi_cmd_sp_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->hw_arbitration = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_dc(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_dc_pkt *cmd; + + cmd = (struct ncsi_cmd_dc_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->ald = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_rc(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_rc_pkt *cmd; + + cmd = (struct ncsi_cmd_rc_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_ae(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_ae_pkt *cmd; + + cmd = (struct ncsi_cmd_ae_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mc_id = nca->bytes[0]; + cmd->mode = htonl(nca->dwords[1]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_sl(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_sl_pkt *cmd; + + cmd = (struct ncsi_cmd_sl_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = htonl(nca->dwords[0]); + cmd->oem_mode = htonl(nca->dwords[1]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_svf(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_svf_pkt *cmd; + + cmd = (struct ncsi_cmd_svf_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->vlan = htons(nca->words[0]); + cmd->index = nca->bytes[2]; + cmd->enable = nca->bytes[3]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_ev(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_ev_pkt *cmd; + + cmd = (struct ncsi_cmd_ev_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_sma(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_sma_pkt *cmd; + int i; + + cmd = (struct ncsi_cmd_sma_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + for (i = 0; i < 6; i++) + cmd->mac[i] = nca->bytes[i]; + cmd->index = nca->bytes[6]; + cmd->at_e = nca->bytes[7]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_ebf(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_ebf_pkt *cmd; + + cmd = (struct ncsi_cmd_ebf_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = htonl(nca->dwords[0]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_egmf(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_egmf_pkt *cmd; + + cmd = (struct ncsi_cmd_egmf_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = htonl(nca->dwords[0]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_snfc(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_snfc_pkt *cmd; + + cmd = (struct ncsi_cmd_snfc_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static struct ncsi_cmd_handler { + unsigned char type; + int payload; + int (*handler)(struct sk_buff *skb, + struct ncsi_cmd_arg *nca); +} ncsi_cmd_handlers[] = { + { NCSI_PKT_CMD_CIS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SP, 4, ncsi_cmd_handler_sp }, + { NCSI_PKT_CMD_DP, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_EC, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_DC, 4, ncsi_cmd_handler_dc }, + { NCSI_PKT_CMD_RC, 4, ncsi_cmd_handler_rc }, + { NCSI_PKT_CMD_ECNT, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_DCNT, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_AE, 8, ncsi_cmd_handler_ae }, + { NCSI_PKT_CMD_SL, 8, ncsi_cmd_handler_sl }, + { NCSI_PKT_CMD_GLS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SVF, 4, ncsi_cmd_handler_svf }, + { NCSI_PKT_CMD_EV, 4, ncsi_cmd_handler_ev }, + { NCSI_PKT_CMD_DV, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SMA, 8, ncsi_cmd_handler_sma }, + { NCSI_PKT_CMD_EBF, 4, ncsi_cmd_handler_ebf }, + { NCSI_PKT_CMD_DBF, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_EGMF, 4, ncsi_cmd_handler_egmf }, + { NCSI_PKT_CMD_DGMF, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SNFC, 4, ncsi_cmd_handler_snfc }, + { NCSI_PKT_CMD_GVI, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GC, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GP, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GCPS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GNS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GNPTS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GPS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_OEM, 0, NULL }, + { NCSI_PKT_CMD_PLDM, 0, NULL }, + { NCSI_PKT_CMD_GPUUID, 0, ncsi_cmd_handler_default } +}; + +static struct ncsi_request *ncsi_alloc_command(struct ncsi_cmd_arg *nca) +{ + struct ncsi_dev_priv *ndp = nca->ndp; + struct ncsi_dev *nd = &ndp->ndev; + struct net_device *dev = nd->dev; + int hlen = LL_RESERVED_SPACE(dev); + int tlen = dev->needed_tailroom; + int len = hlen + tlen; + struct sk_buff *skb; + struct ncsi_request *nr; + + nr = ncsi_alloc_request(ndp, nca->driven); + if (!nr) + return NULL; + + /* NCSI command packet has 16-bytes header, payload, 4 bytes checksum. + * The packet needs padding if its payload is less than 26 bytes to + * meet 64 bytes minimal ethernet frame length. + */ + len += sizeof(struct ncsi_cmd_pkt_hdr) + 4; + if (nca->payload < 26) + len += 26; + else + len += nca->payload; + + /* Allocate skb */ + skb = alloc_skb(len, GFP_ATOMIC); + if (!skb) { + ncsi_free_request(nr); + return NULL; + } + + nr->cmd = skb; + skb_reserve(skb, hlen); + skb_reset_network_header(skb); + + skb->dev = dev; + skb->protocol = htons(ETH_P_NCSI); + + return nr; +} + +int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca) +{ + struct ncsi_request *nr; + struct ethhdr *eh; + struct ncsi_cmd_handler *nch = NULL; + int i, ret; + + /* Search for the handler */ + for (i = 0; i < ARRAY_SIZE(ncsi_cmd_handlers); i++) { + if (ncsi_cmd_handlers[i].type == nca->type) { + if (ncsi_cmd_handlers[i].handler) + nch = &ncsi_cmd_handlers[i]; + else + nch = NULL; + + break; + } + } + + if (!nch) { + netdev_err(nca->ndp->ndev.dev, + "Cannot send packet with type 0x%02x\n", nca->type); + return -ENOENT; + } + + /* Get packet payload length and allocate the request */ + nca->payload = nch->payload; + nr = ncsi_alloc_command(nca); + if (!nr) + return -ENOMEM; + + /* Prepare the packet */ + nca->id = nr->id; + ret = nch->handler(nr->cmd, nca); + if (ret) { + ncsi_free_request(nr); + return ret; + } + + /* Fill the ethernet header */ + eh = (struct ethhdr *)skb_push(nr->cmd, sizeof(*eh)); + eh->h_proto = htons(ETH_P_NCSI); + eth_broadcast_addr(eh->h_dest); + eth_broadcast_addr(eh->h_source); + + /* Start the timer for the request that might not have + * corresponding response. Given NCSI is an internal + * connection a 1 second delay should be sufficient. + */ + nr->enabled = true; + mod_timer(&nr->timer, jiffies + 1 * HZ); + + /* Send NCSI packet */ + skb_get(nr->cmd); + ret = dev_queue_xmit(nr->cmd); + if (ret < 0) { + ncsi_free_request(nr); + return ret; + } + + return 0; +} diff --git a/net/ncsi/ncsi-pkt.h b/net/ncsi/ncsi-pkt.h new file mode 100644 index 000000000000..548145863c49 --- /dev/null +++ b/net/ncsi/ncsi-pkt.h @@ -0,0 +1,171 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __NCSI_PKT_H__ +#define __NCSI_PKT_H__ + +struct ncsi_pkt_hdr { + unsigned char mc_id; /* Management controller ID */ + unsigned char revision; /* NCSI version - 0x01 */ + unsigned char reserved; /* Reserved */ + unsigned char id; /* Packet sequence number */ + unsigned char type; /* Packet type */ + unsigned char channel; /* Network controller ID */ + __be16 length; /* Payload length */ + __be32 reserved1[2]; /* Reserved */ +}; + +struct ncsi_cmd_pkt_hdr { + struct ncsi_pkt_hdr common; /* Common NCSI packet header */ +}; + +/* NCSI common command packet */ +struct ncsi_cmd_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 checksum; /* Checksum */ + unsigned char pad[26]; +}; + +/* Select Package */ +struct ncsi_cmd_sp_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char hw_arbitration; /* HW arbitration */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Disable Channel */ +struct ncsi_cmd_dc_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char ald; /* Allow link down */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Reset Channel */ +struct ncsi_cmd_rc_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 reserved; /* Reserved */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* AEN Enable */ +struct ncsi_cmd_ae_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char mc_id; /* MC ID */ + __be32 mode; /* AEN working mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + +/* Set Link */ +struct ncsi_cmd_sl_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 mode; /* Link working mode */ + __be32 oem_mode; /* OEM link mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + +/* Set VLAN Filter */ +struct ncsi_cmd_svf_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be16 reserved; /* Reserved */ + __be16 vlan; /* VLAN ID */ + __be16 reserved1; /* Reserved */ + unsigned char index; /* VLAN table index */ + unsigned char enable; /* Enable or disable */ + __be32 checksum; /* Checksum */ + unsigned char pad[14]; +}; + +/* Enable VLAN */ +struct ncsi_cmd_ev_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char mode; /* VLAN filter mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Set MAC Address */ +struct ncsi_cmd_sma_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char mac[6]; /* MAC address */ + unsigned char index; /* MAC table index */ + unsigned char at_e; /* Addr type and operation */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + +/* Enable Broadcast Filter */ +struct ncsi_cmd_ebf_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 mode; /* Filter mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Enable Global Multicast Filter */ +struct ncsi_cmd_egmf_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 mode; /* Global MC mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Set NCSI Flow Control */ +struct ncsi_cmd_snfc_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char mode; /* Flow control mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* NCSI packet revision */ +#define NCSI_PKT_REVISION 0x01 + +/* NCSI packet commands */ +#define NCSI_PKT_CMD_CIS 0x00 /* Clear Initial State */ +#define NCSI_PKT_CMD_SP 0x01 /* Select Package */ +#define NCSI_PKT_CMD_DP 0x02 /* Deselect Package */ +#define NCSI_PKT_CMD_EC 0x03 /* Enable Channel */ +#define NCSI_PKT_CMD_DC 0x04 /* Disable Channel */ +#define NCSI_PKT_CMD_RC 0x05 /* Reset Channel */ +#define NCSI_PKT_CMD_ECNT 0x06 /* Enable Channel Network Tx */ +#define NCSI_PKT_CMD_DCNT 0x07 /* Disable Channel Network Tx */ +#define NCSI_PKT_CMD_AE 0x08 /* AEN Enable */ +#define NCSI_PKT_CMD_SL 0x09 /* Set Link */ +#define NCSI_PKT_CMD_GLS 0x0a /* Get Link */ +#define NCSI_PKT_CMD_SVF 0x0b /* Set VLAN Filter */ +#define NCSI_PKT_CMD_EV 0x0c /* Enable VLAN */ +#define NCSI_PKT_CMD_DV 0x0d /* Disable VLAN */ +#define NCSI_PKT_CMD_SMA 0x0e /* Set MAC address */ +#define NCSI_PKT_CMD_EBF 0x10 /* Enable Broadcast Filter */ +#define NCSI_PKT_CMD_DBF 0x11 /* Disable Broadcast Filter */ +#define NCSI_PKT_CMD_EGMF 0x12 /* Enable Global Multicast Filter */ +#define NCSI_PKT_CMD_DGMF 0x13 /* Disable Global Multicast Filter */ +#define NCSI_PKT_CMD_SNFC 0x14 /* Set NCSI Flow Control */ +#define NCSI_PKT_CMD_GVI 0x15 /* Get Version ID */ +#define NCSI_PKT_CMD_GC 0x16 /* Get Capabilities */ +#define NCSI_PKT_CMD_GP 0x17 /* Get Parameters */ +#define NCSI_PKT_CMD_GCPS 0x18 /* Get Controller Packet Statistics */ +#define NCSI_PKT_CMD_GNS 0x19 /* Get NCSI Statistics */ +#define NCSI_PKT_CMD_GNPTS 0x1a /* Get NCSI Pass-throu Statistics */ +#define NCSI_PKT_CMD_GPS 0x1b /* Get package status */ +#define NCSI_PKT_CMD_OEM 0x50 /* OEM */ +#define NCSI_PKT_CMD_PLDM 0x51 /* PLDM request over NCSI over RBT */ +#define NCSI_PKT_CMD_GPUUID 0x52 /* Get package UUID */ + +#endif /* __NCSI_PKT_H__ */ -- cgit v1.2.3 From 6a773a15a1e8874e5eccd2f29190c31085912c95 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:47 -0700 Subject: bpf: add XDP prog type for early driver filter Add a new bpf prog type that is intended to run in early stages of the packet rx path. Only minimal packet metadata will be available, hence a new context type, struct xdp_md, is exposed to userspace. So far only expose the packet start and end pointers, and only in read mode. An XDP program must return one of the well known enum values, all other return codes are reserved for future use. Unfortunately, this restriction is hard to enforce at verification time, so take the approach of warning at runtime when such programs are encountered. Out of bounds return codes should alias to XDP_ABORTED. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/filter.h | 18 +++++++++++ include/uapi/linux/bpf.h | 20 ++++++++++++ kernel/bpf/verifier.c | 1 + net/core/filter.c | 79 ++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 118 insertions(+) (limited to 'include/uapi') diff --git a/include/linux/filter.h b/include/linux/filter.h index 6fc31ef1da2d..15d816a8b755 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -368,6 +368,11 @@ struct bpf_skb_data_end { void *data_end; }; +struct xdp_buff { + void *data; + void *data_end; +}; + /* compute the linear packet data range [data, data_end) which * will be accessed by cls_bpf and act_bpf programs */ @@ -429,6 +434,18 @@ static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog, return BPF_PROG_RUN(prog, skb); } +static inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog, + struct xdp_buff *xdp) +{ + u32 ret; + + rcu_read_lock(); + ret = BPF_PROG_RUN(prog, (void *)xdp); + rcu_read_unlock(); + + return ret; +} + static inline unsigned int bpf_prog_size(unsigned int proglen) { return max(sizeof(struct bpf_prog), @@ -509,6 +526,7 @@ bool bpf_helper_changes_skb_data(void *func); struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, const struct bpf_insn *patch, u32 len); +void bpf_warn_invalid_xdp_action(u32 act); #ifdef CONFIG_BPF_JIT extern int bpf_jit_enable; diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index c4d922439d20..a51786566c2f 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -94,6 +94,7 @@ enum bpf_prog_type { BPF_PROG_TYPE_SCHED_CLS, BPF_PROG_TYPE_SCHED_ACT, BPF_PROG_TYPE_TRACEPOINT, + BPF_PROG_TYPE_XDP, }; #define BPF_PSEUDO_MAP_FD 1 @@ -439,4 +440,23 @@ struct bpf_tunnel_key { __u32 tunnel_label; }; +/* User return codes for XDP prog type. + * A valid XDP program must return one of these defined values. All other + * return codes are reserved for future use. Unknown return codes will result + * in packet drop. + */ +enum xdp_action { + XDP_ABORTED = 0, + XDP_DROP, + XDP_PASS, +}; + +/* user accessible metadata for XDP packet hook + * new fields must be added to the end of this structure + */ +struct xdp_md { + __u32 data; + __u32 data_end; +}; + #endif /* _UAPI__LINUX_BPF_H__ */ diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index e206c2181412..a8d67d097b0d 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -713,6 +713,7 @@ static int check_ptr_alignment(struct verifier_env *env, struct reg_state *reg, switch (env->prog->type) { case BPF_PROG_TYPE_SCHED_CLS: case BPF_PROG_TYPE_SCHED_ACT: + case BPF_PROG_TYPE_XDP: break; default: verbose("verifier is misconfigured\n"); diff --git a/net/core/filter.c b/net/core/filter.c index 22e3992c8b48..6c627bc4be6e 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2410,6 +2410,12 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) } } +static const struct bpf_func_proto * +xdp_func_proto(enum bpf_func_id func_id) +{ + return sk_filter_func_proto(func_id); +} + static bool __is_valid_access(int off, int size, enum bpf_access_type type) { if (off < 0 || off >= sizeof(struct __sk_buff)) @@ -2477,6 +2483,44 @@ static bool tc_cls_act_is_valid_access(int off, int size, return __is_valid_access(off, size, type); } +static bool __is_valid_xdp_access(int off, int size, + enum bpf_access_type type) +{ + if (off < 0 || off >= sizeof(struct xdp_md)) + return false; + if (off % size != 0) + return false; + if (size != 4) + return false; + + return true; +} + +static bool xdp_is_valid_access(int off, int size, + enum bpf_access_type type, + enum bpf_reg_type *reg_type) +{ + if (type == BPF_WRITE) + return false; + + switch (off) { + case offsetof(struct xdp_md, data): + *reg_type = PTR_TO_PACKET; + break; + case offsetof(struct xdp_md, data_end): + *reg_type = PTR_TO_PACKET_END; + break; + } + + return __is_valid_xdp_access(off, size, type); +} + +void bpf_warn_invalid_xdp_action(u32 act) +{ + WARN_ONCE(1, "Illegal XDP return value %u, expect packet loss\n", act); +} +EXPORT_SYMBOL_GPL(bpf_warn_invalid_xdp_action); + static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg, int src_reg, int ctx_off, struct bpf_insn *insn_buf, @@ -2628,6 +2672,29 @@ static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg, return insn - insn_buf; } +static u32 xdp_convert_ctx_access(enum bpf_access_type type, int dst_reg, + int src_reg, int ctx_off, + struct bpf_insn *insn_buf, + struct bpf_prog *prog) +{ + struct bpf_insn *insn = insn_buf; + + switch (ctx_off) { + case offsetof(struct xdp_md, data): + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct xdp_buff, data)), + dst_reg, src_reg, + offsetof(struct xdp_buff, data)); + break; + case offsetof(struct xdp_md, data_end): + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct xdp_buff, data_end)), + dst_reg, src_reg, + offsetof(struct xdp_buff, data_end)); + break; + } + + return insn - insn_buf; +} + static const struct bpf_verifier_ops sk_filter_ops = { .get_func_proto = sk_filter_func_proto, .is_valid_access = sk_filter_is_valid_access, @@ -2640,6 +2707,12 @@ static const struct bpf_verifier_ops tc_cls_act_ops = { .convert_ctx_access = bpf_net_convert_ctx_access, }; +static const struct bpf_verifier_ops xdp_ops = { + .get_func_proto = xdp_func_proto, + .is_valid_access = xdp_is_valid_access, + .convert_ctx_access = xdp_convert_ctx_access, +}; + static struct bpf_prog_type_list sk_filter_type __read_mostly = { .ops = &sk_filter_ops, .type = BPF_PROG_TYPE_SOCKET_FILTER, @@ -2655,11 +2728,17 @@ static struct bpf_prog_type_list sched_act_type __read_mostly = { .type = BPF_PROG_TYPE_SCHED_ACT, }; +static struct bpf_prog_type_list xdp_type __read_mostly = { + .ops = &xdp_ops, + .type = BPF_PROG_TYPE_XDP, +}; + static int __init register_sk_filter_ops(void) { bpf_register_prog_type(&sk_filter_type); bpf_register_prog_type(&sched_cls_type); bpf_register_prog_type(&sched_act_type); + bpf_register_prog_type(&xdp_type); return 0; } -- cgit v1.2.3 From d1fdd9138682e0f272beee0cb08b6328c5478b26 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:49 -0700 Subject: rtnl: add option for setting link xdp prog Sets the bpf program represented by fd as an early filter in the rx path of the netdev. The fd must have been created as BPF_PROG_TYPE_XDP. Providing a negative value as fd clears the program. Getting the fd back via rtnl is not possible, therefore reading of this value merely provides a bool whether the program is valid on the link or not. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller --- include/uapi/linux/if_link.h | 12 +++++++++ net/core/rtnetlink.c | 64 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 76 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 4285ac31e865..a1b5202c5f6b 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -156,6 +156,7 @@ enum { IFLA_GSO_MAX_SEGS, IFLA_GSO_MAX_SIZE, IFLA_PAD, + IFLA_XDP, __IFLA_MAX }; @@ -843,4 +844,15 @@ enum { }; #define LINK_XSTATS_TYPE_MAX (__LINK_XSTATS_TYPE_MAX - 1) +/* XDP section */ + +enum { + IFLA_XDP_UNSPEC, + IFLA_XDP_FD, + IFLA_XDP_ATTACHED, + __IFLA_XDP_MAX, +}; + +#define IFLA_XDP_MAX (__IFLA_XDP_MAX - 1) + #endif /* _UAPI_LINUX_IF_LINK_H */ diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a9e3805af739..eba2b8260dbd 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -891,6 +891,16 @@ static size_t rtnl_port_size(const struct net_device *dev, return port_self_size; } +static size_t rtnl_xdp_size(const struct net_device *dev) +{ + size_t xdp_size = nla_total_size(1); /* XDP_ATTACHED */ + + if (!dev->netdev_ops->ndo_xdp) + return 0; + else + return xdp_size; +} + static noinline size_t if_nlmsg_size(const struct net_device *dev, u32 ext_filter_mask) { @@ -927,6 +937,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */ + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */ + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */ + + rtnl_xdp_size(dev) /* IFLA_XDP */ + nla_total_size(1); /* IFLA_PROTO_DOWN */ } @@ -1211,6 +1222,33 @@ static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev) return 0; } +static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev) +{ + struct netdev_xdp xdp_op = {}; + struct nlattr *xdp; + int err; + + if (!dev->netdev_ops->ndo_xdp) + return 0; + xdp = nla_nest_start(skb, IFLA_XDP); + if (!xdp) + return -EMSGSIZE; + xdp_op.command = XDP_QUERY_PROG; + err = dev->netdev_ops->ndo_xdp(dev, &xdp_op); + if (err) + goto err_cancel; + err = nla_put_u8(skb, IFLA_XDP_ATTACHED, xdp_op.prog_attached); + if (err) + goto err_cancel; + + nla_nest_end(skb, xdp); + return 0; + +err_cancel: + nla_nest_cancel(skb, xdp); + return err; +} + static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, int type, u32 pid, u32 seq, u32 change, unsigned int flags, u32 ext_filter_mask) @@ -1307,6 +1345,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, if (rtnl_port_fill(skb, dev, ext_filter_mask)) goto nla_put_failure; + if (rtnl_xdp_fill(skb, dev)) + goto nla_put_failure; + if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) { if (rtnl_link_fill(skb, dev) < 0) goto nla_put_failure; @@ -1392,6 +1433,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = { [IFLA_PHYS_SWITCH_ID] = { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN }, [IFLA_LINK_NETNSID] = { .type = NLA_S32 }, [IFLA_PROTO_DOWN] = { .type = NLA_U8 }, + [IFLA_XDP] = { .type = NLA_NESTED }, }; static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = { @@ -1429,6 +1471,11 @@ static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = { [IFLA_PORT_RESPONSE] = { .type = NLA_U16, }, }; +static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = { + [IFLA_XDP_FD] = { .type = NLA_S32 }, + [IFLA_XDP_ATTACHED] = { .type = NLA_U8 }, +}; + static const struct rtnl_link_ops *linkinfo_to_kind_ops(const struct nlattr *nla) { const struct rtnl_link_ops *ops = NULL; @@ -2054,6 +2101,23 @@ static int do_setlink(const struct sk_buff *skb, status |= DO_SETLINK_NOTIFY; } + if (tb[IFLA_XDP]) { + struct nlattr *xdp[IFLA_XDP_MAX + 1]; + + err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP], + ifla_xdp_policy); + if (err < 0) + goto errout; + + if (xdp[IFLA_XDP_FD]) { + err = dev_change_xdp_fd(dev, + nla_get_s32(xdp[IFLA_XDP_FD])); + if (err) + goto errout; + status |= DO_SETLINK_NOTIFY; + } + } + errout: if (status & DO_SETLINK_MODIFIED) { if (status & DO_SETLINK_NOTIFY) -- cgit v1.2.3 From 6ce96ca348a9e949f8c43f4d3e98db367d93cffd Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:53 -0700 Subject: bpf: add XDP_TX xdp_action for direct forwarding XDP enabled drivers must transmit received packets back out on the same port they were received on when a program returns this action. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index a51786566c2f..2b7076f5b5ad 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -449,6 +449,7 @@ enum xdp_action { XDP_ABORTED = 0, XDP_DROP, XDP_PASS, + XDP_TX, }; /* user accessible metadata for XDP packet hook -- cgit v1.2.3 From b02b94b331be5097d248d49242bd1fc0649a8092 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 20 Jul 2016 20:17:47 +0200 Subject: bpf, elf: add official ELF machine define for eBPF Add the official BPF ELF e_machine value that was assigned recently [1,2] and will be propagated to glibc, et al. LLVM is switching to it in 3.9 release. [1] https://github.com/llvm-mirror/llvm/commit/36b9c09330bfb5e771914cfe307588f30d5510d2 [2] http://lists.iovisor.org/pipermail/iovisor-dev/2016-June/000266.html Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/elf-em.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/elf-em.h b/include/uapi/linux/elf-em.h index c3fdfe79e5cc..cb5d1a519202 100644 --- a/include/uapi/linux/elf-em.h +++ b/include/uapi/linux/elf-em.h @@ -40,6 +40,7 @@ #define EM_TILEPRO 188 /* Tilera TILEPro */ #define EM_MICROBLAZE 189 /* Xilinx MicroBlaze */ #define EM_TILEGX 191 /* Tilera TILE-Gx */ +#define EM_BPF 247 /* Linux BPF - in-kernel virtual machine */ #define EM_FRV 0x5441 /* Fujitsu FR-V */ #define EM_AVR32 0x18ad /* Atmel AVR32 */ -- cgit v1.2.3 From 545ed20e6df68a4d2584a29a2a28ee8b2f7e9547 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Wed, 22 Jun 2016 17:54:53 -0600 Subject: dm: add infrastructure for DAX support Change mapped device to implement direct_access function, dm_blk_direct_access(), which calls a target direct_access function. 'struct target_type' is extended to have target direct_access interface. This function limits direct accessible size to the dm_target's limit with max_io_len(). Add dm_table_supports_dax() to iterate all targets and associated block devices to check for DAX support. To add DAX support to a DM target the target must only implement the direct_access function. Add a new dm type, DM_TYPE_DAX_BIO_BASED, which indicates that mapped device supports DAX and is bio based. This new type is used to assure that all target devices have DAX support and remain that way after QUEUE_FLAG_DAX is set in mapped device. At initial table load, QUEUE_FLAG_DAX is set to mapped device when setting DM_TYPE_DAX_BIO_BASED to the type. Any subsequent table load to the mapped device must have the same type, or else it fails per the check in table_load(). Signed-off-by: Toshi Kani Signed-off-by: Mike Snitzer --- drivers/md/dm-table.c | 44 ++++++++++++++++++++++++++++++++++++++++++- drivers/md/dm.c | 38 +++++++++++++++++++++++++++++++++++-- drivers/md/dm.h | 1 + include/linux/device-mapper.h | 10 ++++++++++ include/uapi/linux/dm-ioctl.h | 4 ++-- 5 files changed, 92 insertions(+), 5 deletions(-) (limited to 'include/uapi') diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 88f01744ac16..ee6f37eafbc3 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -827,6 +827,12 @@ void dm_consume_args(struct dm_arg_set *as, unsigned num_args) } EXPORT_SYMBOL(dm_consume_args); +static bool __table_type_bio_based(unsigned table_type) +{ + return (table_type == DM_TYPE_BIO_BASED || + table_type == DM_TYPE_DAX_BIO_BASED); +} + static bool __table_type_request_based(unsigned table_type) { return (table_type == DM_TYPE_REQUEST_BASED || @@ -839,6 +845,34 @@ void dm_table_set_type(struct dm_table *t, unsigned type) } EXPORT_SYMBOL_GPL(dm_table_set_type); +static int device_supports_dax(struct dm_target *ti, struct dm_dev *dev, + sector_t start, sector_t len, void *data) +{ + struct request_queue *q = bdev_get_queue(dev->bdev); + + return q && blk_queue_dax(q); +} + +static bool dm_table_supports_dax(struct dm_table *t) +{ + struct dm_target *ti; + unsigned i = 0; + + /* Ensure that all targets support DAX. */ + while (i < dm_table_get_num_targets(t)) { + ti = dm_table_get_target(t, i++); + + if (!ti->type->direct_access) + return false; + + if (!ti->type->iterate_devices || + !ti->type->iterate_devices(ti, device_supports_dax, NULL)) + return false; + } + + return true; +} + static int dm_table_determine_type(struct dm_table *t) { unsigned i; @@ -853,6 +887,7 @@ static int dm_table_determine_type(struct dm_table *t) /* target already set the table's type */ if (t->type == DM_TYPE_BIO_BASED) return 0; + BUG_ON(t->type == DM_TYPE_DAX_BIO_BASED); goto verify_rq_based; } @@ -887,6 +922,8 @@ static int dm_table_determine_type(struct dm_table *t) if (bio_based) { /* We must use this table as bio-based */ t->type = DM_TYPE_BIO_BASED; + if (dm_table_supports_dax(t)) + t->type = DM_TYPE_DAX_BIO_BASED; return 0; } @@ -979,6 +1016,11 @@ struct dm_target *dm_table_get_wildcard_target(struct dm_table *t) return NULL; } +bool dm_table_bio_based(struct dm_table *t) +{ + return __table_type_bio_based(dm_table_get_type(t)); +} + bool dm_table_request_based(struct dm_table *t) { return __table_type_request_based(dm_table_get_type(t)); @@ -1001,7 +1043,7 @@ static int dm_table_alloc_md_mempools(struct dm_table *t, struct mapped_device * return -EINVAL; } - if (type == DM_TYPE_BIO_BASED) + if (__table_type_bio_based(type)) for (i = 0; i < t->num_targets; i++) { tgt = t->targets + i; per_io_data_size = max(per_io_data_size, tgt->per_io_data_size); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 7538b8972820..4dca5a792e4b 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -905,6 +905,33 @@ int dm_set_target_max_io_len(struct dm_target *ti, sector_t len) } EXPORT_SYMBOL_GPL(dm_set_target_max_io_len); +static long dm_blk_direct_access(struct block_device *bdev, sector_t sector, + void __pmem **kaddr, pfn_t *pfn, long size) +{ + struct mapped_device *md = bdev->bd_disk->private_data; + struct dm_table *map; + struct dm_target *ti; + int srcu_idx; + long len, ret = -EIO; + + map = dm_get_live_table(md, &srcu_idx); + if (!map) + goto out; + + ti = dm_table_find_target(map, sector); + if (!dm_target_is_valid(ti)) + goto out; + + len = max_io_len(sector, ti) << SECTOR_SHIFT; + size = min(len, size); + + if (ti->type->direct_access) + ret = ti->type->direct_access(ti, sector, kaddr, pfn, size); +out: + dm_put_live_table(md, srcu_idx); + return min(ret, size); +} + /* * A target may call dm_accept_partial_bio only from the map routine. It is * allowed for all bio types except REQ_PREFLUSH. @@ -1548,7 +1575,7 @@ static void __bind_mempools(struct mapped_device *md, struct dm_table *t) if (md->bs) { /* The md already has necessary mempools. */ - if (dm_table_get_type(t) == DM_TYPE_BIO_BASED) { + if (dm_table_bio_based(t)) { /* * Reload bioset because front_pad may have changed * because a different table was loaded. @@ -1744,8 +1771,9 @@ EXPORT_SYMBOL_GPL(dm_get_queue_limits); int dm_setup_md_queue(struct mapped_device *md, struct dm_table *t) { int r; + unsigned type = dm_get_md_type(md); - switch (dm_get_md_type(md)) { + switch (type) { case DM_TYPE_REQUEST_BASED: r = dm_old_init_request_queue(md); if (r) { @@ -1761,6 +1789,7 @@ int dm_setup_md_queue(struct mapped_device *md, struct dm_table *t) } break; case DM_TYPE_BIO_BASED: + case DM_TYPE_DAX_BIO_BASED: dm_init_normal_md_queue(md); blk_queue_make_request(md->queue, dm_make_request); /* @@ -1769,6 +1798,9 @@ int dm_setup_md_queue(struct mapped_device *md, struct dm_table *t) */ bioset_free(md->queue->bio_split); md->queue->bio_split = NULL; + + if (type == DM_TYPE_DAX_BIO_BASED) + queue_flag_set_unlocked(QUEUE_FLAG_DAX, md->queue); break; } @@ -2465,6 +2497,7 @@ struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, unsigned t switch (type) { case DM_TYPE_BIO_BASED: + case DM_TYPE_DAX_BIO_BASED: cachep = _io_cache; pool_size = dm_get_reserved_bio_based_ios(); front_pad = roundup(per_io_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone); @@ -2691,6 +2724,7 @@ static const struct block_device_operations dm_blk_dops = { .open = dm_blk_open, .release = dm_blk_close, .ioctl = dm_blk_ioctl, + .direct_access = dm_blk_direct_access, .getgeo = dm_blk_getgeo, .pr_ops = &dm_pr_ops, .owner = THIS_MODULE diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 2e0e4a53a312..f0aad08b9654 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -68,6 +68,7 @@ unsigned dm_table_get_type(struct dm_table *t); struct target_type *dm_table_get_immutable_target_type(struct dm_table *t); struct dm_target *dm_table_get_immutable_target(struct dm_table *t); struct dm_target *dm_table_get_wildcard_target(struct dm_table *t); +bool dm_table_bio_based(struct dm_table *t); bool dm_table_request_based(struct dm_table *t); bool dm_table_all_blk_mq_devices(struct dm_table *t); void dm_table_free_md_mempools(struct dm_table *t); diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 2ce339212b6e..b0db857f334b 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -26,6 +26,7 @@ struct bio_vec; #define DM_TYPE_BIO_BASED 1 #define DM_TYPE_REQUEST_BASED 2 #define DM_TYPE_MQ_REQUEST_BASED 3 +#define DM_TYPE_DAX_BIO_BASED 4 typedef enum { STATUSTYPE_INFO, STATUSTYPE_TABLE } status_type_t; @@ -124,6 +125,14 @@ typedef void (*dm_io_hints_fn) (struct dm_target *ti, */ typedef int (*dm_busy_fn) (struct dm_target *ti); +/* + * Returns: + * < 0 : error + * >= 0 : the number of bytes accessible at the address + */ +typedef long (*dm_direct_access_fn) (struct dm_target *ti, sector_t sector, + void __pmem **kaddr, pfn_t *pfn, long size); + void dm_error(const char *message); struct dm_dev { @@ -170,6 +179,7 @@ struct target_type { dm_busy_fn busy; dm_iterate_devices_fn iterate_devices; dm_io_hints_fn io_hints; + dm_direct_access_fn direct_access; /* For internal device-mapper use. */ struct list_head list; diff --git a/include/uapi/linux/dm-ioctl.h b/include/uapi/linux/dm-ioctl.h index 30afd0a23c4b..4bf9f1eabffc 100644 --- a/include/uapi/linux/dm-ioctl.h +++ b/include/uapi/linux/dm-ioctl.h @@ -267,9 +267,9 @@ enum { #define DM_DEV_SET_GEOMETRY _IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl) #define DM_VERSION_MAJOR 4 -#define DM_VERSION_MINOR 34 +#define DM_VERSION_MINOR 35 #define DM_VERSION_PATCHLEVEL 0 -#define DM_VERSION_EXTRA "-ioctl (2015-10-28)" +#define DM_VERSION_EXTRA "-ioctl (2016-06-23)" /* Status bits */ #define DM_READONLY_FLAG (1 << 0) /* In/Out */ -- cgit v1.2.3 From 76a10b86785c5e3fc49bcee355502d035b07e47a Mon Sep 17 00:00:00 2001 From: Eric Auger Date: Fri, 22 Jul 2016 16:20:37 +0000 Subject: KVM: api: Pass the devid in the msi routing entry MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On ARM, the MSI msg (address and data) comes along with out-of-band device ID information. The device ID encodes the device that writes the MSI msg. Let's convey the device id in kvm_irq_routing_msi and use KVM_MSI_VALID_DEVID flag value in kvm_irq_routing_entry to indicate the msi devid is populated. Signed-off-by: Eric Auger Reviewed-by: Andre Przywara Acked-by: Radim Krčmář Signed-off-by: Marc Zyngier --- Documentation/virtual/kvm/api.txt | 19 +++++++++++++++++-- include/uapi/linux/kvm.h | 5 ++++- 2 files changed, 21 insertions(+), 3 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 07049eadb124..415cde1647e9 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -1468,7 +1468,11 @@ struct kvm_irq_routing_entry { #define KVM_IRQ_ROUTING_S390_ADAPTER 3 #define KVM_IRQ_ROUTING_HV_SINT 4 -No flags are specified so far, the corresponding field must be set to zero. +flags: +- KVM_MSI_VALID_DEVID: used along with KVM_IRQ_ROUTING_MSI + routing entry type, specifies that the devid field contains + a valid value. +- zero otherwise struct kvm_irq_routing_irqchip { __u32 irqchip; @@ -1479,9 +1483,20 @@ struct kvm_irq_routing_msi { __u32 address_lo; __u32 address_hi; __u32 data; - __u32 pad; + union { + __u32 pad; + __u32 devid; + }; }; +devid: If KVM_MSI_VALID_DEVID is set, contains a unique device identifier + for the device that wrote the MSI message. + For PCI, this is usually a BFD identifier in the lower 16 bits. + +The per-VM KVM_CAP_MSI_DEVID capability advertises the requirement to +provide the device ID. If this capability is not set, userland cannot +rely on the kernel to allow the KVM_MSI_VALID_DEVID flag being set. + struct kvm_irq_routing_s390_adapter { __u64 ind_addr; __u64 summary_addr; diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index d8c4c324cfae..eb2220895c6e 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -879,7 +879,10 @@ struct kvm_irq_routing_msi { __u32 address_lo; __u32 address_hi; __u32 data; - __u32 pad; + union { + __u32 pad; + __u32 devid; + }; }; struct kvm_irq_routing_s390_adapter { -- cgit v1.2.3 From bf3994d2ed310813da28362d87bfe9f0e1c3e37f Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 21 Jul 2016 12:03:11 +0200 Subject: net/sched: introduce Match-all classifier The matchall classifier matches every packet and allows the user to apply actions on it. This filter is very useful in usecases where every packet should be matched, for example, packet mirroring (SPAN) can be setup very easily using that filter. Signed-off-by: Jiri Pirko Signed-off-by: Yotam Gigi Signed-off-by: David S. Miller --- include/uapi/linux/pkt_cls.h | 11 ++ net/sched/Kconfig | 10 ++ net/sched/Makefile | 1 + net/sched/cls_matchall.c | 248 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 270 insertions(+) create mode 100644 net/sched/cls_matchall.c (limited to 'include/uapi') diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index 5702e933dc07..a32494887e01 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -433,6 +433,17 @@ enum { #define TCA_FLOWER_MAX (__TCA_FLOWER_MAX - 1) +/* Match-all classifier */ + +enum { + TCA_MATCHALL_UNSPEC, + TCA_MATCHALL_CLASSID, + TCA_MATCHALL_ACT, + __TCA_MATCHALL_MAX, +}; + +#define TCA_MATCHALL_MAX (__TCA_MATCHALL_MAX - 1) + /* Extended Matches */ struct tcf_ematch_tree_hdr { diff --git a/net/sched/Kconfig b/net/sched/Kconfig index b148302bbaf2..ccf931b3b94c 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -494,6 +494,16 @@ config NET_CLS_FLOWER To compile this code as a module, choose M here: the module will be called cls_flower. +config NET_CLS_MATCHALL + tristate "Match-all classifier" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets based on + nothing. Every packet will match. + + To compile this code as a module, choose M here: the module will + be called cls_matchall. + config NET_EMATCH bool "Extended Matches" select NET_CLS diff --git a/net/sched/Makefile b/net/sched/Makefile index 84bddb373517..ae088a5a9d95 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -60,6 +60,7 @@ obj-$(CONFIG_NET_CLS_FLOW) += cls_flow.o obj-$(CONFIG_NET_CLS_CGROUP) += cls_cgroup.o obj-$(CONFIG_NET_CLS_BPF) += cls_bpf.o obj-$(CONFIG_NET_CLS_FLOWER) += cls_flower.o +obj-$(CONFIG_NET_CLS_MATCHALL) += cls_matchall.o obj-$(CONFIG_NET_EMATCH) += ematch.o obj-$(CONFIG_NET_EMATCH_CMP) += em_cmp.o obj-$(CONFIG_NET_EMATCH_NBYTE) += em_nbyte.o diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c new file mode 100644 index 000000000000..8a6b4de7a99a --- /dev/null +++ b/net/sched/cls_matchall.c @@ -0,0 +1,248 @@ +/* + * net/sched/cls_matchll.c Match-all classifier + * + * Copyright (c) 2016 Jiri Pirko + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include + +#include +#include + +struct cls_mall_filter { + struct tcf_exts exts; + struct tcf_result res; + u32 handle; + struct rcu_head rcu; +}; + +struct cls_mall_head { + struct cls_mall_filter *filter; + struct rcu_head rcu; +}; + +static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct cls_mall_head *head = rcu_dereference_bh(tp->root); + struct cls_mall_filter *f = head->filter; + + return tcf_exts_exec(skb, &f->exts, res); +} + +static int mall_init(struct tcf_proto *tp) +{ + struct cls_mall_head *head; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOBUFS; + + rcu_assign_pointer(tp->root, head); + + return 0; +} + +static void mall_destroy_filter(struct rcu_head *head) +{ + struct cls_mall_filter *f = container_of(head, struct cls_mall_filter, rcu); + + tcf_exts_destroy(&f->exts); + kfree(f); +} + +static bool mall_destroy(struct tcf_proto *tp, bool force) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + + if (!force && head->filter) + return false; + + if (head->filter) + call_rcu(&head->filter->rcu, mall_destroy_filter); + RCU_INIT_POINTER(tp->root, NULL); + kfree_rcu(head, rcu); + return true; +} + +static unsigned long mall_get(struct tcf_proto *tp, u32 handle) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *f = head->filter; + + if (f && f->handle == handle) + return (unsigned long) f; + return 0; +} + +static const struct nla_policy mall_policy[TCA_MATCHALL_MAX + 1] = { + [TCA_MATCHALL_UNSPEC] = { .type = NLA_UNSPEC }, + [TCA_MATCHALL_CLASSID] = { .type = NLA_U32 }, +}; + +static int mall_set_parms(struct net *net, struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long base, struct nlattr **tb, + struct nlattr *est, bool ovr) +{ + struct tcf_exts e; + int err; + + tcf_exts_init(&e, TCA_MATCHALL_ACT, 0); + err = tcf_exts_validate(net, tp, tb, est, &e, ovr); + if (err < 0) + return err; + + if (tb[TCA_MATCHALL_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_MATCHALL_CLASSID]); + tcf_bind_filter(tp, &f->res, base); + } + + tcf_exts_change(tp, &f->exts, &e); + + return 0; +} + +static int mall_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + unsigned long *arg, bool ovr) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *fold = (struct cls_mall_filter *) *arg; + struct cls_mall_filter *f; + struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + int err; + + if (!tca[TCA_OPTIONS]) + return -EINVAL; + + if (head->filter) + return -EBUSY; + + if (fold) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_MATCHALL_MAX, + tca[TCA_OPTIONS], mall_policy); + if (err < 0) + return err; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + return -ENOBUFS; + + tcf_exts_init(&f->exts, TCA_MATCHALL_ACT, 0); + + if (!handle) + handle = 1; + f->handle = handle; + + err = mall_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); + if (err) + goto errout; + + *arg = (unsigned long) f; + rcu_assign_pointer(head->filter, f); + + return 0; + +errout: + kfree(f); + return err; +} + +static int mall_delete(struct tcf_proto *tp, unsigned long arg) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *f = (struct cls_mall_filter *) arg; + + RCU_INIT_POINTER(head->filter, NULL); + tcf_unbind_filter(tp, &f->res); + call_rcu(&f->rcu, mall_destroy_filter); + return 0; +} + +static void mall_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *f = head->filter; + + if (arg->count < arg->skip) + goto skip; + if (arg->fn(tp, (unsigned long) f, arg) < 0) + arg->stop = 1; +skip: + arg->count++; +} + +static int mall_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct cls_mall_filter *f = (struct cls_mall_filter *) fh; + struct nlattr *nest; + + if (!f) + return skb->len; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + if (f->res.classid && + nla_put_u32(skb, TCA_MATCHALL_CLASSID, f->res.classid)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static struct tcf_proto_ops cls_mall_ops __read_mostly = { + .kind = "matchall", + .classify = mall_classify, + .init = mall_init, + .destroy = mall_destroy, + .get = mall_get, + .change = mall_change, + .delete = mall_delete, + .walk = mall_walk, + .dump = mall_dump, + .owner = THIS_MODULE, +}; + +static int __init cls_mall_init(void) +{ + return register_tcf_proto_ops(&cls_mall_ops); +} + +static void __exit cls_mall_exit(void) +{ + unregister_tcf_proto_ops(&cls_mall_ops); +} + +module_init(cls_mall_init); +module_exit(cls_mall_exit); + +MODULE_AUTHOR("Jiri Pirko "); +MODULE_DESCRIPTION("Match-all classifier"); +MODULE_LICENSE("GPL v2"); -- cgit v1.2.3 From b87f7936a93246804cf70e7e2e0568799c948bb1 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:12 +0200 Subject: net/sched: Add match-all classifier hw offloading. Following the work that have been done on offloading classifiers like u32 and flower, now the match-all classifier hw offloading is possible. if the interface supports tc offloading. To control the offloading, two tc flags have been introduced: skip_sw and skip_hw. Typical usage: tc filter add dev eth25 parent ffff: \ matchall skip_sw \ action mirred egress mirror \ dev eth27 Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ include/net/pkt_cls.h | 11 +++++++ include/uapi/linux/pkt_cls.h | 1 + net/sched/cls_matchall.c | 76 ++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 87 insertions(+), 3 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 43c749b1b619..076df5360ba5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -787,6 +787,7 @@ enum { TC_SETUP_MQPRIO, TC_SETUP_CLSU32, TC_SETUP_CLSFLOWER, + TC_SETUP_MATCHALL, }; struct tc_cls_u32_offload; @@ -797,6 +798,7 @@ struct tc_to_netdev { u8 tc; struct tc_cls_u32_offload *cls_u32; struct tc_cls_flower_offload *cls_flower; + struct tc_cls_matchall_offload *cls_mall; }; }; diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 3722dda0199d..6f8d65342d3a 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -442,4 +442,15 @@ struct tc_cls_flower_offload { struct tcf_exts *exts; }; +enum tc_matchall_command { + TC_CLSMATCHALL_REPLACE, + TC_CLSMATCHALL_DESTROY, +}; + +struct tc_cls_matchall_offload { + enum tc_matchall_command command; + struct tcf_exts *exts; + unsigned long cookie; +}; + #endif diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index a32494887e01..d1c1ccaba787 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -439,6 +439,7 @@ enum { TCA_MATCHALL_UNSPEC, TCA_MATCHALL_CLASSID, TCA_MATCHALL_ACT, + TCA_MATCHALL_FLAGS, __TCA_MATCHALL_MAX, }; diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c index 8a6b4de7a99a..25927b6c4436 100644 --- a/net/sched/cls_matchall.c +++ b/net/sched/cls_matchall.c @@ -21,6 +21,7 @@ struct cls_mall_filter { struct tcf_result res; u32 handle; struct rcu_head rcu; + u32 flags; }; struct cls_mall_head { @@ -34,6 +35,9 @@ static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct cls_mall_head *head = rcu_dereference_bh(tp->root); struct cls_mall_filter *f = head->filter; + if (tc_skip_sw(f->flags)) + return -1; + return tcf_exts_exec(skb, &f->exts, res); } @@ -55,18 +59,61 @@ static void mall_destroy_filter(struct rcu_head *head) struct cls_mall_filter *f = container_of(head, struct cls_mall_filter, rcu); tcf_exts_destroy(&f->exts); + kfree(f); } +static int mall_replace_hw_filter(struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long cookie) +{ + struct net_device *dev = tp->q->dev_queue->dev; + struct tc_to_netdev offload; + struct tc_cls_matchall_offload mall_offload = {0}; + + offload.type = TC_SETUP_MATCHALL; + offload.cls_mall = &mall_offload; + offload.cls_mall->command = TC_CLSMATCHALL_REPLACE; + offload.cls_mall->exts = &f->exts; + offload.cls_mall->cookie = cookie; + + return dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, + &offload); +} + +static void mall_destroy_hw_filter(struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long cookie) +{ + struct net_device *dev = tp->q->dev_queue->dev; + struct tc_to_netdev offload; + struct tc_cls_matchall_offload mall_offload = {0}; + + offload.type = TC_SETUP_MATCHALL; + offload.cls_mall = &mall_offload; + offload.cls_mall->command = TC_CLSMATCHALL_DESTROY; + offload.cls_mall->exts = NULL; + offload.cls_mall->cookie = cookie; + + dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, + &offload); +} + static bool mall_destroy(struct tcf_proto *tp, bool force) { struct cls_mall_head *head = rtnl_dereference(tp->root); + struct net_device *dev = tp->q->dev_queue->dev; + struct cls_mall_filter *f = head->filter; - if (!force && head->filter) + if (!force && f) return false; - if (head->filter) - call_rcu(&head->filter->rcu, mall_destroy_filter); + if (f) { + if (tc_should_offload(dev, tp, f->flags)) + mall_destroy_hw_filter(tp, f, (unsigned long) f); + + call_rcu(&f->rcu, mall_destroy_filter); + } RCU_INIT_POINTER(tp->root, NULL); kfree_rcu(head, rcu); return true; @@ -117,8 +164,10 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, { struct cls_mall_head *head = rtnl_dereference(tp->root); struct cls_mall_filter *fold = (struct cls_mall_filter *) *arg; + struct net_device *dev = tp->q->dev_queue->dev; struct cls_mall_filter *f; struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + u32 flags = 0; int err; if (!tca[TCA_OPTIONS]) @@ -135,6 +184,12 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (err < 0) return err; + if (tb[TCA_MATCHALL_FLAGS]) { + flags = nla_get_u32(tb[TCA_MATCHALL_FLAGS]); + if (!tc_flags_valid(flags)) + return -EINVAL; + } + f = kzalloc(sizeof(*f), GFP_KERNEL); if (!f) return -ENOBUFS; @@ -144,11 +199,22 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (!handle) handle = 1; f->handle = handle; + f->flags = flags; err = mall_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); if (err) goto errout; + if (tc_should_offload(dev, tp, flags)) { + err = mall_replace_hw_filter(tp, f, (unsigned long) f); + if (err) { + if (tc_skip_sw(flags)) + goto errout; + else + err = 0; + } + } + *arg = (unsigned long) f; rcu_assign_pointer(head->filter, f); @@ -163,6 +229,10 @@ static int mall_delete(struct tcf_proto *tp, unsigned long arg) { struct cls_mall_head *head = rtnl_dereference(tp->root); struct cls_mall_filter *f = (struct cls_mall_filter *) arg; + struct net_device *dev = tp->q->dev_queue->dev; + + if (tc_should_offload(dev, tp, f->flags)) + mall_destroy_hw_filter(tp, f, (unsigned long) f); RCU_INIT_POINTER(head->filter, NULL); tcf_unbind_filter(tp, &f->res); -- cgit v1.2.3 From fbc872c38c8fed31948c85683b5326ee5ab9fccc Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Mon, 11 Jul 2016 15:45:51 +0100 Subject: xen/evtchn: add IOCTL_EVTCHN_RESTRICT IOCTL_EVTCHN_RESTRICT limits the file descriptor to being able to bind to interdomain event channels from a specific domain. Event channels that are already bound continue to work for sending and receiving notifications. This is useful as part of deprivileging a user space PV backend or device model (QEMU). e.g., Once the device model as bound to the ioreq server event channels it can restrict the file handle so an exploited DM cannot use it to create or bind to arbitrary event channels. Signed-off-by: David Vrabel Reviewed-by: Boris Ostrovsky --- drivers/xen/evtchn.c | 40 ++++++++++++++++++++++++++++++++++++++++ include/uapi/xen/evtchn.h | 15 +++++++++++++++ 2 files changed, 55 insertions(+) (limited to 'include/uapi') diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c index f4edd6df3df2..7efd1cb9bb40 100644 --- a/drivers/xen/evtchn.c +++ b/drivers/xen/evtchn.c @@ -73,8 +73,12 @@ struct per_user_data { wait_queue_head_t evtchn_wait; struct fasync_struct *evtchn_async_queue; const char *name; + + domid_t restrict_domid; }; +#define UNRESTRICTED_DOMID ((domid_t)-1) + struct user_evtchn { struct rb_node node; struct per_user_data *user; @@ -443,6 +447,10 @@ static long evtchn_ioctl(struct file *file, struct ioctl_evtchn_bind_virq bind; struct evtchn_bind_virq bind_virq; + rc = -EACCES; + if (u->restrict_domid != UNRESTRICTED_DOMID) + break; + rc = -EFAULT; if (copy_from_user(&bind, uarg, sizeof(bind))) break; @@ -468,6 +476,11 @@ static long evtchn_ioctl(struct file *file, if (copy_from_user(&bind, uarg, sizeof(bind))) break; + rc = -EACCES; + if (u->restrict_domid != UNRESTRICTED_DOMID && + u->restrict_domid != bind.remote_domain) + break; + bind_interdomain.remote_dom = bind.remote_domain; bind_interdomain.remote_port = bind.remote_port; rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain, @@ -485,6 +498,10 @@ static long evtchn_ioctl(struct file *file, struct ioctl_evtchn_bind_unbound_port bind; struct evtchn_alloc_unbound alloc_unbound; + rc = -EACCES; + if (u->restrict_domid != UNRESTRICTED_DOMID) + break; + rc = -EFAULT; if (copy_from_user(&bind, uarg, sizeof(bind))) break; @@ -553,6 +570,27 @@ static long evtchn_ioctl(struct file *file, break; } + case IOCTL_EVTCHN_RESTRICT_DOMID: { + struct ioctl_evtchn_restrict_domid ierd; + + rc = -EACCES; + if (u->restrict_domid != UNRESTRICTED_DOMID) + break; + + rc = -EFAULT; + if (copy_from_user(&ierd, uarg, sizeof(ierd))) + break; + + rc = -EINVAL; + if (ierd.domid == 0 || ierd.domid >= DOMID_FIRST_RESERVED) + break; + + u->restrict_domid = ierd.domid; + rc = 0; + + break; + } + default: rc = -ENOSYS; break; @@ -601,6 +639,8 @@ static int evtchn_open(struct inode *inode, struct file *filp) mutex_init(&u->ring_cons_mutex); spin_lock_init(&u->ring_prod_lock); + u->restrict_domid = UNRESTRICTED_DOMID; + filp->private_data = u; return nonseekable_open(inode, filp); diff --git a/include/uapi/xen/evtchn.h b/include/uapi/xen/evtchn.h index 14e833ee4e0b..cb4aa4bb905e 100644 --- a/include/uapi/xen/evtchn.h +++ b/include/uapi/xen/evtchn.h @@ -85,4 +85,19 @@ struct ioctl_evtchn_notify { #define IOCTL_EVTCHN_RESET \ _IOC(_IOC_NONE, 'E', 5, 0) +/* + * Restrict this file descriptor so that it can only be used to bind + * new interdomain events from one domain. + * + * Once a file descriptor has been restricted it cannot be + * de-restricted, and must be closed and re-opened. Event channels + * which were bound before restricting remain bound afterwards, and + * can be notified as usual. + */ +#define IOCTL_EVTCHN_RESTRICT_DOMID \ + _IOC(_IOC_NONE, 'E', 6, sizeof(struct ioctl_evtchn_restrict_domid)) +struct ioctl_evtchn_restrict_domid { + domid_t domid; +}; + #endif /* __LINUX_PUBLIC_EVTCHN_H__ */ -- cgit v1.2.3 From 2ccbe2cb79f2f74ab739252299b6f9ff27586f2c Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 22 Jul 2016 15:07:56 +0200 Subject: macsec: limit ICV length to 16 octets IEEE 802.1AE-2006 standard recommends that the ICV element in a MACsec frame should not exceed 16 octets: add MACSEC_STD_ICV_LEN in uapi definitions accordingly, and avoid accepting configurations where the ICV length exceeds the standard value. Leave definition of MACSEC_MAX_ICV_LEN unchanged for backwards compatibility with userspace programs. Fixes: dece8d2b78d1 ("uapi: add MACsec bits") Signed-off-by: Davide Caratti Signed-off-by: David S. Miller --- drivers/net/macsec.c | 4 ++-- include/uapi/linux/if_macsec.h | 2 ++ 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 0cbb935078da..18cfb46c5911 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -510,7 +510,7 @@ static bool macsec_validate_skb(struct sk_buff *skb, u16 icv_len) } #define MACSEC_NEEDED_HEADROOM (macsec_extra_len(true)) -#define MACSEC_NEEDED_TAILROOM MACSEC_MAX_ICV_LEN +#define MACSEC_NEEDED_TAILROOM MACSEC_STD_ICV_LEN static void macsec_fill_iv(unsigned char *iv, sci_t sci, u32 pn) { @@ -3217,7 +3217,7 @@ static int macsec_validate_attr(struct nlattr *tb[], struct nlattr *data[]) case MACSEC_DEFAULT_CIPHER_ID: case MACSEC_DEFAULT_CIPHER_ALT: if (icv_len < MACSEC_MIN_ICV_LEN || - icv_len > MACSEC_MAX_ICV_LEN) + icv_len > MACSEC_STD_ICV_LEN) return -EINVAL; break; default: diff --git a/include/uapi/linux/if_macsec.h b/include/uapi/linux/if_macsec.h index f7d4831a2cc7..02fc49cb72d8 100644 --- a/include/uapi/linux/if_macsec.h +++ b/include/uapi/linux/if_macsec.h @@ -26,6 +26,8 @@ #define MACSEC_MIN_ICV_LEN 8 #define MACSEC_MAX_ICV_LEN 32 +/* upper limit for ICV length as recommended by IEEE802.1AE-2006 */ +#define MACSEC_STD_ICV_LEN 16 enum macsec_attrs { MACSEC_ATTR_UNSPEC, -- cgit v1.2.3 From 96ae52279594470622ff0585621a13e96b700600 Mon Sep 17 00:00:00 2001 From: Sargun Dhillon Date: Mon, 25 Jul 2016 05:54:46 -0700 Subject: bpf: Add bpf_probe_write_user BPF helper to be called in tracers This allows user memory to be written to during the course of a kprobe. It shouldn't be used to implement any kind of security mechanism because of TOC-TOU attacks, but rather to debug, divert, and manipulate execution of semi-cooperative processes. Although it uses probe_kernel_write, we limit the address space the probe can write into by checking the space with access_ok. We do this as opposed to calling copy_to_user directly, in order to avoid sleeping. In addition we ensure the threads's current fs / segment is USER_DS and the thread isn't exiting nor a kernel thread. Given this feature is meant for experiments, and it has a risk of crashing the system, and running programs, we print a warning on when a proglet that attempts to use this helper is installed, along with the pid and process name. Signed-off-by: Sargun Dhillon Cc: Alexei Starovoitov Cc: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/uapi/linux/bpf.h | 10 ++++++++++ kernel/trace/bpf_trace.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ samples/bpf/bpf_helpers.h | 2 ++ 3 files changed, 57 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 2b7076f5b5ad..da218fec6056 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -365,6 +365,16 @@ enum bpf_func_id { */ BPF_FUNC_get_current_task, + /** + * bpf_probe_write_user(void *dst, void *src, int len) + * safely attempt to write to a location + * @dst: destination address in userspace + * @src: source address on stack + * @len: number of bytes to copy + * Return: 0 on success or negative error + */ + BPF_FUNC_probe_write_user, + __BPF_FUNC_MAX_ID, }; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index a12bbd32c0a6..b20438fdb029 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -81,6 +81,49 @@ static const struct bpf_func_proto bpf_probe_read_proto = { .arg3_type = ARG_ANYTHING, }; +static u64 bpf_probe_write_user(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + void *unsafe_ptr = (void *) (long) r1; + void *src = (void *) (long) r2; + int size = (int) r3; + + /* + * Ensure we're in user context which is safe for the helper to + * run. This helper has no business in a kthread. + * + * access_ok() should prevent writing to non-user memory, but in + * some situations (nommu, temporary switch, etc) access_ok() does + * not provide enough validation, hence the check on KERNEL_DS. + */ + + if (unlikely(in_interrupt() || + current->flags & (PF_KTHREAD | PF_EXITING))) + return -EPERM; + if (unlikely(segment_eq(get_fs(), KERNEL_DS))) + return -EPERM; + if (!access_ok(VERIFY_WRITE, unsafe_ptr, size)) + return -EPERM; + + return probe_kernel_write(unsafe_ptr, src, size); +} + +static const struct bpf_func_proto bpf_probe_write_user_proto = { + .func = bpf_probe_write_user, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_ANYTHING, + .arg2_type = ARG_PTR_TO_STACK, + .arg3_type = ARG_CONST_STACK_SIZE, +}; + +static const struct bpf_func_proto *bpf_get_probe_write_proto(void) +{ + pr_warn_ratelimited("%s[%d] is installing a program with bpf_probe_write_user helper that may corrupt user memory!", + current->comm, task_pid_nr(current)); + + return &bpf_probe_write_user_proto; +} + /* * limited trace_printk() * only %d %u %x %ld %lu %lx %lld %llu %llx %p %s conversion specifiers allowed @@ -362,6 +405,8 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) return &bpf_get_smp_processor_id_proto; case BPF_FUNC_perf_event_read: return &bpf_perf_event_read_proto; + case BPF_FUNC_probe_write_user: + return bpf_get_probe_write_proto(); default: return NULL; } diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h index 84e3fd919a06..217c8d507f2e 100644 --- a/samples/bpf/bpf_helpers.h +++ b/samples/bpf/bpf_helpers.h @@ -41,6 +41,8 @@ static int (*bpf_perf_event_output)(void *ctx, void *map, int index, void *data, (void *) BPF_FUNC_perf_event_output; static int (*bpf_get_stackid)(void *ctx, void *map, int flags) = (void *) BPF_FUNC_get_stackid; +static int (*bpf_probe_write_user)(void *dst, void *src, int size) = + (void *) BPF_FUNC_probe_write_user; /* llvm builtin functions that eBPF C program may use to * emit BPF_LD_ABS and BPF_LD_IND instructions -- cgit v1.2.3 From 7af7c616fa2f1ce6c0d806b89898d2df098b4bd8 Mon Sep 17 00:00:00 2001 From: Hans van Kranenburg Date: Sun, 3 Jul 2016 23:23:06 +0200 Subject: Btrfs: use the correct struct for BTRFS_IOC_LOGICAL_INO BTRFS_IOC_LOGICAL_INO takes a btrfs_ioctl_logical_ino_args as argument, not a btrfs_ioctl_ino_path_args. The lines were probably copy/pasted when the code was written. Since btrfs_ioctl_logical_ino_args and btrfs_ioctl_ino_path_args have the same size, the actual IOCTL definition here does not change. But, it makes the code less confusing for the reader. Signed-off-by: Hans van Kranenburg Signed-off-by: David Sterba --- include/uapi/linux/btrfs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index 2bdd1e3e7007..ac5eacd3055b 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -798,7 +798,7 @@ static inline char *btrfs_err_str(enum btrfs_err_code err_code) #define BTRFS_IOC_INO_PATHS _IOWR(BTRFS_IOCTL_MAGIC, 35, \ struct btrfs_ioctl_ino_path_args) #define BTRFS_IOC_LOGICAL_INO _IOWR(BTRFS_IOCTL_MAGIC, 36, \ - struct btrfs_ioctl_ino_path_args) + struct btrfs_ioctl_logical_ino_args) #define BTRFS_IOC_SET_RECEIVED_SUBVOL _IOWR(BTRFS_IOCTL_MAGIC, 37, \ struct btrfs_ioctl_received_subvol_args) #define BTRFS_IOC_SEND _IOW(BTRFS_IOCTL_MAGIC, 38, struct btrfs_ioctl_send_args) -- cgit v1.2.3 From 9ff26e9fabaf52f28fb5e875c0b9ffc2d1512039 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:18 +0200 Subject: tipc: introduce constants for tipc address validation In this commit, we introduce defines for tipc address size, offset and mask specification for Zone.Cluster.Node. There is no functional change in this commit. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- include/uapi/linux/tipc.h | 30 ++++++++++++++++++++++++++---- net/tipc/addr.h | 5 +---- net/tipc/bearer.c | 4 ++-- 3 files changed, 29 insertions(+), 10 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/tipc.h b/include/uapi/linux/tipc.h index 6f71b9b41595..bf049e8fe31b 100644 --- a/include/uapi/linux/tipc.h +++ b/include/uapi/linux/tipc.h @@ -60,26 +60,48 @@ struct tipc_name_seq { __u32 upper; }; +/* TIPC Address Size, Offset, Mask specification for Z.C.N + */ +#define TIPC_NODE_BITS 12 +#define TIPC_CLUSTER_BITS 12 +#define TIPC_ZONE_BITS 8 + +#define TIPC_NODE_OFFSET 0 +#define TIPC_CLUSTER_OFFSET TIPC_NODE_BITS +#define TIPC_ZONE_OFFSET (TIPC_CLUSTER_OFFSET + TIPC_CLUSTER_BITS) + +#define TIPC_NODE_SIZE ((1UL << TIPC_NODE_BITS) - 1) +#define TIPC_CLUSTER_SIZE ((1UL << TIPC_CLUSTER_BITS) - 1) +#define TIPC_ZONE_SIZE ((1UL << TIPC_ZONE_BITS) - 1) + +#define TIPC_NODE_MASK (TIPC_NODE_SIZE << TIPC_NODE_OFFSET) +#define TIPC_CLUSTER_MASK (TIPC_CLUSTER_SIZE << TIPC_CLUSTER_OFFSET) +#define TIPC_ZONE_MASK (TIPC_ZONE_SIZE << TIPC_ZONE_OFFSET) + +#define TIPC_ZONE_CLUSTER_MASK (TIPC_ZONE_MASK | TIPC_CLUSTER_MASK) + static inline __u32 tipc_addr(unsigned int zone, unsigned int cluster, unsigned int node) { - return (zone << 24) | (cluster << 12) | node; + return (zone << TIPC_ZONE_OFFSET) | + (cluster << TIPC_CLUSTER_OFFSET) | + node; } static inline unsigned int tipc_zone(__u32 addr) { - return addr >> 24; + return addr >> TIPC_ZONE_OFFSET; } static inline unsigned int tipc_cluster(__u32 addr) { - return (addr >> 12) & 0xfff; + return (addr & TIPC_CLUSTER_MASK) >> TIPC_CLUSTER_OFFSET; } static inline unsigned int tipc_node(__u32 addr) { - return addr & 0xfff; + return addr & TIPC_NODE_MASK; } /* diff --git a/net/tipc/addr.h b/net/tipc/addr.h index 64f4004a6fac..bebb347803ce 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -43,9 +43,6 @@ #include #include "core.h" -#define TIPC_ZONE_MASK 0xff000000u -#define TIPC_CLUSTER_MASK 0xfffff000u - static inline u32 tipc_own_addr(struct net *net) { struct tipc_net *tn = net_generic(net, tipc_net_id); @@ -60,7 +57,7 @@ static inline u32 tipc_zone_mask(u32 addr) static inline u32 tipc_cluster_mask(u32 addr) { - return addr & TIPC_CLUSTER_MASK; + return addr & TIPC_ZONE_CLUSTER_MASK; } u32 tipc_own_addr(struct net *net); diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 4131d5a86f55..65b0998a9bab 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -225,7 +225,7 @@ static int tipc_enable_bearer(struct net *net, const char *name, if (tipc_addr_domain_valid(disc_domain) && (disc_domain != tn->own_addr)) { if (tipc_in_scope(disc_domain, tn->own_addr)) { - disc_domain = tn->own_addr & TIPC_CLUSTER_MASK; + disc_domain = tn->own_addr & TIPC_ZONE_CLUSTER_MASK; res = 0; /* accept any node in own cluster */ } else if (in_own_cluster_exact(net, disc_domain)) res = 0; /* accept specified node in own cluster */ @@ -832,7 +832,7 @@ int tipc_nl_bearer_enable(struct sk_buff *skb, struct genl_info *info) u32 prio; prio = TIPC_MEDIA_LINK_PRI; - domain = tn->own_addr & TIPC_CLUSTER_MASK; + domain = tn->own_addr & TIPC_ZONE_CLUSTER_MASK; if (!info->attrs[TIPC_NLA_BEARER]) return -EINVAL; -- cgit v1.2.3 From 7b3f52296493656015f0c0deddb6e90e36b9cda2 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:19 +0200 Subject: tipc: make cluster size threshold for monitoring configurable In this commit, we introduce support to configure the minimum threshold to activate the new link monitoring algorithm. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- include/uapi/linux/tipc_netlink.h | 11 +++++++++++ net/tipc/monitor.c | 12 ++++++++++++ net/tipc/monitor.h | 1 + net/tipc/netlink.c | 15 +++++++++++++-- net/tipc/netlink.h | 1 + net/tipc/node.c | 27 +++++++++++++++++++++++++++ net/tipc/node.h | 1 + 7 files changed, 66 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h index d4c8f142ba63..d387b65a0d97 100644 --- a/include/uapi/linux/tipc_netlink.h +++ b/include/uapi/linux/tipc_netlink.h @@ -56,6 +56,7 @@ enum { TIPC_NL_NET_GET, TIPC_NL_NET_SET, TIPC_NL_NAME_TABLE_GET, + TIPC_NL_MON_SET, __TIPC_NL_CMD_MAX, TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1 @@ -72,6 +73,7 @@ enum { TIPC_NLA_NODE, /* nest */ TIPC_NLA_NET, /* nest */ TIPC_NLA_NAME_TABLE, /* nest */ + TIPC_NLA_MON, /* nest */ __TIPC_NLA_MAX, TIPC_NLA_MAX = __TIPC_NLA_MAX - 1 @@ -166,6 +168,15 @@ enum { TIPC_NLA_NAME_TABLE_MAX = __TIPC_NLA_NAME_TABLE_MAX - 1 }; +/* Monitor info */ +enum { + TIPC_NLA_MON_UNSPEC, + TIPC_NLA_MON_ACTIVATION_THRESHOLD, /* u32 */ + + __TIPC_NLA_MON_MAX, + TIPC_NLA_MON_MAX = __TIPC_NLA_MON_MAX - 1 +}; + /* Publication info */ enum { TIPC_NLA_PUBL_UNSPEC, diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 0d489e81fcca..3892d05b8b45 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -649,3 +649,15 @@ void tipc_mon_delete(struct net *net, int bearer_id) kfree(self); kfree(mon); } + +int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size) +{ + struct tipc_net *tn = tipc_net(net); + + if (cluster_size > TIPC_CLUSTER_SIZE) + return -EINVAL; + + tn->mon_threshold = cluster_size; + + return 0; +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h index 598459cbed5d..91f5dd09432b 100644 --- a/net/tipc/monitor.h +++ b/net/tipc/monitor.h @@ -69,5 +69,6 @@ void tipc_mon_get_state(struct net *net, u32 addr, int bearer_id); void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); +int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size); extern const int tipc_max_domain_size; #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 56935df2167a..1e43ac0200ed 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -52,7 +52,8 @@ static const struct nla_policy tipc_nl_policy[TIPC_NLA_MAX + 1] = { [TIPC_NLA_MEDIA] = { .type = NLA_NESTED, }, [TIPC_NLA_NODE] = { .type = NLA_NESTED, }, [TIPC_NLA_NET] = { .type = NLA_NESTED, }, - [TIPC_NLA_NAME_TABLE] = { .type = NLA_NESTED, } + [TIPC_NLA_NAME_TABLE] = { .type = NLA_NESTED, }, + [TIPC_NLA_MON] = { .type = NLA_NESTED, }, }; const struct nla_policy @@ -61,6 +62,11 @@ tipc_nl_name_table_policy[TIPC_NLA_NAME_TABLE_MAX + 1] = { [TIPC_NLA_NAME_TABLE_PUBL] = { .type = NLA_NESTED } }; +const struct nla_policy tipc_nl_monitor_policy[TIPC_NLA_MON_MAX + 1] = { + [TIPC_NLA_MON_UNSPEC] = { .type = NLA_UNSPEC }, + [TIPC_NLA_MON_ACTIVATION_THRESHOLD] = { .type = NLA_U32 }, +}; + const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = { [TIPC_NLA_SOCK_UNSPEC] = { .type = NLA_UNSPEC }, [TIPC_NLA_SOCK_ADDR] = { .type = NLA_U32 }, @@ -214,7 +220,12 @@ static const struct genl_ops tipc_genl_v2_ops[] = { .cmd = TIPC_NL_NAME_TABLE_GET, .dumpit = tipc_nl_name_table_dump, .policy = tipc_nl_policy, - } + }, + { + .cmd = TIPC_NL_MON_SET, + .doit = tipc_nl_node_set_monitor, + .policy = tipc_nl_policy, + }, }; int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***attr) diff --git a/net/tipc/netlink.h b/net/tipc/netlink.h index ed1dbcb4afbd..4ba0ad422110 100644 --- a/net/tipc/netlink.h +++ b/net/tipc/netlink.h @@ -55,6 +55,7 @@ extern const struct nla_policy tipc_nl_prop_policy[]; extern const struct nla_policy tipc_nl_bearer_policy[]; extern const struct nla_policy tipc_nl_media_policy[]; extern const struct nla_policy tipc_nl_udp_policy[]; +extern const struct nla_policy tipc_nl_monitor_policy[]; int tipc_netlink_start(void); int tipc_netlink_compat_start(void); diff --git a/net/tipc/node.c b/net/tipc/node.c index 95cc78b51532..0fc531d0f709 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1928,3 +1928,30 @@ out: return skb->len; } + +int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info) +{ + struct nlattr *attrs[TIPC_NLA_MON_MAX + 1]; + struct net *net = sock_net(skb->sk); + int err; + + if (!info->attrs[TIPC_NLA_MON]) + return -EINVAL; + + err = nla_parse_nested(attrs, TIPC_NLA_MON_MAX, + info->attrs[TIPC_NLA_MON], + tipc_nl_monitor_policy); + if (err) + return err; + + if (attrs[TIPC_NLA_MON_ACTIVATION_THRESHOLD]) { + u32 val; + + val = nla_get_u32(attrs[TIPC_NLA_MON_ACTIVATION_THRESHOLD]); + err = tipc_nl_monitor_set_threshold(net, val); + if (err) + return err; + } + + return 0; +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 8264b3d97dc4..65aa12ede8a5 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -78,4 +78,5 @@ int tipc_nl_node_reset_link_stats(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info); #endif -- cgit v1.2.3 From bf1035b2ff5296c7c49e262152253ce29d87e82d Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:20 +0200 Subject: tipc: get monitor threshold for the cluster In this commit, we add support to fetch the configured cluster monitoring threshold. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- include/uapi/linux/tipc_netlink.h | 1 + net/tipc/monitor.c | 7 ++++++ net/tipc/monitor.h | 2 ++ net/tipc/netlink.c | 5 ++++ net/tipc/node.c | 52 +++++++++++++++++++++++++++++++++++++++ net/tipc/node.h | 1 + 6 files changed, 68 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h index d387b65a0d97..d07c6ec76062 100644 --- a/include/uapi/linux/tipc_netlink.h +++ b/include/uapi/linux/tipc_netlink.h @@ -57,6 +57,7 @@ enum { TIPC_NL_NET_SET, TIPC_NL_NAME_TABLE_GET, TIPC_NL_MON_SET, + TIPC_NL_MON_GET, __TIPC_NL_CMD_MAX, TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1 diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 3892d05b8b45..3579126e2ac8 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -661,3 +661,10 @@ int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size) return 0; } + +int tipc_nl_monitor_get_threshold(struct net *net) +{ + struct tipc_net *tn = tipc_net(net); + + return tn->mon_threshold; +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h index 91f5dd09432b..aedf62c60bd3 100644 --- a/net/tipc/monitor.h +++ b/net/tipc/monitor.h @@ -70,5 +70,7 @@ void tipc_mon_get_state(struct net *net, u32 addr, void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size); +int tipc_nl_monitor_get_threshold(struct net *net); + extern const int tipc_max_domain_size; #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 1e43ac0200ed..2cfc5f7c6380 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -226,6 +226,11 @@ static const struct genl_ops tipc_genl_v2_ops[] = { .doit = tipc_nl_node_set_monitor, .policy = tipc_nl_policy, }, + { + .cmd = TIPC_NL_MON_GET, + .doit = tipc_nl_node_get_monitor, + .policy = tipc_nl_policy, + }, }; int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***attr) diff --git a/net/tipc/node.c b/net/tipc/node.c index 0fc531d0f709..2a7e74753f9f 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1955,3 +1955,55 @@ int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info) return 0; } + +static int __tipc_nl_add_monitor_prop(struct net *net, struct tipc_nl_msg *msg) +{ + struct nlattr *attrs; + void *hdr; + u32 val; + + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + 0, TIPC_NL_MON_GET); + if (!hdr) + return -EMSGSIZE; + + attrs = nla_nest_start(msg->skb, TIPC_NLA_MON); + if (!attrs) + goto msg_full; + + val = tipc_nl_monitor_get_threshold(net); + + if (nla_put_u32(msg->skb, TIPC_NLA_MON_ACTIVATION_THRESHOLD, val)) + goto attr_msg_full; + + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); + + return 0; + +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + genlmsg_cancel(msg->skb, hdr); + + return -EMSGSIZE; +} + +int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = sock_net(skb->sk); + struct tipc_nl_msg msg; + int err; + + msg.skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + msg.portid = info->snd_portid; + msg.seq = info->snd_seq; + + err = __tipc_nl_add_monitor_prop(net, &msg); + if (err) { + nlmsg_free(msg.skb); + return err; + } + + return genlmsg_reply(msg.skb, info); +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 65aa12ede8a5..216f053b817f 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -79,4 +79,5 @@ int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info); #endif -- cgit v1.2.3 From cf6f7e1d51090772d5ff7355aaf0fcff17f20d1a Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:22 +0200 Subject: tipc: dump monitor attributes In this commit, we dump the monitor attributes when queried. The link monitor attributes are separated into two kinds: 1. general attributes per bearer 2. specific attributes per node/peer This style resembles the socket attributes and the nametable publications per socket. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller --- include/uapi/linux/tipc_netlink.h | 25 +++++++ net/tipc/monitor.c | 133 ++++++++++++++++++++++++++++++++++++++ net/tipc/monitor.h | 6 ++ net/tipc/netlink.c | 7 ++ net/tipc/node.c | 86 ++++++++++++++++++++++++ net/tipc/node.h | 3 + 6 files changed, 260 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h index d07c6ec76062..5f3f6d09fb79 100644 --- a/include/uapi/linux/tipc_netlink.h +++ b/include/uapi/linux/tipc_netlink.h @@ -58,6 +58,7 @@ enum { TIPC_NL_NAME_TABLE_GET, TIPC_NL_MON_SET, TIPC_NL_MON_GET, + TIPC_NL_MON_PEER_GET, __TIPC_NL_CMD_MAX, TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1 @@ -75,6 +76,7 @@ enum { TIPC_NLA_NET, /* nest */ TIPC_NLA_NAME_TABLE, /* nest */ TIPC_NLA_MON, /* nest */ + TIPC_NLA_MON_PEER, /* nest */ __TIPC_NLA_MAX, TIPC_NLA_MAX = __TIPC_NLA_MAX - 1 @@ -173,6 +175,11 @@ enum { enum { TIPC_NLA_MON_UNSPEC, TIPC_NLA_MON_ACTIVATION_THRESHOLD, /* u32 */ + TIPC_NLA_MON_REF, /* u32 */ + TIPC_NLA_MON_ACTIVE, /* flag */ + TIPC_NLA_MON_BEARER_NAME, /* string */ + TIPC_NLA_MON_PEERCNT, /* u32 */ + TIPC_NLA_MON_LISTGEN, /* u32 */ __TIPC_NLA_MON_MAX, TIPC_NLA_MON_MAX = __TIPC_NLA_MON_MAX - 1 @@ -194,6 +201,24 @@ enum { TIPC_NLA_PUBL_MAX = __TIPC_NLA_PUBL_MAX - 1 }; +/* Monitor peer info */ +enum { + TIPC_NLA_MON_PEER_UNSPEC, + + TIPC_NLA_MON_PEER_ADDR, /* u32 */ + TIPC_NLA_MON_PEER_DOMGEN, /* u32 */ + TIPC_NLA_MON_PEER_APPLIED, /* u32 */ + TIPC_NLA_MON_PEER_UPMAP, /* u64 */ + TIPC_NLA_MON_PEER_MEMBERS, /* tlv */ + TIPC_NLA_MON_PEER_UP, /* flag */ + TIPC_NLA_MON_PEER_HEAD, /* flag */ + TIPC_NLA_MON_PEER_LOCAL, /* flag */ + TIPC_NLA_MON_PEER_PAD, /* flag */ + + __TIPC_NLA_MON_PEER_MAX, + TIPC_NLA_MON_PEER_MAX = __TIPC_NLA_MON_PEER_MAX - 1 +}; + /* Nest, connection info */ enum { TIPC_NLA_CON_UNSPEC, diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 3579126e2ac8..be70a57c1ff9 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -33,9 +33,11 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#include #include "core.h" #include "addr.h" #include "monitor.h" +#include "bearer.h" #define MAX_MON_DOMAIN 64 #define MON_TIMEOUT 120000 @@ -668,3 +670,134 @@ int tipc_nl_monitor_get_threshold(struct net *net) return tn->mon_threshold; } + +int __tipc_nl_add_monitor_peer(struct tipc_peer *peer, struct tipc_nl_msg *msg) +{ + struct tipc_mon_domain *dom = peer->domain; + struct nlattr *attrs; + void *hdr; + + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + NLM_F_MULTI, TIPC_NL_MON_PEER_GET); + if (!hdr) + return -EMSGSIZE; + + attrs = nla_nest_start(msg->skb, TIPC_NLA_MON_PEER); + if (!attrs) + goto msg_full; + + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEER_ADDR, peer->addr)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEER_APPLIED, peer->applied)) + goto attr_msg_full; + + if (peer->is_up) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_PEER_UP)) + goto attr_msg_full; + if (peer->is_local) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_PEER_LOCAL)) + goto attr_msg_full; + if (peer->is_head) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_PEER_HEAD)) + goto attr_msg_full; + + if (dom) { + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEER_DOMGEN, dom->gen)) + goto attr_msg_full; + if (nla_put_u64_64bit(msg->skb, TIPC_NLA_MON_PEER_UPMAP, + dom->up_map, TIPC_NLA_MON_PEER_PAD)) + goto attr_msg_full; + if (nla_put(msg->skb, TIPC_NLA_MON_PEER_MEMBERS, + dom->member_cnt * sizeof(u32), &dom->members)) + goto attr_msg_full; + } + + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); + return 0; + +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + genlmsg_cancel(msg->skb, hdr); + + return -EMSGSIZE; +} + +int tipc_nl_add_monitor_peer(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id, u32 *prev_node) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *peer = mon->self; + + if (!mon) + return -EINVAL; + + read_lock_bh(&mon->lock); + do { + if (*prev_node) { + if (peer->addr == *prev_node) + *prev_node = 0; + else + continue; + } + if (__tipc_nl_add_monitor_peer(peer, msg)) { + *prev_node = peer->addr; + read_unlock_bh(&mon->lock); + return -EMSGSIZE; + } + } while ((peer = peer_nxt(peer)) != mon->self); + read_unlock_bh(&mon->lock); + + return 0; +} + +int __tipc_nl_add_monitor(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + char bearer_name[TIPC_MAX_BEARER_NAME]; + struct nlattr *attrs; + void *hdr; + int ret; + + ret = tipc_bearer_get_name(net, bearer_name, bearer_id); + if (ret || !mon) + return -EINVAL; + + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + NLM_F_MULTI, TIPC_NL_MON_GET); + if (!hdr) + return -EMSGSIZE; + + attrs = nla_nest_start(msg->skb, TIPC_NLA_MON); + if (!attrs) + goto msg_full; + + read_lock_bh(&mon->lock); + if (nla_put_u32(msg->skb, TIPC_NLA_MON_REF, bearer_id)) + goto attr_msg_full; + if (tipc_mon_is_active(net, mon)) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_ACTIVE)) + goto attr_msg_full; + if (nla_put_string(msg->skb, TIPC_NLA_MON_BEARER_NAME, bearer_name)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEERCNT, mon->peer_cnt)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_MON_LISTGEN, mon->list_gen)) + goto attr_msg_full; + + read_unlock_bh(&mon->lock); + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); + + return 0; + +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + genlmsg_cancel(msg->skb, hdr); + read_unlock_bh(&mon->lock); + + return -EMSGSIZE; +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h index aedf62c60bd3..2a21b93e0d04 100644 --- a/net/tipc/monitor.h +++ b/net/tipc/monitor.h @@ -36,6 +36,8 @@ #ifndef _TIPC_MONITOR_H #define _TIPC_MONITOR_H +#include "netlink.h" + /* struct tipc_mon_state: link instance's cache of monitor list and domain state * @list_gen: current generation of this node's monitor list * @gen: current generation of this node's local domain @@ -71,6 +73,10 @@ void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size); int tipc_nl_monitor_get_threshold(struct net *net); +int __tipc_nl_add_monitor(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id); +int tipc_nl_add_monitor_peer(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id, u32 *prev_node); extern const int tipc_max_domain_size; #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 2cfc5f7c6380..a84daec0afe9 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -64,6 +64,7 @@ tipc_nl_name_table_policy[TIPC_NLA_NAME_TABLE_MAX + 1] = { const struct nla_policy tipc_nl_monitor_policy[TIPC_NLA_MON_MAX + 1] = { [TIPC_NLA_MON_UNSPEC] = { .type = NLA_UNSPEC }, + [TIPC_NLA_MON_REF] = { .type = NLA_U32 }, [TIPC_NLA_MON_ACTIVATION_THRESHOLD] = { .type = NLA_U32 }, }; @@ -229,6 +230,12 @@ static const struct genl_ops tipc_genl_v2_ops[] = { { .cmd = TIPC_NL_MON_GET, .doit = tipc_nl_node_get_monitor, + .dumpit = tipc_nl_node_dump_monitor, + .policy = tipc_nl_policy, + }, + { + .cmd = TIPC_NL_MON_PEER_GET, + .dumpit = tipc_nl_node_dump_monitor_peer, .policy = tipc_nl_policy, }, }; diff --git a/net/tipc/node.c b/net/tipc/node.c index 2a7e74753f9f..21974191e425 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -2007,3 +2007,89 @@ int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info) return genlmsg_reply(msg.skb, info); } + +int tipc_nl_node_dump_monitor(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + u32 prev_bearer = cb->args[0]; + struct tipc_nl_msg msg; + int err; + int i; + + if (prev_bearer == MAX_BEARERS) + return 0; + + msg.skb = skb; + msg.portid = NETLINK_CB(cb->skb).portid; + msg.seq = cb->nlh->nlmsg_seq; + + rtnl_lock(); + for (i = prev_bearer; i < MAX_BEARERS; i++) { + prev_bearer = i; + err = __tipc_nl_add_monitor(net, &msg, prev_bearer); + if (err) + goto out; + } + +out: + rtnl_unlock(); + cb->args[0] = prev_bearer; + + return skb->len; +} + +int tipc_nl_node_dump_monitor_peer(struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + u32 prev_node = cb->args[1]; + u32 bearer_id = cb->args[2]; + int done = cb->args[0]; + struct tipc_nl_msg msg; + int err; + + if (!prev_node) { + struct nlattr **attrs; + struct nlattr *mon[TIPC_NLA_MON_MAX + 1]; + + err = tipc_nlmsg_parse(cb->nlh, &attrs); + if (err) + return err; + + if (!attrs[TIPC_NLA_MON]) + return -EINVAL; + + err = nla_parse_nested(mon, TIPC_NLA_MON_MAX, + attrs[TIPC_NLA_MON], + tipc_nl_monitor_policy); + if (err) + return err; + + if (!mon[TIPC_NLA_MON_REF]) + return -EINVAL; + + bearer_id = nla_get_u32(mon[TIPC_NLA_MON_REF]); + + if (bearer_id >= MAX_BEARERS) + return -EINVAL; + } + + if (done) + return 0; + + msg.skb = skb; + msg.portid = NETLINK_CB(cb->skb).portid; + msg.seq = cb->nlh->nlmsg_seq; + + rtnl_lock(); + err = tipc_nl_add_monitor_peer(net, &msg, bearer_id, &prev_node); + if (!err) + done = 1; + + rtnl_unlock(); + cb->args[0] = done; + cb->args[1] = prev_node; + cb->args[2] = bearer_id; + + return skb->len; +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 216f053b817f..d69fdfcc0ec9 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -80,4 +80,7 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_dump_monitor(struct sk_buff *skb, struct netlink_callback *cb); +int tipc_nl_node_dump_monitor_peer(struct sk_buff *skb, + struct netlink_callback *cb); #endif -- cgit v1.2.3 From b1123ea6d3b3da25af5c8a9d843bd07ab63213f4 Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Tue, 26 Jul 2016 15:23:09 -0700 Subject: mm: balloon: use general non-lru movable page feature Now, VM has a feature to migrate non-lru movable pages so balloon doesn't need custom migration hooks in migrate.c and compaction.c. Instead, this patch implements the page->mapping->a_ops-> {isolate|migrate|putback} functions. With that, we could remove hooks for ballooning in general migration functions and make balloon compaction simple. [akpm@linux-foundation.org: compaction.h requires that the includer first include node.h] Link: http://lkml.kernel.org/r/1464736881-24886-4-git-send-email-minchan@kernel.org Signed-off-by: Gioh Kim Signed-off-by: Minchan Kim Acked-by: Vlastimil Babka Cc: Rafael Aquini Cc: Konstantin Khlebnikov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/virtio/virtio_balloon.c | 54 +++++++++++++++++++--- include/linux/balloon_compaction.h | 54 +++++++--------------- include/uapi/linux/magic.h | 1 + mm/balloon_compaction.c | 94 +++++++------------------------------- mm/compaction.c | 7 --- mm/migrate.c | 19 +------- mm/vmscan.c | 2 +- 7 files changed, 86 insertions(+), 145 deletions(-) (limited to 'include/uapi') diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c index 476c0e3a7150..88d5609375de 100644 --- a/drivers/virtio/virtio_balloon.c +++ b/drivers/virtio/virtio_balloon.c @@ -30,6 +30,7 @@ #include #include #include +#include /* * Balloon device works in 4K page units. So each page is pointed to by @@ -45,6 +46,10 @@ static int oom_pages = OOM_VBALLOON_DEFAULT_PAGES; module_param(oom_pages, int, S_IRUSR | S_IWUSR); MODULE_PARM_DESC(oom_pages, "pages to free on OOM"); +#ifdef CONFIG_BALLOON_COMPACTION +static struct vfsmount *balloon_mnt; +#endif + struct virtio_balloon { struct virtio_device *vdev; struct virtqueue *inflate_vq, *deflate_vq, *stats_vq; @@ -488,8 +493,26 @@ static int virtballoon_migratepage(struct balloon_dev_info *vb_dev_info, put_page(page); /* balloon reference */ - return MIGRATEPAGE_SUCCESS; + return 0; } + +static struct dentry *balloon_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) +{ + static const struct dentry_operations ops = { + .d_dname = simple_dname, + }; + + return mount_pseudo(fs_type, "balloon-kvm:", NULL, &ops, + BALLOON_KVM_MAGIC); +} + +static struct file_system_type balloon_fs = { + .name = "balloon-kvm", + .mount = balloon_mount, + .kill_sb = kill_anon_super, +}; + #endif /* CONFIG_BALLOON_COMPACTION */ static int virtballoon_probe(struct virtio_device *vdev) @@ -519,9 +542,6 @@ static int virtballoon_probe(struct virtio_device *vdev) vb->vdev = vdev; balloon_devinfo_init(&vb->vb_dev_info); -#ifdef CONFIG_BALLOON_COMPACTION - vb->vb_dev_info.migratepage = virtballoon_migratepage; -#endif err = init_vqs(vb); if (err) @@ -531,13 +551,33 @@ static int virtballoon_probe(struct virtio_device *vdev) vb->nb.priority = VIRTBALLOON_OOM_NOTIFY_PRIORITY; err = register_oom_notifier(&vb->nb); if (err < 0) - goto out_oom_notify; + goto out_del_vqs; + +#ifdef CONFIG_BALLOON_COMPACTION + balloon_mnt = kern_mount(&balloon_fs); + if (IS_ERR(balloon_mnt)) { + err = PTR_ERR(balloon_mnt); + unregister_oom_notifier(&vb->nb); + goto out_del_vqs; + } + + vb->vb_dev_info.migratepage = virtballoon_migratepage; + vb->vb_dev_info.inode = alloc_anon_inode(balloon_mnt->mnt_sb); + if (IS_ERR(vb->vb_dev_info.inode)) { + err = PTR_ERR(vb->vb_dev_info.inode); + kern_unmount(balloon_mnt); + unregister_oom_notifier(&vb->nb); + vb->vb_dev_info.inode = NULL; + goto out_del_vqs; + } + vb->vb_dev_info.inode->i_mapping->a_ops = &balloon_aops; +#endif virtio_device_ready(vdev); return 0; -out_oom_notify: +out_del_vqs: vdev->config->del_vqs(vdev); out_free_vb: kfree(vb); @@ -571,6 +611,8 @@ static void virtballoon_remove(struct virtio_device *vdev) cancel_work_sync(&vb->update_balloon_stats_work); remove_common(vb); + if (vb->vb_dev_info.inode) + iput(vb->vb_dev_info.inode); kfree(vb); } diff --git a/include/linux/balloon_compaction.h b/include/linux/balloon_compaction.h index 9b0a15d06a4f..504bd724e6ab 100644 --- a/include/linux/balloon_compaction.h +++ b/include/linux/balloon_compaction.h @@ -45,9 +45,11 @@ #define _LINUX_BALLOON_COMPACTION_H #include #include -#include +#include +#include #include #include +#include /* * Balloon device information descriptor. @@ -62,6 +64,7 @@ struct balloon_dev_info { struct list_head pages; /* Pages enqueued & handled to Host */ int (*migratepage)(struct balloon_dev_info *, struct page *newpage, struct page *page, enum migrate_mode mode); + struct inode *inode; }; extern struct page *balloon_page_enqueue(struct balloon_dev_info *b_dev_info); @@ -73,44 +76,18 @@ static inline void balloon_devinfo_init(struct balloon_dev_info *balloon) spin_lock_init(&balloon->pages_lock); INIT_LIST_HEAD(&balloon->pages); balloon->migratepage = NULL; + balloon->inode = NULL; } #ifdef CONFIG_BALLOON_COMPACTION -extern bool balloon_page_isolate(struct page *page); +extern const struct address_space_operations balloon_aops; +extern bool balloon_page_isolate(struct page *page, + isolate_mode_t mode); extern void balloon_page_putback(struct page *page); -extern int balloon_page_migrate(struct page *newpage, +extern int balloon_page_migrate(struct address_space *mapping, + struct page *newpage, struct page *page, enum migrate_mode mode); -/* - * __is_movable_balloon_page - helper to perform @page PageBalloon tests - */ -static inline bool __is_movable_balloon_page(struct page *page) -{ - return PageBalloon(page); -} - -/* - * balloon_page_movable - test PageBalloon to identify balloon pages - * and PagePrivate to check that the page is not - * isolated and can be moved by compaction/migration. - * - * As we might return false positives in the case of a balloon page being just - * released under us, this need to be re-tested later, under the page lock. - */ -static inline bool balloon_page_movable(struct page *page) -{ - return PageBalloon(page) && PagePrivate(page); -} - -/* - * isolated_balloon_page - identify an isolated balloon page on private - * compaction/migration page lists. - */ -static inline bool isolated_balloon_page(struct page *page) -{ - return PageBalloon(page); -} - /* * balloon_page_insert - insert a page into the balloon's page list and make * the page->private assignment accordingly. @@ -124,7 +101,7 @@ static inline void balloon_page_insert(struct balloon_dev_info *balloon, struct page *page) { __SetPageBalloon(page); - SetPagePrivate(page); + __SetPageMovable(page, balloon->inode->i_mapping); set_page_private(page, (unsigned long)balloon); list_add(&page->lru, &balloon->pages); } @@ -140,11 +117,14 @@ static inline void balloon_page_insert(struct balloon_dev_info *balloon, static inline void balloon_page_delete(struct page *page) { __ClearPageBalloon(page); + __ClearPageMovable(page); set_page_private(page, 0); - if (PagePrivate(page)) { - ClearPagePrivate(page); + /* + * No touch page.lru field once @page has been isolated + * because VM is using the field. + */ + if (!PageIsolated(page)) list_del(&page->lru); - } } /* diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h index 546b38886e11..d829ce63529d 100644 --- a/include/uapi/linux/magic.h +++ b/include/uapi/linux/magic.h @@ -80,5 +80,6 @@ #define BPF_FS_MAGIC 0xcafe4a11 /* Since UDF 2.01 is ISO 13346 based... */ #define UDF_SUPER_MAGIC 0x15013346 +#define BALLOON_KVM_MAGIC 0x13661366 #endif /* __LINUX_MAGIC_H__ */ diff --git a/mm/balloon_compaction.c b/mm/balloon_compaction.c index 57b3e9bd6bc5..da91df50ba31 100644 --- a/mm/balloon_compaction.c +++ b/mm/balloon_compaction.c @@ -70,7 +70,7 @@ struct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info) */ if (trylock_page(page)) { #ifdef CONFIG_BALLOON_COMPACTION - if (!PagePrivate(page)) { + if (PageIsolated(page)) { /* raced with isolation */ unlock_page(page); continue; @@ -106,110 +106,50 @@ EXPORT_SYMBOL_GPL(balloon_page_dequeue); #ifdef CONFIG_BALLOON_COMPACTION -static inline void __isolate_balloon_page(struct page *page) +bool balloon_page_isolate(struct page *page, isolate_mode_t mode) + { struct balloon_dev_info *b_dev_info = balloon_page_device(page); unsigned long flags; spin_lock_irqsave(&b_dev_info->pages_lock, flags); - ClearPagePrivate(page); list_del(&page->lru); b_dev_info->isolated_pages++; spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); + + return true; } -static inline void __putback_balloon_page(struct page *page) +void balloon_page_putback(struct page *page) { struct balloon_dev_info *b_dev_info = balloon_page_device(page); unsigned long flags; spin_lock_irqsave(&b_dev_info->pages_lock, flags); - SetPagePrivate(page); list_add(&page->lru, &b_dev_info->pages); b_dev_info->isolated_pages--; spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); } -/* __isolate_lru_page() counterpart for a ballooned page */ -bool balloon_page_isolate(struct page *page) -{ - /* - * Avoid burning cycles with pages that are yet under __free_pages(), - * or just got freed under us. - * - * In case we 'win' a race for a balloon page being freed under us and - * raise its refcount preventing __free_pages() from doing its job - * the put_page() at the end of this block will take care of - * release this page, thus avoiding a nasty leakage. - */ - if (likely(get_page_unless_zero(page))) { - /* - * As balloon pages are not isolated from LRU lists, concurrent - * compaction threads can race against page migration functions - * as well as race against the balloon driver releasing a page. - * - * In order to avoid having an already isolated balloon page - * being (wrongly) re-isolated while it is under migration, - * or to avoid attempting to isolate pages being released by - * the balloon driver, lets be sure we have the page lock - * before proceeding with the balloon page isolation steps. - */ - if (likely(trylock_page(page))) { - /* - * A ballooned page, by default, has PagePrivate set. - * Prevent concurrent compaction threads from isolating - * an already isolated balloon page by clearing it. - */ - if (balloon_page_movable(page)) { - __isolate_balloon_page(page); - unlock_page(page); - return true; - } - unlock_page(page); - } - put_page(page); - } - return false; -} - -/* putback_lru_page() counterpart for a ballooned page */ -void balloon_page_putback(struct page *page) -{ - /* - * 'lock_page()' stabilizes the page and prevents races against - * concurrent isolation threads attempting to re-isolate it. - */ - lock_page(page); - - if (__is_movable_balloon_page(page)) { - __putback_balloon_page(page); - /* drop the extra ref count taken for page isolation */ - put_page(page); - } else { - WARN_ON(1); - dump_page(page, "not movable balloon page"); - } - unlock_page(page); -} /* move_to_new_page() counterpart for a ballooned page */ -int balloon_page_migrate(struct page *newpage, - struct page *page, enum migrate_mode mode) +int balloon_page_migrate(struct address_space *mapping, + struct page *newpage, struct page *page, + enum migrate_mode mode) { struct balloon_dev_info *balloon = balloon_page_device(page); - int rc = -EAGAIN; VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(!PageLocked(newpage), newpage); - if (WARN_ON(!__is_movable_balloon_page(page))) { - dump_page(page, "not movable balloon page"); - return rc; - } + return balloon->migratepage(balloon, newpage, page, mode); +} - if (balloon && balloon->migratepage) - rc = balloon->migratepage(balloon, newpage, page, mode); +const struct address_space_operations balloon_aops = { + .migratepage = balloon_page_migrate, + .isolate_page = balloon_page_isolate, + .putback_page = balloon_page_putback, +}; +EXPORT_SYMBOL_GPL(balloon_aops); - return rc; -} #endif /* CONFIG_BALLOON_COMPACTION */ diff --git a/mm/compaction.c b/mm/compaction.c index fe95d8d021c3..d85520647d1d 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -791,13 +791,6 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, * Skip any other type of page */ if (!PageLRU(page)) { - if (unlikely(balloon_page_movable(page))) { - if (balloon_page_isolate(page)) { - /* Successfully isolated */ - goto isolate_success; - } - } - /* * __PageMovable can return false positive so we need * to verify it under page_lock. diff --git a/mm/migrate.c b/mm/migrate.c index 8119fdc563f8..f278005f609c 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -170,14 +170,12 @@ void putback_movable_pages(struct list_head *l) list_del(&page->lru); dec_zone_page_state(page, NR_ISOLATED_ANON + page_is_file_cache(page)); - if (unlikely(isolated_balloon_page(page))) { - balloon_page_putback(page); /* * We isolated non-lru movable page so here we can use * __PageMovable because LRU page's mapping cannot have * PAGE_MAPPING_MOVABLE. */ - } else if (unlikely(__PageMovable(page))) { + if (unlikely(__PageMovable(page))) { VM_BUG_ON_PAGE(!PageIsolated(page), page); lock_page(page); if (PageMovable(page)) @@ -992,18 +990,6 @@ static int __unmap_and_move(struct page *page, struct page *newpage, if (unlikely(!trylock_page(newpage))) goto out_unlock; - if (unlikely(isolated_balloon_page(page))) { - /* - * A ballooned page does not need any special attention from - * physical to virtual reverse mapping procedures. - * Skip any attempt to unmap PTEs or to remap swap cache, - * in order to avoid burning cycles at rmap level, and perform - * the page migration right away (proteced by page lock). - */ - rc = balloon_page_migrate(newpage, page, mode); - goto out_unlock_both; - } - if (unlikely(!is_lru)) { rc = move_to_new_page(newpage, page, mode); goto out_unlock_both; @@ -1058,8 +1044,7 @@ out: * list in here. */ if (rc == MIGRATEPAGE_SUCCESS) { - if (unlikely(__is_movable_balloon_page(newpage) || - __PageMovable(newpage))) + if (unlikely(__PageMovable(newpage))) put_page(newpage); else putback_lru_page(newpage); diff --git a/mm/vmscan.c b/mm/vmscan.c index c4a2f4512fca..93ba33789ac6 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1254,7 +1254,7 @@ unsigned long reclaim_clean_pages_from_list(struct zone *zone, list_for_each_entry_safe(page, next, page_list, lru) { if (page_is_file_cache(page) && !PageDirty(page) && - !isolated_balloon_page(page)) { + !__PageMovable(page)) { ClearPageActive(page); list_move(&page->lru, &clean_pages); } -- cgit v1.2.3 From 48b4800a1c6af2cdda344ea4e2c843dcc1f6afc9 Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Tue, 26 Jul 2016 15:23:31 -0700 Subject: zsmalloc: page migration support This patch introduces run-time migration feature for zspage. For migration, VM uses page.lru field so it would be better to not use page.next field which is unified with page.lru for own purpose. For that, firstly, we can get first object offset of the page via runtime calculation instead of using page.index so we can use page.index as link for page chaining instead of page.next. In case of huge object, it stores handle to page.index instead of next link of page chaining because huge object doesn't need to next link for page chaining. So get_next_page need to identify huge object to return NULL. For it, this patch uses PG_owner_priv_1 flag of the page flag. For migration, it supports three functions * zs_page_isolate It isolates a zspage which includes a subpage VM want to migrate from class so anyone cannot allocate new object from the zspage. We could try to isolate a zspage by the number of subpage so subsequent isolation trial of other subpage of the zpsage shouldn't fail. For that, we introduce zspage.isolated count. With that, zs_page_isolate can know whether zspage is already isolated or not for migration so if it is isolated for migration, subsequent isolation trial can be successful without trying further isolation. * zs_page_migrate First of all, it holds write-side zspage->lock to prevent migrate other subpage in zspage. Then, lock all objects in the page VM want to migrate. The reason we should lock all objects in the page is due to race between zs_map_object and zs_page_migrate. zs_map_object zs_page_migrate pin_tag(handle) obj = handle_to_obj(handle) obj_to_location(obj, &page, &obj_idx); write_lock(&zspage->lock) if (!trypin_tag(handle)) goto unpin_object zspage = get_zspage(page); read_lock(&zspage->lock); If zs_page_migrate doesn't do trypin_tag, zs_map_object's page can be stale by migration so it goes crash. If it locks all of objects successfully, it copies content from old page to new one, finally, create new zspage chain with new page. And if it's last isolated subpage in the zspage, put the zspage back to class. * zs_page_putback It returns isolated zspage to right fullness_group list if it fails to migrate a page. If it find a zspage is ZS_EMPTY, it queues zspage freeing to workqueue. See below about async zspage freeing. This patch introduces asynchronous zspage free. The reason to need it is we need page_lock to clear PG_movable but unfortunately, zs_free path should be atomic so the apporach is try to grab page_lock. If it got page_lock of all of pages successfully, it can free zspage immediately. Otherwise, it queues free request and free zspage via workqueue in process context. If zs_free finds the zspage is isolated when it try to free zspage, it delays the freeing until zs_page_putback finds it so it will free free the zspage finally. In this patch, we expand fullness_list from ZS_EMPTY to ZS_FULL. First of all, it will use ZS_EMPTY list for delay freeing. And with adding ZS_FULL list, it makes to identify whether zspage is isolated or not via list_empty(&zspage->list) test. [minchan@kernel.org: zsmalloc: keep first object offset in struct page] Link: http://lkml.kernel.org/r/1465788015-23195-1-git-send-email-minchan@kernel.org [minchan@kernel.org: zsmalloc: zspage sanity check] Link: http://lkml.kernel.org/r/20160603010129.GC3304@bbox Link: http://lkml.kernel.org/r/1464736881-24886-12-git-send-email-minchan@kernel.org Signed-off-by: Minchan Kim Cc: Sergey Senozhatsky Cc: Vlastimil Babka Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/uapi/linux/magic.h | 1 + mm/zsmalloc.c | 769 ++++++++++++++++++++++++++++++++++++++------- 2 files changed, 654 insertions(+), 116 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h index d829ce63529d..e398beac67b8 100644 --- a/include/uapi/linux/magic.h +++ b/include/uapi/linux/magic.h @@ -81,5 +81,6 @@ /* Since UDF 2.01 is ISO 13346 based... */ #define UDF_SUPER_MAGIC 0x15013346 #define BALLOON_KVM_MAGIC 0x13661366 +#define ZSMALLOC_MAGIC 0x58295829 #endif /* __LINUX_MAGIC_H__ */ diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index c6fb543cfb98..04a4f063b4fd 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -17,14 +17,14 @@ * * Usage of struct page fields: * page->private: points to zspage - * page->index: offset of the first object starting in this page. - * For the first page, this is always 0, so we use this field - * to store handle for huge object. - * page->next: links together all component pages of a zspage + * page->freelist(index): links together all component pages of a zspage + * For the huge page, this is always 0, so we use this field + * to store handle. * * Usage of struct page flags: * PG_private: identifies the first component page * PG_private2: identifies the last component page + * PG_owner_priv_1: indentifies the huge component page * */ @@ -49,6 +49,11 @@ #include #include #include +#include +#include +#include + +#define ZSPAGE_MAGIC 0x58 /* * This must be power of 2 and greater than of equal to sizeof(link_free). @@ -136,25 +141,23 @@ * We do not maintain any list for completely empty or full pages */ enum fullness_group { - ZS_ALMOST_FULL, - ZS_ALMOST_EMPTY, ZS_EMPTY, - ZS_FULL + ZS_ALMOST_EMPTY, + ZS_ALMOST_FULL, + ZS_FULL, + NR_ZS_FULLNESS, }; enum zs_stat_type { + CLASS_EMPTY, + CLASS_ALMOST_EMPTY, + CLASS_ALMOST_FULL, + CLASS_FULL, OBJ_ALLOCATED, OBJ_USED, - CLASS_ALMOST_FULL, - CLASS_ALMOST_EMPTY, + NR_ZS_STAT_TYPE, }; -#ifdef CONFIG_ZSMALLOC_STAT -#define NR_ZS_STAT_TYPE (CLASS_ALMOST_EMPTY + 1) -#else -#define NR_ZS_STAT_TYPE (OBJ_USED + 1) -#endif - struct zs_size_stat { unsigned long objs[NR_ZS_STAT_TYPE]; }; @@ -163,6 +166,10 @@ struct zs_size_stat { static struct dentry *zs_stat_root; #endif +#ifdef CONFIG_COMPACTION +static struct vfsmount *zsmalloc_mnt; +#endif + /* * number of size_classes */ @@ -186,23 +193,36 @@ static const int fullness_threshold_frac = 4; struct size_class { spinlock_t lock; - struct list_head fullness_list[2]; + struct list_head fullness_list[NR_ZS_FULLNESS]; /* * Size of objects stored in this class. Must be multiple * of ZS_ALIGN. */ int size; int objs_per_zspage; - unsigned int index; - - struct zs_size_stat stats; - /* Number of PAGE_SIZE sized pages to combine to form a 'zspage' */ int pages_per_zspage; - /* huge object: pages_per_zspage == 1 && maxobj_per_zspage == 1 */ - bool huge; + + unsigned int index; + struct zs_size_stat stats; }; +/* huge object: pages_per_zspage == 1 && maxobj_per_zspage == 1 */ +static void SetPageHugeObject(struct page *page) +{ + SetPageOwnerPriv1(page); +} + +static void ClearPageHugeObject(struct page *page) +{ + ClearPageOwnerPriv1(page); +} + +static int PageHugeObject(struct page *page) +{ + return PageOwnerPriv1(page); +} + /* * Placed within free objects to form a singly linked list. * For every zspage, zspage->freeobj gives head of this list. @@ -244,6 +264,10 @@ struct zs_pool { #ifdef CONFIG_ZSMALLOC_STAT struct dentry *stat_dentry; #endif +#ifdef CONFIG_COMPACTION + struct inode *inode; + struct work_struct free_work; +#endif }; /* @@ -252,16 +276,23 @@ struct zs_pool { */ #define FULLNESS_BITS 2 #define CLASS_BITS 8 +#define ISOLATED_BITS 3 +#define MAGIC_VAL_BITS 8 struct zspage { struct { unsigned int fullness:FULLNESS_BITS; unsigned int class:CLASS_BITS; + unsigned int isolated:ISOLATED_BITS; + unsigned int magic:MAGIC_VAL_BITS; }; unsigned int inuse; unsigned int freeobj; struct page *first_page; struct list_head list; /* fullness list */ +#ifdef CONFIG_COMPACTION + rwlock_t lock; +#endif }; struct mapping_area { @@ -274,6 +305,28 @@ struct mapping_area { enum zs_mapmode vm_mm; /* mapping mode */ }; +#ifdef CONFIG_COMPACTION +static int zs_register_migration(struct zs_pool *pool); +static void zs_unregister_migration(struct zs_pool *pool); +static void migrate_lock_init(struct zspage *zspage); +static void migrate_read_lock(struct zspage *zspage); +static void migrate_read_unlock(struct zspage *zspage); +static void kick_deferred_free(struct zs_pool *pool); +static void init_deferred_free(struct zs_pool *pool); +static void SetZsPageMovable(struct zs_pool *pool, struct zspage *zspage); +#else +static int zsmalloc_mount(void) { return 0; } +static void zsmalloc_unmount(void) {} +static int zs_register_migration(struct zs_pool *pool) { return 0; } +static void zs_unregister_migration(struct zs_pool *pool) {} +static void migrate_lock_init(struct zspage *zspage) {} +static void migrate_read_lock(struct zspage *zspage) {} +static void migrate_read_unlock(struct zspage *zspage) {} +static void kick_deferred_free(struct zs_pool *pool) {} +static void init_deferred_free(struct zs_pool *pool) {} +static void SetZsPageMovable(struct zs_pool *pool, struct zspage *zspage) {} +#endif + static int create_cache(struct zs_pool *pool) { pool->handle_cachep = kmem_cache_create("zs_handle", ZS_HANDLE_SIZE, @@ -301,7 +354,7 @@ static void destroy_cache(struct zs_pool *pool) static unsigned long cache_alloc_handle(struct zs_pool *pool, gfp_t gfp) { return (unsigned long)kmem_cache_alloc(pool->handle_cachep, - gfp & ~__GFP_HIGHMEM); + gfp & ~(__GFP_HIGHMEM|__GFP_MOVABLE)); } static void cache_free_handle(struct zs_pool *pool, unsigned long handle) @@ -311,7 +364,8 @@ static void cache_free_handle(struct zs_pool *pool, unsigned long handle) static struct zspage *cache_alloc_zspage(struct zs_pool *pool, gfp_t flags) { - return kmem_cache_alloc(pool->zspage_cachep, flags & ~__GFP_HIGHMEM); + return kmem_cache_alloc(pool->zspage_cachep, + flags & ~(__GFP_HIGHMEM|__GFP_MOVABLE)); }; static void cache_free_zspage(struct zs_pool *pool, struct zspage *zspage) @@ -421,11 +475,17 @@ static unsigned int get_maxobj_per_zspage(int size, int pages_per_zspage) /* per-cpu VM mapping areas for zspage accesses that cross page boundaries */ static DEFINE_PER_CPU(struct mapping_area, zs_map_area); +static bool is_zspage_isolated(struct zspage *zspage) +{ + return zspage->isolated; +} + static int is_first_page(struct page *page) { return PagePrivate(page); } +/* Protected by class->lock */ static inline int get_zspage_inuse(struct zspage *zspage) { return zspage->inuse; @@ -441,20 +501,22 @@ static inline void mod_zspage_inuse(struct zspage *zspage, int val) zspage->inuse += val; } -static inline int get_first_obj_offset(struct page *page) +static inline struct page *get_first_page(struct zspage *zspage) { - if (is_first_page(page)) - return 0; + struct page *first_page = zspage->first_page; - return page->index; + VM_BUG_ON_PAGE(!is_first_page(first_page), first_page); + return first_page; } -static inline void set_first_obj_offset(struct page *page, int offset) +static inline int get_first_obj_offset(struct page *page) { - if (is_first_page(page)) - return; + return page->units; +} - page->index = offset; +static inline void set_first_obj_offset(struct page *page, int offset) +{ + page->units = offset; } static inline unsigned int get_freeobj(struct zspage *zspage) @@ -471,6 +533,8 @@ static void get_zspage_mapping(struct zspage *zspage, unsigned int *class_idx, enum fullness_group *fullness) { + BUG_ON(zspage->magic != ZSPAGE_MAGIC); + *fullness = zspage->fullness; *class_idx = zspage->class; } @@ -504,23 +568,19 @@ static int get_size_class_index(int size) static inline void zs_stat_inc(struct size_class *class, enum zs_stat_type type, unsigned long cnt) { - if (type < NR_ZS_STAT_TYPE) - class->stats.objs[type] += cnt; + class->stats.objs[type] += cnt; } static inline void zs_stat_dec(struct size_class *class, enum zs_stat_type type, unsigned long cnt) { - if (type < NR_ZS_STAT_TYPE) - class->stats.objs[type] -= cnt; + class->stats.objs[type] -= cnt; } static inline unsigned long zs_stat_get(struct size_class *class, enum zs_stat_type type) { - if (type < NR_ZS_STAT_TYPE) - return class->stats.objs[type]; - return 0; + return class->stats.objs[type]; } #ifdef CONFIG_ZSMALLOC_STAT @@ -664,6 +724,7 @@ static inline void zs_pool_stat_destroy(struct zs_pool *pool) } #endif + /* * For each size class, zspages are divided into different groups * depending on how "full" they are. This was done so that we could @@ -704,15 +765,9 @@ static void insert_zspage(struct size_class *class, { struct zspage *head; - if (fullness >= ZS_EMPTY) - return; - + zs_stat_inc(class, fullness, 1); head = list_first_entry_or_null(&class->fullness_list[fullness], struct zspage, list); - - zs_stat_inc(class, fullness == ZS_ALMOST_EMPTY ? - CLASS_ALMOST_EMPTY : CLASS_ALMOST_FULL, 1); - /* * We want to see more ZS_FULL pages and less almost empty/full. * Put pages with higher ->inuse first. @@ -734,14 +789,11 @@ static void remove_zspage(struct size_class *class, struct zspage *zspage, enum fullness_group fullness) { - if (fullness >= ZS_EMPTY) - return; - VM_BUG_ON(list_empty(&class->fullness_list[fullness])); + VM_BUG_ON(is_zspage_isolated(zspage)); list_del_init(&zspage->list); - zs_stat_dec(class, fullness == ZS_ALMOST_EMPTY ? - CLASS_ALMOST_EMPTY : CLASS_ALMOST_FULL, 1); + zs_stat_dec(class, fullness, 1); } /* @@ -764,8 +816,11 @@ static enum fullness_group fix_fullness_group(struct size_class *class, if (newfg == currfg) goto out; - remove_zspage(class, zspage, currfg); - insert_zspage(class, zspage, newfg); + if (!is_zspage_isolated(zspage)) { + remove_zspage(class, zspage, currfg); + insert_zspage(class, zspage, newfg); + } + set_zspage_mapping(zspage, class_idx, newfg); out: @@ -808,19 +863,20 @@ static int get_pages_per_zspage(int class_size) return max_usedpc_order; } -static struct page *get_first_page(struct zspage *zspage) -{ - return zspage->first_page; -} - static struct zspage *get_zspage(struct page *page) { - return (struct zspage *)page->private; + struct zspage *zspage = (struct zspage *)page->private; + + BUG_ON(zspage->magic != ZSPAGE_MAGIC); + return zspage; } static struct page *get_next_page(struct page *page) { - return page->next; + if (unlikely(PageHugeObject(page))) + return NULL; + + return page->freelist; } /** @@ -857,16 +913,20 @@ static unsigned long handle_to_obj(unsigned long handle) return *(unsigned long *)handle; } -static unsigned long obj_to_head(struct size_class *class, struct page *page, - void *obj) +static unsigned long obj_to_head(struct page *page, void *obj) { - if (class->huge) { + if (unlikely(PageHugeObject(page))) { VM_BUG_ON_PAGE(!is_first_page(page), page); return page->index; } else return *(unsigned long *)obj; } +static inline int testpin_tag(unsigned long handle) +{ + return bit_spin_is_locked(HANDLE_PIN_BIT, (unsigned long *)handle); +} + static inline int trypin_tag(unsigned long handle) { return bit_spin_trylock(HANDLE_PIN_BIT, (unsigned long *)handle); @@ -884,27 +944,94 @@ static void unpin_tag(unsigned long handle) static void reset_page(struct page *page) { + __ClearPageMovable(page); clear_bit(PG_private, &page->flags); clear_bit(PG_private_2, &page->flags); set_page_private(page, 0); - page->index = 0; + page_mapcount_reset(page); + ClearPageHugeObject(page); + page->freelist = NULL; +} + +/* + * To prevent zspage destroy during migration, zspage freeing should + * hold locks of all pages in the zspage. + */ +void lock_zspage(struct zspage *zspage) +{ + struct page *page = get_first_page(zspage); + + do { + lock_page(page); + } while ((page = get_next_page(page)) != NULL); +} + +int trylock_zspage(struct zspage *zspage) +{ + struct page *cursor, *fail; + + for (cursor = get_first_page(zspage); cursor != NULL; cursor = + get_next_page(cursor)) { + if (!trylock_page(cursor)) { + fail = cursor; + goto unlock; + } + } + + return 1; +unlock: + for (cursor = get_first_page(zspage); cursor != fail; cursor = + get_next_page(cursor)) + unlock_page(cursor); + + return 0; } -static void free_zspage(struct zs_pool *pool, struct zspage *zspage) +static void __free_zspage(struct zs_pool *pool, struct size_class *class, + struct zspage *zspage) { struct page *page, *next; + enum fullness_group fg; + unsigned int class_idx; + + get_zspage_mapping(zspage, &class_idx, &fg); + + assert_spin_locked(&class->lock); VM_BUG_ON(get_zspage_inuse(zspage)); + VM_BUG_ON(fg != ZS_EMPTY); - next = page = zspage->first_page; + next = page = get_first_page(zspage); do { - next = page->next; + VM_BUG_ON_PAGE(!PageLocked(page), page); + next = get_next_page(page); reset_page(page); + unlock_page(page); put_page(page); page = next; } while (page != NULL); cache_free_zspage(pool, zspage); + + zs_stat_dec(class, OBJ_ALLOCATED, get_maxobj_per_zspage( + class->size, class->pages_per_zspage)); + atomic_long_sub(class->pages_per_zspage, + &pool->pages_allocated); +} + +static void free_zspage(struct zs_pool *pool, struct size_class *class, + struct zspage *zspage) +{ + VM_BUG_ON(get_zspage_inuse(zspage)); + VM_BUG_ON(list_empty(&zspage->list)); + + if (!trylock_zspage(zspage)) { + kick_deferred_free(pool); + return; + } + + remove_zspage(class, zspage, ZS_EMPTY); + __free_zspage(pool, class, zspage); } /* Initialize a newly allocated zspage */ @@ -912,7 +1039,7 @@ static void init_zspage(struct size_class *class, struct zspage *zspage) { unsigned int freeobj = 1; unsigned long off = 0; - struct page *page = zspage->first_page; + struct page *page = get_first_page(zspage); while (page) { struct page *next_page; @@ -952,16 +1079,17 @@ static void init_zspage(struct size_class *class, struct zspage *zspage) set_freeobj(zspage, 0); } -static void create_page_chain(struct zspage *zspage, struct page *pages[], - int nr_pages) +static void create_page_chain(struct size_class *class, struct zspage *zspage, + struct page *pages[]) { int i; struct page *page; struct page *prev_page = NULL; + int nr_pages = class->pages_per_zspage; /* * Allocate individual pages and link them together as: - * 1. all pages are linked together using page->next + * 1. all pages are linked together using page->freelist * 2. each sub-page point to zspage using page->private * * we set PG_private to identify the first page (i.e. no other sub-page @@ -970,16 +1098,18 @@ static void create_page_chain(struct zspage *zspage, struct page *pages[], for (i = 0; i < nr_pages; i++) { page = pages[i]; set_page_private(page, (unsigned long)zspage); + page->freelist = NULL; if (i == 0) { zspage->first_page = page; SetPagePrivate(page); + if (unlikely(class->objs_per_zspage == 1 && + class->pages_per_zspage == 1)) + SetPageHugeObject(page); } else { - prev_page->next = page; + prev_page->freelist = page; } - if (i == nr_pages - 1) { + if (i == nr_pages - 1) SetPagePrivate2(page); - page->next = NULL; - } prev_page = page; } } @@ -999,6 +1129,8 @@ static struct zspage *alloc_zspage(struct zs_pool *pool, return NULL; memset(zspage, 0, sizeof(struct zspage)); + zspage->magic = ZSPAGE_MAGIC; + migrate_lock_init(zspage); for (i = 0; i < class->pages_per_zspage; i++) { struct page *page; @@ -1013,7 +1145,7 @@ static struct zspage *alloc_zspage(struct zs_pool *pool, pages[i] = page; } - create_page_chain(zspage, pages, class->pages_per_zspage); + create_page_chain(class, zspage, pages); init_zspage(class, zspage); return zspage; @@ -1024,7 +1156,7 @@ static struct zspage *find_get_zspage(struct size_class *class) int i; struct zspage *zspage; - for (i = ZS_ALMOST_FULL; i <= ZS_ALMOST_EMPTY; i++) { + for (i = ZS_ALMOST_FULL; i >= ZS_EMPTY; i--) { zspage = list_first_entry_or_null(&class->fullness_list[i], struct zspage, list); if (zspage) @@ -1289,6 +1421,10 @@ void *zs_map_object(struct zs_pool *pool, unsigned long handle, obj = handle_to_obj(handle); obj_to_location(obj, &page, &obj_idx); zspage = get_zspage(page); + + /* migration cannot move any subpage in this zspage */ + migrate_read_lock(zspage); + get_zspage_mapping(zspage, &class_idx, &fg); class = pool->size_class[class_idx]; off = (class->size * obj_idx) & ~PAGE_MASK; @@ -1309,7 +1445,7 @@ void *zs_map_object(struct zs_pool *pool, unsigned long handle, ret = __zs_map_object(area, pages, off, class->size); out: - if (!class->huge) + if (likely(!PageHugeObject(page))) ret += ZS_HANDLE_SIZE; return ret; @@ -1348,6 +1484,8 @@ void zs_unmap_object(struct zs_pool *pool, unsigned long handle) __zs_unmap_object(area, pages, off, class->size); } put_cpu_var(zs_map_area); + + migrate_read_unlock(zspage); unpin_tag(handle); } EXPORT_SYMBOL_GPL(zs_unmap_object); @@ -1377,7 +1515,7 @@ static unsigned long obj_malloc(struct size_class *class, vaddr = kmap_atomic(m_page); link = (struct link_free *)vaddr + m_offset / sizeof(*link); set_freeobj(zspage, link->next >> OBJ_ALLOCATED_TAG); - if (!class->huge) + if (likely(!PageHugeObject(m_page))) /* record handle in the header of allocated chunk */ link->handle = handle; else @@ -1407,6 +1545,7 @@ unsigned long zs_malloc(struct zs_pool *pool, size_t size, gfp_t gfp) { unsigned long handle, obj; struct size_class *class; + enum fullness_group newfg; struct zspage *zspage; if (unlikely(!size || size > ZS_MAX_ALLOC_SIZE)) @@ -1422,28 +1561,37 @@ unsigned long zs_malloc(struct zs_pool *pool, size_t size, gfp_t gfp) spin_lock(&class->lock); zspage = find_get_zspage(class); - - if (!zspage) { + if (likely(zspage)) { + obj = obj_malloc(class, zspage, handle); + /* Now move the zspage to another fullness group, if required */ + fix_fullness_group(class, zspage); + record_obj(handle, obj); spin_unlock(&class->lock); - zspage = alloc_zspage(pool, class, gfp); - if (unlikely(!zspage)) { - cache_free_handle(pool, handle); - return 0; - } - set_zspage_mapping(zspage, class->index, ZS_EMPTY); - atomic_long_add(class->pages_per_zspage, - &pool->pages_allocated); + return handle; + } - spin_lock(&class->lock); - zs_stat_inc(class, OBJ_ALLOCATED, get_maxobj_per_zspage( - class->size, class->pages_per_zspage)); + spin_unlock(&class->lock); + + zspage = alloc_zspage(pool, class, gfp); + if (!zspage) { + cache_free_handle(pool, handle); + return 0; } + spin_lock(&class->lock); obj = obj_malloc(class, zspage, handle); - /* Now move the zspage to another fullness group, if required */ - fix_fullness_group(class, zspage); + newfg = get_fullness_group(class, zspage); + insert_zspage(class, zspage, newfg); + set_zspage_mapping(zspage, class->index, newfg); record_obj(handle, obj); + atomic_long_add(class->pages_per_zspage, + &pool->pages_allocated); + zs_stat_inc(class, OBJ_ALLOCATED, get_maxobj_per_zspage( + class->size, class->pages_per_zspage)); + + /* We completely set up zspage so mark them as movable */ + SetZsPageMovable(pool, zspage); spin_unlock(&class->lock); return handle; @@ -1484,6 +1632,7 @@ void zs_free(struct zs_pool *pool, unsigned long handle) int class_idx; struct size_class *class; enum fullness_group fullness; + bool isolated; if (unlikely(!handle)) return; @@ -1493,22 +1642,28 @@ void zs_free(struct zs_pool *pool, unsigned long handle) obj_to_location(obj, &f_page, &f_objidx); zspage = get_zspage(f_page); + migrate_read_lock(zspage); + get_zspage_mapping(zspage, &class_idx, &fullness); class = pool->size_class[class_idx]; spin_lock(&class->lock); obj_free(class, obj); fullness = fix_fullness_group(class, zspage); - if (fullness == ZS_EMPTY) { - zs_stat_dec(class, OBJ_ALLOCATED, get_maxobj_per_zspage( - class->size, class->pages_per_zspage)); - atomic_long_sub(class->pages_per_zspage, - &pool->pages_allocated); - free_zspage(pool, zspage); + if (fullness != ZS_EMPTY) { + migrate_read_unlock(zspage); + goto out; } + + isolated = is_zspage_isolated(zspage); + migrate_read_unlock(zspage); + /* If zspage is isolated, zs_page_putback will free the zspage */ + if (likely(!isolated)) + free_zspage(pool, class, zspage); +out: + spin_unlock(&class->lock); unpin_tag(handle); - cache_free_handle(pool, handle); } EXPORT_SYMBOL_GPL(zs_free); @@ -1592,7 +1747,7 @@ static unsigned long find_alloced_obj(struct size_class *class, offset += class->size * index; while (offset < PAGE_SIZE) { - head = obj_to_head(class, page, addr + offset); + head = obj_to_head(page, addr + offset); if (head & OBJ_ALLOCATED_TAG) { handle = head & ~OBJ_ALLOCATED_TAG; if (trypin_tag(handle)) @@ -1684,6 +1839,7 @@ static struct zspage *isolate_zspage(struct size_class *class, bool source) zspage = list_first_entry_or_null(&class->fullness_list[fg[i]], struct zspage, list); if (zspage) { + VM_BUG_ON(is_zspage_isolated(zspage)); remove_zspage(class, zspage, fg[i]); return zspage; } @@ -1704,6 +1860,8 @@ static enum fullness_group putback_zspage(struct size_class *class, { enum fullness_group fullness; + VM_BUG_ON(is_zspage_isolated(zspage)); + fullness = get_fullness_group(class, zspage); insert_zspage(class, zspage, fullness); set_zspage_mapping(zspage, class->index, fullness); @@ -1711,6 +1869,378 @@ static enum fullness_group putback_zspage(struct size_class *class, return fullness; } +#ifdef CONFIG_COMPACTION +static struct dentry *zs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) +{ + static const struct dentry_operations ops = { + .d_dname = simple_dname, + }; + + return mount_pseudo(fs_type, "zsmalloc:", NULL, &ops, ZSMALLOC_MAGIC); +} + +static struct file_system_type zsmalloc_fs = { + .name = "zsmalloc", + .mount = zs_mount, + .kill_sb = kill_anon_super, +}; + +static int zsmalloc_mount(void) +{ + int ret = 0; + + zsmalloc_mnt = kern_mount(&zsmalloc_fs); + if (IS_ERR(zsmalloc_mnt)) + ret = PTR_ERR(zsmalloc_mnt); + + return ret; +} + +static void zsmalloc_unmount(void) +{ + kern_unmount(zsmalloc_mnt); +} + +static void migrate_lock_init(struct zspage *zspage) +{ + rwlock_init(&zspage->lock); +} + +static void migrate_read_lock(struct zspage *zspage) +{ + read_lock(&zspage->lock); +} + +static void migrate_read_unlock(struct zspage *zspage) +{ + read_unlock(&zspage->lock); +} + +static void migrate_write_lock(struct zspage *zspage) +{ + write_lock(&zspage->lock); +} + +static void migrate_write_unlock(struct zspage *zspage) +{ + write_unlock(&zspage->lock); +} + +/* Number of isolated subpage for *page migration* in this zspage */ +static void inc_zspage_isolation(struct zspage *zspage) +{ + zspage->isolated++; +} + +static void dec_zspage_isolation(struct zspage *zspage) +{ + zspage->isolated--; +} + +static void replace_sub_page(struct size_class *class, struct zspage *zspage, + struct page *newpage, struct page *oldpage) +{ + struct page *page; + struct page *pages[ZS_MAX_PAGES_PER_ZSPAGE] = {NULL, }; + int idx = 0; + + page = get_first_page(zspage); + do { + if (page == oldpage) + pages[idx] = newpage; + else + pages[idx] = page; + idx++; + } while ((page = get_next_page(page)) != NULL); + + create_page_chain(class, zspage, pages); + set_first_obj_offset(newpage, get_first_obj_offset(oldpage)); + if (unlikely(PageHugeObject(oldpage))) + newpage->index = oldpage->index; + __SetPageMovable(newpage, page_mapping(oldpage)); +} + +bool zs_page_isolate(struct page *page, isolate_mode_t mode) +{ + struct zs_pool *pool; + struct size_class *class; + int class_idx; + enum fullness_group fullness; + struct zspage *zspage; + struct address_space *mapping; + + /* + * Page is locked so zspage couldn't be destroyed. For detail, look at + * lock_zspage in free_zspage. + */ + VM_BUG_ON_PAGE(!PageMovable(page), page); + VM_BUG_ON_PAGE(PageIsolated(page), page); + + zspage = get_zspage(page); + + /* + * Without class lock, fullness could be stale while class_idx is okay + * because class_idx is constant unless page is freed so we should get + * fullness again under class lock. + */ + get_zspage_mapping(zspage, &class_idx, &fullness); + mapping = page_mapping(page); + pool = mapping->private_data; + class = pool->size_class[class_idx]; + + spin_lock(&class->lock); + if (get_zspage_inuse(zspage) == 0) { + spin_unlock(&class->lock); + return false; + } + + /* zspage is isolated for object migration */ + if (list_empty(&zspage->list) && !is_zspage_isolated(zspage)) { + spin_unlock(&class->lock); + return false; + } + + /* + * If this is first time isolation for the zspage, isolate zspage from + * size_class to prevent further object allocation from the zspage. + */ + if (!list_empty(&zspage->list) && !is_zspage_isolated(zspage)) { + get_zspage_mapping(zspage, &class_idx, &fullness); + remove_zspage(class, zspage, fullness); + } + + inc_zspage_isolation(zspage); + spin_unlock(&class->lock); + + return true; +} + +int zs_page_migrate(struct address_space *mapping, struct page *newpage, + struct page *page, enum migrate_mode mode) +{ + struct zs_pool *pool; + struct size_class *class; + int class_idx; + enum fullness_group fullness; + struct zspage *zspage; + struct page *dummy; + void *s_addr, *d_addr, *addr; + int offset, pos; + unsigned long handle, head; + unsigned long old_obj, new_obj; + unsigned int obj_idx; + int ret = -EAGAIN; + + VM_BUG_ON_PAGE(!PageMovable(page), page); + VM_BUG_ON_PAGE(!PageIsolated(page), page); + + zspage = get_zspage(page); + + /* Concurrent compactor cannot migrate any subpage in zspage */ + migrate_write_lock(zspage); + get_zspage_mapping(zspage, &class_idx, &fullness); + pool = mapping->private_data; + class = pool->size_class[class_idx]; + offset = get_first_obj_offset(page); + + spin_lock(&class->lock); + if (!get_zspage_inuse(zspage)) { + ret = -EBUSY; + goto unlock_class; + } + + pos = offset; + s_addr = kmap_atomic(page); + while (pos < PAGE_SIZE) { + head = obj_to_head(page, s_addr + pos); + if (head & OBJ_ALLOCATED_TAG) { + handle = head & ~OBJ_ALLOCATED_TAG; + if (!trypin_tag(handle)) + goto unpin_objects; + } + pos += class->size; + } + + /* + * Here, any user cannot access all objects in the zspage so let's move. + */ + d_addr = kmap_atomic(newpage); + memcpy(d_addr, s_addr, PAGE_SIZE); + kunmap_atomic(d_addr); + + for (addr = s_addr + offset; addr < s_addr + pos; + addr += class->size) { + head = obj_to_head(page, addr); + if (head & OBJ_ALLOCATED_TAG) { + handle = head & ~OBJ_ALLOCATED_TAG; + if (!testpin_tag(handle)) + BUG(); + + old_obj = handle_to_obj(handle); + obj_to_location(old_obj, &dummy, &obj_idx); + new_obj = (unsigned long)location_to_obj(newpage, + obj_idx); + new_obj |= BIT(HANDLE_PIN_BIT); + record_obj(handle, new_obj); + } + } + + replace_sub_page(class, zspage, newpage, page); + get_page(newpage); + + dec_zspage_isolation(zspage); + + /* + * Page migration is done so let's putback isolated zspage to + * the list if @page is final isolated subpage in the zspage. + */ + if (!is_zspage_isolated(zspage)) + putback_zspage(class, zspage); + + reset_page(page); + put_page(page); + page = newpage; + + ret = 0; +unpin_objects: + for (addr = s_addr + offset; addr < s_addr + pos; + addr += class->size) { + head = obj_to_head(page, addr); + if (head & OBJ_ALLOCATED_TAG) { + handle = head & ~OBJ_ALLOCATED_TAG; + if (!testpin_tag(handle)) + BUG(); + unpin_tag(handle); + } + } + kunmap_atomic(s_addr); +unlock_class: + spin_unlock(&class->lock); + migrate_write_unlock(zspage); + + return ret; +} + +void zs_page_putback(struct page *page) +{ + struct zs_pool *pool; + struct size_class *class; + int class_idx; + enum fullness_group fg; + struct address_space *mapping; + struct zspage *zspage; + + VM_BUG_ON_PAGE(!PageMovable(page), page); + VM_BUG_ON_PAGE(!PageIsolated(page), page); + + zspage = get_zspage(page); + get_zspage_mapping(zspage, &class_idx, &fg); + mapping = page_mapping(page); + pool = mapping->private_data; + class = pool->size_class[class_idx]; + + spin_lock(&class->lock); + dec_zspage_isolation(zspage); + if (!is_zspage_isolated(zspage)) { + fg = putback_zspage(class, zspage); + /* + * Due to page_lock, we cannot free zspage immediately + * so let's defer. + */ + if (fg == ZS_EMPTY) + schedule_work(&pool->free_work); + } + spin_unlock(&class->lock); +} + +const struct address_space_operations zsmalloc_aops = { + .isolate_page = zs_page_isolate, + .migratepage = zs_page_migrate, + .putback_page = zs_page_putback, +}; + +static int zs_register_migration(struct zs_pool *pool) +{ + pool->inode = alloc_anon_inode(zsmalloc_mnt->mnt_sb); + if (IS_ERR(pool->inode)) { + pool->inode = NULL; + return 1; + } + + pool->inode->i_mapping->private_data = pool; + pool->inode->i_mapping->a_ops = &zsmalloc_aops; + return 0; +} + +static void zs_unregister_migration(struct zs_pool *pool) +{ + flush_work(&pool->free_work); + if (pool->inode) + iput(pool->inode); +} + +/* + * Caller should hold page_lock of all pages in the zspage + * In here, we cannot use zspage meta data. + */ +static void async_free_zspage(struct work_struct *work) +{ + int i; + struct size_class *class; + unsigned int class_idx; + enum fullness_group fullness; + struct zspage *zspage, *tmp; + LIST_HEAD(free_pages); + struct zs_pool *pool = container_of(work, struct zs_pool, + free_work); + + for (i = 0; i < zs_size_classes; i++) { + class = pool->size_class[i]; + if (class->index != i) + continue; + + spin_lock(&class->lock); + list_splice_init(&class->fullness_list[ZS_EMPTY], &free_pages); + spin_unlock(&class->lock); + } + + + list_for_each_entry_safe(zspage, tmp, &free_pages, list) { + list_del(&zspage->list); + lock_zspage(zspage); + + get_zspage_mapping(zspage, &class_idx, &fullness); + VM_BUG_ON(fullness != ZS_EMPTY); + class = pool->size_class[class_idx]; + spin_lock(&class->lock); + __free_zspage(pool, pool->size_class[class_idx], zspage); + spin_unlock(&class->lock); + } +}; + +static void kick_deferred_free(struct zs_pool *pool) +{ + schedule_work(&pool->free_work); +} + +static void init_deferred_free(struct zs_pool *pool) +{ + INIT_WORK(&pool->free_work, async_free_zspage); +} + +static void SetZsPageMovable(struct zs_pool *pool, struct zspage *zspage) +{ + struct page *page = get_first_page(zspage); + + do { + WARN_ON(!trylock_page(page)); + __SetPageMovable(page, pool->inode->i_mapping); + unlock_page(page); + } while ((page = get_next_page(page)) != NULL); +} +#endif + /* * * Based on the number of unused allocated objects calculate @@ -1745,10 +2275,10 @@ static void __zs_compact(struct zs_pool *pool, struct size_class *class) break; cc.index = 0; - cc.s_page = src_zspage->first_page; + cc.s_page = get_first_page(src_zspage); while ((dst_zspage = isolate_zspage(class, false))) { - cc.d_page = dst_zspage->first_page; + cc.d_page = get_first_page(dst_zspage); /* * If there is no more space in dst_page, resched * and see if anyone had allocated another zspage. @@ -1765,11 +2295,7 @@ static void __zs_compact(struct zs_pool *pool, struct size_class *class) putback_zspage(class, dst_zspage); if (putback_zspage(class, src_zspage) == ZS_EMPTY) { - zs_stat_dec(class, OBJ_ALLOCATED, get_maxobj_per_zspage( - class->size, class->pages_per_zspage)); - atomic_long_sub(class->pages_per_zspage, - &pool->pages_allocated); - free_zspage(pool, src_zspage); + free_zspage(pool, class, src_zspage); pool->stats.pages_compacted += class->pages_per_zspage; } spin_unlock(&class->lock); @@ -1885,6 +2411,7 @@ struct zs_pool *zs_create_pool(const char *name) if (!pool) return NULL; + init_deferred_free(pool); pool->size_class = kcalloc(zs_size_classes, sizeof(struct size_class *), GFP_KERNEL); if (!pool->size_class) { @@ -1939,12 +2466,10 @@ struct zs_pool *zs_create_pool(const char *name) class->pages_per_zspage = pages_per_zspage; class->objs_per_zspage = class->pages_per_zspage * PAGE_SIZE / class->size; - if (pages_per_zspage == 1 && class->objs_per_zspage == 1) - class->huge = true; spin_lock_init(&class->lock); pool->size_class[i] = class; - for (fullness = ZS_ALMOST_FULL; fullness <= ZS_ALMOST_EMPTY; - fullness++) + for (fullness = ZS_EMPTY; fullness < NR_ZS_FULLNESS; + fullness++) INIT_LIST_HEAD(&class->fullness_list[fullness]); prev_class = class; @@ -1953,6 +2478,9 @@ struct zs_pool *zs_create_pool(const char *name) /* debug only, don't abort if it fails */ zs_pool_stat_create(pool, name); + if (zs_register_migration(pool)) + goto err; + /* * Not critical, we still can use the pool * and user can trigger compaction manually. @@ -1972,6 +2500,7 @@ void zs_destroy_pool(struct zs_pool *pool) int i; zs_unregister_shrinker(pool); + zs_unregister_migration(pool); zs_pool_stat_destroy(pool); for (i = 0; i < zs_size_classes; i++) { @@ -1984,7 +2513,7 @@ void zs_destroy_pool(struct zs_pool *pool) if (class->index != i) continue; - for (fg = ZS_ALMOST_FULL; fg <= ZS_ALMOST_EMPTY; fg++) { + for (fg = ZS_EMPTY; fg < NR_ZS_FULLNESS; fg++) { if (!list_empty(&class->fullness_list[fg])) { pr_info("Freeing non-empty class with size %db, fullness group %d\n", class->size, fg); @@ -2002,7 +2531,13 @@ EXPORT_SYMBOL_GPL(zs_destroy_pool); static int __init zs_init(void) { - int ret = zs_register_cpu_notifier(); + int ret; + + ret = zsmalloc_mount(); + if (ret) + goto out; + + ret = zs_register_cpu_notifier(); if (ret) goto notifier_fail; @@ -2019,7 +2554,8 @@ static int __init zs_init(void) notifier_fail: zs_unregister_cpu_notifier(); - + zsmalloc_unmount(); +out: return ret; } @@ -2028,6 +2564,7 @@ static void __exit zs_exit(void) #ifdef CONFIG_ZPOOL zpool_unregister_driver(&zs_zpool_driver); #endif + zsmalloc_unmount(); zs_unregister_cpu_notifier(); zs_stat_exit(); -- cgit v1.2.3 From 0dc696bcf2e86f48a23fb95ca2f40c8708241e7e Mon Sep 17 00:00:00 2001 From: Anshuman Khandual Date: Thu, 28 Jul 2016 10:57:30 +0800 Subject: elf: Add powerpc specific core note sections This patch adds twelve ELF core note sections for powerpc architecture for various registers and register sets which need to be accessed from ptrace interface and then gdb. These additions include special purpose registers like TAR, PPR, DSCR, TM running and checkpointed state for various register sets, EBB related register set, performance monitor register set etc. Addition of these new ELF core note sections extends the existing ELF ABI on powerpc arch without affecting it in any manner. Signed-off-by: Anshuman Khandual Signed-off-by: Simon Guo Signed-off-by: Michael Ellerman --- include/uapi/linux/elf.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/elf.h b/include/uapi/linux/elf.h index cb4a72f888d5..1be3c5f6183b 100644 --- a/include/uapi/linux/elf.h +++ b/include/uapi/linux/elf.h @@ -381,6 +381,19 @@ typedef struct elf64_shdr { #define NT_PPC_VMX 0x100 /* PowerPC Altivec/VMX registers */ #define NT_PPC_SPE 0x101 /* PowerPC SPE/EVR registers */ #define NT_PPC_VSX 0x102 /* PowerPC VSX registers */ +#define NT_PPC_TAR 0x103 /* Target Address Register */ +#define NT_PPC_PPR 0x104 /* Program Priority Register */ +#define NT_PPC_DSCR 0x105 /* Data Stream Control Register */ +#define NT_PPC_EBB 0x106 /* Event Based Branch Registers */ +#define NT_PPC_PMU 0x107 /* Performance Monitor Registers */ +#define NT_PPC_TM_CGPR 0x108 /* TM checkpointed GPR Registers */ +#define NT_PPC_TM_CFPR 0x109 /* TM checkpointed FPR Registers */ +#define NT_PPC_TM_CVMX 0x10a /* TM checkpointed VMX Registers */ +#define NT_PPC_TM_CVSX 0x10b /* TM checkpointed VSX Registers */ +#define NT_PPC_TM_SPR 0x10c /* TM Special Purpose Registers */ +#define NT_PPC_TM_CTAR 0x10d /* TM checkpointed Target Address Register */ +#define NT_PPC_TM_CPPR 0x10e /* TM checkpointed Program Priority Register */ +#define NT_PPC_TM_CDSCR 0x10f /* TM checkpointed Data Stream Control Register */ #define NT_386_TLS 0x200 /* i386 TLS slots (struct user_desc) */ #define NT_386_IOPERM 0x201 /* x86 io permission bitmap (1=deny) */ #define NT_X86_XSTATE 0x202 /* x86 extended state using xsave */ -- cgit v1.2.3 From 23528bb21ee2c9b27f3feddd77a2a3351a8df148 Mon Sep 17 00:00:00 2001 From: Sam Bobroff Date: Wed, 20 Jul 2016 13:41:36 +1000 Subject: KVM: PPC: Introduce KVM_CAP_PPC_HTM Introduce a new KVM capability, KVM_CAP_PPC_HTM, that can be queried to determine if a PowerPC KVM guest should use HTM (Hardware Transactional Memory). This will be used by QEMU to populate the pa-features bits in the guest's device tree. Signed-off-by: Sam Bobroff Reviewed-by: David Gibson Signed-off-by: Paolo Bonzini --- arch/powerpc/kvm/powerpc.c | 4 ++++ include/uapi/linux/kvm.h | 1 + 2 files changed, 5 insertions(+) (limited to 'include/uapi') diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 1ac036e45ed4..6ce40dd6fe51 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -588,6 +588,10 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = 1; break; #endif + case KVM_CAP_PPC_HTM: + r = cpu_has_feature(CPU_FTR_TM_COMP) && + is_kvmppc_hv_enabled(kvm); + break; default: r = 0; break; diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 8f2756c263d4..e98bb4cce639 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -869,6 +869,7 @@ struct kvm_ppc_smmu_info { #define KVM_CAP_X2APIC_API 129 #define KVM_CAP_S390_USER_INSTR0 130 #define KVM_CAP_MSI_DEVID 131 +#define KVM_CAP_PPC_HTM 132 #ifdef KVM_CAP_IRQ_ROUTING -- cgit v1.2.3 From 1a937693993ff10d7e80cca6ddd55f3000aa6376 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 18 Apr 2016 12:58:14 +0300 Subject: virtio: new feature to detect IOMMU device quirk The interaction between virtio and IOMMUs is messy. On most systems with virtio, physical addresses match bus addresses, and it doesn't particularly matter which one we use to program the device. On some systems, including Xen and any system with a physical device that speaks virtio behind a physical IOMMU, we must program the IOMMU for virtio DMA to work at all. On other systems, including SPARC and PPC64, virtio-pci devices are enumerated as though they are behind an IOMMU, but the virtio host ignores the IOMMU, so we must either pretend that the IOMMU isn't there or somehow map everything as the identity. Add a feature bit to detect that quirk: VIRTIO_F_IOMMU_PLATFORM. Any device with this feature bit set to 0 needs a quirk and has to be passed physical addresses (as opposed to bus addresses) even though the device is behind an IOMMU. Note: it has to be a per-device quirk because for example, there could be a mix of passed-through and virtual virtio devices. As another example, some devices could be implemented by an out of process hypervisor backend (in case of qemu vhost, or vhost-user) and so support for an IOMMU needs to be coded up separately. It would be cleanest to handle this in IOMMU core code, but that needs per-device DMA ops. While we are waiting for that to be implemented, use a work-around in virtio core. Note: a "noiommu" feature is a quirk - add a wrapper to make that clear. Signed-off-by: Michael S. Tsirkin --- drivers/virtio/virtio_ring.c | 15 ++++++++++++++- include/linux/virtio_config.h | 13 +++++++++++++ include/uapi/linux/virtio_config.h | 10 +++++++++- 3 files changed, 36 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index ca6bfddaacad..114a0c88afb8 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -117,7 +117,10 @@ struct vring_virtqueue { #define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq) /* - * The interaction between virtio and a possible IOMMU is a mess. + * Modern virtio devices have feature bits to specify whether they need a + * quirk and bypass the IOMMU. If not there, just use the DMA API. + * + * If there, the interaction between virtio and DMA API is messy. * * On most systems with virtio, physical addresses match bus addresses, * and it doesn't particularly matter whether we use the DMA API. @@ -133,10 +136,18 @@ struct vring_virtqueue { * * For the time being, we preserve historic behavior and bypass the DMA * API. + * + * TODO: install a per-device DMA ops structure that does the right thing + * taking into account all the above quirks, and use the DMA API + * unconditionally on data path. */ static bool vring_use_dma_api(struct virtio_device *vdev) { + if (!virtio_has_iommu_quirk(vdev)) + return true; + + /* Otherwise, we are left to guess. */ /* * In theory, it's possible to have a buggy QEMU-supposed * emulated Q35 IOMMU and Xen enabled at the same time. On @@ -1099,6 +1110,8 @@ void vring_transport_features(struct virtio_device *vdev) break; case VIRTIO_F_VERSION_1: break; + case VIRTIO_F_IOMMU_PLATFORM: + break; default: /* We don't understand this bit. */ __virtio_clear_bit(vdev, i); diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h index 6e6cb0c9d7cb..26c155bb639b 100644 --- a/include/linux/virtio_config.h +++ b/include/linux/virtio_config.h @@ -149,6 +149,19 @@ static inline bool virtio_has_feature(const struct virtio_device *vdev, return __virtio_test_bit(vdev, fbit); } +/** + * virtio_has_iommu_quirk - determine whether this device has the iommu quirk + * @vdev: the device + */ +static inline bool virtio_has_iommu_quirk(const struct virtio_device *vdev) +{ + /* + * Note the reverse polarity of the quirk feature (compared to most + * other features), this is for compatibility with legacy systems. + */ + return !virtio_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM); +} + static inline struct virtqueue *virtio_find_single_vq(struct virtio_device *vdev, vq_callback_t *c, const char *n) diff --git a/include/uapi/linux/virtio_config.h b/include/uapi/linux/virtio_config.h index 4cb65bbfa654..308e2096291f 100644 --- a/include/uapi/linux/virtio_config.h +++ b/include/uapi/linux/virtio_config.h @@ -49,7 +49,7 @@ * transport being used (eg. virtio_ring), the rest are per-device feature * bits. */ #define VIRTIO_TRANSPORT_F_START 28 -#define VIRTIO_TRANSPORT_F_END 33 +#define VIRTIO_TRANSPORT_F_END 34 #ifndef VIRTIO_CONFIG_NO_LEGACY /* Do we get callbacks when the ring is completely used, even if we've @@ -63,4 +63,12 @@ /* v1.0 compliant. */ #define VIRTIO_F_VERSION_1 32 +/* + * If clear - device has the IOMMU bypass quirk feature. + * If set - use platform tools to detect the IOMMU. + * + * Note the reverse polarity (compared to most other features), + * this is for compatibility with legacy systems. + */ +#define VIRTIO_F_IOMMU_PLATFORM 33 #endif /* _UAPI_LINUX_VIRTIO_CONFIG_H */ -- cgit v1.2.3 From 06a8fc78367d070720af960dcecec917d3ae5f3b Mon Sep 17 00:00:00 2001 From: Asias He Date: Thu, 28 Jul 2016 15:36:32 +0100 Subject: VSOCK: Introduce virtio_vsock_common.ko This module contains the common code and header files for the following virtio_transporto and vhost_vsock kernel modules. Signed-off-by: Asias He Signed-off-by: Claudio Imbrenda Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- MAINTAINERS | 10 + include/linux/virtio_vsock.h | 154 ++++ include/net/af_vsock.h | 2 + .../trace/events/vsock_virtio_transport_common.h | 144 +++ include/uapi/linux/Kbuild | 1 + include/uapi/linux/virtio_ids.h | 1 + include/uapi/linux/virtio_vsock.h | 94 ++ net/vmw_vsock/virtio_transport_common.c | 992 +++++++++++++++++++++ 8 files changed, 1398 insertions(+) create mode 100644 include/linux/virtio_vsock.h create mode 100644 include/trace/events/vsock_virtio_transport_common.h create mode 100644 include/uapi/linux/virtio_vsock.h create mode 100644 net/vmw_vsock/virtio_transport_common.c (limited to 'include/uapi') diff --git a/MAINTAINERS b/MAINTAINERS index 8c20323d1277..b49ffb86ebf0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -12138,6 +12138,16 @@ S: Maintained F: drivers/media/v4l2-core/videobuf2-* F: include/media/videobuf2-* +VIRTIO AND VHOST VSOCK DRIVER +M: Stefan Hajnoczi +L: kvm@vger.kernel.org +L: virtualization@lists.linux-foundation.org +L: netdev@vger.kernel.org +S: Maintained +F: include/linux/virtio_vsock.h +F: include/uapi/linux/virtio_vsock.h +F: net/vmw_vsock/virtio_transport_common.c + VIRTUAL SERIO DEVICE DRIVER M: Stephen Chandler Paul S: Maintained diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h new file mode 100644 index 000000000000..9638bfeb0d1f --- /dev/null +++ b/include/linux/virtio_vsock.h @@ -0,0 +1,154 @@ +#ifndef _LINUX_VIRTIO_VSOCK_H +#define _LINUX_VIRTIO_VSOCK_H + +#include +#include +#include +#include + +#define VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE 128 +#define VIRTIO_VSOCK_DEFAULT_BUF_SIZE (1024 * 256) +#define VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE (1024 * 256) +#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE (1024 * 4) +#define VIRTIO_VSOCK_MAX_BUF_SIZE 0xFFFFFFFFUL +#define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE (1024 * 64) + +enum { + VSOCK_VQ_RX = 0, /* for host to guest data */ + VSOCK_VQ_TX = 1, /* for guest to host data */ + VSOCK_VQ_EVENT = 2, + VSOCK_VQ_MAX = 3, +}; + +/* Per-socket state (accessed via vsk->trans) */ +struct virtio_vsock_sock { + struct vsock_sock *vsk; + + /* Protected by lock_sock(sk_vsock(trans->vsk)) */ + u32 buf_size; + u32 buf_size_min; + u32 buf_size_max; + + spinlock_t tx_lock; + spinlock_t rx_lock; + + /* Protected by tx_lock */ + u32 tx_cnt; + u32 buf_alloc; + u32 peer_fwd_cnt; + u32 peer_buf_alloc; + + /* Protected by rx_lock */ + u32 fwd_cnt; + u32 rx_bytes; + struct list_head rx_queue; +}; + +struct virtio_vsock_pkt { + struct virtio_vsock_hdr hdr; + struct work_struct work; + struct list_head list; + void *buf; + u32 len; + u32 off; + bool reply; +}; + +struct virtio_vsock_pkt_info { + u32 remote_cid, remote_port; + struct msghdr *msg; + u32 pkt_len; + u16 type; + u16 op; + u32 flags; + bool reply; +}; + +struct virtio_transport { + /* This must be the first field */ + struct vsock_transport transport; + + /* Takes ownership of the packet */ + int (*send_pkt)(struct virtio_vsock_pkt *pkt); +}; + +ssize_t +virtio_transport_stream_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, + int type); +int +virtio_transport_dgram_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags); + +s64 virtio_transport_stream_has_data(struct vsock_sock *vsk); +s64 virtio_transport_stream_has_space(struct vsock_sock *vsk); + +int virtio_transport_do_socket_init(struct vsock_sock *vsk, + struct vsock_sock *psk); +u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk); +u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk); +u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk); +void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val); +void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val); +void virtio_transport_set_max_buffer_size(struct vsock_sock *vs, u64 val); +int +virtio_transport_notify_poll_in(struct vsock_sock *vsk, + size_t target, + bool *data_ready_now); +int +virtio_transport_notify_poll_out(struct vsock_sock *vsk, + size_t target, + bool *space_available_now); + +int virtio_transport_notify_recv_init(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, + size_t target, ssize_t copied, bool data_read, + struct vsock_transport_recv_notify_data *data); +int virtio_transport_notify_send_init(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data); +int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data); +int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data); +int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, + ssize_t written, struct vsock_transport_send_notify_data *data); + +u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk); +bool virtio_transport_stream_is_active(struct vsock_sock *vsk); +bool virtio_transport_stream_allow(u32 cid, u32 port); +int virtio_transport_dgram_bind(struct vsock_sock *vsk, + struct sockaddr_vm *addr); +bool virtio_transport_dgram_allow(u32 cid, u32 port); + +int virtio_transport_connect(struct vsock_sock *vsk); + +int virtio_transport_shutdown(struct vsock_sock *vsk, int mode); + +void virtio_transport_release(struct vsock_sock *vsk); + +ssize_t +virtio_transport_stream_enqueue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len); +int +virtio_transport_dgram_enqueue(struct vsock_sock *vsk, + struct sockaddr_vm *remote_addr, + struct msghdr *msg, + size_t len); + +void virtio_transport_destruct(struct vsock_sock *vsk); + +void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt); +void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt); +void virtio_transport_inc_tx_pkt(struct virtio_vsock_sock *vvs, struct virtio_vsock_pkt *pkt); +u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 wanted); +void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit); + +#endif /* _LINUX_VIRTIO_VSOCK_H */ diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index 3af0b224f754..f2758964ce6f 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -63,6 +63,8 @@ struct vsock_sock { struct list_head accept_queue; bool rejected; struct delayed_work dwork; + struct delayed_work close_work; + bool close_work_scheduled; u32 peer_shutdown; bool sent_request; bool ignore_connecting_rst; diff --git a/include/trace/events/vsock_virtio_transport_common.h b/include/trace/events/vsock_virtio_transport_common.h new file mode 100644 index 000000000000..b7f1d6278280 --- /dev/null +++ b/include/trace/events/vsock_virtio_transport_common.h @@ -0,0 +1,144 @@ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM vsock + +#if !defined(_TRACE_VSOCK_VIRTIO_TRANSPORT_COMMON_H) || \ + defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_VSOCK_VIRTIO_TRANSPORT_COMMON_H + +#include + +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_TYPE_STREAM); + +#define show_type(val) \ + __print_symbolic(val, { VIRTIO_VSOCK_TYPE_STREAM, "STREAM" }) + +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_INVALID); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_REQUEST); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_RESPONSE); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_RST); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_SHUTDOWN); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_RW); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_CREDIT_UPDATE); +TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_CREDIT_REQUEST); + +#define show_op(val) \ + __print_symbolic(val, \ + { VIRTIO_VSOCK_OP_INVALID, "INVALID" }, \ + { VIRTIO_VSOCK_OP_REQUEST, "REQUEST" }, \ + { VIRTIO_VSOCK_OP_RESPONSE, "RESPONSE" }, \ + { VIRTIO_VSOCK_OP_RST, "RST" }, \ + { VIRTIO_VSOCK_OP_SHUTDOWN, "SHUTDOWN" }, \ + { VIRTIO_VSOCK_OP_RW, "RW" }, \ + { VIRTIO_VSOCK_OP_CREDIT_UPDATE, "CREDIT_UPDATE" }, \ + { VIRTIO_VSOCK_OP_CREDIT_REQUEST, "CREDIT_REQUEST" }) + +TRACE_EVENT(virtio_transport_alloc_pkt, + TP_PROTO( + __u32 src_cid, __u32 src_port, + __u32 dst_cid, __u32 dst_port, + __u32 len, + __u16 type, + __u16 op, + __u32 flags + ), + TP_ARGS( + src_cid, src_port, + dst_cid, dst_port, + len, + type, + op, + flags + ), + TP_STRUCT__entry( + __field(__u32, src_cid) + __field(__u32, src_port) + __field(__u32, dst_cid) + __field(__u32, dst_port) + __field(__u32, len) + __field(__u16, type) + __field(__u16, op) + __field(__u32, flags) + ), + TP_fast_assign( + __entry->src_cid = src_cid; + __entry->src_port = src_port; + __entry->dst_cid = dst_cid; + __entry->dst_port = dst_port; + __entry->len = len; + __entry->type = type; + __entry->op = op; + __entry->flags = flags; + ), + TP_printk("%u:%u -> %u:%u len=%u type=%s op=%s flags=%#x", + __entry->src_cid, __entry->src_port, + __entry->dst_cid, __entry->dst_port, + __entry->len, + show_type(__entry->type), + show_op(__entry->op), + __entry->flags) +); + +TRACE_EVENT(virtio_transport_recv_pkt, + TP_PROTO( + __u32 src_cid, __u32 src_port, + __u32 dst_cid, __u32 dst_port, + __u32 len, + __u16 type, + __u16 op, + __u32 flags, + __u32 buf_alloc, + __u32 fwd_cnt + ), + TP_ARGS( + src_cid, src_port, + dst_cid, dst_port, + len, + type, + op, + flags, + buf_alloc, + fwd_cnt + ), + TP_STRUCT__entry( + __field(__u32, src_cid) + __field(__u32, src_port) + __field(__u32, dst_cid) + __field(__u32, dst_port) + __field(__u32, len) + __field(__u16, type) + __field(__u16, op) + __field(__u32, flags) + __field(__u32, buf_alloc) + __field(__u32, fwd_cnt) + ), + TP_fast_assign( + __entry->src_cid = src_cid; + __entry->src_port = src_port; + __entry->dst_cid = dst_cid; + __entry->dst_port = dst_port; + __entry->len = len; + __entry->type = type; + __entry->op = op; + __entry->flags = flags; + __entry->buf_alloc = buf_alloc; + __entry->fwd_cnt = fwd_cnt; + ), + TP_printk("%u:%u -> %u:%u len=%u type=%s op=%s flags=%#x " + "buf_alloc=%u fwd_cnt=%u", + __entry->src_cid, __entry->src_port, + __entry->dst_cid, __entry->dst_port, + __entry->len, + show_type(__entry->type), + show_op(__entry->op), + __entry->flags, + __entry->buf_alloc, + __entry->fwd_cnt) +); + +#endif /* _TRACE_VSOCK_VIRTIO_TRANSPORT_COMMON_H */ + +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE vsock_virtio_transport_common + +/* This part must be outside protection */ +#include diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild index ec10cfef166a..3cf0116d9c2b 100644 --- a/include/uapi/linux/Kbuild +++ b/include/uapi/linux/Kbuild @@ -453,6 +453,7 @@ header-y += virtio_ring.h header-y += virtio_rng.h header-y += virtio_scsi.h header-y += virtio_types.h +header-y += virtio_vsock.h header-y += vm_sockets.h header-y += vt.h header-y += wait.h diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h index 77925f587b15..3228d582234a 100644 --- a/include/uapi/linux/virtio_ids.h +++ b/include/uapi/linux/virtio_ids.h @@ -41,5 +41,6 @@ #define VIRTIO_ID_CAIF 12 /* Virtio caif */ #define VIRTIO_ID_GPU 16 /* virtio GPU */ #define VIRTIO_ID_INPUT 18 /* virtio input */ +#define VIRTIO_ID_VSOCK 19 /* virtio vsock transport */ #endif /* _LINUX_VIRTIO_IDS_H */ diff --git a/include/uapi/linux/virtio_vsock.h b/include/uapi/linux/virtio_vsock.h new file mode 100644 index 000000000000..6b011c19b50f --- /dev/null +++ b/include/uapi/linux/virtio_vsock.h @@ -0,0 +1,94 @@ +/* + * This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so + * anyone can use the definitions to implement compatible drivers/servers: + * + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of IBM nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS IS'' + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL IBM OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + * + * Copyright (C) Red Hat, Inc., 2013-2015 + * Copyright (C) Asias He , 2013 + * Copyright (C) Stefan Hajnoczi , 2015 + */ + +#ifndef _UAPI_LINUX_VIRTIO_VSOCK_H +#define _UAPI_LINUX_VIRTIO_VOSCK_H + +#include +#include +#include + +struct virtio_vsock_config { + __le64 guest_cid; +} __attribute__((packed)); + +enum virtio_vsock_event_id { + VIRTIO_VSOCK_EVENT_TRANSPORT_RESET = 0, +}; + +struct virtio_vsock_event { + __le32 id; +} __attribute__((packed)); + +struct virtio_vsock_hdr { + __le64 src_cid; + __le64 dst_cid; + __le32 src_port; + __le32 dst_port; + __le32 len; + __le16 type; /* enum virtio_vsock_type */ + __le16 op; /* enum virtio_vsock_op */ + __le32 flags; + __le32 buf_alloc; + __le32 fwd_cnt; +} __attribute__((packed)); + +enum virtio_vsock_type { + VIRTIO_VSOCK_TYPE_STREAM = 1, +}; + +enum virtio_vsock_op { + VIRTIO_VSOCK_OP_INVALID = 0, + + /* Connect operations */ + VIRTIO_VSOCK_OP_REQUEST = 1, + VIRTIO_VSOCK_OP_RESPONSE = 2, + VIRTIO_VSOCK_OP_RST = 3, + VIRTIO_VSOCK_OP_SHUTDOWN = 4, + + /* To send payload */ + VIRTIO_VSOCK_OP_RW = 5, + + /* Tell the peer our credit info */ + VIRTIO_VSOCK_OP_CREDIT_UPDATE = 6, + /* Request the peer to send the credit info to us */ + VIRTIO_VSOCK_OP_CREDIT_REQUEST = 7, +}; + +/* VIRTIO_VSOCK_OP_SHUTDOWN flags values */ +enum virtio_vsock_shutdown { + VIRTIO_VSOCK_SHUTDOWN_RCV = 1, + VIRTIO_VSOCK_SHUTDOWN_SEND = 2, +}; + +#endif /* _UAPI_LINUX_VIRTIO_VSOCK_H */ diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c new file mode 100644 index 000000000000..a53b3a16b4f1 --- /dev/null +++ b/net/vmw_vsock/virtio_transport_common.c @@ -0,0 +1,992 @@ +/* + * common code for virtio vsock + * + * Copyright (C) 2013-2015 Red Hat, Inc. + * Author: Asias He + * Stefan Hajnoczi + * + * This work is licensed under the terms of the GNU GPL, version 2. + */ +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#define CREATE_TRACE_POINTS +#include + +/* How long to wait for graceful shutdown of a connection */ +#define VSOCK_CLOSE_TIMEOUT (8 * HZ) + +static const struct virtio_transport *virtio_transport_get_ops(void) +{ + const struct vsock_transport *t = vsock_core_get_transport(); + + return container_of(t, struct virtio_transport, transport); +} + +struct virtio_vsock_pkt * +virtio_transport_alloc_pkt(struct virtio_vsock_pkt_info *info, + size_t len, + u32 src_cid, + u32 src_port, + u32 dst_cid, + u32 dst_port) +{ + struct virtio_vsock_pkt *pkt; + int err; + + pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); + if (!pkt) + return NULL; + + pkt->hdr.type = cpu_to_le16(info->type); + pkt->hdr.op = cpu_to_le16(info->op); + pkt->hdr.src_cid = cpu_to_le64(src_cid); + pkt->hdr.dst_cid = cpu_to_le64(dst_cid); + pkt->hdr.src_port = cpu_to_le32(src_port); + pkt->hdr.dst_port = cpu_to_le32(dst_port); + pkt->hdr.flags = cpu_to_le32(info->flags); + pkt->len = len; + pkt->hdr.len = cpu_to_le32(len); + pkt->reply = info->reply; + + if (info->msg && len > 0) { + pkt->buf = kmalloc(len, GFP_KERNEL); + if (!pkt->buf) + goto out_pkt; + err = memcpy_from_msg(pkt->buf, info->msg, len); + if (err) + goto out; + } + + trace_virtio_transport_alloc_pkt(src_cid, src_port, + dst_cid, dst_port, + len, + info->type, + info->op, + info->flags); + + return pkt; + +out: + kfree(pkt->buf); +out_pkt: + kfree(pkt); + return NULL; +} +EXPORT_SYMBOL_GPL(virtio_transport_alloc_pkt); + +static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, + struct virtio_vsock_pkt_info *info) +{ + u32 src_cid, src_port, dst_cid, dst_port; + struct virtio_vsock_sock *vvs; + struct virtio_vsock_pkt *pkt; + u32 pkt_len = info->pkt_len; + + src_cid = vm_sockets_get_local_cid(); + src_port = vsk->local_addr.svm_port; + if (!info->remote_cid) { + dst_cid = vsk->remote_addr.svm_cid; + dst_port = vsk->remote_addr.svm_port; + } else { + dst_cid = info->remote_cid; + dst_port = info->remote_port; + } + + vvs = vsk->trans; + + /* we can send less than pkt_len bytes */ + if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE) + pkt_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE; + + /* virtio_transport_get_credit might return less than pkt_len credit */ + pkt_len = virtio_transport_get_credit(vvs, pkt_len); + + /* Do not send zero length OP_RW pkt */ + if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW) + return pkt_len; + + pkt = virtio_transport_alloc_pkt(info, pkt_len, + src_cid, src_port, + dst_cid, dst_port); + if (!pkt) { + virtio_transport_put_credit(vvs, pkt_len); + return -ENOMEM; + } + + virtio_transport_inc_tx_pkt(vvs, pkt); + + return virtio_transport_get_ops()->send_pkt(pkt); +} + +static void virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, + struct virtio_vsock_pkt *pkt) +{ + vvs->rx_bytes += pkt->len; +} + +static void virtio_transport_dec_rx_pkt(struct virtio_vsock_sock *vvs, + struct virtio_vsock_pkt *pkt) +{ + vvs->rx_bytes -= pkt->len; + vvs->fwd_cnt += pkt->len; +} + +void virtio_transport_inc_tx_pkt(struct virtio_vsock_sock *vvs, struct virtio_vsock_pkt *pkt) +{ + spin_lock_bh(&vvs->tx_lock); + pkt->hdr.fwd_cnt = cpu_to_le32(vvs->fwd_cnt); + pkt->hdr.buf_alloc = cpu_to_le32(vvs->buf_alloc); + spin_unlock_bh(&vvs->tx_lock); +} +EXPORT_SYMBOL_GPL(virtio_transport_inc_tx_pkt); + +u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 credit) +{ + u32 ret; + + spin_lock_bh(&vvs->tx_lock); + ret = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt); + if (ret > credit) + ret = credit; + vvs->tx_cnt += ret; + spin_unlock_bh(&vvs->tx_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_credit); + +void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit) +{ + spin_lock_bh(&vvs->tx_lock); + vvs->tx_cnt -= credit; + spin_unlock_bh(&vvs->tx_lock); +} +EXPORT_SYMBOL_GPL(virtio_transport_put_credit); + +static int virtio_transport_send_credit_update(struct vsock_sock *vsk, + int type, + struct virtio_vsock_hdr *hdr) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_CREDIT_UPDATE, + .type = type, + }; + + return virtio_transport_send_pkt_info(vsk, &info); +} + +static ssize_t +virtio_transport_stream_do_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + struct virtio_vsock_pkt *pkt; + size_t bytes, total = 0; + int err = -EFAULT; + + spin_lock_bh(&vvs->rx_lock); + while (total < len && !list_empty(&vvs->rx_queue)) { + pkt = list_first_entry(&vvs->rx_queue, + struct virtio_vsock_pkt, list); + + bytes = len - total; + if (bytes > pkt->len - pkt->off) + bytes = pkt->len - pkt->off; + + /* sk_lock is held by caller so no one else can dequeue. + * Unlock rx_lock since memcpy_to_msg() may sleep. + */ + spin_unlock_bh(&vvs->rx_lock); + + err = memcpy_to_msg(msg, pkt->buf + pkt->off, bytes); + if (err) + goto out; + + spin_lock_bh(&vvs->rx_lock); + + total += bytes; + pkt->off += bytes; + if (pkt->off == pkt->len) { + virtio_transport_dec_rx_pkt(vvs, pkt); + list_del(&pkt->list); + virtio_transport_free_pkt(pkt); + } + } + spin_unlock_bh(&vvs->rx_lock); + + /* Send a credit pkt to peer */ + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM, + NULL); + + return total; + +out: + if (total) + err = total; + return err; +} + +ssize_t +virtio_transport_stream_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags) +{ + if (flags & MSG_PEEK) + return -EOPNOTSUPP; + + return virtio_transport_stream_do_dequeue(vsk, msg, len); +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue); + +int +virtio_transport_dgram_dequeue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len, int flags) +{ + return -EOPNOTSUPP; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_dequeue); + +s64 virtio_transport_stream_has_data(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + s64 bytes; + + spin_lock_bh(&vvs->rx_lock); + bytes = vvs->rx_bytes; + spin_unlock_bh(&vvs->rx_lock); + + return bytes; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_has_data); + +static s64 virtio_transport_has_space(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + s64 bytes; + + bytes = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt); + if (bytes < 0) + bytes = 0; + + return bytes; +} + +s64 virtio_transport_stream_has_space(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + s64 bytes; + + spin_lock_bh(&vvs->tx_lock); + bytes = virtio_transport_has_space(vsk); + spin_unlock_bh(&vvs->tx_lock); + + return bytes; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_has_space); + +int virtio_transport_do_socket_init(struct vsock_sock *vsk, + struct vsock_sock *psk) +{ + struct virtio_vsock_sock *vvs; + + vvs = kzalloc(sizeof(*vvs), GFP_KERNEL); + if (!vvs) + return -ENOMEM; + + vsk->trans = vvs; + vvs->vsk = vsk; + if (psk) { + struct virtio_vsock_sock *ptrans = psk->trans; + + vvs->buf_size = ptrans->buf_size; + vvs->buf_size_min = ptrans->buf_size_min; + vvs->buf_size_max = ptrans->buf_size_max; + vvs->peer_buf_alloc = ptrans->peer_buf_alloc; + } else { + vvs->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE; + vvs->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE; + vvs->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE; + } + + vvs->buf_alloc = vvs->buf_size; + + spin_lock_init(&vvs->rx_lock); + spin_lock_init(&vvs->tx_lock); + INIT_LIST_HEAD(&vvs->rx_queue); + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init); + +u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + return vvs->buf_size; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size); + +u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + return vvs->buf_size_min; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size); + +u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + return vvs->buf_size_max; +} +EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size); + +void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val < vvs->buf_size_min) + vvs->buf_size_min = val; + if (val > vvs->buf_size_max) + vvs->buf_size_max = val; + vvs->buf_size = val; + vvs->buf_alloc = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size); + +void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val > vvs->buf_size) + vvs->buf_size = val; + vvs->buf_size_min = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size); + +void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + if (val > VIRTIO_VSOCK_MAX_BUF_SIZE) + val = VIRTIO_VSOCK_MAX_BUF_SIZE; + if (val < vvs->buf_size) + vvs->buf_size = val; + vvs->buf_size_max = val; +} +EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size); + +int +virtio_transport_notify_poll_in(struct vsock_sock *vsk, + size_t target, + bool *data_ready_now) +{ + if (vsock_stream_has_data(vsk)) + *data_ready_now = true; + else + *data_ready_now = false; + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_in); + +int +virtio_transport_notify_poll_out(struct vsock_sock *vsk, + size_t target, + bool *space_avail_now) +{ + s64 free_space; + + free_space = vsock_stream_has_space(vsk); + if (free_space > 0) + *space_avail_now = true; + else if (free_space == 0) + *space_avail_now = false; + + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_out); + +int virtio_transport_notify_recv_init(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_init); + +int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_block); + +int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk, + size_t target, struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_dequeue); + +int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk, + size_t target, ssize_t copied, bool data_read, + struct vsock_transport_recv_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_post_dequeue); + +int virtio_transport_notify_send_init(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_init); + +int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_block); + +int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk, + struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_enqueue); + +int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk, + ssize_t written, struct vsock_transport_send_notify_data *data) +{ + return 0; +} +EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue); + +u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + return vvs->buf_size; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat); + +bool virtio_transport_stream_is_active(struct vsock_sock *vsk) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_is_active); + +bool virtio_transport_stream_allow(u32 cid, u32 port) +{ + return true; +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_allow); + +int virtio_transport_dgram_bind(struct vsock_sock *vsk, + struct sockaddr_vm *addr) +{ + return -EOPNOTSUPP; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_bind); + +bool virtio_transport_dgram_allow(u32 cid, u32 port) +{ + return false; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_allow); + +int virtio_transport_connect(struct vsock_sock *vsk) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_REQUEST, + .type = VIRTIO_VSOCK_TYPE_STREAM, + }; + + return virtio_transport_send_pkt_info(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_connect); + +int virtio_transport_shutdown(struct vsock_sock *vsk, int mode) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_SHUTDOWN, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .flags = (mode & RCV_SHUTDOWN ? + VIRTIO_VSOCK_SHUTDOWN_RCV : 0) | + (mode & SEND_SHUTDOWN ? + VIRTIO_VSOCK_SHUTDOWN_SEND : 0), + }; + + return virtio_transport_send_pkt_info(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_shutdown); + +int +virtio_transport_dgram_enqueue(struct vsock_sock *vsk, + struct sockaddr_vm *remote_addr, + struct msghdr *msg, + size_t dgram_len) +{ + return -EOPNOTSUPP; +} +EXPORT_SYMBOL_GPL(virtio_transport_dgram_enqueue); + +ssize_t +virtio_transport_stream_enqueue(struct vsock_sock *vsk, + struct msghdr *msg, + size_t len) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RW, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .msg = msg, + .pkt_len = len, + }; + + return virtio_transport_send_pkt_info(vsk, &info); +} +EXPORT_SYMBOL_GPL(virtio_transport_stream_enqueue); + +void virtio_transport_destruct(struct vsock_sock *vsk) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + + kfree(vvs); +} +EXPORT_SYMBOL_GPL(virtio_transport_destruct); + +static int virtio_transport_reset(struct vsock_sock *vsk, + struct virtio_vsock_pkt *pkt) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RST, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .reply = !!pkt, + }; + + /* Send RST only if the original pkt is not a RST pkt */ + if (pkt && le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST) + return 0; + + return virtio_transport_send_pkt_info(vsk, &info); +} + +/* Normally packets are associated with a socket. There may be no socket if an + * attempt was made to connect to a socket that does not exist. + */ +static int virtio_transport_reset_no_sock(struct virtio_vsock_pkt *pkt) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RST, + .type = le16_to_cpu(pkt->hdr.type), + .reply = true, + }; + + /* Send RST only if the original pkt is not a RST pkt */ + if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST) + return 0; + + pkt = virtio_transport_alloc_pkt(&info, 0, + le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port), + le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port)); + if (!pkt) + return -ENOMEM; + + return virtio_transport_get_ops()->send_pkt(pkt); +} + +static void virtio_transport_wait_close(struct sock *sk, long timeout) +{ + if (timeout) { + DEFINE_WAIT(wait); + + do { + prepare_to_wait(sk_sleep(sk), &wait, + TASK_INTERRUPTIBLE); + if (sk_wait_event(sk, &timeout, + sock_flag(sk, SOCK_DONE))) + break; + } while (!signal_pending(current) && timeout); + + finish_wait(sk_sleep(sk), &wait); + } +} + +static void virtio_transport_do_close(struct vsock_sock *vsk, + bool cancel_timeout) +{ + struct sock *sk = sk_vsock(vsk); + + sock_set_flag(sk, SOCK_DONE); + vsk->peer_shutdown = SHUTDOWN_MASK; + if (vsock_stream_has_data(vsk) <= 0) + sk->sk_state = SS_DISCONNECTING; + sk->sk_state_change(sk); + + if (vsk->close_work_scheduled && + (!cancel_timeout || cancel_delayed_work(&vsk->close_work))) { + vsk->close_work_scheduled = false; + + vsock_remove_sock(vsk); + + /* Release refcnt obtained when we scheduled the timeout */ + sock_put(sk); + } +} + +static void virtio_transport_close_timeout(struct work_struct *work) +{ + struct vsock_sock *vsk = + container_of(work, struct vsock_sock, close_work.work); + struct sock *sk = sk_vsock(vsk); + + sock_hold(sk); + lock_sock(sk); + + if (!sock_flag(sk, SOCK_DONE)) { + (void)virtio_transport_reset(vsk, NULL); + + virtio_transport_do_close(vsk, false); + } + + vsk->close_work_scheduled = false; + + release_sock(sk); + sock_put(sk); +} + +/* User context, vsk->sk is locked */ +static bool virtio_transport_close(struct vsock_sock *vsk) +{ + struct sock *sk = &vsk->sk; + + if (!(sk->sk_state == SS_CONNECTED || + sk->sk_state == SS_DISCONNECTING)) + return true; + + /* Already received SHUTDOWN from peer, reply with RST */ + if ((vsk->peer_shutdown & SHUTDOWN_MASK) == SHUTDOWN_MASK) { + (void)virtio_transport_reset(vsk, NULL); + return true; + } + + if ((sk->sk_shutdown & SHUTDOWN_MASK) != SHUTDOWN_MASK) + (void)virtio_transport_shutdown(vsk, SHUTDOWN_MASK); + + if (sock_flag(sk, SOCK_LINGER) && !(current->flags & PF_EXITING)) + virtio_transport_wait_close(sk, sk->sk_lingertime); + + if (sock_flag(sk, SOCK_DONE)) { + return true; + } + + sock_hold(sk); + INIT_DELAYED_WORK(&vsk->close_work, + virtio_transport_close_timeout); + vsk->close_work_scheduled = true; + schedule_delayed_work(&vsk->close_work, VSOCK_CLOSE_TIMEOUT); + return false; +} + +void virtio_transport_release(struct vsock_sock *vsk) +{ + struct sock *sk = &vsk->sk; + bool remove_sock = true; + + lock_sock(sk); + if (sk->sk_type == SOCK_STREAM) + remove_sock = virtio_transport_close(vsk); + release_sock(sk); + + if (remove_sock) + vsock_remove_sock(vsk); +} +EXPORT_SYMBOL_GPL(virtio_transport_release); + +static int +virtio_transport_recv_connecting(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + int err; + int skerr; + + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_RESPONSE: + sk->sk_state = SS_CONNECTED; + sk->sk_socket->state = SS_CONNECTED; + vsock_insert_connected(vsk); + sk->sk_state_change(sk); + break; + case VIRTIO_VSOCK_OP_INVALID: + break; + case VIRTIO_VSOCK_OP_RST: + skerr = ECONNRESET; + err = 0; + goto destroy; + default: + skerr = EPROTO; + err = -EINVAL; + goto destroy; + } + return 0; + +destroy: + virtio_transport_reset(vsk, pkt); + sk->sk_state = SS_UNCONNECTED; + sk->sk_err = skerr; + sk->sk_error_report(sk); + return err; +} + +static int +virtio_transport_recv_connected(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct virtio_vsock_sock *vvs = vsk->trans; + int err = 0; + + switch (le16_to_cpu(pkt->hdr.op)) { + case VIRTIO_VSOCK_OP_RW: + pkt->len = le32_to_cpu(pkt->hdr.len); + pkt->off = 0; + + spin_lock_bh(&vvs->rx_lock); + virtio_transport_inc_rx_pkt(vvs, pkt); + list_add_tail(&pkt->list, &vvs->rx_queue); + spin_unlock_bh(&vvs->rx_lock); + + sk->sk_data_ready(sk); + return err; + case VIRTIO_VSOCK_OP_CREDIT_UPDATE: + sk->sk_write_space(sk); + break; + case VIRTIO_VSOCK_OP_SHUTDOWN: + if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_RCV) + vsk->peer_shutdown |= RCV_SHUTDOWN; + if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_SEND) + vsk->peer_shutdown |= SEND_SHUTDOWN; + if (vsk->peer_shutdown == SHUTDOWN_MASK && + vsock_stream_has_data(vsk) <= 0) + sk->sk_state = SS_DISCONNECTING; + if (le32_to_cpu(pkt->hdr.flags)) + sk->sk_state_change(sk); + break; + case VIRTIO_VSOCK_OP_RST: + virtio_transport_do_close(vsk, true); + break; + default: + err = -EINVAL; + break; + } + + virtio_transport_free_pkt(pkt); + return err; +} + +static void +virtio_transport_recv_disconnecting(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + + if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST) + virtio_transport_do_close(vsk, true); +} + +static int +virtio_transport_send_response(struct vsock_sock *vsk, + struct virtio_vsock_pkt *pkt) +{ + struct virtio_vsock_pkt_info info = { + .op = VIRTIO_VSOCK_OP_RESPONSE, + .type = VIRTIO_VSOCK_TYPE_STREAM, + .remote_cid = le32_to_cpu(pkt->hdr.src_cid), + .remote_port = le32_to_cpu(pkt->hdr.src_port), + .reply = true, + }; + + return virtio_transport_send_pkt_info(vsk, &info); +} + +/* Handle server socket */ +static int +virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct vsock_sock *vchild; + struct sock *child; + + if (le16_to_cpu(pkt->hdr.op) != VIRTIO_VSOCK_OP_REQUEST) { + virtio_transport_reset(vsk, pkt); + return -EINVAL; + } + + if (sk_acceptq_is_full(sk)) { + virtio_transport_reset(vsk, pkt); + return -ENOMEM; + } + + child = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL, + sk->sk_type, 0); + if (!child) { + virtio_transport_reset(vsk, pkt); + return -ENOMEM; + } + + sk->sk_ack_backlog++; + + lock_sock_nested(child, SINGLE_DEPTH_NESTING); + + child->sk_state = SS_CONNECTED; + + vchild = vsock_sk(child); + vsock_addr_init(&vchild->local_addr, le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port)); + vsock_addr_init(&vchild->remote_addr, le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port)); + + vsock_insert_connected(vchild); + vsock_enqueue_accept(sk, child); + virtio_transport_send_response(vchild, pkt); + + release_sock(child); + + sk->sk_data_ready(sk); + return 0; +} + +static bool virtio_transport_space_update(struct sock *sk, + struct virtio_vsock_pkt *pkt) +{ + struct vsock_sock *vsk = vsock_sk(sk); + struct virtio_vsock_sock *vvs = vsk->trans; + bool space_available; + + /* buf_alloc and fwd_cnt is always included in the hdr */ + spin_lock_bh(&vvs->tx_lock); + vvs->peer_buf_alloc = le32_to_cpu(pkt->hdr.buf_alloc); + vvs->peer_fwd_cnt = le32_to_cpu(pkt->hdr.fwd_cnt); + space_available = virtio_transport_has_space(vsk); + spin_unlock_bh(&vvs->tx_lock); + return space_available; +} + +/* We are under the virtio-vsock's vsock->rx_lock or vhost-vsock's vq->mutex + * lock. + */ +void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt) +{ + struct sockaddr_vm src, dst; + struct vsock_sock *vsk; + struct sock *sk; + bool space_available; + + vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid), + le32_to_cpu(pkt->hdr.src_port)); + vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid), + le32_to_cpu(pkt->hdr.dst_port)); + + trace_virtio_transport_recv_pkt(src.svm_cid, src.svm_port, + dst.svm_cid, dst.svm_port, + le32_to_cpu(pkt->hdr.len), + le16_to_cpu(pkt->hdr.type), + le16_to_cpu(pkt->hdr.op), + le32_to_cpu(pkt->hdr.flags), + le32_to_cpu(pkt->hdr.buf_alloc), + le32_to_cpu(pkt->hdr.fwd_cnt)); + + if (le16_to_cpu(pkt->hdr.type) != VIRTIO_VSOCK_TYPE_STREAM) { + (void)virtio_transport_reset_no_sock(pkt); + goto free_pkt; + } + + /* The socket must be in connected or bound table + * otherwise send reset back + */ + sk = vsock_find_connected_socket(&src, &dst); + if (!sk) { + sk = vsock_find_bound_socket(&dst); + if (!sk) { + (void)virtio_transport_reset_no_sock(pkt); + goto free_pkt; + } + } + + vsk = vsock_sk(sk); + + space_available = virtio_transport_space_update(sk, pkt); + + lock_sock(sk); + + /* Update CID in case it has changed after a transport reset event */ + vsk->local_addr.svm_cid = dst.svm_cid; + + if (space_available) + sk->sk_write_space(sk); + + switch (sk->sk_state) { + case VSOCK_SS_LISTEN: + virtio_transport_recv_listen(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + case SS_CONNECTING: + virtio_transport_recv_connecting(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + case SS_CONNECTED: + virtio_transport_recv_connected(sk, pkt); + break; + case SS_DISCONNECTING: + virtio_transport_recv_disconnecting(sk, pkt); + virtio_transport_free_pkt(pkt); + break; + default: + virtio_transport_free_pkt(pkt); + break; + } + release_sock(sk); + + /* Release refcnt obtained when we fetched this socket out of the + * bound or connected list. + */ + sock_put(sk); + return; + +free_pkt: + virtio_transport_free_pkt(pkt); +} +EXPORT_SYMBOL_GPL(virtio_transport_recv_pkt); + +void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt) +{ + kfree(pkt->buf); + kfree(pkt); +} +EXPORT_SYMBOL_GPL(virtio_transport_free_pkt); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Asias He"); +MODULE_DESCRIPTION("common code for virtio vsock"); -- cgit v1.2.3 From 433fc58e6bf2c8bd97e57153ed28e64fd78207b8 Mon Sep 17 00:00:00 2001 From: Asias He Date: Thu, 28 Jul 2016 15:36:34 +0100 Subject: VSOCK: Introduce vhost_vsock.ko VM sockets vhost transport implementation. This driver runs on the host. Signed-off-by: Asias He Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- MAINTAINERS | 2 + drivers/vhost/vsock.c | 722 +++++++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/vhost.h | 5 + 3 files changed, 729 insertions(+) create mode 100644 drivers/vhost/vsock.c (limited to 'include/uapi') diff --git a/MAINTAINERS b/MAINTAINERS index 7302663ae4c3..12c79e5dc27e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -12148,6 +12148,8 @@ F: include/linux/virtio_vsock.h F: include/uapi/linux/virtio_vsock.h F: net/vmw_vsock/virtio_transport_common.c F: net/vmw_vsock/virtio_transport.c +F: drivers/vhost/vsock.c +F: drivers/vhost/vsock.h VIRTUAL SERIO DEVICE DRIVER M: Stephen Chandler Paul diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c new file mode 100644 index 000000000000..028ca16c2d36 --- /dev/null +++ b/drivers/vhost/vsock.c @@ -0,0 +1,722 @@ +/* + * vhost transport for vsock + * + * Copyright (C) 2013-2015 Red Hat, Inc. + * Author: Asias He + * Stefan Hajnoczi + * + * This work is licensed under the terms of the GNU GPL, version 2. + */ +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include "vhost.h" + +#define VHOST_VSOCK_DEFAULT_HOST_CID 2 + +enum { + VHOST_VSOCK_FEATURES = VHOST_FEATURES, +}; + +/* Used to track all the vhost_vsock instances on the system. */ +static DEFINE_SPINLOCK(vhost_vsock_lock); +static LIST_HEAD(vhost_vsock_list); + +struct vhost_vsock { + struct vhost_dev dev; + struct vhost_virtqueue vqs[2]; + + /* Link to global vhost_vsock_list, protected by vhost_vsock_lock */ + struct list_head list; + + struct vhost_work send_pkt_work; + spinlock_t send_pkt_list_lock; + struct list_head send_pkt_list; /* host->guest pending packets */ + + atomic_t queued_replies; + + u32 guest_cid; +}; + +static u32 vhost_transport_get_local_cid(void) +{ + return VHOST_VSOCK_DEFAULT_HOST_CID; +} + +static struct vhost_vsock *vhost_vsock_get(u32 guest_cid) +{ + struct vhost_vsock *vsock; + + spin_lock_bh(&vhost_vsock_lock); + list_for_each_entry(vsock, &vhost_vsock_list, list) { + u32 other_cid = vsock->guest_cid; + + /* Skip instances that have no CID yet */ + if (other_cid == 0) + continue; + + if (other_cid == guest_cid) { + spin_unlock_bh(&vhost_vsock_lock); + return vsock; + } + } + spin_unlock_bh(&vhost_vsock_lock); + + return NULL; +} + +static void +vhost_transport_do_send_pkt(struct vhost_vsock *vsock, + struct vhost_virtqueue *vq) +{ + struct vhost_virtqueue *tx_vq = &vsock->vqs[VSOCK_VQ_TX]; + bool added = false; + bool restart_tx = false; + + mutex_lock(&vq->mutex); + + if (!vq->private_data) + goto out; + + /* Avoid further vmexits, we're already processing the virtqueue */ + vhost_disable_notify(&vsock->dev, vq); + + for (;;) { + struct virtio_vsock_pkt *pkt; + struct iov_iter iov_iter; + unsigned out, in; + size_t nbytes; + size_t len; + int head; + + spin_lock_bh(&vsock->send_pkt_list_lock); + if (list_empty(&vsock->send_pkt_list)) { + spin_unlock_bh(&vsock->send_pkt_list_lock); + vhost_enable_notify(&vsock->dev, vq); + break; + } + + pkt = list_first_entry(&vsock->send_pkt_list, + struct virtio_vsock_pkt, list); + list_del_init(&pkt->list); + spin_unlock_bh(&vsock->send_pkt_list_lock); + + head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov), + &out, &in, NULL, NULL); + if (head < 0) { + spin_lock_bh(&vsock->send_pkt_list_lock); + list_add(&pkt->list, &vsock->send_pkt_list); + spin_unlock_bh(&vsock->send_pkt_list_lock); + break; + } + + if (head == vq->num) { + spin_lock_bh(&vsock->send_pkt_list_lock); + list_add(&pkt->list, &vsock->send_pkt_list); + spin_unlock_bh(&vsock->send_pkt_list_lock); + + /* We cannot finish yet if more buffers snuck in while + * re-enabling notify. + */ + if (unlikely(vhost_enable_notify(&vsock->dev, vq))) { + vhost_disable_notify(&vsock->dev, vq); + continue; + } + break; + } + + if (out) { + virtio_transport_free_pkt(pkt); + vq_err(vq, "Expected 0 output buffers, got %u\n", out); + break; + } + + len = iov_length(&vq->iov[out], in); + iov_iter_init(&iov_iter, READ, &vq->iov[out], in, len); + + nbytes = copy_to_iter(&pkt->hdr, sizeof(pkt->hdr), &iov_iter); + if (nbytes != sizeof(pkt->hdr)) { + virtio_transport_free_pkt(pkt); + vq_err(vq, "Faulted on copying pkt hdr\n"); + break; + } + + nbytes = copy_to_iter(pkt->buf, pkt->len, &iov_iter); + if (nbytes != pkt->len) { + virtio_transport_free_pkt(pkt); + vq_err(vq, "Faulted on copying pkt buf\n"); + break; + } + + vhost_add_used(vq, head, sizeof(pkt->hdr) + pkt->len); + added = true; + + if (pkt->reply) { + int val; + + val = atomic_dec_return(&vsock->queued_replies); + + /* Do we have resources to resume tx processing? */ + if (val + 1 == tx_vq->num) + restart_tx = true; + } + + virtio_transport_free_pkt(pkt); + } + if (added) + vhost_signal(&vsock->dev, vq); + +out: + mutex_unlock(&vq->mutex); + + if (restart_tx) + vhost_poll_queue(&tx_vq->poll); +} + +static void vhost_transport_send_pkt_work(struct vhost_work *work) +{ + struct vhost_virtqueue *vq; + struct vhost_vsock *vsock; + + vsock = container_of(work, struct vhost_vsock, send_pkt_work); + vq = &vsock->vqs[VSOCK_VQ_RX]; + + vhost_transport_do_send_pkt(vsock, vq); +} + +static int +vhost_transport_send_pkt(struct virtio_vsock_pkt *pkt) +{ + struct vhost_vsock *vsock; + struct vhost_virtqueue *vq; + int len = pkt->len; + + /* Find the vhost_vsock according to guest context id */ + vsock = vhost_vsock_get(le64_to_cpu(pkt->hdr.dst_cid)); + if (!vsock) { + virtio_transport_free_pkt(pkt); + return -ENODEV; + } + + vq = &vsock->vqs[VSOCK_VQ_RX]; + + if (pkt->reply) + atomic_inc(&vsock->queued_replies); + + spin_lock_bh(&vsock->send_pkt_list_lock); + list_add_tail(&pkt->list, &vsock->send_pkt_list); + spin_unlock_bh(&vsock->send_pkt_list_lock); + + vhost_work_queue(&vsock->dev, &vsock->send_pkt_work); + return len; +} + +static struct virtio_vsock_pkt * +vhost_vsock_alloc_pkt(struct vhost_virtqueue *vq, + unsigned int out, unsigned int in) +{ + struct virtio_vsock_pkt *pkt; + struct iov_iter iov_iter; + size_t nbytes; + size_t len; + + if (in != 0) { + vq_err(vq, "Expected 0 input buffers, got %u\n", in); + return NULL; + } + + pkt = kzalloc(sizeof(*pkt), GFP_KERNEL); + if (!pkt) + return NULL; + + len = iov_length(vq->iov, out); + iov_iter_init(&iov_iter, WRITE, vq->iov, out, len); + + nbytes = copy_from_iter(&pkt->hdr, sizeof(pkt->hdr), &iov_iter); + if (nbytes != sizeof(pkt->hdr)) { + vq_err(vq, "Expected %zu bytes for pkt->hdr, got %zu bytes\n", + sizeof(pkt->hdr), nbytes); + kfree(pkt); + return NULL; + } + + if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_STREAM) + pkt->len = le32_to_cpu(pkt->hdr.len); + + /* No payload */ + if (!pkt->len) + return pkt; + + /* The pkt is too big */ + if (pkt->len > VIRTIO_VSOCK_MAX_PKT_BUF_SIZE) { + kfree(pkt); + return NULL; + } + + pkt->buf = kmalloc(pkt->len, GFP_KERNEL); + if (!pkt->buf) { + kfree(pkt); + return NULL; + } + + nbytes = copy_from_iter(pkt->buf, pkt->len, &iov_iter); + if (nbytes != pkt->len) { + vq_err(vq, "Expected %u byte payload, got %zu bytes\n", + pkt->len, nbytes); + virtio_transport_free_pkt(pkt); + return NULL; + } + + return pkt; +} + +/* Is there space left for replies to rx packets? */ +static bool vhost_vsock_more_replies(struct vhost_vsock *vsock) +{ + struct vhost_virtqueue *vq = &vsock->vqs[VSOCK_VQ_TX]; + int val; + + smp_rmb(); /* paired with atomic_inc() and atomic_dec_return() */ + val = atomic_read(&vsock->queued_replies); + + return val < vq->num; +} + +static void vhost_vsock_handle_tx_kick(struct vhost_work *work) +{ + struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue, + poll.work); + struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock, + dev); + struct virtio_vsock_pkt *pkt; + int head; + unsigned int out, in; + bool added = false; + + mutex_lock(&vq->mutex); + + if (!vq->private_data) + goto out; + + vhost_disable_notify(&vsock->dev, vq); + for (;;) { + if (!vhost_vsock_more_replies(vsock)) { + /* Stop tx until the device processes already + * pending replies. Leave tx virtqueue + * callbacks disabled. + */ + goto no_more_replies; + } + + head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov), + &out, &in, NULL, NULL); + if (head < 0) + break; + + if (head == vq->num) { + if (unlikely(vhost_enable_notify(&vsock->dev, vq))) { + vhost_disable_notify(&vsock->dev, vq); + continue; + } + break; + } + + pkt = vhost_vsock_alloc_pkt(vq, out, in); + if (!pkt) { + vq_err(vq, "Faulted on pkt\n"); + continue; + } + + /* Only accept correctly addressed packets */ + if (le64_to_cpu(pkt->hdr.src_cid) == vsock->guest_cid) + virtio_transport_recv_pkt(pkt); + else + virtio_transport_free_pkt(pkt); + + vhost_add_used(vq, head, sizeof(pkt->hdr) + pkt->len); + added = true; + } + +no_more_replies: + if (added) + vhost_signal(&vsock->dev, vq); + +out: + mutex_unlock(&vq->mutex); +} + +static void vhost_vsock_handle_rx_kick(struct vhost_work *work) +{ + struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue, + poll.work); + struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock, + dev); + + vhost_transport_do_send_pkt(vsock, vq); +} + +static int vhost_vsock_start(struct vhost_vsock *vsock) +{ + size_t i; + int ret; + + mutex_lock(&vsock->dev.mutex); + + ret = vhost_dev_check_owner(&vsock->dev); + if (ret) + goto err; + + for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) { + struct vhost_virtqueue *vq = &vsock->vqs[i]; + + mutex_lock(&vq->mutex); + + if (!vhost_vq_access_ok(vq)) { + ret = -EFAULT; + mutex_unlock(&vq->mutex); + goto err_vq; + } + + if (!vq->private_data) { + vq->private_data = vsock; + vhost_vq_init_access(vq); + } + + mutex_unlock(&vq->mutex); + } + + mutex_unlock(&vsock->dev.mutex); + return 0; + +err_vq: + for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) { + struct vhost_virtqueue *vq = &vsock->vqs[i]; + + mutex_lock(&vq->mutex); + vq->private_data = NULL; + mutex_unlock(&vq->mutex); + } +err: + mutex_unlock(&vsock->dev.mutex); + return ret; +} + +static int vhost_vsock_stop(struct vhost_vsock *vsock) +{ + size_t i; + int ret; + + mutex_lock(&vsock->dev.mutex); + + ret = vhost_dev_check_owner(&vsock->dev); + if (ret) + goto err; + + for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) { + struct vhost_virtqueue *vq = &vsock->vqs[i]; + + mutex_lock(&vq->mutex); + vq->private_data = NULL; + mutex_unlock(&vq->mutex); + } + +err: + mutex_unlock(&vsock->dev.mutex); + return ret; +} + +static void vhost_vsock_free(struct vhost_vsock *vsock) +{ + if (is_vmalloc_addr(vsock)) + vfree(vsock); + else + kfree(vsock); +} + +static int vhost_vsock_dev_open(struct inode *inode, struct file *file) +{ + struct vhost_virtqueue **vqs; + struct vhost_vsock *vsock; + int ret; + + /* This struct is large and allocation could fail, fall back to vmalloc + * if there is no other way. + */ + vsock = kzalloc(sizeof(*vsock), GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT); + if (!vsock) { + vsock = vmalloc(sizeof(*vsock)); + if (!vsock) + return -ENOMEM; + } + + vqs = kmalloc_array(ARRAY_SIZE(vsock->vqs), sizeof(*vqs), GFP_KERNEL); + if (!vqs) { + ret = -ENOMEM; + goto out; + } + + atomic_set(&vsock->queued_replies, 0); + + vqs[VSOCK_VQ_TX] = &vsock->vqs[VSOCK_VQ_TX]; + vqs[VSOCK_VQ_RX] = &vsock->vqs[VSOCK_VQ_RX]; + vsock->vqs[VSOCK_VQ_TX].handle_kick = vhost_vsock_handle_tx_kick; + vsock->vqs[VSOCK_VQ_RX].handle_kick = vhost_vsock_handle_rx_kick; + + vhost_dev_init(&vsock->dev, vqs, ARRAY_SIZE(vsock->vqs)); + + file->private_data = vsock; + spin_lock_init(&vsock->send_pkt_list_lock); + INIT_LIST_HEAD(&vsock->send_pkt_list); + vhost_work_init(&vsock->send_pkt_work, vhost_transport_send_pkt_work); + + spin_lock_bh(&vhost_vsock_lock); + list_add_tail(&vsock->list, &vhost_vsock_list); + spin_unlock_bh(&vhost_vsock_lock); + return 0; + +out: + vhost_vsock_free(vsock); + return ret; +} + +static void vhost_vsock_flush(struct vhost_vsock *vsock) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) + if (vsock->vqs[i].handle_kick) + vhost_poll_flush(&vsock->vqs[i].poll); + vhost_work_flush(&vsock->dev, &vsock->send_pkt_work); +} + +static void vhost_vsock_reset_orphans(struct sock *sk) +{ + struct vsock_sock *vsk = vsock_sk(sk); + + /* vmci_transport.c doesn't take sk_lock here either. At least we're + * under vsock_table_lock so the sock cannot disappear while we're + * executing. + */ + + if (!vhost_vsock_get(vsk->local_addr.svm_cid)) { + sock_set_flag(sk, SOCK_DONE); + vsk->peer_shutdown = SHUTDOWN_MASK; + sk->sk_state = SS_UNCONNECTED; + sk->sk_err = ECONNRESET; + sk->sk_error_report(sk); + } +} + +static int vhost_vsock_dev_release(struct inode *inode, struct file *file) +{ + struct vhost_vsock *vsock = file->private_data; + + spin_lock_bh(&vhost_vsock_lock); + list_del(&vsock->list); + spin_unlock_bh(&vhost_vsock_lock); + + /* Iterating over all connections for all CIDs to find orphans is + * inefficient. Room for improvement here. */ + vsock_for_each_connected_socket(vhost_vsock_reset_orphans); + + vhost_vsock_stop(vsock); + vhost_vsock_flush(vsock); + vhost_dev_stop(&vsock->dev); + + spin_lock_bh(&vsock->send_pkt_list_lock); + while (!list_empty(&vsock->send_pkt_list)) { + struct virtio_vsock_pkt *pkt; + + pkt = list_first_entry(&vsock->send_pkt_list, + struct virtio_vsock_pkt, list); + list_del_init(&pkt->list); + virtio_transport_free_pkt(pkt); + } + spin_unlock_bh(&vsock->send_pkt_list_lock); + + vhost_dev_cleanup(&vsock->dev, false); + kfree(vsock->dev.vqs); + vhost_vsock_free(vsock); + return 0; +} + +static int vhost_vsock_set_cid(struct vhost_vsock *vsock, u64 guest_cid) +{ + struct vhost_vsock *other; + + /* Refuse reserved CIDs */ + if (guest_cid <= VMADDR_CID_HOST || + guest_cid == U32_MAX) + return -EINVAL; + + /* 64-bit CIDs are not yet supported */ + if (guest_cid > U32_MAX) + return -EINVAL; + + /* Refuse if CID is already in use */ + other = vhost_vsock_get(guest_cid); + if (other && other != vsock) + return -EADDRINUSE; + + spin_lock_bh(&vhost_vsock_lock); + vsock->guest_cid = guest_cid; + spin_unlock_bh(&vhost_vsock_lock); + + return 0; +} + +static int vhost_vsock_set_features(struct vhost_vsock *vsock, u64 features) +{ + struct vhost_virtqueue *vq; + int i; + + if (features & ~VHOST_VSOCK_FEATURES) + return -EOPNOTSUPP; + + mutex_lock(&vsock->dev.mutex); + if ((features & (1 << VHOST_F_LOG_ALL)) && + !vhost_log_access_ok(&vsock->dev)) { + mutex_unlock(&vsock->dev.mutex); + return -EFAULT; + } + + for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) { + vq = &vsock->vqs[i]; + mutex_lock(&vq->mutex); + vq->acked_features = features; + mutex_unlock(&vq->mutex); + } + mutex_unlock(&vsock->dev.mutex); + return 0; +} + +static long vhost_vsock_dev_ioctl(struct file *f, unsigned int ioctl, + unsigned long arg) +{ + struct vhost_vsock *vsock = f->private_data; + void __user *argp = (void __user *)arg; + u64 guest_cid; + u64 features; + int start; + int r; + + switch (ioctl) { + case VHOST_VSOCK_SET_GUEST_CID: + if (copy_from_user(&guest_cid, argp, sizeof(guest_cid))) + return -EFAULT; + return vhost_vsock_set_cid(vsock, guest_cid); + case VHOST_VSOCK_SET_RUNNING: + if (copy_from_user(&start, argp, sizeof(start))) + return -EFAULT; + if (start) + return vhost_vsock_start(vsock); + else + return vhost_vsock_stop(vsock); + case VHOST_GET_FEATURES: + features = VHOST_VSOCK_FEATURES; + if (copy_to_user(argp, &features, sizeof(features))) + return -EFAULT; + return 0; + case VHOST_SET_FEATURES: + if (copy_from_user(&features, argp, sizeof(features))) + return -EFAULT; + return vhost_vsock_set_features(vsock, features); + default: + mutex_lock(&vsock->dev.mutex); + r = vhost_dev_ioctl(&vsock->dev, ioctl, argp); + if (r == -ENOIOCTLCMD) + r = vhost_vring_ioctl(&vsock->dev, ioctl, argp); + else + vhost_vsock_flush(vsock); + mutex_unlock(&vsock->dev.mutex); + return r; + } +} + +static const struct file_operations vhost_vsock_fops = { + .owner = THIS_MODULE, + .open = vhost_vsock_dev_open, + .release = vhost_vsock_dev_release, + .llseek = noop_llseek, + .unlocked_ioctl = vhost_vsock_dev_ioctl, +}; + +static struct miscdevice vhost_vsock_misc = { + .minor = MISC_DYNAMIC_MINOR, + .name = "vhost-vsock", + .fops = &vhost_vsock_fops, +}; + +static struct virtio_transport vhost_transport = { + .transport = { + .get_local_cid = vhost_transport_get_local_cid, + + .init = virtio_transport_do_socket_init, + .destruct = virtio_transport_destruct, + .release = virtio_transport_release, + .connect = virtio_transport_connect, + .shutdown = virtio_transport_shutdown, + + .dgram_enqueue = virtio_transport_dgram_enqueue, + .dgram_dequeue = virtio_transport_dgram_dequeue, + .dgram_bind = virtio_transport_dgram_bind, + .dgram_allow = virtio_transport_dgram_allow, + + .stream_enqueue = virtio_transport_stream_enqueue, + .stream_dequeue = virtio_transport_stream_dequeue, + .stream_has_data = virtio_transport_stream_has_data, + .stream_has_space = virtio_transport_stream_has_space, + .stream_rcvhiwat = virtio_transport_stream_rcvhiwat, + .stream_is_active = virtio_transport_stream_is_active, + .stream_allow = virtio_transport_stream_allow, + + .notify_poll_in = virtio_transport_notify_poll_in, + .notify_poll_out = virtio_transport_notify_poll_out, + .notify_recv_init = virtio_transport_notify_recv_init, + .notify_recv_pre_block = virtio_transport_notify_recv_pre_block, + .notify_recv_pre_dequeue = virtio_transport_notify_recv_pre_dequeue, + .notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue, + .notify_send_init = virtio_transport_notify_send_init, + .notify_send_pre_block = virtio_transport_notify_send_pre_block, + .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, + .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, + + .set_buffer_size = virtio_transport_set_buffer_size, + .set_min_buffer_size = virtio_transport_set_min_buffer_size, + .set_max_buffer_size = virtio_transport_set_max_buffer_size, + .get_buffer_size = virtio_transport_get_buffer_size, + .get_min_buffer_size = virtio_transport_get_min_buffer_size, + .get_max_buffer_size = virtio_transport_get_max_buffer_size, + }, + + .send_pkt = vhost_transport_send_pkt, +}; + +static int __init vhost_vsock_init(void) +{ + int ret; + + ret = vsock_core_init(&vhost_transport.transport); + if (ret < 0) + return ret; + return misc_register(&vhost_vsock_misc); +}; + +static void __exit vhost_vsock_exit(void) +{ + misc_deregister(&vhost_vsock_misc); + vsock_core_exit(); +}; + +module_init(vhost_vsock_init); +module_exit(vhost_vsock_exit); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Asias He"); +MODULE_DESCRIPTION("vhost transport for vsock "); diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h index 61a8777178c6..c4400b267716 100644 --- a/include/uapi/linux/vhost.h +++ b/include/uapi/linux/vhost.h @@ -175,4 +175,9 @@ struct vhost_scsi_target { #define VHOST_SCSI_SET_EVENTS_MISSED _IOW(VHOST_VIRTIO, 0x43, __u32) #define VHOST_SCSI_GET_EVENTS_MISSED _IOW(VHOST_VIRTIO, 0x44, __u32) +/* VHOST_VSOCK specific defines */ + +#define VHOST_VSOCK_SET_GUEST_CID _IOW(VHOST_VIRTIO, 0x60, __u64) +#define VHOST_VSOCK_SET_RUNNING _IOW(VHOST_VIRTIO, 0x61, int) + #endif -- cgit v1.2.3 From 6b1e6cc7855b09a0a9bfa1d9f30172ba366f161c Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 23 Jun 2016 02:04:32 -0400 Subject: vhost: new device IOTLB API This patch tries to implement an device IOTLB for vhost. This could be used with userspace(qemu) implementation of DMA remapping to emulate an IOMMU for the guest. The idea is simple, cache the translation in a software device IOTLB (which is implemented as an interval tree) in vhost and use vhost_net file descriptor for reporting IOTLB miss and IOTLB update/invalidation. When vhost meets an IOTLB miss, the fault address, size and access can be read from the file. After userspace finishes the translation, it writes the translated address to the vhost_net file to update the device IOTLB. When device IOTLB is enabled by setting VIRTIO_F_IOMMU_PLATFORM all vq addresses set by ioctl are treated as iova instead of virtual address and the accessing can only be done through IOTLB instead of direct userspace memory access. Before each round or vq processing, all vq metadata is prefetched in device IOTLB to make sure no translation fault happens during vq processing. In most cases, virtqueues are contiguous even in virtual address space. The IOTLB translation for virtqueue itself may make it a little slower. We might add fast path cache on top of this patch. Signed-off-by: Jason Wang [mst: use virtio feature bit: VHOST_F_DEVICE_IOTLB -> VIRTIO_F_IOMMU_PLATFORM ] [mst: fix build warnings ] Signed-off-by: Michael S. Tsirkin [ weiyj.lk: missing unlock on error ] Signed-off-by: Wei Yongjun --- drivers/vhost/net.c | 59 ++++- drivers/vhost/vhost.c | 636 ++++++++++++++++++++++++++++++++++++++++++--- drivers/vhost/vhost.h | 32 ++- include/uapi/linux/vhost.h | 28 ++ 4 files changed, 705 insertions(+), 50 deletions(-) (limited to 'include/uapi') diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index a6b270aff9ef..0965f869dc57 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -61,7 +61,8 @@ MODULE_PARM_DESC(experimental_zcopytx, "Enable Zero Copy TX;" enum { VHOST_NET_FEATURES = VHOST_FEATURES | (1ULL << VHOST_NET_F_VIRTIO_NET_HDR) | - (1ULL << VIRTIO_NET_F_MRG_RXBUF) + (1ULL << VIRTIO_NET_F_MRG_RXBUF) | + (1ULL << VIRTIO_F_IOMMU_PLATFORM) }; enum { @@ -308,7 +309,7 @@ static int vhost_net_tx_get_vq_desc(struct vhost_net *net, { unsigned long uninitialized_var(endtime); int r = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov), - out_num, in_num, NULL, NULL); + out_num, in_num, NULL, NULL); if (r == vq->num && vq->busyloop_timeout) { preempt_disable(); @@ -318,7 +319,7 @@ static int vhost_net_tx_get_vq_desc(struct vhost_net *net, cpu_relax_lowlatency(); preempt_enable(); r = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov), - out_num, in_num, NULL, NULL); + out_num, in_num, NULL, NULL); } return r; @@ -351,6 +352,9 @@ static void handle_tx(struct vhost_net *net) if (!sock) goto out; + if (!vq_iotlb_prefetch(vq)) + goto out; + vhost_disable_notify(&net->dev, vq); hdr_size = nvq->vhost_hlen; @@ -612,6 +616,10 @@ static void handle_rx(struct vhost_net *net) sock = vq->private_data; if (!sock) goto out; + + if (!vq_iotlb_prefetch(vq)) + goto out; + vhost_disable_notify(&net->dev, vq); vhost_hlen = nvq->vhost_hlen; @@ -1080,10 +1088,14 @@ static int vhost_net_set_features(struct vhost_net *n, u64 features) } mutex_lock(&n->dev.mutex); if ((features & (1 << VHOST_F_LOG_ALL)) && - !vhost_log_access_ok(&n->dev)) { - mutex_unlock(&n->dev.mutex); - return -EFAULT; + !vhost_log_access_ok(&n->dev)) + goto out_unlock; + + if ((features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))) { + if (vhost_init_device_iotlb(&n->dev, true)) + goto out_unlock; } + for (i = 0; i < VHOST_NET_VQ_MAX; ++i) { mutex_lock(&n->vqs[i].vq.mutex); n->vqs[i].vq.acked_features = features; @@ -1093,6 +1105,10 @@ static int vhost_net_set_features(struct vhost_net *n, u64 features) } mutex_unlock(&n->dev.mutex); return 0; + +out_unlock: + mutex_unlock(&n->dev.mutex); + return -EFAULT; } static long vhost_net_set_owner(struct vhost_net *n) @@ -1166,9 +1182,40 @@ static long vhost_net_compat_ioctl(struct file *f, unsigned int ioctl, } #endif +static ssize_t vhost_net_chr_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + struct file *file = iocb->ki_filp; + struct vhost_net *n = file->private_data; + struct vhost_dev *dev = &n->dev; + int noblock = file->f_flags & O_NONBLOCK; + + return vhost_chr_read_iter(dev, to, noblock); +} + +static ssize_t vhost_net_chr_write_iter(struct kiocb *iocb, + struct iov_iter *from) +{ + struct file *file = iocb->ki_filp; + struct vhost_net *n = file->private_data; + struct vhost_dev *dev = &n->dev; + + return vhost_chr_write_iter(dev, from); +} + +static unsigned int vhost_net_chr_poll(struct file *file, poll_table *wait) +{ + struct vhost_net *n = file->private_data; + struct vhost_dev *dev = &n->dev; + + return vhost_chr_poll(file, dev, wait); +} + static const struct file_operations vhost_net_fops = { .owner = THIS_MODULE, .release = vhost_net_release, + .read_iter = vhost_net_chr_read_iter, + .write_iter = vhost_net_chr_write_iter, + .poll = vhost_net_chr_poll, .unlocked_ioctl = vhost_net_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = vhost_net_compat_ioctl, diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 8071f3638db9..d02c1614921f 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -35,6 +35,10 @@ static ushort max_mem_regions = 64; module_param(max_mem_regions, ushort, 0444); MODULE_PARM_DESC(max_mem_regions, "Maximum number of memory regions in memory map. (default: 64)"); +static int max_iotlb_entries = 2048; +module_param(max_iotlb_entries, int, 0444); +MODULE_PARM_DESC(max_iotlb_entries, + "Maximum number of iotlb entries. (default: 2048)"); enum { VHOST_MEMORY_F_LOG = 0x1, @@ -306,6 +310,7 @@ static void vhost_vq_reset(struct vhost_dev *dev, vhost_disable_cross_endian(vq); vq->busyloop_timeout = 0; vq->umem = NULL; + vq->iotlb = NULL; } static int vhost_worker(void *data) @@ -400,9 +405,14 @@ void vhost_dev_init(struct vhost_dev *dev, dev->log_ctx = NULL; dev->log_file = NULL; dev->umem = NULL; + dev->iotlb = NULL; dev->mm = NULL; dev->worker = NULL; init_llist_head(&dev->work_list); + init_waitqueue_head(&dev->wait); + INIT_LIST_HEAD(&dev->read_list); + INIT_LIST_HEAD(&dev->pending_list); + spin_lock_init(&dev->iotlb_lock); for (i = 0; i < dev->nvqs; ++i) { @@ -550,6 +560,15 @@ void vhost_dev_stop(struct vhost_dev *dev) } EXPORT_SYMBOL_GPL(vhost_dev_stop); +static void vhost_umem_free(struct vhost_umem *umem, + struct vhost_umem_node *node) +{ + vhost_umem_interval_tree_remove(node, &umem->umem_tree); + list_del(&node->link); + kfree(node); + umem->numem--; +} + static void vhost_umem_clean(struct vhost_umem *umem) { struct vhost_umem_node *node, *tmp; @@ -557,14 +576,31 @@ static void vhost_umem_clean(struct vhost_umem *umem) if (!umem) return; - list_for_each_entry_safe(node, tmp, &umem->umem_list, link) { - vhost_umem_interval_tree_remove(node, &umem->umem_tree); - list_del(&node->link); - kvfree(node); - } + list_for_each_entry_safe(node, tmp, &umem->umem_list, link) + vhost_umem_free(umem, node); + kvfree(umem); } +static void vhost_clear_msg(struct vhost_dev *dev) +{ + struct vhost_msg_node *node, *n; + + spin_lock(&dev->iotlb_lock); + + list_for_each_entry_safe(node, n, &dev->read_list, node) { + list_del(&node->node); + kfree(node); + } + + list_for_each_entry_safe(node, n, &dev->pending_list, node) { + list_del(&node->node); + kfree(node); + } + + spin_unlock(&dev->iotlb_lock); +} + /* Caller should have device mutex if and only if locked is set */ void vhost_dev_cleanup(struct vhost_dev *dev, bool locked) { @@ -593,6 +629,10 @@ void vhost_dev_cleanup(struct vhost_dev *dev, bool locked) /* No one will access memory at this point */ vhost_umem_clean(dev->umem); dev->umem = NULL; + vhost_umem_clean(dev->iotlb); + dev->iotlb = NULL; + vhost_clear_msg(dev); + wake_up_interruptible_poll(&dev->wait, POLLIN | POLLRDNORM); WARN_ON(!llist_empty(&dev->work_list)); if (dev->worker) { kthread_stop(dev->worker); @@ -668,28 +708,381 @@ static int memory_access_ok(struct vhost_dev *d, struct vhost_umem *umem, return 1; } -#define vhost_put_user(vq, x, ptr) __put_user(x, ptr) +static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len, + struct iovec iov[], int iov_size, int access); static int vhost_copy_to_user(struct vhost_virtqueue *vq, void *to, const void *from, unsigned size) { - return __copy_to_user(to, from, size); -} + int ret; -#define vhost_get_user(vq, x, ptr) __get_user(x, ptr) + if (!vq->iotlb) + return __copy_to_user(to, from, size); + else { + /* This function should be called after iotlb + * prefetch, which means we're sure that all vq + * could be access through iotlb. So -EAGAIN should + * not happen in this case. + */ + /* TODO: more fast path */ + struct iov_iter t; + ret = translate_desc(vq, (u64)(uintptr_t)to, size, vq->iotlb_iov, + ARRAY_SIZE(vq->iotlb_iov), + VHOST_ACCESS_WO); + if (ret < 0) + goto out; + iov_iter_init(&t, WRITE, vq->iotlb_iov, ret, size); + ret = copy_to_iter(from, size, &t); + if (ret == size) + ret = 0; + } +out: + return ret; +} static int vhost_copy_from_user(struct vhost_virtqueue *vq, void *to, void *from, unsigned size) { - return __copy_from_user(to, from, size); + int ret; + + if (!vq->iotlb) + return __copy_from_user(to, from, size); + else { + /* This function should be called after iotlb + * prefetch, which means we're sure that vq + * could be access through iotlb. So -EAGAIN should + * not happen in this case. + */ + /* TODO: more fast path */ + struct iov_iter f; + ret = translate_desc(vq, (u64)(uintptr_t)from, size, vq->iotlb_iov, + ARRAY_SIZE(vq->iotlb_iov), + VHOST_ACCESS_RO); + if (ret < 0) { + vq_err(vq, "IOTLB translation failure: uaddr " + "%p size 0x%llx\n", from, + (unsigned long long) size); + goto out; + } + iov_iter_init(&f, READ, vq->iotlb_iov, ret, size); + ret = copy_from_iter(to, size, &f); + if (ret == size) + ret = 0; + } + +out: + return ret; +} + +static void __user *__vhost_get_user(struct vhost_virtqueue *vq, + void *addr, unsigned size) +{ + int ret; + + /* This function should be called after iotlb + * prefetch, which means we're sure that vq + * could be access through iotlb. So -EAGAIN should + * not happen in this case. + */ + /* TODO: more fast path */ + ret = translate_desc(vq, (u64)(uintptr_t)addr, size, vq->iotlb_iov, + ARRAY_SIZE(vq->iotlb_iov), + VHOST_ACCESS_RO); + if (ret < 0) { + vq_err(vq, "IOTLB translation failure: uaddr " + "%p size 0x%llx\n", addr, + (unsigned long long) size); + return NULL; + } + + if (ret != 1 || vq->iotlb_iov[0].iov_len != size) { + vq_err(vq, "Non atomic userspace memory access: uaddr " + "%p size 0x%llx\n", addr, + (unsigned long long) size); + return NULL; + } + + return vq->iotlb_iov[0].iov_base; +} + +#define vhost_put_user(vq, x, ptr) \ +({ \ + int ret = -EFAULT; \ + if (!vq->iotlb) { \ + ret = __put_user(x, ptr); \ + } else { \ + __typeof__(ptr) to = \ + (__typeof__(ptr)) __vhost_get_user(vq, ptr, sizeof(*ptr)); \ + if (to != NULL) \ + ret = __put_user(x, to); \ + else \ + ret = -EFAULT; \ + } \ + ret; \ +}) + +#define vhost_get_user(vq, x, ptr) \ +({ \ + int ret; \ + if (!vq->iotlb) { \ + ret = __get_user(x, ptr); \ + } else { \ + __typeof__(ptr) from = \ + (__typeof__(ptr)) __vhost_get_user(vq, ptr, sizeof(*ptr)); \ + if (from != NULL) \ + ret = __get_user(x, from); \ + else \ + ret = -EFAULT; \ + } \ + ret; \ +}) + +static void vhost_dev_lock_vqs(struct vhost_dev *d) +{ + int i = 0; + for (i = 0; i < d->nvqs; ++i) + mutex_lock(&d->vqs[i]->mutex); +} + +static void vhost_dev_unlock_vqs(struct vhost_dev *d) +{ + int i = 0; + for (i = 0; i < d->nvqs; ++i) + mutex_unlock(&d->vqs[i]->mutex); +} + +static int vhost_new_umem_range(struct vhost_umem *umem, + u64 start, u64 size, u64 end, + u64 userspace_addr, int perm) +{ + struct vhost_umem_node *tmp, *node = kmalloc(sizeof(*node), GFP_ATOMIC); + + if (!node) + return -ENOMEM; + + if (umem->numem == max_iotlb_entries) { + tmp = list_first_entry(&umem->umem_list, typeof(*tmp), link); + vhost_umem_free(umem, tmp); + } + + node->start = start; + node->size = size; + node->last = end; + node->userspace_addr = userspace_addr; + node->perm = perm; + INIT_LIST_HEAD(&node->link); + list_add_tail(&node->link, &umem->umem_list); + vhost_umem_interval_tree_insert(node, &umem->umem_tree); + umem->numem++; + + return 0; +} + +static void vhost_del_umem_range(struct vhost_umem *umem, + u64 start, u64 end) +{ + struct vhost_umem_node *node; + + while ((node = vhost_umem_interval_tree_iter_first(&umem->umem_tree, + start, end))) + vhost_umem_free(umem, node); +} + +static void vhost_iotlb_notify_vq(struct vhost_dev *d, + struct vhost_iotlb_msg *msg) +{ + struct vhost_msg_node *node, *n; + + spin_lock(&d->iotlb_lock); + + list_for_each_entry_safe(node, n, &d->pending_list, node) { + struct vhost_iotlb_msg *vq_msg = &node->msg.iotlb; + if (msg->iova <= vq_msg->iova && + msg->iova + msg->size - 1 > vq_msg->iova && + vq_msg->type == VHOST_IOTLB_MISS) { + vhost_poll_queue(&node->vq->poll); + list_del(&node->node); + kfree(node); + } + } + + spin_unlock(&d->iotlb_lock); +} + +static int umem_access_ok(u64 uaddr, u64 size, int access) +{ + unsigned long a = uaddr; + + if ((access & VHOST_ACCESS_RO) && + !access_ok(VERIFY_READ, (void __user *)a, size)) + return -EFAULT; + if ((access & VHOST_ACCESS_WO) && + !access_ok(VERIFY_WRITE, (void __user *)a, size)) + return -EFAULT; + return 0; +} + +int vhost_process_iotlb_msg(struct vhost_dev *dev, + struct vhost_iotlb_msg *msg) +{ + int ret = 0; + + vhost_dev_lock_vqs(dev); + switch (msg->type) { + case VHOST_IOTLB_UPDATE: + if (!dev->iotlb) { + ret = -EFAULT; + break; + } + if (umem_access_ok(msg->uaddr, msg->size, msg->perm)) { + ret = -EFAULT; + break; + } + if (vhost_new_umem_range(dev->iotlb, msg->iova, msg->size, + msg->iova + msg->size - 1, + msg->uaddr, msg->perm)) { + ret = -ENOMEM; + break; + } + vhost_iotlb_notify_vq(dev, msg); + break; + case VHOST_IOTLB_INVALIDATE: + vhost_del_umem_range(dev->iotlb, msg->iova, + msg->iova + msg->size - 1); + break; + default: + ret = -EINVAL; + break; + } + + vhost_dev_unlock_vqs(dev); + return ret; +} +ssize_t vhost_chr_write_iter(struct vhost_dev *dev, + struct iov_iter *from) +{ + struct vhost_msg_node node; + unsigned size = sizeof(struct vhost_msg); + size_t ret; + int err; + + if (iov_iter_count(from) < size) + return 0; + ret = copy_from_iter(&node.msg, size, from); + if (ret != size) + goto done; + + switch (node.msg.type) { + case VHOST_IOTLB_MSG: + err = vhost_process_iotlb_msg(dev, &node.msg.iotlb); + if (err) + ret = err; + break; + default: + ret = -EINVAL; + break; + } + +done: + return ret; +} +EXPORT_SYMBOL(vhost_chr_write_iter); + +unsigned int vhost_chr_poll(struct file *file, struct vhost_dev *dev, + poll_table *wait) +{ + unsigned int mask = 0; + + poll_wait(file, &dev->wait, wait); + + if (!list_empty(&dev->read_list)) + mask |= POLLIN | POLLRDNORM; + + return mask; +} +EXPORT_SYMBOL(vhost_chr_poll); + +ssize_t vhost_chr_read_iter(struct vhost_dev *dev, struct iov_iter *to, + int noblock) +{ + DEFINE_WAIT(wait); + struct vhost_msg_node *node; + ssize_t ret = 0; + unsigned size = sizeof(struct vhost_msg); + + if (iov_iter_count(to) < size) + return 0; + + while (1) { + if (!noblock) + prepare_to_wait(&dev->wait, &wait, + TASK_INTERRUPTIBLE); + + node = vhost_dequeue_msg(dev, &dev->read_list); + if (node) + break; + if (noblock) { + ret = -EAGAIN; + break; + } + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + if (!dev->iotlb) { + ret = -EBADFD; + break; + } + + schedule(); + } + + if (!noblock) + finish_wait(&dev->wait, &wait); + + if (node) { + ret = copy_to_iter(&node->msg, size, to); + + if (ret != size || node->msg.type != VHOST_IOTLB_MISS) { + kfree(node); + return ret; + } + + vhost_enqueue_msg(dev, &dev->pending_list, node); + } + + return ret; +} +EXPORT_SYMBOL_GPL(vhost_chr_read_iter); + +static int vhost_iotlb_miss(struct vhost_virtqueue *vq, u64 iova, int access) +{ + struct vhost_dev *dev = vq->dev; + struct vhost_msg_node *node; + struct vhost_iotlb_msg *msg; + + node = vhost_new_msg(vq, VHOST_IOTLB_MISS); + if (!node) + return -ENOMEM; + + msg = &node->msg.iotlb; + msg->type = VHOST_IOTLB_MISS; + msg->iova = iova; + msg->perm = access; + + vhost_enqueue_msg(dev, &dev->read_list, node); + + return 0; } static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num, struct vring_desc __user *desc, struct vring_avail __user *avail, struct vring_used __user *used) + { size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0; + return access_ok(VERIFY_READ, desc, num * sizeof *desc) && access_ok(VERIFY_READ, avail, sizeof *avail + num * sizeof *avail->ring + s) && @@ -697,6 +1090,54 @@ static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num, sizeof *used + num * sizeof *used->ring + s); } +static int iotlb_access_ok(struct vhost_virtqueue *vq, + int access, u64 addr, u64 len) +{ + const struct vhost_umem_node *node; + struct vhost_umem *umem = vq->iotlb; + u64 s = 0, size; + + while (len > s) { + node = vhost_umem_interval_tree_iter_first(&umem->umem_tree, + addr, + addr + len - 1); + if (node == NULL || node->start > addr) { + vhost_iotlb_miss(vq, addr, access); + return false; + } else if (!(node->perm & access)) { + /* Report the possible access violation by + * request another translation from userspace. + */ + return false; + } + + size = node->size - addr + node->start; + s += size; + addr += size; + } + + return true; +} + +int vq_iotlb_prefetch(struct vhost_virtqueue *vq) +{ + size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0; + unsigned int num = vq->num; + + if (!vq->iotlb) + return 1; + + return iotlb_access_ok(vq, VHOST_ACCESS_RO, (u64)(uintptr_t)vq->desc, + num * sizeof *vq->desc) && + iotlb_access_ok(vq, VHOST_ACCESS_RO, (u64)(uintptr_t)vq->avail, + sizeof *vq->avail + + num * sizeof *vq->avail->ring + s) && + iotlb_access_ok(vq, VHOST_ACCESS_WO, (u64)(uintptr_t)vq->used, + sizeof *vq->used + + num * sizeof *vq->used->ring + s); +} +EXPORT_SYMBOL_GPL(vq_iotlb_prefetch); + /* Can we log writes? */ /* Caller should have device mutex but not vq mutex */ int vhost_log_access_ok(struct vhost_dev *dev) @@ -723,16 +1164,35 @@ static int vq_log_access_ok(struct vhost_virtqueue *vq, /* Caller should have vq mutex and device mutex */ int vhost_vq_access_ok(struct vhost_virtqueue *vq) { + if (vq->iotlb) { + /* When device IOTLB was used, the access validation + * will be validated during prefetching. + */ + return 1; + } return vq_access_ok(vq, vq->num, vq->desc, vq->avail, vq->used) && vq_log_access_ok(vq, vq->log_base); } EXPORT_SYMBOL_GPL(vhost_vq_access_ok); +static struct vhost_umem *vhost_umem_alloc(void) +{ + struct vhost_umem *umem = vhost_kvzalloc(sizeof(*umem)); + + if (!umem) + return NULL; + + umem->umem_tree = RB_ROOT; + umem->numem = 0; + INIT_LIST_HEAD(&umem->umem_list); + + return umem; +} + static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) { struct vhost_memory mem, *newmem; struct vhost_memory_region *region; - struct vhost_umem_node *node; struct vhost_umem *newumem, *oldumem; unsigned long size = offsetof(struct vhost_memory, regions); int i; @@ -754,28 +1214,23 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) return -EFAULT; } - newumem = vhost_kvzalloc(sizeof(*newumem)); + newumem = vhost_umem_alloc(); if (!newumem) { kvfree(newmem); return -ENOMEM; } - newumem->umem_tree = RB_ROOT; - INIT_LIST_HEAD(&newumem->umem_list); - for (region = newmem->regions; region < newmem->regions + mem.nregions; region++) { - node = vhost_kvzalloc(sizeof(*node)); - if (!node) + if (vhost_new_umem_range(newumem, + region->guest_phys_addr, + region->memory_size, + region->guest_phys_addr + + region->memory_size - 1, + region->userspace_addr, + VHOST_ACCESS_RW)) goto err; - node->start = region->guest_phys_addr; - node->size = region->memory_size; - node->last = node->start + node->size - 1; - node->userspace_addr = region->userspace_addr; - INIT_LIST_HEAD(&node->link); - list_add_tail(&node->link, &newumem->umem_list); - vhost_umem_interval_tree_insert(node, &newumem->umem_tree); } if (!memory_access_ok(d, newumem, 0)) @@ -1019,6 +1474,30 @@ long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp) } EXPORT_SYMBOL_GPL(vhost_vring_ioctl); +int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled) +{ + struct vhost_umem *niotlb, *oiotlb; + int i; + + niotlb = vhost_umem_alloc(); + if (!niotlb) + return -ENOMEM; + + oiotlb = d->iotlb; + d->iotlb = niotlb; + + for (i = 0; i < d->nvqs; ++i) { + mutex_lock(&d->vqs[i]->mutex); + d->vqs[i]->iotlb = niotlb; + mutex_unlock(&d->vqs[i]->mutex); + } + + vhost_umem_clean(oiotlb); + + return 0; +} +EXPORT_SYMBOL_GPL(vhost_init_device_iotlb); + /* Caller must have device mutex */ long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp) { @@ -1233,15 +1712,20 @@ int vhost_vq_init_access(struct vhost_virtqueue *vq) if (r) goto err; vq->signalled_used_valid = false; - if (!access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx)) { + if (!vq->iotlb && + !access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx)) { r = -EFAULT; goto err; } r = vhost_get_user(vq, last_used_idx, &vq->used->idx); - if (r) + if (r) { + vq_err(vq, "Can't access used idx at %p\n", + &vq->used->idx); goto err; + } vq->last_used_idx = vhost16_to_cpu(vq, last_used_idx); return 0; + err: vq->is_le = is_le; return r; @@ -1249,10 +1733,11 @@ err: EXPORT_SYMBOL_GPL(vhost_vq_init_access); static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len, - struct iovec iov[], int iov_size) + struct iovec iov[], int iov_size, int access) { const struct vhost_umem_node *node; - struct vhost_umem *umem = vq->umem; + struct vhost_dev *dev = vq->dev; + struct vhost_umem *umem = dev->iotlb ? dev->iotlb : dev->umem; struct iovec *_iov; u64 s = 0; int ret = 0; @@ -1263,12 +1748,21 @@ static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len, ret = -ENOBUFS; break; } + node = vhost_umem_interval_tree_iter_first(&umem->umem_tree, addr, addr + len - 1); if (node == NULL || node->start > addr) { - ret = -EFAULT; + if (umem != dev->iotlb) { + ret = -EFAULT; + break; + } + ret = -EAGAIN; + break; + } else if (!(node->perm & access)) { + ret = -EPERM; break; } + _iov = iov + ret; size = node->size - addr + node->start; _iov->iov_len = min((u64)len - s, size); @@ -1279,6 +1773,8 @@ static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len, ++ret; } + if (ret == -EAGAIN) + vhost_iotlb_miss(vq, addr, access); return ret; } @@ -1313,7 +1809,7 @@ static int get_indirect(struct vhost_virtqueue *vq, unsigned int i = 0, count, found = 0; u32 len = vhost32_to_cpu(vq, indirect->len); struct iov_iter from; - int ret; + int ret, access; /* Sanity check */ if (unlikely(len % sizeof desc)) { @@ -1325,9 +1821,10 @@ static int get_indirect(struct vhost_virtqueue *vq, } ret = translate_desc(vq, vhost64_to_cpu(vq, indirect->addr), len, vq->indirect, - UIO_MAXIOV); + UIO_MAXIOV, VHOST_ACCESS_RO); if (unlikely(ret < 0)) { - vq_err(vq, "Translation failure %d in indirect.\n", ret); + if (ret != -EAGAIN) + vq_err(vq, "Translation failure %d in indirect.\n", ret); return ret; } iov_iter_init(&from, READ, vq->indirect, ret, len); @@ -1365,16 +1862,22 @@ static int get_indirect(struct vhost_virtqueue *vq, return -EINVAL; } + if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE)) + access = VHOST_ACCESS_WO; + else + access = VHOST_ACCESS_RO; + ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr), vhost32_to_cpu(vq, desc.len), iov + iov_count, - iov_size - iov_count); + iov_size - iov_count, access); if (unlikely(ret < 0)) { - vq_err(vq, "Translation failure %d indirect idx %d\n", - ret, i); + if (ret != -EAGAIN) + vq_err(vq, "Translation failure %d indirect idx %d\n", + ret, i); return ret; } /* If this is an input descriptor, increment that count. */ - if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE)) { + if (access == VHOST_ACCESS_WO) { *in_num += ret; if (unlikely(log)) { log[*log_num].addr = vhost64_to_cpu(vq, desc.addr); @@ -1413,7 +1916,7 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq, u16 last_avail_idx; __virtio16 avail_idx; __virtio16 ring_head; - int ret; + int ret, access; /* Check it isn't doing very strange things with descriptor numbers. */ last_avail_idx = vq->last_avail_idx; @@ -1487,22 +1990,28 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq, out_num, in_num, log, log_num, &desc); if (unlikely(ret < 0)) { - vq_err(vq, "Failure detected " - "in indirect descriptor at idx %d\n", i); + if (ret != -EAGAIN) + vq_err(vq, "Failure detected " + "in indirect descriptor at idx %d\n", i); return ret; } continue; } + if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE)) + access = VHOST_ACCESS_WO; + else + access = VHOST_ACCESS_RO; ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr), vhost32_to_cpu(vq, desc.len), iov + iov_count, - iov_size - iov_count); + iov_size - iov_count, access); if (unlikely(ret < 0)) { - vq_err(vq, "Translation failure %d descriptor idx %d\n", - ret, i); + if (ret != -EAGAIN) + vq_err(vq, "Translation failure %d descriptor idx %d\n", + ret, i); return ret; } - if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE)) { + if (access == VHOST_ACCESS_WO) { /* If this is an input descriptor, * increment that count. */ *in_num += ret; @@ -1768,6 +2277,47 @@ void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq) } EXPORT_SYMBOL_GPL(vhost_disable_notify); +/* Create a new message. */ +struct vhost_msg_node *vhost_new_msg(struct vhost_virtqueue *vq, int type) +{ + struct vhost_msg_node *node = kmalloc(sizeof *node, GFP_KERNEL); + if (!node) + return NULL; + node->vq = vq; + node->msg.type = type; + return node; +} +EXPORT_SYMBOL_GPL(vhost_new_msg); + +void vhost_enqueue_msg(struct vhost_dev *dev, struct list_head *head, + struct vhost_msg_node *node) +{ + spin_lock(&dev->iotlb_lock); + list_add_tail(&node->node, head); + spin_unlock(&dev->iotlb_lock); + + wake_up_interruptible_poll(&dev->wait, POLLIN | POLLRDNORM); +} +EXPORT_SYMBOL_GPL(vhost_enqueue_msg); + +struct vhost_msg_node *vhost_dequeue_msg(struct vhost_dev *dev, + struct list_head *head) +{ + struct vhost_msg_node *node = NULL; + + spin_lock(&dev->iotlb_lock); + if (!list_empty(head)) { + node = list_first_entry(head, struct vhost_msg_node, + node); + list_del(&node->node); + } + spin_unlock(&dev->iotlb_lock); + + return node; +} +EXPORT_SYMBOL_GPL(vhost_dequeue_msg); + + static int __init vhost_init(void) { return 0; diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h index eaaf6df72218..78f3c5fc02e4 100644 --- a/drivers/vhost/vhost.h +++ b/drivers/vhost/vhost.h @@ -65,13 +65,15 @@ struct vhost_umem_node { __u64 last; __u64 size; __u64 userspace_addr; - __u64 flags_padding; + __u32 perm; + __u32 flags_padding; __u64 __subtree_last; }; struct vhost_umem { struct rb_root umem_tree; struct list_head umem_list; + int numem; }; /* The virtqueue structure describes a queue attached to a device. */ @@ -119,10 +121,12 @@ struct vhost_virtqueue { u64 log_addr; struct iovec iov[UIO_MAXIOV]; + struct iovec iotlb_iov[64]; struct iovec *indirect; struct vring_used_elem *heads; /* Protected by virtqueue mutex. */ struct vhost_umem *umem; + struct vhost_umem *iotlb; void *private_data; u64 acked_features; /* Log write descriptors */ @@ -139,6 +143,12 @@ struct vhost_virtqueue { u32 busyloop_timeout; }; +struct vhost_msg_node { + struct vhost_msg msg; + struct vhost_virtqueue *vq; + struct list_head node; +}; + struct vhost_dev { struct mm_struct *mm; struct mutex mutex; @@ -149,6 +159,11 @@ struct vhost_dev { struct llist_head work_list; struct task_struct *worker; struct vhost_umem *umem; + struct vhost_umem *iotlb; + spinlock_t iotlb_lock; + struct list_head read_list; + struct list_head pending_list; + wait_queue_head_t wait; }; void vhost_dev_init(struct vhost_dev *, struct vhost_virtqueue **vqs, int nvqs); @@ -185,6 +200,21 @@ bool vhost_enable_notify(struct vhost_dev *, struct vhost_virtqueue *); int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log, unsigned int log_num, u64 len); +int vq_iotlb_prefetch(struct vhost_virtqueue *vq); + +struct vhost_msg_node *vhost_new_msg(struct vhost_virtqueue *vq, int type); +void vhost_enqueue_msg(struct vhost_dev *dev, + struct list_head *head, + struct vhost_msg_node *node); +struct vhost_msg_node *vhost_dequeue_msg(struct vhost_dev *dev, + struct list_head *head); +unsigned int vhost_chr_poll(struct file *file, struct vhost_dev *dev, + poll_table *wait); +ssize_t vhost_chr_read_iter(struct vhost_dev *dev, struct iov_iter *to, + int noblock); +ssize_t vhost_chr_write_iter(struct vhost_dev *dev, + struct iov_iter *from); +int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled); #define vq_err(vq, fmt, ...) do { \ pr_debug(pr_fmt(fmt), ##__VA_ARGS__); \ diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h index c4400b267716..56b7ab584cc0 100644 --- a/include/uapi/linux/vhost.h +++ b/include/uapi/linux/vhost.h @@ -47,6 +47,32 @@ struct vhost_vring_addr { __u64 log_guest_addr; }; +/* no alignment requirement */ +struct vhost_iotlb_msg { + __u64 iova; + __u64 size; + __u64 uaddr; +#define VHOST_ACCESS_RO 0x1 +#define VHOST_ACCESS_WO 0x2 +#define VHOST_ACCESS_RW 0x3 + __u8 perm; +#define VHOST_IOTLB_MISS 1 +#define VHOST_IOTLB_UPDATE 2 +#define VHOST_IOTLB_INVALIDATE 3 +#define VHOST_IOTLB_ACCESS_FAIL 4 + __u8 type; +}; + +#define VHOST_IOTLB_MSG 0x1 + +struct vhost_msg { + int type; + union { + struct vhost_iotlb_msg iotlb; + __u8 padding[64]; + }; +}; + struct vhost_memory_region { __u64 guest_phys_addr; __u64 memory_size; /* bytes */ @@ -146,6 +172,8 @@ struct vhost_memory { #define VHOST_F_LOG_ALL 26 /* vhost-net should add virtio_net_hdr for RX, and strip for TX packets. */ #define VHOST_NET_F_VIRTIO_NET_HDR 27 +/* Vhost have device IOTLB */ +#define VHOST_F_DEVICE_IOTLB 63 /* VHOST_SCSI specific definitions */ -- cgit v1.2.3 From db3f60012482756f46cc4d7d9ad7d793ae30360c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 2 Aug 2016 14:03:36 -0700 Subject: uapi: move forward declarations of internal structures Don't user forward declarations of internal kernel structures in headers exported to userspace. Move "struct completion;". Move "struct task_struct;". Link: http://lkml.kernel.org/r/20160713215808.GA22486@p183.telecom.by Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/capability.h | 1 + include/linux/sysctl.h | 1 + include/uapi/linux/capability.h | 2 -- include/uapi/linux/sysctl.h | 2 -- 4 files changed, 2 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/capability.h b/include/linux/capability.h index 5f3c63dde2d5..dbc21c719ce6 100644 --- a/include/linux/capability.h +++ b/include/linux/capability.h @@ -38,6 +38,7 @@ struct cpu_vfs_cap_data { struct file; struct inode; struct dentry; +struct task_struct; struct user_namespace; extern const kernel_cap_t __cap_empty_set; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index fa7bc29925c9..697e160c78d0 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -28,6 +28,7 @@ #include /* For the /proc/sys support */ +struct completion; struct ctl_table; struct nsproxy; struct ctl_table_root; diff --git a/include/uapi/linux/capability.h b/include/uapi/linux/capability.h index 12c37a197d24..49bc06295398 100644 --- a/include/uapi/linux/capability.h +++ b/include/uapi/linux/capability.h @@ -15,8 +15,6 @@ #include -struct task_struct; - /* User-level do most of the mapping between kernel and user capabilities based on the version tag given by the kernel. The kernel might be somewhat backwards compatible, but don't bet on diff --git a/include/uapi/linux/sysctl.h b/include/uapi/linux/sysctl.h index 0956373b56db..d2b12152e358 100644 --- a/include/uapi/linux/sysctl.h +++ b/include/uapi/linux/sysctl.h @@ -26,8 +26,6 @@ #include #include -struct completion; - #define CTL_MAXNAME 10 /* how many path components do we allow in a call to sysctl? In other words, what is the largest acceptable value for the nlen -- cgit v1.2.3 From e63e88bc53bac7e4c3f592f8126c51a7569be673 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Tue, 2 Aug 2016 14:05:30 -0700 Subject: nilfs2: move ioctl interface and disk layout to uapi separately The header file "include/linux/nilfs2_fs.h" is composed of parts for ioctl and disk format, and both are intended to be shared with user space programs. This moves them to the uapi directory "include/uapi/linux" splitting the file to "nilfs2_api.h" and "nilfs2_ondisk.h". The following minor changes are accompanied by this migration: - nilfs_direct_node struct in nilfs2/direct.h is converged to nilfs2_ondisk.h because it's an on-disk structure. - inline functions nilfs_rec_len_from_disk() and nilfs_rec_len_to_disk() are moved to nilfs2/dir.c. Link: http://lkml.kernel.org/r/1465825507-3407-4-git-send-email-konishi.ryusuke@lab.ntt.co.jp Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/nilfs2.txt | 3 +- Documentation/ioctl/ioctl-number.txt | 2 +- MAINTAINERS | 3 +- fs/nilfs2/bmap.h | 2 +- fs/nilfs2/btree.h | 2 +- fs/nilfs2/cpfile.c | 1 - fs/nilfs2/cpfile.h | 3 +- fs/nilfs2/dat.h | 1 + fs/nilfs2/dir.c | 22 + fs/nilfs2/direct.h | 10 - fs/nilfs2/ifile.h | 1 - fs/nilfs2/ioctl.c | 1 - fs/nilfs2/nilfs.h | 3 +- fs/nilfs2/segment.h | 1 - fs/nilfs2/sufile.c | 1 - fs/nilfs2/sufile.h | 1 - include/linux/nilfs2_fs.h | 934 ----------------------------------- include/uapi/linux/nilfs2_api.h | 292 +++++++++++ include/uapi/linux/nilfs2_ondisk.h | 650 ++++++++++++++++++++++++ 19 files changed, 976 insertions(+), 957 deletions(-) delete mode 100644 include/linux/nilfs2_fs.h create mode 100644 include/uapi/linux/nilfs2_api.h create mode 100644 include/uapi/linux/nilfs2_ondisk.h (limited to 'include/uapi') diff --git a/Documentation/filesystems/nilfs2.txt b/Documentation/filesystems/nilfs2.txt index 5b21ef76f751..c0727dc36271 100644 --- a/Documentation/filesystems/nilfs2.txt +++ b/Documentation/filesystems/nilfs2.txt @@ -267,7 +267,8 @@ among NILFS2 files can be depicted as follows: `-- file (ino=yy) ( regular file, directory, or symlink ) -For detail on the format of each file, please see include/linux/nilfs2_fs.h. +For detail on the format of each file, please see nilfs2_ondisk.h +located at include/uapi/linux directory. There are no patents or other intellectual property that we protect with regard to the design of NILFS2. It is allowed to replicate the diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt index 56af5e43e9c0..81c7f2bb7daf 100644 --- a/Documentation/ioctl/ioctl-number.txt +++ b/Documentation/ioctl/ioctl-number.txt @@ -248,7 +248,7 @@ Code Seq#(hex) Include File Comments 'm' 00 drivers/scsi/megaraid/megaraid_ioctl.h conflict! 'm' 00-1F net/irda/irmod.h conflict! 'n' 00-7F linux/ncp_fs.h and fs/ncpfs/ioctl.c -'n' 80-8F linux/nilfs2_fs.h NILFS2 +'n' 80-8F uapi/linux/nilfs2_api.h NILFS2 'n' E0-FF linux/matroxfb.h matroxfb 'o' 00-1F fs/ocfs2/ocfs2_fs.h OCFS2 'o' 00-03 mtd/ubi-user.h conflict! (OCFS2 and UBI overlaps) diff --git a/MAINTAINERS b/MAINTAINERS index bb51bbbc9e1d..e9eacacf0f08 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8258,8 +8258,9 @@ T: git git://github.com/konis/nilfs2.git S: Supported F: Documentation/filesystems/nilfs2.txt F: fs/nilfs2/ -F: include/linux/nilfs2_fs.h F: include/trace/events/nilfs2.h +F: include/uapi/linux/nilfs2_api.h +F: include/uapi/linux/nilfs2_ondisk.h NINJA SCSI-3 / NINJA SCSI-32Bi (16bit/CardBus) PCMCIA SCSI HOST ADAPTER DRIVER M: YOKOTA Hiroshi diff --git a/fs/nilfs2/bmap.h b/fs/nilfs2/bmap.h index b6a4c8f93ac8..2b6ffbe5997a 100644 --- a/fs/nilfs2/bmap.h +++ b/fs/nilfs2/bmap.h @@ -22,7 +22,7 @@ #include #include #include -#include +#include /* nilfs_binfo, nilfs_inode, etc */ #include "alloc.h" #include "dat.h" diff --git a/fs/nilfs2/btree.h b/fs/nilfs2/btree.h index df1a25faa83b..2184e47fa4bf 100644 --- a/fs/nilfs2/btree.h +++ b/fs/nilfs2/btree.h @@ -22,7 +22,7 @@ #include #include #include -#include +#include /* nilfs_btree_node */ #include "btnode.h" #include "bmap.h" diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c index 19d9f4ae8347..a15a1601e931 100644 --- a/fs/nilfs2/cpfile.c +++ b/fs/nilfs2/cpfile.c @@ -21,7 +21,6 @@ #include #include #include -#include #include "mdt.h" #include "cpfile.h" diff --git a/fs/nilfs2/cpfile.h b/fs/nilfs2/cpfile.h index 0249744ae234..6eca972f9673 100644 --- a/fs/nilfs2/cpfile.h +++ b/fs/nilfs2/cpfile.h @@ -21,7 +21,8 @@ #include #include -#include +#include /* nilfs_cpstat */ +#include /* nilfs_inode, nilfs_checkpoint */ int nilfs_cpfile_get_checkpoint(struct inode *, __u64, int, diff --git a/fs/nilfs2/dat.h b/fs/nilfs2/dat.h index abbfdabcabea..57dc6cf466d0 100644 --- a/fs/nilfs2/dat.h +++ b/fs/nilfs2/dat.h @@ -22,6 +22,7 @@ #include #include #include +#include /* nilfs_inode, nilfs_checkpoint */ struct nilfs_palloc_req; diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c index 746956d2937a..908ebbf0ac7e 100644 --- a/fs/nilfs2/dir.c +++ b/fs/nilfs2/dir.c @@ -42,6 +42,28 @@ #include "nilfs.h" #include "page.h" +static inline unsigned int nilfs_rec_len_from_disk(__le16 dlen) +{ + unsigned int len = le16_to_cpu(dlen); + +#if (PAGE_SIZE >= 65536) + if (len == NILFS_MAX_REC_LEN) + return 1 << 16; +#endif + return len; +} + +static inline __le16 nilfs_rec_len_to_disk(unsigned int len) +{ +#if (PAGE_SIZE >= 65536) + if (len == (1 << 16)) + return cpu_to_le16(NILFS_MAX_REC_LEN); + + BUG_ON(len > (1 << 16)); +#endif + return cpu_to_le16(len); +} + /* * nilfs uses block-sized chunks. Arguably, sector-sized ones would be * more robust, but we have what we have diff --git a/fs/nilfs2/direct.h b/fs/nilfs2/direct.h index 3015a6e78724..cfe85e848bba 100644 --- a/fs/nilfs2/direct.h +++ b/fs/nilfs2/direct.h @@ -24,16 +24,6 @@ #include "bmap.h" -/** - * struct nilfs_direct_node - direct node - * @dn_flags: flags - * @dn_pad: padding - */ -struct nilfs_direct_node { - __u8 dn_flags; - __u8 pad[7]; -}; - #define NILFS_DIRECT_NBLOCKS (NILFS_BMAP_SIZE / sizeof(__le64) - 1) #define NILFS_DIRECT_KEY_MIN 0 #define NILFS_DIRECT_KEY_MAX (NILFS_DIRECT_NBLOCKS - 1) diff --git a/fs/nilfs2/ifile.h b/fs/nilfs2/ifile.h index 23ad2f091e76..188b94fe0ec5 100644 --- a/fs/nilfs2/ifile.h +++ b/fs/nilfs2/ifile.h @@ -23,7 +23,6 @@ #include #include -#include #include "mdt.h" #include "alloc.h" diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 827283fe9525..f1d7989459fd 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -25,7 +25,6 @@ #include /* compat_ptr() */ #include /* mnt_want_write_file(), mnt_drop_write_file() */ #include -#include #include "nilfs.h" #include "segment.h" #include "bmap.h" diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index 2ba8a146af1f..33f8c8fc96e8 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -23,7 +23,8 @@ #include #include #include -#include +#include +#include #include "the_nilfs.h" #include "bmap.h" diff --git a/fs/nilfs2/segment.h b/fs/nilfs2/segment.h index 6565c10b7b76..1060949d7dd2 100644 --- a/fs/nilfs2/segment.h +++ b/fs/nilfs2/segment.h @@ -23,7 +23,6 @@ #include #include #include -#include #include "nilfs.h" struct nilfs_root; diff --git a/fs/nilfs2/sufile.c b/fs/nilfs2/sufile.c index 12d11de93602..1541a1e9221a 100644 --- a/fs/nilfs2/sufile.c +++ b/fs/nilfs2/sufile.c @@ -22,7 +22,6 @@ #include #include #include -#include #include "mdt.h" #include "sufile.h" diff --git a/fs/nilfs2/sufile.h b/fs/nilfs2/sufile.h index 46e89872294c..158a9190c8ec 100644 --- a/fs/nilfs2/sufile.h +++ b/fs/nilfs2/sufile.h @@ -21,7 +21,6 @@ #include #include -#include #include "mdt.h" diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h deleted file mode 100644 index 5988dd57ba66..000000000000 --- a/include/linux/nilfs2_fs.h +++ /dev/null @@ -1,934 +0,0 @@ -/* - * nilfs2_fs.h - NILFS2 on-disk structures and common declarations. - * - * Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU Lesser General Public License as published - * by the Free Software Foundation; either version 2.1 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU Lesser General Public License for more details. - * - * Written by Koji Sato and Ryusuke Konishi. - */ -/* - * linux/include/linux/ext2_fs.h - * - * Copyright (C) 1992, 1993, 1994, 1995 - * Remy Card (card@masi.ibp.fr) - * Laboratoire MASI - Institut Blaise Pascal - * Universite Pierre et Marie Curie (Paris VI) - * - * from - * - * linux/include/linux/minix_fs.h - * - * Copyright (C) 1991, 1992 Linus Torvalds - */ - -#ifndef _LINUX_NILFS_FS_H -#define _LINUX_NILFS_FS_H - -#include -#include -#include -#include - - -#define NILFS_INODE_BMAP_SIZE 7 -/** - * struct nilfs_inode - structure of an inode on disk - * @i_blocks: blocks count - * @i_size: size in bytes - * @i_ctime: creation time (seconds) - * @i_mtime: modification time (seconds) - * @i_ctime_nsec: creation time (nano seconds) - * @i_mtime_nsec: modification time (nano seconds) - * @i_uid: user id - * @i_gid: group id - * @i_mode: file mode - * @i_links_count: links count - * @i_flags: file flags - * @i_bmap: block mapping - * @i_xattr: extended attributes - * @i_generation: file generation (for NFS) - * @i_pad: padding - */ -struct nilfs_inode { - __le64 i_blocks; - __le64 i_size; - __le64 i_ctime; - __le64 i_mtime; - __le32 i_ctime_nsec; - __le32 i_mtime_nsec; - __le32 i_uid; - __le32 i_gid; - __le16 i_mode; - __le16 i_links_count; - __le32 i_flags; - __le64 i_bmap[NILFS_INODE_BMAP_SIZE]; -#define i_device_code i_bmap[0] - __le64 i_xattr; - __le32 i_generation; - __le32 i_pad; -}; - -#define NILFS_MIN_INODE_SIZE 128 - -/** - * struct nilfs_super_root - structure of super root - * @sr_sum: check sum - * @sr_bytes: byte count of the structure - * @sr_flags: flags (reserved) - * @sr_nongc_ctime: write time of the last segment not for cleaner operation - * @sr_dat: DAT file inode - * @sr_cpfile: checkpoint file inode - * @sr_sufile: segment usage file inode - */ -struct nilfs_super_root { - __le32 sr_sum; - __le16 sr_bytes; - __le16 sr_flags; - __le64 sr_nongc_ctime; - struct nilfs_inode sr_dat; - struct nilfs_inode sr_cpfile; - struct nilfs_inode sr_sufile; -}; - -#define NILFS_SR_MDT_OFFSET(inode_size, i) \ - ((unsigned long)&((struct nilfs_super_root *)0)->sr_dat + \ - (inode_size) * (i)) -#define NILFS_SR_DAT_OFFSET(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 0) -#define NILFS_SR_CPFILE_OFFSET(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 1) -#define NILFS_SR_SUFILE_OFFSET(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 2) -#define NILFS_SR_BYTES(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 3) - -/* - * Maximal mount counts - */ -#define NILFS_DFL_MAX_MNT_COUNT 50 /* 50 mounts */ - -/* - * File system states (sbp->s_state, nilfs->ns_mount_state) - */ -#define NILFS_VALID_FS 0x0001 /* Unmounted cleanly */ -#define NILFS_ERROR_FS 0x0002 /* Errors detected */ -#define NILFS_RESIZE_FS 0x0004 /* Resize required */ - -/* - * Mount flags (sbi->s_mount_opt) - */ -#define NILFS_MOUNT_ERROR_MODE 0x0070 /* Error mode mask */ -#define NILFS_MOUNT_ERRORS_CONT 0x0010 /* Continue on errors */ -#define NILFS_MOUNT_ERRORS_RO 0x0020 /* Remount fs ro on errors */ -#define NILFS_MOUNT_ERRORS_PANIC 0x0040 /* Panic on errors */ -#define NILFS_MOUNT_BARRIER 0x1000 /* Use block barriers */ -#define NILFS_MOUNT_STRICT_ORDER 0x2000 /* - * Apply strict in-order - * semantics also for data - */ -#define NILFS_MOUNT_NORECOVERY 0x4000 /* - * Disable write access during - * mount-time recovery - */ -#define NILFS_MOUNT_DISCARD 0x8000 /* Issue DISCARD requests */ - - -/** - * struct nilfs_super_block - structure of super block on disk - */ -struct nilfs_super_block { -/*00*/ __le32 s_rev_level; /* Revision level */ - __le16 s_minor_rev_level; /* minor revision level */ - __le16 s_magic; /* Magic signature */ - - __le16 s_bytes; /* - * Bytes count of CRC calculation - * for this structure. s_reserved - * is excluded. - */ - __le16 s_flags; /* flags */ - __le32 s_crc_seed; /* Seed value of CRC calculation */ -/*10*/ __le32 s_sum; /* Check sum of super block */ - - __le32 s_log_block_size; /* - * Block size represented as follows - * blocksize = - * 1 << (s_log_block_size + 10) - */ - __le64 s_nsegments; /* Number of segments in filesystem */ -/*20*/ __le64 s_dev_size; /* block device size in bytes */ - __le64 s_first_data_block; /* 1st seg disk block number */ -/*30*/ __le32 s_blocks_per_segment; /* number of blocks per full segment */ - __le32 s_r_segments_percentage; /* Reserved segments percentage */ - - __le64 s_last_cno; /* Last checkpoint number */ -/*40*/ __le64 s_last_pseg; /* disk block addr pseg written last */ - __le64 s_last_seq; /* seq. number of seg written last */ -/*50*/ __le64 s_free_blocks_count; /* Free blocks count */ - - __le64 s_ctime; /* - * Creation time (execution time of - * newfs) - */ -/*60*/ __le64 s_mtime; /* Mount time */ - __le64 s_wtime; /* Write time */ -/*70*/ __le16 s_mnt_count; /* Mount count */ - __le16 s_max_mnt_count; /* Maximal mount count */ - __le16 s_state; /* File system state */ - __le16 s_errors; /* Behaviour when detecting errors */ - __le64 s_lastcheck; /* time of last check */ - -/*80*/ __le32 s_checkinterval; /* max. time between checks */ - __le32 s_creator_os; /* OS */ - __le16 s_def_resuid; /* Default uid for reserved blocks */ - __le16 s_def_resgid; /* Default gid for reserved blocks */ - __le32 s_first_ino; /* First non-reserved inode */ - -/*90*/ __le16 s_inode_size; /* Size of an inode */ - __le16 s_dat_entry_size; /* Size of a dat entry */ - __le16 s_checkpoint_size; /* Size of a checkpoint */ - __le16 s_segment_usage_size; /* Size of a segment usage */ - -/*98*/ __u8 s_uuid[16]; /* 128-bit uuid for volume */ -/*A8*/ char s_volume_name[80]; /* volume name */ - -/*F8*/ __le32 s_c_interval; /* Commit interval of segment */ - __le32 s_c_block_max; /* - * Threshold of data amount for - * the segment construction - */ -/*100*/ __le64 s_feature_compat; /* Compatible feature set */ - __le64 s_feature_compat_ro; /* Read-only compatible feature set */ - __le64 s_feature_incompat; /* Incompatible feature set */ - __u32 s_reserved[186]; /* padding to the end of the block */ -}; - -/* - * Codes for operating systems - */ -#define NILFS_OS_LINUX 0 -/* Codes from 1 to 4 are reserved to keep compatibility with ext2 creator-OS */ - -/* - * Revision levels - */ -#define NILFS_CURRENT_REV 2 /* current major revision */ -#define NILFS_MINOR_REV 0 /* minor revision */ -#define NILFS_MIN_SUPP_REV 2 /* minimum supported revision */ - -/* - * Feature set definitions - * - * If there is a bit set in the incompatible feature set that the kernel - * doesn't know about, it should refuse to mount the filesystem. - */ -#define NILFS_FEATURE_COMPAT_RO_BLOCK_COUNT 0x00000001ULL - -#define NILFS_FEATURE_COMPAT_SUPP 0ULL -#define NILFS_FEATURE_COMPAT_RO_SUPP NILFS_FEATURE_COMPAT_RO_BLOCK_COUNT -#define NILFS_FEATURE_INCOMPAT_SUPP 0ULL - -/* - * Bytes count of super_block for CRC-calculation - */ -#define NILFS_SB_BYTES \ - ((long)&((struct nilfs_super_block *)0)->s_reserved) - -/* - * Special inode number - */ -#define NILFS_ROOT_INO 2 /* Root file inode */ -#define NILFS_DAT_INO 3 /* DAT file */ -#define NILFS_CPFILE_INO 4 /* checkpoint file */ -#define NILFS_SUFILE_INO 5 /* segment usage file */ -#define NILFS_IFILE_INO 6 /* ifile */ -#define NILFS_ATIME_INO 7 /* Atime file (reserved) */ -#define NILFS_XATTR_INO 8 /* Xattribute file (reserved) */ -#define NILFS_SKETCH_INO 10 /* Sketch file */ -#define NILFS_USER_INO 11 /* Fisrt user's file inode number */ - -#define NILFS_SB_OFFSET_BYTES 1024 /* byte offset of nilfs superblock */ - -#define NILFS_SEG_MIN_BLOCKS 16 /* - * Minimum number of blocks in - * a full segment - */ -#define NILFS_PSEG_MIN_BLOCKS 2 /* - * Minimum number of blocks in - * a partial segment - */ -#define NILFS_MIN_NRSVSEGS 8 /* - * Minimum number of reserved - * segments - */ - -/* - * We call DAT, cpfile, and sufile root metadata files. Inodes of - * these files are written in super root block instead of ifile, and - * garbage collector doesn't keep any past versions of these files. - */ -#define NILFS_ROOT_METADATA_FILE(ino) \ - ((ino) >= NILFS_DAT_INO && (ino) <= NILFS_SUFILE_INO) - -/* - * bytes offset of secondary super block - */ -#define NILFS_SB2_OFFSET_BYTES(devsize) ((((devsize) >> 12) - 1) << 12) - -/* - * Maximal count of links to a file - */ -#define NILFS_LINK_MAX 32000 - -/* - * Structure of a directory entry - * (Same as ext2) - */ - -#define NILFS_NAME_LEN 255 - -/* - * Block size limitations - */ -#define NILFS_MIN_BLOCK_SIZE 1024 -#define NILFS_MAX_BLOCK_SIZE 65536 - -/* - * The new version of the directory entry. Since V0 structures are - * stored in intel byte order, and the name_len field could never be - * bigger than 255 chars, it's safe to reclaim the extra byte for the - * file_type field. - */ -struct nilfs_dir_entry { - __le64 inode; /* Inode number */ - __le16 rec_len; /* Directory entry length */ - __u8 name_len; /* Name length */ - __u8 file_type; /* Dir entry type (file, dir, etc) */ - char name[NILFS_NAME_LEN]; /* File name */ - char pad; -}; - -/* - * NILFS directory file types. Only the low 3 bits are used. The - * other bits are reserved for now. - */ -enum { - NILFS_FT_UNKNOWN, - NILFS_FT_REG_FILE, - NILFS_FT_DIR, - NILFS_FT_CHRDEV, - NILFS_FT_BLKDEV, - NILFS_FT_FIFO, - NILFS_FT_SOCK, - NILFS_FT_SYMLINK, - NILFS_FT_MAX -}; - -/* - * NILFS_DIR_PAD defines the directory entries boundaries - * - * NOTE: It must be a multiple of 8 - */ -#define NILFS_DIR_PAD 8 -#define NILFS_DIR_ROUND (NILFS_DIR_PAD - 1) -#define NILFS_DIR_REC_LEN(name_len) (((name_len) + 12 + NILFS_DIR_ROUND) & \ - ~NILFS_DIR_ROUND) -#define NILFS_MAX_REC_LEN ((1<<16)-1) - -static inline unsigned int nilfs_rec_len_from_disk(__le16 dlen) -{ - unsigned int len = le16_to_cpu(dlen); - -#if !defined(__KERNEL__) || (PAGE_SIZE >= 65536) - if (len == NILFS_MAX_REC_LEN) - return 1 << 16; -#endif - return len; -} - -static inline __le16 nilfs_rec_len_to_disk(unsigned int len) -{ -#if !defined(__KERNEL__) || (PAGE_SIZE >= 65536) - if (len == (1 << 16)) - return cpu_to_le16(NILFS_MAX_REC_LEN); - else if (len > (1 << 16)) - BUG(); -#endif - return cpu_to_le16(len); -} - -/** - * struct nilfs_finfo - file information - * @fi_ino: inode number - * @fi_cno: checkpoint number - * @fi_nblocks: number of blocks (including intermediate blocks) - * @fi_ndatablk: number of file data blocks - */ -struct nilfs_finfo { - __le64 fi_ino; - __le64 fi_cno; - __le32 fi_nblocks; - __le32 fi_ndatablk; - /* array of virtual block numbers */ -}; - -/** - * struct nilfs_binfo_v - information for the block to which a virtual block number is assigned - * @bi_vblocknr: virtual block number - * @bi_blkoff: block offset - */ -struct nilfs_binfo_v { - __le64 bi_vblocknr; - __le64 bi_blkoff; -}; - -/** - * struct nilfs_binfo_dat - information for the block which belongs to the DAT file - * @bi_blkoff: block offset - * @bi_level: level - * @bi_pad: padding - */ -struct nilfs_binfo_dat { - __le64 bi_blkoff; - __u8 bi_level; - __u8 bi_pad[7]; -}; - -/** - * union nilfs_binfo: block information - * @bi_v: nilfs_binfo_v structure - * @bi_dat: nilfs_binfo_dat structure - */ -union nilfs_binfo { - struct nilfs_binfo_v bi_v; - struct nilfs_binfo_dat bi_dat; -}; - -/** - * struct nilfs_segment_summary - segment summary header - * @ss_datasum: checksum of data - * @ss_sumsum: checksum of segment summary - * @ss_magic: magic number - * @ss_bytes: size of this structure in bytes - * @ss_flags: flags - * @ss_seq: sequence number - * @ss_create: creation timestamp - * @ss_next: next segment - * @ss_nblocks: number of blocks - * @ss_nfinfo: number of finfo structures - * @ss_sumbytes: total size of segment summary in bytes - * @ss_pad: padding - * @ss_cno: checkpoint number - */ -struct nilfs_segment_summary { - __le32 ss_datasum; - __le32 ss_sumsum; - __le32 ss_magic; - __le16 ss_bytes; - __le16 ss_flags; - __le64 ss_seq; - __le64 ss_create; - __le64 ss_next; - __le32 ss_nblocks; - __le32 ss_nfinfo; - __le32 ss_sumbytes; - __le32 ss_pad; - __le64 ss_cno; - /* array of finfo structures */ -}; - -#define NILFS_SEGSUM_MAGIC 0x1eaffa11 /* segment summary magic number */ - -/* - * Segment summary flags - */ -#define NILFS_SS_LOGBGN 0x0001 /* begins a logical segment */ -#define NILFS_SS_LOGEND 0x0002 /* ends a logical segment */ -#define NILFS_SS_SR 0x0004 /* has super root */ -#define NILFS_SS_SYNDT 0x0008 /* includes data only updates */ -#define NILFS_SS_GC 0x0010 /* segment written for cleaner operation */ - -/** - * struct nilfs_btree_node - B-tree node - * @bn_flags: flags - * @bn_level: level - * @bn_nchildren: number of children - * @bn_pad: padding - */ -struct nilfs_btree_node { - __u8 bn_flags; - __u8 bn_level; - __le16 bn_nchildren; - __le32 bn_pad; -}; - -/* flags */ -#define NILFS_BTREE_NODE_ROOT 0x01 - -/* level */ -#define NILFS_BTREE_LEVEL_DATA 0 -#define NILFS_BTREE_LEVEL_NODE_MIN (NILFS_BTREE_LEVEL_DATA + 1) -#define NILFS_BTREE_LEVEL_MAX 14 /* Max level (exclusive) */ - -/** - * struct nilfs_palloc_group_desc - block group descriptor - * @pg_nfrees: number of free entries in block group - */ -struct nilfs_palloc_group_desc { - __le32 pg_nfrees; -}; - -/** - * struct nilfs_dat_entry - disk address translation entry - * @de_blocknr: block number - * @de_start: start checkpoint number - * @de_end: end checkpoint number - * @de_rsv: reserved for future use - */ -struct nilfs_dat_entry { - __le64 de_blocknr; - __le64 de_start; - __le64 de_end; - __le64 de_rsv; -}; - -#define NILFS_MIN_DAT_ENTRY_SIZE 32 - -/** - * struct nilfs_snapshot_list - snapshot list - * @ssl_next: next checkpoint number on snapshot list - * @ssl_prev: previous checkpoint number on snapshot list - */ -struct nilfs_snapshot_list { - __le64 ssl_next; - __le64 ssl_prev; -}; - -/** - * struct nilfs_checkpoint - checkpoint structure - * @cp_flags: flags - * @cp_checkpoints_count: checkpoints count in a block - * @cp_snapshot_list: snapshot list - * @cp_cno: checkpoint number - * @cp_create: creation timestamp - * @cp_nblk_inc: number of blocks incremented by this checkpoint - * @cp_inodes_count: inodes count - * @cp_blocks_count: blocks count - * @cp_ifile_inode: inode of ifile - */ -struct nilfs_checkpoint { - __le32 cp_flags; - __le32 cp_checkpoints_count; - struct nilfs_snapshot_list cp_snapshot_list; - __le64 cp_cno; - __le64 cp_create; - __le64 cp_nblk_inc; - __le64 cp_inodes_count; - __le64 cp_blocks_count; - - /* - * Do not change the byte offset of ifile inode. - * To keep the compatibility of the disk format, - * additional fields should be added behind cp_ifile_inode. - */ - struct nilfs_inode cp_ifile_inode; -}; - -#define NILFS_MIN_CHECKPOINT_SIZE (64 + NILFS_MIN_INODE_SIZE) - -/* checkpoint flags */ -enum { - NILFS_CHECKPOINT_SNAPSHOT, - NILFS_CHECKPOINT_INVALID, - NILFS_CHECKPOINT_SKETCH, - NILFS_CHECKPOINT_MINOR, -}; - -#define NILFS_CHECKPOINT_FNS(flag, name) \ -static inline void \ -nilfs_checkpoint_set_##name(struct nilfs_checkpoint *cp) \ -{ \ - cp->cp_flags = cpu_to_le32(le32_to_cpu(cp->cp_flags) | \ - (1UL << NILFS_CHECKPOINT_##flag)); \ -} \ -static inline void \ -nilfs_checkpoint_clear_##name(struct nilfs_checkpoint *cp) \ -{ \ - cp->cp_flags = cpu_to_le32(le32_to_cpu(cp->cp_flags) & \ - ~(1UL << NILFS_CHECKPOINT_##flag)); \ -} \ -static inline int \ -nilfs_checkpoint_##name(const struct nilfs_checkpoint *cp) \ -{ \ - return !!(le32_to_cpu(cp->cp_flags) & \ - (1UL << NILFS_CHECKPOINT_##flag)); \ -} - -NILFS_CHECKPOINT_FNS(SNAPSHOT, snapshot) -NILFS_CHECKPOINT_FNS(INVALID, invalid) -NILFS_CHECKPOINT_FNS(MINOR, minor) - -/** - * struct nilfs_cpinfo - checkpoint information - * @ci_flags: flags - * @ci_pad: padding - * @ci_cno: checkpoint number - * @ci_create: creation timestamp - * @ci_nblk_inc: number of blocks incremented by this checkpoint - * @ci_inodes_count: inodes count - * @ci_blocks_count: blocks count - * @ci_next: next checkpoint number in snapshot list - */ -struct nilfs_cpinfo { - __u32 ci_flags; - __u32 ci_pad; - __u64 ci_cno; - __u64 ci_create; - __u64 ci_nblk_inc; - __u64 ci_inodes_count; - __u64 ci_blocks_count; - __u64 ci_next; -}; - -#define NILFS_CPINFO_FNS(flag, name) \ -static inline int \ -nilfs_cpinfo_##name(const struct nilfs_cpinfo *cpinfo) \ -{ \ - return !!(cpinfo->ci_flags & (1UL << NILFS_CHECKPOINT_##flag)); \ -} - -NILFS_CPINFO_FNS(SNAPSHOT, snapshot) -NILFS_CPINFO_FNS(INVALID, invalid) -NILFS_CPINFO_FNS(MINOR, minor) - - -/** - * struct nilfs_cpfile_header - checkpoint file header - * @ch_ncheckpoints: number of checkpoints - * @ch_nsnapshots: number of snapshots - * @ch_snapshot_list: snapshot list - */ -struct nilfs_cpfile_header { - __le64 ch_ncheckpoints; - __le64 ch_nsnapshots; - struct nilfs_snapshot_list ch_snapshot_list; -}; - -#define NILFS_CPFILE_FIRST_CHECKPOINT_OFFSET \ - ((sizeof(struct nilfs_cpfile_header) + \ - sizeof(struct nilfs_checkpoint) - 1) / \ - sizeof(struct nilfs_checkpoint)) - -/** - * struct nilfs_segment_usage - segment usage - * @su_lastmod: last modified timestamp - * @su_nblocks: number of blocks in segment - * @su_flags: flags - */ -struct nilfs_segment_usage { - __le64 su_lastmod; - __le32 su_nblocks; - __le32 su_flags; -}; - -#define NILFS_MIN_SEGMENT_USAGE_SIZE 16 - -/* segment usage flag */ -enum { - NILFS_SEGMENT_USAGE_ACTIVE, - NILFS_SEGMENT_USAGE_DIRTY, - NILFS_SEGMENT_USAGE_ERROR, - - /* ... */ -}; - -#define NILFS_SEGMENT_USAGE_FNS(flag, name) \ -static inline void \ -nilfs_segment_usage_set_##name(struct nilfs_segment_usage *su) \ -{ \ - su->su_flags = cpu_to_le32(le32_to_cpu(su->su_flags) | \ - (1UL << NILFS_SEGMENT_USAGE_##flag));\ -} \ -static inline void \ -nilfs_segment_usage_clear_##name(struct nilfs_segment_usage *su) \ -{ \ - su->su_flags = \ - cpu_to_le32(le32_to_cpu(su->su_flags) & \ - ~(1UL << NILFS_SEGMENT_USAGE_##flag)); \ -} \ -static inline int \ -nilfs_segment_usage_##name(const struct nilfs_segment_usage *su) \ -{ \ - return !!(le32_to_cpu(su->su_flags) & \ - (1UL << NILFS_SEGMENT_USAGE_##flag)); \ -} - -NILFS_SEGMENT_USAGE_FNS(ACTIVE, active) -NILFS_SEGMENT_USAGE_FNS(DIRTY, dirty) -NILFS_SEGMENT_USAGE_FNS(ERROR, error) - -static inline void -nilfs_segment_usage_set_clean(struct nilfs_segment_usage *su) -{ - su->su_lastmod = cpu_to_le64(0); - su->su_nblocks = cpu_to_le32(0); - su->su_flags = cpu_to_le32(0); -} - -static inline int -nilfs_segment_usage_clean(const struct nilfs_segment_usage *su) -{ - return !le32_to_cpu(su->su_flags); -} - -/** - * struct nilfs_sufile_header - segment usage file header - * @sh_ncleansegs: number of clean segments - * @sh_ndirtysegs: number of dirty segments - * @sh_last_alloc: last allocated segment number - */ -struct nilfs_sufile_header { - __le64 sh_ncleansegs; - __le64 sh_ndirtysegs; - __le64 sh_last_alloc; - /* ... */ -}; - -#define NILFS_SUFILE_FIRST_SEGMENT_USAGE_OFFSET \ - ((sizeof(struct nilfs_sufile_header) + \ - sizeof(struct nilfs_segment_usage) - 1) / \ - sizeof(struct nilfs_segment_usage)) - -/** - * nilfs_suinfo - segment usage information - * @sui_lastmod: timestamp of last modification - * @sui_nblocks: number of written blocks in segment - * @sui_flags: segment usage flags - */ -struct nilfs_suinfo { - __u64 sui_lastmod; - __u32 sui_nblocks; - __u32 sui_flags; -}; - -#define NILFS_SUINFO_FNS(flag, name) \ -static inline int \ -nilfs_suinfo_##name(const struct nilfs_suinfo *si) \ -{ \ - return si->sui_flags & (1UL << NILFS_SEGMENT_USAGE_##flag); \ -} - -NILFS_SUINFO_FNS(ACTIVE, active) -NILFS_SUINFO_FNS(DIRTY, dirty) -NILFS_SUINFO_FNS(ERROR, error) - -static inline int nilfs_suinfo_clean(const struct nilfs_suinfo *si) -{ - return !si->sui_flags; -} - -/* ioctl */ -/** - * nilfs_suinfo_update - segment usage information update - * @sup_segnum: segment number - * @sup_flags: flags for which fields are active in sup_sui - * @sup_reserved: reserved necessary for alignment - * @sup_sui: segment usage information - */ -struct nilfs_suinfo_update { - __u64 sup_segnum; - __u32 sup_flags; - __u32 sup_reserved; - struct nilfs_suinfo sup_sui; -}; - -enum { - NILFS_SUINFO_UPDATE_LASTMOD, - NILFS_SUINFO_UPDATE_NBLOCKS, - NILFS_SUINFO_UPDATE_FLAGS, - __NR_NILFS_SUINFO_UPDATE_FIELDS, -}; - -#define NILFS_SUINFO_UPDATE_FNS(flag, name) \ -static inline void \ -nilfs_suinfo_update_set_##name(struct nilfs_suinfo_update *sup) \ -{ \ - sup->sup_flags |= 1UL << NILFS_SUINFO_UPDATE_##flag; \ -} \ -static inline void \ -nilfs_suinfo_update_clear_##name(struct nilfs_suinfo_update *sup) \ -{ \ - sup->sup_flags &= ~(1UL << NILFS_SUINFO_UPDATE_##flag); \ -} \ -static inline int \ -nilfs_suinfo_update_##name(const struct nilfs_suinfo_update *sup) \ -{ \ - return !!(sup->sup_flags & (1UL << NILFS_SUINFO_UPDATE_##flag));\ -} - -NILFS_SUINFO_UPDATE_FNS(LASTMOD, lastmod) -NILFS_SUINFO_UPDATE_FNS(NBLOCKS, nblocks) -NILFS_SUINFO_UPDATE_FNS(FLAGS, flags) - -enum { - NILFS_CHECKPOINT, - NILFS_SNAPSHOT, -}; - -/** - * struct nilfs_cpmode - change checkpoint mode structure - * @cm_cno: checkpoint number - * @cm_mode: mode of checkpoint - * @cm_pad: padding - */ -struct nilfs_cpmode { - __u64 cm_cno; - __u32 cm_mode; - __u32 cm_pad; -}; - -/** - * struct nilfs_argv - argument vector - * @v_base: pointer on data array from userspace - * @v_nmembs: number of members in data array - * @v_size: size of data array in bytes - * @v_flags: flags - * @v_index: start number of target data items - */ -struct nilfs_argv { - __u64 v_base; - __u32 v_nmembs; /* number of members */ - __u16 v_size; /* size of members */ - __u16 v_flags; - __u64 v_index; -}; - -/** - * struct nilfs_period - period of checkpoint numbers - * @p_start: start checkpoint number (inclusive) - * @p_end: end checkpoint number (exclusive) - */ -struct nilfs_period { - __u64 p_start; - __u64 p_end; -}; - -/** - * struct nilfs_cpstat - checkpoint statistics - * @cs_cno: checkpoint number - * @cs_ncps: number of checkpoints - * @cs_nsss: number of snapshots - */ -struct nilfs_cpstat { - __u64 cs_cno; - __u64 cs_ncps; - __u64 cs_nsss; -}; - -/** - * struct nilfs_sustat - segment usage statistics - * @ss_nsegs: number of segments - * @ss_ncleansegs: number of clean segments - * @ss_ndirtysegs: number of dirty segments - * @ss_ctime: creation time of the last segment - * @ss_nongc_ctime: creation time of the last segment not for GC - * @ss_prot_seq: least sequence number of segments which must not be reclaimed - */ -struct nilfs_sustat { - __u64 ss_nsegs; - __u64 ss_ncleansegs; - __u64 ss_ndirtysegs; - __u64 ss_ctime; - __u64 ss_nongc_ctime; - __u64 ss_prot_seq; -}; - -/** - * struct nilfs_vinfo - virtual block number information - * @vi_vblocknr: virtual block number - * @vi_start: start checkpoint number (inclusive) - * @vi_end: end checkpoint number (exclusive) - * @vi_blocknr: disk block number - */ -struct nilfs_vinfo { - __u64 vi_vblocknr; - __u64 vi_start; - __u64 vi_end; - __u64 vi_blocknr; -}; - -/** - * struct nilfs_vdesc - descriptor of virtual block number - * @vd_ino: inode number - * @vd_cno: checkpoint number - * @vd_vblocknr: virtual block number - * @vd_period: period of checkpoint numbers - * @vd_blocknr: disk block number - * @vd_offset: logical block offset inside a file - * @vd_flags: flags (data or node block) - * @vd_pad: padding - */ -struct nilfs_vdesc { - __u64 vd_ino; - __u64 vd_cno; - __u64 vd_vblocknr; - struct nilfs_period vd_period; - __u64 vd_blocknr; - __u64 vd_offset; - __u32 vd_flags; - __u32 vd_pad; -}; - -/** - * struct nilfs_bdesc - descriptor of disk block number - * @bd_ino: inode number - * @bd_oblocknr: disk block address (for skipping dead blocks) - * @bd_blocknr: disk block address - * @bd_offset: logical block offset inside a file - * @bd_level: level in the b-tree organization - * @bd_pad: padding - */ -struct nilfs_bdesc { - __u64 bd_ino; - __u64 bd_oblocknr; - __u64 bd_blocknr; - __u64 bd_offset; - __u32 bd_level; - __u32 bd_pad; -}; - -#define NILFS_IOCTL_IDENT 'n' - -#define NILFS_IOCTL_CHANGE_CPMODE \ - _IOW(NILFS_IOCTL_IDENT, 0x80, struct nilfs_cpmode) -#define NILFS_IOCTL_DELETE_CHECKPOINT \ - _IOW(NILFS_IOCTL_IDENT, 0x81, __u64) -#define NILFS_IOCTL_GET_CPINFO \ - _IOR(NILFS_IOCTL_IDENT, 0x82, struct nilfs_argv) -#define NILFS_IOCTL_GET_CPSTAT \ - _IOR(NILFS_IOCTL_IDENT, 0x83, struct nilfs_cpstat) -#define NILFS_IOCTL_GET_SUINFO \ - _IOR(NILFS_IOCTL_IDENT, 0x84, struct nilfs_argv) -#define NILFS_IOCTL_GET_SUSTAT \ - _IOR(NILFS_IOCTL_IDENT, 0x85, struct nilfs_sustat) -#define NILFS_IOCTL_GET_VINFO \ - _IOWR(NILFS_IOCTL_IDENT, 0x86, struct nilfs_argv) -#define NILFS_IOCTL_GET_BDESCS \ - _IOWR(NILFS_IOCTL_IDENT, 0x87, struct nilfs_argv) -#define NILFS_IOCTL_CLEAN_SEGMENTS \ - _IOW(NILFS_IOCTL_IDENT, 0x88, struct nilfs_argv[5]) -#define NILFS_IOCTL_SYNC \ - _IOR(NILFS_IOCTL_IDENT, 0x8A, __u64) -#define NILFS_IOCTL_RESIZE \ - _IOW(NILFS_IOCTL_IDENT, 0x8B, __u64) -#define NILFS_IOCTL_SET_ALLOC_RANGE \ - _IOW(NILFS_IOCTL_IDENT, 0x8C, __u64[2]) -#define NILFS_IOCTL_SET_SUINFO \ - _IOW(NILFS_IOCTL_IDENT, 0x8D, struct nilfs_argv) - -#endif /* _LINUX_NILFS_FS_H */ diff --git a/include/uapi/linux/nilfs2_api.h b/include/uapi/linux/nilfs2_api.h new file mode 100644 index 000000000000..ef4c1de89b11 --- /dev/null +++ b/include/uapi/linux/nilfs2_api.h @@ -0,0 +1,292 @@ +/* + * nilfs2_api.h - NILFS2 user space API + * + * Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + */ + +#ifndef _LINUX_NILFS2_API_H +#define _LINUX_NILFS2_API_H + +#include +#include + +/** + * struct nilfs_cpinfo - checkpoint information + * @ci_flags: flags + * @ci_pad: padding + * @ci_cno: checkpoint number + * @ci_create: creation timestamp + * @ci_nblk_inc: number of blocks incremented by this checkpoint + * @ci_inodes_count: inodes count + * @ci_blocks_count: blocks count + * @ci_next: next checkpoint number in snapshot list + */ +struct nilfs_cpinfo { + __u32 ci_flags; + __u32 ci_pad; + __u64 ci_cno; + __u64 ci_create; + __u64 ci_nblk_inc; + __u64 ci_inodes_count; + __u64 ci_blocks_count; + __u64 ci_next; +}; + +/* checkpoint flags */ +enum { + NILFS_CPINFO_SNAPSHOT, + NILFS_CPINFO_INVALID, + NILFS_CPINFO_SKETCH, + NILFS_CPINFO_MINOR, +}; + +#define NILFS_CPINFO_FNS(flag, name) \ +static inline int \ +nilfs_cpinfo_##name(const struct nilfs_cpinfo *cpinfo) \ +{ \ + return !!(cpinfo->ci_flags & (1UL << NILFS_CPINFO_##flag)); \ +} + +NILFS_CPINFO_FNS(SNAPSHOT, snapshot) +NILFS_CPINFO_FNS(INVALID, invalid) +NILFS_CPINFO_FNS(MINOR, minor) + +/** + * nilfs_suinfo - segment usage information + * @sui_lastmod: timestamp of last modification + * @sui_nblocks: number of written blocks in segment + * @sui_flags: segment usage flags + */ +struct nilfs_suinfo { + __u64 sui_lastmod; + __u32 sui_nblocks; + __u32 sui_flags; +}; + +/* segment usage flags */ +enum { + NILFS_SUINFO_ACTIVE, + NILFS_SUINFO_DIRTY, + NILFS_SUINFO_ERROR, +}; + +#define NILFS_SUINFO_FNS(flag, name) \ +static inline int \ +nilfs_suinfo_##name(const struct nilfs_suinfo *si) \ +{ \ + return si->sui_flags & (1UL << NILFS_SUINFO_##flag); \ +} + +NILFS_SUINFO_FNS(ACTIVE, active) +NILFS_SUINFO_FNS(DIRTY, dirty) +NILFS_SUINFO_FNS(ERROR, error) + +static inline int nilfs_suinfo_clean(const struct nilfs_suinfo *si) +{ + return !si->sui_flags; +} + +/** + * nilfs_suinfo_update - segment usage information update + * @sup_segnum: segment number + * @sup_flags: flags for which fields are active in sup_sui + * @sup_reserved: reserved necessary for alignment + * @sup_sui: segment usage information + */ +struct nilfs_suinfo_update { + __u64 sup_segnum; + __u32 sup_flags; + __u32 sup_reserved; + struct nilfs_suinfo sup_sui; +}; + +enum { + NILFS_SUINFO_UPDATE_LASTMOD, + NILFS_SUINFO_UPDATE_NBLOCKS, + NILFS_SUINFO_UPDATE_FLAGS, + __NR_NILFS_SUINFO_UPDATE_FIELDS, +}; + +#define NILFS_SUINFO_UPDATE_FNS(flag, name) \ +static inline void \ +nilfs_suinfo_update_set_##name(struct nilfs_suinfo_update *sup) \ +{ \ + sup->sup_flags |= 1UL << NILFS_SUINFO_UPDATE_##flag; \ +} \ +static inline void \ +nilfs_suinfo_update_clear_##name(struct nilfs_suinfo_update *sup) \ +{ \ + sup->sup_flags &= ~(1UL << NILFS_SUINFO_UPDATE_##flag); \ +} \ +static inline int \ +nilfs_suinfo_update_##name(const struct nilfs_suinfo_update *sup) \ +{ \ + return !!(sup->sup_flags & (1UL << NILFS_SUINFO_UPDATE_##flag));\ +} + +NILFS_SUINFO_UPDATE_FNS(LASTMOD, lastmod) +NILFS_SUINFO_UPDATE_FNS(NBLOCKS, nblocks) +NILFS_SUINFO_UPDATE_FNS(FLAGS, flags) + +enum { + NILFS_CHECKPOINT, + NILFS_SNAPSHOT, +}; + +/** + * struct nilfs_cpmode - change checkpoint mode structure + * @cm_cno: checkpoint number + * @cm_mode: mode of checkpoint + * @cm_pad: padding + */ +struct nilfs_cpmode { + __u64 cm_cno; + __u32 cm_mode; + __u32 cm_pad; +}; + +/** + * struct nilfs_argv - argument vector + * @v_base: pointer on data array from userspace + * @v_nmembs: number of members in data array + * @v_size: size of data array in bytes + * @v_flags: flags + * @v_index: start number of target data items + */ +struct nilfs_argv { + __u64 v_base; + __u32 v_nmembs; /* number of members */ + __u16 v_size; /* size of members */ + __u16 v_flags; + __u64 v_index; +}; + +/** + * struct nilfs_period - period of checkpoint numbers + * @p_start: start checkpoint number (inclusive) + * @p_end: end checkpoint number (exclusive) + */ +struct nilfs_period { + __u64 p_start; + __u64 p_end; +}; + +/** + * struct nilfs_cpstat - checkpoint statistics + * @cs_cno: checkpoint number + * @cs_ncps: number of checkpoints + * @cs_nsss: number of snapshots + */ +struct nilfs_cpstat { + __u64 cs_cno; + __u64 cs_ncps; + __u64 cs_nsss; +}; + +/** + * struct nilfs_sustat - segment usage statistics + * @ss_nsegs: number of segments + * @ss_ncleansegs: number of clean segments + * @ss_ndirtysegs: number of dirty segments + * @ss_ctime: creation time of the last segment + * @ss_nongc_ctime: creation time of the last segment not for GC + * @ss_prot_seq: least sequence number of segments which must not be reclaimed + */ +struct nilfs_sustat { + __u64 ss_nsegs; + __u64 ss_ncleansegs; + __u64 ss_ndirtysegs; + __u64 ss_ctime; + __u64 ss_nongc_ctime; + __u64 ss_prot_seq; +}; + +/** + * struct nilfs_vinfo - virtual block number information + * @vi_vblocknr: virtual block number + * @vi_start: start checkpoint number (inclusive) + * @vi_end: end checkpoint number (exclusive) + * @vi_blocknr: disk block number + */ +struct nilfs_vinfo { + __u64 vi_vblocknr; + __u64 vi_start; + __u64 vi_end; + __u64 vi_blocknr; +}; + +/** + * struct nilfs_vdesc - descriptor of virtual block number + * @vd_ino: inode number + * @vd_cno: checkpoint number + * @vd_vblocknr: virtual block number + * @vd_period: period of checkpoint numbers + * @vd_blocknr: disk block number + * @vd_offset: logical block offset inside a file + * @vd_flags: flags (data or node block) + * @vd_pad: padding + */ +struct nilfs_vdesc { + __u64 vd_ino; + __u64 vd_cno; + __u64 vd_vblocknr; + struct nilfs_period vd_period; + __u64 vd_blocknr; + __u64 vd_offset; + __u32 vd_flags; + __u32 vd_pad; +}; + +/** + * struct nilfs_bdesc - descriptor of disk block number + * @bd_ino: inode number + * @bd_oblocknr: disk block address (for skipping dead blocks) + * @bd_blocknr: disk block address + * @bd_offset: logical block offset inside a file + * @bd_level: level in the b-tree organization + * @bd_pad: padding + */ +struct nilfs_bdesc { + __u64 bd_ino; + __u64 bd_oblocknr; + __u64 bd_blocknr; + __u64 bd_offset; + __u32 bd_level; + __u32 bd_pad; +}; + +#define NILFS_IOCTL_IDENT 'n' + +#define NILFS_IOCTL_CHANGE_CPMODE \ + _IOW(NILFS_IOCTL_IDENT, 0x80, struct nilfs_cpmode) +#define NILFS_IOCTL_DELETE_CHECKPOINT \ + _IOW(NILFS_IOCTL_IDENT, 0x81, __u64) +#define NILFS_IOCTL_GET_CPINFO \ + _IOR(NILFS_IOCTL_IDENT, 0x82, struct nilfs_argv) +#define NILFS_IOCTL_GET_CPSTAT \ + _IOR(NILFS_IOCTL_IDENT, 0x83, struct nilfs_cpstat) +#define NILFS_IOCTL_GET_SUINFO \ + _IOR(NILFS_IOCTL_IDENT, 0x84, struct nilfs_argv) +#define NILFS_IOCTL_GET_SUSTAT \ + _IOR(NILFS_IOCTL_IDENT, 0x85, struct nilfs_sustat) +#define NILFS_IOCTL_GET_VINFO \ + _IOWR(NILFS_IOCTL_IDENT, 0x86, struct nilfs_argv) +#define NILFS_IOCTL_GET_BDESCS \ + _IOWR(NILFS_IOCTL_IDENT, 0x87, struct nilfs_argv) +#define NILFS_IOCTL_CLEAN_SEGMENTS \ + _IOW(NILFS_IOCTL_IDENT, 0x88, struct nilfs_argv[5]) +#define NILFS_IOCTL_SYNC \ + _IOR(NILFS_IOCTL_IDENT, 0x8A, __u64) +#define NILFS_IOCTL_RESIZE \ + _IOW(NILFS_IOCTL_IDENT, 0x8B, __u64) +#define NILFS_IOCTL_SET_ALLOC_RANGE \ + _IOW(NILFS_IOCTL_IDENT, 0x8C, __u64[2]) +#define NILFS_IOCTL_SET_SUINFO \ + _IOW(NILFS_IOCTL_IDENT, 0x8D, struct nilfs_argv) + +#endif /* _LINUX_NILFS2_API_H */ diff --git a/include/uapi/linux/nilfs2_ondisk.h b/include/uapi/linux/nilfs2_ondisk.h new file mode 100644 index 000000000000..2a8a3addb675 --- /dev/null +++ b/include/uapi/linux/nilfs2_ondisk.h @@ -0,0 +1,650 @@ +/* + * nilfs2_ondisk.h - NILFS2 on-disk structures + * + * Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + */ +/* + * linux/include/linux/ext2_fs.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _LINUX_NILFS2_ONDISK_H +#define _LINUX_NILFS2_ONDISK_H + +#include +#include + + +#define NILFS_INODE_BMAP_SIZE 7 + +/** + * struct nilfs_inode - structure of an inode on disk + * @i_blocks: blocks count + * @i_size: size in bytes + * @i_ctime: creation time (seconds) + * @i_mtime: modification time (seconds) + * @i_ctime_nsec: creation time (nano seconds) + * @i_mtime_nsec: modification time (nano seconds) + * @i_uid: user id + * @i_gid: group id + * @i_mode: file mode + * @i_links_count: links count + * @i_flags: file flags + * @i_bmap: block mapping + * @i_xattr: extended attributes + * @i_generation: file generation (for NFS) + * @i_pad: padding + */ +struct nilfs_inode { + __le64 i_blocks; + __le64 i_size; + __le64 i_ctime; + __le64 i_mtime; + __le32 i_ctime_nsec; + __le32 i_mtime_nsec; + __le32 i_uid; + __le32 i_gid; + __le16 i_mode; + __le16 i_links_count; + __le32 i_flags; + __le64 i_bmap[NILFS_INODE_BMAP_SIZE]; +#define i_device_code i_bmap[0] + __le64 i_xattr; + __le32 i_generation; + __le32 i_pad; +}; + +#define NILFS_MIN_INODE_SIZE 128 + +/** + * struct nilfs_super_root - structure of super root + * @sr_sum: check sum + * @sr_bytes: byte count of the structure + * @sr_flags: flags (reserved) + * @sr_nongc_ctime: write time of the last segment not for cleaner operation + * @sr_dat: DAT file inode + * @sr_cpfile: checkpoint file inode + * @sr_sufile: segment usage file inode + */ +struct nilfs_super_root { + __le32 sr_sum; + __le16 sr_bytes; + __le16 sr_flags; + __le64 sr_nongc_ctime; + struct nilfs_inode sr_dat; + struct nilfs_inode sr_cpfile; + struct nilfs_inode sr_sufile; +}; + +#define NILFS_SR_MDT_OFFSET(inode_size, i) \ + ((unsigned long)&((struct nilfs_super_root *)0)->sr_dat + \ + (inode_size) * (i)) +#define NILFS_SR_DAT_OFFSET(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 0) +#define NILFS_SR_CPFILE_OFFSET(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 1) +#define NILFS_SR_SUFILE_OFFSET(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 2) +#define NILFS_SR_BYTES(inode_size) NILFS_SR_MDT_OFFSET(inode_size, 3) + +/* + * Maximal mount counts + */ +#define NILFS_DFL_MAX_MNT_COUNT 50 /* 50 mounts */ + +/* + * File system states (sbp->s_state, nilfs->ns_mount_state) + */ +#define NILFS_VALID_FS 0x0001 /* Unmounted cleanly */ +#define NILFS_ERROR_FS 0x0002 /* Errors detected */ +#define NILFS_RESIZE_FS 0x0004 /* Resize required */ + +/* + * Mount flags (sbi->s_mount_opt) + */ +#define NILFS_MOUNT_ERROR_MODE 0x0070 /* Error mode mask */ +#define NILFS_MOUNT_ERRORS_CONT 0x0010 /* Continue on errors */ +#define NILFS_MOUNT_ERRORS_RO 0x0020 /* Remount fs ro on errors */ +#define NILFS_MOUNT_ERRORS_PANIC 0x0040 /* Panic on errors */ +#define NILFS_MOUNT_BARRIER 0x1000 /* Use block barriers */ +#define NILFS_MOUNT_STRICT_ORDER 0x2000 /* + * Apply strict in-order + * semantics also for data + */ +#define NILFS_MOUNT_NORECOVERY 0x4000 /* + * Disable write access during + * mount-time recovery + */ +#define NILFS_MOUNT_DISCARD 0x8000 /* Issue DISCARD requests */ + + +/** + * struct nilfs_super_block - structure of super block on disk + */ +struct nilfs_super_block { +/*00*/ __le32 s_rev_level; /* Revision level */ + __le16 s_minor_rev_level; /* minor revision level */ + __le16 s_magic; /* Magic signature */ + + __le16 s_bytes; /* + * Bytes count of CRC calculation + * for this structure. s_reserved + * is excluded. + */ + __le16 s_flags; /* flags */ + __le32 s_crc_seed; /* Seed value of CRC calculation */ +/*10*/ __le32 s_sum; /* Check sum of super block */ + + __le32 s_log_block_size; /* + * Block size represented as follows + * blocksize = + * 1 << (s_log_block_size + 10) + */ + __le64 s_nsegments; /* Number of segments in filesystem */ +/*20*/ __le64 s_dev_size; /* block device size in bytes */ + __le64 s_first_data_block; /* 1st seg disk block number */ +/*30*/ __le32 s_blocks_per_segment; /* number of blocks per full segment */ + __le32 s_r_segments_percentage; /* Reserved segments percentage */ + + __le64 s_last_cno; /* Last checkpoint number */ +/*40*/ __le64 s_last_pseg; /* disk block addr pseg written last */ + __le64 s_last_seq; /* seq. number of seg written last */ +/*50*/ __le64 s_free_blocks_count; /* Free blocks count */ + + __le64 s_ctime; /* + * Creation time (execution time of + * newfs) + */ +/*60*/ __le64 s_mtime; /* Mount time */ + __le64 s_wtime; /* Write time */ +/*70*/ __le16 s_mnt_count; /* Mount count */ + __le16 s_max_mnt_count; /* Maximal mount count */ + __le16 s_state; /* File system state */ + __le16 s_errors; /* Behaviour when detecting errors */ + __le64 s_lastcheck; /* time of last check */ + +/*80*/ __le32 s_checkinterval; /* max. time between checks */ + __le32 s_creator_os; /* OS */ + __le16 s_def_resuid; /* Default uid for reserved blocks */ + __le16 s_def_resgid; /* Default gid for reserved blocks */ + __le32 s_first_ino; /* First non-reserved inode */ + +/*90*/ __le16 s_inode_size; /* Size of an inode */ + __le16 s_dat_entry_size; /* Size of a dat entry */ + __le16 s_checkpoint_size; /* Size of a checkpoint */ + __le16 s_segment_usage_size; /* Size of a segment usage */ + +/*98*/ __u8 s_uuid[16]; /* 128-bit uuid for volume */ +/*A8*/ char s_volume_name[80]; /* volume name */ + +/*F8*/ __le32 s_c_interval; /* Commit interval of segment */ + __le32 s_c_block_max; /* + * Threshold of data amount for + * the segment construction + */ +/*100*/ __le64 s_feature_compat; /* Compatible feature set */ + __le64 s_feature_compat_ro; /* Read-only compatible feature set */ + __le64 s_feature_incompat; /* Incompatible feature set */ + __u32 s_reserved[186]; /* padding to the end of the block */ +}; + +/* + * Codes for operating systems + */ +#define NILFS_OS_LINUX 0 +/* Codes from 1 to 4 are reserved to keep compatibility with ext2 creator-OS */ + +/* + * Revision levels + */ +#define NILFS_CURRENT_REV 2 /* current major revision */ +#define NILFS_MINOR_REV 0 /* minor revision */ +#define NILFS_MIN_SUPP_REV 2 /* minimum supported revision */ + +/* + * Feature set definitions + * + * If there is a bit set in the incompatible feature set that the kernel + * doesn't know about, it should refuse to mount the filesystem. + */ +#define NILFS_FEATURE_COMPAT_RO_BLOCK_COUNT 0x00000001ULL + +#define NILFS_FEATURE_COMPAT_SUPP 0ULL +#define NILFS_FEATURE_COMPAT_RO_SUPP NILFS_FEATURE_COMPAT_RO_BLOCK_COUNT +#define NILFS_FEATURE_INCOMPAT_SUPP 0ULL + +/* + * Bytes count of super_block for CRC-calculation + */ +#define NILFS_SB_BYTES \ + ((long)&((struct nilfs_super_block *)0)->s_reserved) + +/* + * Special inode number + */ +#define NILFS_ROOT_INO 2 /* Root file inode */ +#define NILFS_DAT_INO 3 /* DAT file */ +#define NILFS_CPFILE_INO 4 /* checkpoint file */ +#define NILFS_SUFILE_INO 5 /* segment usage file */ +#define NILFS_IFILE_INO 6 /* ifile */ +#define NILFS_ATIME_INO 7 /* Atime file (reserved) */ +#define NILFS_XATTR_INO 8 /* Xattribute file (reserved) */ +#define NILFS_SKETCH_INO 10 /* Sketch file */ +#define NILFS_USER_INO 11 /* Fisrt user's file inode number */ + +#define NILFS_SB_OFFSET_BYTES 1024 /* byte offset of nilfs superblock */ + +#define NILFS_SEG_MIN_BLOCKS 16 /* + * Minimum number of blocks in + * a full segment + */ +#define NILFS_PSEG_MIN_BLOCKS 2 /* + * Minimum number of blocks in + * a partial segment + */ +#define NILFS_MIN_NRSVSEGS 8 /* + * Minimum number of reserved + * segments + */ + +/* + * We call DAT, cpfile, and sufile root metadata files. Inodes of + * these files are written in super root block instead of ifile, and + * garbage collector doesn't keep any past versions of these files. + */ +#define NILFS_ROOT_METADATA_FILE(ino) \ + ((ino) >= NILFS_DAT_INO && (ino) <= NILFS_SUFILE_INO) + +/* + * bytes offset of secondary super block + */ +#define NILFS_SB2_OFFSET_BYTES(devsize) ((((devsize) >> 12) - 1) << 12) + +/* + * Maximal count of links to a file + */ +#define NILFS_LINK_MAX 32000 + +/* + * Structure of a directory entry + * (Same as ext2) + */ + +#define NILFS_NAME_LEN 255 + +/* + * Block size limitations + */ +#define NILFS_MIN_BLOCK_SIZE 1024 +#define NILFS_MAX_BLOCK_SIZE 65536 + +/* + * The new version of the directory entry. Since V0 structures are + * stored in intel byte order, and the name_len field could never be + * bigger than 255 chars, it's safe to reclaim the extra byte for the + * file_type field. + */ +struct nilfs_dir_entry { + __le64 inode; /* Inode number */ + __le16 rec_len; /* Directory entry length */ + __u8 name_len; /* Name length */ + __u8 file_type; /* Dir entry type (file, dir, etc) */ + char name[NILFS_NAME_LEN]; /* File name */ + char pad; +}; + +/* + * NILFS directory file types. Only the low 3 bits are used. The + * other bits are reserved for now. + */ +enum { + NILFS_FT_UNKNOWN, + NILFS_FT_REG_FILE, + NILFS_FT_DIR, + NILFS_FT_CHRDEV, + NILFS_FT_BLKDEV, + NILFS_FT_FIFO, + NILFS_FT_SOCK, + NILFS_FT_SYMLINK, + NILFS_FT_MAX +}; + +/* + * NILFS_DIR_PAD defines the directory entries boundaries + * + * NOTE: It must be a multiple of 8 + */ +#define NILFS_DIR_PAD 8 +#define NILFS_DIR_ROUND (NILFS_DIR_PAD - 1) +#define NILFS_DIR_REC_LEN(name_len) (((name_len) + 12 + NILFS_DIR_ROUND) & \ + ~NILFS_DIR_ROUND) +#define NILFS_MAX_REC_LEN ((1 << 16) - 1) + +/** + * struct nilfs_finfo - file information + * @fi_ino: inode number + * @fi_cno: checkpoint number + * @fi_nblocks: number of blocks (including intermediate blocks) + * @fi_ndatablk: number of file data blocks + */ +struct nilfs_finfo { + __le64 fi_ino; + __le64 fi_cno; + __le32 fi_nblocks; + __le32 fi_ndatablk; +}; + +/** + * struct nilfs_binfo_v - information on a data block (except DAT) + * @bi_vblocknr: virtual block number + * @bi_blkoff: block offset + */ +struct nilfs_binfo_v { + __le64 bi_vblocknr; + __le64 bi_blkoff; +}; + +/** + * struct nilfs_binfo_dat - information on a DAT node block + * @bi_blkoff: block offset + * @bi_level: level + * @bi_pad: padding + */ +struct nilfs_binfo_dat { + __le64 bi_blkoff; + __u8 bi_level; + __u8 bi_pad[7]; +}; + +/** + * union nilfs_binfo: block information + * @bi_v: nilfs_binfo_v structure + * @bi_dat: nilfs_binfo_dat structure + */ +union nilfs_binfo { + struct nilfs_binfo_v bi_v; + struct nilfs_binfo_dat bi_dat; +}; + +/** + * struct nilfs_segment_summary - segment summary header + * @ss_datasum: checksum of data + * @ss_sumsum: checksum of segment summary + * @ss_magic: magic number + * @ss_bytes: size of this structure in bytes + * @ss_flags: flags + * @ss_seq: sequence number + * @ss_create: creation timestamp + * @ss_next: next segment + * @ss_nblocks: number of blocks + * @ss_nfinfo: number of finfo structures + * @ss_sumbytes: total size of segment summary in bytes + * @ss_pad: padding + * @ss_cno: checkpoint number + */ +struct nilfs_segment_summary { + __le32 ss_datasum; + __le32 ss_sumsum; + __le32 ss_magic; + __le16 ss_bytes; + __le16 ss_flags; + __le64 ss_seq; + __le64 ss_create; + __le64 ss_next; + __le32 ss_nblocks; + __le32 ss_nfinfo; + __le32 ss_sumbytes; + __le32 ss_pad; + __le64 ss_cno; + /* array of finfo structures */ +}; + +#define NILFS_SEGSUM_MAGIC 0x1eaffa11 /* segment summary magic number */ + +/* + * Segment summary flags + */ +#define NILFS_SS_LOGBGN 0x0001 /* begins a logical segment */ +#define NILFS_SS_LOGEND 0x0002 /* ends a logical segment */ +#define NILFS_SS_SR 0x0004 /* has super root */ +#define NILFS_SS_SYNDT 0x0008 /* includes data only updates */ +#define NILFS_SS_GC 0x0010 /* segment written for cleaner operation */ + +/** + * struct nilfs_btree_node - header of B-tree node block + * @bn_flags: flags + * @bn_level: level + * @bn_nchildren: number of children + * @bn_pad: padding + */ +struct nilfs_btree_node { + __u8 bn_flags; + __u8 bn_level; + __le16 bn_nchildren; + __le32 bn_pad; +}; + +/* flags */ +#define NILFS_BTREE_NODE_ROOT 0x01 + +/* level */ +#define NILFS_BTREE_LEVEL_DATA 0 +#define NILFS_BTREE_LEVEL_NODE_MIN (NILFS_BTREE_LEVEL_DATA + 1) +#define NILFS_BTREE_LEVEL_MAX 14 /* Max level (exclusive) */ + +/** + * struct nilfs_direct_node - header of built-in bmap array + * @dn_flags: flags + * @dn_pad: padding + */ +struct nilfs_direct_node { + __u8 dn_flags; + __u8 pad[7]; +}; + +/** + * struct nilfs_palloc_group_desc - block group descriptor + * @pg_nfrees: number of free entries in block group + */ +struct nilfs_palloc_group_desc { + __le32 pg_nfrees; +}; + +/** + * struct nilfs_dat_entry - disk address translation entry + * @de_blocknr: block number + * @de_start: start checkpoint number + * @de_end: end checkpoint number + * @de_rsv: reserved for future use + */ +struct nilfs_dat_entry { + __le64 de_blocknr; + __le64 de_start; + __le64 de_end; + __le64 de_rsv; +}; + +#define NILFS_MIN_DAT_ENTRY_SIZE 32 + +/** + * struct nilfs_snapshot_list - snapshot list + * @ssl_next: next checkpoint number on snapshot list + * @ssl_prev: previous checkpoint number on snapshot list + */ +struct nilfs_snapshot_list { + __le64 ssl_next; + __le64 ssl_prev; +}; + +/** + * struct nilfs_checkpoint - checkpoint structure + * @cp_flags: flags + * @cp_checkpoints_count: checkpoints count in a block + * @cp_snapshot_list: snapshot list + * @cp_cno: checkpoint number + * @cp_create: creation timestamp + * @cp_nblk_inc: number of blocks incremented by this checkpoint + * @cp_inodes_count: inodes count + * @cp_blocks_count: blocks count + * @cp_ifile_inode: inode of ifile + */ +struct nilfs_checkpoint { + __le32 cp_flags; + __le32 cp_checkpoints_count; + struct nilfs_snapshot_list cp_snapshot_list; + __le64 cp_cno; + __le64 cp_create; + __le64 cp_nblk_inc; + __le64 cp_inodes_count; + __le64 cp_blocks_count; + + /* + * Do not change the byte offset of ifile inode. + * To keep the compatibility of the disk format, + * additional fields should be added behind cp_ifile_inode. + */ + struct nilfs_inode cp_ifile_inode; +}; + +#define NILFS_MIN_CHECKPOINT_SIZE (64 + NILFS_MIN_INODE_SIZE) + +/* checkpoint flags */ +enum { + NILFS_CHECKPOINT_SNAPSHOT, + NILFS_CHECKPOINT_INVALID, + NILFS_CHECKPOINT_SKETCH, + NILFS_CHECKPOINT_MINOR, +}; + +#define NILFS_CHECKPOINT_FNS(flag, name) \ +static inline void \ +nilfs_checkpoint_set_##name(struct nilfs_checkpoint *cp) \ +{ \ + cp->cp_flags = cpu_to_le32(le32_to_cpu(cp->cp_flags) | \ + (1UL << NILFS_CHECKPOINT_##flag)); \ +} \ +static inline void \ +nilfs_checkpoint_clear_##name(struct nilfs_checkpoint *cp) \ +{ \ + cp->cp_flags = cpu_to_le32(le32_to_cpu(cp->cp_flags) & \ + ~(1UL << NILFS_CHECKPOINT_##flag)); \ +} \ +static inline int \ +nilfs_checkpoint_##name(const struct nilfs_checkpoint *cp) \ +{ \ + return !!(le32_to_cpu(cp->cp_flags) & \ + (1UL << NILFS_CHECKPOINT_##flag)); \ +} + +NILFS_CHECKPOINT_FNS(SNAPSHOT, snapshot) +NILFS_CHECKPOINT_FNS(INVALID, invalid) +NILFS_CHECKPOINT_FNS(MINOR, minor) + +/** + * struct nilfs_cpfile_header - checkpoint file header + * @ch_ncheckpoints: number of checkpoints + * @ch_nsnapshots: number of snapshots + * @ch_snapshot_list: snapshot list + */ +struct nilfs_cpfile_header { + __le64 ch_ncheckpoints; + __le64 ch_nsnapshots; + struct nilfs_snapshot_list ch_snapshot_list; +}; + +#define NILFS_CPFILE_FIRST_CHECKPOINT_OFFSET \ + ((sizeof(struct nilfs_cpfile_header) + \ + sizeof(struct nilfs_checkpoint) - 1) / \ + sizeof(struct nilfs_checkpoint)) + +/** + * struct nilfs_segment_usage - segment usage + * @su_lastmod: last modified timestamp + * @su_nblocks: number of blocks in segment + * @su_flags: flags + */ +struct nilfs_segment_usage { + __le64 su_lastmod; + __le32 su_nblocks; + __le32 su_flags; +}; + +#define NILFS_MIN_SEGMENT_USAGE_SIZE 16 + +/* segment usage flag */ +enum { + NILFS_SEGMENT_USAGE_ACTIVE, + NILFS_SEGMENT_USAGE_DIRTY, + NILFS_SEGMENT_USAGE_ERROR, +}; + +#define NILFS_SEGMENT_USAGE_FNS(flag, name) \ +static inline void \ +nilfs_segment_usage_set_##name(struct nilfs_segment_usage *su) \ +{ \ + su->su_flags = cpu_to_le32(le32_to_cpu(su->su_flags) | \ + (1UL << NILFS_SEGMENT_USAGE_##flag));\ +} \ +static inline void \ +nilfs_segment_usage_clear_##name(struct nilfs_segment_usage *su) \ +{ \ + su->su_flags = \ + cpu_to_le32(le32_to_cpu(su->su_flags) & \ + ~(1UL << NILFS_SEGMENT_USAGE_##flag)); \ +} \ +static inline int \ +nilfs_segment_usage_##name(const struct nilfs_segment_usage *su) \ +{ \ + return !!(le32_to_cpu(su->su_flags) & \ + (1UL << NILFS_SEGMENT_USAGE_##flag)); \ +} + +NILFS_SEGMENT_USAGE_FNS(ACTIVE, active) +NILFS_SEGMENT_USAGE_FNS(DIRTY, dirty) +NILFS_SEGMENT_USAGE_FNS(ERROR, error) + +static inline void +nilfs_segment_usage_set_clean(struct nilfs_segment_usage *su) +{ + su->su_lastmod = cpu_to_le64(0); + su->su_nblocks = cpu_to_le32(0); + su->su_flags = cpu_to_le32(0); +} + +static inline int +nilfs_segment_usage_clean(const struct nilfs_segment_usage *su) +{ + return !le32_to_cpu(su->su_flags); +} + +/** + * struct nilfs_sufile_header - segment usage file header + * @sh_ncleansegs: number of clean segments + * @sh_ndirtysegs: number of dirty segments + * @sh_last_alloc: last allocated segment number + */ +struct nilfs_sufile_header { + __le64 sh_ncleansegs; + __le64 sh_ndirtysegs; + __le64 sh_last_alloc; + /* ... */ +}; + +#define NILFS_SUFILE_FIRST_SEGMENT_USAGE_OFFSET \ + ((sizeof(struct nilfs_sufile_header) + \ + sizeof(struct nilfs_segment_usage) - 1) / \ + sizeof(struct nilfs_segment_usage)) + +#endif /* _LINUX_NILFS2_ONDISK_H */ -- cgit v1.2.3 From b6e8d4aa1110306378af0f3472a6b85a1f039a16 Mon Sep 17 00:00:00 2001 From: Alexandre Bounine Date: Tue, 2 Aug 2016 14:06:25 -0700 Subject: rapidio: add RapidIO channelized messaging driver Add channelized messaging driver to support native RapidIO messaging exchange between multiple senders/recipients on devices that use kernel RapidIO subsystem services. This device driver is the result of collaboration within the RapidIO.org Software Task Group (STG) between Texas Instruments, Prodrive Technologies, Nokia Networks, BAE and IDT. Additional input was received from other members of RapidIO.org. The objective was to create a character mode driver interface which exposes messaging capabilities of RapidIO endpoint devices (mports) directly to applications, in a manner that allows the numerous and varied RapidIO implementations to interoperate. This char mode device driver allows user-space applications to setup messaging communication channels using single shared RapidIO messaging mailbox. By default this driver uses RapidIO MBOX_1 (MBOX_0 is reserved for use by RIONET Ethernet emulation driver). [weiyj.lk@gmail.com: rapidio/rio_cm: fix return value check in riocm_init()] Link: http://lkml.kernel.org/r/1469198221-21970-1-git-send-email-alexandre.bounine@idt.com Link: http://lkml.kernel.org/r/1468952862-18056-1-git-send-email-alexandre.bounine@idt.com Signed-off-by: Alexandre Bounine Tested-by: Barry Wood Cc: Matt Porter Cc: Aurelien Jacquiot Cc: Andre van Herk Cc: Barry Wood Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/rapidio/rio_cm.txt | 119 ++ drivers/rapidio/Kconfig | 9 + drivers/rapidio/Makefile | 1 + drivers/rapidio/rio_cm.c | 2366 ++++++++++++++++++++++++++++++++++++++ include/uapi/linux/Kbuild | 1 + include/uapi/linux/rio_cm_cdev.h | 78 ++ 6 files changed, 2574 insertions(+) create mode 100644 Documentation/rapidio/rio_cm.txt create mode 100644 drivers/rapidio/rio_cm.c create mode 100644 include/uapi/linux/rio_cm_cdev.h (limited to 'include/uapi') diff --git a/Documentation/rapidio/rio_cm.txt b/Documentation/rapidio/rio_cm.txt new file mode 100644 index 000000000000..27aa401f1126 --- /dev/null +++ b/Documentation/rapidio/rio_cm.txt @@ -0,0 +1,119 @@ +RapidIO subsystem Channelized Messaging character device driver (rio_cm.c) +========================================================================== + +Version History: +---------------- + 1.0.0 - Initial driver release. + +========================================================================== + +I. Overview + +This device driver is the result of collaboration within the RapidIO.org +Software Task Group (STG) between Texas Instruments, Prodrive Technologies, +Nokia Networks, BAE and IDT. Additional input was received from other members +of RapidIO.org. + +The objective was to create a character mode driver interface which exposes +messaging capabilities of RapidIO endpoint devices (mports) directly +to applications, in a manner that allows the numerous and varied RapidIO +implementations to interoperate. + +This driver (RIO_CM) provides to user-space applications shared access to +RapidIO mailbox messaging resources. + +RapidIO specification (Part 2) defines that endpoint devices may have up to four +messaging mailboxes in case of multi-packet message (up to 4KB) and +up to 64 mailboxes if single-packet messages (up to 256 B) are used. In addition +to protocol definition limitations, a particular hardware implementation can +have reduced number of messaging mailboxes. RapidIO aware applications must +therefore share the messaging resources of a RapidIO endpoint. + +Main purpose of this device driver is to provide RapidIO mailbox messaging +capability to large number of user-space processes by introducing socket-like +operations using a single messaging mailbox. This allows applications to +use the limited RapidIO messaging hardware resources efficiently. + +Most of device driver's operations are supported through 'ioctl' system calls. + +When loaded this device driver creates a single file system node named rio_cm +in /dev directory common for all registered RapidIO mport devices. + +Following ioctl commands are available to user-space applications: + +- RIO_CM_MPORT_GET_LIST : Returns to caller list of local mport devices that + support messaging operations (number of entries up to RIO_MAX_MPORTS). + Each list entry is combination of mport's index in the system and RapidIO + destination ID assigned to the port. +- RIO_CM_EP_GET_LIST_SIZE : Returns number of messaging capable remote endpoints + in a RapidIO network associated with the specified mport device. +- RIO_CM_EP_GET_LIST : Returns list of RapidIO destination IDs for messaging + capable remote endpoints (peers) available in a RapidIO network associated + with the specified mport device. +- RIO_CM_CHAN_CREATE : Creates RapidIO message exchange channel data structure + with channel ID assigned automatically or as requested by a caller. +- RIO_CM_CHAN_BIND : Binds the specified channel data structure to the specified + mport device. +- RIO_CM_CHAN_LISTEN : Enables listening for connection requests on the specified + channel. +- RIO_CM_CHAN_ACCEPT : Accepts a connection request from peer on the specified + channel. If wait timeout for this request is specified by a caller it is + a blocking call. If timeout set to 0 this is non-blocking call - ioctl + handler checks for a pending connection request and if one is not available + exits with -EGAIN error status immediately. +- RIO_CM_CHAN_CONNECT : Sends a connection request to a remote peer/channel. +- RIO_CM_CHAN_SEND : Sends a data message through the specified channel. + The handler for this request assumes that message buffer specified by + a caller includes the reserved space for a packet header required by + this driver. +- RIO_CM_CHAN_RECEIVE : Receives a data message through a connected channel. + If the channel does not have an incoming message ready to return this ioctl + handler will wait for new message until timeout specified by a caller + expires. If timeout value is set to 0, ioctl handler uses a default value + defined by MAX_SCHEDULE_TIMEOUT. +- RIO_CM_CHAN_CLOSE : Closes a specified channel and frees associated buffers. + If the specified channel is in the CONNECTED state, sends close notification + to the remote peer. + +The ioctl command codes and corresponding data structures intended for use by +user-space applications are defined in 'include/uapi/linux/rio_cm_cdev.h'. + +II. Hardware Compatibility + +This device driver uses standard interfaces defined by kernel RapidIO subsystem +and therefore it can be used with any mport device driver registered by RapidIO +subsystem with limitations set by available mport HW implementation of messaging +mailboxes. + +III. Module parameters + +- 'dbg_level' - This parameter allows to control amount of debug information + generated by this device driver. This parameter is formed by set of + bit masks that correspond to the specific functional block. + For mask definitions see 'drivers/rapidio/devices/rio_cm.c' + This parameter can be changed dynamically. + Use CONFIG_RAPIDIO_DEBUG=y to enable debug output at the top level. + +- 'cmbox' - Number of RapidIO mailbox to use (default value is 1). + This parameter allows to set messaging mailbox number that will be used + within entire RapidIO network. It can be used when default mailbox is + used by other device drivers or is not supported by some nodes in the + RapidIO network. + +- 'chstart' - Start channel number for dynamic assignment. Default value - 256. + Allows to exclude channel numbers below this parameter from dynamic + allocation to avoid conflicts with software components that use + reserved predefined channel numbers. + +IV. Known problems + + None. + +V. User-space Applications and API Library + +Messaging API library and applications that use this device driver are available +from RapidIO.org. + +VI. TODO List + +- Add support for system notification messages (reserved channel 0). diff --git a/drivers/rapidio/Kconfig b/drivers/rapidio/Kconfig index b5a10d3c92c7..d6d2f20c4597 100644 --- a/drivers/rapidio/Kconfig +++ b/drivers/rapidio/Kconfig @@ -67,6 +67,15 @@ config RAPIDIO_ENUM_BASIC endchoice +config RAPIDIO_CHMAN + tristate "RapidIO Channelized Messaging driver" + depends on RAPIDIO + help + This option includes RapidIO channelized messaging driver which + provides socket-like interface to allow sharing of single RapidIO + messaging mailbox between multiple user-space applications. + See "Documentation/rapidio/rio_cm.txt" for driver description. + config RAPIDIO_MPORT_CDEV tristate "RapidIO /dev mport device driver" depends on RAPIDIO diff --git a/drivers/rapidio/Makefile b/drivers/rapidio/Makefile index 6271ada6993f..74dcea45ad49 100644 --- a/drivers/rapidio/Makefile +++ b/drivers/rapidio/Makefile @@ -5,6 +5,7 @@ obj-$(CONFIG_RAPIDIO) += rapidio.o rapidio-y := rio.o rio-access.o rio-driver.o rio-sysfs.o obj-$(CONFIG_RAPIDIO_ENUM_BASIC) += rio-scan.o +obj-$(CONFIG_RAPIDIO_CHMAN) += rio_cm.o obj-$(CONFIG_RAPIDIO) += switches/ obj-$(CONFIG_RAPIDIO) += devices/ diff --git a/drivers/rapidio/rio_cm.c b/drivers/rapidio/rio_cm.c new file mode 100644 index 000000000000..cecc15a880de --- /dev/null +++ b/drivers/rapidio/rio_cm.c @@ -0,0 +1,2366 @@ +/* + * rio_cm - RapidIO Channelized Messaging Driver + * + * Copyright 2013-2016 Integrated Device Technology, Inc. + * Copyright (c) 2015, Prodrive Technologies + * Copyright (c) 2015, RapidIO Trade Association + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * THIS PROGRAM IS DISTRIBUTED IN THE HOPE THAT IT WILL BE USEFUL, + * BUT WITHOUT ANY WARRANTY; WITHOUT EVEN THE IMPLIED WARRANTY OF + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. SEE THE + * GNU GENERAL PUBLIC LICENSE FOR MORE DETAILS. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define DRV_NAME "rio_cm" +#define DRV_VERSION "1.0.0" +#define DRV_AUTHOR "Alexandre Bounine " +#define DRV_DESC "RapidIO Channelized Messaging Driver" +#define DEV_NAME "rio_cm" + +/* Debug output filtering masks */ +enum { + DBG_NONE = 0, + DBG_INIT = BIT(0), /* driver init */ + DBG_EXIT = BIT(1), /* driver exit */ + DBG_MPORT = BIT(2), /* mport add/remove */ + DBG_RDEV = BIT(3), /* RapidIO device add/remove */ + DBG_CHOP = BIT(4), /* channel operations */ + DBG_WAIT = BIT(5), /* waiting for events */ + DBG_TX = BIT(6), /* message TX */ + DBG_TX_EVENT = BIT(7), /* message TX event */ + DBG_RX_DATA = BIT(8), /* inbound data messages */ + DBG_RX_CMD = BIT(9), /* inbound REQ/ACK/NACK messages */ + DBG_ALL = ~0, +}; + +#ifdef DEBUG +#define riocm_debug(level, fmt, arg...) \ + do { \ + if (DBG_##level & dbg_level) \ + pr_debug(DRV_NAME ": %s " fmt "\n", \ + __func__, ##arg); \ + } while (0) +#else +#define riocm_debug(level, fmt, arg...) \ + no_printk(KERN_DEBUG pr_fmt(DRV_NAME fmt "\n"), ##arg) +#endif + +#define riocm_warn(fmt, arg...) \ + pr_warn(DRV_NAME ": %s WARNING " fmt "\n", __func__, ##arg) + +#define riocm_error(fmt, arg...) \ + pr_err(DRV_NAME ": %s ERROR " fmt "\n", __func__, ##arg) + + +static int cmbox = 1; +module_param(cmbox, int, S_IRUGO); +MODULE_PARM_DESC(cmbox, "RapidIO Mailbox number (default 1)"); + +static int chstart = 256; +module_param(chstart, int, S_IRUGO); +MODULE_PARM_DESC(chstart, + "Start channel number for dynamic allocation (default 256)"); + +#ifdef DEBUG +static u32 dbg_level = DBG_NONE; +module_param(dbg_level, uint, S_IWUSR | S_IRUGO); +MODULE_PARM_DESC(dbg_level, "Debugging output level (default 0 = none)"); +#endif + +MODULE_AUTHOR(DRV_AUTHOR); +MODULE_DESCRIPTION(DRV_DESC); +MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_VERSION); + +#define RIOCM_TX_RING_SIZE 128 +#define RIOCM_RX_RING_SIZE 128 +#define RIOCM_CONNECT_TO 3 /* connect response TO (in sec) */ + +#define RIOCM_MAX_CHNUM 0xffff /* Use full range of u16 field */ +#define RIOCM_CHNUM_AUTO 0 +#define RIOCM_MAX_EP_COUNT 0x10000 /* Max number of endpoints */ + +enum rio_cm_state { + RIO_CM_IDLE, + RIO_CM_CONNECT, + RIO_CM_CONNECTED, + RIO_CM_DISCONNECT, + RIO_CM_CHAN_BOUND, + RIO_CM_LISTEN, + RIO_CM_DESTROYING, +}; + +enum rio_cm_pkt_type { + RIO_CM_SYS = 0xaa, + RIO_CM_CHAN = 0x55, +}; + +enum rio_cm_chop { + CM_CONN_REQ, + CM_CONN_ACK, + CM_CONN_CLOSE, + CM_DATA_MSG, +}; + +struct rio_ch_base_bhdr { + u32 src_id; + u32 dst_id; +#define RIO_HDR_LETTER_MASK 0xffff0000 +#define RIO_HDR_MBOX_MASK 0x0000ffff + u8 src_mbox; + u8 dst_mbox; + u8 type; +} __attribute__((__packed__)); + +struct rio_ch_chan_hdr { + struct rio_ch_base_bhdr bhdr; + u8 ch_op; + u16 dst_ch; + u16 src_ch; + u16 msg_len; + u16 rsrvd; +} __attribute__((__packed__)); + +struct tx_req { + struct list_head node; + struct rio_dev *rdev; + void *buffer; + size_t len; +}; + +struct cm_dev { + struct list_head list; + struct rio_mport *mport; + void *rx_buf[RIOCM_RX_RING_SIZE]; + int rx_slots; + struct mutex rx_lock; + + void *tx_buf[RIOCM_TX_RING_SIZE]; + int tx_slot; + int tx_cnt; + int tx_ack_slot; + struct list_head tx_reqs; + spinlock_t tx_lock; + + struct list_head peers; + u32 npeers; + struct workqueue_struct *rx_wq; + struct work_struct rx_work; +}; + +struct chan_rx_ring { + void *buf[RIOCM_RX_RING_SIZE]; + int head; + int tail; + int count; + + /* Tracking RX buffers reported to upper level */ + void *inuse[RIOCM_RX_RING_SIZE]; + int inuse_cnt; +}; + +struct rio_channel { + u16 id; /* local channel ID */ + struct kref ref; /* channel refcount */ + struct file *filp; + struct cm_dev *cmdev; /* associated CM device object */ + struct rio_dev *rdev; /* remote RapidIO device */ + enum rio_cm_state state; + int error; + spinlock_t lock; + void *context; + u32 loc_destid; /* local destID */ + u32 rem_destid; /* remote destID */ + u16 rem_channel; /* remote channel ID */ + struct list_head accept_queue; + struct list_head ch_node; + struct completion comp; + struct completion comp_close; + struct chan_rx_ring rx_ring; +}; + +struct cm_peer { + struct list_head node; + struct rio_dev *rdev; +}; + +struct rio_cm_work { + struct work_struct work; + struct cm_dev *cm; + void *data; +}; + +struct conn_req { + struct list_head node; + u32 destid; /* requester destID */ + u16 chan; /* requester channel ID */ + struct cm_dev *cmdev; +}; + +/* + * A channel_dev structure represents a CM_CDEV + * @cdev Character device + * @dev Associated device object + */ +struct channel_dev { + struct cdev cdev; + struct device *dev; +}; + +static struct rio_channel *riocm_ch_alloc(u16 ch_num); +static void riocm_ch_free(struct kref *ref); +static int riocm_post_send(struct cm_dev *cm, struct rio_dev *rdev, + void *buffer, size_t len); +static int riocm_ch_close(struct rio_channel *ch); + +static DEFINE_SPINLOCK(idr_lock); +static DEFINE_IDR(ch_idr); + +static LIST_HEAD(cm_dev_list); +static DECLARE_RWSEM(rdev_sem); + +static struct class *dev_class; +static unsigned int dev_major; +static unsigned int dev_minor_base; +static dev_t dev_number; +static struct channel_dev riocm_cdev; + +#define is_msg_capable(src_ops, dst_ops) \ + ((src_ops & RIO_SRC_OPS_DATA_MSG) && \ + (dst_ops & RIO_DST_OPS_DATA_MSG)) +#define dev_cm_capable(dev) \ + is_msg_capable(dev->src_ops, dev->dst_ops) + +static int riocm_cmp(struct rio_channel *ch, enum rio_cm_state cmp) +{ + int ret; + + spin_lock_bh(&ch->lock); + ret = (ch->state == cmp); + spin_unlock_bh(&ch->lock); + return ret; +} + +static int riocm_cmp_exch(struct rio_channel *ch, + enum rio_cm_state cmp, enum rio_cm_state exch) +{ + int ret; + + spin_lock_bh(&ch->lock); + ret = (ch->state == cmp); + if (ret) + ch->state = exch; + spin_unlock_bh(&ch->lock); + return ret; +} + +static enum rio_cm_state riocm_exch(struct rio_channel *ch, + enum rio_cm_state exch) +{ + enum rio_cm_state old; + + spin_lock_bh(&ch->lock); + old = ch->state; + ch->state = exch; + spin_unlock_bh(&ch->lock); + return old; +} + +static struct rio_channel *riocm_get_channel(u16 nr) +{ + struct rio_channel *ch; + + spin_lock_bh(&idr_lock); + ch = idr_find(&ch_idr, nr); + if (ch) + kref_get(&ch->ref); + spin_unlock_bh(&idr_lock); + return ch; +} + +static void riocm_put_channel(struct rio_channel *ch) +{ + kref_put(&ch->ref, riocm_ch_free); +} + +static void *riocm_rx_get_msg(struct cm_dev *cm) +{ + void *msg; + int i; + + msg = rio_get_inb_message(cm->mport, cmbox); + if (msg) { + for (i = 0; i < RIOCM_RX_RING_SIZE; i++) { + if (cm->rx_buf[i] == msg) { + cm->rx_buf[i] = NULL; + cm->rx_slots++; + break; + } + } + + if (i == RIOCM_RX_RING_SIZE) + riocm_warn("no record for buffer 0x%p", msg); + } + + return msg; +} + +/* + * riocm_rx_fill - fills a ring of receive buffers for given cm device + * @cm: cm_dev object + * @nent: max number of entries to fill + * + * Returns: none + */ +static void riocm_rx_fill(struct cm_dev *cm, int nent) +{ + int i; + + if (cm->rx_slots == 0) + return; + + for (i = 0; i < RIOCM_RX_RING_SIZE && cm->rx_slots && nent; i++) { + if (cm->rx_buf[i] == NULL) { + cm->rx_buf[i] = kmalloc(RIO_MAX_MSG_SIZE, GFP_KERNEL); + if (cm->rx_buf[i] == NULL) + break; + rio_add_inb_buffer(cm->mport, cmbox, cm->rx_buf[i]); + cm->rx_slots--; + nent--; + } + } +} + +/* + * riocm_rx_free - frees all receive buffers associated with given cm device + * @cm: cm_dev object + * + * Returns: none + */ +static void riocm_rx_free(struct cm_dev *cm) +{ + int i; + + for (i = 0; i < RIOCM_RX_RING_SIZE; i++) { + if (cm->rx_buf[i] != NULL) { + kfree(cm->rx_buf[i]); + cm->rx_buf[i] = NULL; + } + } +} + +/* + * riocm_req_handler - connection request handler + * @cm: cm_dev object + * @req_data: pointer to the request packet + * + * Returns: 0 if success, or + * -EINVAL if channel is not in correct state, + * -ENODEV if cannot find a channel with specified ID, + * -ENOMEM if unable to allocate memory to store the request + */ +static int riocm_req_handler(struct cm_dev *cm, void *req_data) +{ + struct rio_channel *ch; + struct conn_req *req; + struct rio_ch_chan_hdr *hh = req_data; + u16 chnum; + + chnum = ntohs(hh->dst_ch); + + ch = riocm_get_channel(chnum); + + if (!ch) + return -ENODEV; + + if (ch->state != RIO_CM_LISTEN) { + riocm_debug(RX_CMD, "channel %d is not in listen state", chnum); + riocm_put_channel(ch); + return -EINVAL; + } + + req = kzalloc(sizeof(*req), GFP_KERNEL); + if (!req) { + riocm_put_channel(ch); + return -ENOMEM; + } + + req->destid = ntohl(hh->bhdr.src_id); + req->chan = ntohs(hh->src_ch); + req->cmdev = cm; + + spin_lock_bh(&ch->lock); + list_add_tail(&req->node, &ch->accept_queue); + spin_unlock_bh(&ch->lock); + complete(&ch->comp); + riocm_put_channel(ch); + + return 0; +} + +/* + * riocm_resp_handler - response to connection request handler + * @resp_data: pointer to the response packet + * + * Returns: 0 if success, or + * -EINVAL if channel is not in correct state, + * -ENODEV if cannot find a channel with specified ID, + */ +static int riocm_resp_handler(void *resp_data) +{ + struct rio_channel *ch; + struct rio_ch_chan_hdr *hh = resp_data; + u16 chnum; + + chnum = ntohs(hh->dst_ch); + ch = riocm_get_channel(chnum); + if (!ch) + return -ENODEV; + + if (ch->state != RIO_CM_CONNECT) { + riocm_put_channel(ch); + return -EINVAL; + } + + riocm_exch(ch, RIO_CM_CONNECTED); + ch->rem_channel = ntohs(hh->src_ch); + complete(&ch->comp); + riocm_put_channel(ch); + + return 0; +} + +/* + * riocm_close_handler - channel close request handler + * @req_data: pointer to the request packet + * + * Returns: 0 if success, or + * -ENODEV if cannot find a channel with specified ID, + * + error codes returned by riocm_ch_close. + */ +static int riocm_close_handler(void *data) +{ + struct rio_channel *ch; + struct rio_ch_chan_hdr *hh = data; + int ret; + + riocm_debug(RX_CMD, "for ch=%d", ntohs(hh->dst_ch)); + + spin_lock_bh(&idr_lock); + ch = idr_find(&ch_idr, ntohs(hh->dst_ch)); + if (!ch) { + spin_unlock_bh(&idr_lock); + return -ENODEV; + } + idr_remove(&ch_idr, ch->id); + spin_unlock_bh(&idr_lock); + + riocm_exch(ch, RIO_CM_DISCONNECT); + + ret = riocm_ch_close(ch); + if (ret) + riocm_debug(RX_CMD, "riocm_ch_close() returned %d", ret); + + return 0; +} + +/* + * rio_cm_handler - function that services request (non-data) packets + * @cm: cm_dev object + * @data: pointer to the packet + */ +static void rio_cm_handler(struct cm_dev *cm, void *data) +{ + struct rio_ch_chan_hdr *hdr; + + if (!rio_mport_is_running(cm->mport)) + goto out; + + hdr = data; + + riocm_debug(RX_CMD, "OP=%x for ch=%d from %d", + hdr->ch_op, ntohs(hdr->dst_ch), ntohs(hdr->src_ch)); + + switch (hdr->ch_op) { + case CM_CONN_REQ: + riocm_req_handler(cm, data); + break; + case CM_CONN_ACK: + riocm_resp_handler(data); + break; + case CM_CONN_CLOSE: + riocm_close_handler(data); + break; + default: + riocm_error("Invalid packet header"); + break; + } +out: + kfree(data); +} + +/* + * rio_rx_data_handler - received data packet handler + * @cm: cm_dev object + * @buf: data packet + * + * Returns: 0 if success, or + * -ENODEV if cannot find a channel with specified ID, + * -EIO if channel is not in CONNECTED state, + * -ENOMEM if channel RX queue is full (packet discarded) + */ +static int rio_rx_data_handler(struct cm_dev *cm, void *buf) +{ + struct rio_ch_chan_hdr *hdr; + struct rio_channel *ch; + + hdr = buf; + + riocm_debug(RX_DATA, "for ch=%d", ntohs(hdr->dst_ch)); + + ch = riocm_get_channel(ntohs(hdr->dst_ch)); + if (!ch) { + /* Discard data message for non-existing channel */ + kfree(buf); + return -ENODEV; + } + + /* Place pointer to the buffer into channel's RX queue */ + spin_lock(&ch->lock); + + if (ch->state != RIO_CM_CONNECTED) { + /* Channel is not ready to receive data, discard a packet */ + riocm_debug(RX_DATA, "ch=%d is in wrong state=%d", + ch->id, ch->state); + spin_unlock(&ch->lock); + kfree(buf); + riocm_put_channel(ch); + return -EIO; + } + + if (ch->rx_ring.count == RIOCM_RX_RING_SIZE) { + /* If RX ring is full, discard a packet */ + riocm_debug(RX_DATA, "ch=%d is full", ch->id); + spin_unlock(&ch->lock); + kfree(buf); + riocm_put_channel(ch); + return -ENOMEM; + } + + ch->rx_ring.buf[ch->rx_ring.head] = buf; + ch->rx_ring.head++; + ch->rx_ring.count++; + ch->rx_ring.head %= RIOCM_RX_RING_SIZE; + + complete(&ch->comp); + + spin_unlock(&ch->lock); + riocm_put_channel(ch); + + return 0; +} + +/* + * rio_ibmsg_handler - inbound message packet handler + */ +static void rio_ibmsg_handler(struct work_struct *work) +{ + struct cm_dev *cm = container_of(work, struct cm_dev, rx_work); + void *data; + struct rio_ch_chan_hdr *hdr; + + if (!rio_mport_is_running(cm->mport)) + return; + + while (1) { + mutex_lock(&cm->rx_lock); + data = riocm_rx_get_msg(cm); + if (data) + riocm_rx_fill(cm, 1); + mutex_unlock(&cm->rx_lock); + + if (data == NULL) + break; + + hdr = data; + + if (hdr->bhdr.type != RIO_CM_CHAN) { + /* For now simply discard packets other than channel */ + riocm_error("Unsupported TYPE code (0x%x). Msg dropped", + hdr->bhdr.type); + kfree(data); + continue; + } + + /* Process a channel message */ + if (hdr->ch_op == CM_DATA_MSG) + rio_rx_data_handler(cm, data); + else + rio_cm_handler(cm, data); + } +} + +static void riocm_inb_msg_event(struct rio_mport *mport, void *dev_id, + int mbox, int slot) +{ + struct cm_dev *cm = dev_id; + + if (rio_mport_is_running(cm->mport) && !work_pending(&cm->rx_work)) + queue_work(cm->rx_wq, &cm->rx_work); +} + +/* + * rio_txcq_handler - TX completion handler + * @cm: cm_dev object + * @slot: TX queue slot + * + * TX completion handler also ensures that pending request packets are placed + * into transmit queue as soon as a free slot becomes available. This is done + * to give higher priority to request packets during high intensity data flow. + */ +static void rio_txcq_handler(struct cm_dev *cm, int slot) +{ + int ack_slot; + + /* ATTN: Add TX completion notification if/when direct buffer + * transfer is implemented. At this moment only correct tracking + * of tx_count is important. + */ + riocm_debug(TX_EVENT, "for mport_%d slot %d tx_cnt %d", + cm->mport->id, slot, cm->tx_cnt); + + spin_lock(&cm->tx_lock); + ack_slot = cm->tx_ack_slot; + + if (ack_slot == slot) + riocm_debug(TX_EVENT, "slot == ack_slot"); + + while (cm->tx_cnt && ((ack_slot != slot) || + (cm->tx_cnt == RIOCM_TX_RING_SIZE))) { + + cm->tx_buf[ack_slot] = NULL; + ++ack_slot; + ack_slot &= (RIOCM_TX_RING_SIZE - 1); + cm->tx_cnt--; + } + + if (cm->tx_cnt < 0 || cm->tx_cnt > RIOCM_TX_RING_SIZE) + riocm_error("tx_cnt %d out of sync", cm->tx_cnt); + + WARN_ON((cm->tx_cnt < 0) || (cm->tx_cnt > RIOCM_TX_RING_SIZE)); + + cm->tx_ack_slot = ack_slot; + + /* + * If there are pending requests, insert them into transmit queue + */ + if (!list_empty(&cm->tx_reqs) && (cm->tx_cnt < RIOCM_TX_RING_SIZE)) { + struct tx_req *req, *_req; + int rc; + + list_for_each_entry_safe(req, _req, &cm->tx_reqs, node) { + list_del(&req->node); + cm->tx_buf[cm->tx_slot] = req->buffer; + rc = rio_add_outb_message(cm->mport, req->rdev, cmbox, + req->buffer, req->len); + kfree(req->buffer); + kfree(req); + + ++cm->tx_cnt; + ++cm->tx_slot; + cm->tx_slot &= (RIOCM_TX_RING_SIZE - 1); + if (cm->tx_cnt == RIOCM_TX_RING_SIZE) + break; + } + } + + spin_unlock(&cm->tx_lock); +} + +static void riocm_outb_msg_event(struct rio_mport *mport, void *dev_id, + int mbox, int slot) +{ + struct cm_dev *cm = dev_id; + + if (cm && rio_mport_is_running(cm->mport)) + rio_txcq_handler(cm, slot); +} + +static int riocm_queue_req(struct cm_dev *cm, struct rio_dev *rdev, + void *buffer, size_t len) +{ + unsigned long flags; + struct tx_req *treq; + + treq = kzalloc(sizeof(*treq), GFP_KERNEL); + if (treq == NULL) + return -ENOMEM; + + treq->rdev = rdev; + treq->buffer = buffer; + treq->len = len; + + spin_lock_irqsave(&cm->tx_lock, flags); + list_add_tail(&treq->node, &cm->tx_reqs); + spin_unlock_irqrestore(&cm->tx_lock, flags); + return 0; +} + +/* + * riocm_post_send - helper function that places packet into msg TX queue + * @cm: cm_dev object + * @rdev: target RapidIO device object (required by outbound msg interface) + * @buffer: pointer to a packet buffer to send + * @len: length of data to transfer + * @req: request priority flag + * + * Returns: 0 if success, or error code otherwise. + */ +static int riocm_post_send(struct cm_dev *cm, struct rio_dev *rdev, + void *buffer, size_t len) +{ + int rc; + unsigned long flags; + + spin_lock_irqsave(&cm->tx_lock, flags); + + if (cm->mport == NULL) { + rc = -ENODEV; + goto err_out; + } + + if (cm->tx_cnt == RIOCM_TX_RING_SIZE) { + riocm_debug(TX, "Tx Queue is full"); + rc = -EBUSY; + goto err_out; + } + + cm->tx_buf[cm->tx_slot] = buffer; + rc = rio_add_outb_message(cm->mport, rdev, cmbox, buffer, len); + + riocm_debug(TX, "Add buf@%p destid=%x tx_slot=%d tx_cnt=%d", + buffer, rdev->destid, cm->tx_slot, cm->tx_cnt); + + ++cm->tx_cnt; + ++cm->tx_slot; + cm->tx_slot &= (RIOCM_TX_RING_SIZE - 1); + +err_out: + spin_unlock_irqrestore(&cm->tx_lock, flags); + return rc; +} + +/* + * riocm_ch_send - sends a data packet to a remote device + * @ch_id: local channel ID + * @buf: pointer to a data buffer to send (including CM header) + * @len: length of data to transfer (including CM header) + * + * ATTN: ASSUMES THAT THE HEADER SPACE IS RESERVED PART OF THE DATA PACKET + * + * Returns: 0 if success, or + * -EINVAL if one or more input parameters is/are not valid, + * -ENODEV if cannot find a channel with specified ID, + * -EAGAIN if a channel is not in CONNECTED state, + * + error codes returned by HW send routine. + */ +static int riocm_ch_send(u16 ch_id, void *buf, int len) +{ + struct rio_channel *ch; + struct rio_ch_chan_hdr *hdr; + int ret; + + if (buf == NULL || ch_id == 0 || len == 0 || len > RIO_MAX_MSG_SIZE) + return -EINVAL; + + ch = riocm_get_channel(ch_id); + if (!ch) { + riocm_error("%s(%d) ch_%d not found", current->comm, + task_pid_nr(current), ch_id); + return -ENODEV; + } + + if (!riocm_cmp(ch, RIO_CM_CONNECTED)) { + ret = -EAGAIN; + goto err_out; + } + + /* + * Fill buffer header section with corresponding channel data + */ + hdr = buf; + + hdr->bhdr.src_id = htonl(ch->loc_destid); + hdr->bhdr.dst_id = htonl(ch->rem_destid); + hdr->bhdr.src_mbox = cmbox; + hdr->bhdr.dst_mbox = cmbox; + hdr->bhdr.type = RIO_CM_CHAN; + hdr->ch_op = CM_DATA_MSG; + hdr->dst_ch = htons(ch->rem_channel); + hdr->src_ch = htons(ch->id); + hdr->msg_len = htons((u16)len); + + /* ATTN: the function call below relies on the fact that underlying + * HW-specific add_outb_message() routine copies TX data into its own + * internal transfer buffer (true for all RIONET compatible mport + * drivers). Must be reviewed if mport driver uses the buffer directly. + */ + + ret = riocm_post_send(ch->cmdev, ch->rdev, buf, len); + if (ret) + riocm_debug(TX, "ch %d send_err=%d", ch->id, ret); +err_out: + riocm_put_channel(ch); + return ret; +} + +static int riocm_ch_free_rxbuf(struct rio_channel *ch, void *buf) +{ + int i, ret = -EINVAL; + + spin_lock_bh(&ch->lock); + + for (i = 0; i < RIOCM_RX_RING_SIZE; i++) { + if (ch->rx_ring.inuse[i] == buf) { + ch->rx_ring.inuse[i] = NULL; + ch->rx_ring.inuse_cnt--; + ret = 0; + break; + } + } + + spin_unlock_bh(&ch->lock); + + if (!ret) + kfree(buf); + + return ret; +} + +/* + * riocm_ch_receive - fetch a data packet received for the specified channel + * @ch: local channel ID + * @buf: pointer to a packet buffer + * @timeout: timeout to wait for incoming packet (in jiffies) + * + * Returns: 0 and valid buffer pointer if success, or NULL pointer and one of: + * -EAGAIN if a channel is not in CONNECTED state, + * -ENOMEM if in-use tracking queue is full, + * -ETIME if wait timeout expired, + * -EINTR if wait was interrupted. + */ +static int riocm_ch_receive(struct rio_channel *ch, void **buf, long timeout) +{ + void *rxmsg = NULL; + int i, ret = 0; + long wret; + + if (!riocm_cmp(ch, RIO_CM_CONNECTED)) { + ret = -EAGAIN; + goto out; + } + + if (ch->rx_ring.inuse_cnt == RIOCM_RX_RING_SIZE) { + /* If we do not have entries to track buffers given to upper + * layer, reject request. + */ + ret = -ENOMEM; + goto out; + } + + wret = wait_for_completion_interruptible_timeout(&ch->comp, timeout); + + riocm_debug(WAIT, "wait on %d returned %ld", ch->id, wret); + + if (!wret) + ret = -ETIME; + else if (wret == -ERESTARTSYS) + ret = -EINTR; + else + ret = riocm_cmp(ch, RIO_CM_CONNECTED) ? 0 : -ECONNRESET; + + if (ret) + goto out; + + spin_lock_bh(&ch->lock); + + rxmsg = ch->rx_ring.buf[ch->rx_ring.tail]; + ch->rx_ring.buf[ch->rx_ring.tail] = NULL; + ch->rx_ring.count--; + ch->rx_ring.tail++; + ch->rx_ring.tail %= RIOCM_RX_RING_SIZE; + ret = -ENOMEM; + + for (i = 0; i < RIOCM_RX_RING_SIZE; i++) { + if (ch->rx_ring.inuse[i] == NULL) { + ch->rx_ring.inuse[i] = rxmsg; + ch->rx_ring.inuse_cnt++; + ret = 0; + break; + } + } + + if (ret) { + /* We have no entry to store pending message: drop it */ + kfree(rxmsg); + rxmsg = NULL; + } + + spin_unlock_bh(&ch->lock); +out: + *buf = rxmsg; + return ret; +} + +/* + * riocm_ch_connect - sends a connect request to a remote device + * @loc_ch: local channel ID + * @cm: CM device to send connect request + * @peer: target RapidIO device + * @rem_ch: remote channel ID + * + * Returns: 0 if success, or + * -EINVAL if the channel is not in IDLE state, + * -EAGAIN if no connection request available immediately, + * -ETIME if ACK response timeout expired, + * -EINTR if wait for response was interrupted. + */ +static int riocm_ch_connect(u16 loc_ch, struct cm_dev *cm, + struct cm_peer *peer, u16 rem_ch) +{ + struct rio_channel *ch = NULL; + struct rio_ch_chan_hdr *hdr; + int ret; + long wret; + + ch = riocm_get_channel(loc_ch); + if (!ch) + return -ENODEV; + + if (!riocm_cmp_exch(ch, RIO_CM_IDLE, RIO_CM_CONNECT)) { + ret = -EINVAL; + goto conn_done; + } + + ch->cmdev = cm; + ch->rdev = peer->rdev; + ch->context = NULL; + ch->loc_destid = cm->mport->host_deviceid; + ch->rem_channel = rem_ch; + + /* + * Send connect request to the remote RapidIO device + */ + + hdr = kzalloc(sizeof(*hdr), GFP_KERNEL); + if (hdr == NULL) { + ret = -ENOMEM; + goto conn_done; + } + + hdr->bhdr.src_id = htonl(ch->loc_destid); + hdr->bhdr.dst_id = htonl(peer->rdev->destid); + hdr->bhdr.src_mbox = cmbox; + hdr->bhdr.dst_mbox = cmbox; + hdr->bhdr.type = RIO_CM_CHAN; + hdr->ch_op = CM_CONN_REQ; + hdr->dst_ch = htons(rem_ch); + hdr->src_ch = htons(loc_ch); + + /* ATTN: the function call below relies on the fact that underlying + * HW-specific add_outb_message() routine copies TX data into its + * internal transfer buffer. Must be reviewed if mport driver uses + * this buffer directly. + */ + ret = riocm_post_send(cm, peer->rdev, hdr, sizeof(*hdr)); + + if (ret != -EBUSY) { + kfree(hdr); + } else { + ret = riocm_queue_req(cm, peer->rdev, hdr, sizeof(*hdr)); + if (ret) + kfree(hdr); + } + + if (ret) { + riocm_cmp_exch(ch, RIO_CM_CONNECT, RIO_CM_IDLE); + goto conn_done; + } + + /* Wait for connect response from the remote device */ + wret = wait_for_completion_interruptible_timeout(&ch->comp, + RIOCM_CONNECT_TO * HZ); + riocm_debug(WAIT, "wait on %d returns %ld", ch->id, wret); + + if (!wret) + ret = -ETIME; + else if (wret == -ERESTARTSYS) + ret = -EINTR; + else + ret = riocm_cmp(ch, RIO_CM_CONNECTED) ? 0 : -1; + +conn_done: + riocm_put_channel(ch); + return ret; +} + +static int riocm_send_ack(struct rio_channel *ch) +{ + struct rio_ch_chan_hdr *hdr; + int ret; + + hdr = kzalloc(sizeof(*hdr), GFP_KERNEL); + if (hdr == NULL) + return -ENOMEM; + + hdr->bhdr.src_id = htonl(ch->loc_destid); + hdr->bhdr.dst_id = htonl(ch->rem_destid); + hdr->dst_ch = htons(ch->rem_channel); + hdr->src_ch = htons(ch->id); + hdr->bhdr.src_mbox = cmbox; + hdr->bhdr.dst_mbox = cmbox; + hdr->bhdr.type = RIO_CM_CHAN; + hdr->ch_op = CM_CONN_ACK; + + /* ATTN: the function call below relies on the fact that underlying + * add_outb_message() routine copies TX data into its internal transfer + * buffer. Review if switching to direct buffer version. + */ + ret = riocm_post_send(ch->cmdev, ch->rdev, hdr, sizeof(*hdr)); + + if (ret == -EBUSY && !riocm_queue_req(ch->cmdev, + ch->rdev, hdr, sizeof(*hdr))) + return 0; + kfree(hdr); + + if (ret) + riocm_error("send ACK to ch_%d on %s failed (ret=%d)", + ch->id, rio_name(ch->rdev), ret); + return ret; +} + +/* + * riocm_ch_accept - accept incoming connection request + * @ch_id: channel ID + * @new_ch_id: local mport device + * @timeout: wait timeout (if 0 non-blocking call, do not wait if connection + * request is not available). + * + * Returns: pointer to new channel struct if success, or error-valued pointer: + * -ENODEV - cannot find specified channel or mport, + * -EINVAL - the channel is not in IDLE state, + * -EAGAIN - no connection request available immediately (timeout=0), + * -ENOMEM - unable to allocate new channel, + * -ETIME - wait timeout expired, + * -EINTR - wait was interrupted. + */ +static struct rio_channel *riocm_ch_accept(u16 ch_id, u16 *new_ch_id, + long timeout) +{ + struct rio_channel *ch = NULL; + struct rio_channel *new_ch = NULL; + struct conn_req *req; + struct cm_peer *peer; + int found = 0; + int err = 0; + long wret; + + ch = riocm_get_channel(ch_id); + if (!ch) + return ERR_PTR(-EINVAL); + + if (!riocm_cmp(ch, RIO_CM_LISTEN)) { + err = -EINVAL; + goto err_put; + } + + /* Don't sleep if this is a non blocking call */ + if (!timeout) { + if (!try_wait_for_completion(&ch->comp)) { + err = -EAGAIN; + goto err_put; + } + } else { + riocm_debug(WAIT, "on %d", ch->id); + + wret = wait_for_completion_interruptible_timeout(&ch->comp, + timeout); + if (!wret) { + err = -ETIME; + goto err_put; + } else if (wret == -ERESTARTSYS) { + err = -EINTR; + goto err_put; + } + } + + spin_lock_bh(&ch->lock); + + if (ch->state != RIO_CM_LISTEN) { + err = -ECANCELED; + } else if (list_empty(&ch->accept_queue)) { + riocm_debug(WAIT, "on %d accept_queue is empty on completion", + ch->id); + err = -EIO; + } + + spin_unlock_bh(&ch->lock); + + if (err) { + riocm_debug(WAIT, "on %d returns %d", ch->id, err); + goto err_put; + } + + /* Create new channel for this connection */ + new_ch = riocm_ch_alloc(RIOCM_CHNUM_AUTO); + + if (IS_ERR(new_ch)) { + riocm_error("failed to get channel for new req (%ld)", + PTR_ERR(new_ch)); + err = -ENOMEM; + goto err_put; + } + + spin_lock_bh(&ch->lock); + + req = list_first_entry(&ch->accept_queue, struct conn_req, node); + list_del(&req->node); + new_ch->cmdev = ch->cmdev; + new_ch->loc_destid = ch->loc_destid; + new_ch->rem_destid = req->destid; + new_ch->rem_channel = req->chan; + + spin_unlock_bh(&ch->lock); + riocm_put_channel(ch); + kfree(req); + + down_read(&rdev_sem); + /* Find requester's device object */ + list_for_each_entry(peer, &new_ch->cmdev->peers, node) { + if (peer->rdev->destid == new_ch->rem_destid) { + riocm_debug(RX_CMD, "found matching device(%s)", + rio_name(peer->rdev)); + found = 1; + break; + } + } + up_read(&rdev_sem); + + if (!found) { + /* If peer device object not found, simply ignore the request */ + err = -ENODEV; + goto err_nodev; + } + + new_ch->rdev = peer->rdev; + new_ch->state = RIO_CM_CONNECTED; + spin_lock_init(&new_ch->lock); + + /* Acknowledge the connection request. */ + riocm_send_ack(new_ch); + + *new_ch_id = new_ch->id; + return new_ch; +err_put: + riocm_put_channel(ch); +err_nodev: + if (new_ch) { + spin_lock_bh(&idr_lock); + idr_remove(&ch_idr, new_ch->id); + spin_unlock_bh(&idr_lock); + riocm_put_channel(new_ch); + } + *new_ch_id = 0; + return ERR_PTR(err); +} + +/* + * riocm_ch_listen - puts a channel into LISTEN state + * @ch_id: channel ID + * + * Returns: 0 if success, or + * -EINVAL if the specified channel does not exists or + * is not in CHAN_BOUND state. + */ +static int riocm_ch_listen(u16 ch_id) +{ + struct rio_channel *ch = NULL; + int ret = 0; + + riocm_debug(CHOP, "(ch_%d)", ch_id); + + ch = riocm_get_channel(ch_id); + if (!ch || !riocm_cmp_exch(ch, RIO_CM_CHAN_BOUND, RIO_CM_LISTEN)) + ret = -EINVAL; + riocm_put_channel(ch); + return ret; +} + +/* + * riocm_ch_bind - associate a channel object and an mport device + * @ch_id: channel ID + * @mport_id: local mport device ID + * @context: pointer to the additional caller's context + * + * Returns: 0 if success, or + * -ENODEV if cannot find specified mport, + * -EINVAL if the specified channel does not exist or + * is not in IDLE state. + */ +static int riocm_ch_bind(u16 ch_id, u8 mport_id, void *context) +{ + struct rio_channel *ch = NULL; + struct cm_dev *cm; + int rc = -ENODEV; + + riocm_debug(CHOP, "ch_%d to mport_%d", ch_id, mport_id); + + /* Find matching cm_dev object */ + down_read(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) { + if ((cm->mport->id == mport_id) && + rio_mport_is_running(cm->mport)) { + rc = 0; + break; + } + } + + if (rc) + goto exit; + + ch = riocm_get_channel(ch_id); + if (!ch) { + rc = -EINVAL; + goto exit; + } + + spin_lock_bh(&ch->lock); + if (ch->state != RIO_CM_IDLE) { + spin_unlock_bh(&ch->lock); + rc = -EINVAL; + goto err_put; + } + + ch->cmdev = cm; + ch->loc_destid = cm->mport->host_deviceid; + ch->context = context; + ch->state = RIO_CM_CHAN_BOUND; + spin_unlock_bh(&ch->lock); +err_put: + riocm_put_channel(ch); +exit: + up_read(&rdev_sem); + return rc; +} + +/* + * riocm_ch_alloc - channel object allocation helper routine + * @ch_num: channel ID (1 ... RIOCM_MAX_CHNUM, 0 = automatic) + * + * Return value: pointer to newly created channel object, + * or error-valued pointer + */ +static struct rio_channel *riocm_ch_alloc(u16 ch_num) +{ + int id; + int start, end; + struct rio_channel *ch; + + ch = kzalloc(sizeof(*ch), GFP_KERNEL); + if (!ch) + return ERR_PTR(-ENOMEM); + + if (ch_num) { + /* If requested, try to obtain the specified channel ID */ + start = ch_num; + end = ch_num + 1; + } else { + /* Obtain channel ID from the dynamic allocation range */ + start = chstart; + end = RIOCM_MAX_CHNUM + 1; + } + + idr_preload(GFP_KERNEL); + spin_lock_bh(&idr_lock); + id = idr_alloc_cyclic(&ch_idr, ch, start, end, GFP_NOWAIT); + spin_unlock_bh(&idr_lock); + idr_preload_end(); + + if (id < 0) { + kfree(ch); + return ERR_PTR(id == -ENOSPC ? -EBUSY : id); + } + + ch->id = (u16)id; + ch->state = RIO_CM_IDLE; + spin_lock_init(&ch->lock); + INIT_LIST_HEAD(&ch->accept_queue); + INIT_LIST_HEAD(&ch->ch_node); + init_completion(&ch->comp); + init_completion(&ch->comp_close); + kref_init(&ch->ref); + ch->rx_ring.head = 0; + ch->rx_ring.tail = 0; + ch->rx_ring.count = 0; + ch->rx_ring.inuse_cnt = 0; + + return ch; +} + +/* + * riocm_ch_create - creates a new channel object and allocates ID for it + * @ch_num: channel ID (1 ... RIOCM_MAX_CHNUM, 0 = automatic) + * + * Allocates and initializes a new channel object. If the parameter ch_num > 0 + * and is within the valid range, riocm_ch_create tries to allocate the + * specified ID for the new channel. If ch_num = 0, channel ID will be assigned + * automatically from the range (chstart ... RIOCM_MAX_CHNUM). + * Module parameter 'chstart' defines start of an ID range available for dynamic + * allocation. Range below 'chstart' is reserved for pre-defined ID numbers. + * Available channel numbers are limited by 16-bit size of channel numbers used + * in the packet header. + * + * Return value: PTR to rio_channel structure if successful (with channel number + * updated via pointer) or error-valued pointer if error. + */ +static struct rio_channel *riocm_ch_create(u16 *ch_num) +{ + struct rio_channel *ch = NULL; + + ch = riocm_ch_alloc(*ch_num); + + if (IS_ERR(ch)) + riocm_debug(CHOP, "Failed to allocate channel %d (err=%ld)", + *ch_num, PTR_ERR(ch)); + else + *ch_num = ch->id; + + return ch; +} + +/* + * riocm_ch_free - channel object release routine + * @ref: pointer to a channel's kref structure + */ +static void riocm_ch_free(struct kref *ref) +{ + struct rio_channel *ch = container_of(ref, struct rio_channel, ref); + int i; + + riocm_debug(CHOP, "(ch_%d)", ch->id); + + if (ch->rx_ring.inuse_cnt) { + for (i = 0; + i < RIOCM_RX_RING_SIZE && ch->rx_ring.inuse_cnt; i++) { + if (ch->rx_ring.inuse[i] != NULL) { + kfree(ch->rx_ring.inuse[i]); + ch->rx_ring.inuse_cnt--; + } + } + } + + if (ch->rx_ring.count) + for (i = 0; i < RIOCM_RX_RING_SIZE && ch->rx_ring.count; i++) { + if (ch->rx_ring.buf[i] != NULL) { + kfree(ch->rx_ring.buf[i]); + ch->rx_ring.count--; + } + } + + complete(&ch->comp_close); +} + +static int riocm_send_close(struct rio_channel *ch) +{ + struct rio_ch_chan_hdr *hdr; + int ret; + + /* + * Send CH_CLOSE notification to the remote RapidIO device + */ + + hdr = kzalloc(sizeof(*hdr), GFP_KERNEL); + if (hdr == NULL) + return -ENOMEM; + + hdr->bhdr.src_id = htonl(ch->loc_destid); + hdr->bhdr.dst_id = htonl(ch->rem_destid); + hdr->bhdr.src_mbox = cmbox; + hdr->bhdr.dst_mbox = cmbox; + hdr->bhdr.type = RIO_CM_CHAN; + hdr->ch_op = CM_CONN_CLOSE; + hdr->dst_ch = htons(ch->rem_channel); + hdr->src_ch = htons(ch->id); + + /* ATTN: the function call below relies on the fact that underlying + * add_outb_message() routine copies TX data into its internal transfer + * buffer. Needs to be reviewed if switched to direct buffer mode. + */ + ret = riocm_post_send(ch->cmdev, ch->rdev, hdr, sizeof(*hdr)); + + if (ret == -EBUSY && !riocm_queue_req(ch->cmdev, ch->rdev, + hdr, sizeof(*hdr))) + return 0; + kfree(hdr); + + if (ret) + riocm_error("ch(%d) send CLOSE failed (ret=%d)", ch->id, ret); + + return ret; +} + +/* + * riocm_ch_close - closes a channel object with specified ID (by local request) + * @ch: channel to be closed + */ +static int riocm_ch_close(struct rio_channel *ch) +{ + unsigned long tmo = msecs_to_jiffies(3000); + enum rio_cm_state state; + long wret; + int ret = 0; + + riocm_debug(CHOP, "ch_%d by %s(%d)", + ch->id, current->comm, task_pid_nr(current)); + + state = riocm_exch(ch, RIO_CM_DESTROYING); + if (state == RIO_CM_CONNECTED) + riocm_send_close(ch); + + complete_all(&ch->comp); + + riocm_put_channel(ch); + wret = wait_for_completion_interruptible_timeout(&ch->comp_close, tmo); + + riocm_debug(WAIT, "wait on %d returns %ld", ch->id, wret); + + if (wret == 0) { + /* Timeout on wait occurred */ + riocm_debug(CHOP, "%s(%d) timed out waiting for ch %d", + current->comm, task_pid_nr(current), ch->id); + ret = -ETIMEDOUT; + } else if (wret == -ERESTARTSYS) { + /* Wait_for_completion was interrupted by a signal */ + riocm_debug(CHOP, "%s(%d) wait for ch %d was interrupted", + current->comm, task_pid_nr(current), ch->id); + ret = -EINTR; + } + + if (!ret) { + riocm_debug(CHOP, "ch_%d resources released", ch->id); + kfree(ch); + } else { + riocm_debug(CHOP, "failed to release ch_%d resources", ch->id); + } + + return ret; +} + +/* + * riocm_cdev_open() - Open character device + */ +static int riocm_cdev_open(struct inode *inode, struct file *filp) +{ + riocm_debug(INIT, "by %s(%d) filp=%p ", + current->comm, task_pid_nr(current), filp); + + if (list_empty(&cm_dev_list)) + return -ENODEV; + + return 0; +} + +/* + * riocm_cdev_release() - Release character device + */ +static int riocm_cdev_release(struct inode *inode, struct file *filp) +{ + struct rio_channel *ch, *_c; + unsigned int i; + LIST_HEAD(list); + + riocm_debug(EXIT, "by %s(%d) filp=%p", + current->comm, task_pid_nr(current), filp); + + /* Check if there are channels associated with this file descriptor */ + spin_lock_bh(&idr_lock); + idr_for_each_entry(&ch_idr, ch, i) { + if (ch && ch->filp == filp) { + riocm_debug(EXIT, "ch_%d not released by %s(%d)", + ch->id, current->comm, + task_pid_nr(current)); + idr_remove(&ch_idr, ch->id); + list_add(&ch->ch_node, &list); + } + } + spin_unlock_bh(&idr_lock); + + if (!list_empty(&list)) { + list_for_each_entry_safe(ch, _c, &list, ch_node) { + list_del(&ch->ch_node); + riocm_ch_close(ch); + } + } + + return 0; +} + +/* + * cm_ep_get_list_size() - Reports number of endpoints in the network + */ +static int cm_ep_get_list_size(void __user *arg) +{ + u32 __user *p = arg; + u32 mport_id; + u32 count = 0; + struct cm_dev *cm; + + if (get_user(mport_id, p)) + return -EFAULT; + if (mport_id >= RIO_MAX_MPORTS) + return -EINVAL; + + /* Find a matching cm_dev object */ + down_read(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) { + if (cm->mport->id == mport_id) { + count = cm->npeers; + up_read(&rdev_sem); + if (copy_to_user(arg, &count, sizeof(u32))) + return -EFAULT; + return 0; + } + } + up_read(&rdev_sem); + + return -ENODEV; +} + +/* + * cm_ep_get_list() - Returns list of attached endpoints + */ +static int cm_ep_get_list(void __user *arg) +{ + struct cm_dev *cm; + struct cm_peer *peer; + u32 info[2]; + void *buf; + u32 nent; + u32 *entry_ptr; + u32 i = 0; + int ret = 0; + + if (copy_from_user(&info, arg, sizeof(info))) + return -EFAULT; + + if (info[1] >= RIO_MAX_MPORTS || info[0] > RIOCM_MAX_EP_COUNT) + return -EINVAL; + + /* Find a matching cm_dev object */ + down_read(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) + if (cm->mport->id == (u8)info[1]) + goto found; + + up_read(&rdev_sem); + return -ENODEV; + +found: + nent = min(info[0], cm->npeers); + buf = kcalloc(nent + 2, sizeof(u32), GFP_KERNEL); + if (!buf) { + up_read(&rdev_sem); + return -ENOMEM; + } + + entry_ptr = (u32 *)((uintptr_t)buf + 2*sizeof(u32)); + + list_for_each_entry(peer, &cm->peers, node) { + *entry_ptr = (u32)peer->rdev->destid; + entry_ptr++; + if (++i == nent) + break; + } + up_read(&rdev_sem); + + ((u32 *)buf)[0] = i; /* report an updated number of entries */ + ((u32 *)buf)[1] = info[1]; /* put back an mport ID */ + if (copy_to_user(arg, buf, sizeof(u32) * (info[0] + 2))) + ret = -EFAULT; + + kfree(buf); + return ret; +} + +/* + * cm_mport_get_list() - Returns list of available local mport devices + */ +static int cm_mport_get_list(void __user *arg) +{ + int ret = 0; + u32 entries; + void *buf; + struct cm_dev *cm; + u32 *entry_ptr; + int count = 0; + + if (copy_from_user(&entries, arg, sizeof(entries))) + return -EFAULT; + if (entries == 0 || entries > RIO_MAX_MPORTS) + return -EINVAL; + buf = kcalloc(entries + 1, sizeof(u32), GFP_KERNEL); + if (!buf) + return -ENOMEM; + + /* Scan all registered cm_dev objects */ + entry_ptr = (u32 *)((uintptr_t)buf + sizeof(u32)); + down_read(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) { + if (count++ < entries) { + *entry_ptr = (cm->mport->id << 16) | + cm->mport->host_deviceid; + entry_ptr++; + } + } + up_read(&rdev_sem); + + *((u32 *)buf) = count; /* report a real number of entries */ + if (copy_to_user(arg, buf, sizeof(u32) * (count + 1))) + ret = -EFAULT; + + kfree(buf); + return ret; +} + +/* + * cm_chan_create() - Create a message exchange channel + */ +static int cm_chan_create(struct file *filp, void __user *arg) +{ + u16 __user *p = arg; + u16 ch_num; + struct rio_channel *ch; + + if (get_user(ch_num, p)) + return -EFAULT; + + riocm_debug(CHOP, "ch_%d requested by %s(%d)", + ch_num, current->comm, task_pid_nr(current)); + ch = riocm_ch_create(&ch_num); + if (IS_ERR(ch)) + return PTR_ERR(ch); + + ch->filp = filp; + riocm_debug(CHOP, "ch_%d created by %s(%d)", + ch_num, current->comm, task_pid_nr(current)); + return put_user(ch_num, p); +} + +/* + * cm_chan_close() - Close channel + * @filp: Pointer to file object + * @arg: Channel to close + */ +static int cm_chan_close(struct file *filp, void __user *arg) +{ + u16 __user *p = arg; + u16 ch_num; + struct rio_channel *ch; + + if (get_user(ch_num, p)) + return -EFAULT; + + riocm_debug(CHOP, "ch_%d by %s(%d)", + ch_num, current->comm, task_pid_nr(current)); + + spin_lock_bh(&idr_lock); + ch = idr_find(&ch_idr, ch_num); + if (!ch) { + spin_unlock_bh(&idr_lock); + return 0; + } + if (ch->filp != filp) { + spin_unlock_bh(&idr_lock); + return -EINVAL; + } + idr_remove(&ch_idr, ch->id); + spin_unlock_bh(&idr_lock); + + return riocm_ch_close(ch); +} + +/* + * cm_chan_bind() - Bind channel + * @arg: Channel number + */ +static int cm_chan_bind(void __user *arg) +{ + struct rio_cm_channel chan; + + if (copy_from_user(&chan, arg, sizeof(chan))) + return -EFAULT; + if (chan.mport_id >= RIO_MAX_MPORTS) + return -EINVAL; + + return riocm_ch_bind(chan.id, chan.mport_id, NULL); +} + +/* + * cm_chan_listen() - Listen on channel + * @arg: Channel number + */ +static int cm_chan_listen(void __user *arg) +{ + u16 __user *p = arg; + u16 ch_num; + + if (get_user(ch_num, p)) + return -EFAULT; + + return riocm_ch_listen(ch_num); +} + +/* + * cm_chan_accept() - Accept incoming connection + * @filp: Pointer to file object + * @arg: Channel number + */ +static int cm_chan_accept(struct file *filp, void __user *arg) +{ + struct rio_cm_accept param; + long accept_to; + struct rio_channel *ch; + + if (copy_from_user(¶m, arg, sizeof(param))) + return -EFAULT; + + riocm_debug(CHOP, "on ch_%d by %s(%d)", + param.ch_num, current->comm, task_pid_nr(current)); + + accept_to = param.wait_to ? + msecs_to_jiffies(param.wait_to) : 0; + + ch = riocm_ch_accept(param.ch_num, ¶m.ch_num, accept_to); + if (IS_ERR(ch)) + return PTR_ERR(ch); + ch->filp = filp; + + riocm_debug(CHOP, "new ch_%d for %s(%d)", + ch->id, current->comm, task_pid_nr(current)); + + if (copy_to_user(arg, ¶m, sizeof(param))) + return -EFAULT; + return 0; +} + +/* + * cm_chan_connect() - Connect on channel + * @arg: Channel information + */ +static int cm_chan_connect(void __user *arg) +{ + struct rio_cm_channel chan; + struct cm_dev *cm; + struct cm_peer *peer; + int ret = -ENODEV; + + if (copy_from_user(&chan, arg, sizeof(chan))) + return -EFAULT; + if (chan.mport_id >= RIO_MAX_MPORTS) + return -EINVAL; + + down_read(&rdev_sem); + + /* Find matching cm_dev object */ + list_for_each_entry(cm, &cm_dev_list, list) { + if (cm->mport->id == chan.mport_id) { + ret = 0; + break; + } + } + + if (ret) + goto err_out; + + if (chan.remote_destid >= RIO_ANY_DESTID(cm->mport->sys_size)) { + ret = -EINVAL; + goto err_out; + } + + /* Find corresponding RapidIO endpoint device object */ + ret = -ENODEV; + + list_for_each_entry(peer, &cm->peers, node) { + if (peer->rdev->destid == chan.remote_destid) { + ret = 0; + break; + } + } + + if (ret) + goto err_out; + + up_read(&rdev_sem); + + return riocm_ch_connect(chan.id, cm, peer, chan.remote_channel); +err_out: + up_read(&rdev_sem); + return ret; +} + +/* + * cm_chan_msg_send() - Send a message through channel + * @arg: Outbound message information + */ +static int cm_chan_msg_send(void __user *arg) +{ + struct rio_cm_msg msg; + void *buf; + int ret = 0; + + if (copy_from_user(&msg, arg, sizeof(msg))) + return -EFAULT; + if (msg.size > RIO_MAX_MSG_SIZE) + return -EINVAL; + + buf = kmalloc(msg.size, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + if (copy_from_user(buf, (void __user *)(uintptr_t)msg.msg, msg.size)) { + ret = -EFAULT; + goto out; + } + + ret = riocm_ch_send(msg.ch_num, buf, msg.size); +out: + kfree(buf); + return ret; +} + +/* + * cm_chan_msg_rcv() - Receive a message through channel + * @arg: Inbound message information + */ +static int cm_chan_msg_rcv(void __user *arg) +{ + struct rio_cm_msg msg; + struct rio_channel *ch; + void *buf; + long rxto; + int ret = 0, msg_size; + + if (copy_from_user(&msg, arg, sizeof(msg))) + return -EFAULT; + + if (msg.ch_num == 0 || msg.size == 0) + return -EINVAL; + + ch = riocm_get_channel(msg.ch_num); + if (!ch) + return -ENODEV; + + rxto = msg.rxto ? msecs_to_jiffies(msg.rxto) : MAX_SCHEDULE_TIMEOUT; + + ret = riocm_ch_receive(ch, &buf, rxto); + if (ret) + goto out; + + msg_size = min(msg.size, (u16)(RIO_MAX_MSG_SIZE)); + + if (copy_to_user((void __user *)(uintptr_t)msg.msg, buf, msg_size)) + ret = -EFAULT; + + riocm_ch_free_rxbuf(ch, buf); +out: + riocm_put_channel(ch); + return ret; +} + +/* + * riocm_cdev_ioctl() - IOCTL requests handler + */ +static long +riocm_cdev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) +{ + switch (cmd) { + case RIO_CM_EP_GET_LIST_SIZE: + return cm_ep_get_list_size((void __user *)arg); + case RIO_CM_EP_GET_LIST: + return cm_ep_get_list((void __user *)arg); + case RIO_CM_CHAN_CREATE: + return cm_chan_create(filp, (void __user *)arg); + case RIO_CM_CHAN_CLOSE: + return cm_chan_close(filp, (void __user *)arg); + case RIO_CM_CHAN_BIND: + return cm_chan_bind((void __user *)arg); + case RIO_CM_CHAN_LISTEN: + return cm_chan_listen((void __user *)arg); + case RIO_CM_CHAN_ACCEPT: + return cm_chan_accept(filp, (void __user *)arg); + case RIO_CM_CHAN_CONNECT: + return cm_chan_connect((void __user *)arg); + case RIO_CM_CHAN_SEND: + return cm_chan_msg_send((void __user *)arg); + case RIO_CM_CHAN_RECEIVE: + return cm_chan_msg_rcv((void __user *)arg); + case RIO_CM_MPORT_GET_LIST: + return cm_mport_get_list((void __user *)arg); + default: + break; + } + + return -EINVAL; +} + +static const struct file_operations riocm_cdev_fops = { + .owner = THIS_MODULE, + .open = riocm_cdev_open, + .release = riocm_cdev_release, + .unlocked_ioctl = riocm_cdev_ioctl, +}; + +/* + * riocm_add_dev - add new remote RapidIO device into channel management core + * @dev: device object associated with RapidIO device + * @sif: subsystem interface + * + * Adds the specified RapidIO device (if applicable) into peers list of + * the corresponding channel management device (cm_dev). + */ +static int riocm_add_dev(struct device *dev, struct subsys_interface *sif) +{ + struct cm_peer *peer; + struct rio_dev *rdev = to_rio_dev(dev); + struct cm_dev *cm; + + /* Check if the remote device has capabilities required to support CM */ + if (!dev_cm_capable(rdev)) + return 0; + + riocm_debug(RDEV, "(%s)", rio_name(rdev)); + + peer = kmalloc(sizeof(*peer), GFP_KERNEL); + if (!peer) + return -ENOMEM; + + /* Find a corresponding cm_dev object */ + down_write(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) { + if (cm->mport == rdev->net->hport) + goto found; + } + + up_write(&rdev_sem); + kfree(peer); + return -ENODEV; + +found: + peer->rdev = rdev; + list_add_tail(&peer->node, &cm->peers); + cm->npeers++; + + up_write(&rdev_sem); + return 0; +} + +/* + * riocm_remove_dev - remove remote RapidIO device from channel management core + * @dev: device object associated with RapidIO device + * @sif: subsystem interface + * + * Removes the specified RapidIO device (if applicable) from peers list of + * the corresponding channel management device (cm_dev). + */ +static void riocm_remove_dev(struct device *dev, struct subsys_interface *sif) +{ + struct rio_dev *rdev = to_rio_dev(dev); + struct cm_dev *cm; + struct cm_peer *peer; + struct rio_channel *ch, *_c; + unsigned int i; + bool found = false; + LIST_HEAD(list); + + /* Check if the remote device has capabilities required to support CM */ + if (!dev_cm_capable(rdev)) + return; + + riocm_debug(RDEV, "(%s)", rio_name(rdev)); + + /* Find matching cm_dev object */ + down_write(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) { + if (cm->mport == rdev->net->hport) { + found = true; + break; + } + } + + if (!found) { + up_write(&rdev_sem); + return; + } + + /* Remove remote device from the list of peers */ + found = false; + list_for_each_entry(peer, &cm->peers, node) { + if (peer->rdev == rdev) { + riocm_debug(RDEV, "removing peer %s", rio_name(rdev)); + found = true; + list_del(&peer->node); + cm->npeers--; + kfree(peer); + break; + } + } + + up_write(&rdev_sem); + + if (!found) + return; + + /* + * Release channels associated with this peer + */ + + spin_lock_bh(&idr_lock); + idr_for_each_entry(&ch_idr, ch, i) { + if (ch && ch->rdev == rdev) { + if (atomic_read(&rdev->state) != RIO_DEVICE_SHUTDOWN) + riocm_exch(ch, RIO_CM_DISCONNECT); + idr_remove(&ch_idr, ch->id); + list_add(&ch->ch_node, &list); + } + } + spin_unlock_bh(&idr_lock); + + if (!list_empty(&list)) { + list_for_each_entry_safe(ch, _c, &list, ch_node) { + list_del(&ch->ch_node); + riocm_ch_close(ch); + } + } +} + +/* + * riocm_cdev_add() - Create rio_cm char device + * @devno: device number assigned to device (MAJ + MIN) + */ +static int riocm_cdev_add(dev_t devno) +{ + int ret; + + cdev_init(&riocm_cdev.cdev, &riocm_cdev_fops); + riocm_cdev.cdev.owner = THIS_MODULE; + ret = cdev_add(&riocm_cdev.cdev, devno, 1); + if (ret < 0) { + riocm_error("Cannot register a device with error %d", ret); + return ret; + } + + riocm_cdev.dev = device_create(dev_class, NULL, devno, NULL, DEV_NAME); + if (IS_ERR(riocm_cdev.dev)) { + cdev_del(&riocm_cdev.cdev); + return PTR_ERR(riocm_cdev.dev); + } + + riocm_debug(MPORT, "Added %s cdev(%d:%d)", + DEV_NAME, MAJOR(devno), MINOR(devno)); + + return 0; +} + +/* + * riocm_add_mport - add new local mport device into channel management core + * @dev: device object associated with mport + * @class_intf: class interface + * + * When a new mport device is added, CM immediately reserves inbound and + * outbound RapidIO mailboxes that will be used. + */ +static int riocm_add_mport(struct device *dev, + struct class_interface *class_intf) +{ + int rc; + int i; + struct cm_dev *cm; + struct rio_mport *mport = to_rio_mport(dev); + + riocm_debug(MPORT, "add mport %s", mport->name); + + cm = kzalloc(sizeof(*cm), GFP_KERNEL); + if (!cm) + return -ENOMEM; + + cm->mport = mport; + + rc = rio_request_outb_mbox(mport, cm, cmbox, + RIOCM_TX_RING_SIZE, riocm_outb_msg_event); + if (rc) { + riocm_error("failed to allocate OBMBOX_%d on %s", + cmbox, mport->name); + kfree(cm); + return -ENODEV; + } + + rc = rio_request_inb_mbox(mport, cm, cmbox, + RIOCM_RX_RING_SIZE, riocm_inb_msg_event); + if (rc) { + riocm_error("failed to allocate IBMBOX_%d on %s", + cmbox, mport->name); + rio_release_outb_mbox(mport, cmbox); + kfree(cm); + return -ENODEV; + } + + /* + * Allocate and register inbound messaging buffers to be ready + * to receive channel and system management requests + */ + for (i = 0; i < RIOCM_RX_RING_SIZE; i++) + cm->rx_buf[i] = NULL; + + cm->rx_slots = RIOCM_RX_RING_SIZE; + mutex_init(&cm->rx_lock); + riocm_rx_fill(cm, RIOCM_RX_RING_SIZE); + cm->rx_wq = create_workqueue(DRV_NAME "/rxq"); + INIT_WORK(&cm->rx_work, rio_ibmsg_handler); + + cm->tx_slot = 0; + cm->tx_cnt = 0; + cm->tx_ack_slot = 0; + spin_lock_init(&cm->tx_lock); + + INIT_LIST_HEAD(&cm->peers); + cm->npeers = 0; + INIT_LIST_HEAD(&cm->tx_reqs); + + down_write(&rdev_sem); + list_add_tail(&cm->list, &cm_dev_list); + up_write(&rdev_sem); + + return 0; +} + +/* + * riocm_remove_mport - remove local mport device from channel management core + * @dev: device object associated with mport + * @class_intf: class interface + * + * Removes a local mport device from the list of registered devices that provide + * channel management services. Returns an error if the specified mport is not + * registered with the CM core. + */ +static void riocm_remove_mport(struct device *dev, + struct class_interface *class_intf) +{ + struct rio_mport *mport = to_rio_mport(dev); + struct cm_dev *cm; + struct cm_peer *peer, *temp; + struct rio_channel *ch, *_c; + unsigned int i; + bool found = false; + LIST_HEAD(list); + + riocm_debug(MPORT, "%s", mport->name); + + /* Find a matching cm_dev object */ + down_write(&rdev_sem); + list_for_each_entry(cm, &cm_dev_list, list) { + if (cm->mport == mport) { + list_del(&cm->list); + found = true; + break; + } + } + up_write(&rdev_sem); + if (!found) + return; + + flush_workqueue(cm->rx_wq); + destroy_workqueue(cm->rx_wq); + + /* Release channels bound to this mport */ + spin_lock_bh(&idr_lock); + idr_for_each_entry(&ch_idr, ch, i) { + if (ch->cmdev == cm) { + riocm_debug(RDEV, "%s drop ch_%d", + mport->name, ch->id); + idr_remove(&ch_idr, ch->id); + list_add(&ch->ch_node, &list); + } + } + spin_unlock_bh(&idr_lock); + + if (!list_empty(&list)) { + list_for_each_entry_safe(ch, _c, &list, ch_node) { + list_del(&ch->ch_node); + riocm_ch_close(ch); + } + } + + rio_release_inb_mbox(mport, cmbox); + rio_release_outb_mbox(mport, cmbox); + + /* Remove and free peer entries */ + if (!list_empty(&cm->peers)) + riocm_debug(RDEV, "ATTN: peer list not empty"); + list_for_each_entry_safe(peer, temp, &cm->peers, node) { + riocm_debug(RDEV, "removing peer %s", rio_name(peer->rdev)); + list_del(&peer->node); + kfree(peer); + } + + riocm_rx_free(cm); + kfree(cm); + riocm_debug(MPORT, "%s done", mport->name); +} + +static int rio_cm_shutdown(struct notifier_block *nb, unsigned long code, + void *unused) +{ + struct rio_channel *ch; + unsigned int i; + + riocm_debug(EXIT, "."); + + spin_lock_bh(&idr_lock); + idr_for_each_entry(&ch_idr, ch, i) { + riocm_debug(EXIT, "close ch %d", ch->id); + if (ch->state == RIO_CM_CONNECTED) + riocm_send_close(ch); + } + spin_unlock_bh(&idr_lock); + + return NOTIFY_DONE; +} + +/* + * riocm_interface handles addition/removal of remote RapidIO devices + */ +static struct subsys_interface riocm_interface = { + .name = "rio_cm", + .subsys = &rio_bus_type, + .add_dev = riocm_add_dev, + .remove_dev = riocm_remove_dev, +}; + +/* + * rio_mport_interface handles addition/removal local mport devices + */ +static struct class_interface rio_mport_interface __refdata = { + .class = &rio_mport_class, + .add_dev = riocm_add_mport, + .remove_dev = riocm_remove_mport, +}; + +static struct notifier_block rio_cm_notifier = { + .notifier_call = rio_cm_shutdown, +}; + +static int __init riocm_init(void) +{ + int ret; + + /* Create device class needed by udev */ + dev_class = class_create(THIS_MODULE, DRV_NAME); + if (IS_ERR(dev_class)) { + riocm_error("Cannot create " DRV_NAME " class"); + return PTR_ERR(dev_class); + } + + ret = alloc_chrdev_region(&dev_number, 0, 1, DRV_NAME); + if (ret) { + class_destroy(dev_class); + return ret; + } + + dev_major = MAJOR(dev_number); + dev_minor_base = MINOR(dev_number); + riocm_debug(INIT, "Registered class with %d major", dev_major); + + /* + * Register as rapidio_port class interface to get notifications about + * mport additions and removals. + */ + ret = class_interface_register(&rio_mport_interface); + if (ret) { + riocm_error("class_interface_register error: %d", ret); + goto err_reg; + } + + /* + * Register as RapidIO bus interface to get notifications about + * addition/removal of remote RapidIO devices. + */ + ret = subsys_interface_register(&riocm_interface); + if (ret) { + riocm_error("subsys_interface_register error: %d", ret); + goto err_cl; + } + + ret = register_reboot_notifier(&rio_cm_notifier); + if (ret) { + riocm_error("failed to register reboot notifier (err=%d)", ret); + goto err_sif; + } + + ret = riocm_cdev_add(dev_number); + if (ret) { + unregister_reboot_notifier(&rio_cm_notifier); + ret = -ENODEV; + goto err_sif; + } + + return 0; +err_sif: + subsys_interface_unregister(&riocm_interface); +err_cl: + class_interface_unregister(&rio_mport_interface); +err_reg: + unregister_chrdev_region(dev_number, 1); + class_destroy(dev_class); + return ret; +} + +static void __exit riocm_exit(void) +{ + riocm_debug(EXIT, "enter"); + unregister_reboot_notifier(&rio_cm_notifier); + subsys_interface_unregister(&riocm_interface); + class_interface_unregister(&rio_mport_interface); + idr_destroy(&ch_idr); + + device_unregister(riocm_cdev.dev); + cdev_del(&(riocm_cdev.cdev)); + + class_destroy(dev_class); + unregister_chrdev_region(dev_number, 1); +} + +late_initcall(riocm_init); +module_exit(riocm_exit); diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild index 6d4e92ccdc91..c44747c0796a 100644 --- a/include/uapi/linux/Kbuild +++ b/include/uapi/linux/Kbuild @@ -357,6 +357,7 @@ header-y += reiserfs_fs.h header-y += reiserfs_xattr.h header-y += resource.h header-y += rfkill.h +header-y += rio_cm_cdev.h header-y += rio_mport_cdev.h header-y += romfs_fs.h header-y += rose.h diff --git a/include/uapi/linux/rio_cm_cdev.h b/include/uapi/linux/rio_cm_cdev.h new file mode 100644 index 000000000000..6edb900d318d --- /dev/null +++ b/include/uapi/linux/rio_cm_cdev.h @@ -0,0 +1,78 @@ +/* + * Copyright (c) 2015, Integrated Device Technology Inc. + * Copyright (c) 2015, Prodrive Technologies + * Copyright (c) 2015, RapidIO Trade Association + * All rights reserved. + * + * This software is available to you under a choice of one of two licenses. + * You may choose to be licensed under the terms of the GNU General Public + * License(GPL) Version 2, or the BSD-3 Clause license below: + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright notice, + * this list of conditions and the following disclaimer in the documentation + * and/or other materials provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of its contributors + * may be used to endorse or promote products derived from this software without + * specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR + * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, + * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, + * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, + * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR + * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _RIO_CM_CDEV_H_ +#define _RIO_CM_CDEV_H_ + +#include + +struct rio_cm_channel { + __u16 id; + __u16 remote_channel; + __u16 remote_destid; + __u8 mport_id; +}; + +struct rio_cm_msg { + __u16 ch_num; + __u16 size; + __u32 rxto; /* receive timeout in mSec. 0 = blocking */ + __u64 msg; +}; + +struct rio_cm_accept { + __u16 ch_num; + __u16 pad0; + __u32 wait_to; /* accept timeout in mSec. 0 = blocking */ +}; + +/* RapidIO Channelized Messaging Driver IOCTLs */ +#define RIO_CM_IOC_MAGIC 'c' + +#define RIO_CM_EP_GET_LIST_SIZE _IOWR(RIO_CM_IOC_MAGIC, 1, __u32) +#define RIO_CM_EP_GET_LIST _IOWR(RIO_CM_IOC_MAGIC, 2, __u32) +#define RIO_CM_CHAN_CREATE _IOWR(RIO_CM_IOC_MAGIC, 3, __u16) +#define RIO_CM_CHAN_CLOSE _IOW(RIO_CM_IOC_MAGIC, 4, __u16) +#define RIO_CM_CHAN_BIND _IOW(RIO_CM_IOC_MAGIC, 5, struct rio_cm_channel) +#define RIO_CM_CHAN_LISTEN _IOW(RIO_CM_IOC_MAGIC, 6, __u16) +#define RIO_CM_CHAN_ACCEPT _IOWR(RIO_CM_IOC_MAGIC, 7, struct rio_cm_accept) +#define RIO_CM_CHAN_CONNECT _IOW(RIO_CM_IOC_MAGIC, 8, struct rio_cm_channel) +#define RIO_CM_CHAN_SEND _IOW(RIO_CM_IOC_MAGIC, 9, struct rio_cm_msg) +#define RIO_CM_CHAN_RECEIVE _IOWR(RIO_CM_IOC_MAGIC, 10, struct rio_cm_msg) +#define RIO_CM_MPORT_GET_LIST _IOWR(RIO_CM_IOC_MAGIC, 11, __u32) + +#endif /* _RIO_CM_CDEV_H_ */ -- cgit v1.2.3 From c49298026908a8ce9dcf01ed68734ad171cef98b Mon Sep 17 00:00:00 2001 From: Ira Weiny Date: Wed, 27 Jul 2016 21:08:42 -0400 Subject: IB/hfi1: Allow for non-double word multiple message sizes for user SDMA The driver pads non-double word multiple message sizes but it doesn't account for this padding when the packet length is calculated. Also, the data length is miscalculated for message sizes less than 4 bytes due to the bit representation in LRH. And there's a check for non-double word multiple message sizes that prevents these messages from being sent. This patch fixes length miscalculations and enables the functionality to send non-double word multiple message sizes. Reviewed-by: Harish Chegondi Signed-off-by: Sebastian Sanchez Signed-off-by: Ira Weiny Signed-off-by: Doug Ledford --- drivers/infiniband/hw/hfi1/user_sdma.c | 31 ++++++++++++++++++++++--------- include/uapi/rdma/hfi/hfi1_user.h | 2 +- 2 files changed, 23 insertions(+), 10 deletions(-) (limited to 'include/uapi') diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c index d16ed52a2cb1..1e266c95056a 100644 --- a/drivers/infiniband/hw/hfi1/user_sdma.c +++ b/drivers/infiniband/hw/hfi1/user_sdma.c @@ -793,14 +793,21 @@ static inline u32 compute_data_length(struct user_sdma_request *req, * The size of the data of the first packet is in the header * template. However, it includes the header and ICRC, which need * to be subtracted. + * The minimum representable packet data length in a header is 4 bytes, + * therefore, when the data length request is less than 4 bytes, there's + * only one packet, and the packet data length is equal to that of the + * request data length. * The size of the remaining packets is the minimum of the frag * size (MTU) or remaining data in the request. */ u32 len; if (!req->seqnum) { - len = ((be16_to_cpu(req->hdr.lrh[2]) << 2) - - (sizeof(tx->hdr) - 4)); + if (req->data_len < sizeof(u32)) + len = req->data_len; + else + len = ((be16_to_cpu(req->hdr.lrh[2]) << 2) - + (sizeof(tx->hdr) - 4)); } else if (req_opcode(req->info.ctrl) == EXPECTED) { u32 tidlen = EXP_TID_GET(req->tids[req->tididx], LEN) * PAGE_SIZE; @@ -830,6 +837,13 @@ static inline u32 compute_data_length(struct user_sdma_request *req, return len; } +static inline u32 pad_len(u32 len) +{ + if (len & (sizeof(u32) - 1)) + len += sizeof(u32) - (len & (sizeof(u32) - 1)); + return len; +} + static inline u32 get_lrh_len(struct hfi1_pkt_header hdr, u32 len) { /* (Size of complete header - size of PBC) + 4B ICRC + data length */ @@ -921,7 +935,8 @@ static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts) if (test_bit(SDMA_REQ_HAVE_AHG, &req->flags)) { if (!req->seqnum) { u16 pbclen = le16_to_cpu(req->hdr.pbc[0]); - u32 lrhlen = get_lrh_len(req->hdr, datalen); + u32 lrhlen = get_lrh_len(req->hdr, + pad_len(datalen)); /* * Copy the request header into the tx header * because the HW needs a cacheline-aligned @@ -1219,16 +1234,14 @@ static int check_header_template(struct user_sdma_request *req, /* * Perform safety checks for any type of packet: * - transfer size is multiple of 64bytes - * - packet length is multiple of 4bytes - * - entire request length is multiple of 4bytes + * - packet length is multiple of 4 bytes * - packet length is not larger than MTU size * * These checks are only done for the first packet of the * transfer since the header is "given" to us by user space. * For the remainder of the packets we compute the values. */ - if (req->info.fragsize % PIO_BLOCK_SIZE || - lrhlen & 0x3 || req->data_len & 0x3 || + if (req->info.fragsize % PIO_BLOCK_SIZE || lrhlen & 0x3 || lrhlen > get_lrh_len(*hdr, req->info.fragsize)) return -EINVAL; @@ -1290,7 +1303,7 @@ static int set_txreq_header(struct user_sdma_request *req, struct hfi1_pkt_header *hdr = &tx->hdr; u16 pbclen; int ret; - u32 tidval = 0, lrhlen = get_lrh_len(*hdr, datalen); + u32 tidval = 0, lrhlen = get_lrh_len(*hdr, pad_len(datalen)); /* Copy the header template to the request before modification */ memcpy(hdr, &req->hdr, sizeof(*hdr)); @@ -1401,7 +1414,7 @@ static int set_txreq_header_ahg(struct user_sdma_request *req, struct hfi1_user_sdma_pkt_q *pq = req->pq; struct hfi1_pkt_header *hdr = &req->hdr; u16 pbclen = le16_to_cpu(hdr->pbc[0]); - u32 val32, tidval = 0, lrhlen = get_lrh_len(*hdr, len); + u32 val32, tidval = 0, lrhlen = get_lrh_len(*hdr, pad_len(len)); if (PBC2LRH(pbclen) != lrhlen) { /* PBC.PbcLengthDWs */ diff --git a/include/uapi/rdma/hfi/hfi1_user.h b/include/uapi/rdma/hfi/hfi1_user.h index 98bebf8bef55..d15e7289d835 100644 --- a/include/uapi/rdma/hfi/hfi1_user.h +++ b/include/uapi/rdma/hfi/hfi1_user.h @@ -75,7 +75,7 @@ * may not be implemented; the user code must deal with this if it * cares, or it must abort after initialization reports the difference. */ -#define HFI1_USER_SWMINOR 1 +#define HFI1_USER_SWMINOR 2 /* * We will encode the major/minor inside a single 32bit version number. -- cgit v1.2.3 From 444d13ff10fb13bc3e64859c3cf9ce43dcfeb075 Mon Sep 17 00:00:00 2001 From: Jessica Yu Date: Wed, 27 Jul 2016 12:06:21 +0930 Subject: modules: add ro_after_init support Add ro_after_init support for modules by adding a new page-aligned section in the module layout (after rodata) for ro_after_init data and enabling RO protection for that section after module init runs. Signed-off-by: Jessica Yu Acked-by: Kees Cook Signed-off-by: Rusty Russell --- include/linux/module.h | 6 +++-- include/uapi/linux/elf.h | 1 + kernel/livepatch/core.c | 2 +- kernel/module.c | 66 +++++++++++++++++++++++++++++++++++++++--------- 4 files changed, 60 insertions(+), 15 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/module.h b/include/linux/module.h index f95ed243a4de..0c3207d26ac0 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -298,6 +298,8 @@ struct module_layout { unsigned int text_size; /* Size of RO section of the module (text+rodata) */ unsigned int ro_size; + /* Size of RO after init section */ + unsigned int ro_after_init_size; #ifdef CONFIG_MODULES_TREE_LOOKUP struct mod_tree_node mtn; @@ -765,12 +767,12 @@ extern int module_sysfs_initialized; #ifdef CONFIG_DEBUG_SET_MODULE_RONX extern void set_all_modules_text_rw(void); extern void set_all_modules_text_ro(void); -extern void module_enable_ro(const struct module *mod); +extern void module_enable_ro(const struct module *mod, bool after_init); extern void module_disable_ro(const struct module *mod); #else static inline void set_all_modules_text_rw(void) { } static inline void set_all_modules_text_ro(void) { } -static inline void module_enable_ro(const struct module *mod) { } +static inline void module_enable_ro(const struct module *mod, bool after_init) { } static inline void module_disable_ro(const struct module *mod) { } #endif diff --git a/include/uapi/linux/elf.h b/include/uapi/linux/elf.h index cb4a72f888d5..70b172ba41ce 100644 --- a/include/uapi/linux/elf.h +++ b/include/uapi/linux/elf.h @@ -286,6 +286,7 @@ typedef struct elf64_phdr { #define SHF_ALLOC 0x2 #define SHF_EXECINSTR 0x4 #define SHF_RELA_LIVEPATCH 0x00100000 +#define SHF_RO_AFTER_INIT 0x00200000 #define SHF_MASKPROC 0xf0000000 /* special section indexes */ diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c index 5c2bc1052691..8bbe50704621 100644 --- a/kernel/livepatch/core.c +++ b/kernel/livepatch/core.c @@ -309,7 +309,7 @@ static int klp_write_object_relocations(struct module *pmod, break; } - module_enable_ro(pmod); + module_enable_ro(pmod, true); return ret; } diff --git a/kernel/module.c b/kernel/module.c index c91c2fdca2e6..205a71a97852 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1857,10 +1857,11 @@ static void mod_sysfs_teardown(struct module *mod) * from modification and any data from execution. * * General layout of module is: - * [text] [read-only-data] [writable data] - * text_size -----^ ^ ^ - * ro_size ------------------------| | - * size -------------------------------------------| + * [text] [read-only-data] [ro-after-init] [writable data] + * text_size -----^ ^ ^ ^ + * ro_size ------------------------| | | + * ro_after_init_size -----------------------------| | + * size -----------------------------------------------------------| * * These values are always page-aligned (as is base) */ @@ -1883,14 +1884,24 @@ static void frob_rodata(const struct module_layout *layout, (layout->ro_size - layout->text_size) >> PAGE_SHIFT); } +static void frob_ro_after_init(const struct module_layout *layout, + int (*set_memory)(unsigned long start, int num_pages)) +{ + BUG_ON((unsigned long)layout->base & (PAGE_SIZE-1)); + BUG_ON((unsigned long)layout->ro_size & (PAGE_SIZE-1)); + BUG_ON((unsigned long)layout->ro_after_init_size & (PAGE_SIZE-1)); + set_memory((unsigned long)layout->base + layout->ro_size, + (layout->ro_after_init_size - layout->ro_size) >> PAGE_SHIFT); +} + static void frob_writable_data(const struct module_layout *layout, int (*set_memory)(unsigned long start, int num_pages)) { BUG_ON((unsigned long)layout->base & (PAGE_SIZE-1)); - BUG_ON((unsigned long)layout->ro_size & (PAGE_SIZE-1)); + BUG_ON((unsigned long)layout->ro_after_init_size & (PAGE_SIZE-1)); BUG_ON((unsigned long)layout->size & (PAGE_SIZE-1)); - set_memory((unsigned long)layout->base + layout->ro_size, - (layout->size - layout->ro_size) >> PAGE_SHIFT); + set_memory((unsigned long)layout->base + layout->ro_after_init_size, + (layout->size - layout->ro_after_init_size) >> PAGE_SHIFT); } /* livepatching wants to disable read-only so it can frob module. */ @@ -1898,21 +1909,26 @@ void module_disable_ro(const struct module *mod) { frob_text(&mod->core_layout, set_memory_rw); frob_rodata(&mod->core_layout, set_memory_rw); + frob_ro_after_init(&mod->core_layout, set_memory_rw); frob_text(&mod->init_layout, set_memory_rw); frob_rodata(&mod->init_layout, set_memory_rw); } -void module_enable_ro(const struct module *mod) +void module_enable_ro(const struct module *mod, bool after_init) { frob_text(&mod->core_layout, set_memory_ro); frob_rodata(&mod->core_layout, set_memory_ro); frob_text(&mod->init_layout, set_memory_ro); frob_rodata(&mod->init_layout, set_memory_ro); + + if (after_init) + frob_ro_after_init(&mod->core_layout, set_memory_ro); } static void module_enable_nx(const struct module *mod) { frob_rodata(&mod->core_layout, set_memory_nx); + frob_ro_after_init(&mod->core_layout, set_memory_nx); frob_writable_data(&mod->core_layout, set_memory_nx); frob_rodata(&mod->init_layout, set_memory_nx); frob_writable_data(&mod->init_layout, set_memory_nx); @@ -1921,6 +1937,7 @@ static void module_enable_nx(const struct module *mod) static void module_disable_nx(const struct module *mod) { frob_rodata(&mod->core_layout, set_memory_x); + frob_ro_after_init(&mod->core_layout, set_memory_x); frob_writable_data(&mod->core_layout, set_memory_x); frob_rodata(&mod->init_layout, set_memory_x); frob_writable_data(&mod->init_layout, set_memory_x); @@ -1963,6 +1980,8 @@ static void disable_ro_nx(const struct module_layout *layout) frob_text(layout, set_memory_rw); frob_rodata(layout, set_memory_rw); frob_rodata(layout, set_memory_x); + frob_ro_after_init(layout, set_memory_rw); + frob_ro_after_init(layout, set_memory_x); frob_writable_data(layout, set_memory_x); } @@ -2305,6 +2324,7 @@ static void layout_sections(struct module *mod, struct load_info *info) * finder in the two loops below */ { SHF_EXECINSTR | SHF_ALLOC, ARCH_SHF_SMALL }, { SHF_ALLOC, SHF_WRITE | ARCH_SHF_SMALL }, + { SHF_RO_AFTER_INIT | SHF_ALLOC, ARCH_SHF_SMALL }, { SHF_WRITE | SHF_ALLOC, ARCH_SHF_SMALL }, { ARCH_SHF_SMALL | SHF_ALLOC, 0 } }; @@ -2336,7 +2356,11 @@ static void layout_sections(struct module *mod, struct load_info *info) mod->core_layout.size = debug_align(mod->core_layout.size); mod->core_layout.ro_size = mod->core_layout.size; break; - case 3: /* whole core */ + case 2: /* RO after init */ + mod->core_layout.size = debug_align(mod->core_layout.size); + mod->core_layout.ro_after_init_size = mod->core_layout.size; + break; + case 4: /* whole core */ mod->core_layout.size = debug_align(mod->core_layout.size); break; } @@ -2366,7 +2390,14 @@ static void layout_sections(struct module *mod, struct load_info *info) mod->init_layout.size = debug_align(mod->init_layout.size); mod->init_layout.ro_size = mod->init_layout.size; break; - case 3: /* whole init */ + case 2: + /* + * RO after init doesn't apply to init_layout (only + * core_layout), so it just takes the value of ro_size. + */ + mod->init_layout.ro_after_init_size = mod->init_layout.ro_size; + break; + case 4: /* whole init */ mod->init_layout.size = debug_align(mod->init_layout.size); break; } @@ -3193,6 +3224,7 @@ static struct module *layout_and_allocate(struct load_info *info, int flags) { /* Module within temporary copy. */ struct module *mod; + unsigned int ndx; int err; mod = setup_load_info(info, flags); @@ -3215,6 +3247,15 @@ static struct module *layout_and_allocate(struct load_info *info, int flags) /* We will do a special allocation for per-cpu sections later. */ info->sechdrs[info->index.pcpu].sh_flags &= ~(unsigned long)SHF_ALLOC; + /* + * Mark ro_after_init section with SHF_RO_AFTER_INIT so that + * layout_sections() can put it in the right place. + * Note: ro_after_init sections also have SHF_{WRITE,ALLOC} set. + */ + ndx = find_sec(info, ".data..ro_after_init"); + if (ndx) + info->sechdrs[ndx].sh_flags |= SHF_RO_AFTER_INIT; + /* Determine total sizes, and put offsets in sh_entsize. For now this is done generically; there doesn't appear to be any special cases for the architectures. */ @@ -3381,12 +3422,14 @@ static noinline int do_init_module(struct module *mod) /* Switch to core kallsyms now init is done: kallsyms may be walking! */ rcu_assign_pointer(mod->kallsyms, &mod->core_kallsyms); #endif + module_enable_ro(mod, true); mod_tree_remove_init(mod); disable_ro_nx(&mod->init_layout); module_arch_freeing_init(mod); mod->init_layout.base = NULL; mod->init_layout.size = 0; mod->init_layout.ro_size = 0; + mod->init_layout.ro_after_init_size = 0; mod->init_layout.text_size = 0; /* * We want to free module_init, but be aware that kallsyms may be @@ -3478,8 +3521,7 @@ static int complete_formation(struct module *mod, struct load_info *info) /* This relies on module_mutex for list integrity. */ module_bug_finalize(info->hdr, info->sechdrs, mod); - /* Set RO and NX regions */ - module_enable_ro(mod); + module_enable_ro(mod, false); module_enable_nx(mod); /* Mark state as coming so strong_try_module_get() ignores us, -- cgit v1.2.3 From ab15c95a17b3fe8c0e01bb7ce1dd0b657598eb61 Mon Sep 17 00:00:00 2001 From: Alex Vesker Date: Wed, 6 Jul 2016 16:36:35 +0300 Subject: IB/core: Support for CMA multicast join flags Added UCMA and CMA support for multicast join flags. Flags are passed using UCMA CM join command previously reserved fields. Currently supporting two join flags indicating two different multicast JoinStates: 1. Full Member: The initiator creates the Multicast group(MCG) if it wasn't previously created, can send Multicast messages to the group and receive messages from the MCG. 2. Send Only Full Member: The initiator creates the Multicast group(MCG) if it wasn't previously created, can send Multicast messages to the group but doesn't receive any messages from the MCG. IB: Send Only Full Member requires a query of ClassPortInfo to determine if SM/SA supports this option. If SM/SA doesn't support Send-Only there will be no join request sent and an error will be returned. ETH: When Send Only Full Member is requested no IGMP join will be sent. Signed-off-by: Alex Vesker Reviewed by: Hal Rosenstock Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/core/cma.c | 100 +++++++++++++++++++++++++++++++++--- drivers/infiniband/core/multicast.c | 12 ----- drivers/infiniband/core/ucma.c | 18 +++++-- include/rdma/ib_sa.h | 13 +++++ include/rdma/rdma_cm.h | 4 +- include/uapi/rdma/rdma_user_cm.h | 9 +++- 6 files changed, 131 insertions(+), 25 deletions(-) (limited to 'include/uapi') diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index f0c91ba3178a..0451307bea18 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -68,6 +68,7 @@ MODULE_DESCRIPTION("Generic RDMA CM Agent"); MODULE_LICENSE("Dual BSD/GPL"); #define CMA_CM_RESPONSE_TIMEOUT 20 +#define CMA_QUERY_CLASSPORT_INFO_TIMEOUT 3000 #define CMA_MAX_CM_RETRIES 15 #define CMA_CM_MRA_SETTING (IB_CM_MRA_FLAG_DELAY | 24) #define CMA_IBOE_PACKET_LIFETIME 18 @@ -162,6 +163,14 @@ struct rdma_bind_list { unsigned short port; }; +struct class_port_info_context { + struct ib_class_port_info *class_port_info; + struct ib_device *device; + struct completion done; + struct ib_sa_query *sa_query; + u8 port_num; +}; + static int cma_ps_alloc(struct net *net, enum rdma_port_space ps, struct rdma_bind_list *bind_list, int snum) { @@ -306,6 +315,7 @@ struct cma_multicast { struct sockaddr_storage addr; struct kref mcref; bool igmp_joined; + u8 join_state; }; struct cma_work { @@ -3754,10 +3764,63 @@ static void cma_set_mgid(struct rdma_id_private *id_priv, } } +static void cma_query_sa_classport_info_cb(int status, + struct ib_class_port_info *rec, + void *context) +{ + struct class_port_info_context *cb_ctx = context; + + WARN_ON(!context); + + if (status || !rec) { + pr_debug("RDMA CM: %s port %u failed query ClassPortInfo status: %d\n", + cb_ctx->device->name, cb_ctx->port_num, status); + goto out; + } + + memcpy(cb_ctx->class_port_info, rec, sizeof(struct ib_class_port_info)); + +out: + complete(&cb_ctx->done); +} + +static int cma_query_sa_classport_info(struct ib_device *device, u8 port_num, + struct ib_class_port_info *class_port_info) +{ + struct class_port_info_context *cb_ctx; + int ret; + + cb_ctx = kmalloc(sizeof(*cb_ctx), GFP_KERNEL); + if (!cb_ctx) + return -ENOMEM; + + cb_ctx->device = device; + cb_ctx->class_port_info = class_port_info; + cb_ctx->port_num = port_num; + init_completion(&cb_ctx->done); + + ret = ib_sa_classport_info_rec_query(&sa_client, device, port_num, + CMA_QUERY_CLASSPORT_INFO_TIMEOUT, + GFP_KERNEL, cma_query_sa_classport_info_cb, + cb_ctx, &cb_ctx->sa_query); + if (ret < 0) { + pr_err("RDMA CM: %s port %u failed to send ClassPortInfo query, ret: %d\n", + device->name, port_num, ret); + goto out; + } + + wait_for_completion(&cb_ctx->done); + +out: + kfree(cb_ctx); + return ret; +} + static int cma_join_ib_multicast(struct rdma_id_private *id_priv, struct cma_multicast *mc) { struct ib_sa_mcmember_rec rec; + struct ib_class_port_info class_port_info; struct rdma_dev_addr *dev_addr = &id_priv->id.route.addr.dev_addr; ib_sa_comp_mask comp_mask; int ret; @@ -3776,7 +3839,24 @@ static int cma_join_ib_multicast(struct rdma_id_private *id_priv, rec.qkey = cpu_to_be32(id_priv->qkey); rdma_addr_get_sgid(dev_addr, &rec.port_gid); rec.pkey = cpu_to_be16(ib_addr_get_pkey(dev_addr)); - rec.join_state = 1; + rec.join_state = mc->join_state; + + if (rec.join_state == BIT(SENDONLY_FULLMEMBER_JOIN)) { + ret = cma_query_sa_classport_info(id_priv->id.device, + id_priv->id.port_num, + &class_port_info); + + if (ret) + return ret; + + if (!(ib_get_cpi_capmask2(&class_port_info) & + IB_SA_CAP_MASK2_SENDONLY_FULL_MEM_SUPPORT)) { + pr_warn("RDMA CM: %s port %u Unable to multicast join\n" + "RDMA CM: SM doesn't support Send Only Full Member option\n", + id_priv->id.device->name, id_priv->id.port_num); + return -EOPNOTSUPP; + } + } comp_mask = IB_SA_MCMEMBER_REC_MGID | IB_SA_MCMEMBER_REC_PORT_GID | IB_SA_MCMEMBER_REC_PKEY | IB_SA_MCMEMBER_REC_JOIN_STATE | @@ -3845,6 +3925,9 @@ static int cma_iboe_join_multicast(struct rdma_id_private *id_priv, struct sockaddr *addr = (struct sockaddr *)&mc->addr; struct net_device *ndev = NULL; enum ib_gid_type gid_type; + bool send_only; + + send_only = mc->join_state == BIT(SENDONLY_FULLMEMBER_JOIN); if (cma_zero_addr((struct sockaddr *)&mc->addr)) return -EINVAL; @@ -3878,12 +3961,14 @@ static int cma_iboe_join_multicast(struct rdma_id_private *id_priv, gid_type = id_priv->cma_dev->default_gid_type[id_priv->id.port_num - rdma_start_port(id_priv->cma_dev->device)]; if (addr->sa_family == AF_INET) { - if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) - err = cma_igmp_send(ndev, &mc->multicast.ib->rec.mgid, - true); - if (!err) { - mc->igmp_joined = true; + if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) { mc->multicast.ib->rec.hop_limit = IPV6_DEFAULT_HOPLIMIT; + if (!send_only) { + err = cma_igmp_send(ndev, &mc->multicast.ib->rec.mgid, + true); + if (!err) + mc->igmp_joined = true; + } } } else { if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) @@ -3913,7 +3998,7 @@ out1: } int rdma_join_multicast(struct rdma_cm_id *id, struct sockaddr *addr, - void *context) + u8 join_state, void *context) { struct rdma_id_private *id_priv; struct cma_multicast *mc; @@ -3932,6 +4017,7 @@ int rdma_join_multicast(struct rdma_cm_id *id, struct sockaddr *addr, mc->context = context; mc->id_priv = id_priv; mc->igmp_joined = false; + mc->join_state = join_state; spin_lock(&id_priv->lock); list_add(&mc->list, &id_priv->mc_list); spin_unlock(&id_priv->lock); diff --git a/drivers/infiniband/core/multicast.c b/drivers/infiniband/core/multicast.c index a83ec28a147b..3a3c5d73bbfc 100644 --- a/drivers/infiniband/core/multicast.c +++ b/drivers/infiniband/core/multicast.c @@ -93,18 +93,6 @@ enum { struct mcast_member; -/* -* There are 4 types of join states: -* FullMember, NonMember, SendOnlyNonMember, SendOnlyFullMember. -*/ -enum { - FULLMEMBER_JOIN, - NONMEMBER_JOIN, - SENDONLY_NONMEBER_JOIN, - SENDONLY_FULLMEMBER_JOIN, - NUM_JOIN_MEMBERSHIP_TYPES, -}; - struct mcast_group { struct ib_sa_mcmember_rec rec; struct rb_node node; diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c index c0f3826abb30..2825ece91d3c 100644 --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -106,6 +106,7 @@ struct ucma_multicast { int events_reported; u64 uid; + u8 join_state; struct list_head list; struct sockaddr_storage addr; }; @@ -1317,12 +1318,20 @@ static ssize_t ucma_process_join(struct ucma_file *file, struct ucma_multicast *mc; struct sockaddr *addr; int ret; + u8 join_state; if (out_len < sizeof(resp)) return -ENOSPC; addr = (struct sockaddr *) &cmd->addr; - if (cmd->reserved || !cmd->addr_size || (cmd->addr_size != rdma_addr_size(addr))) + if (!cmd->addr_size || (cmd->addr_size != rdma_addr_size(addr))) + return -EINVAL; + + if (cmd->join_flags == RDMA_MC_JOIN_FLAG_FULLMEMBER) + join_state = BIT(FULLMEMBER_JOIN); + else if (cmd->join_flags == RDMA_MC_JOIN_FLAG_SENDONLY_FULLMEMBER) + join_state = BIT(SENDONLY_FULLMEMBER_JOIN); + else return -EINVAL; ctx = ucma_get_ctx(file, cmd->id); @@ -1335,10 +1344,11 @@ static ssize_t ucma_process_join(struct ucma_file *file, ret = -ENOMEM; goto err1; } - + mc->join_state = join_state; mc->uid = cmd->uid; memcpy(&mc->addr, addr, cmd->addr_size); - ret = rdma_join_multicast(ctx->cm_id, (struct sockaddr *) &mc->addr, mc); + ret = rdma_join_multicast(ctx->cm_id, (struct sockaddr *)&mc->addr, + join_state, mc); if (ret) goto err2; @@ -1382,7 +1392,7 @@ static ssize_t ucma_join_ip_multicast(struct ucma_file *file, join_cmd.uid = cmd.uid; join_cmd.id = cmd.id; join_cmd.addr_size = rdma_addr_size((struct sockaddr *) &cmd.addr); - join_cmd.reserved = 0; + join_cmd.join_flags = RDMA_MC_JOIN_FLAG_FULLMEMBER; memcpy(&join_cmd.addr, &cmd.addr, join_cmd.addr_size); return ucma_process_join(file, &join_cmd, out_len); diff --git a/include/rdma/ib_sa.h b/include/rdma/ib_sa.h index 384041669489..5ee7aab95eb8 100644 --- a/include/rdma/ib_sa.h +++ b/include/rdma/ib_sa.h @@ -94,6 +94,19 @@ enum ib_sa_selector { IB_SA_BEST = 3 }; +/* + * There are 4 types of join states: + * FullMember, NonMember, SendOnlyNonMember, SendOnlyFullMember. + * The order corresponds to JoinState bits in MCMemberRecord. + */ +enum ib_sa_mc_join_states { + FULLMEMBER_JOIN, + NONMEMBER_JOIN, + SENDONLY_NONMEBER_JOIN, + SENDONLY_FULLMEMBER_JOIN, + NUM_JOIN_MEMBERSHIP_TYPES, +}; + #define IB_SA_CAP_MASK2_SENDONLY_FULL_MEM_SUPPORT BIT(12) /* diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index afe44fde72a5..81fb1d15e8bb 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -333,11 +333,13 @@ int rdma_disconnect(struct rdma_cm_id *id); * address. * @id: Communication identifier associated with the request. * @addr: Multicast address identifying the group to join. + * @join_state: Multicast JoinState bitmap requested by port. + * Bitmap is based on IB_SA_MCMEMBER_REC_JOIN_STATE bits. * @context: User-defined context associated with the join request, returned * to the user through the private_data pointer in multicast events. */ int rdma_join_multicast(struct rdma_cm_id *id, struct sockaddr *addr, - void *context); + u8 join_state, void *context); /** * rdma_leave_multicast - Leave the multicast group specified by the given diff --git a/include/uapi/rdma/rdma_user_cm.h b/include/uapi/rdma/rdma_user_cm.h index 3066718eb120..01923d463673 100644 --- a/include/uapi/rdma/rdma_user_cm.h +++ b/include/uapi/rdma/rdma_user_cm.h @@ -244,12 +244,19 @@ struct rdma_ucm_join_ip_mcast { __u32 id; }; +/* Multicast join flags */ +enum { + RDMA_MC_JOIN_FLAG_FULLMEMBER, + RDMA_MC_JOIN_FLAG_SENDONLY_FULLMEMBER, + RDMA_MC_JOIN_FLAG_RESERVED, +}; + struct rdma_ucm_join_mcast { __u64 response; /* rdma_ucma_create_id_resp */ __u64 uid; __u32 id; __u16 addr_size; - __u16 reserved; + __u16 join_flags; struct sockaddr_storage addr; }; -- cgit v1.2.3 From 8700e3e7c4857d28ebaa824509934556da0b3e76 Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Thu, 16 Jun 2016 16:45:23 +0300 Subject: Soft RoCE driver Soft RoCE (RXE) - The software RoCE driver ib_rxe implements the RDMA transport and registers to the RDMA core device as a kernel verbs provider. It also implements the packet IO layer. On the other hand ib_rxe registers to the Linux netdev stack as a udp encapsulating protocol, in that case RDMA, for sending and receiving packets over any Ethernet device. This yields a RDMA transport over the UDP/Ethernet network layer forming a RoCEv2 compatible device. The configuration procedure of the Soft RoCE drivers requires binding to any existing Ethernet network device. This is done with /sys interface. A userspace Soft RoCE library (librxe) provides user applications the ability to run with Soft RoCE devices. The use of rxe verbs ins user space requires the inclusion of librxe as a device specifics plug-in to libibverbs. librxe is packaged separately. Architecture: +-----------------------------------------------------------+ | Application | +-----------------------------------------------------------+ +-----------------------------------+ | libibverbs | User +-----------------------------------+ +----------------+ +----------------+ | librxe | | HW RoCE lib | +----------------+ +----------------+ +---------------------------------------------------------------+ +--------------+ +------------+ | Sockets | | RDMA ULP | +--------------+ +------------+ +--------------+ +---------------------+ | TCP/IP | | ib_core | +--------------+ +---------------------+ +------------+ +----------------+ Kernel | ib_rxe | | HW RoCE driver | +------------+ +----------------+ +------------------------------------+ | NIC driver | +------------------------------------+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +-----------------------------------------------------------+ | Application | +-----------------------------------------------------------+ +-----------------------------------+ | libibverbs | User +-----------------------------------+ +----------------+ +----------------+ | librxe | | HW RoCE lib | +----------------+ +----------------+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +--------------+ +------------+ | Sockets | | RDMA ULP | +--------------+ +------------+ +--------------+ +---------------------+ | TCP/IP | | ib_core | +--------------+ +---------------------+ +------------+ +----------------+ Kernel | ib_rxe | | HW RoCE driver | +------------+ +----------------+ +------------------------------------+ | NIC driver | +------------------------------------+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Soft RoCE resources: [1[ https://github.com/SoftRoCE/librxe-dev librxe - source code in Github [2] https://github.com/SoftRoCE/rxe-dev/wiki/rxe-dev:-Home - Soft RoCE Wiki page [3] https://github.com/SoftRoCE/librxe-dev - Soft RoCE userspace library Signed-off-by: Kamal Heib Signed-off-by: Amir Vadai Signed-off-by: Moni Shoua Reviewed-by: Haggai Eran Signed-off-by: Doug Ledford --- MAINTAINERS | 9 + drivers/infiniband/Kconfig | 1 + drivers/infiniband/sw/Makefile | 1 + drivers/infiniband/sw/rxe/Kconfig | 24 + drivers/infiniband/sw/rxe/Makefile | 24 + drivers/infiniband/sw/rxe/rxe.c | 386 +++++++++ drivers/infiniband/sw/rxe/rxe.h | 77 ++ drivers/infiniband/sw/rxe/rxe_av.c | 98 +++ drivers/infiniband/sw/rxe/rxe_comp.c | 734 +++++++++++++++++ drivers/infiniband/sw/rxe/rxe_cq.c | 165 ++++ drivers/infiniband/sw/rxe/rxe_dma.c | 166 ++++ drivers/infiniband/sw/rxe/rxe_hdr.h | 952 ++++++++++++++++++++++ drivers/infiniband/sw/rxe/rxe_icrc.c | 96 +++ drivers/infiniband/sw/rxe/rxe_loc.h | 286 +++++++ drivers/infiniband/sw/rxe/rxe_mcast.c | 190 +++++ drivers/infiniband/sw/rxe/rxe_mmap.c | 173 ++++ drivers/infiniband/sw/rxe/rxe_mr.c | 643 +++++++++++++++ drivers/infiniband/sw/rxe/rxe_net.c | 708 ++++++++++++++++ drivers/infiniband/sw/rxe/rxe_net.h | 53 ++ drivers/infiniband/sw/rxe/rxe_opcode.c | 961 ++++++++++++++++++++++ drivers/infiniband/sw/rxe/rxe_opcode.h | 129 +++ drivers/infiniband/sw/rxe/rxe_param.h | 172 ++++ drivers/infiniband/sw/rxe/rxe_pool.c | 502 ++++++++++++ drivers/infiniband/sw/rxe/rxe_pool.h | 163 ++++ drivers/infiniband/sw/rxe/rxe_qp.c | 851 ++++++++++++++++++++ drivers/infiniband/sw/rxe/rxe_queue.c | 217 +++++ drivers/infiniband/sw/rxe/rxe_queue.h | 178 ++++ drivers/infiniband/sw/rxe/rxe_recv.c | 420 ++++++++++ drivers/infiniband/sw/rxe/rxe_req.c | 726 +++++++++++++++++ drivers/infiniband/sw/rxe/rxe_resp.c | 1380 ++++++++++++++++++++++++++++++++ drivers/infiniband/sw/rxe/rxe_srq.c | 193 +++++ drivers/infiniband/sw/rxe/rxe_sysfs.c | 157 ++++ drivers/infiniband/sw/rxe/rxe_task.c | 154 ++++ drivers/infiniband/sw/rxe/rxe_task.h | 95 +++ drivers/infiniband/sw/rxe/rxe_verbs.c | 1330 ++++++++++++++++++++++++++++++ drivers/infiniband/sw/rxe/rxe_verbs.h | 480 +++++++++++ include/uapi/rdma/Kbuild | 1 + include/uapi/rdma/rdma_user_rxe.h | 144 ++++ 38 files changed, 13039 insertions(+) create mode 100644 drivers/infiniband/sw/rxe/Kconfig create mode 100644 drivers/infiniband/sw/rxe/Makefile create mode 100644 drivers/infiniband/sw/rxe/rxe.c create mode 100644 drivers/infiniband/sw/rxe/rxe.h create mode 100644 drivers/infiniband/sw/rxe/rxe_av.c create mode 100644 drivers/infiniband/sw/rxe/rxe_comp.c create mode 100644 drivers/infiniband/sw/rxe/rxe_cq.c create mode 100644 drivers/infiniband/sw/rxe/rxe_dma.c create mode 100644 drivers/infiniband/sw/rxe/rxe_hdr.h create mode 100644 drivers/infiniband/sw/rxe/rxe_icrc.c create mode 100644 drivers/infiniband/sw/rxe/rxe_loc.h create mode 100644 drivers/infiniband/sw/rxe/rxe_mcast.c create mode 100644 drivers/infiniband/sw/rxe/rxe_mmap.c create mode 100644 drivers/infiniband/sw/rxe/rxe_mr.c create mode 100644 drivers/infiniband/sw/rxe/rxe_net.c create mode 100644 drivers/infiniband/sw/rxe/rxe_net.h create mode 100644 drivers/infiniband/sw/rxe/rxe_opcode.c create mode 100644 drivers/infiniband/sw/rxe/rxe_opcode.h create mode 100644 drivers/infiniband/sw/rxe/rxe_param.h create mode 100644 drivers/infiniband/sw/rxe/rxe_pool.c create mode 100644 drivers/infiniband/sw/rxe/rxe_pool.h create mode 100644 drivers/infiniband/sw/rxe/rxe_qp.c create mode 100644 drivers/infiniband/sw/rxe/rxe_queue.c create mode 100644 drivers/infiniband/sw/rxe/rxe_queue.h create mode 100644 drivers/infiniband/sw/rxe/rxe_recv.c create mode 100644 drivers/infiniband/sw/rxe/rxe_req.c create mode 100644 drivers/infiniband/sw/rxe/rxe_resp.c create mode 100644 drivers/infiniband/sw/rxe/rxe_srq.c create mode 100644 drivers/infiniband/sw/rxe/rxe_sysfs.c create mode 100644 drivers/infiniband/sw/rxe/rxe_task.c create mode 100644 drivers/infiniband/sw/rxe/rxe_task.h create mode 100644 drivers/infiniband/sw/rxe/rxe_verbs.c create mode 100644 drivers/infiniband/sw/rxe/rxe_verbs.h create mode 100644 include/uapi/rdma/rdma_user_rxe.h (limited to 'include/uapi') diff --git a/MAINTAINERS b/MAINTAINERS index e1b090f86e0d..4daa6712eac1 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7444,6 +7444,15 @@ W: http://www.mellanox.com Q: http://patchwork.ozlabs.org/project/netdev/list/ F: drivers/net/ethernet/mellanox/mlxsw/ +SOFT-ROCE DRIVER (rxe) +M: Moni Shoua +L: linux-rdma@vger.kernel.org +S: Supported +W: https://github.com/SoftRoCE/rxe-dev/wiki/rxe-dev:-Home +Q: http://patchwork.kernel.org/project/linux-rdma/list/ +F: drivers/infiniband/hw/rxe/ +F: include/uapi/rdma/rdma_user_rxe.h + MEMBARRIER SUPPORT M: Mathieu Desnoyers M: "Paul E. McKenney" diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig index 2137adfbd8c3..e9b7dc037ff8 100644 --- a/drivers/infiniband/Kconfig +++ b/drivers/infiniband/Kconfig @@ -84,6 +84,7 @@ source "drivers/infiniband/ulp/iser/Kconfig" source "drivers/infiniband/ulp/isert/Kconfig" source "drivers/infiniband/sw/rdmavt/Kconfig" +source "drivers/infiniband/sw/rxe/Kconfig" source "drivers/infiniband/hw/hfi1/Kconfig" diff --git a/drivers/infiniband/sw/Makefile b/drivers/infiniband/sw/Makefile index 988b6a0101a4..8b095b27db87 100644 --- a/drivers/infiniband/sw/Makefile +++ b/drivers/infiniband/sw/Makefile @@ -1 +1,2 @@ obj-$(CONFIG_INFINIBAND_RDMAVT) += rdmavt/ +obj-$(CONFIG_RDMA_RXE) += rxe/ diff --git a/drivers/infiniband/sw/rxe/Kconfig b/drivers/infiniband/sw/rxe/Kconfig new file mode 100644 index 000000000000..1e4e628fe7b0 --- /dev/null +++ b/drivers/infiniband/sw/rxe/Kconfig @@ -0,0 +1,24 @@ +config RDMA_RXE + tristate "Software RDMA over Ethernet (RoCE) driver" + depends on INET && PCI && INFINIBAND + depends on NET_UDP_TUNNEL + ---help--- + This driver implements the InfiniBand RDMA transport over + the Linux network stack. It enables a system with a + standard Ethernet adapter to interoperate with a RoCE + adapter or with another system running the RXE driver. + Documentation on InfiniBand and RoCE can be downloaded at + www.infinibandta.org and www.openfabrics.org. (See also + siw which is a similar software driver for iWARP.) + + The driver is split into two layers, one interfaces with the + Linux RDMA stack and implements a kernel or user space + verbs API. The user space verbs API requires a support + library named librxe which is loaded by the generic user + space verbs API, libibverbs. The other layer interfaces + with the Linux network stack at layer 3. + + To configure and work with soft-RoCE driver please use the + following wiki page under "configure Soft-RoCE (RXE)" section: + + https://github.com/SoftRoCE/rxe-dev/wiki/rxe-dev:-Home diff --git a/drivers/infiniband/sw/rxe/Makefile b/drivers/infiniband/sw/rxe/Makefile new file mode 100644 index 000000000000..3b3fb9d1c470 --- /dev/null +++ b/drivers/infiniband/sw/rxe/Makefile @@ -0,0 +1,24 @@ +obj-$(CONFIG_RDMA_RXE) += rdma_rxe.o + +rdma_rxe-y := \ + rxe.o \ + rxe_comp.o \ + rxe_req.o \ + rxe_resp.o \ + rxe_recv.o \ + rxe_pool.o \ + rxe_queue.o \ + rxe_verbs.o \ + rxe_av.o \ + rxe_srq.o \ + rxe_qp.o \ + rxe_cq.o \ + rxe_mr.o \ + rxe_dma.o \ + rxe_opcode.o \ + rxe_mmap.o \ + rxe_icrc.o \ + rxe_mcast.o \ + rxe_task.o \ + rxe_net.o \ + rxe_sysfs.o diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c new file mode 100644 index 000000000000..55f0e8f0ca79 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe.c @@ -0,0 +1,386 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +MODULE_AUTHOR("Bob Pearson, Frank Zago, John Groves, Kamal Heib"); +MODULE_DESCRIPTION("Soft RDMA transport"); +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_VERSION("0.2"); + +/* free resources for all ports on a device */ +static void rxe_cleanup_ports(struct rxe_dev *rxe) +{ + kfree(rxe->port.pkey_tbl); + rxe->port.pkey_tbl = NULL; + +} + +/* free resources for a rxe device all objects created for this device must + * have been destroyed + */ +static void rxe_cleanup(struct rxe_dev *rxe) +{ + rxe_pool_cleanup(&rxe->uc_pool); + rxe_pool_cleanup(&rxe->pd_pool); + rxe_pool_cleanup(&rxe->ah_pool); + rxe_pool_cleanup(&rxe->srq_pool); + rxe_pool_cleanup(&rxe->qp_pool); + rxe_pool_cleanup(&rxe->cq_pool); + rxe_pool_cleanup(&rxe->mr_pool); + rxe_pool_cleanup(&rxe->mw_pool); + rxe_pool_cleanup(&rxe->mc_grp_pool); + rxe_pool_cleanup(&rxe->mc_elem_pool); + + rxe_cleanup_ports(rxe); +} + +/* called when all references have been dropped */ +void rxe_release(struct kref *kref) +{ + struct rxe_dev *rxe = container_of(kref, struct rxe_dev, ref_cnt); + + rxe_cleanup(rxe); + ib_dealloc_device(&rxe->ib_dev); +} + +void rxe_dev_put(struct rxe_dev *rxe) +{ + kref_put(&rxe->ref_cnt, rxe_release); +} +EXPORT_SYMBOL_GPL(rxe_dev_put); + +/* initialize rxe device parameters */ +static int rxe_init_device_param(struct rxe_dev *rxe) +{ + rxe->max_inline_data = RXE_MAX_INLINE_DATA; + + rxe->attr.fw_ver = RXE_FW_VER; + rxe->attr.max_mr_size = RXE_MAX_MR_SIZE; + rxe->attr.page_size_cap = RXE_PAGE_SIZE_CAP; + rxe->attr.vendor_id = RXE_VENDOR_ID; + rxe->attr.vendor_part_id = RXE_VENDOR_PART_ID; + rxe->attr.hw_ver = RXE_HW_VER; + rxe->attr.max_qp = RXE_MAX_QP; + rxe->attr.max_qp_wr = RXE_MAX_QP_WR; + rxe->attr.device_cap_flags = RXE_DEVICE_CAP_FLAGS; + rxe->attr.max_sge = RXE_MAX_SGE; + rxe->attr.max_sge_rd = RXE_MAX_SGE_RD; + rxe->attr.max_cq = RXE_MAX_CQ; + rxe->attr.max_cqe = (1 << RXE_MAX_LOG_CQE) - 1; + rxe->attr.max_mr = RXE_MAX_MR; + rxe->attr.max_pd = RXE_MAX_PD; + rxe->attr.max_qp_rd_atom = RXE_MAX_QP_RD_ATOM; + rxe->attr.max_ee_rd_atom = RXE_MAX_EE_RD_ATOM; + rxe->attr.max_res_rd_atom = RXE_MAX_RES_RD_ATOM; + rxe->attr.max_qp_init_rd_atom = RXE_MAX_QP_INIT_RD_ATOM; + rxe->attr.max_ee_init_rd_atom = RXE_MAX_EE_INIT_RD_ATOM; + rxe->attr.atomic_cap = RXE_ATOMIC_CAP; + rxe->attr.max_ee = RXE_MAX_EE; + rxe->attr.max_rdd = RXE_MAX_RDD; + rxe->attr.max_mw = RXE_MAX_MW; + rxe->attr.max_raw_ipv6_qp = RXE_MAX_RAW_IPV6_QP; + rxe->attr.max_raw_ethy_qp = RXE_MAX_RAW_ETHY_QP; + rxe->attr.max_mcast_grp = RXE_MAX_MCAST_GRP; + rxe->attr.max_mcast_qp_attach = RXE_MAX_MCAST_QP_ATTACH; + rxe->attr.max_total_mcast_qp_attach = RXE_MAX_TOT_MCAST_QP_ATTACH; + rxe->attr.max_ah = RXE_MAX_AH; + rxe->attr.max_fmr = RXE_MAX_FMR; + rxe->attr.max_map_per_fmr = RXE_MAX_MAP_PER_FMR; + rxe->attr.max_srq = RXE_MAX_SRQ; + rxe->attr.max_srq_wr = RXE_MAX_SRQ_WR; + rxe->attr.max_srq_sge = RXE_MAX_SRQ_SGE; + rxe->attr.max_fast_reg_page_list_len = RXE_MAX_FMR_PAGE_LIST_LEN; + rxe->attr.max_pkeys = RXE_MAX_PKEYS; + rxe->attr.local_ca_ack_delay = RXE_LOCAL_CA_ACK_DELAY; + + rxe->max_ucontext = RXE_MAX_UCONTEXT; + + return 0; +} + +/* initialize port attributes */ +static int rxe_init_port_param(struct rxe_port *port) +{ + port->attr.state = RXE_PORT_STATE; + port->attr.max_mtu = RXE_PORT_MAX_MTU; + port->attr.active_mtu = RXE_PORT_ACTIVE_MTU; + port->attr.gid_tbl_len = RXE_PORT_GID_TBL_LEN; + port->attr.port_cap_flags = RXE_PORT_PORT_CAP_FLAGS; + port->attr.max_msg_sz = RXE_PORT_MAX_MSG_SZ; + port->attr.bad_pkey_cntr = RXE_PORT_BAD_PKEY_CNTR; + port->attr.qkey_viol_cntr = RXE_PORT_QKEY_VIOL_CNTR; + port->attr.pkey_tbl_len = RXE_PORT_PKEY_TBL_LEN; + port->attr.lid = RXE_PORT_LID; + port->attr.sm_lid = RXE_PORT_SM_LID; + port->attr.lmc = RXE_PORT_LMC; + port->attr.max_vl_num = RXE_PORT_MAX_VL_NUM; + port->attr.sm_sl = RXE_PORT_SM_SL; + port->attr.subnet_timeout = RXE_PORT_SUBNET_TIMEOUT; + port->attr.init_type_reply = RXE_PORT_INIT_TYPE_REPLY; + port->attr.active_width = RXE_PORT_ACTIVE_WIDTH; + port->attr.active_speed = RXE_PORT_ACTIVE_SPEED; + port->attr.phys_state = RXE_PORT_PHYS_STATE; + port->mtu_cap = + ib_mtu_enum_to_int(RXE_PORT_ACTIVE_MTU); + port->subnet_prefix = cpu_to_be64(RXE_PORT_SUBNET_PREFIX); + + return 0; +} + +/* initialize port state, note IB convention that HCA ports are always + * numbered from 1 + */ +static int rxe_init_ports(struct rxe_dev *rxe) +{ + struct rxe_port *port = &rxe->port; + + rxe_init_port_param(port); + + if (!port->attr.pkey_tbl_len || !port->attr.gid_tbl_len) + return -EINVAL; + + port->pkey_tbl = kcalloc(port->attr.pkey_tbl_len, + sizeof(*port->pkey_tbl), GFP_KERNEL); + + if (!port->pkey_tbl) + return -ENOMEM; + + port->pkey_tbl[0] = 0xffff; + port->port_guid = rxe->ifc_ops->port_guid(rxe); + + spin_lock_init(&port->port_lock); + + return 0; +} + +/* init pools of managed objects */ +static int rxe_init_pools(struct rxe_dev *rxe) +{ + int err; + + err = rxe_pool_init(rxe, &rxe->uc_pool, RXE_TYPE_UC, + rxe->max_ucontext); + if (err) + goto err1; + + err = rxe_pool_init(rxe, &rxe->pd_pool, RXE_TYPE_PD, + rxe->attr.max_pd); + if (err) + goto err2; + + err = rxe_pool_init(rxe, &rxe->ah_pool, RXE_TYPE_AH, + rxe->attr.max_ah); + if (err) + goto err3; + + err = rxe_pool_init(rxe, &rxe->srq_pool, RXE_TYPE_SRQ, + rxe->attr.max_srq); + if (err) + goto err4; + + err = rxe_pool_init(rxe, &rxe->qp_pool, RXE_TYPE_QP, + rxe->attr.max_qp); + if (err) + goto err5; + + err = rxe_pool_init(rxe, &rxe->cq_pool, RXE_TYPE_CQ, + rxe->attr.max_cq); + if (err) + goto err6; + + err = rxe_pool_init(rxe, &rxe->mr_pool, RXE_TYPE_MR, + rxe->attr.max_mr); + if (err) + goto err7; + + err = rxe_pool_init(rxe, &rxe->mw_pool, RXE_TYPE_MW, + rxe->attr.max_mw); + if (err) + goto err8; + + err = rxe_pool_init(rxe, &rxe->mc_grp_pool, RXE_TYPE_MC_GRP, + rxe->attr.max_mcast_grp); + if (err) + goto err9; + + err = rxe_pool_init(rxe, &rxe->mc_elem_pool, RXE_TYPE_MC_ELEM, + rxe->attr.max_total_mcast_qp_attach); + if (err) + goto err10; + + return 0; + +err10: + rxe_pool_cleanup(&rxe->mc_grp_pool); +err9: + rxe_pool_cleanup(&rxe->mw_pool); +err8: + rxe_pool_cleanup(&rxe->mr_pool); +err7: + rxe_pool_cleanup(&rxe->cq_pool); +err6: + rxe_pool_cleanup(&rxe->qp_pool); +err5: + rxe_pool_cleanup(&rxe->srq_pool); +err4: + rxe_pool_cleanup(&rxe->ah_pool); +err3: + rxe_pool_cleanup(&rxe->pd_pool); +err2: + rxe_pool_cleanup(&rxe->uc_pool); +err1: + return err; +} + +/* initialize rxe device state */ +static int rxe_init(struct rxe_dev *rxe) +{ + int err; + + /* init default device parameters */ + rxe_init_device_param(rxe); + + err = rxe_init_ports(rxe); + if (err) + goto err1; + + err = rxe_init_pools(rxe); + if (err) + goto err2; + + /* init pending mmap list */ + spin_lock_init(&rxe->mmap_offset_lock); + spin_lock_init(&rxe->pending_lock); + INIT_LIST_HEAD(&rxe->pending_mmaps); + INIT_LIST_HEAD(&rxe->list); + + mutex_init(&rxe->usdev_lock); + + return 0; + +err2: + rxe_cleanup_ports(rxe); +err1: + return err; +} + +int rxe_set_mtu(struct rxe_dev *rxe, unsigned int ndev_mtu) +{ + struct rxe_port *port = &rxe->port; + enum ib_mtu mtu; + + mtu = eth_mtu_int_to_enum(ndev_mtu); + + /* Make sure that new MTU in range */ + mtu = mtu ? min_t(enum ib_mtu, mtu, RXE_PORT_MAX_MTU) : IB_MTU_256; + + port->attr.active_mtu = mtu; + port->mtu_cap = ib_mtu_enum_to_int(mtu); + + return 0; +} +EXPORT_SYMBOL(rxe_set_mtu); + +/* called by ifc layer to create new rxe device. + * The caller should allocate memory for rxe by calling ib_alloc_device. + */ +int rxe_add(struct rxe_dev *rxe, unsigned int mtu) +{ + int err; + + kref_init(&rxe->ref_cnt); + + err = rxe_init(rxe); + if (err) + goto err1; + + err = rxe_set_mtu(rxe, mtu); + if (err) + goto err1; + + err = rxe_register_device(rxe); + if (err) + goto err1; + + return 0; + +err1: + rxe_dev_put(rxe); + return err; +} +EXPORT_SYMBOL(rxe_add); + +/* called by the ifc layer to remove a device */ +void rxe_remove(struct rxe_dev *rxe) +{ + rxe_unregister_device(rxe); + + rxe_dev_put(rxe); +} +EXPORT_SYMBOL(rxe_remove); + +static int __init rxe_module_init(void) +{ + int err; + + /* initialize slab caches for managed objects */ + err = rxe_cache_init(); + if (err) { + pr_err("rxe: unable to init object pools\n"); + return err; + } + + err = rxe_net_init(); + if (err) { + pr_err("rxe: unable to init\n"); + rxe_cache_exit(); + return err; + } + pr_info("rxe: loaded\n"); + + return 0; +} + +static void __exit rxe_module_exit(void) +{ + rxe_remove_all(); + rxe_net_exit(); + rxe_cache_exit(); + + pr_info("rxe: unloaded\n"); +} + +module_init(rxe_module_init); +module_exit(rxe_module_exit); diff --git a/drivers/infiniband/sw/rxe/rxe.h b/drivers/infiniband/sw/rxe/rxe.h new file mode 100644 index 000000000000..12c71c549f97 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe.h @@ -0,0 +1,77 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_H +#define RXE_H + +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#include "rxe_net.h" +#include "rxe_opcode.h" +#include "rxe_hdr.h" +#include "rxe_param.h" +#include "rxe_verbs.h" + +#define RXE_UVERBS_ABI_VERSION (1) + +#define IB_PHYS_STATE_LINK_UP (5) +#define IB_PHYS_STATE_LINK_DOWN (3) + +#define RXE_ROCE_V2_SPORT (0xc000) + +int rxe_set_mtu(struct rxe_dev *rxe, unsigned int dev_mtu); + +int rxe_add(struct rxe_dev *rxe, unsigned int mtu); +void rxe_remove(struct rxe_dev *rxe); +void rxe_remove_all(void); + +int rxe_rcv(struct sk_buff *skb); + +void rxe_dev_put(struct rxe_dev *rxe); +struct rxe_dev *net_to_rxe(struct net_device *ndev); +struct rxe_dev *get_rxe_by_name(const char* name); + +void rxe_port_up(struct rxe_dev *rxe); +void rxe_port_down(struct rxe_dev *rxe); + +#endif /* RXE_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_av.c b/drivers/infiniband/sw/rxe/rxe_av.c new file mode 100644 index 000000000000..5c9474212d4e --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_av.c @@ -0,0 +1,98 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +int rxe_av_chk_attr(struct rxe_dev *rxe, struct ib_ah_attr *attr) +{ + struct rxe_port *port; + + if (attr->port_num != 1) { + pr_info("rxe: invalid port_num = %d\n", attr->port_num); + return -EINVAL; + } + + port = &rxe->port; + + if (attr->ah_flags & IB_AH_GRH) { + if (attr->grh.sgid_index > port->attr.gid_tbl_len) { + pr_info("rxe: invalid sgid index = %d\n", + attr->grh.sgid_index); + return -EINVAL; + } + } + + return 0; +} + +int rxe_av_from_attr(struct rxe_dev *rxe, u8 port_num, + struct rxe_av *av, struct ib_ah_attr *attr) +{ + memset(av, 0, sizeof(*av)); + memcpy(&av->grh, &attr->grh, sizeof(attr->grh)); + av->port_num = port_num; + return 0; +} + +int rxe_av_to_attr(struct rxe_dev *rxe, struct rxe_av *av, + struct ib_ah_attr *attr) +{ + memcpy(&attr->grh, &av->grh, sizeof(av->grh)); + attr->port_num = av->port_num; + return 0; +} + +int rxe_av_fill_ip_info(struct rxe_dev *rxe, + struct rxe_av *av, + struct ib_ah_attr *attr, + struct ib_gid_attr *sgid_attr, + union ib_gid *sgid) +{ + rdma_gid2ip(&av->sgid_addr._sockaddr, sgid); + rdma_gid2ip(&av->dgid_addr._sockaddr, &attr->grh.dgid); + av->network_type = ib_gid_to_network_type(sgid_attr->gid_type, sgid); + + return 0; +} + +struct rxe_av *rxe_get_av(struct rxe_pkt_info *pkt) +{ + if (!pkt || !pkt->qp) + return NULL; + + if (qp_type(pkt->qp) == IB_QPT_RC || qp_type(pkt->qp) == IB_QPT_UC) + return &pkt->qp->pri_av; + + return (pkt->wqe) ? &pkt->wqe->av : NULL; +} diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c new file mode 100644 index 000000000000..36f67de44095 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -0,0 +1,734 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" +#include "rxe_task.h" + +enum comp_state { + COMPST_GET_ACK, + COMPST_GET_WQE, + COMPST_COMP_WQE, + COMPST_COMP_ACK, + COMPST_CHECK_PSN, + COMPST_CHECK_ACK, + COMPST_READ, + COMPST_ATOMIC, + COMPST_WRITE_SEND, + COMPST_UPDATE_COMP, + COMPST_ERROR_RETRY, + COMPST_RNR_RETRY, + COMPST_ERROR, + COMPST_EXIT, /* We have an issue, and we want to rerun the completer */ + COMPST_DONE, /* The completer finished successflly */ +}; + +static char *comp_state_name[] = { + [COMPST_GET_ACK] = "GET ACK", + [COMPST_GET_WQE] = "GET WQE", + [COMPST_COMP_WQE] = "COMP WQE", + [COMPST_COMP_ACK] = "COMP ACK", + [COMPST_CHECK_PSN] = "CHECK PSN", + [COMPST_CHECK_ACK] = "CHECK ACK", + [COMPST_READ] = "READ", + [COMPST_ATOMIC] = "ATOMIC", + [COMPST_WRITE_SEND] = "WRITE/SEND", + [COMPST_UPDATE_COMP] = "UPDATE COMP", + [COMPST_ERROR_RETRY] = "ERROR RETRY", + [COMPST_RNR_RETRY] = "RNR RETRY", + [COMPST_ERROR] = "ERROR", + [COMPST_EXIT] = "EXIT", + [COMPST_DONE] = "DONE", +}; + +static unsigned long rnrnak_usec[32] = { + [IB_RNR_TIMER_655_36] = 655360, + [IB_RNR_TIMER_000_01] = 10, + [IB_RNR_TIMER_000_02] = 20, + [IB_RNR_TIMER_000_03] = 30, + [IB_RNR_TIMER_000_04] = 40, + [IB_RNR_TIMER_000_06] = 60, + [IB_RNR_TIMER_000_08] = 80, + [IB_RNR_TIMER_000_12] = 120, + [IB_RNR_TIMER_000_16] = 160, + [IB_RNR_TIMER_000_24] = 240, + [IB_RNR_TIMER_000_32] = 320, + [IB_RNR_TIMER_000_48] = 480, + [IB_RNR_TIMER_000_64] = 640, + [IB_RNR_TIMER_000_96] = 960, + [IB_RNR_TIMER_001_28] = 1280, + [IB_RNR_TIMER_001_92] = 1920, + [IB_RNR_TIMER_002_56] = 2560, + [IB_RNR_TIMER_003_84] = 3840, + [IB_RNR_TIMER_005_12] = 5120, + [IB_RNR_TIMER_007_68] = 7680, + [IB_RNR_TIMER_010_24] = 10240, + [IB_RNR_TIMER_015_36] = 15360, + [IB_RNR_TIMER_020_48] = 20480, + [IB_RNR_TIMER_030_72] = 30720, + [IB_RNR_TIMER_040_96] = 40960, + [IB_RNR_TIMER_061_44] = 61410, + [IB_RNR_TIMER_081_92] = 81920, + [IB_RNR_TIMER_122_88] = 122880, + [IB_RNR_TIMER_163_84] = 163840, + [IB_RNR_TIMER_245_76] = 245760, + [IB_RNR_TIMER_327_68] = 327680, + [IB_RNR_TIMER_491_52] = 491520, +}; + +static inline unsigned long rnrnak_jiffies(u8 timeout) +{ + return max_t(unsigned long, + usecs_to_jiffies(rnrnak_usec[timeout]), 1); +} + +static enum ib_wc_opcode wr_to_wc_opcode(enum ib_wr_opcode opcode) +{ + switch (opcode) { + case IB_WR_RDMA_WRITE: return IB_WC_RDMA_WRITE; + case IB_WR_RDMA_WRITE_WITH_IMM: return IB_WC_RDMA_WRITE; + case IB_WR_SEND: return IB_WC_SEND; + case IB_WR_SEND_WITH_IMM: return IB_WC_SEND; + case IB_WR_RDMA_READ: return IB_WC_RDMA_READ; + case IB_WR_ATOMIC_CMP_AND_SWP: return IB_WC_COMP_SWAP; + case IB_WR_ATOMIC_FETCH_AND_ADD: return IB_WC_FETCH_ADD; + case IB_WR_LSO: return IB_WC_LSO; + case IB_WR_SEND_WITH_INV: return IB_WC_SEND; + case IB_WR_RDMA_READ_WITH_INV: return IB_WC_RDMA_READ; + case IB_WR_LOCAL_INV: return IB_WC_LOCAL_INV; + case IB_WR_REG_MR: return IB_WC_REG_MR; + + default: + return 0xff; + } +} + +void retransmit_timer(unsigned long data) +{ + struct rxe_qp *qp = (struct rxe_qp *)data; + + if (qp->valid) { + qp->comp.timeout = 1; + rxe_run_task(&qp->comp.task, 1); + } +} + +void rxe_comp_queue_pkt(struct rxe_dev *rxe, struct rxe_qp *qp, + struct sk_buff *skb) +{ + int must_sched; + + skb_queue_tail(&qp->resp_pkts, skb); + + must_sched = skb_queue_len(&qp->resp_pkts) > 1; + rxe_run_task(&qp->comp.task, must_sched); +} + +static inline enum comp_state get_wqe(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe **wqe_p) +{ + struct rxe_send_wqe *wqe; + + /* we come here whether or not we found a response packet to see if + * there are any posted WQEs + */ + wqe = queue_head(qp->sq.queue); + *wqe_p = wqe; + + /* no WQE or requester has not started it yet */ + if (!wqe || wqe->state == wqe_state_posted) + return pkt ? COMPST_DONE : COMPST_EXIT; + + /* WQE does not require an ack */ + if (wqe->state == wqe_state_done) + return COMPST_COMP_WQE; + + /* WQE caused an error */ + if (wqe->state == wqe_state_error) + return COMPST_ERROR; + + /* we have a WQE, if we also have an ack check its PSN */ + return pkt ? COMPST_CHECK_PSN : COMPST_EXIT; +} + +static inline void reset_retry_counters(struct rxe_qp *qp) +{ + qp->comp.retry_cnt = qp->attr.retry_cnt; + qp->comp.rnr_retry = qp->attr.rnr_retry; +} + +static inline enum comp_state check_psn(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe *wqe) +{ + s32 diff; + + /* check to see if response is past the oldest WQE. if it is, complete + * send/write or error read/atomic + */ + diff = psn_compare(pkt->psn, wqe->last_psn); + if (diff > 0) { + if (wqe->state == wqe_state_pending) { + if (wqe->mask & WR_ATOMIC_OR_READ_MASK) + return COMPST_ERROR_RETRY; + + reset_retry_counters(qp); + return COMPST_COMP_WQE; + } else { + return COMPST_DONE; + } + } + + /* compare response packet to expected response */ + diff = psn_compare(pkt->psn, qp->comp.psn); + if (diff < 0) { + /* response is most likely a retried packet if it matches an + * uncompleted WQE go complete it else ignore it + */ + if (pkt->psn == wqe->last_psn) + return COMPST_COMP_ACK; + else + return COMPST_DONE; + } else if ((diff > 0) && (wqe->mask & WR_ATOMIC_OR_READ_MASK)) { + return COMPST_ERROR_RETRY; + } else { + return COMPST_CHECK_ACK; + } +} + +static inline enum comp_state check_ack(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe *wqe) +{ + unsigned int mask = pkt->mask; + u8 syn; + + /* Check the sequence only */ + switch (qp->comp.opcode) { + case -1: + /* Will catch all *_ONLY cases. */ + if (!(mask & RXE_START_MASK)) + return COMPST_ERROR; + + break; + + case IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST: + case IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE: + if (pkt->opcode != IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE && + pkt->opcode != IB_OPCODE_RC_RDMA_READ_RESPONSE_LAST) { + return COMPST_ERROR; + } + break; + default: + WARN_ON(1); + } + + /* Check operation validity. */ + switch (pkt->opcode) { + case IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST: + case IB_OPCODE_RC_RDMA_READ_RESPONSE_LAST: + case IB_OPCODE_RC_RDMA_READ_RESPONSE_ONLY: + syn = aeth_syn(pkt); + + if ((syn & AETH_TYPE_MASK) != AETH_ACK) + return COMPST_ERROR; + + /* Fall through (IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE + * doesn't have an AETH) + */ + case IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE: + if (wqe->wr.opcode != IB_WR_RDMA_READ && + wqe->wr.opcode != IB_WR_RDMA_READ_WITH_INV) { + return COMPST_ERROR; + } + reset_retry_counters(qp); + return COMPST_READ; + + case IB_OPCODE_RC_ATOMIC_ACKNOWLEDGE: + syn = aeth_syn(pkt); + + if ((syn & AETH_TYPE_MASK) != AETH_ACK) + return COMPST_ERROR; + + if (wqe->wr.opcode != IB_WR_ATOMIC_CMP_AND_SWP && + wqe->wr.opcode != IB_WR_ATOMIC_FETCH_AND_ADD) + return COMPST_ERROR; + reset_retry_counters(qp); + return COMPST_ATOMIC; + + case IB_OPCODE_RC_ACKNOWLEDGE: + syn = aeth_syn(pkt); + switch (syn & AETH_TYPE_MASK) { + case AETH_ACK: + reset_retry_counters(qp); + return COMPST_WRITE_SEND; + + case AETH_RNR_NAK: + return COMPST_RNR_RETRY; + + case AETH_NAK: + switch (syn) { + case AETH_NAK_PSN_SEQ_ERROR: + /* a nak implicitly acks all packets with psns + * before + */ + if (psn_compare(pkt->psn, qp->comp.psn) > 0) { + qp->comp.psn = pkt->psn; + if (qp->req.wait_psn) { + qp->req.wait_psn = 0; + rxe_run_task(&qp->req.task, 1); + } + } + return COMPST_ERROR_RETRY; + + case AETH_NAK_INVALID_REQ: + wqe->status = IB_WC_REM_INV_REQ_ERR; + return COMPST_ERROR; + + case AETH_NAK_REM_ACC_ERR: + wqe->status = IB_WC_REM_ACCESS_ERR; + return COMPST_ERROR; + + case AETH_NAK_REM_OP_ERR: + wqe->status = IB_WC_REM_OP_ERR; + return COMPST_ERROR; + + default: + pr_warn("unexpected nak %x\n", syn); + wqe->status = IB_WC_REM_OP_ERR; + return COMPST_ERROR; + } + + default: + return COMPST_ERROR; + } + break; + + default: + pr_warn("unexpected opcode\n"); + } + + return COMPST_ERROR; +} + +static inline enum comp_state do_read(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe *wqe) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + int ret; + + ret = copy_data(rxe, qp->pd, IB_ACCESS_LOCAL_WRITE, + &wqe->dma, payload_addr(pkt), + payload_size(pkt), to_mem_obj, NULL); + if (ret) + return COMPST_ERROR; + + if (wqe->dma.resid == 0 && (pkt->mask & RXE_END_MASK)) + return COMPST_COMP_ACK; + else + return COMPST_UPDATE_COMP; +} + +static inline enum comp_state do_atomic(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe *wqe) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + int ret; + + u64 atomic_orig = atmack_orig(pkt); + + ret = copy_data(rxe, qp->pd, IB_ACCESS_LOCAL_WRITE, + &wqe->dma, &atomic_orig, + sizeof(u64), to_mem_obj, NULL); + if (ret) + return COMPST_ERROR; + else + return COMPST_COMP_ACK; +} + +static void make_send_cqe(struct rxe_qp *qp, struct rxe_send_wqe *wqe, + struct rxe_cqe *cqe) +{ + memset(cqe, 0, sizeof(*cqe)); + + if (!qp->is_user) { + struct ib_wc *wc = &cqe->ibwc; + + wc->wr_id = wqe->wr.wr_id; + wc->status = wqe->status; + wc->opcode = wr_to_wc_opcode(wqe->wr.opcode); + if (wqe->wr.opcode == IB_WR_RDMA_WRITE_WITH_IMM || + wqe->wr.opcode == IB_WR_SEND_WITH_IMM) + wc->wc_flags = IB_WC_WITH_IMM; + wc->byte_len = wqe->dma.length; + wc->qp = &qp->ibqp; + } else { + struct ib_uverbs_wc *uwc = &cqe->uibwc; + + uwc->wr_id = wqe->wr.wr_id; + uwc->status = wqe->status; + uwc->opcode = wr_to_wc_opcode(wqe->wr.opcode); + if (wqe->wr.opcode == IB_WR_RDMA_WRITE_WITH_IMM || + wqe->wr.opcode == IB_WR_SEND_WITH_IMM) + uwc->wc_flags = IB_WC_WITH_IMM; + uwc->byte_len = wqe->dma.length; + uwc->qp_num = qp->ibqp.qp_num; + } +} + +static void do_complete(struct rxe_qp *qp, struct rxe_send_wqe *wqe) +{ + struct rxe_cqe cqe; + + if ((qp->sq_sig_type == IB_SIGNAL_ALL_WR) || + (wqe->wr.send_flags & IB_SEND_SIGNALED) || + (qp->req.state == QP_STATE_ERROR)) { + make_send_cqe(qp, wqe, &cqe); + rxe_cq_post(qp->scq, &cqe, 0); + } + + advance_consumer(qp->sq.queue); + + /* + * we completed something so let req run again + * if it is trying to fence + */ + if (qp->req.wait_fence) { + qp->req.wait_fence = 0; + rxe_run_task(&qp->req.task, 1); + } +} + +static inline enum comp_state complete_ack(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe *wqe) +{ + unsigned long flags; + + if (wqe->has_rd_atomic) { + wqe->has_rd_atomic = 0; + atomic_inc(&qp->req.rd_atomic); + if (qp->req.need_rd_atomic) { + qp->comp.timeout_retry = 0; + qp->req.need_rd_atomic = 0; + rxe_run_task(&qp->req.task, 1); + } + } + + if (unlikely(qp->req.state == QP_STATE_DRAIN)) { + /* state_lock used by requester & completer */ + spin_lock_irqsave(&qp->state_lock, flags); + if ((qp->req.state == QP_STATE_DRAIN) && + (qp->comp.psn == qp->req.psn)) { + qp->req.state = QP_STATE_DRAINED; + spin_unlock_irqrestore(&qp->state_lock, flags); + + if (qp->ibqp.event_handler) { + struct ib_event ev; + + ev.device = qp->ibqp.device; + ev.element.qp = &qp->ibqp; + ev.event = IB_EVENT_SQ_DRAINED; + qp->ibqp.event_handler(&ev, + qp->ibqp.qp_context); + } + } else { + spin_unlock_irqrestore(&qp->state_lock, flags); + } + } + + do_complete(qp, wqe); + + if (psn_compare(pkt->psn, qp->comp.psn) >= 0) + return COMPST_UPDATE_COMP; + else + return COMPST_DONE; +} + +static inline enum comp_state complete_wqe(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_send_wqe *wqe) +{ + qp->comp.opcode = -1; + + if (pkt) { + if (psn_compare(pkt->psn, qp->comp.psn) >= 0) + qp->comp.psn = (pkt->psn + 1) & BTH_PSN_MASK; + + if (qp->req.wait_psn) { + qp->req.wait_psn = 0; + rxe_run_task(&qp->req.task, 1); + } + } + + do_complete(qp, wqe); + + return COMPST_GET_WQE; +} + +int rxe_completer(void *arg) +{ + struct rxe_qp *qp = (struct rxe_qp *)arg; + struct rxe_send_wqe *wqe = wqe; + struct sk_buff *skb = NULL; + struct rxe_pkt_info *pkt = NULL; + enum comp_state state; + + if (!qp->valid) { + while ((skb = skb_dequeue(&qp->resp_pkts))) { + rxe_drop_ref(qp); + kfree_skb(skb); + } + skb = NULL; + pkt = NULL; + + while (queue_head(qp->sq.queue)) + advance_consumer(qp->sq.queue); + + goto exit; + } + + if (qp->req.state == QP_STATE_ERROR) { + while ((skb = skb_dequeue(&qp->resp_pkts))) { + rxe_drop_ref(qp); + kfree_skb(skb); + } + skb = NULL; + pkt = NULL; + + while ((wqe = queue_head(qp->sq.queue))) { + wqe->status = IB_WC_WR_FLUSH_ERR; + do_complete(qp, wqe); + } + + goto exit; + } + + if (qp->req.state == QP_STATE_RESET) { + while ((skb = skb_dequeue(&qp->resp_pkts))) { + rxe_drop_ref(qp); + kfree_skb(skb); + } + skb = NULL; + pkt = NULL; + + while (queue_head(qp->sq.queue)) + advance_consumer(qp->sq.queue); + + goto exit; + } + + if (qp->comp.timeout) { + qp->comp.timeout_retry = 1; + qp->comp.timeout = 0; + } else { + qp->comp.timeout_retry = 0; + } + + if (qp->req.need_retry) + goto exit; + + state = COMPST_GET_ACK; + + while (1) { + pr_debug("state = %s\n", comp_state_name[state]); + switch (state) { + case COMPST_GET_ACK: + skb = skb_dequeue(&qp->resp_pkts); + if (skb) { + pkt = SKB_TO_PKT(skb); + qp->comp.timeout_retry = 0; + } + state = COMPST_GET_WQE; + break; + + case COMPST_GET_WQE: + state = get_wqe(qp, pkt, &wqe); + break; + + case COMPST_CHECK_PSN: + state = check_psn(qp, pkt, wqe); + break; + + case COMPST_CHECK_ACK: + state = check_ack(qp, pkt, wqe); + break; + + case COMPST_READ: + state = do_read(qp, pkt, wqe); + break; + + case COMPST_ATOMIC: + state = do_atomic(qp, pkt, wqe); + break; + + case COMPST_WRITE_SEND: + if (wqe->state == wqe_state_pending && + wqe->last_psn == pkt->psn) + state = COMPST_COMP_ACK; + else + state = COMPST_UPDATE_COMP; + break; + + case COMPST_COMP_ACK: + state = complete_ack(qp, pkt, wqe); + break; + + case COMPST_COMP_WQE: + state = complete_wqe(qp, pkt, wqe); + break; + + case COMPST_UPDATE_COMP: + if (pkt->mask & RXE_END_MASK) + qp->comp.opcode = -1; + else + qp->comp.opcode = pkt->opcode; + + if (psn_compare(pkt->psn, qp->comp.psn) >= 0) + qp->comp.psn = (pkt->psn + 1) & BTH_PSN_MASK; + + if (qp->req.wait_psn) { + qp->req.wait_psn = 0; + rxe_run_task(&qp->req.task, 1); + } + + state = COMPST_DONE; + break; + + case COMPST_DONE: + if (pkt) { + rxe_drop_ref(pkt->qp); + kfree_skb(skb); + } + goto done; + + case COMPST_EXIT: + if (qp->comp.timeout_retry && wqe) { + state = COMPST_ERROR_RETRY; + break; + } + + /* re reset the timeout counter if + * (1) QP is type RC + * (2) the QP is alive + * (3) there is a packet sent by the requester that + * might be acked (we still might get spurious + * timeouts but try to keep them as few as possible) + * (4) the timeout parameter is set + */ + if ((qp_type(qp) == IB_QPT_RC) && + (qp->req.state == QP_STATE_READY) && + (psn_compare(qp->req.psn, qp->comp.psn) > 0) && + qp->qp_timeout_jiffies) + mod_timer(&qp->retrans_timer, + jiffies + qp->qp_timeout_jiffies); + goto exit; + + case COMPST_ERROR_RETRY: + /* we come here if the retry timer fired and we did + * not receive a response packet. try to retry the send + * queue if that makes sense and the limits have not + * been exceeded. remember that some timeouts are + * spurious since we do not reset the timer but kick + * it down the road or let it expire + */ + + /* there is nothing to retry in this case */ + if (!wqe || (wqe->state == wqe_state_posted)) + goto exit; + + if (qp->comp.retry_cnt > 0) { + if (qp->comp.retry_cnt != 7) + qp->comp.retry_cnt--; + + /* no point in retrying if we have already + * seen the last ack that the requester could + * have caused + */ + if (psn_compare(qp->req.psn, + qp->comp.psn) > 0) { + /* tell the requester to retry the + * send send queue next time around + */ + qp->req.need_retry = 1; + rxe_run_task(&qp->req.task, 1); + } + goto exit; + } else { + wqe->status = IB_WC_RETRY_EXC_ERR; + state = COMPST_ERROR; + } + break; + + case COMPST_RNR_RETRY: + if (qp->comp.rnr_retry > 0) { + if (qp->comp.rnr_retry != 7) + qp->comp.rnr_retry--; + + qp->req.need_retry = 1; + pr_debug("set rnr nak timer\n"); + mod_timer(&qp->rnr_nak_timer, + jiffies + rnrnak_jiffies(aeth_syn(pkt) + & ~AETH_TYPE_MASK)); + goto exit; + } else { + wqe->status = IB_WC_RNR_RETRY_EXC_ERR; + state = COMPST_ERROR; + } + break; + + case COMPST_ERROR: + do_complete(qp, wqe); + rxe_qp_error(qp); + goto exit; + } + } + +exit: + /* we come here if we are done with processing and want the task to + * exit from the loop calling us + */ + return -EAGAIN; + +done: + /* we come here if we have processed a packet we want the task to call + * us again to see if there is anything else to do + */ + return 0; +} diff --git a/drivers/infiniband/sw/rxe/rxe_cq.c b/drivers/infiniband/sw/rxe/rxe_cq.c new file mode 100644 index 000000000000..e5e6a5e7dee9 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_cq.c @@ -0,0 +1,165 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +int rxe_cq_chk_attr(struct rxe_dev *rxe, struct rxe_cq *cq, + int cqe, int comp_vector, struct ib_udata *udata) +{ + int count; + + if (cqe <= 0) { + pr_warn("cqe(%d) <= 0\n", cqe); + goto err1; + } + + if (cqe > rxe->attr.max_cqe) { + pr_warn("cqe(%d) > max_cqe(%d)\n", + cqe, rxe->attr.max_cqe); + goto err1; + } + + if (cq) { + count = queue_count(cq->queue); + if (cqe < count) { + pr_warn("cqe(%d) < current # elements in queue (%d)", + cqe, count); + goto err1; + } + } + + return 0; + +err1: + return -EINVAL; +} + +static void rxe_send_complete(unsigned long data) +{ + struct rxe_cq *cq = (struct rxe_cq *)data; + + cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context); +} + +int rxe_cq_from_init(struct rxe_dev *rxe, struct rxe_cq *cq, int cqe, + int comp_vector, struct ib_ucontext *context, + struct ib_udata *udata) +{ + int err; + + cq->queue = rxe_queue_init(rxe, &cqe, + sizeof(struct rxe_cqe)); + if (!cq->queue) { + pr_warn("unable to create cq\n"); + return -ENOMEM; + } + + err = do_mmap_info(rxe, udata, false, context, cq->queue->buf, + cq->queue->buf_size, &cq->queue->ip); + if (err) { + kvfree(cq->queue->buf); + kfree(cq->queue); + return err; + } + + if (udata) + cq->is_user = 1; + + tasklet_init(&cq->comp_task, rxe_send_complete, (unsigned long)cq); + + spin_lock_init(&cq->cq_lock); + cq->ibcq.cqe = cqe; + return 0; +} + +int rxe_cq_resize_queue(struct rxe_cq *cq, int cqe, struct ib_udata *udata) +{ + int err; + + err = rxe_queue_resize(cq->queue, (unsigned int *)&cqe, + sizeof(struct rxe_cqe), + cq->queue->ip ? cq->queue->ip->context : NULL, + udata, NULL, &cq->cq_lock); + if (!err) + cq->ibcq.cqe = cqe; + + return err; +} + +int rxe_cq_post(struct rxe_cq *cq, struct rxe_cqe *cqe, int solicited) +{ + struct ib_event ev; + unsigned long flags; + + spin_lock_irqsave(&cq->cq_lock, flags); + + if (unlikely(queue_full(cq->queue))) { + spin_unlock_irqrestore(&cq->cq_lock, flags); + if (cq->ibcq.event_handler) { + ev.device = cq->ibcq.device; + ev.element.cq = &cq->ibcq; + ev.event = IB_EVENT_CQ_ERR; + cq->ibcq.event_handler(&ev, cq->ibcq.cq_context); + } + + return -EBUSY; + } + + memcpy(producer_addr(cq->queue), cqe, sizeof(*cqe)); + + /* make sure all changes to the CQ are written before we update the + * producer pointer + */ + smp_wmb(); + + advance_producer(cq->queue); + spin_unlock_irqrestore(&cq->cq_lock, flags); + + if ((cq->notify == IB_CQ_NEXT_COMP) || + (cq->notify == IB_CQ_SOLICITED && solicited)) { + cq->notify = 0; + tasklet_schedule(&cq->comp_task); + } + + return 0; +} + +void rxe_cq_cleanup(void *arg) +{ + struct rxe_cq *cq = arg; + + if (cq->queue) + rxe_queue_cleanup(cq->queue); +} diff --git a/drivers/infiniband/sw/rxe/rxe_dma.c b/drivers/infiniband/sw/rxe/rxe_dma.c new file mode 100644 index 000000000000..7634c1a81b2b --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_dma.c @@ -0,0 +1,166 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +#define DMA_BAD_ADDER ((u64)0) + +static int rxe_mapping_error(struct ib_device *dev, u64 dma_addr) +{ + return dma_addr == DMA_BAD_ADDER; +} + +static u64 rxe_dma_map_single(struct ib_device *dev, + void *cpu_addr, size_t size, + enum dma_data_direction direction) +{ + WARN_ON(!valid_dma_direction(direction)); + return (uintptr_t)cpu_addr; +} + +static void rxe_dma_unmap_single(struct ib_device *dev, + u64 addr, size_t size, + enum dma_data_direction direction) +{ + WARN_ON(!valid_dma_direction(direction)); +} + +static u64 rxe_dma_map_page(struct ib_device *dev, + struct page *page, + unsigned long offset, + size_t size, enum dma_data_direction direction) +{ + u64 addr; + + WARN_ON(!valid_dma_direction(direction)); + + if (offset + size > PAGE_SIZE) { + addr = DMA_BAD_ADDER; + goto done; + } + + addr = (uintptr_t)page_address(page); + if (addr) + addr += offset; + +done: + return addr; +} + +static void rxe_dma_unmap_page(struct ib_device *dev, + u64 addr, size_t size, + enum dma_data_direction direction) +{ + WARN_ON(!valid_dma_direction(direction)); +} + +static int rxe_map_sg(struct ib_device *dev, struct scatterlist *sgl, + int nents, enum dma_data_direction direction) +{ + struct scatterlist *sg; + u64 addr; + int i; + int ret = nents; + + WARN_ON(!valid_dma_direction(direction)); + + for_each_sg(sgl, sg, nents, i) { + addr = (uintptr_t)page_address(sg_page(sg)); + if (!addr) { + ret = 0; + break; + } + sg->dma_address = addr + sg->offset; +#ifdef CONFIG_NEED_SG_DMA_LENGTH + sg->dma_length = sg->length; +#endif + } + + return ret; +} + +static void rxe_unmap_sg(struct ib_device *dev, + struct scatterlist *sg, int nents, + enum dma_data_direction direction) +{ + WARN_ON(!valid_dma_direction(direction)); +} + +static void rxe_sync_single_for_cpu(struct ib_device *dev, + u64 addr, + size_t size, enum dma_data_direction dir) +{ +} + +static void rxe_sync_single_for_device(struct ib_device *dev, + u64 addr, + size_t size, enum dma_data_direction dir) +{ +} + +static void *rxe_dma_alloc_coherent(struct ib_device *dev, size_t size, + u64 *dma_handle, gfp_t flag) +{ + struct page *p; + void *addr = NULL; + + p = alloc_pages(flag, get_order(size)); + if (p) + addr = page_address(p); + + if (dma_handle) + *dma_handle = (uintptr_t)addr; + + return addr; +} + +static void rxe_dma_free_coherent(struct ib_device *dev, size_t size, + void *cpu_addr, u64 dma_handle) +{ + free_pages((unsigned long)cpu_addr, get_order(size)); +} + +struct ib_dma_mapping_ops rxe_dma_mapping_ops = { + .mapping_error = rxe_mapping_error, + .map_single = rxe_dma_map_single, + .unmap_single = rxe_dma_unmap_single, + .map_page = rxe_dma_map_page, + .unmap_page = rxe_dma_unmap_page, + .map_sg = rxe_map_sg, + .unmap_sg = rxe_unmap_sg, + .sync_single_for_cpu = rxe_sync_single_for_cpu, + .sync_single_for_device = rxe_sync_single_for_device, + .alloc_coherent = rxe_dma_alloc_coherent, + .free_coherent = rxe_dma_free_coherent +}; diff --git a/drivers/infiniband/sw/rxe/rxe_hdr.h b/drivers/infiniband/sw/rxe/rxe_hdr.h new file mode 100644 index 000000000000..d57b5e956ceb --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_hdr.h @@ -0,0 +1,952 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_HDR_H +#define RXE_HDR_H + +/* extracted information about a packet carried in an sk_buff struct fits in + * the skbuff cb array. Must be at most 48 bytes. stored in control block of + * sk_buff for received packets. + */ +struct rxe_pkt_info { + struct rxe_dev *rxe; /* device that owns packet */ + struct rxe_qp *qp; /* qp that owns packet */ + struct rxe_send_wqe *wqe; /* send wqe */ + u8 *hdr; /* points to bth */ + u32 mask; /* useful info about pkt */ + u32 psn; /* bth psn of packet */ + u16 pkey_index; /* partition of pkt */ + u16 paylen; /* length of bth - icrc */ + u8 port_num; /* port pkt received on */ + u8 opcode; /* bth opcode of packet */ + u8 offset; /* bth offset from pkt->hdr */ +}; + +/* Macros should be used only for received skb */ +#define SKB_TO_PKT(skb) ((struct rxe_pkt_info *)(skb)->cb) +#define PKT_TO_SKB(pkt) container_of((void *)(pkt), struct sk_buff, cb) + +/* + * IBA header types and methods + * + * Some of these are for reference and completeness only since + * rxe does not currently support RD transport + * most of this could be moved into IB core. ib_pack.h has + * part of this but is incomplete + * + * Header specific routines to insert/extract values to/from headers + * the routines that are named __hhh_(set_)fff() take a pointer to a + * hhh header and get(set) the fff field. The routines named + * hhh_(set_)fff take a packet info struct and find the + * header and field based on the opcode in the packet. + * Conversion to/from network byte order from cpu order is also done. + */ + +#define RXE_ICRC_SIZE (4) +#define RXE_MAX_HDR_LENGTH (80) + +/****************************************************************************** + * Base Transport Header + ******************************************************************************/ +struct rxe_bth { + u8 opcode; + u8 flags; + __be16 pkey; + __be32 qpn; + __be32 apsn; +}; + +#define BTH_TVER (0) +#define BTH_DEF_PKEY (0xffff) + +#define BTH_SE_MASK (0x80) +#define BTH_MIG_MASK (0x40) +#define BTH_PAD_MASK (0x30) +#define BTH_TVER_MASK (0x0f) +#define BTH_FECN_MASK (0x80000000) +#define BTH_BECN_MASK (0x40000000) +#define BTH_RESV6A_MASK (0x3f000000) +#define BTH_QPN_MASK (0x00ffffff) +#define BTH_ACK_MASK (0x80000000) +#define BTH_RESV7_MASK (0x7f000000) +#define BTH_PSN_MASK (0x00ffffff) + +static inline u8 __bth_opcode(void *arg) +{ + struct rxe_bth *bth = arg; + + return bth->opcode; +} + +static inline void __bth_set_opcode(void *arg, u8 opcode) +{ + struct rxe_bth *bth = arg; + + bth->opcode = opcode; +} + +static inline u8 __bth_se(void *arg) +{ + struct rxe_bth *bth = arg; + + return 0 != (BTH_SE_MASK & bth->flags); +} + +static inline void __bth_set_se(void *arg, int se) +{ + struct rxe_bth *bth = arg; + + if (se) + bth->flags |= BTH_SE_MASK; + else + bth->flags &= ~BTH_SE_MASK; +} + +static inline u8 __bth_mig(void *arg) +{ + struct rxe_bth *bth = arg; + + return 0 != (BTH_MIG_MASK & bth->flags); +} + +static inline void __bth_set_mig(void *arg, u8 mig) +{ + struct rxe_bth *bth = arg; + + if (mig) + bth->flags |= BTH_MIG_MASK; + else + bth->flags &= ~BTH_MIG_MASK; +} + +static inline u8 __bth_pad(void *arg) +{ + struct rxe_bth *bth = arg; + + return (BTH_PAD_MASK & bth->flags) >> 4; +} + +static inline void __bth_set_pad(void *arg, u8 pad) +{ + struct rxe_bth *bth = arg; + + bth->flags = (BTH_PAD_MASK & (pad << 4)) | + (~BTH_PAD_MASK & bth->flags); +} + +static inline u8 __bth_tver(void *arg) +{ + struct rxe_bth *bth = arg; + + return BTH_TVER_MASK & bth->flags; +} + +static inline void __bth_set_tver(void *arg, u8 tver) +{ + struct rxe_bth *bth = arg; + + bth->flags = (BTH_TVER_MASK & tver) | + (~BTH_TVER_MASK & bth->flags); +} + +static inline u16 __bth_pkey(void *arg) +{ + struct rxe_bth *bth = arg; + + return be16_to_cpu(bth->pkey); +} + +static inline void __bth_set_pkey(void *arg, u16 pkey) +{ + struct rxe_bth *bth = arg; + + bth->pkey = cpu_to_be16(pkey); +} + +static inline u32 __bth_qpn(void *arg) +{ + struct rxe_bth *bth = arg; + + return BTH_QPN_MASK & be32_to_cpu(bth->qpn); +} + +static inline void __bth_set_qpn(void *arg, u32 qpn) +{ + struct rxe_bth *bth = arg; + u32 resvqpn = be32_to_cpu(bth->qpn); + + bth->qpn = cpu_to_be32((BTH_QPN_MASK & qpn) | + (~BTH_QPN_MASK & resvqpn)); +} + +static inline int __bth_fecn(void *arg) +{ + struct rxe_bth *bth = arg; + + return 0 != (cpu_to_be32(BTH_FECN_MASK) & bth->qpn); +} + +static inline void __bth_set_fecn(void *arg, int fecn) +{ + struct rxe_bth *bth = arg; + + if (fecn) + bth->qpn |= cpu_to_be32(BTH_FECN_MASK); + else + bth->qpn &= ~cpu_to_be32(BTH_FECN_MASK); +} + +static inline int __bth_becn(void *arg) +{ + struct rxe_bth *bth = arg; + + return 0 != (cpu_to_be32(BTH_BECN_MASK) & bth->qpn); +} + +static inline void __bth_set_becn(void *arg, int becn) +{ + struct rxe_bth *bth = arg; + + if (becn) + bth->qpn |= cpu_to_be32(BTH_BECN_MASK); + else + bth->qpn &= ~cpu_to_be32(BTH_BECN_MASK); +} + +static inline u8 __bth_resv6a(void *arg) +{ + struct rxe_bth *bth = arg; + + return (BTH_RESV6A_MASK & be32_to_cpu(bth->qpn)) >> 24; +} + +static inline void __bth_set_resv6a(void *arg) +{ + struct rxe_bth *bth = arg; + + bth->qpn = cpu_to_be32(~BTH_RESV6A_MASK); +} + +static inline int __bth_ack(void *arg) +{ + struct rxe_bth *bth = arg; + + return 0 != (cpu_to_be32(BTH_ACK_MASK) & bth->apsn); +} + +static inline void __bth_set_ack(void *arg, int ack) +{ + struct rxe_bth *bth = arg; + + if (ack) + bth->apsn |= cpu_to_be32(BTH_ACK_MASK); + else + bth->apsn &= ~cpu_to_be32(BTH_ACK_MASK); +} + +static inline void __bth_set_resv7(void *arg) +{ + struct rxe_bth *bth = arg; + + bth->apsn &= ~cpu_to_be32(BTH_RESV7_MASK); +} + +static inline u32 __bth_psn(void *arg) +{ + struct rxe_bth *bth = arg; + + return BTH_PSN_MASK & be32_to_cpu(bth->apsn); +} + +static inline void __bth_set_psn(void *arg, u32 psn) +{ + struct rxe_bth *bth = arg; + u32 apsn = be32_to_cpu(bth->apsn); + + bth->apsn = cpu_to_be32((BTH_PSN_MASK & psn) | + (~BTH_PSN_MASK & apsn)); +} + +static inline u8 bth_opcode(struct rxe_pkt_info *pkt) +{ + return __bth_opcode(pkt->hdr + pkt->offset); +} + +static inline void bth_set_opcode(struct rxe_pkt_info *pkt, u8 opcode) +{ + __bth_set_opcode(pkt->hdr + pkt->offset, opcode); +} + +static inline u8 bth_se(struct rxe_pkt_info *pkt) +{ + return __bth_se(pkt->hdr + pkt->offset); +} + +static inline void bth_set_se(struct rxe_pkt_info *pkt, int se) +{ + __bth_set_se(pkt->hdr + pkt->offset, se); +} + +static inline u8 bth_mig(struct rxe_pkt_info *pkt) +{ + return __bth_mig(pkt->hdr + pkt->offset); +} + +static inline void bth_set_mig(struct rxe_pkt_info *pkt, u8 mig) +{ + __bth_set_mig(pkt->hdr + pkt->offset, mig); +} + +static inline u8 bth_pad(struct rxe_pkt_info *pkt) +{ + return __bth_pad(pkt->hdr + pkt->offset); +} + +static inline void bth_set_pad(struct rxe_pkt_info *pkt, u8 pad) +{ + __bth_set_pad(pkt->hdr + pkt->offset, pad); +} + +static inline u8 bth_tver(struct rxe_pkt_info *pkt) +{ + return __bth_tver(pkt->hdr + pkt->offset); +} + +static inline void bth_set_tver(struct rxe_pkt_info *pkt, u8 tver) +{ + __bth_set_tver(pkt->hdr + pkt->offset, tver); +} + +static inline u16 bth_pkey(struct rxe_pkt_info *pkt) +{ + return __bth_pkey(pkt->hdr + pkt->offset); +} + +static inline void bth_set_pkey(struct rxe_pkt_info *pkt, u16 pkey) +{ + __bth_set_pkey(pkt->hdr + pkt->offset, pkey); +} + +static inline u32 bth_qpn(struct rxe_pkt_info *pkt) +{ + return __bth_qpn(pkt->hdr + pkt->offset); +} + +static inline void bth_set_qpn(struct rxe_pkt_info *pkt, u32 qpn) +{ + __bth_set_qpn(pkt->hdr + pkt->offset, qpn); +} + +static inline int bth_fecn(struct rxe_pkt_info *pkt) +{ + return __bth_fecn(pkt->hdr + pkt->offset); +} + +static inline void bth_set_fecn(struct rxe_pkt_info *pkt, int fecn) +{ + __bth_set_fecn(pkt->hdr + pkt->offset, fecn); +} + +static inline int bth_becn(struct rxe_pkt_info *pkt) +{ + return __bth_becn(pkt->hdr + pkt->offset); +} + +static inline void bth_set_becn(struct rxe_pkt_info *pkt, int becn) +{ + __bth_set_becn(pkt->hdr + pkt->offset, becn); +} + +static inline u8 bth_resv6a(struct rxe_pkt_info *pkt) +{ + return __bth_resv6a(pkt->hdr + pkt->offset); +} + +static inline void bth_set_resv6a(struct rxe_pkt_info *pkt) +{ + __bth_set_resv6a(pkt->hdr + pkt->offset); +} + +static inline int bth_ack(struct rxe_pkt_info *pkt) +{ + return __bth_ack(pkt->hdr + pkt->offset); +} + +static inline void bth_set_ack(struct rxe_pkt_info *pkt, int ack) +{ + __bth_set_ack(pkt->hdr + pkt->offset, ack); +} + +static inline void bth_set_resv7(struct rxe_pkt_info *pkt) +{ + __bth_set_resv7(pkt->hdr + pkt->offset); +} + +static inline u32 bth_psn(struct rxe_pkt_info *pkt) +{ + return __bth_psn(pkt->hdr + pkt->offset); +} + +static inline void bth_set_psn(struct rxe_pkt_info *pkt, u32 psn) +{ + __bth_set_psn(pkt->hdr + pkt->offset, psn); +} + +static inline void bth_init(struct rxe_pkt_info *pkt, u8 opcode, int se, + int mig, int pad, u16 pkey, u32 qpn, int ack_req, + u32 psn) +{ + struct rxe_bth *bth = (struct rxe_bth *)(pkt->hdr + pkt->offset); + + bth->opcode = opcode; + bth->flags = (pad << 4) & BTH_PAD_MASK; + if (se) + bth->flags |= BTH_SE_MASK; + if (mig) + bth->flags |= BTH_MIG_MASK; + bth->pkey = cpu_to_be16(pkey); + bth->qpn = cpu_to_be32(qpn & BTH_QPN_MASK); + psn &= BTH_PSN_MASK; + if (ack_req) + psn |= BTH_ACK_MASK; + bth->apsn = cpu_to_be32(psn); +} + +/****************************************************************************** + * Reliable Datagram Extended Transport Header + ******************************************************************************/ +struct rxe_rdeth { + __be32 een; +}; + +#define RDETH_EEN_MASK (0x00ffffff) + +static inline u8 __rdeth_een(void *arg) +{ + struct rxe_rdeth *rdeth = arg; + + return RDETH_EEN_MASK & be32_to_cpu(rdeth->een); +} + +static inline void __rdeth_set_een(void *arg, u32 een) +{ + struct rxe_rdeth *rdeth = arg; + + rdeth->een = cpu_to_be32(RDETH_EEN_MASK & een); +} + +static inline u8 rdeth_een(struct rxe_pkt_info *pkt) +{ + return __rdeth_een(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RDETH]); +} + +static inline void rdeth_set_een(struct rxe_pkt_info *pkt, u32 een) +{ + __rdeth_set_een(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RDETH], een); +} + +/****************************************************************************** + * Datagram Extended Transport Header + ******************************************************************************/ +struct rxe_deth { + __be32 qkey; + __be32 sqp; +}; + +#define GSI_QKEY (0x80010000) +#define DETH_SQP_MASK (0x00ffffff) + +static inline u32 __deth_qkey(void *arg) +{ + struct rxe_deth *deth = arg; + + return be32_to_cpu(deth->qkey); +} + +static inline void __deth_set_qkey(void *arg, u32 qkey) +{ + struct rxe_deth *deth = arg; + + deth->qkey = cpu_to_be32(qkey); +} + +static inline u32 __deth_sqp(void *arg) +{ + struct rxe_deth *deth = arg; + + return DETH_SQP_MASK & be32_to_cpu(deth->sqp); +} + +static inline void __deth_set_sqp(void *arg, u32 sqp) +{ + struct rxe_deth *deth = arg; + + deth->sqp = cpu_to_be32(DETH_SQP_MASK & sqp); +} + +static inline u32 deth_qkey(struct rxe_pkt_info *pkt) +{ + return __deth_qkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_DETH]); +} + +static inline void deth_set_qkey(struct rxe_pkt_info *pkt, u32 qkey) +{ + __deth_set_qkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_DETH], qkey); +} + +static inline u32 deth_sqp(struct rxe_pkt_info *pkt) +{ + return __deth_sqp(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_DETH]); +} + +static inline void deth_set_sqp(struct rxe_pkt_info *pkt, u32 sqp) +{ + __deth_set_sqp(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_DETH], sqp); +} + +/****************************************************************************** + * RDMA Extended Transport Header + ******************************************************************************/ +struct rxe_reth { + __be64 va; + __be32 rkey; + __be32 len; +}; + +static inline u64 __reth_va(void *arg) +{ + struct rxe_reth *reth = arg; + + return be64_to_cpu(reth->va); +} + +static inline void __reth_set_va(void *arg, u64 va) +{ + struct rxe_reth *reth = arg; + + reth->va = cpu_to_be64(va); +} + +static inline u32 __reth_rkey(void *arg) +{ + struct rxe_reth *reth = arg; + + return be32_to_cpu(reth->rkey); +} + +static inline void __reth_set_rkey(void *arg, u32 rkey) +{ + struct rxe_reth *reth = arg; + + reth->rkey = cpu_to_be32(rkey); +} + +static inline u32 __reth_len(void *arg) +{ + struct rxe_reth *reth = arg; + + return be32_to_cpu(reth->len); +} + +static inline void __reth_set_len(void *arg, u32 len) +{ + struct rxe_reth *reth = arg; + + reth->len = cpu_to_be32(len); +} + +static inline u64 reth_va(struct rxe_pkt_info *pkt) +{ + return __reth_va(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RETH]); +} + +static inline void reth_set_va(struct rxe_pkt_info *pkt, u64 va) +{ + __reth_set_va(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RETH], va); +} + +static inline u32 reth_rkey(struct rxe_pkt_info *pkt) +{ + return __reth_rkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RETH]); +} + +static inline void reth_set_rkey(struct rxe_pkt_info *pkt, u32 rkey) +{ + __reth_set_rkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RETH], rkey); +} + +static inline u32 reth_len(struct rxe_pkt_info *pkt) +{ + return __reth_len(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RETH]); +} + +static inline void reth_set_len(struct rxe_pkt_info *pkt, u32 len) +{ + __reth_set_len(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_RETH], len); +} + +/****************************************************************************** + * Atomic Extended Transport Header + ******************************************************************************/ +struct rxe_atmeth { + __be64 va; + __be32 rkey; + __be64 swap_add; + __be64 comp; +} __attribute__((__packed__)); + +static inline u64 __atmeth_va(void *arg) +{ + struct rxe_atmeth *atmeth = arg; + + return be64_to_cpu(atmeth->va); +} + +static inline void __atmeth_set_va(void *arg, u64 va) +{ + struct rxe_atmeth *atmeth = arg; + + atmeth->va = cpu_to_be64(va); +} + +static inline u32 __atmeth_rkey(void *arg) +{ + struct rxe_atmeth *atmeth = arg; + + return be32_to_cpu(atmeth->rkey); +} + +static inline void __atmeth_set_rkey(void *arg, u32 rkey) +{ + struct rxe_atmeth *atmeth = arg; + + atmeth->rkey = cpu_to_be32(rkey); +} + +static inline u64 __atmeth_swap_add(void *arg) +{ + struct rxe_atmeth *atmeth = arg; + + return be64_to_cpu(atmeth->swap_add); +} + +static inline void __atmeth_set_swap_add(void *arg, u64 swap_add) +{ + struct rxe_atmeth *atmeth = arg; + + atmeth->swap_add = cpu_to_be64(swap_add); +} + +static inline u64 __atmeth_comp(void *arg) +{ + struct rxe_atmeth *atmeth = arg; + + return be64_to_cpu(atmeth->comp); +} + +static inline void __atmeth_set_comp(void *arg, u64 comp) +{ + struct rxe_atmeth *atmeth = arg; + + atmeth->comp = cpu_to_be64(comp); +} + +static inline u64 atmeth_va(struct rxe_pkt_info *pkt) +{ + return __atmeth_va(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH]); +} + +static inline void atmeth_set_va(struct rxe_pkt_info *pkt, u64 va) +{ + __atmeth_set_va(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH], va); +} + +static inline u32 atmeth_rkey(struct rxe_pkt_info *pkt) +{ + return __atmeth_rkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH]); +} + +static inline void atmeth_set_rkey(struct rxe_pkt_info *pkt, u32 rkey) +{ + __atmeth_set_rkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH], rkey); +} + +static inline u64 atmeth_swap_add(struct rxe_pkt_info *pkt) +{ + return __atmeth_swap_add(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH]); +} + +static inline void atmeth_set_swap_add(struct rxe_pkt_info *pkt, u64 swap_add) +{ + __atmeth_set_swap_add(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH], swap_add); +} + +static inline u64 atmeth_comp(struct rxe_pkt_info *pkt) +{ + return __atmeth_comp(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH]); +} + +static inline void atmeth_set_comp(struct rxe_pkt_info *pkt, u64 comp) +{ + __atmeth_set_comp(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMETH], comp); +} + +/****************************************************************************** + * Ack Extended Transport Header + ******************************************************************************/ +struct rxe_aeth { + __be32 smsn; +}; + +#define AETH_SYN_MASK (0xff000000) +#define AETH_MSN_MASK (0x00ffffff) + +enum aeth_syndrome { + AETH_TYPE_MASK = 0xe0, + AETH_ACK = 0x00, + AETH_RNR_NAK = 0x20, + AETH_RSVD = 0x40, + AETH_NAK = 0x60, + AETH_ACK_UNLIMITED = 0x1f, + AETH_NAK_PSN_SEQ_ERROR = 0x60, + AETH_NAK_INVALID_REQ = 0x61, + AETH_NAK_REM_ACC_ERR = 0x62, + AETH_NAK_REM_OP_ERR = 0x63, + AETH_NAK_INV_RD_REQ = 0x64, +}; + +static inline u8 __aeth_syn(void *arg) +{ + struct rxe_aeth *aeth = arg; + + return (AETH_SYN_MASK & be32_to_cpu(aeth->smsn)) >> 24; +} + +static inline void __aeth_set_syn(void *arg, u8 syn) +{ + struct rxe_aeth *aeth = arg; + u32 smsn = be32_to_cpu(aeth->smsn); + + aeth->smsn = cpu_to_be32((AETH_SYN_MASK & (syn << 24)) | + (~AETH_SYN_MASK & smsn)); +} + +static inline u32 __aeth_msn(void *arg) +{ + struct rxe_aeth *aeth = arg; + + return AETH_MSN_MASK & be32_to_cpu(aeth->smsn); +} + +static inline void __aeth_set_msn(void *arg, u32 msn) +{ + struct rxe_aeth *aeth = arg; + u32 smsn = be32_to_cpu(aeth->smsn); + + aeth->smsn = cpu_to_be32((AETH_MSN_MASK & msn) | + (~AETH_MSN_MASK & smsn)); +} + +static inline u8 aeth_syn(struct rxe_pkt_info *pkt) +{ + return __aeth_syn(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_AETH]); +} + +static inline void aeth_set_syn(struct rxe_pkt_info *pkt, u8 syn) +{ + __aeth_set_syn(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_AETH], syn); +} + +static inline u32 aeth_msn(struct rxe_pkt_info *pkt) +{ + return __aeth_msn(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_AETH]); +} + +static inline void aeth_set_msn(struct rxe_pkt_info *pkt, u32 msn) +{ + __aeth_set_msn(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_AETH], msn); +} + +/****************************************************************************** + * Atomic Ack Extended Transport Header + ******************************************************************************/ +struct rxe_atmack { + __be64 orig; +}; + +static inline u64 __atmack_orig(void *arg) +{ + struct rxe_atmack *atmack = arg; + + return be64_to_cpu(atmack->orig); +} + +static inline void __atmack_set_orig(void *arg, u64 orig) +{ + struct rxe_atmack *atmack = arg; + + atmack->orig = cpu_to_be64(orig); +} + +static inline u64 atmack_orig(struct rxe_pkt_info *pkt) +{ + return __atmack_orig(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMACK]); +} + +static inline void atmack_set_orig(struct rxe_pkt_info *pkt, u64 orig) +{ + __atmack_set_orig(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_ATMACK], orig); +} + +/****************************************************************************** + * Immediate Extended Transport Header + ******************************************************************************/ +struct rxe_immdt { + __be32 imm; +}; + +static inline __be32 __immdt_imm(void *arg) +{ + struct rxe_immdt *immdt = arg; + + return immdt->imm; +} + +static inline void __immdt_set_imm(void *arg, __be32 imm) +{ + struct rxe_immdt *immdt = arg; + + immdt->imm = imm; +} + +static inline __be32 immdt_imm(struct rxe_pkt_info *pkt) +{ + return __immdt_imm(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_IMMDT]); +} + +static inline void immdt_set_imm(struct rxe_pkt_info *pkt, __be32 imm) +{ + __immdt_set_imm(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_IMMDT], imm); +} + +/****************************************************************************** + * Invalidate Extended Transport Header + ******************************************************************************/ +struct rxe_ieth { + __be32 rkey; +}; + +static inline u32 __ieth_rkey(void *arg) +{ + struct rxe_ieth *ieth = arg; + + return be32_to_cpu(ieth->rkey); +} + +static inline void __ieth_set_rkey(void *arg, u32 rkey) +{ + struct rxe_ieth *ieth = arg; + + ieth->rkey = cpu_to_be32(rkey); +} + +static inline u32 ieth_rkey(struct rxe_pkt_info *pkt) +{ + return __ieth_rkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_IETH]); +} + +static inline void ieth_set_rkey(struct rxe_pkt_info *pkt, u32 rkey) +{ + __ieth_set_rkey(pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_IETH], rkey); +} + +enum rxe_hdr_length { + RXE_BTH_BYTES = sizeof(struct rxe_bth), + RXE_DETH_BYTES = sizeof(struct rxe_deth), + RXE_IMMDT_BYTES = sizeof(struct rxe_immdt), + RXE_RETH_BYTES = sizeof(struct rxe_reth), + RXE_AETH_BYTES = sizeof(struct rxe_aeth), + RXE_ATMACK_BYTES = sizeof(struct rxe_atmack), + RXE_ATMETH_BYTES = sizeof(struct rxe_atmeth), + RXE_IETH_BYTES = sizeof(struct rxe_ieth), + RXE_RDETH_BYTES = sizeof(struct rxe_rdeth), +}; + +static inline size_t header_size(struct rxe_pkt_info *pkt) +{ + return pkt->offset + rxe_opcode[pkt->opcode].length; +} + +static inline void *payload_addr(struct rxe_pkt_info *pkt) +{ + return pkt->hdr + pkt->offset + + rxe_opcode[pkt->opcode].offset[RXE_PAYLOAD]; +} + +static inline size_t payload_size(struct rxe_pkt_info *pkt) +{ + return pkt->paylen - rxe_opcode[pkt->opcode].offset[RXE_PAYLOAD] + - bth_pad(pkt) - RXE_ICRC_SIZE; +} + +#endif /* RXE_HDR_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_icrc.c b/drivers/infiniband/sw/rxe/rxe_icrc.c new file mode 100644 index 000000000000..413b56b23a06 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_icrc.c @@ -0,0 +1,96 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +/* Compute a partial ICRC for all the IB transport headers. */ +u32 rxe_icrc_hdr(struct rxe_pkt_info *pkt, struct sk_buff *skb) +{ + unsigned int bth_offset = 0; + struct iphdr *ip4h = NULL; + struct ipv6hdr *ip6h = NULL; + struct udphdr *udph; + struct rxe_bth *bth; + int crc; + int length; + int hdr_size = sizeof(struct udphdr) + + (skb->protocol == htons(ETH_P_IP) ? + sizeof(struct iphdr) : sizeof(struct ipv6hdr)); + /* pseudo header buffer size is calculate using ipv6 header size since + * it is bigger than ipv4 + */ + u8 pshdr[sizeof(struct udphdr) + + sizeof(struct ipv6hdr) + + RXE_BTH_BYTES]; + + /* This seed is the result of computing a CRC with a seed of + * 0xfffffff and 8 bytes of 0xff representing a masked LRH. + */ + crc = 0xdebb20e3; + + if (skb->protocol == htons(ETH_P_IP)) { /* IPv4 */ + memcpy(pshdr, ip_hdr(skb), hdr_size); + ip4h = (struct iphdr *)pshdr; + udph = (struct udphdr *)(ip4h + 1); + + ip4h->ttl = 0xff; + ip4h->check = CSUM_MANGLED_0; + ip4h->tos = 0xff; + } else { /* IPv6 */ + memcpy(pshdr, ipv6_hdr(skb), hdr_size); + ip6h = (struct ipv6hdr *)pshdr; + udph = (struct udphdr *)(ip6h + 1); + + memset(ip6h->flow_lbl, 0xff, sizeof(ip6h->flow_lbl)); + ip6h->priority = 0xf; + ip6h->hop_limit = 0xff; + } + udph->check = CSUM_MANGLED_0; + + bth_offset += hdr_size; + + memcpy(&pshdr[bth_offset], pkt->hdr, RXE_BTH_BYTES); + bth = (struct rxe_bth *)&pshdr[bth_offset]; + + /* exclude bth.resv8a */ + bth->qpn |= cpu_to_be32(~BTH_QPN_MASK); + + length = hdr_size + RXE_BTH_BYTES; + crc = crc32_le(crc, pshdr, length); + + /* And finish to compute the CRC on the remainder of the headers. */ + crc = crc32_le(crc, pkt->hdr + RXE_BTH_BYTES, + rxe_opcode[pkt->opcode].length - RXE_BTH_BYTES); + return crc; +} diff --git a/drivers/infiniband/sw/rxe/rxe_loc.h b/drivers/infiniband/sw/rxe/rxe_loc.h new file mode 100644 index 000000000000..4a5484ef604f --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_loc.h @@ -0,0 +1,286 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_LOC_H +#define RXE_LOC_H + +/* rxe_av.c */ + +int rxe_av_chk_attr(struct rxe_dev *rxe, struct ib_ah_attr *attr); + +int rxe_av_from_attr(struct rxe_dev *rxe, u8 port_num, + struct rxe_av *av, struct ib_ah_attr *attr); + +int rxe_av_to_attr(struct rxe_dev *rxe, struct rxe_av *av, + struct ib_ah_attr *attr); + +int rxe_av_fill_ip_info(struct rxe_dev *rxe, + struct rxe_av *av, + struct ib_ah_attr *attr, + struct ib_gid_attr *sgid_attr, + union ib_gid *sgid); + +struct rxe_av *rxe_get_av(struct rxe_pkt_info *pkt); + +/* rxe_cq.c */ +int rxe_cq_chk_attr(struct rxe_dev *rxe, struct rxe_cq *cq, + int cqe, int comp_vector, struct ib_udata *udata); + +int rxe_cq_from_init(struct rxe_dev *rxe, struct rxe_cq *cq, int cqe, + int comp_vector, struct ib_ucontext *context, + struct ib_udata *udata); + +int rxe_cq_resize_queue(struct rxe_cq *cq, int new_cqe, struct ib_udata *udata); + +int rxe_cq_post(struct rxe_cq *cq, struct rxe_cqe *cqe, int solicited); + +void rxe_cq_cleanup(void *arg); + +/* rxe_mcast.c */ +int rxe_mcast_get_grp(struct rxe_dev *rxe, union ib_gid *mgid, + struct rxe_mc_grp **grp_p); + +int rxe_mcast_add_grp_elem(struct rxe_dev *rxe, struct rxe_qp *qp, + struct rxe_mc_grp *grp); + +int rxe_mcast_drop_grp_elem(struct rxe_dev *rxe, struct rxe_qp *qp, + union ib_gid *mgid); + +void rxe_drop_all_mcast_groups(struct rxe_qp *qp); + +void rxe_mc_cleanup(void *arg); + +/* rxe_mmap.c */ +struct rxe_mmap_info { + struct list_head pending_mmaps; + struct ib_ucontext *context; + struct kref ref; + void *obj; + + struct mminfo info; +}; + +void rxe_mmap_release(struct kref *ref); + +struct rxe_mmap_info *rxe_create_mmap_info(struct rxe_dev *dev, + u32 size, + struct ib_ucontext *context, + void *obj); + +int rxe_mmap(struct ib_ucontext *context, struct vm_area_struct *vma); + +/* rxe_mr.c */ +enum copy_direction { + to_mem_obj, + from_mem_obj, +}; + +int rxe_mem_init_dma(struct rxe_dev *rxe, struct rxe_pd *pd, + int access, struct rxe_mem *mem); + +int rxe_mem_init_user(struct rxe_dev *rxe, struct rxe_pd *pd, u64 start, + u64 length, u64 iova, int access, struct ib_udata *udata, + struct rxe_mem *mr); + +int rxe_mem_init_fast(struct rxe_dev *rxe, struct rxe_pd *pd, + int max_pages, struct rxe_mem *mem); + +int rxe_mem_copy(struct rxe_mem *mem, u64 iova, void *addr, + int length, enum copy_direction dir, u32 *crcp); + +int copy_data(struct rxe_dev *rxe, struct rxe_pd *pd, int access, + struct rxe_dma_info *dma, void *addr, int length, + enum copy_direction dir, u32 *crcp); + +void *iova_to_vaddr(struct rxe_mem *mem, u64 iova, int length); + +enum lookup_type { + lookup_local, + lookup_remote, +}; + +struct rxe_mem *lookup_mem(struct rxe_pd *pd, int access, u32 key, + enum lookup_type type); + +int mem_check_range(struct rxe_mem *mem, u64 iova, size_t length); + +int rxe_mem_map_pages(struct rxe_dev *rxe, struct rxe_mem *mem, + u64 *page, int num_pages, u64 iova); + +void rxe_mem_cleanup(void *arg); + +int advance_dma_data(struct rxe_dma_info *dma, unsigned int length); + +/* rxe_qp.c */ +int rxe_qp_chk_init(struct rxe_dev *rxe, struct ib_qp_init_attr *init); + +int rxe_qp_from_init(struct rxe_dev *rxe, struct rxe_qp *qp, struct rxe_pd *pd, + struct ib_qp_init_attr *init, struct ib_udata *udata, + struct ib_pd *ibpd); + +int rxe_qp_to_init(struct rxe_qp *qp, struct ib_qp_init_attr *init); + +int rxe_qp_chk_attr(struct rxe_dev *rxe, struct rxe_qp *qp, + struct ib_qp_attr *attr, int mask); + +int rxe_qp_from_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, + int mask, struct ib_udata *udata); + +int rxe_qp_to_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask); + +void rxe_qp_error(struct rxe_qp *qp); + +void rxe_qp_destroy(struct rxe_qp *qp); + +void rxe_qp_cleanup(void *arg); + +static inline int qp_num(struct rxe_qp *qp) +{ + return qp->ibqp.qp_num; +} + +static inline enum ib_qp_type qp_type(struct rxe_qp *qp) +{ + return qp->ibqp.qp_type; +} + +static inline enum ib_qp_state qp_state(struct rxe_qp *qp) +{ + return qp->attr.qp_state; +} + +static inline int qp_mtu(struct rxe_qp *qp) +{ + if (qp->ibqp.qp_type == IB_QPT_RC || qp->ibqp.qp_type == IB_QPT_UC) + return qp->attr.path_mtu; + else + return RXE_PORT_MAX_MTU; +} + +static inline int rcv_wqe_size(int max_sge) +{ + return sizeof(struct rxe_recv_wqe) + + max_sge * sizeof(struct ib_sge); +} + +void free_rd_atomic_resource(struct rxe_qp *qp, struct resp_res *res); + +static inline void rxe_advance_resp_resource(struct rxe_qp *qp) +{ + qp->resp.res_head++; + if (unlikely(qp->resp.res_head == qp->attr.max_rd_atomic)) + qp->resp.res_head = 0; +} + +void retransmit_timer(unsigned long data); +void rnr_nak_timer(unsigned long data); + +void dump_qp(struct rxe_qp *qp); + +/* rxe_srq.c */ +#define IB_SRQ_INIT_MASK (~IB_SRQ_LIMIT) + +int rxe_srq_chk_attr(struct rxe_dev *rxe, struct rxe_srq *srq, + struct ib_srq_attr *attr, enum ib_srq_attr_mask mask); + +int rxe_srq_from_init(struct rxe_dev *rxe, struct rxe_srq *srq, + struct ib_srq_init_attr *init, + struct ib_ucontext *context, struct ib_udata *udata); + +int rxe_srq_from_attr(struct rxe_dev *rxe, struct rxe_srq *srq, + struct ib_srq_attr *attr, enum ib_srq_attr_mask mask, + struct ib_udata *udata); + +extern struct ib_dma_mapping_ops rxe_dma_mapping_ops; + +void rxe_release(struct kref *kref); + +int rxe_completer(void *arg); +int rxe_requester(void *arg); +int rxe_responder(void *arg); + +u32 rxe_icrc_hdr(struct rxe_pkt_info *pkt, struct sk_buff *skb); + +void rxe_resp_queue_pkt(struct rxe_dev *rxe, + struct rxe_qp *qp, struct sk_buff *skb); + +void rxe_comp_queue_pkt(struct rxe_dev *rxe, + struct rxe_qp *qp, struct sk_buff *skb); + +static inline unsigned wr_opcode_mask(int opcode, struct rxe_qp *qp) +{ + return rxe_wr_opcode_info[opcode].mask[qp->ibqp.qp_type]; +} + +static inline int rxe_xmit_packet(struct rxe_dev *rxe, struct rxe_qp *qp, + struct rxe_pkt_info *pkt, struct sk_buff *skb) +{ + int err; + int is_request = pkt->mask & RXE_REQ_MASK; + + if ((is_request && (qp->req.state != QP_STATE_READY)) || + (!is_request && (qp->resp.state != QP_STATE_READY))) { + pr_info("Packet dropped. QP is not in ready state\n"); + goto drop; + } + + if (pkt->mask & RXE_LOOPBACK_MASK) { + memcpy(SKB_TO_PKT(skb), pkt, sizeof(*pkt)); + err = rxe->ifc_ops->loopback(skb); + } else { + err = rxe->ifc_ops->send(rxe, pkt, skb); + } + + if (err) { + rxe->xmit_errors++; + return err; + } + + atomic_inc(&qp->skb_out); + + if ((qp_type(qp) != IB_QPT_RC) && + (pkt->mask & RXE_END_MASK)) { + pkt->wqe->state = wqe_state_done; + rxe_run_task(&qp->comp.task, 1); + } + + goto done; + +drop: + kfree_skb(skb); + err = 0; +done: + return err; +} + +#endif /* RXE_LOC_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_mcast.c b/drivers/infiniband/sw/rxe/rxe_mcast.c new file mode 100644 index 000000000000..fa95544ca7e0 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_mcast.c @@ -0,0 +1,190 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +int rxe_mcast_get_grp(struct rxe_dev *rxe, union ib_gid *mgid, + struct rxe_mc_grp **grp_p) +{ + int err; + struct rxe_mc_grp *grp; + + if (rxe->attr.max_mcast_qp_attach == 0) { + err = -EINVAL; + goto err1; + } + + grp = rxe_pool_get_key(&rxe->mc_grp_pool, mgid); + if (grp) + goto done; + + grp = rxe_alloc(&rxe->mc_grp_pool); + if (!grp) { + err = -ENOMEM; + goto err1; + } + + INIT_LIST_HEAD(&grp->qp_list); + spin_lock_init(&grp->mcg_lock); + grp->rxe = rxe; + + rxe_add_key(grp, mgid); + + err = rxe->ifc_ops->mcast_add(rxe, mgid); + if (err) + goto err2; + +done: + *grp_p = grp; + return 0; + +err2: + rxe_drop_ref(grp); +err1: + return err; +} + +int rxe_mcast_add_grp_elem(struct rxe_dev *rxe, struct rxe_qp *qp, + struct rxe_mc_grp *grp) +{ + int err; + struct rxe_mc_elem *elem; + + /* check to see of the qp is already a member of the group */ + spin_lock_bh(&qp->grp_lock); + spin_lock_bh(&grp->mcg_lock); + list_for_each_entry(elem, &grp->qp_list, qp_list) { + if (elem->qp == qp) { + err = 0; + goto out; + } + } + + if (grp->num_qp >= rxe->attr.max_mcast_qp_attach) { + err = -ENOMEM; + goto out; + } + + elem = rxe_alloc(&rxe->mc_elem_pool); + if (!elem) { + err = -ENOMEM; + goto out; + } + + /* each qp holds a ref on the grp */ + rxe_add_ref(grp); + + grp->num_qp++; + elem->qp = qp; + elem->grp = grp; + + list_add(&elem->qp_list, &grp->qp_list); + list_add(&elem->grp_list, &qp->grp_list); + + err = 0; +out: + spin_unlock_bh(&grp->mcg_lock); + spin_unlock_bh(&qp->grp_lock); + return err; +} + +int rxe_mcast_drop_grp_elem(struct rxe_dev *rxe, struct rxe_qp *qp, + union ib_gid *mgid) +{ + struct rxe_mc_grp *grp; + struct rxe_mc_elem *elem, *tmp; + + grp = rxe_pool_get_key(&rxe->mc_grp_pool, mgid); + if (!grp) + goto err1; + + spin_lock_bh(&qp->grp_lock); + spin_lock_bh(&grp->mcg_lock); + + list_for_each_entry_safe(elem, tmp, &grp->qp_list, qp_list) { + if (elem->qp == qp) { + list_del(&elem->qp_list); + list_del(&elem->grp_list); + grp->num_qp--; + + spin_unlock_bh(&grp->mcg_lock); + spin_unlock_bh(&qp->grp_lock); + rxe_drop_ref(elem); + rxe_drop_ref(grp); /* ref held by QP */ + rxe_drop_ref(grp); /* ref from get_key */ + return 0; + } + } + + spin_unlock_bh(&grp->mcg_lock); + spin_unlock_bh(&qp->grp_lock); + rxe_drop_ref(grp); /* ref from get_key */ +err1: + return -EINVAL; +} + +void rxe_drop_all_mcast_groups(struct rxe_qp *qp) +{ + struct rxe_mc_grp *grp; + struct rxe_mc_elem *elem; + + while (1) { + spin_lock_bh(&qp->grp_lock); + if (list_empty(&qp->grp_list)) { + spin_unlock_bh(&qp->grp_lock); + break; + } + elem = list_first_entry(&qp->grp_list, struct rxe_mc_elem, + grp_list); + list_del(&elem->grp_list); + spin_unlock_bh(&qp->grp_lock); + + grp = elem->grp; + spin_lock_bh(&grp->mcg_lock); + list_del(&elem->qp_list); + grp->num_qp--; + spin_unlock_bh(&grp->mcg_lock); + rxe_drop_ref(grp); + rxe_drop_ref(elem); + } +} + +void rxe_mc_cleanup(void *arg) +{ + struct rxe_mc_grp *grp = arg; + struct rxe_dev *rxe = grp->rxe; + + rxe_drop_key(grp); + rxe->ifc_ops->mcast_delete(rxe, &grp->mgid); +} diff --git a/drivers/infiniband/sw/rxe/rxe_mmap.c b/drivers/infiniband/sw/rxe/rxe_mmap.c new file mode 100644 index 000000000000..54b3c7c99eff --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_mmap.c @@ -0,0 +1,173 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +void rxe_mmap_release(struct kref *ref) +{ + struct rxe_mmap_info *ip = container_of(ref, + struct rxe_mmap_info, ref); + struct rxe_dev *rxe = to_rdev(ip->context->device); + + spin_lock_bh(&rxe->pending_lock); + + if (!list_empty(&ip->pending_mmaps)) + list_del(&ip->pending_mmaps); + + spin_unlock_bh(&rxe->pending_lock); + + vfree(ip->obj); /* buf */ + kfree(ip); +} + +/* + * open and close keep track of how many times the memory region is mapped, + * to avoid releasing it. + */ +static void rxe_vma_open(struct vm_area_struct *vma) +{ + struct rxe_mmap_info *ip = vma->vm_private_data; + + kref_get(&ip->ref); +} + +static void rxe_vma_close(struct vm_area_struct *vma) +{ + struct rxe_mmap_info *ip = vma->vm_private_data; + + kref_put(&ip->ref, rxe_mmap_release); +} + +static struct vm_operations_struct rxe_vm_ops = { + .open = rxe_vma_open, + .close = rxe_vma_close, +}; + +/** + * rxe_mmap - create a new mmap region + * @context: the IB user context of the process making the mmap() call + * @vma: the VMA to be initialized + * Return zero if the mmap is OK. Otherwise, return an errno. + */ +int rxe_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) +{ + struct rxe_dev *rxe = to_rdev(context->device); + unsigned long offset = vma->vm_pgoff << PAGE_SHIFT; + unsigned long size = vma->vm_end - vma->vm_start; + struct rxe_mmap_info *ip, *pp; + int ret; + + /* + * Search the device's list of objects waiting for a mmap call. + * Normally, this list is very short since a call to create a + * CQ, QP, or SRQ is soon followed by a call to mmap(). + */ + spin_lock_bh(&rxe->pending_lock); + list_for_each_entry_safe(ip, pp, &rxe->pending_mmaps, pending_mmaps) { + if (context != ip->context || (__u64)offset != ip->info.offset) + continue; + + /* Don't allow a mmap larger than the object. */ + if (size > ip->info.size) { + pr_err("mmap region is larger than the object!\n"); + spin_unlock_bh(&rxe->pending_lock); + ret = -EINVAL; + goto done; + } + + goto found_it; + } + pr_warn("unable to find pending mmap info\n"); + spin_unlock_bh(&rxe->pending_lock); + ret = -EINVAL; + goto done; + +found_it: + list_del_init(&ip->pending_mmaps); + spin_unlock_bh(&rxe->pending_lock); + + ret = remap_vmalloc_range(vma, ip->obj, 0); + if (ret) { + pr_err("rxe: err %d from remap_vmalloc_range\n", ret); + goto done; + } + + vma->vm_ops = &rxe_vm_ops; + vma->vm_private_data = ip; + rxe_vma_open(vma); +done: + return ret; +} + +/* + * Allocate information for rxe_mmap + */ +struct rxe_mmap_info *rxe_create_mmap_info(struct rxe_dev *rxe, + u32 size, + struct ib_ucontext *context, + void *obj) +{ + struct rxe_mmap_info *ip; + + ip = kmalloc(sizeof(*ip), GFP_KERNEL); + if (!ip) + return NULL; + + size = PAGE_ALIGN(size); + + spin_lock_bh(&rxe->mmap_offset_lock); + + if (rxe->mmap_offset == 0) + rxe->mmap_offset = PAGE_SIZE; + + ip->info.offset = rxe->mmap_offset; + rxe->mmap_offset += size; + + spin_unlock_bh(&rxe->mmap_offset_lock); + + INIT_LIST_HEAD(&ip->pending_mmaps); + ip->info.size = size; + ip->context = context; + ip->obj = obj; + kref_init(&ip->ref); + + return ip; +} diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c new file mode 100644 index 000000000000..f3dab6574504 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_mr.c @@ -0,0 +1,643 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +/* + * lfsr (linear feedback shift register) with period 255 + */ +static u8 rxe_get_key(void) +{ + static unsigned key = 1; + + key = key << 1; + + key |= (0 != (key & 0x100)) ^ (0 != (key & 0x10)) + ^ (0 != (key & 0x80)) ^ (0 != (key & 0x40)); + + key &= 0xff; + + return key; +} + +int mem_check_range(struct rxe_mem *mem, u64 iova, size_t length) +{ + switch (mem->type) { + case RXE_MEM_TYPE_DMA: + return 0; + + case RXE_MEM_TYPE_MR: + case RXE_MEM_TYPE_FMR: + return ((iova < mem->iova) || + ((iova + length) > (mem->iova + mem->length))) ? + -EFAULT : 0; + + default: + return -EFAULT; + } +} + +#define IB_ACCESS_REMOTE (IB_ACCESS_REMOTE_READ \ + | IB_ACCESS_REMOTE_WRITE \ + | IB_ACCESS_REMOTE_ATOMIC) + +static void rxe_mem_init(int access, struct rxe_mem *mem) +{ + u32 lkey = mem->pelem.index << 8 | rxe_get_key(); + u32 rkey = (access & IB_ACCESS_REMOTE) ? lkey : 0; + + if (mem->pelem.pool->type == RXE_TYPE_MR) { + mem->ibmr.lkey = lkey; + mem->ibmr.rkey = rkey; + } + + mem->lkey = lkey; + mem->rkey = rkey; + mem->state = RXE_MEM_STATE_INVALID; + mem->type = RXE_MEM_TYPE_NONE; + mem->map_shift = ilog2(RXE_BUF_PER_MAP); +} + +void rxe_mem_cleanup(void *arg) +{ + struct rxe_mem *mem = arg; + int i; + + if (mem->umem) + ib_umem_release(mem->umem); + + if (mem->map) { + for (i = 0; i < mem->num_map; i++) + kfree(mem->map[i]); + + kfree(mem->map); + } +} + +static int rxe_mem_alloc(struct rxe_dev *rxe, struct rxe_mem *mem, int num_buf) +{ + int i; + int num_map; + struct rxe_map **map = mem->map; + + num_map = (num_buf + RXE_BUF_PER_MAP - 1) / RXE_BUF_PER_MAP; + + mem->map = kmalloc_array(num_map, sizeof(*map), GFP_KERNEL); + if (!mem->map) + goto err1; + + for (i = 0; i < num_map; i++) { + mem->map[i] = kmalloc(sizeof(**map), GFP_KERNEL); + if (!mem->map[i]) + goto err2; + } + + WARN_ON(!is_power_of_2(RXE_BUF_PER_MAP)); + + mem->map_shift = ilog2(RXE_BUF_PER_MAP); + mem->map_mask = RXE_BUF_PER_MAP - 1; + + mem->num_buf = num_buf; + mem->num_map = num_map; + mem->max_buf = num_map * RXE_BUF_PER_MAP; + + return 0; + +err2: + for (i--; i >= 0; i--) + kfree(mem->map[i]); + + kfree(mem->map); +err1: + return -ENOMEM; +} + +int rxe_mem_init_dma(struct rxe_dev *rxe, struct rxe_pd *pd, + int access, struct rxe_mem *mem) +{ + rxe_mem_init(access, mem); + + mem->pd = pd; + mem->access = access; + mem->state = RXE_MEM_STATE_VALID; + mem->type = RXE_MEM_TYPE_DMA; + + return 0; +} + +int rxe_mem_init_user(struct rxe_dev *rxe, struct rxe_pd *pd, u64 start, + u64 length, u64 iova, int access, struct ib_udata *udata, + struct rxe_mem *mem) +{ + int entry; + struct rxe_map **map; + struct rxe_phys_buf *buf = NULL; + struct ib_umem *umem; + struct scatterlist *sg; + int num_buf; + void *vaddr; + int err; + + umem = ib_umem_get(pd->ibpd.uobject->context, start, length, access, 0); + if (IS_ERR(umem)) { + pr_warn("err %d from rxe_umem_get\n", + (int)PTR_ERR(umem)); + err = -EINVAL; + goto err1; + } + + mem->umem = umem; + num_buf = umem->nmap; + + rxe_mem_init(access, mem); + + err = rxe_mem_alloc(rxe, mem, num_buf); + if (err) { + pr_warn("err %d from rxe_mem_alloc\n", err); + ib_umem_release(umem); + goto err1; + } + + WARN_ON(!is_power_of_2(umem->page_size)); + + mem->page_shift = ilog2(umem->page_size); + mem->page_mask = umem->page_size - 1; + + num_buf = 0; + map = mem->map; + if (length > 0) { + buf = map[0]->buf; + + for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { + vaddr = page_address(sg_page(sg)); + if (!vaddr) { + pr_warn("null vaddr\n"); + err = -ENOMEM; + goto err1; + } + + buf->addr = (uintptr_t)vaddr; + buf->size = umem->page_size; + num_buf++; + buf++; + + if (num_buf >= RXE_BUF_PER_MAP) { + map++; + buf = map[0]->buf; + num_buf = 0; + } + } + } + + mem->pd = pd; + mem->umem = umem; + mem->access = access; + mem->length = length; + mem->iova = iova; + mem->va = start; + mem->offset = ib_umem_offset(umem); + mem->state = RXE_MEM_STATE_VALID; + mem->type = RXE_MEM_TYPE_MR; + + return 0; + +err1: + return err; +} + +int rxe_mem_init_fast(struct rxe_dev *rxe, struct rxe_pd *pd, + int max_pages, struct rxe_mem *mem) +{ + int err; + + rxe_mem_init(0, mem); + + /* In fastreg, we also set the rkey */ + mem->ibmr.rkey = mem->ibmr.lkey; + + err = rxe_mem_alloc(rxe, mem, max_pages); + if (err) + goto err1; + + mem->pd = pd; + mem->max_buf = max_pages; + mem->state = RXE_MEM_STATE_FREE; + mem->type = RXE_MEM_TYPE_MR; + + return 0; + +err1: + return err; +} + +static void lookup_iova( + struct rxe_mem *mem, + u64 iova, + int *m_out, + int *n_out, + size_t *offset_out) +{ + size_t offset = iova - mem->iova + mem->offset; + int map_index; + int buf_index; + u64 length; + + if (likely(mem->page_shift)) { + *offset_out = offset & mem->page_mask; + offset >>= mem->page_shift; + *n_out = offset & mem->map_mask; + *m_out = offset >> mem->map_shift; + } else { + map_index = 0; + buf_index = 0; + + length = mem->map[map_index]->buf[buf_index].size; + + while (offset >= length) { + offset -= length; + buf_index++; + + if (buf_index == RXE_BUF_PER_MAP) { + map_index++; + buf_index = 0; + } + length = mem->map[map_index]->buf[buf_index].size; + } + + *m_out = map_index; + *n_out = buf_index; + *offset_out = offset; + } +} + +void *iova_to_vaddr(struct rxe_mem *mem, u64 iova, int length) +{ + size_t offset; + int m, n; + void *addr; + + if (mem->state != RXE_MEM_STATE_VALID) { + pr_warn("mem not in valid state\n"); + addr = NULL; + goto out; + } + + if (!mem->map) { + addr = (void *)(uintptr_t)iova; + goto out; + } + + if (mem_check_range(mem, iova, length)) { + pr_warn("range violation\n"); + addr = NULL; + goto out; + } + + lookup_iova(mem, iova, &m, &n, &offset); + + if (offset + length > mem->map[m]->buf[n].size) { + pr_warn("crosses page boundary\n"); + addr = NULL; + goto out; + } + + addr = (void *)(uintptr_t)mem->map[m]->buf[n].addr + offset; + +out: + return addr; +} + +/* copy data from a range (vaddr, vaddr+length-1) to or from + * a mem object starting at iova. Compute incremental value of + * crc32 if crcp is not zero. caller must hold a reference to mem + */ +int rxe_mem_copy(struct rxe_mem *mem, u64 iova, void *addr, int length, + enum copy_direction dir, u32 *crcp) +{ + int err; + int bytes; + u8 *va; + struct rxe_map **map; + struct rxe_phys_buf *buf; + int m; + int i; + size_t offset; + u32 crc = crcp ? (*crcp) : 0; + + if (mem->type == RXE_MEM_TYPE_DMA) { + u8 *src, *dest; + + src = (dir == to_mem_obj) ? + addr : ((void *)(uintptr_t)iova); + + dest = (dir == to_mem_obj) ? + ((void *)(uintptr_t)iova) : addr; + + if (crcp) + *crcp = crc32_le(*crcp, src, length); + + memcpy(dest, src, length); + + return 0; + } + + WARN_ON(!mem->map); + + err = mem_check_range(mem, iova, length); + if (err) { + err = -EFAULT; + goto err1; + } + + lookup_iova(mem, iova, &m, &i, &offset); + + map = mem->map + m; + buf = map[0]->buf + i; + + while (length > 0) { + u8 *src, *dest; + + va = (u8 *)(uintptr_t)buf->addr + offset; + src = (dir == to_mem_obj) ? addr : va; + dest = (dir == to_mem_obj) ? va : addr; + + bytes = buf->size - offset; + + if (bytes > length) + bytes = length; + + if (crcp) + crc = crc32_le(crc, src, bytes); + + memcpy(dest, src, bytes); + + length -= bytes; + addr += bytes; + + offset = 0; + buf++; + i++; + + if (i == RXE_BUF_PER_MAP) { + i = 0; + map++; + buf = map[0]->buf; + } + } + + if (crcp) + *crcp = crc; + + return 0; + +err1: + return err; +} + +/* copy data in or out of a wqe, i.e. sg list + * under the control of a dma descriptor + */ +int copy_data( + struct rxe_dev *rxe, + struct rxe_pd *pd, + int access, + struct rxe_dma_info *dma, + void *addr, + int length, + enum copy_direction dir, + u32 *crcp) +{ + int bytes; + struct rxe_sge *sge = &dma->sge[dma->cur_sge]; + int offset = dma->sge_offset; + int resid = dma->resid; + struct rxe_mem *mem = NULL; + u64 iova; + int err; + + if (length == 0) + return 0; + + if (length > resid) { + err = -EINVAL; + goto err2; + } + + if (sge->length && (offset < sge->length)) { + mem = lookup_mem(pd, access, sge->lkey, lookup_local); + if (!mem) { + err = -EINVAL; + goto err1; + } + } + + while (length > 0) { + bytes = length; + + if (offset >= sge->length) { + if (mem) { + rxe_drop_ref(mem); + mem = NULL; + } + sge++; + dma->cur_sge++; + offset = 0; + + if (dma->cur_sge >= dma->num_sge) { + err = -ENOSPC; + goto err2; + } + + if (sge->length) { + mem = lookup_mem(pd, access, sge->lkey, + lookup_local); + if (!mem) { + err = -EINVAL; + goto err1; + } + } else { + continue; + } + } + + if (bytes > sge->length - offset) + bytes = sge->length - offset; + + if (bytes > 0) { + iova = sge->addr + offset; + + err = rxe_mem_copy(mem, iova, addr, bytes, dir, crcp); + if (err) + goto err2; + + offset += bytes; + resid -= bytes; + length -= bytes; + addr += bytes; + } + } + + dma->sge_offset = offset; + dma->resid = resid; + + if (mem) + rxe_drop_ref(mem); + + return 0; + +err2: + if (mem) + rxe_drop_ref(mem); +err1: + return err; +} + +int advance_dma_data(struct rxe_dma_info *dma, unsigned int length) +{ + struct rxe_sge *sge = &dma->sge[dma->cur_sge]; + int offset = dma->sge_offset; + int resid = dma->resid; + + while (length) { + unsigned int bytes; + + if (offset >= sge->length) { + sge++; + dma->cur_sge++; + offset = 0; + if (dma->cur_sge >= dma->num_sge) + return -ENOSPC; + } + + bytes = length; + + if (bytes > sge->length - offset) + bytes = sge->length - offset; + + offset += bytes; + resid -= bytes; + length -= bytes; + } + + dma->sge_offset = offset; + dma->resid = resid; + + return 0; +} + +/* (1) find the mem (mr or mw) corresponding to lkey/rkey + * depending on lookup_type + * (2) verify that the (qp) pd matches the mem pd + * (3) verify that the mem can support the requested access + * (4) verify that mem state is valid + */ +struct rxe_mem *lookup_mem(struct rxe_pd *pd, int access, u32 key, + enum lookup_type type) +{ + struct rxe_mem *mem; + struct rxe_dev *rxe = to_rdev(pd->ibpd.device); + int index = key >> 8; + + if (index >= RXE_MIN_MR_INDEX && index <= RXE_MAX_MR_INDEX) { + mem = rxe_pool_get_index(&rxe->mr_pool, index); + if (!mem) + goto err1; + } else { + goto err1; + } + + if ((type == lookup_local && mem->lkey != key) || + (type == lookup_remote && mem->rkey != key)) + goto err2; + + if (mem->pd != pd) + goto err2; + + if (access && !(access & mem->access)) + goto err2; + + if (mem->state != RXE_MEM_STATE_VALID) + goto err2; + + return mem; + +err2: + rxe_drop_ref(mem); +err1: + return NULL; +} + +int rxe_mem_map_pages(struct rxe_dev *rxe, struct rxe_mem *mem, + u64 *page, int num_pages, u64 iova) +{ + int i; + int num_buf; + int err; + struct rxe_map **map; + struct rxe_phys_buf *buf; + int page_size; + + if (num_pages > mem->max_buf) { + err = -EINVAL; + goto err1; + } + + num_buf = 0; + page_size = 1 << mem->page_shift; + map = mem->map; + buf = map[0]->buf; + + for (i = 0; i < num_pages; i++) { + buf->addr = *page++; + buf->size = page_size; + buf++; + num_buf++; + + if (num_buf == RXE_BUF_PER_MAP) { + map++; + buf = map[0]->buf; + num_buf = 0; + } + } + + mem->iova = iova; + mem->va = iova; + mem->length = num_pages << mem->page_shift; + mem->state = RXE_MEM_STATE_VALID; + + return 0; + +err1: + return err; +} diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c new file mode 100644 index 000000000000..0b8d2ea8b41d --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -0,0 +1,708 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "rxe.h" +#include "rxe_net.h" +#include "rxe_loc.h" + +static LIST_HEAD(rxe_dev_list); +static spinlock_t dev_list_lock; /* spinlock for device list */ + +struct rxe_dev *net_to_rxe(struct net_device *ndev) +{ + struct rxe_dev *rxe; + struct rxe_dev *found = NULL; + + spin_lock_bh(&dev_list_lock); + list_for_each_entry(rxe, &rxe_dev_list, list) { + if (rxe->ndev == ndev) { + found = rxe; + break; + } + } + spin_unlock_bh(&dev_list_lock); + + return found; +} + +struct rxe_dev *get_rxe_by_name(const char* name) +{ + struct rxe_dev *rxe; + struct rxe_dev *found = NULL; + + spin_lock_bh(&dev_list_lock); + list_for_each_entry(rxe, &rxe_dev_list, list) { + if (!strcmp(name, rxe->ib_dev.name)) { + found = rxe; + break; + } + } + spin_unlock_bh(&dev_list_lock); + return found; +} + + +struct rxe_recv_sockets recv_sockets; + +static __be64 rxe_mac_to_eui64(struct net_device *ndev) +{ + unsigned char *mac_addr = ndev->dev_addr; + __be64 eui64; + unsigned char *dst = (unsigned char *)&eui64; + + dst[0] = mac_addr[0] ^ 2; + dst[1] = mac_addr[1]; + dst[2] = mac_addr[2]; + dst[3] = 0xff; + dst[4] = 0xfe; + dst[5] = mac_addr[3]; + dst[6] = mac_addr[4]; + dst[7] = mac_addr[5]; + + return eui64; +} + +static __be64 node_guid(struct rxe_dev *rxe) +{ + return rxe_mac_to_eui64(rxe->ndev); +} + +static __be64 port_guid(struct rxe_dev *rxe) +{ + return rxe_mac_to_eui64(rxe->ndev); +} + +static struct device *dma_device(struct rxe_dev *rxe) +{ + struct net_device *ndev; + + ndev = rxe->ndev; + + if (ndev->priv_flags & IFF_802_1Q_VLAN) + ndev = vlan_dev_real_dev(ndev); + + return ndev->dev.parent; +} + +static int mcast_add(struct rxe_dev *rxe, union ib_gid *mgid) +{ + int err; + unsigned char ll_addr[ETH_ALEN]; + + ipv6_eth_mc_map((struct in6_addr *)mgid->raw, ll_addr); + err = dev_mc_add(rxe->ndev, ll_addr); + + return err; +} + +static int mcast_delete(struct rxe_dev *rxe, union ib_gid *mgid) +{ + int err; + unsigned char ll_addr[ETH_ALEN]; + + ipv6_eth_mc_map((struct in6_addr *)mgid->raw, ll_addr); + err = dev_mc_del(rxe->ndev, ll_addr); + + return err; +} + +static struct dst_entry *rxe_find_route4(struct net_device *ndev, + struct in_addr *saddr, + struct in_addr *daddr) +{ + struct rtable *rt; + struct flowi4 fl = { { 0 } }; + + memset(&fl, 0, sizeof(fl)); + fl.flowi4_oif = ndev->ifindex; + memcpy(&fl.saddr, saddr, sizeof(*saddr)); + memcpy(&fl.daddr, daddr, sizeof(*daddr)); + fl.flowi4_proto = IPPROTO_UDP; + + rt = ip_route_output_key(&init_net, &fl); + if (IS_ERR(rt)) { + pr_err_ratelimited("no route to %pI4\n", &daddr->s_addr); + return NULL; + } + + return &rt->dst; +} + +#if IS_ENABLED(CONFIG_IPV6) +static struct dst_entry *rxe_find_route6(struct net_device *ndev, + struct in6_addr *saddr, + struct in6_addr *daddr) +{ + struct dst_entry *ndst; + struct flowi6 fl6 = { { 0 } }; + + memset(&fl6, 0, sizeof(fl6)); + fl6.flowi6_oif = ndev->ifindex; + memcpy(&fl6.saddr, saddr, sizeof(*saddr)); + memcpy(&fl6.daddr, daddr, sizeof(*daddr)); + fl6.flowi6_proto = IPPROTO_UDP; + + if (unlikely(ipv6_stub->ipv6_dst_lookup(sock_net(recv_sockets.sk6->sk), + recv_sockets.sk6->sk, &ndst, &fl6))) { + pr_err_ratelimited("no route to %pI6\n", daddr); + goto put; + } + + if (unlikely(ndst->error)) { + pr_err("no route to %pI6\n", daddr); + goto put; + } + + return ndst; +put: + dst_release(ndst); + return NULL; +} + +#else + +static struct dst_entry *rxe_find_route6(struct net_device *ndev, + struct in6_addr *saddr, + struct in6_addr *daddr) +{ + return NULL; +} + +#endif + +static int rxe_udp_encap_recv(struct sock *sk, struct sk_buff *skb) +{ + struct udphdr *udph; + struct net_device *ndev = skb->dev; + struct rxe_dev *rxe = net_to_rxe(ndev); + struct rxe_pkt_info *pkt = SKB_TO_PKT(skb); + + if (!rxe) + goto drop; + + if (skb_linearize(skb)) { + pr_err("skb_linearize failed\n"); + goto drop; + } + + udph = udp_hdr(skb); + pkt->rxe = rxe; + pkt->port_num = 1; + pkt->hdr = (u8 *)(udph + 1); + pkt->mask = RXE_GRH_MASK; + pkt->paylen = be16_to_cpu(udph->len) - sizeof(*udph); + + return rxe_rcv(skb); +drop: + kfree_skb(skb); + return 0; +} + +static struct socket *rxe_setup_udp_tunnel(struct net *net, __be16 port, + bool ipv6) +{ + int err; + struct socket *sock; + struct udp_port_cfg udp_cfg; + struct udp_tunnel_sock_cfg tnl_cfg; + + memset(&udp_cfg, 0, sizeof(udp_cfg)); + + if (ipv6) { + udp_cfg.family = AF_INET6; + udp_cfg.ipv6_v6only = 1; + } else { + udp_cfg.family = AF_INET; + } + + udp_cfg.local_udp_port = port; + + /* Create UDP socket */ + err = udp_sock_create(net, &udp_cfg, &sock); + if (err < 0) { + pr_err("failed to create udp socket. err = %d\n", err); + return ERR_PTR(err); + } + + tnl_cfg.sk_user_data = NULL; + tnl_cfg.encap_type = 1; + tnl_cfg.encap_rcv = rxe_udp_encap_recv; + tnl_cfg.encap_destroy = NULL; + + /* Setup UDP tunnel */ + setup_udp_tunnel_sock(net, sock, &tnl_cfg); + + return sock; +} + +static void rxe_release_udp_tunnel(struct socket *sk) +{ + udp_tunnel_sock_release(sk); +} + +static void prepare_udp_hdr(struct sk_buff *skb, __be16 src_port, + __be16 dst_port) +{ + struct udphdr *udph; + + __skb_push(skb, sizeof(*udph)); + skb_reset_transport_header(skb); + udph = udp_hdr(skb); + + udph->dest = dst_port; + udph->source = src_port; + udph->len = htons(skb->len); + udph->check = 0; +} + +static void prepare_ipv4_hdr(struct dst_entry *dst, struct sk_buff *skb, + __be32 saddr, __be32 daddr, __u8 proto, + __u8 tos, __u8 ttl, __be16 df, bool xnet) +{ + struct iphdr *iph; + + skb_scrub_packet(skb, xnet); + + skb_clear_hash(skb); + skb_dst_set(skb, dst); + memset(IPCB(skb), 0, sizeof(*IPCB(skb))); + + skb_push(skb, sizeof(struct iphdr)); + skb_reset_network_header(skb); + + iph = ip_hdr(skb); + + iph->version = IPVERSION; + iph->ihl = sizeof(struct iphdr) >> 2; + iph->frag_off = df; + iph->protocol = proto; + iph->tos = tos; + iph->daddr = daddr; + iph->saddr = saddr; + iph->ttl = ttl; + __ip_select_ident(dev_net(dst->dev), iph, + skb_shinfo(skb)->gso_segs ?: 1); + iph->tot_len = htons(skb->len); + ip_send_check(iph); +} + +static void prepare_ipv6_hdr(struct dst_entry *dst, struct sk_buff *skb, + struct in6_addr *saddr, struct in6_addr *daddr, + __u8 proto, __u8 prio, __u8 ttl) +{ + struct ipv6hdr *ip6h; + + memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); + IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED + | IPSKB_REROUTED); + skb_dst_set(skb, dst); + + __skb_push(skb, sizeof(*ip6h)); + skb_reset_network_header(skb); + ip6h = ipv6_hdr(skb); + ip6_flow_hdr(ip6h, prio, htonl(0)); + ip6h->payload_len = htons(skb->len); + ip6h->nexthdr = proto; + ip6h->hop_limit = ttl; + ip6h->daddr = *daddr; + ip6h->saddr = *saddr; + ip6h->payload_len = htons(skb->len - sizeof(*ip6h)); +} + +static int prepare4(struct rxe_dev *rxe, struct sk_buff *skb, struct rxe_av *av) +{ + struct dst_entry *dst; + bool xnet = false; + __be16 df = htons(IP_DF); + struct in_addr *saddr = &av->sgid_addr._sockaddr_in.sin_addr; + struct in_addr *daddr = &av->dgid_addr._sockaddr_in.sin_addr; + struct rxe_pkt_info *pkt = SKB_TO_PKT(skb); + + dst = rxe_find_route4(rxe->ndev, saddr, daddr); + if (!dst) { + pr_err("Host not reachable\n"); + return -EHOSTUNREACH; + } + + if (!memcmp(saddr, daddr, sizeof(*daddr))) + pkt->mask |= RXE_LOOPBACK_MASK; + + prepare_udp_hdr(skb, htons(RXE_ROCE_V2_SPORT), + htons(ROCE_V2_UDP_DPORT)); + + prepare_ipv4_hdr(dst, skb, saddr->s_addr, daddr->s_addr, IPPROTO_UDP, + av->grh.traffic_class, av->grh.hop_limit, df, xnet); + return 0; +} + +static int prepare6(struct rxe_dev *rxe, struct sk_buff *skb, struct rxe_av *av) +{ + struct dst_entry *dst; + struct in6_addr *saddr = &av->sgid_addr._sockaddr_in6.sin6_addr; + struct in6_addr *daddr = &av->dgid_addr._sockaddr_in6.sin6_addr; + struct rxe_pkt_info *pkt = SKB_TO_PKT(skb); + + dst = rxe_find_route6(rxe->ndev, saddr, daddr); + if (!dst) { + pr_err("Host not reachable\n"); + return -EHOSTUNREACH; + } + + if (!memcmp(saddr, daddr, sizeof(*daddr))) + pkt->mask |= RXE_LOOPBACK_MASK; + + prepare_udp_hdr(skb, htons(RXE_ROCE_V2_SPORT), + htons(ROCE_V2_UDP_DPORT)); + + prepare_ipv6_hdr(dst, skb, saddr, daddr, IPPROTO_UDP, + av->grh.traffic_class, + av->grh.hop_limit); + return 0; +} + +static int prepare(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb, u32 *crc) +{ + int err = 0; + struct rxe_av *av = rxe_get_av(pkt); + + if (av->network_type == RDMA_NETWORK_IPV4) + err = prepare4(rxe, skb, av); + else if (av->network_type == RDMA_NETWORK_IPV6) + err = prepare6(rxe, skb, av); + + *crc = rxe_icrc_hdr(pkt, skb); + + return err; +} + +static void rxe_skb_tx_dtor(struct sk_buff *skb) +{ + struct sock *sk = skb->sk; + struct rxe_qp *qp = sk->sk_user_data; + int skb_out = atomic_dec_return(&qp->skb_out); + + if (unlikely(qp->need_req_skb && + skb_out < RXE_INFLIGHT_SKBS_PER_QP_LOW)) + rxe_run_task(&qp->req.task, 1); +} + +static int send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb) +{ + struct sk_buff *nskb; + struct rxe_av *av; + int err; + + av = rxe_get_av(pkt); + + nskb = skb_clone(skb, GFP_ATOMIC); + if (!nskb) + return -ENOMEM; + + nskb->destructor = rxe_skb_tx_dtor; + nskb->sk = pkt->qp->sk->sk; + + if (av->network_type == RDMA_NETWORK_IPV4) { + err = ip_local_out(dev_net(skb_dst(skb)->dev), nskb->sk, nskb); + } else if (av->network_type == RDMA_NETWORK_IPV6) { + err = ip6_local_out(dev_net(skb_dst(skb)->dev), nskb->sk, nskb); + } else { + pr_err("Unknown layer 3 protocol: %d\n", av->network_type); + kfree_skb(nskb); + return -EINVAL; + } + + if (unlikely(net_xmit_eval(err))) { + pr_debug("error sending packet: %d\n", err); + return -EAGAIN; + } + + kfree_skb(skb); + + return 0; +} + +static int loopback(struct sk_buff *skb) +{ + return rxe_rcv(skb); +} + +static inline int addr_same(struct rxe_dev *rxe, struct rxe_av *av) +{ + return rxe->port.port_guid == av->grh.dgid.global.interface_id; +} + +static struct sk_buff *init_packet(struct rxe_dev *rxe, struct rxe_av *av, + int paylen, struct rxe_pkt_info *pkt) +{ + unsigned int hdr_len; + struct sk_buff *skb; + + if (av->network_type == RDMA_NETWORK_IPV4) + hdr_len = ETH_HLEN + sizeof(struct udphdr) + + sizeof(struct iphdr); + else + hdr_len = ETH_HLEN + sizeof(struct udphdr) + + sizeof(struct ipv6hdr); + + skb = alloc_skb(paylen + hdr_len + LL_RESERVED_SPACE(rxe->ndev), + GFP_ATOMIC); + if (unlikely(!skb)) + return NULL; + + skb_reserve(skb, hdr_len + LL_RESERVED_SPACE(rxe->ndev)); + + skb->dev = rxe->ndev; + if (av->network_type == RDMA_NETWORK_IPV4) + skb->protocol = htons(ETH_P_IP); + else + skb->protocol = htons(ETH_P_IPV6); + + pkt->rxe = rxe; + pkt->port_num = 1; + pkt->hdr = skb_put(skb, paylen); + pkt->mask |= RXE_GRH_MASK; + + memset(pkt->hdr, 0, paylen); + + return skb; +} + +/* + * this is required by rxe_cfg to match rxe devices in + * /sys/class/infiniband up with their underlying ethernet devices + */ +static char *parent_name(struct rxe_dev *rxe, unsigned int port_num) +{ + return rxe->ndev->name; +} + +static enum rdma_link_layer link_layer(struct rxe_dev *rxe, + unsigned int port_num) +{ + return IB_LINK_LAYER_ETHERNET; +} + +static struct rxe_ifc_ops ifc_ops = { + .node_guid = node_guid, + .port_guid = port_guid, + .dma_device = dma_device, + .mcast_add = mcast_add, + .mcast_delete = mcast_delete, + .prepare = prepare, + .send = send, + .loopback = loopback, + .init_packet = init_packet, + .parent_name = parent_name, + .link_layer = link_layer, +}; + +struct rxe_dev *rxe_net_add(struct net_device *ndev) +{ + int err; + struct rxe_dev *rxe = NULL; + + rxe = (struct rxe_dev *)ib_alloc_device(sizeof(*rxe)); + if (!rxe) + return NULL; + + rxe->ifc_ops = &ifc_ops; + rxe->ndev = ndev; + + err = rxe_add(rxe, ndev->mtu); + if (err) { + ib_dealloc_device(&rxe->ib_dev); + return NULL; + } + + spin_lock_bh(&dev_list_lock); + list_add_tail(&rxe_dev_list, &rxe->list); + spin_unlock_bh(&dev_list_lock); + return rxe; +} + +void rxe_remove_all(void) +{ + spin_lock_bh(&dev_list_lock); + while (!list_empty(&rxe_dev_list)) { + struct rxe_dev *rxe = + list_first_entry(&rxe_dev_list, struct rxe_dev, list); + + list_del(&rxe->list); + spin_unlock_bh(&dev_list_lock); + rxe_remove(rxe); + spin_lock_bh(&dev_list_lock); + } + spin_unlock_bh(&dev_list_lock); +} +EXPORT_SYMBOL(rxe_remove_all); + +static void rxe_port_event(struct rxe_dev *rxe, + enum ib_event_type event) +{ + struct ib_event ev; + + ev.device = &rxe->ib_dev; + ev.element.port_num = 1; + ev.event = event; + + ib_dispatch_event(&ev); +} + +/* Caller must hold net_info_lock */ +void rxe_port_up(struct rxe_dev *rxe) +{ + struct rxe_port *port; + + port = &rxe->port; + port->attr.state = IB_PORT_ACTIVE; + port->attr.phys_state = IB_PHYS_STATE_LINK_UP; + + rxe_port_event(rxe, IB_EVENT_PORT_ACTIVE); + pr_info("rxe: set %s active\n", rxe->ib_dev.name); + return; +} + +/* Caller must hold net_info_lock */ +void rxe_port_down(struct rxe_dev *rxe) +{ + struct rxe_port *port; + + port = &rxe->port; + port->attr.state = IB_PORT_DOWN; + port->attr.phys_state = IB_PHYS_STATE_LINK_DOWN; + + rxe_port_event(rxe, IB_EVENT_PORT_ERR); + pr_info("rxe: set %s down\n", rxe->ib_dev.name); + return; +} + +static int rxe_notify(struct notifier_block *not_blk, + unsigned long event, + void *arg) +{ + struct net_device *ndev = netdev_notifier_info_to_dev(arg); + struct rxe_dev *rxe = net_to_rxe(ndev); + + if (!rxe) + goto out; + + switch (event) { + case NETDEV_UNREGISTER: + list_del(&rxe->list); + rxe_remove(rxe); + break; + case NETDEV_UP: + rxe_port_up(rxe); + break; + case NETDEV_DOWN: + rxe_port_down(rxe); + break; + case NETDEV_CHANGEMTU: + pr_info("rxe: %s changed mtu to %d\n", ndev->name, ndev->mtu); + rxe_set_mtu(rxe, ndev->mtu); + break; + case NETDEV_REBOOT: + case NETDEV_CHANGE: + case NETDEV_GOING_DOWN: + case NETDEV_CHANGEADDR: + case NETDEV_CHANGENAME: + case NETDEV_FEAT_CHANGE: + default: + pr_info("rxe: ignoring netdev event = %ld for %s\n", + event, ndev->name); + break; + } +out: + return NOTIFY_OK; +} + +static struct notifier_block rxe_net_notifier = { + .notifier_call = rxe_notify, +}; + +int rxe_net_init(void) +{ + int err; + + spin_lock_init(&dev_list_lock); + + recv_sockets.sk6 = rxe_setup_udp_tunnel(&init_net, + htons(ROCE_V2_UDP_DPORT), true); + if (IS_ERR(recv_sockets.sk6)) { + recv_sockets.sk6 = NULL; + pr_err("rxe: Failed to create IPv6 UDP tunnel\n"); + return -1; + } + + recv_sockets.sk4 = rxe_setup_udp_tunnel(&init_net, + htons(ROCE_V2_UDP_DPORT), false); + if (IS_ERR(recv_sockets.sk4)) { + rxe_release_udp_tunnel(recv_sockets.sk6); + recv_sockets.sk4 = NULL; + recv_sockets.sk6 = NULL; + pr_err("rxe: Failed to create IPv4 UDP tunnel\n"); + return -1; + } + + err = register_netdevice_notifier(&rxe_net_notifier); + if (err) { + rxe_release_udp_tunnel(recv_sockets.sk6); + rxe_release_udp_tunnel(recv_sockets.sk4); + pr_err("rxe: Failed to rigister netdev notifier\n"); + } + + return err; +} + +void rxe_net_exit(void) +{ + if (recv_sockets.sk6) + rxe_release_udp_tunnel(recv_sockets.sk6); + + if (recv_sockets.sk4) + rxe_release_udp_tunnel(recv_sockets.sk4); + + unregister_netdevice_notifier(&rxe_net_notifier); +} diff --git a/drivers/infiniband/sw/rxe/rxe_net.h b/drivers/infiniband/sw/rxe/rxe_net.h new file mode 100644 index 000000000000..7b06f76d16cc --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_net.h @@ -0,0 +1,53 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_NET_H +#define RXE_NET_H + +#include +#include +#include + +struct rxe_recv_sockets { + struct socket *sk4; + struct socket *sk6; +}; + +extern struct rxe_recv_sockets recv_sockets; + +struct rxe_dev *rxe_net_add(struct net_device *ndev); + +int rxe_net_init(void); +void rxe_net_exit(void); + +#endif /* RXE_NET_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_opcode.c b/drivers/infiniband/sw/rxe/rxe_opcode.c new file mode 100644 index 000000000000..61927c165b59 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_opcode.c @@ -0,0 +1,961 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include "rxe_opcode.h" +#include "rxe_hdr.h" + +/* useful information about work request opcodes and pkt opcodes in + * table form + */ +struct rxe_wr_opcode_info rxe_wr_opcode_info[] = { + [IB_WR_RDMA_WRITE] = { + .name = "IB_WR_RDMA_WRITE", + .mask = { + [IB_QPT_RC] = WR_INLINE_MASK | WR_WRITE_MASK, + [IB_QPT_UC] = WR_INLINE_MASK | WR_WRITE_MASK, + }, + }, + [IB_WR_RDMA_WRITE_WITH_IMM] = { + .name = "IB_WR_RDMA_WRITE_WITH_IMM", + .mask = { + [IB_QPT_RC] = WR_INLINE_MASK | WR_WRITE_MASK, + [IB_QPT_UC] = WR_INLINE_MASK | WR_WRITE_MASK, + }, + }, + [IB_WR_SEND] = { + .name = "IB_WR_SEND", + .mask = { + [IB_QPT_SMI] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_GSI] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_RC] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_UC] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_UD] = WR_INLINE_MASK | WR_SEND_MASK, + }, + }, + [IB_WR_SEND_WITH_IMM] = { + .name = "IB_WR_SEND_WITH_IMM", + .mask = { + [IB_QPT_SMI] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_GSI] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_RC] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_UC] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_UD] = WR_INLINE_MASK | WR_SEND_MASK, + }, + }, + [IB_WR_RDMA_READ] = { + .name = "IB_WR_RDMA_READ", + .mask = { + [IB_QPT_RC] = WR_READ_MASK, + }, + }, + [IB_WR_ATOMIC_CMP_AND_SWP] = { + .name = "IB_WR_ATOMIC_CMP_AND_SWP", + .mask = { + [IB_QPT_RC] = WR_ATOMIC_MASK, + }, + }, + [IB_WR_ATOMIC_FETCH_AND_ADD] = { + .name = "IB_WR_ATOMIC_FETCH_AND_ADD", + .mask = { + [IB_QPT_RC] = WR_ATOMIC_MASK, + }, + }, + [IB_WR_LSO] = { + .name = "IB_WR_LSO", + .mask = { + /* not supported */ + }, + }, + [IB_WR_SEND_WITH_INV] = { + .name = "IB_WR_SEND_WITH_INV", + .mask = { + [IB_QPT_RC] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_UC] = WR_INLINE_MASK | WR_SEND_MASK, + [IB_QPT_UD] = WR_INLINE_MASK | WR_SEND_MASK, + }, + }, + [IB_WR_RDMA_READ_WITH_INV] = { + .name = "IB_WR_RDMA_READ_WITH_INV", + .mask = { + [IB_QPT_RC] = WR_READ_MASK, + }, + }, + [IB_WR_LOCAL_INV] = { + .name = "IB_WR_LOCAL_INV", + .mask = { + [IB_QPT_RC] = WR_REG_MASK, + }, + }, + [IB_WR_REG_MR] = { + .name = "IB_WR_REG_MR", + .mask = { + [IB_QPT_RC] = WR_REG_MASK, + }, + }, +}; + +struct rxe_opcode_info rxe_opcode[RXE_NUM_OPCODE] = { + [IB_OPCODE_RC_SEND_FIRST] = { + .name = "IB_OPCODE_RC_SEND_FIRST", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_RWR_MASK + | RXE_SEND_MASK | RXE_START_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_SEND_MIDDLE] = { + .name = "IB_OPCODE_RC_SEND_MIDDLE]", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_SEND_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_SEND_LAST] = { + .name = "IB_OPCODE_RC_SEND_LAST", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_COMP_MASK + | RXE_SEND_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_SEND_LAST_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RC_SEND_LAST_WITH_IMMEDIATE", + .mask = RXE_IMMDT_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_SEND_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IMMDT] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RC_SEND_ONLY] = { + .name = "IB_OPCODE_RC_SEND_ONLY", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_COMP_MASK + | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_SEND_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RC_SEND_ONLY_WITH_IMMEDIATE", + .mask = RXE_IMMDT_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IMMDT] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_WRITE_FIRST] = { + .name = "IB_OPCODE_RC_RDMA_WRITE_FIRST", + .mask = RXE_RETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_START_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_WRITE_MIDDLE] = { + .name = "IB_OPCODE_RC_RDMA_WRITE_MIDDLE", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_WRITE_LAST] = { + .name = "IB_OPCODE_RC_RDMA_WRITE_LAST", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_WRITE_LAST_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RC_RDMA_WRITE_LAST_WITH_IMMEDIATE", + .mask = RXE_IMMDT_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IMMDT] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_WRITE_ONLY] = { + .name = "IB_OPCODE_RC_RDMA_WRITE_ONLY", + .mask = RXE_RETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_START_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_WRITE_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RC_RDMA_WRITE_ONLY_WITH_IMMEDIATE", + .mask = RXE_RETH_MASK | RXE_IMMDT_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_READ_REQUEST] = { + .name = "IB_OPCODE_RC_RDMA_READ_REQUEST", + .mask = RXE_RETH_MASK | RXE_REQ_MASK | RXE_READ_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST] = { + .name = "IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST", + .mask = RXE_AETH_MASK | RXE_PAYLOAD_MASK | RXE_ACK_MASK + | RXE_START_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_AETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE] = { + .name = "IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE", + .mask = RXE_PAYLOAD_MASK | RXE_ACK_MASK | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_READ_RESPONSE_LAST] = { + .name = "IB_OPCODE_RC_RDMA_READ_RESPONSE_LAST", + .mask = RXE_AETH_MASK | RXE_PAYLOAD_MASK | RXE_ACK_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_AETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RC_RDMA_READ_RESPONSE_ONLY] = { + .name = "IB_OPCODE_RC_RDMA_READ_RESPONSE_ONLY", + .mask = RXE_AETH_MASK | RXE_PAYLOAD_MASK | RXE_ACK_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_AETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RC_ACKNOWLEDGE] = { + .name = "IB_OPCODE_RC_ACKNOWLEDGE", + .mask = RXE_AETH_MASK | RXE_ACK_MASK | RXE_START_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_AETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RC_ATOMIC_ACKNOWLEDGE] = { + .name = "IB_OPCODE_RC_ATOMIC_ACKNOWLEDGE", + .mask = RXE_AETH_MASK | RXE_ATMACK_MASK | RXE_ACK_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_ATMACK_BYTES + RXE_AETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_AETH] = RXE_BTH_BYTES, + [RXE_ATMACK] = RXE_BTH_BYTES + + RXE_AETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_ATMACK_BYTES + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RC_COMPARE_SWAP] = { + .name = "IB_OPCODE_RC_COMPARE_SWAP", + .mask = RXE_ATMETH_MASK | RXE_REQ_MASK | RXE_ATOMIC_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_ATMETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_ATMETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_ATMETH_BYTES, + } + }, + [IB_OPCODE_RC_FETCH_ADD] = { + .name = "IB_OPCODE_RC_FETCH_ADD", + .mask = RXE_ATMETH_MASK | RXE_REQ_MASK | RXE_ATOMIC_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_ATMETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_ATMETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_ATMETH_BYTES, + } + }, + [IB_OPCODE_RC_SEND_LAST_WITH_INVALIDATE] = { + .name = "IB_OPCODE_RC_SEND_LAST_WITH_INVALIDATE", + .mask = RXE_IETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_SEND_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IETH_BYTES, + } + }, + [IB_OPCODE_RC_SEND_ONLY_WITH_INVALIDATE] = { + .name = "IB_OPCODE_RC_SEND_ONLY_INV", + .mask = RXE_IETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IETH_BYTES, + } + }, + + /* UC */ + [IB_OPCODE_UC_SEND_FIRST] = { + .name = "IB_OPCODE_UC_SEND_FIRST", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_RWR_MASK + | RXE_SEND_MASK | RXE_START_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_UC_SEND_MIDDLE] = { + .name = "IB_OPCODE_UC_SEND_MIDDLE", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_SEND_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_UC_SEND_LAST] = { + .name = "IB_OPCODE_UC_SEND_LAST", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_COMP_MASK + | RXE_SEND_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_UC_SEND_LAST_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_UC_SEND_LAST_WITH_IMMEDIATE", + .mask = RXE_IMMDT_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_SEND_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IMMDT] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_UC_SEND_ONLY] = { + .name = "IB_OPCODE_UC_SEND_ONLY", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_COMP_MASK + | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_UC_SEND_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_UC_SEND_ONLY_WITH_IMMEDIATE", + .mask = RXE_IMMDT_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IMMDT] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_UC_RDMA_WRITE_FIRST] = { + .name = "IB_OPCODE_UC_RDMA_WRITE_FIRST", + .mask = RXE_RETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_START_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_UC_RDMA_WRITE_MIDDLE] = { + .name = "IB_OPCODE_UC_RDMA_WRITE_MIDDLE", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_UC_RDMA_WRITE_LAST] = { + .name = "IB_OPCODE_UC_RDMA_WRITE_LAST", + .mask = RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_PAYLOAD] = RXE_BTH_BYTES, + } + }, + [IB_OPCODE_UC_RDMA_WRITE_LAST_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_UC_RDMA_WRITE_LAST_WITH_IMMEDIATE", + .mask = RXE_IMMDT_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_IMMDT] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_UC_RDMA_WRITE_ONLY] = { + .name = "IB_OPCODE_UC_RDMA_WRITE_ONLY", + .mask = RXE_RETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_START_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_UC_RDMA_WRITE_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_UC_RDMA_WRITE_ONLY_WITH_IMMEDIATE", + .mask = RXE_RETH_MASK | RXE_IMMDT_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_RETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RETH] = RXE_BTH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_RETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + + /* RD */ + [IB_OPCODE_RD_SEND_FIRST] = { + .name = "IB_OPCODE_RD_SEND_FIRST", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_RD_SEND_MIDDLE] = { + .name = "IB_OPCODE_RD_SEND_MIDDLE", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_SEND_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_RD_SEND_LAST] = { + .name = "IB_OPCODE_RD_SEND_LAST", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_COMP_MASK | RXE_SEND_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_RD_SEND_LAST_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RD_SEND_LAST_WITH_IMMEDIATE", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_IMMDT_MASK + | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_SEND_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RD_SEND_ONLY] = { + .name = "IB_OPCODE_RD_SEND_ONLY", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_SEND_MASK | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_RD_SEND_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RD_SEND_ONLY_WITH_IMMEDIATE", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_IMMDT_MASK + | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_WRITE_FIRST] = { + .name = "IB_OPCODE_RD_RDMA_WRITE_FIRST", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_RETH_MASK + | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_START_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_RETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_WRITE_MIDDLE] = { + .name = "IB_OPCODE_RD_RDMA_WRITE_MIDDLE", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_WRITE_LAST] = { + .name = "IB_OPCODE_RD_RDMA_WRITE_LAST", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_WRITE_LAST_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RD_RDMA_WRITE_LAST_WITH_IMMEDIATE", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_IMMDT_MASK + | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_WRITE_ONLY] = { + .name = "IB_OPCODE_RD_RDMA_WRITE_ONLY", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_RETH_MASK + | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_WRITE_MASK | RXE_START_MASK + | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_RETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_RETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_WRITE_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_RD_RDMA_WRITE_ONLY_WITH_IMMEDIATE", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_RETH_MASK + | RXE_IMMDT_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_WRITE_MASK + | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_RETH_BYTES + + RXE_DETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_RETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_RETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES + + RXE_RETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_READ_REQUEST] = { + .name = "IB_OPCODE_RD_RDMA_READ_REQUEST", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_RETH_MASK + | RXE_REQ_MASK | RXE_READ_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_RETH_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_RETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RETH_BYTES + + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_READ_RESPONSE_FIRST] = { + .name = "IB_OPCODE_RD_RDMA_READ_RESPONSE_FIRST", + .mask = RXE_RDETH_MASK | RXE_AETH_MASK + | RXE_PAYLOAD_MASK | RXE_ACK_MASK + | RXE_START_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_AETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_READ_RESPONSE_MIDDLE] = { + .name = "IB_OPCODE_RD_RDMA_READ_RESPONSE_MIDDLE", + .mask = RXE_RDETH_MASK | RXE_PAYLOAD_MASK | RXE_ACK_MASK + | RXE_MIDDLE_MASK, + .length = RXE_BTH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_READ_RESPONSE_LAST] = { + .name = "IB_OPCODE_RD_RDMA_READ_RESPONSE_LAST", + .mask = RXE_RDETH_MASK | RXE_AETH_MASK | RXE_PAYLOAD_MASK + | RXE_ACK_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_AETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RD_RDMA_READ_RESPONSE_ONLY] = { + .name = "IB_OPCODE_RD_RDMA_READ_RESPONSE_ONLY", + .mask = RXE_RDETH_MASK | RXE_AETH_MASK | RXE_PAYLOAD_MASK + | RXE_ACK_MASK | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_AETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RD_ACKNOWLEDGE] = { + .name = "IB_OPCODE_RD_ACKNOWLEDGE", + .mask = RXE_RDETH_MASK | RXE_AETH_MASK | RXE_ACK_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_AETH_BYTES + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_AETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + } + }, + [IB_OPCODE_RD_ATOMIC_ACKNOWLEDGE] = { + .name = "IB_OPCODE_RD_ATOMIC_ACKNOWLEDGE", + .mask = RXE_RDETH_MASK | RXE_AETH_MASK | RXE_ATMACK_MASK + | RXE_ACK_MASK | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_ATMACK_BYTES + RXE_AETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_AETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_ATMACK] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_AETH_BYTES, + } + }, + [IB_OPCODE_RD_COMPARE_SWAP] = { + .name = "RD_COMPARE_SWAP", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_ATMETH_MASK + | RXE_REQ_MASK | RXE_ATOMIC_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_ATMETH_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_ATMETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + + RXE_ATMETH_BYTES + + RXE_DETH_BYTES + + + RXE_RDETH_BYTES, + } + }, + [IB_OPCODE_RD_FETCH_ADD] = { + .name = "IB_OPCODE_RD_FETCH_ADD", + .mask = RXE_RDETH_MASK | RXE_DETH_MASK | RXE_ATMETH_MASK + | RXE_REQ_MASK | RXE_ATOMIC_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_ATMETH_BYTES + RXE_DETH_BYTES + + RXE_RDETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_RDETH] = RXE_BTH_BYTES, + [RXE_DETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES, + [RXE_ATMETH] = RXE_BTH_BYTES + + RXE_RDETH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + + RXE_ATMETH_BYTES + + RXE_DETH_BYTES + + + RXE_RDETH_BYTES, + } + }, + + /* UD */ + [IB_OPCODE_UD_SEND_ONLY] = { + .name = "IB_OPCODE_UD_SEND_ONLY", + .mask = RXE_DETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK + | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK + | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_DETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_DETH] = RXE_BTH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_DETH_BYTES, + } + }, + [IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE] = { + .name = "IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE", + .mask = RXE_DETH_MASK | RXE_IMMDT_MASK | RXE_PAYLOAD_MASK + | RXE_REQ_MASK | RXE_COMP_MASK | RXE_RWR_MASK + | RXE_SEND_MASK | RXE_START_MASK | RXE_END_MASK, + .length = RXE_BTH_BYTES + RXE_IMMDT_BYTES + RXE_DETH_BYTES, + .offset = { + [RXE_BTH] = 0, + [RXE_DETH] = RXE_BTH_BYTES, + [RXE_IMMDT] = RXE_BTH_BYTES + + RXE_DETH_BYTES, + [RXE_PAYLOAD] = RXE_BTH_BYTES + + RXE_DETH_BYTES + + RXE_IMMDT_BYTES, + } + }, + +}; diff --git a/drivers/infiniband/sw/rxe/rxe_opcode.h b/drivers/infiniband/sw/rxe/rxe_opcode.h new file mode 100644 index 000000000000..307604e9c78d --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_opcode.h @@ -0,0 +1,129 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_OPCODE_H +#define RXE_OPCODE_H + +/* + * contains header bit mask definitions and header lengths + * declaration of the rxe_opcode_info struct and + * rxe_wr_opcode_info struct + */ + +enum rxe_wr_mask { + WR_INLINE_MASK = BIT(0), + WR_ATOMIC_MASK = BIT(1), + WR_SEND_MASK = BIT(2), + WR_READ_MASK = BIT(3), + WR_WRITE_MASK = BIT(4), + WR_LOCAL_MASK = BIT(5), + WR_REG_MASK = BIT(6), + + WR_READ_OR_WRITE_MASK = WR_READ_MASK | WR_WRITE_MASK, + WR_READ_WRITE_OR_SEND_MASK = WR_READ_OR_WRITE_MASK | WR_SEND_MASK, + WR_WRITE_OR_SEND_MASK = WR_WRITE_MASK | WR_SEND_MASK, + WR_ATOMIC_OR_READ_MASK = WR_ATOMIC_MASK | WR_READ_MASK, +}; + +#define WR_MAX_QPT (8) + +struct rxe_wr_opcode_info { + char *name; + enum rxe_wr_mask mask[WR_MAX_QPT]; +}; + +extern struct rxe_wr_opcode_info rxe_wr_opcode_info[]; + +enum rxe_hdr_type { + RXE_LRH, + RXE_GRH, + RXE_BTH, + RXE_RETH, + RXE_AETH, + RXE_ATMETH, + RXE_ATMACK, + RXE_IETH, + RXE_RDETH, + RXE_DETH, + RXE_IMMDT, + RXE_PAYLOAD, + NUM_HDR_TYPES +}; + +enum rxe_hdr_mask { + RXE_LRH_MASK = BIT(RXE_LRH), + RXE_GRH_MASK = BIT(RXE_GRH), + RXE_BTH_MASK = BIT(RXE_BTH), + RXE_IMMDT_MASK = BIT(RXE_IMMDT), + RXE_RETH_MASK = BIT(RXE_RETH), + RXE_AETH_MASK = BIT(RXE_AETH), + RXE_ATMETH_MASK = BIT(RXE_ATMETH), + RXE_ATMACK_MASK = BIT(RXE_ATMACK), + RXE_IETH_MASK = BIT(RXE_IETH), + RXE_RDETH_MASK = BIT(RXE_RDETH), + RXE_DETH_MASK = BIT(RXE_DETH), + RXE_PAYLOAD_MASK = BIT(RXE_PAYLOAD), + + RXE_REQ_MASK = BIT(NUM_HDR_TYPES + 0), + RXE_ACK_MASK = BIT(NUM_HDR_TYPES + 1), + RXE_SEND_MASK = BIT(NUM_HDR_TYPES + 2), + RXE_WRITE_MASK = BIT(NUM_HDR_TYPES + 3), + RXE_READ_MASK = BIT(NUM_HDR_TYPES + 4), + RXE_ATOMIC_MASK = BIT(NUM_HDR_TYPES + 5), + + RXE_RWR_MASK = BIT(NUM_HDR_TYPES + 6), + RXE_COMP_MASK = BIT(NUM_HDR_TYPES + 7), + + RXE_START_MASK = BIT(NUM_HDR_TYPES + 8), + RXE_MIDDLE_MASK = BIT(NUM_HDR_TYPES + 9), + RXE_END_MASK = BIT(NUM_HDR_TYPES + 10), + + RXE_LOOPBACK_MASK = BIT(NUM_HDR_TYPES + 12), + + RXE_READ_OR_ATOMIC = (RXE_READ_MASK | RXE_ATOMIC_MASK), + RXE_WRITE_OR_SEND = (RXE_WRITE_MASK | RXE_SEND_MASK), +}; + +#define OPCODE_NONE (-1) +#define RXE_NUM_OPCODE 256 + +struct rxe_opcode_info { + char *name; + enum rxe_hdr_mask mask; + int length; + int offset[NUM_HDR_TYPES]; +}; + +extern struct rxe_opcode_info rxe_opcode[RXE_NUM_OPCODE]; + +#endif /* RXE_OPCODE_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_param.h b/drivers/infiniband/sw/rxe/rxe_param.h new file mode 100644 index 000000000000..f459c43a77c8 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_param.h @@ -0,0 +1,172 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_PARAM_H +#define RXE_PARAM_H + +static inline enum ib_mtu rxe_mtu_int_to_enum(int mtu) +{ + if (mtu < 256) + return 0; + else if (mtu < 512) + return IB_MTU_256; + else if (mtu < 1024) + return IB_MTU_512; + else if (mtu < 2048) + return IB_MTU_1024; + else if (mtu < 4096) + return IB_MTU_2048; + else + return IB_MTU_4096; +} + +/* Find the IB mtu for a given network MTU. */ +static inline enum ib_mtu eth_mtu_int_to_enum(int mtu) +{ + mtu -= RXE_MAX_HDR_LENGTH; + + return rxe_mtu_int_to_enum(mtu); +} + +/* default/initial rxe device parameter settings */ +enum rxe_device_param { + RXE_FW_VER = 0, + RXE_MAX_MR_SIZE = -1ull, + RXE_PAGE_SIZE_CAP = 0xfffff000, + RXE_VENDOR_ID = 0, + RXE_VENDOR_PART_ID = 0, + RXE_HW_VER = 0, + RXE_MAX_QP = 0x10000, + RXE_MAX_QP_WR = 0x4000, + RXE_MAX_INLINE_DATA = 400, + RXE_DEVICE_CAP_FLAGS = IB_DEVICE_BAD_PKEY_CNTR + | IB_DEVICE_BAD_QKEY_CNTR + | IB_DEVICE_AUTO_PATH_MIG + | IB_DEVICE_CHANGE_PHY_PORT + | IB_DEVICE_UD_AV_PORT_ENFORCE + | IB_DEVICE_PORT_ACTIVE_EVENT + | IB_DEVICE_SYS_IMAGE_GUID + | IB_DEVICE_RC_RNR_NAK_GEN + | IB_DEVICE_SRQ_RESIZE + | IB_DEVICE_MEM_MGT_EXTENSIONS, + RXE_MAX_SGE = 32, + RXE_MAX_SGE_RD = 32, + RXE_MAX_CQ = 16384, + RXE_MAX_LOG_CQE = 13, + RXE_MAX_MR = 2 * 1024, + RXE_MAX_PD = 0x7ffc, + RXE_MAX_QP_RD_ATOM = 128, + RXE_MAX_EE_RD_ATOM = 0, + RXE_MAX_RES_RD_ATOM = 0x3f000, + RXE_MAX_QP_INIT_RD_ATOM = 128, + RXE_MAX_EE_INIT_RD_ATOM = 0, + RXE_ATOMIC_CAP = 1, + RXE_MAX_EE = 0, + RXE_MAX_RDD = 0, + RXE_MAX_MW = 0, + RXE_MAX_RAW_IPV6_QP = 0, + RXE_MAX_RAW_ETHY_QP = 0, + RXE_MAX_MCAST_GRP = 8192, + RXE_MAX_MCAST_QP_ATTACH = 56, + RXE_MAX_TOT_MCAST_QP_ATTACH = 0x70000, + RXE_MAX_AH = 100, + RXE_MAX_FMR = 0, + RXE_MAX_MAP_PER_FMR = 0, + RXE_MAX_SRQ = 960, + RXE_MAX_SRQ_WR = 0x4000, + RXE_MIN_SRQ_WR = 1, + RXE_MAX_SRQ_SGE = 27, + RXE_MIN_SRQ_SGE = 1, + RXE_MAX_FMR_PAGE_LIST_LEN = 512, + RXE_MAX_PKEYS = 64, + RXE_LOCAL_CA_ACK_DELAY = 15, + + RXE_MAX_UCONTEXT = 512, + + RXE_NUM_PORT = 1, + RXE_NUM_COMP_VECTORS = 1, + + RXE_MIN_QP_INDEX = 16, + RXE_MAX_QP_INDEX = 0x00020000, + + RXE_MIN_SRQ_INDEX = 0x00020001, + RXE_MAX_SRQ_INDEX = 0x00040000, + + RXE_MIN_MR_INDEX = 0x00000001, + RXE_MAX_MR_INDEX = 0x00040000, + RXE_MIN_MW_INDEX = 0x00040001, + RXE_MAX_MW_INDEX = 0x00060000, + RXE_MAX_PKT_PER_ACK = 64, + + RXE_MAX_UNACKED_PSNS = 128, + + /* Max inflight SKBs per queue pair */ + RXE_INFLIGHT_SKBS_PER_QP_HIGH = 64, + RXE_INFLIGHT_SKBS_PER_QP_LOW = 16, + + /* Delay before calling arbiter timer */ + RXE_NSEC_ARB_TIMER_DELAY = 200, +}; + +/* default/initial rxe port parameters */ +enum rxe_port_param { + RXE_PORT_STATE = IB_PORT_DOWN, + RXE_PORT_MAX_MTU = IB_MTU_4096, + RXE_PORT_ACTIVE_MTU = IB_MTU_256, + RXE_PORT_GID_TBL_LEN = 1024, + RXE_PORT_PORT_CAP_FLAGS = RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP, + RXE_PORT_MAX_MSG_SZ = 0x800000, + RXE_PORT_BAD_PKEY_CNTR = 0, + RXE_PORT_QKEY_VIOL_CNTR = 0, + RXE_PORT_LID = 0, + RXE_PORT_SM_LID = 0, + RXE_PORT_SM_SL = 0, + RXE_PORT_LMC = 0, + RXE_PORT_MAX_VL_NUM = 1, + RXE_PORT_SUBNET_TIMEOUT = 0, + RXE_PORT_INIT_TYPE_REPLY = 0, + RXE_PORT_ACTIVE_WIDTH = IB_WIDTH_1X, + RXE_PORT_ACTIVE_SPEED = 1, + RXE_PORT_PKEY_TBL_LEN = 64, + RXE_PORT_PHYS_STATE = 2, + RXE_PORT_SUBNET_PREFIX = 0xfe80000000000000ULL, +}; + +/* default/initial port info parameters */ +enum rxe_port_info_param { + RXE_PORT_INFO_VL_CAP = 4, /* 1-8 */ + RXE_PORT_INFO_MTU_CAP = 5, /* 4096 */ + RXE_PORT_INFO_OPER_VL = 1, /* 1 */ +}; + +#endif /* RXE_PARAM_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_pool.c b/drivers/infiniband/sw/rxe/rxe_pool.c new file mode 100644 index 000000000000..6bac0717c540 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_pool.c @@ -0,0 +1,502 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" + +/* info about object pools + * note that mr and mw share a single index space + * so that one can map an lkey to the correct type of object + */ +struct rxe_type_info rxe_type_info[RXE_NUM_TYPES] = { + [RXE_TYPE_UC] = { + .name = "rxe-uc", + .size = sizeof(struct rxe_ucontext), + }, + [RXE_TYPE_PD] = { + .name = "rxe-pd", + .size = sizeof(struct rxe_pd), + }, + [RXE_TYPE_AH] = { + .name = "rxe-ah", + .size = sizeof(struct rxe_ah), + .flags = RXE_POOL_ATOMIC, + }, + [RXE_TYPE_SRQ] = { + .name = "rxe-srq", + .size = sizeof(struct rxe_srq), + .flags = RXE_POOL_INDEX, + .min_index = RXE_MIN_SRQ_INDEX, + .max_index = RXE_MAX_SRQ_INDEX, + }, + [RXE_TYPE_QP] = { + .name = "rxe-qp", + .size = sizeof(struct rxe_qp), + .cleanup = rxe_qp_cleanup, + .flags = RXE_POOL_INDEX, + .min_index = RXE_MIN_QP_INDEX, + .max_index = RXE_MAX_QP_INDEX, + }, + [RXE_TYPE_CQ] = { + .name = "rxe-cq", + .size = sizeof(struct rxe_cq), + .cleanup = rxe_cq_cleanup, + }, + [RXE_TYPE_MR] = { + .name = "rxe-mr", + .size = sizeof(struct rxe_mem), + .cleanup = rxe_mem_cleanup, + .flags = RXE_POOL_INDEX, + .max_index = RXE_MAX_MR_INDEX, + .min_index = RXE_MIN_MR_INDEX, + }, + [RXE_TYPE_MW] = { + .name = "rxe-mw", + .size = sizeof(struct rxe_mem), + .flags = RXE_POOL_INDEX, + .max_index = RXE_MAX_MW_INDEX, + .min_index = RXE_MIN_MW_INDEX, + }, + [RXE_TYPE_MC_GRP] = { + .name = "rxe-mc_grp", + .size = sizeof(struct rxe_mc_grp), + .cleanup = rxe_mc_cleanup, + .flags = RXE_POOL_KEY, + .key_offset = offsetof(struct rxe_mc_grp, mgid), + .key_size = sizeof(union ib_gid), + }, + [RXE_TYPE_MC_ELEM] = { + .name = "rxe-mc_elem", + .size = sizeof(struct rxe_mc_elem), + .flags = RXE_POOL_ATOMIC, + }, +}; + +static inline char *pool_name(struct rxe_pool *pool) +{ + return rxe_type_info[pool->type].name; +} + +static inline struct kmem_cache *pool_cache(struct rxe_pool *pool) +{ + return rxe_type_info[pool->type].cache; +} + +static inline enum rxe_elem_type rxe_type(void *arg) +{ + struct rxe_pool_entry *elem = arg; + + return elem->pool->type; +} + +int rxe_cache_init(void) +{ + int err; + int i; + size_t size; + struct rxe_type_info *type; + + for (i = 0; i < RXE_NUM_TYPES; i++) { + type = &rxe_type_info[i]; + size = ALIGN(type->size, RXE_POOL_ALIGN); + type->cache = kmem_cache_create(type->name, size, + RXE_POOL_ALIGN, + RXE_POOL_CACHE_FLAGS, NULL); + if (!type->cache) { + pr_err("Unable to init kmem cache for %s\n", + type->name); + err = -ENOMEM; + goto err1; + } + } + + return 0; + +err1: + while (--i >= 0) { + kmem_cache_destroy(type->cache); + type->cache = NULL; + } + + return err; +} + +void rxe_cache_exit(void) +{ + int i; + struct rxe_type_info *type; + + for (i = 0; i < RXE_NUM_TYPES; i++) { + type = &rxe_type_info[i]; + kmem_cache_destroy(type->cache); + type->cache = NULL; + } +} + +static int rxe_pool_init_index(struct rxe_pool *pool, u32 max, u32 min) +{ + int err = 0; + size_t size; + + if ((max - min + 1) < pool->max_elem) { + pr_warn("not enough indices for max_elem\n"); + err = -EINVAL; + goto out; + } + + pool->max_index = max; + pool->min_index = min; + + size = BITS_TO_LONGS(max - min + 1) * sizeof(long); + pool->table = kmalloc(size, GFP_KERNEL); + if (!pool->table) { + pr_warn("no memory for bit table\n"); + err = -ENOMEM; + goto out; + } + + pool->table_size = size; + bitmap_zero(pool->table, max - min + 1); + +out: + return err; +} + +int rxe_pool_init( + struct rxe_dev *rxe, + struct rxe_pool *pool, + enum rxe_elem_type type, + unsigned max_elem) +{ + int err = 0; + size_t size = rxe_type_info[type].size; + + memset(pool, 0, sizeof(*pool)); + + pool->rxe = rxe; + pool->type = type; + pool->max_elem = max_elem; + pool->elem_size = ALIGN(size, RXE_POOL_ALIGN); + pool->flags = rxe_type_info[type].flags; + pool->tree = RB_ROOT; + pool->cleanup = rxe_type_info[type].cleanup; + + atomic_set(&pool->num_elem, 0); + + kref_init(&pool->ref_cnt); + + spin_lock_init(&pool->pool_lock); + + if (rxe_type_info[type].flags & RXE_POOL_INDEX) { + err = rxe_pool_init_index(pool, + rxe_type_info[type].max_index, + rxe_type_info[type].min_index); + if (err) + goto out; + } + + if (rxe_type_info[type].flags & RXE_POOL_KEY) { + pool->key_offset = rxe_type_info[type].key_offset; + pool->key_size = rxe_type_info[type].key_size; + } + + pool->state = rxe_pool_valid; + +out: + return err; +} + +static void rxe_pool_release(struct kref *kref) +{ + struct rxe_pool *pool = container_of(kref, struct rxe_pool, ref_cnt); + + pool->state = rxe_pool_invalid; + kfree(pool->table); +} + +static void rxe_pool_put(struct rxe_pool *pool) +{ + kref_put(&pool->ref_cnt, rxe_pool_release); +} + +int rxe_pool_cleanup(struct rxe_pool *pool) +{ + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + pool->state = rxe_pool_invalid; + if (atomic_read(&pool->num_elem) > 0) + pr_warn("%s pool destroyed with unfree'd elem\n", + pool_name(pool)); + spin_unlock_irqrestore(&pool->pool_lock, flags); + + rxe_pool_put(pool); + + return 0; +} + +static u32 alloc_index(struct rxe_pool *pool) +{ + u32 index; + u32 range = pool->max_index - pool->min_index + 1; + + index = find_next_zero_bit(pool->table, range, pool->last); + if (index >= range) + index = find_first_zero_bit(pool->table, range); + + set_bit(index, pool->table); + pool->last = index; + return index + pool->min_index; +} + +static void insert_index(struct rxe_pool *pool, struct rxe_pool_entry *new) +{ + struct rb_node **link = &pool->tree.rb_node; + struct rb_node *parent = NULL; + struct rxe_pool_entry *elem; + + while (*link) { + parent = *link; + elem = rb_entry(parent, struct rxe_pool_entry, node); + + if (elem->index == new->index) { + pr_warn("element already exists!\n"); + goto out; + } + + if (elem->index > new->index) + link = &(*link)->rb_left; + else + link = &(*link)->rb_right; + } + + rb_link_node(&new->node, parent, link); + rb_insert_color(&new->node, &pool->tree); +out: + return; +} + +static void insert_key(struct rxe_pool *pool, struct rxe_pool_entry *new) +{ + struct rb_node **link = &pool->tree.rb_node; + struct rb_node *parent = NULL; + struct rxe_pool_entry *elem; + int cmp; + + while (*link) { + parent = *link; + elem = rb_entry(parent, struct rxe_pool_entry, node); + + cmp = memcmp((u8 *)elem + pool->key_offset, + (u8 *)new + pool->key_offset, pool->key_size); + + if (cmp == 0) { + pr_warn("key already exists!\n"); + goto out; + } + + if (cmp > 0) + link = &(*link)->rb_left; + else + link = &(*link)->rb_right; + } + + rb_link_node(&new->node, parent, link); + rb_insert_color(&new->node, &pool->tree); +out: + return; +} + +void rxe_add_key(void *arg, void *key) +{ + struct rxe_pool_entry *elem = arg; + struct rxe_pool *pool = elem->pool; + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + memcpy((u8 *)elem + pool->key_offset, key, pool->key_size); + insert_key(pool, elem); + spin_unlock_irqrestore(&pool->pool_lock, flags); +} + +void rxe_drop_key(void *arg) +{ + struct rxe_pool_entry *elem = arg; + struct rxe_pool *pool = elem->pool; + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + rb_erase(&elem->node, &pool->tree); + spin_unlock_irqrestore(&pool->pool_lock, flags); +} + +void rxe_add_index(void *arg) +{ + struct rxe_pool_entry *elem = arg; + struct rxe_pool *pool = elem->pool; + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + elem->index = alloc_index(pool); + insert_index(pool, elem); + spin_unlock_irqrestore(&pool->pool_lock, flags); +} + +void rxe_drop_index(void *arg) +{ + struct rxe_pool_entry *elem = arg; + struct rxe_pool *pool = elem->pool; + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + clear_bit(elem->index - pool->min_index, pool->table); + rb_erase(&elem->node, &pool->tree); + spin_unlock_irqrestore(&pool->pool_lock, flags); +} + +void *rxe_alloc(struct rxe_pool *pool) +{ + struct rxe_pool_entry *elem; + unsigned long flags; + + might_sleep_if(!(pool->flags & RXE_POOL_ATOMIC)); + + spin_lock_irqsave(&pool->pool_lock, flags); + if (pool->state != rxe_pool_valid) { + spin_unlock_irqrestore(&pool->pool_lock, flags); + return NULL; + } + kref_get(&pool->ref_cnt); + spin_unlock_irqrestore(&pool->pool_lock, flags); + + kref_get(&pool->rxe->ref_cnt); + + if (atomic_inc_return(&pool->num_elem) > pool->max_elem) { + atomic_dec(&pool->num_elem); + rxe_dev_put(pool->rxe); + rxe_pool_put(pool); + return NULL; + } + + elem = kmem_cache_zalloc(pool_cache(pool), + (pool->flags & RXE_POOL_ATOMIC) ? + GFP_ATOMIC : GFP_KERNEL); + + elem->pool = pool; + kref_init(&elem->ref_cnt); + + return elem; +} + +void rxe_elem_release(struct kref *kref) +{ + struct rxe_pool_entry *elem = + container_of(kref, struct rxe_pool_entry, ref_cnt); + struct rxe_pool *pool = elem->pool; + + if (pool->cleanup) + pool->cleanup(elem); + + kmem_cache_free(pool_cache(pool), elem); + atomic_dec(&pool->num_elem); + rxe_dev_put(pool->rxe); + rxe_pool_put(pool); +} + +void *rxe_pool_get_index(struct rxe_pool *pool, u32 index) +{ + struct rb_node *node = NULL; + struct rxe_pool_entry *elem = NULL; + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + + if (pool->state != rxe_pool_valid) + goto out; + + node = pool->tree.rb_node; + + while (node) { + elem = rb_entry(node, struct rxe_pool_entry, node); + + if (elem->index > index) + node = node->rb_left; + else if (elem->index < index) + node = node->rb_right; + else + break; + } + + if (node) + kref_get(&elem->ref_cnt); + +out: + spin_unlock_irqrestore(&pool->pool_lock, flags); + return node ? (void *)elem : NULL; +} + +void *rxe_pool_get_key(struct rxe_pool *pool, void *key) +{ + struct rb_node *node = NULL; + struct rxe_pool_entry *elem = NULL; + int cmp; + unsigned long flags; + + spin_lock_irqsave(&pool->pool_lock, flags); + + if (pool->state != rxe_pool_valid) + goto out; + + node = pool->tree.rb_node; + + while (node) { + elem = rb_entry(node, struct rxe_pool_entry, node); + + cmp = memcmp((u8 *)elem + pool->key_offset, + key, pool->key_size); + + if (cmp > 0) + node = node->rb_left; + else if (cmp < 0) + node = node->rb_right; + else + break; + } + + if (node) + kref_get(&elem->ref_cnt); + +out: + spin_unlock_irqrestore(&pool->pool_lock, flags); + return node ? ((void *)elem) : NULL; +} diff --git a/drivers/infiniband/sw/rxe/rxe_pool.h b/drivers/infiniband/sw/rxe/rxe_pool.h new file mode 100644 index 000000000000..4d04830adcae --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_pool.h @@ -0,0 +1,163 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_POOL_H +#define RXE_POOL_H + +#define RXE_POOL_ALIGN (16) +#define RXE_POOL_CACHE_FLAGS (0) + +enum rxe_pool_flags { + RXE_POOL_ATOMIC = BIT(0), + RXE_POOL_INDEX = BIT(1), + RXE_POOL_KEY = BIT(2), +}; + +enum rxe_elem_type { + RXE_TYPE_UC, + RXE_TYPE_PD, + RXE_TYPE_AH, + RXE_TYPE_SRQ, + RXE_TYPE_QP, + RXE_TYPE_CQ, + RXE_TYPE_MR, + RXE_TYPE_MW, + RXE_TYPE_MC_GRP, + RXE_TYPE_MC_ELEM, + RXE_NUM_TYPES, /* keep me last */ +}; + +struct rxe_type_info { + char *name; + size_t size; + void (*cleanup)(void *obj); + enum rxe_pool_flags flags; + u32 max_index; + u32 min_index; + size_t key_offset; + size_t key_size; + struct kmem_cache *cache; +}; + +extern struct rxe_type_info rxe_type_info[]; + +enum rxe_pool_state { + rxe_pool_invalid, + rxe_pool_valid, +}; + +struct rxe_pool_entry { + struct rxe_pool *pool; + struct kref ref_cnt; + struct list_head list; + + /* only used if indexed or keyed */ + struct rb_node node; + u32 index; +}; + +struct rxe_pool { + struct rxe_dev *rxe; + spinlock_t pool_lock; /* pool spinlock */ + size_t elem_size; + struct kref ref_cnt; + void (*cleanup)(void *obj); + enum rxe_pool_state state; + enum rxe_pool_flags flags; + enum rxe_elem_type type; + + unsigned int max_elem; + atomic_t num_elem; + + /* only used if indexed or keyed */ + struct rb_root tree; + unsigned long *table; + size_t table_size; + u32 max_index; + u32 min_index; + u32 last; + size_t key_offset; + size_t key_size; +}; + +/* initialize slab caches for managed objects */ +int rxe_cache_init(void); + +/* cleanup slab caches for managed objects */ +void rxe_cache_exit(void); + +/* initialize a pool of objects with given limit on + * number of elements. gets parameters from rxe_type_info + * pool elements will be allocated out of a slab cache + */ +int rxe_pool_init(struct rxe_dev *rxe, struct rxe_pool *pool, + enum rxe_elem_type type, u32 max_elem); + +/* free resources from object pool */ +int rxe_pool_cleanup(struct rxe_pool *pool); + +/* allocate an object from pool */ +void *rxe_alloc(struct rxe_pool *pool); + +/* assign an index to an indexed object and insert object into + * pool's rb tree + */ +void rxe_add_index(void *elem); + +/* drop an index and remove object from rb tree */ +void rxe_drop_index(void *elem); + +/* assign a key to a keyed object and insert object into + * pool's rb tree + */ +void rxe_add_key(void *elem, void *key); + +/* remove elem from rb tree */ +void rxe_drop_key(void *elem); + +/* lookup an indexed object from index. takes a reference on object */ +void *rxe_pool_get_index(struct rxe_pool *pool, u32 index); + +/* lookup keyed object from key. takes a reference on the object */ +void *rxe_pool_get_key(struct rxe_pool *pool, void *key); + +/* cleanup an object when all references are dropped */ +void rxe_elem_release(struct kref *kref); + +/* take a reference on an object */ +#define rxe_add_ref(elem) kref_get(&(elem)->pelem.ref_cnt) + +/* drop a reference on an object */ +#define rxe_drop_ref(elem) kref_put(&(elem)->pelem.ref_cnt, rxe_elem_release) + +#endif /* RXE_POOL_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_qp.c b/drivers/infiniband/sw/rxe/rxe_qp.c new file mode 100644 index 000000000000..22ba24f2a2c1 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -0,0 +1,851 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" +#include "rxe_task.h" + +char *rxe_qp_state_name[] = { + [QP_STATE_RESET] = "RESET", + [QP_STATE_INIT] = "INIT", + [QP_STATE_READY] = "READY", + [QP_STATE_DRAIN] = "DRAIN", + [QP_STATE_DRAINED] = "DRAINED", + [QP_STATE_ERROR] = "ERROR", +}; + +static int rxe_qp_chk_cap(struct rxe_dev *rxe, struct ib_qp_cap *cap, + int has_srq) +{ + if (cap->max_send_wr > rxe->attr.max_qp_wr) { + pr_warn("invalid send wr = %d > %d\n", + cap->max_send_wr, rxe->attr.max_qp_wr); + goto err1; + } + + if (cap->max_send_sge > rxe->attr.max_sge) { + pr_warn("invalid send sge = %d > %d\n", + cap->max_send_sge, rxe->attr.max_sge); + goto err1; + } + + if (!has_srq) { + if (cap->max_recv_wr > rxe->attr.max_qp_wr) { + pr_warn("invalid recv wr = %d > %d\n", + cap->max_recv_wr, rxe->attr.max_qp_wr); + goto err1; + } + + if (cap->max_recv_sge > rxe->attr.max_sge) { + pr_warn("invalid recv sge = %d > %d\n", + cap->max_recv_sge, rxe->attr.max_sge); + goto err1; + } + } + + if (cap->max_inline_data > rxe->max_inline_data) { + pr_warn("invalid max inline data = %d > %d\n", + cap->max_inline_data, rxe->max_inline_data); + goto err1; + } + + return 0; + +err1: + return -EINVAL; +} + +int rxe_qp_chk_init(struct rxe_dev *rxe, struct ib_qp_init_attr *init) +{ + struct ib_qp_cap *cap = &init->cap; + struct rxe_port *port; + int port_num = init->port_num; + + if (!init->recv_cq || !init->send_cq) { + pr_warn("missing cq\n"); + goto err1; + } + + if (rxe_qp_chk_cap(rxe, cap, !!init->srq)) + goto err1; + + if (init->qp_type == IB_QPT_SMI || init->qp_type == IB_QPT_GSI) { + if (port_num != 1) { + pr_warn("invalid port = %d\n", port_num); + goto err1; + } + + port = &rxe->port; + + if (init->qp_type == IB_QPT_SMI && port->qp_smi_index) { + pr_warn("SMI QP exists for port %d\n", port_num); + goto err1; + } + + if (init->qp_type == IB_QPT_GSI && port->qp_gsi_index) { + pr_warn("GSI QP exists for port %d\n", port_num); + goto err1; + } + } + + return 0; + +err1: + return -EINVAL; +} + +static int alloc_rd_atomic_resources(struct rxe_qp *qp, unsigned int n) +{ + qp->resp.res_head = 0; + qp->resp.res_tail = 0; + qp->resp.resources = kcalloc(n, sizeof(struct resp_res), GFP_KERNEL); + + if (!qp->resp.resources) + return -ENOMEM; + + return 0; +} + +static void free_rd_atomic_resources(struct rxe_qp *qp) +{ + if (qp->resp.resources) { + int i; + + for (i = 0; i < qp->attr.max_rd_atomic; i++) { + struct resp_res *res = &qp->resp.resources[i]; + + free_rd_atomic_resource(qp, res); + } + kfree(qp->resp.resources); + qp->resp.resources = NULL; + } +} + +void free_rd_atomic_resource(struct rxe_qp *qp, struct resp_res *res) +{ + if (res->type == RXE_ATOMIC_MASK) { + rxe_drop_ref(qp); + kfree_skb(res->atomic.skb); + } else if (res->type == RXE_READ_MASK) { + if (res->read.mr) + rxe_drop_ref(res->read.mr); + } + res->type = 0; +} + +static void cleanup_rd_atomic_resources(struct rxe_qp *qp) +{ + int i; + struct resp_res *res; + + if (qp->resp.resources) { + for (i = 0; i < qp->attr.max_rd_atomic; i++) { + res = &qp->resp.resources[i]; + free_rd_atomic_resource(qp, res); + } + } +} + +static void rxe_qp_init_misc(struct rxe_dev *rxe, struct rxe_qp *qp, + struct ib_qp_init_attr *init) +{ + struct rxe_port *port; + u32 qpn; + + qp->sq_sig_type = init->sq_sig_type; + qp->attr.path_mtu = 1; + qp->mtu = ib_mtu_enum_to_int(qp->attr.path_mtu); + + qpn = qp->pelem.index; + port = &rxe->port; + + switch (init->qp_type) { + case IB_QPT_SMI: + qp->ibqp.qp_num = 0; + port->qp_smi_index = qpn; + qp->attr.port_num = init->port_num; + break; + + case IB_QPT_GSI: + qp->ibqp.qp_num = 1; + port->qp_gsi_index = qpn; + qp->attr.port_num = init->port_num; + break; + + default: + qp->ibqp.qp_num = qpn; + break; + } + + INIT_LIST_HEAD(&qp->grp_list); + + skb_queue_head_init(&qp->send_pkts); + + spin_lock_init(&qp->grp_lock); + spin_lock_init(&qp->state_lock); + + atomic_set(&qp->ssn, 0); + atomic_set(&qp->skb_out, 0); +} + +static int rxe_qp_init_req(struct rxe_dev *rxe, struct rxe_qp *qp, + struct ib_qp_init_attr *init, + struct ib_ucontext *context, struct ib_udata *udata) +{ + int err; + int wqe_size; + + err = sock_create_kern(&init_net, AF_INET, SOCK_DGRAM, 0, &qp->sk); + if (err < 0) + return err; + qp->sk->sk->sk_user_data = qp; + + qp->sq.max_wr = init->cap.max_send_wr; + qp->sq.max_sge = init->cap.max_send_sge; + qp->sq.max_inline = init->cap.max_inline_data; + + wqe_size = max_t(int, sizeof(struct rxe_send_wqe) + + qp->sq.max_sge * sizeof(struct ib_sge), + sizeof(struct rxe_send_wqe) + + qp->sq.max_inline); + + qp->sq.queue = rxe_queue_init(rxe, + &qp->sq.max_wr, + wqe_size); + if (!qp->sq.queue) + return -ENOMEM; + + err = do_mmap_info(rxe, udata, true, + context, qp->sq.queue->buf, + qp->sq.queue->buf_size, &qp->sq.queue->ip); + + if (err) { + kvfree(qp->sq.queue->buf); + kfree(qp->sq.queue); + return err; + } + + qp->req.wqe_index = producer_index(qp->sq.queue); + qp->req.state = QP_STATE_RESET; + qp->req.opcode = -1; + qp->comp.opcode = -1; + + spin_lock_init(&qp->sq.sq_lock); + skb_queue_head_init(&qp->req_pkts); + + rxe_init_task(rxe, &qp->req.task, qp, + rxe_requester, "req"); + rxe_init_task(rxe, &qp->comp.task, qp, + rxe_completer, "comp"); + + init_timer(&qp->rnr_nak_timer); + qp->rnr_nak_timer.function = rnr_nak_timer; + qp->rnr_nak_timer.data = (unsigned long)qp; + + init_timer(&qp->retrans_timer); + qp->retrans_timer.function = retransmit_timer; + qp->retrans_timer.data = (unsigned long)qp; + qp->qp_timeout_jiffies = 0; /* Can't be set for UD/UC in modify_qp */ + + return 0; +} + +static int rxe_qp_init_resp(struct rxe_dev *rxe, struct rxe_qp *qp, + struct ib_qp_init_attr *init, + struct ib_ucontext *context, struct ib_udata *udata) +{ + int err; + int wqe_size; + + if (!qp->srq) { + qp->rq.max_wr = init->cap.max_recv_wr; + qp->rq.max_sge = init->cap.max_recv_sge; + + wqe_size = rcv_wqe_size(qp->rq.max_sge); + + pr_debug("max_wr = %d, max_sge = %d, wqe_size = %d\n", + qp->rq.max_wr, qp->rq.max_sge, wqe_size); + + qp->rq.queue = rxe_queue_init(rxe, + &qp->rq.max_wr, + wqe_size); + if (!qp->rq.queue) + return -ENOMEM; + + err = do_mmap_info(rxe, udata, false, context, + qp->rq.queue->buf, + qp->rq.queue->buf_size, + &qp->rq.queue->ip); + if (err) { + kvfree(qp->rq.queue->buf); + kfree(qp->rq.queue); + return err; + } + } + + spin_lock_init(&qp->rq.producer_lock); + spin_lock_init(&qp->rq.consumer_lock); + + skb_queue_head_init(&qp->resp_pkts); + + rxe_init_task(rxe, &qp->resp.task, qp, + rxe_responder, "resp"); + + qp->resp.opcode = OPCODE_NONE; + qp->resp.msn = 0; + qp->resp.state = QP_STATE_RESET; + + return 0; +} + +/* called by the create qp verb */ +int rxe_qp_from_init(struct rxe_dev *rxe, struct rxe_qp *qp, struct rxe_pd *pd, + struct ib_qp_init_attr *init, struct ib_udata *udata, + struct ib_pd *ibpd) +{ + int err; + struct rxe_cq *rcq = to_rcq(init->recv_cq); + struct rxe_cq *scq = to_rcq(init->send_cq); + struct rxe_srq *srq = init->srq ? to_rsrq(init->srq) : NULL; + struct ib_ucontext *context = udata ? ibpd->uobject->context : NULL; + + rxe_add_ref(pd); + rxe_add_ref(rcq); + rxe_add_ref(scq); + if (srq) + rxe_add_ref(srq); + + qp->pd = pd; + qp->rcq = rcq; + qp->scq = scq; + qp->srq = srq; + + rxe_qp_init_misc(rxe, qp, init); + + err = rxe_qp_init_req(rxe, qp, init, context, udata); + if (err) + goto err1; + + err = rxe_qp_init_resp(rxe, qp, init, context, udata); + if (err) + goto err2; + + qp->attr.qp_state = IB_QPS_RESET; + qp->valid = 1; + + return 0; + +err2: + rxe_queue_cleanup(qp->sq.queue); +err1: + if (srq) + rxe_drop_ref(srq); + rxe_drop_ref(scq); + rxe_drop_ref(rcq); + rxe_drop_ref(pd); + + return err; +} + +/* called by the query qp verb */ +int rxe_qp_to_init(struct rxe_qp *qp, struct ib_qp_init_attr *init) +{ + init->event_handler = qp->ibqp.event_handler; + init->qp_context = qp->ibqp.qp_context; + init->send_cq = qp->ibqp.send_cq; + init->recv_cq = qp->ibqp.recv_cq; + init->srq = qp->ibqp.srq; + + init->cap.max_send_wr = qp->sq.max_wr; + init->cap.max_send_sge = qp->sq.max_sge; + init->cap.max_inline_data = qp->sq.max_inline; + + if (!qp->srq) { + init->cap.max_recv_wr = qp->rq.max_wr; + init->cap.max_recv_sge = qp->rq.max_sge; + } + + init->sq_sig_type = qp->sq_sig_type; + + init->qp_type = qp->ibqp.qp_type; + init->port_num = 1; + + return 0; +} + +/* called by the modify qp verb, this routine checks all the parameters before + * making any changes + */ +int rxe_qp_chk_attr(struct rxe_dev *rxe, struct rxe_qp *qp, + struct ib_qp_attr *attr, int mask) +{ + enum ib_qp_state cur_state = (mask & IB_QP_CUR_STATE) ? + attr->cur_qp_state : qp->attr.qp_state; + enum ib_qp_state new_state = (mask & IB_QP_STATE) ? + attr->qp_state : cur_state; + + if (!ib_modify_qp_is_ok(cur_state, new_state, qp_type(qp), mask, + IB_LINK_LAYER_ETHERNET)) { + pr_warn("invalid mask or state for qp\n"); + goto err1; + } + + if (mask & IB_QP_STATE) { + if (cur_state == IB_QPS_SQD) { + if (qp->req.state == QP_STATE_DRAIN && + new_state != IB_QPS_ERR) + goto err1; + } + } + + if (mask & IB_QP_PORT) { + if (attr->port_num != 1) { + pr_warn("invalid port %d\n", attr->port_num); + goto err1; + } + } + + if (mask & IB_QP_CAP && rxe_qp_chk_cap(rxe, &attr->cap, !!qp->srq)) + goto err1; + + if (mask & IB_QP_AV && rxe_av_chk_attr(rxe, &attr->ah_attr)) + goto err1; + + if (mask & IB_QP_ALT_PATH) { + if (rxe_av_chk_attr(rxe, &attr->alt_ah_attr)) + goto err1; + if (attr->alt_port_num != 1) { + pr_warn("invalid alt port %d\n", attr->alt_port_num); + goto err1; + } + if (attr->alt_timeout > 31) { + pr_warn("invalid QP alt timeout %d > 31\n", + attr->alt_timeout); + goto err1; + } + } + + if (mask & IB_QP_PATH_MTU) { + struct rxe_port *port = &rxe->port; + + enum ib_mtu max_mtu = port->attr.max_mtu; + enum ib_mtu mtu = attr->path_mtu; + + if (mtu > max_mtu) { + pr_debug("invalid mtu (%d) > (%d)\n", + ib_mtu_enum_to_int(mtu), + ib_mtu_enum_to_int(max_mtu)); + goto err1; + } + } + + if (mask & IB_QP_MAX_QP_RD_ATOMIC) { + if (attr->max_rd_atomic > rxe->attr.max_qp_rd_atom) { + pr_warn("invalid max_rd_atomic %d > %d\n", + attr->max_rd_atomic, + rxe->attr.max_qp_rd_atom); + goto err1; + } + } + + if (mask & IB_QP_TIMEOUT) { + if (attr->timeout > 31) { + pr_warn("invalid QP timeout %d > 31\n", + attr->timeout); + goto err1; + } + } + + return 0; + +err1: + return -EINVAL; +} + +/* move the qp to the reset state */ +static void rxe_qp_reset(struct rxe_qp *qp) +{ + /* stop tasks from running */ + rxe_disable_task(&qp->resp.task); + + /* stop request/comp */ + if (qp->sq.queue) { + if (qp_type(qp) == IB_QPT_RC) + rxe_disable_task(&qp->comp.task); + rxe_disable_task(&qp->req.task); + } + + /* move qp to the reset state */ + qp->req.state = QP_STATE_RESET; + qp->resp.state = QP_STATE_RESET; + + /* let state machines reset themselves drain work and packet queues + * etc. + */ + __rxe_do_task(&qp->resp.task); + + if (qp->sq.queue) { + __rxe_do_task(&qp->comp.task); + __rxe_do_task(&qp->req.task); + } + + /* cleanup attributes */ + atomic_set(&qp->ssn, 0); + qp->req.opcode = -1; + qp->req.need_retry = 0; + qp->req.noack_pkts = 0; + qp->resp.msn = 0; + qp->resp.opcode = -1; + qp->resp.drop_msg = 0; + qp->resp.goto_error = 0; + qp->resp.sent_psn_nak = 0; + + if (qp->resp.mr) { + rxe_drop_ref(qp->resp.mr); + qp->resp.mr = NULL; + } + + cleanup_rd_atomic_resources(qp); + + /* reenable tasks */ + rxe_enable_task(&qp->resp.task); + + if (qp->sq.queue) { + if (qp_type(qp) == IB_QPT_RC) + rxe_enable_task(&qp->comp.task); + + rxe_enable_task(&qp->req.task); + } +} + +/* drain the send queue */ +static void rxe_qp_drain(struct rxe_qp *qp) +{ + if (qp->sq.queue) { + if (qp->req.state != QP_STATE_DRAINED) { + qp->req.state = QP_STATE_DRAIN; + if (qp_type(qp) == IB_QPT_RC) + rxe_run_task(&qp->comp.task, 1); + else + __rxe_do_task(&qp->comp.task); + rxe_run_task(&qp->req.task, 1); + } + } +} + +/* move the qp to the error state */ +void rxe_qp_error(struct rxe_qp *qp) +{ + qp->req.state = QP_STATE_ERROR; + qp->resp.state = QP_STATE_ERROR; + + /* drain work and packet queues */ + rxe_run_task(&qp->resp.task, 1); + + if (qp_type(qp) == IB_QPT_RC) + rxe_run_task(&qp->comp.task, 1); + else + __rxe_do_task(&qp->comp.task); + rxe_run_task(&qp->req.task, 1); +} + +/* called by the modify qp verb */ +int rxe_qp_from_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask, + struct ib_udata *udata) +{ + int err; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + union ib_gid sgid; + struct ib_gid_attr sgid_attr; + + if (mask & IB_QP_MAX_QP_RD_ATOMIC) { + int max_rd_atomic = __roundup_pow_of_two(attr->max_rd_atomic); + + free_rd_atomic_resources(qp); + + err = alloc_rd_atomic_resources(qp, max_rd_atomic); + if (err) + return err; + + qp->attr.max_rd_atomic = max_rd_atomic; + atomic_set(&qp->req.rd_atomic, max_rd_atomic); + } + + if (mask & IB_QP_CUR_STATE) + qp->attr.cur_qp_state = attr->qp_state; + + if (mask & IB_QP_EN_SQD_ASYNC_NOTIFY) + qp->attr.en_sqd_async_notify = attr->en_sqd_async_notify; + + if (mask & IB_QP_ACCESS_FLAGS) + qp->attr.qp_access_flags = attr->qp_access_flags; + + if (mask & IB_QP_PKEY_INDEX) + qp->attr.pkey_index = attr->pkey_index; + + if (mask & IB_QP_PORT) + qp->attr.port_num = attr->port_num; + + if (mask & IB_QP_QKEY) + qp->attr.qkey = attr->qkey; + + if (mask & IB_QP_AV) { + ib_get_cached_gid(&rxe->ib_dev, 1, + attr->ah_attr.grh.sgid_index, &sgid, + &sgid_attr); + rxe_av_from_attr(rxe, attr->port_num, &qp->pri_av, + &attr->ah_attr); + rxe_av_fill_ip_info(rxe, &qp->pri_av, &attr->ah_attr, + &sgid_attr, &sgid); + if (sgid_attr.ndev) + dev_put(sgid_attr.ndev); + } + + if (mask & IB_QP_ALT_PATH) { + ib_get_cached_gid(&rxe->ib_dev, 1, + attr->alt_ah_attr.grh.sgid_index, &sgid, + &sgid_attr); + + rxe_av_from_attr(rxe, attr->alt_port_num, &qp->alt_av, + &attr->alt_ah_attr); + rxe_av_fill_ip_info(rxe, &qp->alt_av, &attr->alt_ah_attr, + &sgid_attr, &sgid); + if (sgid_attr.ndev) + dev_put(sgid_attr.ndev); + + qp->attr.alt_port_num = attr->alt_port_num; + qp->attr.alt_pkey_index = attr->alt_pkey_index; + qp->attr.alt_timeout = attr->alt_timeout; + } + + if (mask & IB_QP_PATH_MTU) { + qp->attr.path_mtu = attr->path_mtu; + qp->mtu = ib_mtu_enum_to_int(attr->path_mtu); + } + + if (mask & IB_QP_TIMEOUT) { + qp->attr.timeout = attr->timeout; + if (attr->timeout == 0) { + qp->qp_timeout_jiffies = 0; + } else { + /* According to the spec, timeout = 4.096 * 2 ^ attr->timeout [us] */ + int j = nsecs_to_jiffies(4096ULL << attr->timeout); + + qp->qp_timeout_jiffies = j ? j : 1; + } + } + + if (mask & IB_QP_RETRY_CNT) { + qp->attr.retry_cnt = attr->retry_cnt; + qp->comp.retry_cnt = attr->retry_cnt; + pr_debug("set retry count = %d\n", attr->retry_cnt); + } + + if (mask & IB_QP_RNR_RETRY) { + qp->attr.rnr_retry = attr->rnr_retry; + qp->comp.rnr_retry = attr->rnr_retry; + pr_debug("set rnr retry count = %d\n", attr->rnr_retry); + } + + if (mask & IB_QP_RQ_PSN) { + qp->attr.rq_psn = (attr->rq_psn & BTH_PSN_MASK); + qp->resp.psn = qp->attr.rq_psn; + pr_debug("set resp psn = 0x%x\n", qp->resp.psn); + } + + if (mask & IB_QP_MIN_RNR_TIMER) { + qp->attr.min_rnr_timer = attr->min_rnr_timer; + pr_debug("set min rnr timer = 0x%x\n", + attr->min_rnr_timer); + } + + if (mask & IB_QP_SQ_PSN) { + qp->attr.sq_psn = (attr->sq_psn & BTH_PSN_MASK); + qp->req.psn = qp->attr.sq_psn; + qp->comp.psn = qp->attr.sq_psn; + pr_debug("set req psn = 0x%x\n", qp->req.psn); + } + + if (mask & IB_QP_MAX_DEST_RD_ATOMIC) { + qp->attr.max_dest_rd_atomic = + __roundup_pow_of_two(attr->max_dest_rd_atomic); + } + + if (mask & IB_QP_PATH_MIG_STATE) + qp->attr.path_mig_state = attr->path_mig_state; + + if (mask & IB_QP_DEST_QPN) + qp->attr.dest_qp_num = attr->dest_qp_num; + + if (mask & IB_QP_STATE) { + qp->attr.qp_state = attr->qp_state; + + switch (attr->qp_state) { + case IB_QPS_RESET: + pr_debug("qp state -> RESET\n"); + rxe_qp_reset(qp); + break; + + case IB_QPS_INIT: + pr_debug("qp state -> INIT\n"); + qp->req.state = QP_STATE_INIT; + qp->resp.state = QP_STATE_INIT; + break; + + case IB_QPS_RTR: + pr_debug("qp state -> RTR\n"); + qp->resp.state = QP_STATE_READY; + break; + + case IB_QPS_RTS: + pr_debug("qp state -> RTS\n"); + qp->req.state = QP_STATE_READY; + break; + + case IB_QPS_SQD: + pr_debug("qp state -> SQD\n"); + rxe_qp_drain(qp); + break; + + case IB_QPS_SQE: + pr_warn("qp state -> SQE !!?\n"); + /* Not possible from modify_qp. */ + break; + + case IB_QPS_ERR: + pr_debug("qp state -> ERR\n"); + rxe_qp_error(qp); + break; + } + } + + return 0; +} + +/* called by the query qp verb */ +int rxe_qp_to_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + + *attr = qp->attr; + + attr->rq_psn = qp->resp.psn; + attr->sq_psn = qp->req.psn; + + attr->cap.max_send_wr = qp->sq.max_wr; + attr->cap.max_send_sge = qp->sq.max_sge; + attr->cap.max_inline_data = qp->sq.max_inline; + + if (!qp->srq) { + attr->cap.max_recv_wr = qp->rq.max_wr; + attr->cap.max_recv_sge = qp->rq.max_sge; + } + + rxe_av_to_attr(rxe, &qp->pri_av, &attr->ah_attr); + rxe_av_to_attr(rxe, &qp->alt_av, &attr->alt_ah_attr); + + if (qp->req.state == QP_STATE_DRAIN) { + attr->sq_draining = 1; + /* applications that get this state + * typically spin on it. yield the + * processor + */ + cond_resched(); + } else { + attr->sq_draining = 0; + } + + pr_debug("attr->sq_draining = %d\n", attr->sq_draining); + + return 0; +} + +/* called by the destroy qp verb */ +void rxe_qp_destroy(struct rxe_qp *qp) +{ + qp->valid = 0; + qp->qp_timeout_jiffies = 0; + rxe_cleanup_task(&qp->resp.task); + + del_timer_sync(&qp->retrans_timer); + del_timer_sync(&qp->rnr_nak_timer); + + rxe_cleanup_task(&qp->req.task); + if (qp_type(qp) == IB_QPT_RC) + rxe_cleanup_task(&qp->comp.task); + + /* flush out any receive wr's or pending requests */ + __rxe_do_task(&qp->req.task); + if (qp->sq.queue) { + __rxe_do_task(&qp->comp.task); + __rxe_do_task(&qp->req.task); + } +} + +/* called when the last reference to the qp is dropped */ +void rxe_qp_cleanup(void *arg) +{ + struct rxe_qp *qp = arg; + + rxe_drop_all_mcast_groups(qp); + + if (qp->sq.queue) + rxe_queue_cleanup(qp->sq.queue); + + if (qp->srq) + rxe_drop_ref(qp->srq); + + if (qp->rq.queue) + rxe_queue_cleanup(qp->rq.queue); + + if (qp->scq) + rxe_drop_ref(qp->scq); + if (qp->rcq) + rxe_drop_ref(qp->rcq); + if (qp->pd) + rxe_drop_ref(qp->pd); + + if (qp->resp.mr) { + rxe_drop_ref(qp->resp.mr); + qp->resp.mr = NULL; + } + + free_rd_atomic_resources(qp); + + kernel_sock_shutdown(qp->sk, SHUT_RDWR); +} diff --git a/drivers/infiniband/sw/rxe/rxe_queue.c b/drivers/infiniband/sw/rxe/rxe_queue.c new file mode 100644 index 000000000000..08274254eb88 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_queue.c @@ -0,0 +1,217 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must retailuce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +int do_mmap_info(struct rxe_dev *rxe, + struct ib_udata *udata, + bool is_req, + struct ib_ucontext *context, + struct rxe_queue_buf *buf, + size_t buf_size, + struct rxe_mmap_info **ip_p) +{ + int err; + u32 len, offset; + struct rxe_mmap_info *ip = NULL; + + if (udata) { + if (is_req) { + len = udata->outlen - sizeof(struct mminfo); + offset = sizeof(struct mminfo); + } else { + len = udata->outlen; + offset = 0; + } + + if (len < sizeof(ip->info)) + goto err1; + + ip = rxe_create_mmap_info(rxe, buf_size, context, buf); + if (!ip) + goto err1; + + err = copy_to_user(udata->outbuf + offset, &ip->info, + sizeof(ip->info)); + if (err) + goto err2; + + spin_lock_bh(&rxe->pending_lock); + list_add(&ip->pending_mmaps, &rxe->pending_mmaps); + spin_unlock_bh(&rxe->pending_lock); + } + + *ip_p = ip; + + return 0; + +err2: + kfree(ip); +err1: + return -EINVAL; +} + +struct rxe_queue *rxe_queue_init(struct rxe_dev *rxe, + int *num_elem, + unsigned int elem_size) +{ + struct rxe_queue *q; + size_t buf_size; + unsigned int num_slots; + + /* num_elem == 0 is allowed, but uninteresting */ + if (*num_elem < 0) + goto err1; + + q = kmalloc(sizeof(*q), GFP_KERNEL); + if (!q) + goto err1; + + q->rxe = rxe; + + /* used in resize, only need to copy used part of queue */ + q->elem_size = elem_size; + + /* pad element up to at least a cacheline and always a power of 2 */ + if (elem_size < cache_line_size()) + elem_size = cache_line_size(); + elem_size = roundup_pow_of_two(elem_size); + + q->log2_elem_size = order_base_2(elem_size); + + num_slots = *num_elem + 1; + num_slots = roundup_pow_of_two(num_slots); + q->index_mask = num_slots - 1; + + buf_size = sizeof(struct rxe_queue_buf) + num_slots * elem_size; + + q->buf = vmalloc_user(buf_size); + if (!q->buf) + goto err2; + + q->buf->log2_elem_size = q->log2_elem_size; + q->buf->index_mask = q->index_mask; + + q->buf_size = buf_size; + + *num_elem = num_slots - 1; + return q; + +err2: + kfree(q); +err1: + return NULL; +} + +/* copies elements from original q to new q and then swaps the contents of the + * two q headers. This is so that if anyone is holding a pointer to q it will + * still work + */ +static int resize_finish(struct rxe_queue *q, struct rxe_queue *new_q, + unsigned int num_elem) +{ + if (!queue_empty(q) && (num_elem < queue_count(q))) + return -EINVAL; + + while (!queue_empty(q)) { + memcpy(producer_addr(new_q), consumer_addr(q), + new_q->elem_size); + advance_producer(new_q); + advance_consumer(q); + } + + swap(*q, *new_q); + + return 0; +} + +int rxe_queue_resize(struct rxe_queue *q, + unsigned int *num_elem_p, + unsigned int elem_size, + struct ib_ucontext *context, + struct ib_udata *udata, + spinlock_t *producer_lock, + spinlock_t *consumer_lock) +{ + struct rxe_queue *new_q; + unsigned int num_elem = *num_elem_p; + int err; + unsigned long flags = 0, flags1; + + new_q = rxe_queue_init(q->rxe, &num_elem, elem_size); + if (!new_q) + return -ENOMEM; + + err = do_mmap_info(new_q->rxe, udata, false, context, new_q->buf, + new_q->buf_size, &new_q->ip); + if (err) { + vfree(new_q->buf); + kfree(new_q); + goto err1; + } + + spin_lock_irqsave(consumer_lock, flags1); + + if (producer_lock) { + spin_lock_irqsave(producer_lock, flags); + err = resize_finish(q, new_q, num_elem); + spin_unlock_irqrestore(producer_lock, flags); + } else { + err = resize_finish(q, new_q, num_elem); + } + + spin_unlock_irqrestore(consumer_lock, flags1); + + rxe_queue_cleanup(new_q); /* new/old dep on err */ + if (err) + goto err1; + + *num_elem_p = num_elem; + return 0; + +err1: + return err; +} + +void rxe_queue_cleanup(struct rxe_queue *q) +{ + if (q->ip) + kref_put(&q->ip->ref, rxe_mmap_release); + else + vfree(q->buf); + + kfree(q); +} diff --git a/drivers/infiniband/sw/rxe/rxe_queue.h b/drivers/infiniband/sw/rxe/rxe_queue.h new file mode 100644 index 000000000000..239fd609c31e --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_queue.h @@ -0,0 +1,178 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_QUEUE_H +#define RXE_QUEUE_H + +/* implements a simple circular buffer that can optionally be + * shared between user space and the kernel and can be resized + + * the requested element size is rounded up to a power of 2 + * and the number of elements in the buffer is also rounded + * up to a power of 2. Since the queue is empty when the + * producer and consumer indices match the maximum capacity + * of the queue is one less than the number of element slots + */ + +/* this data structure is shared between user space and kernel + * space for those cases where the queue is shared. It contains + * the producer and consumer indices. Is also contains a copy + * of the queue size parameters for user space to use but the + * kernel must use the parameters in the rxe_queue struct + * this MUST MATCH the corresponding librxe struct + * for performance reasons arrange to have producer and consumer + * pointers in separate cache lines + * the kernel should always mask the indices to avoid accessing + * memory outside of the data area + */ +struct rxe_queue_buf { + __u32 log2_elem_size; + __u32 index_mask; + __u32 pad_1[30]; + __u32 producer_index; + __u32 pad_2[31]; + __u32 consumer_index; + __u32 pad_3[31]; + __u8 data[0]; +}; + +struct rxe_queue { + struct rxe_dev *rxe; + struct rxe_queue_buf *buf; + struct rxe_mmap_info *ip; + size_t buf_size; + size_t elem_size; + unsigned int log2_elem_size; + unsigned int index_mask; +}; + +int do_mmap_info(struct rxe_dev *rxe, + struct ib_udata *udata, + bool is_req, + struct ib_ucontext *context, + struct rxe_queue_buf *buf, + size_t buf_size, + struct rxe_mmap_info **ip_p); + +struct rxe_queue *rxe_queue_init(struct rxe_dev *rxe, + int *num_elem, + unsigned int elem_size); + +int rxe_queue_resize(struct rxe_queue *q, + unsigned int *num_elem_p, + unsigned int elem_size, + struct ib_ucontext *context, + struct ib_udata *udata, + /* Protect producers while resizing queue */ + spinlock_t *producer_lock, + /* Protect consumers while resizing queue */ + spinlock_t *consumer_lock); + +void rxe_queue_cleanup(struct rxe_queue *queue); + +static inline int next_index(struct rxe_queue *q, int index) +{ + return (index + 1) & q->buf->index_mask; +} + +static inline int queue_empty(struct rxe_queue *q) +{ + return ((q->buf->producer_index - q->buf->consumer_index) + & q->index_mask) == 0; +} + +static inline int queue_full(struct rxe_queue *q) +{ + return ((q->buf->producer_index + 1 - q->buf->consumer_index) + & q->index_mask) == 0; +} + +static inline void advance_producer(struct rxe_queue *q) +{ + q->buf->producer_index = (q->buf->producer_index + 1) + & q->index_mask; +} + +static inline void advance_consumer(struct rxe_queue *q) +{ + q->buf->consumer_index = (q->buf->consumer_index + 1) + & q->index_mask; +} + +static inline void *producer_addr(struct rxe_queue *q) +{ + return q->buf->data + ((q->buf->producer_index & q->index_mask) + << q->log2_elem_size); +} + +static inline void *consumer_addr(struct rxe_queue *q) +{ + return q->buf->data + ((q->buf->consumer_index & q->index_mask) + << q->log2_elem_size); +} + +static inline unsigned int producer_index(struct rxe_queue *q) +{ + return q->buf->producer_index; +} + +static inline unsigned int consumer_index(struct rxe_queue *q) +{ + return q->buf->consumer_index; +} + +static inline void *addr_from_index(struct rxe_queue *q, unsigned int index) +{ + return q->buf->data + ((index & q->index_mask) + << q->buf->log2_elem_size); +} + +static inline unsigned int index_from_addr(const struct rxe_queue *q, + const void *addr) +{ + return (((u8 *)addr - q->buf->data) >> q->log2_elem_size) + & q->index_mask; +} + +static inline unsigned int queue_count(const struct rxe_queue *q) +{ + return (q->buf->producer_index - q->buf->consumer_index) + & q->index_mask; +} + +static inline void *queue_head(struct rxe_queue *q) +{ + return queue_empty(q) ? NULL : consumer_addr(q); +} + +#endif /* RXE_QUEUE_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_recv.c b/drivers/infiniband/sw/rxe/rxe_recv.c new file mode 100644 index 000000000000..3d464c23e08b --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_recv.c @@ -0,0 +1,420 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include + +#include "rxe.h" +#include "rxe_loc.h" + +static int check_type_state(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct rxe_qp *qp) +{ + if (unlikely(!qp->valid)) + goto err1; + + switch (qp_type(qp)) { + case IB_QPT_RC: + if (unlikely((pkt->opcode & IB_OPCODE_RC) != 0)) { + pr_warn_ratelimited("bad qp type\n"); + goto err1; + } + break; + case IB_QPT_UC: + if (unlikely(!(pkt->opcode & IB_OPCODE_UC))) { + pr_warn_ratelimited("bad qp type\n"); + goto err1; + } + break; + case IB_QPT_UD: + case IB_QPT_SMI: + case IB_QPT_GSI: + if (unlikely(!(pkt->opcode & IB_OPCODE_UD))) { + pr_warn_ratelimited("bad qp type\n"); + goto err1; + } + break; + default: + pr_warn_ratelimited("unsupported qp type\n"); + goto err1; + } + + if (pkt->mask & RXE_REQ_MASK) { + if (unlikely(qp->resp.state != QP_STATE_READY)) + goto err1; + } else if (unlikely(qp->req.state < QP_STATE_READY || + qp->req.state > QP_STATE_DRAINED)) { + goto err1; + } + + return 0; + +err1: + return -EINVAL; +} + +static void set_bad_pkey_cntr(struct rxe_port *port) +{ + spin_lock_bh(&port->port_lock); + port->attr.bad_pkey_cntr = min((u32)0xffff, + port->attr.bad_pkey_cntr + 1); + spin_unlock_bh(&port->port_lock); +} + +static void set_qkey_viol_cntr(struct rxe_port *port) +{ + spin_lock_bh(&port->port_lock); + port->attr.qkey_viol_cntr = min((u32)0xffff, + port->attr.qkey_viol_cntr + 1); + spin_unlock_bh(&port->port_lock); +} + +static int check_keys(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + u32 qpn, struct rxe_qp *qp) +{ + int i; + int found_pkey = 0; + struct rxe_port *port = &rxe->port; + u16 pkey = bth_pkey(pkt); + + pkt->pkey_index = 0; + + if (qpn == 1) { + for (i = 0; i < port->attr.pkey_tbl_len; i++) { + if (pkey_match(pkey, port->pkey_tbl[i])) { + pkt->pkey_index = i; + found_pkey = 1; + break; + } + } + + if (!found_pkey) { + pr_warn_ratelimited("bad pkey = 0x%x\n", pkey); + set_bad_pkey_cntr(port); + goto err1; + } + } else if (qpn != 0) { + if (unlikely(!pkey_match(pkey, + port->pkey_tbl[qp->attr.pkey_index] + ))) { + pr_warn_ratelimited("bad pkey = 0x%0x\n", pkey); + set_bad_pkey_cntr(port); + goto err1; + } + pkt->pkey_index = qp->attr.pkey_index; + } + + if ((qp_type(qp) == IB_QPT_UD || qp_type(qp) == IB_QPT_GSI) && + qpn != 0 && pkt->mask) { + u32 qkey = (qpn == 1) ? GSI_QKEY : qp->attr.qkey; + + if (unlikely(deth_qkey(pkt) != qkey)) { + pr_warn_ratelimited("bad qkey, got 0x%x expected 0x%x for qpn 0x%x\n", + deth_qkey(pkt), qkey, qpn); + set_qkey_viol_cntr(port); + goto err1; + } + } + + return 0; + +err1: + return -EINVAL; +} + +static int check_addr(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct rxe_qp *qp) +{ + struct sk_buff *skb = PKT_TO_SKB(pkt); + + if (qp_type(qp) != IB_QPT_RC && qp_type(qp) != IB_QPT_UC) + goto done; + + if (unlikely(pkt->port_num != qp->attr.port_num)) { + pr_warn_ratelimited("port %d != qp port %d\n", + pkt->port_num, qp->attr.port_num); + goto err1; + } + + if (skb->protocol == htons(ETH_P_IP)) { + struct in_addr *saddr = + &qp->pri_av.sgid_addr._sockaddr_in.sin_addr; + struct in_addr *daddr = + &qp->pri_av.dgid_addr._sockaddr_in.sin_addr; + + if (ip_hdr(skb)->daddr != saddr->s_addr) { + pr_warn_ratelimited("dst addr %pI4 != qp source addr %pI4\n", + &ip_hdr(skb)->daddr, + &saddr->s_addr); + goto err1; + } + + if (ip_hdr(skb)->saddr != daddr->s_addr) { + pr_warn_ratelimited("source addr %pI4 != qp dst addr %pI4\n", + &ip_hdr(skb)->saddr, + &daddr->s_addr); + goto err1; + } + + } else if (skb->protocol == htons(ETH_P_IPV6)) { + struct in6_addr *saddr = + &qp->pri_av.sgid_addr._sockaddr_in6.sin6_addr; + struct in6_addr *daddr = + &qp->pri_av.dgid_addr._sockaddr_in6.sin6_addr; + + if (memcmp(&ipv6_hdr(skb)->daddr, saddr, sizeof(*saddr))) { + pr_warn_ratelimited("dst addr %pI6 != qp source addr %pI6\n", + &ipv6_hdr(skb)->daddr, saddr); + goto err1; + } + + if (memcmp(&ipv6_hdr(skb)->saddr, daddr, sizeof(*daddr))) { + pr_warn_ratelimited("source addr %pI6 != qp dst addr %pI6\n", + &ipv6_hdr(skb)->saddr, daddr); + goto err1; + } + } + +done: + return 0; + +err1: + return -EINVAL; +} + +static int hdr_check(struct rxe_pkt_info *pkt) +{ + struct rxe_dev *rxe = pkt->rxe; + struct rxe_port *port = &rxe->port; + struct rxe_qp *qp = NULL; + u32 qpn = bth_qpn(pkt); + int index; + int err; + + if (unlikely(bth_tver(pkt) != BTH_TVER)) { + pr_warn_ratelimited("bad tver\n"); + goto err1; + } + + if (qpn != IB_MULTICAST_QPN) { + index = (qpn == 0) ? port->qp_smi_index : + ((qpn == 1) ? port->qp_gsi_index : qpn); + qp = rxe_pool_get_index(&rxe->qp_pool, index); + if (unlikely(!qp)) { + pr_warn_ratelimited("no qp matches qpn 0x%x\n", qpn); + goto err1; + } + + err = check_type_state(rxe, pkt, qp); + if (unlikely(err)) + goto err2; + + err = check_addr(rxe, pkt, qp); + if (unlikely(err)) + goto err2; + + err = check_keys(rxe, pkt, qpn, qp); + if (unlikely(err)) + goto err2; + } else { + if (unlikely((pkt->mask & RXE_GRH_MASK) == 0)) { + pr_warn_ratelimited("no grh for mcast qpn\n"); + goto err1; + } + } + + pkt->qp = qp; + return 0; + +err2: + if (qp) + rxe_drop_ref(qp); +err1: + return -EINVAL; +} + +static inline void rxe_rcv_pkt(struct rxe_dev *rxe, + struct rxe_pkt_info *pkt, + struct sk_buff *skb) +{ + if (pkt->mask & RXE_REQ_MASK) + rxe_resp_queue_pkt(rxe, pkt->qp, skb); + else + rxe_comp_queue_pkt(rxe, pkt->qp, skb); +} + +static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb) +{ + struct rxe_pkt_info *pkt = SKB_TO_PKT(skb); + struct rxe_mc_grp *mcg; + struct sk_buff *skb_copy; + struct rxe_mc_elem *mce; + struct rxe_qp *qp; + union ib_gid dgid; + int err; + + if (skb->protocol == htons(ETH_P_IP)) + ipv6_addr_set_v4mapped(ip_hdr(skb)->daddr, + (struct in6_addr *)&dgid); + else if (skb->protocol == htons(ETH_P_IPV6)) + memcpy(&dgid, &ipv6_hdr(skb)->daddr, sizeof(dgid)); + + /* lookup mcast group corresponding to mgid, takes a ref */ + mcg = rxe_pool_get_key(&rxe->mc_grp_pool, &dgid); + if (!mcg) + goto err1; /* mcast group not registered */ + + spin_lock_bh(&mcg->mcg_lock); + + list_for_each_entry(mce, &mcg->qp_list, qp_list) { + qp = mce->qp; + pkt = SKB_TO_PKT(skb); + + /* validate qp for incoming packet */ + err = check_type_state(rxe, pkt, qp); + if (err) + continue; + + err = check_keys(rxe, pkt, bth_qpn(pkt), qp); + if (err) + continue; + + /* if *not* the last qp in the list + * make a copy of the skb to post to the next qp + */ + skb_copy = (mce->qp_list.next != &mcg->qp_list) ? + skb_clone(skb, GFP_KERNEL) : NULL; + + pkt->qp = qp; + rxe_add_ref(qp); + rxe_rcv_pkt(rxe, pkt, skb); + + skb = skb_copy; + if (!skb) + break; + } + + spin_unlock_bh(&mcg->mcg_lock); + + rxe_drop_ref(mcg); /* drop ref from rxe_pool_get_key. */ + +err1: + if (skb) + kfree_skb(skb); +} + +static int rxe_match_dgid(struct rxe_dev *rxe, struct sk_buff *skb) +{ + union ib_gid dgid; + union ib_gid *pdgid; + u16 index; + + if (skb->protocol == htons(ETH_P_IP)) { + ipv6_addr_set_v4mapped(ip_hdr(skb)->daddr, + (struct in6_addr *)&dgid); + pdgid = &dgid; + } else { + pdgid = (union ib_gid *)&ipv6_hdr(skb)->daddr; + } + + return ib_find_cached_gid_by_port(&rxe->ib_dev, pdgid, + IB_GID_TYPE_ROCE_UDP_ENCAP, + 1, rxe->ndev, &index); +} + +/* rxe_rcv is called from the interface driver */ +int rxe_rcv(struct sk_buff *skb) +{ + int err; + struct rxe_pkt_info *pkt = SKB_TO_PKT(skb); + struct rxe_dev *rxe = pkt->rxe; + __be32 *icrcp; + u32 calc_icrc, pack_icrc; + + pkt->offset = 0; + + if (unlikely(skb->len < pkt->offset + RXE_BTH_BYTES)) + goto drop; + + if (unlikely(rxe_match_dgid(rxe, skb) < 0)) { + pr_warn_ratelimited("failed matching dgid\n"); + goto drop; + } + + pkt->opcode = bth_opcode(pkt); + pkt->psn = bth_psn(pkt); + pkt->qp = NULL; + pkt->mask |= rxe_opcode[pkt->opcode].mask; + + if (unlikely(skb->len < header_size(pkt))) + goto drop; + + err = hdr_check(pkt); + if (unlikely(err)) + goto drop; + + /* Verify ICRC */ + icrcp = (__be32 *)(pkt->hdr + pkt->paylen - RXE_ICRC_SIZE); + pack_icrc = be32_to_cpu(*icrcp); + + calc_icrc = rxe_icrc_hdr(pkt, skb); + calc_icrc = crc32_le(calc_icrc, (u8 *)payload_addr(pkt), payload_size(pkt)); + calc_icrc = cpu_to_be32(~calc_icrc); + if (unlikely(calc_icrc != pack_icrc)) { + char saddr[sizeof(struct in6_addr)]; + + if (skb->protocol == htons(ETH_P_IPV6)) + sprintf(saddr, "%pI6", &ipv6_hdr(skb)->saddr); + else if (skb->protocol == htons(ETH_P_IP)) + sprintf(saddr, "%pI4", &ip_hdr(skb)->saddr); + else + sprintf(saddr, "unknown"); + + pr_warn_ratelimited("bad ICRC from %s\n", saddr); + goto drop; + } + + if (unlikely(bth_qpn(pkt) == IB_MULTICAST_QPN)) + rxe_rcv_mcast_pkt(rxe, skb); + else + rxe_rcv_pkt(rxe, pkt, skb); + + return 0; + +drop: + if (pkt->qp) + rxe_drop_ref(pkt->qp); + + kfree_skb(skb); + return 0; +} +EXPORT_SYMBOL(rxe_rcv); diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c new file mode 100644 index 000000000000..33b2d9d77021 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -0,0 +1,726 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +static int next_opcode(struct rxe_qp *qp, struct rxe_send_wqe *wqe, + unsigned opcode); + +static inline void retry_first_write_send(struct rxe_qp *qp, + struct rxe_send_wqe *wqe, + unsigned mask, int npsn) +{ + int i; + + for (i = 0; i < npsn; i++) { + int to_send = (wqe->dma.resid > qp->mtu) ? + qp->mtu : wqe->dma.resid; + + qp->req.opcode = next_opcode(qp, wqe, + wqe->wr.opcode); + + if (wqe->wr.send_flags & IB_SEND_INLINE) { + wqe->dma.resid -= to_send; + wqe->dma.sge_offset += to_send; + } else { + advance_dma_data(&wqe->dma, to_send); + } + if (mask & WR_WRITE_MASK) + wqe->iova += qp->mtu; + } +} + +static void req_retry(struct rxe_qp *qp) +{ + struct rxe_send_wqe *wqe; + unsigned int wqe_index; + unsigned int mask; + int npsn; + int first = 1; + + wqe = queue_head(qp->sq.queue); + npsn = (qp->comp.psn - wqe->first_psn) & BTH_PSN_MASK; + + qp->req.wqe_index = consumer_index(qp->sq.queue); + qp->req.psn = qp->comp.psn; + qp->req.opcode = -1; + + for (wqe_index = consumer_index(qp->sq.queue); + wqe_index != producer_index(qp->sq.queue); + wqe_index = next_index(qp->sq.queue, wqe_index)) { + wqe = addr_from_index(qp->sq.queue, wqe_index); + mask = wr_opcode_mask(wqe->wr.opcode, qp); + + if (wqe->state == wqe_state_posted) + break; + + if (wqe->state == wqe_state_done) + continue; + + wqe->iova = (mask & WR_ATOMIC_MASK) ? + wqe->wr.wr.atomic.remote_addr : + (mask & WR_READ_OR_WRITE_MASK) ? + wqe->wr.wr.rdma.remote_addr : + 0; + + if (!first || (mask & WR_READ_MASK) == 0) { + wqe->dma.resid = wqe->dma.length; + wqe->dma.cur_sge = 0; + wqe->dma.sge_offset = 0; + } + + if (first) { + first = 0; + + if (mask & WR_WRITE_OR_SEND_MASK) + retry_first_write_send(qp, wqe, mask, npsn); + + if (mask & WR_READ_MASK) + wqe->iova += npsn * qp->mtu; + } + + wqe->state = wqe_state_posted; + } +} + +void rnr_nak_timer(unsigned long data) +{ + struct rxe_qp *qp = (struct rxe_qp *)data; + + pr_debug("rnr nak timer fired\n"); + rxe_run_task(&qp->req.task, 1); +} + +static struct rxe_send_wqe *req_next_wqe(struct rxe_qp *qp) +{ + struct rxe_send_wqe *wqe = queue_head(qp->sq.queue); + unsigned long flags; + + if (unlikely(qp->req.state == QP_STATE_DRAIN)) { + /* check to see if we are drained; + * state_lock used by requester and completer + */ + spin_lock_irqsave(&qp->state_lock, flags); + do { + if (qp->req.state != QP_STATE_DRAIN) { + /* comp just finished */ + spin_unlock_irqrestore(&qp->state_lock, + flags); + break; + } + + if (wqe && ((qp->req.wqe_index != + consumer_index(qp->sq.queue)) || + (wqe->state != wqe_state_posted))) { + /* comp not done yet */ + spin_unlock_irqrestore(&qp->state_lock, + flags); + break; + } + + qp->req.state = QP_STATE_DRAINED; + spin_unlock_irqrestore(&qp->state_lock, flags); + + if (qp->ibqp.event_handler) { + struct ib_event ev; + + ev.device = qp->ibqp.device; + ev.element.qp = &qp->ibqp; + ev.event = IB_EVENT_SQ_DRAINED; + qp->ibqp.event_handler(&ev, + qp->ibqp.qp_context); + } + } while (0); + } + + if (qp->req.wqe_index == producer_index(qp->sq.queue)) + return NULL; + + wqe = addr_from_index(qp->sq.queue, qp->req.wqe_index); + + if (unlikely((qp->req.state == QP_STATE_DRAIN || + qp->req.state == QP_STATE_DRAINED) && + (wqe->state != wqe_state_processing))) + return NULL; + + if (unlikely((wqe->wr.send_flags & IB_SEND_FENCE) && + (qp->req.wqe_index != consumer_index(qp->sq.queue)))) { + qp->req.wait_fence = 1; + return NULL; + } + + wqe->mask = wr_opcode_mask(wqe->wr.opcode, qp); + return wqe; +} + +static int next_opcode_rc(struct rxe_qp *qp, unsigned opcode, int fits) +{ + switch (opcode) { + case IB_WR_RDMA_WRITE: + if (qp->req.opcode == IB_OPCODE_RC_RDMA_WRITE_FIRST || + qp->req.opcode == IB_OPCODE_RC_RDMA_WRITE_MIDDLE) + return fits ? + IB_OPCODE_RC_RDMA_WRITE_LAST : + IB_OPCODE_RC_RDMA_WRITE_MIDDLE; + else + return fits ? + IB_OPCODE_RC_RDMA_WRITE_ONLY : + IB_OPCODE_RC_RDMA_WRITE_FIRST; + + case IB_WR_RDMA_WRITE_WITH_IMM: + if (qp->req.opcode == IB_OPCODE_RC_RDMA_WRITE_FIRST || + qp->req.opcode == IB_OPCODE_RC_RDMA_WRITE_MIDDLE) + return fits ? + IB_OPCODE_RC_RDMA_WRITE_LAST_WITH_IMMEDIATE : + IB_OPCODE_RC_RDMA_WRITE_MIDDLE; + else + return fits ? + IB_OPCODE_RC_RDMA_WRITE_ONLY_WITH_IMMEDIATE : + IB_OPCODE_RC_RDMA_WRITE_FIRST; + + case IB_WR_SEND: + if (qp->req.opcode == IB_OPCODE_RC_SEND_FIRST || + qp->req.opcode == IB_OPCODE_RC_SEND_MIDDLE) + return fits ? + IB_OPCODE_RC_SEND_LAST : + IB_OPCODE_RC_SEND_MIDDLE; + else + return fits ? + IB_OPCODE_RC_SEND_ONLY : + IB_OPCODE_RC_SEND_FIRST; + + case IB_WR_SEND_WITH_IMM: + if (qp->req.opcode == IB_OPCODE_RC_SEND_FIRST || + qp->req.opcode == IB_OPCODE_RC_SEND_MIDDLE) + return fits ? + IB_OPCODE_RC_SEND_LAST_WITH_IMMEDIATE : + IB_OPCODE_RC_SEND_MIDDLE; + else + return fits ? + IB_OPCODE_RC_SEND_ONLY_WITH_IMMEDIATE : + IB_OPCODE_RC_SEND_FIRST; + + case IB_WR_RDMA_READ: + return IB_OPCODE_RC_RDMA_READ_REQUEST; + + case IB_WR_ATOMIC_CMP_AND_SWP: + return IB_OPCODE_RC_COMPARE_SWAP; + + case IB_WR_ATOMIC_FETCH_AND_ADD: + return IB_OPCODE_RC_FETCH_ADD; + + case IB_WR_SEND_WITH_INV: + if (qp->req.opcode == IB_OPCODE_RC_SEND_FIRST || + qp->req.opcode == IB_OPCODE_RC_SEND_MIDDLE) + return fits ? IB_OPCODE_RC_SEND_LAST_WITH_INVALIDATE : + IB_OPCODE_RC_SEND_MIDDLE; + else + return fits ? IB_OPCODE_RC_SEND_ONLY_WITH_INVALIDATE : + IB_OPCODE_RC_SEND_FIRST; + case IB_WR_REG_MR: + case IB_WR_LOCAL_INV: + return opcode; + } + + return -EINVAL; +} + +static int next_opcode_uc(struct rxe_qp *qp, unsigned opcode, int fits) +{ + switch (opcode) { + case IB_WR_RDMA_WRITE: + if (qp->req.opcode == IB_OPCODE_UC_RDMA_WRITE_FIRST || + qp->req.opcode == IB_OPCODE_UC_RDMA_WRITE_MIDDLE) + return fits ? + IB_OPCODE_UC_RDMA_WRITE_LAST : + IB_OPCODE_UC_RDMA_WRITE_MIDDLE; + else + return fits ? + IB_OPCODE_UC_RDMA_WRITE_ONLY : + IB_OPCODE_UC_RDMA_WRITE_FIRST; + + case IB_WR_RDMA_WRITE_WITH_IMM: + if (qp->req.opcode == IB_OPCODE_UC_RDMA_WRITE_FIRST || + qp->req.opcode == IB_OPCODE_UC_RDMA_WRITE_MIDDLE) + return fits ? + IB_OPCODE_UC_RDMA_WRITE_LAST_WITH_IMMEDIATE : + IB_OPCODE_UC_RDMA_WRITE_MIDDLE; + else + return fits ? + IB_OPCODE_UC_RDMA_WRITE_ONLY_WITH_IMMEDIATE : + IB_OPCODE_UC_RDMA_WRITE_FIRST; + + case IB_WR_SEND: + if (qp->req.opcode == IB_OPCODE_UC_SEND_FIRST || + qp->req.opcode == IB_OPCODE_UC_SEND_MIDDLE) + return fits ? + IB_OPCODE_UC_SEND_LAST : + IB_OPCODE_UC_SEND_MIDDLE; + else + return fits ? + IB_OPCODE_UC_SEND_ONLY : + IB_OPCODE_UC_SEND_FIRST; + + case IB_WR_SEND_WITH_IMM: + if (qp->req.opcode == IB_OPCODE_UC_SEND_FIRST || + qp->req.opcode == IB_OPCODE_UC_SEND_MIDDLE) + return fits ? + IB_OPCODE_UC_SEND_LAST_WITH_IMMEDIATE : + IB_OPCODE_UC_SEND_MIDDLE; + else + return fits ? + IB_OPCODE_UC_SEND_ONLY_WITH_IMMEDIATE : + IB_OPCODE_UC_SEND_FIRST; + } + + return -EINVAL; +} + +static int next_opcode(struct rxe_qp *qp, struct rxe_send_wqe *wqe, + unsigned opcode) +{ + int fits = (wqe->dma.resid <= qp->mtu); + + switch (qp_type(qp)) { + case IB_QPT_RC: + return next_opcode_rc(qp, opcode, fits); + + case IB_QPT_UC: + return next_opcode_uc(qp, opcode, fits); + + case IB_QPT_SMI: + case IB_QPT_UD: + case IB_QPT_GSI: + switch (opcode) { + case IB_WR_SEND: + return IB_OPCODE_UD_SEND_ONLY; + + case IB_WR_SEND_WITH_IMM: + return IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE; + } + break; + + default: + break; + } + + return -EINVAL; +} + +static inline int check_init_depth(struct rxe_qp *qp, struct rxe_send_wqe *wqe) +{ + int depth; + + if (wqe->has_rd_atomic) + return 0; + + qp->req.need_rd_atomic = 1; + depth = atomic_dec_return(&qp->req.rd_atomic); + + if (depth >= 0) { + qp->req.need_rd_atomic = 0; + wqe->has_rd_atomic = 1; + return 0; + } + + atomic_inc(&qp->req.rd_atomic); + return -EAGAIN; +} + +static inline int get_mtu(struct rxe_qp *qp, struct rxe_send_wqe *wqe) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct rxe_port *port; + struct rxe_av *av; + + if ((qp_type(qp) == IB_QPT_RC) || (qp_type(qp) == IB_QPT_UC)) + return qp->mtu; + + av = &wqe->av; + port = &rxe->port; + + return port->mtu_cap; +} + +static struct sk_buff *init_req_packet(struct rxe_qp *qp, + struct rxe_send_wqe *wqe, + int opcode, int payload, + struct rxe_pkt_info *pkt) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct rxe_port *port = &rxe->port; + struct sk_buff *skb; + struct rxe_send_wr *ibwr = &wqe->wr; + struct rxe_av *av; + int pad = (-payload) & 0x3; + int paylen; + int solicited; + u16 pkey; + u32 qp_num; + int ack_req; + + /* length from start of bth to end of icrc */ + paylen = rxe_opcode[opcode].length + payload + pad + RXE_ICRC_SIZE; + + /* pkt->hdr, rxe, port_num and mask are initialized in ifc + * layer + */ + pkt->opcode = opcode; + pkt->qp = qp; + pkt->psn = qp->req.psn; + pkt->mask = rxe_opcode[opcode].mask; + pkt->paylen = paylen; + pkt->offset = 0; + pkt->wqe = wqe; + + /* init skb */ + av = rxe_get_av(pkt); + skb = rxe->ifc_ops->init_packet(rxe, av, paylen, pkt); + if (unlikely(!skb)) + return NULL; + + /* init bth */ + solicited = (ibwr->send_flags & IB_SEND_SOLICITED) && + (pkt->mask & RXE_END_MASK) && + ((pkt->mask & (RXE_SEND_MASK)) || + (pkt->mask & (RXE_WRITE_MASK | RXE_IMMDT_MASK)) == + (RXE_WRITE_MASK | RXE_IMMDT_MASK)); + + pkey = (qp_type(qp) == IB_QPT_GSI) ? + port->pkey_tbl[ibwr->wr.ud.pkey_index] : + port->pkey_tbl[qp->attr.pkey_index]; + + qp_num = (pkt->mask & RXE_DETH_MASK) ? ibwr->wr.ud.remote_qpn : + qp->attr.dest_qp_num; + + ack_req = ((pkt->mask & RXE_END_MASK) || + (qp->req.noack_pkts++ > RXE_MAX_PKT_PER_ACK)); + if (ack_req) + qp->req.noack_pkts = 0; + + bth_init(pkt, pkt->opcode, solicited, 0, pad, pkey, qp_num, + ack_req, pkt->psn); + + /* init optional headers */ + if (pkt->mask & RXE_RETH_MASK) { + reth_set_rkey(pkt, ibwr->wr.rdma.rkey); + reth_set_va(pkt, wqe->iova); + reth_set_len(pkt, wqe->dma.length); + } + + if (pkt->mask & RXE_IMMDT_MASK) + immdt_set_imm(pkt, ibwr->ex.imm_data); + + if (pkt->mask & RXE_IETH_MASK) + ieth_set_rkey(pkt, ibwr->ex.invalidate_rkey); + + if (pkt->mask & RXE_ATMETH_MASK) { + atmeth_set_va(pkt, wqe->iova); + if (opcode == IB_OPCODE_RC_COMPARE_SWAP || + opcode == IB_OPCODE_RD_COMPARE_SWAP) { + atmeth_set_swap_add(pkt, ibwr->wr.atomic.swap); + atmeth_set_comp(pkt, ibwr->wr.atomic.compare_add); + } else { + atmeth_set_swap_add(pkt, ibwr->wr.atomic.compare_add); + } + atmeth_set_rkey(pkt, ibwr->wr.atomic.rkey); + } + + if (pkt->mask & RXE_DETH_MASK) { + if (qp->ibqp.qp_num == 1) + deth_set_qkey(pkt, GSI_QKEY); + else + deth_set_qkey(pkt, ibwr->wr.ud.remote_qkey); + deth_set_sqp(pkt, qp->ibqp.qp_num); + } + + return skb; +} + +static int fill_packet(struct rxe_qp *qp, struct rxe_send_wqe *wqe, + struct rxe_pkt_info *pkt, struct sk_buff *skb, + int paylen) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + u32 crc = 0; + u32 *p; + int err; + + err = rxe->ifc_ops->prepare(rxe, pkt, skb, &crc); + if (err) + return err; + + if (pkt->mask & RXE_WRITE_OR_SEND) { + if (wqe->wr.send_flags & IB_SEND_INLINE) { + u8 *tmp = &wqe->dma.inline_data[wqe->dma.sge_offset]; + + crc = crc32_le(crc, tmp, paylen); + + memcpy(payload_addr(pkt), tmp, paylen); + + wqe->dma.resid -= paylen; + wqe->dma.sge_offset += paylen; + } else { + err = copy_data(rxe, qp->pd, 0, &wqe->dma, + payload_addr(pkt), paylen, + from_mem_obj, + &crc); + if (err) + return err; + } + } + p = payload_addr(pkt) + paylen + bth_pad(pkt); + + *p = ~crc; + + return 0; +} + +static void update_wqe_state(struct rxe_qp *qp, + struct rxe_send_wqe *wqe, + struct rxe_pkt_info *pkt, + enum wqe_state *prev_state) +{ + enum wqe_state prev_state_ = wqe->state; + + if (pkt->mask & RXE_END_MASK) { + if (qp_type(qp) == IB_QPT_RC) + wqe->state = wqe_state_pending; + } else { + wqe->state = wqe_state_processing; + } + + *prev_state = prev_state_; +} + +static void update_state(struct rxe_qp *qp, struct rxe_send_wqe *wqe, + struct rxe_pkt_info *pkt, int payload) +{ + /* number of packets left to send including current one */ + int num_pkt = (wqe->dma.resid + payload + qp->mtu - 1) / qp->mtu; + + /* handle zero length packet case */ + if (num_pkt == 0) + num_pkt = 1; + + if (pkt->mask & RXE_START_MASK) { + wqe->first_psn = qp->req.psn; + wqe->last_psn = (qp->req.psn + num_pkt - 1) & BTH_PSN_MASK; + } + + if (pkt->mask & RXE_READ_MASK) + qp->req.psn = (wqe->first_psn + num_pkt) & BTH_PSN_MASK; + else + qp->req.psn = (qp->req.psn + 1) & BTH_PSN_MASK; + + qp->req.opcode = pkt->opcode; + + + if (pkt->mask & RXE_END_MASK) + qp->req.wqe_index = next_index(qp->sq.queue, qp->req.wqe_index); + + qp->need_req_skb = 0; + + if (qp->qp_timeout_jiffies && !timer_pending(&qp->retrans_timer)) + mod_timer(&qp->retrans_timer, + jiffies + qp->qp_timeout_jiffies); +} + +int rxe_requester(void *arg) +{ + struct rxe_qp *qp = (struct rxe_qp *)arg; + struct rxe_pkt_info pkt; + struct sk_buff *skb; + struct rxe_send_wqe *wqe; + unsigned mask; + int payload; + int mtu; + int opcode; + int ret; + enum wqe_state prev_state; + +next_wqe: + if (unlikely(!qp->valid || qp->req.state == QP_STATE_ERROR)) + goto exit; + + if (unlikely(qp->req.state == QP_STATE_RESET)) { + qp->req.wqe_index = consumer_index(qp->sq.queue); + qp->req.opcode = -1; + qp->req.need_rd_atomic = 0; + qp->req.wait_psn = 0; + qp->req.need_retry = 0; + goto exit; + } + + if (unlikely(qp->req.need_retry)) { + req_retry(qp); + qp->req.need_retry = 0; + } + + wqe = req_next_wqe(qp); + if (unlikely(!wqe)) + goto exit; + + if (wqe->mask & WR_REG_MASK) { + if (wqe->wr.opcode == IB_WR_LOCAL_INV) { + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct rxe_mem *rmr; + + rmr = rxe_pool_get_index(&rxe->mr_pool, + wqe->wr.ex.invalidate_rkey >> 8); + if (!rmr) { + pr_err("No mr for key %#x\n", wqe->wr.ex.invalidate_rkey); + wqe->state = wqe_state_error; + wqe->status = IB_WC_MW_BIND_ERR; + goto exit; + } + rmr->state = RXE_MEM_STATE_FREE; + wqe->state = wqe_state_done; + wqe->status = IB_WC_SUCCESS; + } else if (wqe->wr.opcode == IB_WR_REG_MR) { + struct rxe_mem *rmr = to_rmr(wqe->wr.wr.reg.mr); + + rmr->state = RXE_MEM_STATE_VALID; + rmr->access = wqe->wr.wr.reg.access; + rmr->lkey = wqe->wr.wr.reg.key; + rmr->rkey = wqe->wr.wr.reg.key; + wqe->state = wqe_state_done; + wqe->status = IB_WC_SUCCESS; + } else { + goto exit; + } + qp->req.wqe_index = next_index(qp->sq.queue, + qp->req.wqe_index); + goto next_wqe; + } + + if (unlikely(qp_type(qp) == IB_QPT_RC && + qp->req.psn > (qp->comp.psn + RXE_MAX_UNACKED_PSNS))) { + qp->req.wait_psn = 1; + goto exit; + } + + /* Limit the number of inflight SKBs per QP */ + if (unlikely(atomic_read(&qp->skb_out) > + RXE_INFLIGHT_SKBS_PER_QP_HIGH)) { + qp->need_req_skb = 1; + goto exit; + } + + opcode = next_opcode(qp, wqe, wqe->wr.opcode); + if (unlikely(opcode < 0)) { + wqe->status = IB_WC_LOC_QP_OP_ERR; + goto exit; + } + + mask = rxe_opcode[opcode].mask; + if (unlikely(mask & RXE_READ_OR_ATOMIC)) { + if (check_init_depth(qp, wqe)) + goto exit; + } + + mtu = get_mtu(qp, wqe); + payload = (mask & RXE_WRITE_OR_SEND) ? wqe->dma.resid : 0; + if (payload > mtu) { + if (qp_type(qp) == IB_QPT_UD) { + /* C10-93.1.1: If the total sum of all the buffer lengths specified for a + * UD message exceeds the MTU of the port as returned by QueryHCA, the CI + * shall not emit any packets for this message. Further, the CI shall not + * generate an error due to this condition. + */ + + /* fake a successful UD send */ + wqe->first_psn = qp->req.psn; + wqe->last_psn = qp->req.psn; + qp->req.psn = (qp->req.psn + 1) & BTH_PSN_MASK; + qp->req.opcode = IB_OPCODE_UD_SEND_ONLY; + qp->req.wqe_index = next_index(qp->sq.queue, + qp->req.wqe_index); + wqe->state = wqe_state_done; + wqe->status = IB_WC_SUCCESS; + goto complete; + } + payload = mtu; + } + + skb = init_req_packet(qp, wqe, opcode, payload, &pkt); + if (unlikely(!skb)) { + pr_err("Failed allocating skb\n"); + goto err; + } + + if (fill_packet(qp, wqe, &pkt, skb, payload)) { + pr_debug("Error during fill packet\n"); + goto err; + } + + update_wqe_state(qp, wqe, &pkt, &prev_state); + ret = rxe_xmit_packet(to_rdev(qp->ibqp.device), qp, &pkt, skb); + if (ret) { + qp->need_req_skb = 1; + kfree_skb(skb); + + wqe->state = prev_state; + + if (ret == -EAGAIN) { + rxe_run_task(&qp->req.task, 1); + goto exit; + } + + goto err; + } + + update_state(qp, wqe, &pkt, payload); + + goto next_wqe; + +err: + kfree_skb(skb); + wqe->status = IB_WC_LOC_PROT_ERR; + wqe->state = wqe_state_error; + +complete: + if (qp_type(qp) != IB_QPT_RC) { + while (rxe_completer(qp) == 0) + ; + } + + return 0; + +exit: + return -EAGAIN; +} diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c new file mode 100644 index 000000000000..ebb03b46e2ad --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -0,0 +1,1380 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +enum resp_states { + RESPST_NONE, + RESPST_GET_REQ, + RESPST_CHK_PSN, + RESPST_CHK_OP_SEQ, + RESPST_CHK_OP_VALID, + RESPST_CHK_RESOURCE, + RESPST_CHK_LENGTH, + RESPST_CHK_RKEY, + RESPST_EXECUTE, + RESPST_READ_REPLY, + RESPST_COMPLETE, + RESPST_ACKNOWLEDGE, + RESPST_CLEANUP, + RESPST_DUPLICATE_REQUEST, + RESPST_ERR_MALFORMED_WQE, + RESPST_ERR_UNSUPPORTED_OPCODE, + RESPST_ERR_MISALIGNED_ATOMIC, + RESPST_ERR_PSN_OUT_OF_SEQ, + RESPST_ERR_MISSING_OPCODE_FIRST, + RESPST_ERR_MISSING_OPCODE_LAST_C, + RESPST_ERR_MISSING_OPCODE_LAST_D1E, + RESPST_ERR_TOO_MANY_RDMA_ATM_REQ, + RESPST_ERR_RNR, + RESPST_ERR_RKEY_VIOLATION, + RESPST_ERR_LENGTH, + RESPST_ERR_CQ_OVERFLOW, + RESPST_ERROR, + RESPST_RESET, + RESPST_DONE, + RESPST_EXIT, +}; + +static char *resp_state_name[] = { + [RESPST_NONE] = "NONE", + [RESPST_GET_REQ] = "GET_REQ", + [RESPST_CHK_PSN] = "CHK_PSN", + [RESPST_CHK_OP_SEQ] = "CHK_OP_SEQ", + [RESPST_CHK_OP_VALID] = "CHK_OP_VALID", + [RESPST_CHK_RESOURCE] = "CHK_RESOURCE", + [RESPST_CHK_LENGTH] = "CHK_LENGTH", + [RESPST_CHK_RKEY] = "CHK_RKEY", + [RESPST_EXECUTE] = "EXECUTE", + [RESPST_READ_REPLY] = "READ_REPLY", + [RESPST_COMPLETE] = "COMPLETE", + [RESPST_ACKNOWLEDGE] = "ACKNOWLEDGE", + [RESPST_CLEANUP] = "CLEANUP", + [RESPST_DUPLICATE_REQUEST] = "DUPLICATE_REQUEST", + [RESPST_ERR_MALFORMED_WQE] = "ERR_MALFORMED_WQE", + [RESPST_ERR_UNSUPPORTED_OPCODE] = "ERR_UNSUPPORTED_OPCODE", + [RESPST_ERR_MISALIGNED_ATOMIC] = "ERR_MISALIGNED_ATOMIC", + [RESPST_ERR_PSN_OUT_OF_SEQ] = "ERR_PSN_OUT_OF_SEQ", + [RESPST_ERR_MISSING_OPCODE_FIRST] = "ERR_MISSING_OPCODE_FIRST", + [RESPST_ERR_MISSING_OPCODE_LAST_C] = "ERR_MISSING_OPCODE_LAST_C", + [RESPST_ERR_MISSING_OPCODE_LAST_D1E] = "ERR_MISSING_OPCODE_LAST_D1E", + [RESPST_ERR_TOO_MANY_RDMA_ATM_REQ] = "ERR_TOO_MANY_RDMA_ATM_REQ", + [RESPST_ERR_RNR] = "ERR_RNR", + [RESPST_ERR_RKEY_VIOLATION] = "ERR_RKEY_VIOLATION", + [RESPST_ERR_LENGTH] = "ERR_LENGTH", + [RESPST_ERR_CQ_OVERFLOW] = "ERR_CQ_OVERFLOW", + [RESPST_ERROR] = "ERROR", + [RESPST_RESET] = "RESET", + [RESPST_DONE] = "DONE", + [RESPST_EXIT] = "EXIT", +}; + +/* rxe_recv calls here to add a request packet to the input queue */ +void rxe_resp_queue_pkt(struct rxe_dev *rxe, struct rxe_qp *qp, + struct sk_buff *skb) +{ + int must_sched; + struct rxe_pkt_info *pkt = SKB_TO_PKT(skb); + + skb_queue_tail(&qp->req_pkts, skb); + + must_sched = (pkt->opcode == IB_OPCODE_RC_RDMA_READ_REQUEST) || + (skb_queue_len(&qp->req_pkts) > 1); + + rxe_run_task(&qp->resp.task, must_sched); +} + +static inline enum resp_states get_req(struct rxe_qp *qp, + struct rxe_pkt_info **pkt_p) +{ + struct sk_buff *skb; + + if (qp->resp.state == QP_STATE_ERROR) { + skb = skb_dequeue(&qp->req_pkts); + if (skb) { + /* drain request packet queue */ + rxe_drop_ref(qp); + kfree_skb(skb); + return RESPST_GET_REQ; + } + + /* go drain recv wr queue */ + return RESPST_CHK_RESOURCE; + } + + skb = skb_peek(&qp->req_pkts); + if (!skb) + return RESPST_EXIT; + + *pkt_p = SKB_TO_PKT(skb); + + return (qp->resp.res) ? RESPST_READ_REPLY : RESPST_CHK_PSN; +} + +static enum resp_states check_psn(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + int diff = psn_compare(pkt->psn, qp->resp.psn); + + switch (qp_type(qp)) { + case IB_QPT_RC: + if (diff > 0) { + if (qp->resp.sent_psn_nak) + return RESPST_CLEANUP; + + qp->resp.sent_psn_nak = 1; + return RESPST_ERR_PSN_OUT_OF_SEQ; + + } else if (diff < 0) { + return RESPST_DUPLICATE_REQUEST; + } + + if (qp->resp.sent_psn_nak) + qp->resp.sent_psn_nak = 0; + + break; + + case IB_QPT_UC: + if (qp->resp.drop_msg || diff != 0) { + if (pkt->mask & RXE_START_MASK) { + qp->resp.drop_msg = 0; + return RESPST_CHK_OP_SEQ; + } + + qp->resp.drop_msg = 1; + return RESPST_CLEANUP; + } + break; + default: + break; + } + + return RESPST_CHK_OP_SEQ; +} + +static enum resp_states check_op_seq(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + switch (qp_type(qp)) { + case IB_QPT_RC: + switch (qp->resp.opcode) { + case IB_OPCODE_RC_SEND_FIRST: + case IB_OPCODE_RC_SEND_MIDDLE: + switch (pkt->opcode) { + case IB_OPCODE_RC_SEND_MIDDLE: + case IB_OPCODE_RC_SEND_LAST: + case IB_OPCODE_RC_SEND_LAST_WITH_IMMEDIATE: + case IB_OPCODE_RC_SEND_LAST_WITH_INVALIDATE: + return RESPST_CHK_OP_VALID; + default: + return RESPST_ERR_MISSING_OPCODE_LAST_C; + } + + case IB_OPCODE_RC_RDMA_WRITE_FIRST: + case IB_OPCODE_RC_RDMA_WRITE_MIDDLE: + switch (pkt->opcode) { + case IB_OPCODE_RC_RDMA_WRITE_MIDDLE: + case IB_OPCODE_RC_RDMA_WRITE_LAST: + case IB_OPCODE_RC_RDMA_WRITE_LAST_WITH_IMMEDIATE: + return RESPST_CHK_OP_VALID; + default: + return RESPST_ERR_MISSING_OPCODE_LAST_C; + } + + default: + switch (pkt->opcode) { + case IB_OPCODE_RC_SEND_MIDDLE: + case IB_OPCODE_RC_SEND_LAST: + case IB_OPCODE_RC_SEND_LAST_WITH_IMMEDIATE: + case IB_OPCODE_RC_SEND_LAST_WITH_INVALIDATE: + case IB_OPCODE_RC_RDMA_WRITE_MIDDLE: + case IB_OPCODE_RC_RDMA_WRITE_LAST: + case IB_OPCODE_RC_RDMA_WRITE_LAST_WITH_IMMEDIATE: + return RESPST_ERR_MISSING_OPCODE_FIRST; + default: + return RESPST_CHK_OP_VALID; + } + } + break; + + case IB_QPT_UC: + switch (qp->resp.opcode) { + case IB_OPCODE_UC_SEND_FIRST: + case IB_OPCODE_UC_SEND_MIDDLE: + switch (pkt->opcode) { + case IB_OPCODE_UC_SEND_MIDDLE: + case IB_OPCODE_UC_SEND_LAST: + case IB_OPCODE_UC_SEND_LAST_WITH_IMMEDIATE: + return RESPST_CHK_OP_VALID; + default: + return RESPST_ERR_MISSING_OPCODE_LAST_D1E; + } + + case IB_OPCODE_UC_RDMA_WRITE_FIRST: + case IB_OPCODE_UC_RDMA_WRITE_MIDDLE: + switch (pkt->opcode) { + case IB_OPCODE_UC_RDMA_WRITE_MIDDLE: + case IB_OPCODE_UC_RDMA_WRITE_LAST: + case IB_OPCODE_UC_RDMA_WRITE_LAST_WITH_IMMEDIATE: + return RESPST_CHK_OP_VALID; + default: + return RESPST_ERR_MISSING_OPCODE_LAST_D1E; + } + + default: + switch (pkt->opcode) { + case IB_OPCODE_UC_SEND_MIDDLE: + case IB_OPCODE_UC_SEND_LAST: + case IB_OPCODE_UC_SEND_LAST_WITH_IMMEDIATE: + case IB_OPCODE_UC_RDMA_WRITE_MIDDLE: + case IB_OPCODE_UC_RDMA_WRITE_LAST: + case IB_OPCODE_UC_RDMA_WRITE_LAST_WITH_IMMEDIATE: + qp->resp.drop_msg = 1; + return RESPST_CLEANUP; + default: + return RESPST_CHK_OP_VALID; + } + } + break; + + default: + return RESPST_CHK_OP_VALID; + } +} + +static enum resp_states check_op_valid(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + switch (qp_type(qp)) { + case IB_QPT_RC: + if (((pkt->mask & RXE_READ_MASK) && + !(qp->attr.qp_access_flags & IB_ACCESS_REMOTE_READ)) || + ((pkt->mask & RXE_WRITE_MASK) && + !(qp->attr.qp_access_flags & IB_ACCESS_REMOTE_WRITE)) || + ((pkt->mask & RXE_ATOMIC_MASK) && + !(qp->attr.qp_access_flags & IB_ACCESS_REMOTE_ATOMIC))) { + return RESPST_ERR_UNSUPPORTED_OPCODE; + } + + break; + + case IB_QPT_UC: + if ((pkt->mask & RXE_WRITE_MASK) && + !(qp->attr.qp_access_flags & IB_ACCESS_REMOTE_WRITE)) { + qp->resp.drop_msg = 1; + return RESPST_CLEANUP; + } + + break; + + case IB_QPT_UD: + case IB_QPT_SMI: + case IB_QPT_GSI: + break; + + default: + WARN_ON(1); + break; + } + + return RESPST_CHK_RESOURCE; +} + +static enum resp_states get_srq_wqe(struct rxe_qp *qp) +{ + struct rxe_srq *srq = qp->srq; + struct rxe_queue *q = srq->rq.queue; + struct rxe_recv_wqe *wqe; + struct ib_event ev; + + if (srq->error) + return RESPST_ERR_RNR; + + spin_lock_bh(&srq->rq.consumer_lock); + + wqe = queue_head(q); + if (!wqe) { + spin_unlock_bh(&srq->rq.consumer_lock); + return RESPST_ERR_RNR; + } + + /* note kernel and user space recv wqes have same size */ + memcpy(&qp->resp.srq_wqe, wqe, sizeof(qp->resp.srq_wqe)); + + qp->resp.wqe = &qp->resp.srq_wqe.wqe; + advance_consumer(q); + + if (srq->limit && srq->ibsrq.event_handler && + (queue_count(q) < srq->limit)) { + srq->limit = 0; + goto event; + } + + spin_unlock_bh(&srq->rq.consumer_lock); + return RESPST_CHK_LENGTH; + +event: + spin_unlock_bh(&srq->rq.consumer_lock); + ev.device = qp->ibqp.device; + ev.element.srq = qp->ibqp.srq; + ev.event = IB_EVENT_SRQ_LIMIT_REACHED; + srq->ibsrq.event_handler(&ev, srq->ibsrq.srq_context); + return RESPST_CHK_LENGTH; +} + +static enum resp_states check_resource(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + struct rxe_srq *srq = qp->srq; + + if (qp->resp.state == QP_STATE_ERROR) { + if (qp->resp.wqe) { + qp->resp.status = IB_WC_WR_FLUSH_ERR; + return RESPST_COMPLETE; + } else if (!srq) { + qp->resp.wqe = queue_head(qp->rq.queue); + if (qp->resp.wqe) { + qp->resp.status = IB_WC_WR_FLUSH_ERR; + return RESPST_COMPLETE; + } else { + return RESPST_EXIT; + } + } else { + return RESPST_EXIT; + } + } + + if (pkt->mask & RXE_READ_OR_ATOMIC) { + /* it is the requesters job to not send + * too many read/atomic ops, we just + * recycle the responder resource queue + */ + if (likely(qp->attr.max_rd_atomic > 0)) + return RESPST_CHK_LENGTH; + else + return RESPST_ERR_TOO_MANY_RDMA_ATM_REQ; + } + + if (pkt->mask & RXE_RWR_MASK) { + if (srq) + return get_srq_wqe(qp); + + qp->resp.wqe = queue_head(qp->rq.queue); + return (qp->resp.wqe) ? RESPST_CHK_LENGTH : RESPST_ERR_RNR; + } + + return RESPST_CHK_LENGTH; +} + +static enum resp_states check_length(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + switch (qp_type(qp)) { + case IB_QPT_RC: + return RESPST_CHK_RKEY; + + case IB_QPT_UC: + return RESPST_CHK_RKEY; + + default: + return RESPST_CHK_RKEY; + } +} + +static enum resp_states check_rkey(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + struct rxe_mem *mem; + u64 va; + u32 rkey; + u32 resid; + u32 pktlen; + int mtu = qp->mtu; + enum resp_states state; + int access; + + if (pkt->mask & (RXE_READ_MASK | RXE_WRITE_MASK)) { + if (pkt->mask & RXE_RETH_MASK) { + qp->resp.va = reth_va(pkt); + qp->resp.rkey = reth_rkey(pkt); + qp->resp.resid = reth_len(pkt); + } + access = (pkt->mask & RXE_READ_MASK) ? IB_ACCESS_REMOTE_READ + : IB_ACCESS_REMOTE_WRITE; + } else if (pkt->mask & RXE_ATOMIC_MASK) { + qp->resp.va = atmeth_va(pkt); + qp->resp.rkey = atmeth_rkey(pkt); + qp->resp.resid = sizeof(u64); + access = IB_ACCESS_REMOTE_ATOMIC; + } else { + return RESPST_EXECUTE; + } + + va = qp->resp.va; + rkey = qp->resp.rkey; + resid = qp->resp.resid; + pktlen = payload_size(pkt); + + mem = lookup_mem(qp->pd, access, rkey, lookup_remote); + if (!mem) { + state = RESPST_ERR_RKEY_VIOLATION; + goto err1; + } + + if (unlikely(mem->state == RXE_MEM_STATE_FREE)) { + state = RESPST_ERR_RKEY_VIOLATION; + goto err1; + } + + if (mem_check_range(mem, va, resid)) { + state = RESPST_ERR_RKEY_VIOLATION; + goto err2; + } + + if (pkt->mask & RXE_WRITE_MASK) { + if (resid > mtu) { + if (pktlen != mtu || bth_pad(pkt)) { + state = RESPST_ERR_LENGTH; + goto err2; + } + + resid = mtu; + } else { + if (pktlen != resid) { + state = RESPST_ERR_LENGTH; + goto err2; + } + if ((bth_pad(pkt) != (0x3 & (-resid)))) { + /* This case may not be exactly that + * but nothing else fits. + */ + state = RESPST_ERR_LENGTH; + goto err2; + } + } + } + + WARN_ON(qp->resp.mr); + + qp->resp.mr = mem; + return RESPST_EXECUTE; + +err2: + rxe_drop_ref(mem); +err1: + return state; +} + +static enum resp_states send_data_in(struct rxe_qp *qp, void *data_addr, + int data_len) +{ + int err; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + + err = copy_data(rxe, qp->pd, IB_ACCESS_LOCAL_WRITE, &qp->resp.wqe->dma, + data_addr, data_len, to_mem_obj, NULL); + if (unlikely(err)) + return (err == -ENOSPC) ? RESPST_ERR_LENGTH + : RESPST_ERR_MALFORMED_WQE; + + return RESPST_NONE; +} + +static enum resp_states write_data_in(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + enum resp_states rc = RESPST_NONE; + int err; + int data_len = payload_size(pkt); + + err = rxe_mem_copy(qp->resp.mr, qp->resp.va, payload_addr(pkt), + data_len, to_mem_obj, NULL); + if (err) { + rc = RESPST_ERR_RKEY_VIOLATION; + goto out; + } + + qp->resp.va += data_len; + qp->resp.resid -= data_len; + +out: + return rc; +} + +/* Guarantee atomicity of atomic operations at the machine level. */ +static DEFINE_SPINLOCK(atomic_ops_lock); + +static enum resp_states process_atomic(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + u64 iova = atmeth_va(pkt); + u64 *vaddr; + enum resp_states ret; + struct rxe_mem *mr = qp->resp.mr; + + if (mr->state != RXE_MEM_STATE_VALID) { + ret = RESPST_ERR_RKEY_VIOLATION; + goto out; + } + + vaddr = iova_to_vaddr(mr, iova, sizeof(u64)); + + /* check vaddr is 8 bytes aligned. */ + if (!vaddr || (uintptr_t)vaddr & 7) { + ret = RESPST_ERR_MISALIGNED_ATOMIC; + goto out; + } + + spin_lock_bh(&atomic_ops_lock); + + qp->resp.atomic_orig = *vaddr; + + if (pkt->opcode == IB_OPCODE_RC_COMPARE_SWAP || + pkt->opcode == IB_OPCODE_RD_COMPARE_SWAP) { + if (*vaddr == atmeth_comp(pkt)) + *vaddr = atmeth_swap_add(pkt); + } else { + *vaddr += atmeth_swap_add(pkt); + } + + spin_unlock_bh(&atomic_ops_lock); + + ret = RESPST_NONE; +out: + return ret; +} + +static struct sk_buff *prepare_ack_packet(struct rxe_qp *qp, + struct rxe_pkt_info *pkt, + struct rxe_pkt_info *ack, + int opcode, + int payload, + u32 psn, + u8 syndrome, + u32 *crcp) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct sk_buff *skb; + u32 crc = 0; + u32 *p; + int paylen; + int pad; + int err; + + /* + * allocate packet + */ + pad = (-payload) & 0x3; + paylen = rxe_opcode[opcode].length + payload + pad + RXE_ICRC_SIZE; + + skb = rxe->ifc_ops->init_packet(rxe, &qp->pri_av, paylen, ack); + if (!skb) + return NULL; + + ack->qp = qp; + ack->opcode = opcode; + ack->mask = rxe_opcode[opcode].mask; + ack->offset = pkt->offset; + ack->paylen = paylen; + + /* fill in bth using the request packet headers */ + memcpy(ack->hdr, pkt->hdr, pkt->offset + RXE_BTH_BYTES); + + bth_set_opcode(ack, opcode); + bth_set_qpn(ack, qp->attr.dest_qp_num); + bth_set_pad(ack, pad); + bth_set_se(ack, 0); + bth_set_psn(ack, psn); + bth_set_ack(ack, 0); + ack->psn = psn; + + if (ack->mask & RXE_AETH_MASK) { + aeth_set_syn(ack, syndrome); + aeth_set_msn(ack, qp->resp.msn); + } + + if (ack->mask & RXE_ATMACK_MASK) + atmack_set_orig(ack, qp->resp.atomic_orig); + + err = rxe->ifc_ops->prepare(rxe, ack, skb, &crc); + if (err) { + kfree_skb(skb); + return NULL; + } + + if (crcp) { + /* CRC computation will be continued by the caller */ + *crcp = crc; + } else { + p = payload_addr(ack) + payload + bth_pad(ack); + *p = ~crc; + } + + return skb; +} + +/* RDMA read response. If res is not NULL, then we have a current RDMA request + * being processed or replayed. + */ +static enum resp_states read_reply(struct rxe_qp *qp, + struct rxe_pkt_info *req_pkt) +{ + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct rxe_pkt_info ack_pkt; + struct sk_buff *skb; + int mtu = qp->mtu; + enum resp_states state; + int payload; + int opcode; + int err; + struct resp_res *res = qp->resp.res; + u32 icrc; + u32 *p; + + if (!res) { + /* This is the first time we process that request. Get a + * resource + */ + res = &qp->resp.resources[qp->resp.res_head]; + + free_rd_atomic_resource(qp, res); + rxe_advance_resp_resource(qp); + + res->type = RXE_READ_MASK; + + res->read.va = qp->resp.va; + res->read.va_org = qp->resp.va; + + res->first_psn = req_pkt->psn; + res->last_psn = req_pkt->psn + + (reth_len(req_pkt) + mtu - 1) / + mtu - 1; + res->cur_psn = req_pkt->psn; + + res->read.resid = qp->resp.resid; + res->read.length = qp->resp.resid; + res->read.rkey = qp->resp.rkey; + + /* note res inherits the reference to mr from qp */ + res->read.mr = qp->resp.mr; + qp->resp.mr = NULL; + + qp->resp.res = res; + res->state = rdatm_res_state_new; + } + + if (res->state == rdatm_res_state_new) { + if (res->read.resid <= mtu) + opcode = IB_OPCODE_RC_RDMA_READ_RESPONSE_ONLY; + else + opcode = IB_OPCODE_RC_RDMA_READ_RESPONSE_FIRST; + } else { + if (res->read.resid > mtu) + opcode = IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE; + else + opcode = IB_OPCODE_RC_RDMA_READ_RESPONSE_LAST; + } + + res->state = rdatm_res_state_next; + + payload = min_t(int, res->read.resid, mtu); + + skb = prepare_ack_packet(qp, req_pkt, &ack_pkt, opcode, payload, + res->cur_psn, AETH_ACK_UNLIMITED, &icrc); + if (!skb) + return RESPST_ERR_RNR; + + err = rxe_mem_copy(res->read.mr, res->read.va, payload_addr(&ack_pkt), + payload, from_mem_obj, &icrc); + if (err) + pr_err("Failed copying memory\n"); + + p = payload_addr(&ack_pkt) + payload + bth_pad(&ack_pkt); + *p = ~icrc; + + err = rxe_xmit_packet(rxe, qp, &ack_pkt, skb); + if (err) { + pr_err("Failed sending RDMA reply.\n"); + kfree_skb(skb); + return RESPST_ERR_RNR; + } + + res->read.va += payload; + res->read.resid -= payload; + res->cur_psn = (res->cur_psn + 1) & BTH_PSN_MASK; + + if (res->read.resid > 0) { + state = RESPST_DONE; + } else { + qp->resp.res = NULL; + qp->resp.opcode = -1; + qp->resp.psn = res->cur_psn; + state = RESPST_CLEANUP; + } + + return state; +} + +/* Executes a new request. A retried request never reach that function (send + * and writes are discarded, and reads and atomics are retried elsewhere. + */ +static enum resp_states execute(struct rxe_qp *qp, struct rxe_pkt_info *pkt) +{ + enum resp_states err; + + if (pkt->mask & RXE_SEND_MASK) { + if (qp_type(qp) == IB_QPT_UD || + qp_type(qp) == IB_QPT_SMI || + qp_type(qp) == IB_QPT_GSI) { + union rdma_network_hdr hdr; + struct sk_buff *skb = PKT_TO_SKB(pkt); + + memset(&hdr, 0, sizeof(hdr)); + if (skb->protocol == htons(ETH_P_IP)) + memcpy(&hdr.roce4grh, ip_hdr(skb), sizeof(hdr.roce4grh)); + else if (skb->protocol == htons(ETH_P_IPV6)) + memcpy(&hdr.ibgrh, ipv6_hdr(skb), sizeof(hdr.ibgrh)); + + err = send_data_in(qp, &hdr, sizeof(hdr)); + if (err) + return err; + } + err = send_data_in(qp, payload_addr(pkt), payload_size(pkt)); + if (err) + return err; + } else if (pkt->mask & RXE_WRITE_MASK) { + err = write_data_in(qp, pkt); + if (err) + return err; + } else if (pkt->mask & RXE_READ_MASK) { + /* For RDMA Read we can increment the msn now. See C9-148. */ + qp->resp.msn++; + return RESPST_READ_REPLY; + } else if (pkt->mask & RXE_ATOMIC_MASK) { + err = process_atomic(qp, pkt); + if (err) + return err; + } else + /* Unreachable */ + WARN_ON(1); + + /* We successfully processed this new request. */ + qp->resp.msn++; + + /* next expected psn, read handles this separately */ + qp->resp.psn = (pkt->psn + 1) & BTH_PSN_MASK; + + qp->resp.opcode = pkt->opcode; + qp->resp.status = IB_WC_SUCCESS; + + if (pkt->mask & RXE_COMP_MASK) + return RESPST_COMPLETE; + else if (qp_type(qp) == IB_QPT_RC) + return RESPST_ACKNOWLEDGE; + else + return RESPST_CLEANUP; +} + +static enum resp_states do_complete(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + struct rxe_cqe cqe; + struct ib_wc *wc = &cqe.ibwc; + struct ib_uverbs_wc *uwc = &cqe.uibwc; + struct rxe_recv_wqe *wqe = qp->resp.wqe; + + if (unlikely(!wqe)) + return RESPST_CLEANUP; + + memset(&cqe, 0, sizeof(cqe)); + + wc->wr_id = wqe->wr_id; + wc->status = qp->resp.status; + wc->qp = &qp->ibqp; + + /* fields after status are not required for errors */ + if (wc->status == IB_WC_SUCCESS) { + wc->opcode = (pkt->mask & RXE_IMMDT_MASK && + pkt->mask & RXE_WRITE_MASK) ? + IB_WC_RECV_RDMA_WITH_IMM : IB_WC_RECV; + wc->vendor_err = 0; + wc->byte_len = wqe->dma.length - wqe->dma.resid; + + /* fields after byte_len are different between kernel and user + * space + */ + if (qp->rcq->is_user) { + uwc->wc_flags = IB_WC_GRH; + + if (pkt->mask & RXE_IMMDT_MASK) { + uwc->wc_flags |= IB_WC_WITH_IMM; + uwc->ex.imm_data = + (__u32 __force)immdt_imm(pkt); + } + + if (pkt->mask & RXE_IETH_MASK) { + uwc->wc_flags |= IB_WC_WITH_INVALIDATE; + uwc->ex.invalidate_rkey = ieth_rkey(pkt); + } + + uwc->qp_num = qp->ibqp.qp_num; + + if (pkt->mask & RXE_DETH_MASK) + uwc->src_qp = deth_sqp(pkt); + + uwc->port_num = qp->attr.port_num; + } else { + struct sk_buff *skb = PKT_TO_SKB(pkt); + + wc->wc_flags = IB_WC_GRH | IB_WC_WITH_NETWORK_HDR_TYPE; + if (skb->protocol == htons(ETH_P_IP)) + wc->network_hdr_type = RDMA_NETWORK_IPV4; + else + wc->network_hdr_type = RDMA_NETWORK_IPV6; + + if (pkt->mask & RXE_IMMDT_MASK) { + wc->wc_flags |= IB_WC_WITH_IMM; + wc->ex.imm_data = immdt_imm(pkt); + } + + if (pkt->mask & RXE_IETH_MASK) { + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct rxe_mem *rmr; + + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + wc->ex.invalidate_rkey = ieth_rkey(pkt); + + rmr = rxe_pool_get_index(&rxe->mr_pool, + wc->ex.invalidate_rkey >> 8); + if (unlikely(!rmr)) { + pr_err("Bad rkey %#x invalidation\n", wc->ex.invalidate_rkey); + return RESPST_ERROR; + } + rmr->state = RXE_MEM_STATE_FREE; + } + + wc->qp = &qp->ibqp; + + if (pkt->mask & RXE_DETH_MASK) + wc->src_qp = deth_sqp(pkt); + + wc->port_num = qp->attr.port_num; + } + } + + /* have copy for srq and reference for !srq */ + if (!qp->srq) + advance_consumer(qp->rq.queue); + + qp->resp.wqe = NULL; + + if (rxe_cq_post(qp->rcq, &cqe, pkt ? bth_se(pkt) : 1)) + return RESPST_ERR_CQ_OVERFLOW; + + if (qp->resp.state == QP_STATE_ERROR) + return RESPST_CHK_RESOURCE; + + if (!pkt) + return RESPST_DONE; + else if (qp_type(qp) == IB_QPT_RC) + return RESPST_ACKNOWLEDGE; + else + return RESPST_CLEANUP; +} + +static int send_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt, + u8 syndrome, u32 psn) +{ + int err = 0; + struct rxe_pkt_info ack_pkt; + struct sk_buff *skb; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + + skb = prepare_ack_packet(qp, pkt, &ack_pkt, IB_OPCODE_RC_ACKNOWLEDGE, + 0, psn, syndrome, NULL); + if (!skb) { + err = -ENOMEM; + goto err1; + } + + err = rxe_xmit_packet(rxe, qp, &ack_pkt, skb); + if (err) { + pr_err_ratelimited("Failed sending ack\n"); + kfree_skb(skb); + } + +err1: + return err; +} + +static int send_atomic_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt, + u8 syndrome) +{ + int rc = 0; + struct rxe_pkt_info ack_pkt; + struct sk_buff *skb; + struct sk_buff *skb_copy; + struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + struct resp_res *res; + + skb = prepare_ack_packet(qp, pkt, &ack_pkt, + IB_OPCODE_RC_ATOMIC_ACKNOWLEDGE, 0, pkt->psn, + syndrome, NULL); + if (!skb) { + rc = -ENOMEM; + goto out; + } + + skb_copy = skb_clone(skb, GFP_ATOMIC); + if (skb_copy) + rxe_add_ref(qp); /* for the new SKB */ + else { + pr_warn("Could not clone atomic response\n"); + rc = -ENOMEM; + goto out; + } + + res = &qp->resp.resources[qp->resp.res_head]; + free_rd_atomic_resource(qp, res); + rxe_advance_resp_resource(qp); + + res->type = RXE_ATOMIC_MASK; + res->atomic.skb = skb; + res->first_psn = qp->resp.psn; + res->last_psn = qp->resp.psn; + res->cur_psn = qp->resp.psn; + + rc = rxe_xmit_packet(rxe, qp, &ack_pkt, skb_copy); + if (rc) { + pr_err_ratelimited("Failed sending ack\n"); + rxe_drop_ref(qp); + kfree_skb(skb_copy); + } + +out: + return rc; +} + +static enum resp_states acknowledge(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + if (qp_type(qp) != IB_QPT_RC) + return RESPST_CLEANUP; + + if (qp->resp.aeth_syndrome != AETH_ACK_UNLIMITED) + send_ack(qp, pkt, qp->resp.aeth_syndrome, pkt->psn); + else if (pkt->mask & RXE_ATOMIC_MASK) + send_atomic_ack(qp, pkt, AETH_ACK_UNLIMITED); + else if (bth_ack(pkt)) + send_ack(qp, pkt, AETH_ACK_UNLIMITED, pkt->psn); + + return RESPST_CLEANUP; +} + +static enum resp_states cleanup(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + struct sk_buff *skb; + + if (pkt) { + skb = skb_dequeue(&qp->req_pkts); + rxe_drop_ref(qp); + kfree_skb(skb); + } + + if (qp->resp.mr) { + rxe_drop_ref(qp->resp.mr); + qp->resp.mr = NULL; + } + + return RESPST_DONE; +} + +static struct resp_res *find_resource(struct rxe_qp *qp, u32 psn) +{ + int i; + + for (i = 0; i < qp->attr.max_rd_atomic; i++) { + struct resp_res *res = &qp->resp.resources[i]; + + if (res->type == 0) + continue; + + if (psn_compare(psn, res->first_psn) >= 0 && + psn_compare(psn, res->last_psn) <= 0) { + return res; + } + } + + return NULL; +} + +static enum resp_states duplicate_request(struct rxe_qp *qp, + struct rxe_pkt_info *pkt) +{ + enum resp_states rc; + + if (pkt->mask & RXE_SEND_MASK || + pkt->mask & RXE_WRITE_MASK) { + /* SEND. Ack again and cleanup. C9-105. */ + if (bth_ack(pkt)) + send_ack(qp, pkt, AETH_ACK_UNLIMITED, qp->resp.psn - 1); + rc = RESPST_CLEANUP; + goto out; + } else if (pkt->mask & RXE_READ_MASK) { + struct resp_res *res; + + res = find_resource(qp, pkt->psn); + if (!res) { + /* Resource not found. Class D error. Drop the + * request. + */ + rc = RESPST_CLEANUP; + goto out; + } else { + /* Ensure this new request is the same as the previous + * one or a subset of it. + */ + u64 iova = reth_va(pkt); + u32 resid = reth_len(pkt); + + if (iova < res->read.va_org || + resid > res->read.length || + (iova + resid) > (res->read.va_org + + res->read.length)) { + rc = RESPST_CLEANUP; + goto out; + } + + if (reth_rkey(pkt) != res->read.rkey) { + rc = RESPST_CLEANUP; + goto out; + } + + res->cur_psn = pkt->psn; + res->state = (pkt->psn == res->first_psn) ? + rdatm_res_state_new : + rdatm_res_state_replay; + + /* Reset the resource, except length. */ + res->read.va_org = iova; + res->read.va = iova; + res->read.resid = resid; + + /* Replay the RDMA read reply. */ + qp->resp.res = res; + rc = RESPST_READ_REPLY; + goto out; + } + } else { + struct resp_res *res; + + /* Find the operation in our list of responder resources. */ + res = find_resource(qp, pkt->psn); + if (res) { + struct sk_buff *skb_copy; + + skb_copy = skb_clone(res->atomic.skb, GFP_ATOMIC); + if (skb_copy) { + rxe_add_ref(qp); /* for the new SKB */ + } else { + pr_warn("Couldn't clone atomic resp\n"); + rc = RESPST_CLEANUP; + goto out; + } + bth_set_psn(SKB_TO_PKT(skb_copy), + qp->resp.psn - 1); + /* Resend the result. */ + rc = rxe_xmit_packet(to_rdev(qp->ibqp.device), qp, + pkt, skb_copy); + if (rc) { + pr_err("Failed resending result. This flow is not handled - skb ignored\n"); + kfree_skb(skb_copy); + rc = RESPST_CLEANUP; + goto out; + } + } + + /* Resource not found. Class D error. Drop the request. */ + rc = RESPST_CLEANUP; + goto out; + } +out: + return rc; +} + +/* Process a class A or C. Both are treated the same in this implementation. */ +static void do_class_ac_error(struct rxe_qp *qp, u8 syndrome, + enum ib_wc_status status) +{ + qp->resp.aeth_syndrome = syndrome; + qp->resp.status = status; + + /* indicate that we should go through the ERROR state */ + qp->resp.goto_error = 1; +} + +static enum resp_states do_class_d1e_error(struct rxe_qp *qp) +{ + /* UC */ + if (qp->srq) { + /* Class E */ + qp->resp.drop_msg = 1; + if (qp->resp.wqe) { + qp->resp.status = IB_WC_REM_INV_REQ_ERR; + return RESPST_COMPLETE; + } else { + return RESPST_CLEANUP; + } + } else { + /* Class D1. This packet may be the start of a + * new message and could be valid. The previous + * message is invalid and ignored. reset the + * recv wr to its original state + */ + if (qp->resp.wqe) { + qp->resp.wqe->dma.resid = qp->resp.wqe->dma.length; + qp->resp.wqe->dma.cur_sge = 0; + qp->resp.wqe->dma.sge_offset = 0; + qp->resp.opcode = -1; + } + + if (qp->resp.mr) { + rxe_drop_ref(qp->resp.mr); + qp->resp.mr = NULL; + } + + return RESPST_CLEANUP; + } +} + +int rxe_responder(void *arg) +{ + struct rxe_qp *qp = (struct rxe_qp *)arg; + enum resp_states state; + struct rxe_pkt_info *pkt = NULL; + int ret = 0; + + qp->resp.aeth_syndrome = AETH_ACK_UNLIMITED; + + if (!qp->valid) { + ret = -EINVAL; + goto done; + } + + switch (qp->resp.state) { + case QP_STATE_RESET: + state = RESPST_RESET; + break; + + default: + state = RESPST_GET_REQ; + break; + } + + while (1) { + pr_debug("state = %s\n", resp_state_name[state]); + switch (state) { + case RESPST_GET_REQ: + state = get_req(qp, &pkt); + break; + case RESPST_CHK_PSN: + state = check_psn(qp, pkt); + break; + case RESPST_CHK_OP_SEQ: + state = check_op_seq(qp, pkt); + break; + case RESPST_CHK_OP_VALID: + state = check_op_valid(qp, pkt); + break; + case RESPST_CHK_RESOURCE: + state = check_resource(qp, pkt); + break; + case RESPST_CHK_LENGTH: + state = check_length(qp, pkt); + break; + case RESPST_CHK_RKEY: + state = check_rkey(qp, pkt); + break; + case RESPST_EXECUTE: + state = execute(qp, pkt); + break; + case RESPST_COMPLETE: + state = do_complete(qp, pkt); + break; + case RESPST_READ_REPLY: + state = read_reply(qp, pkt); + break; + case RESPST_ACKNOWLEDGE: + state = acknowledge(qp, pkt); + break; + case RESPST_CLEANUP: + state = cleanup(qp, pkt); + break; + case RESPST_DUPLICATE_REQUEST: + state = duplicate_request(qp, pkt); + break; + case RESPST_ERR_PSN_OUT_OF_SEQ: + /* RC only - Class B. Drop packet. */ + send_ack(qp, pkt, AETH_NAK_PSN_SEQ_ERROR, qp->resp.psn); + state = RESPST_CLEANUP; + break; + + case RESPST_ERR_TOO_MANY_RDMA_ATM_REQ: + case RESPST_ERR_MISSING_OPCODE_FIRST: + case RESPST_ERR_MISSING_OPCODE_LAST_C: + case RESPST_ERR_UNSUPPORTED_OPCODE: + case RESPST_ERR_MISALIGNED_ATOMIC: + /* RC Only - Class C. */ + do_class_ac_error(qp, AETH_NAK_INVALID_REQ, + IB_WC_REM_INV_REQ_ERR); + state = RESPST_COMPLETE; + break; + + case RESPST_ERR_MISSING_OPCODE_LAST_D1E: + state = do_class_d1e_error(qp); + break; + case RESPST_ERR_RNR: + if (qp_type(qp) == IB_QPT_RC) { + /* RC - class B */ + send_ack(qp, pkt, AETH_RNR_NAK | + (~AETH_TYPE_MASK & + qp->attr.min_rnr_timer), + pkt->psn); + } else { + /* UD/UC - class D */ + qp->resp.drop_msg = 1; + } + state = RESPST_CLEANUP; + break; + + case RESPST_ERR_RKEY_VIOLATION: + if (qp_type(qp) == IB_QPT_RC) { + /* Class C */ + do_class_ac_error(qp, AETH_NAK_REM_ACC_ERR, + IB_WC_REM_ACCESS_ERR); + state = RESPST_COMPLETE; + } else { + qp->resp.drop_msg = 1; + if (qp->srq) { + /* UC/SRQ Class D */ + qp->resp.status = IB_WC_REM_ACCESS_ERR; + state = RESPST_COMPLETE; + } else { + /* UC/non-SRQ Class E. */ + state = RESPST_CLEANUP; + } + } + break; + + case RESPST_ERR_LENGTH: + if (qp_type(qp) == IB_QPT_RC) { + /* Class C */ + do_class_ac_error(qp, AETH_NAK_INVALID_REQ, + IB_WC_REM_INV_REQ_ERR); + state = RESPST_COMPLETE; + } else if (qp->srq) { + /* UC/UD - class E */ + qp->resp.status = IB_WC_REM_INV_REQ_ERR; + state = RESPST_COMPLETE; + } else { + /* UC/UD - class D */ + qp->resp.drop_msg = 1; + state = RESPST_CLEANUP; + } + break; + + case RESPST_ERR_MALFORMED_WQE: + /* All, Class A. */ + do_class_ac_error(qp, AETH_NAK_REM_OP_ERR, + IB_WC_LOC_QP_OP_ERR); + state = RESPST_COMPLETE; + break; + + case RESPST_ERR_CQ_OVERFLOW: + /* All - Class G */ + state = RESPST_ERROR; + break; + + case RESPST_DONE: + if (qp->resp.goto_error) { + state = RESPST_ERROR; + break; + } + + goto done; + + case RESPST_EXIT: + if (qp->resp.goto_error) { + state = RESPST_ERROR; + break; + } + + goto exit; + + case RESPST_RESET: { + struct sk_buff *skb; + + while ((skb = skb_dequeue(&qp->req_pkts))) { + rxe_drop_ref(qp); + kfree_skb(skb); + } + + while (!qp->srq && qp->rq.queue && + queue_head(qp->rq.queue)) + advance_consumer(qp->rq.queue); + + qp->resp.wqe = NULL; + goto exit; + } + + case RESPST_ERROR: + qp->resp.goto_error = 0; + pr_warn("qp#%d moved to error state\n", qp_num(qp)); + rxe_qp_error(qp); + goto exit; + + default: + WARN_ON(1); + } + } + +exit: + ret = -EAGAIN; +done: + return ret; +} diff --git a/drivers/infiniband/sw/rxe/rxe_srq.c b/drivers/infiniband/sw/rxe/rxe_srq.c new file mode 100644 index 000000000000..2a6e3cd2d4e8 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_srq.c @@ -0,0 +1,193 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +int rxe_srq_chk_attr(struct rxe_dev *rxe, struct rxe_srq *srq, + struct ib_srq_attr *attr, enum ib_srq_attr_mask mask) +{ + if (srq && srq->error) { + pr_warn("srq in error state\n"); + goto err1; + } + + if (mask & IB_SRQ_MAX_WR) { + if (attr->max_wr > rxe->attr.max_srq_wr) { + pr_warn("max_wr(%d) > max_srq_wr(%d)\n", + attr->max_wr, rxe->attr.max_srq_wr); + goto err1; + } + + if (attr->max_wr <= 0) { + pr_warn("max_wr(%d) <= 0\n", attr->max_wr); + goto err1; + } + + if (srq && srq->limit && (attr->max_wr < srq->limit)) { + pr_warn("max_wr (%d) < srq->limit (%d)\n", + attr->max_wr, srq->limit); + goto err1; + } + + if (attr->max_wr < RXE_MIN_SRQ_WR) + attr->max_wr = RXE_MIN_SRQ_WR; + } + + if (mask & IB_SRQ_LIMIT) { + if (attr->srq_limit > rxe->attr.max_srq_wr) { + pr_warn("srq_limit(%d) > max_srq_wr(%d)\n", + attr->srq_limit, rxe->attr.max_srq_wr); + goto err1; + } + + if (srq && (attr->srq_limit > srq->rq.queue->buf->index_mask)) { + pr_warn("srq_limit (%d) > cur limit(%d)\n", + attr->srq_limit, + srq->rq.queue->buf->index_mask); + goto err1; + } + } + + if (mask == IB_SRQ_INIT_MASK) { + if (attr->max_sge > rxe->attr.max_srq_sge) { + pr_warn("max_sge(%d) > max_srq_sge(%d)\n", + attr->max_sge, rxe->attr.max_srq_sge); + goto err1; + } + + if (attr->max_sge < RXE_MIN_SRQ_SGE) + attr->max_sge = RXE_MIN_SRQ_SGE; + } + + return 0; + +err1: + return -EINVAL; +} + +int rxe_srq_from_init(struct rxe_dev *rxe, struct rxe_srq *srq, + struct ib_srq_init_attr *init, + struct ib_ucontext *context, struct ib_udata *udata) +{ + int err; + int srq_wqe_size; + struct rxe_queue *q; + + srq->ibsrq.event_handler = init->event_handler; + srq->ibsrq.srq_context = init->srq_context; + srq->limit = init->attr.srq_limit; + srq->srq_num = srq->pelem.index; + srq->rq.max_wr = init->attr.max_wr; + srq->rq.max_sge = init->attr.max_sge; + + srq_wqe_size = rcv_wqe_size(srq->rq.max_sge); + + spin_lock_init(&srq->rq.producer_lock); + spin_lock_init(&srq->rq.consumer_lock); + + q = rxe_queue_init(rxe, &srq->rq.max_wr, + srq_wqe_size); + if (!q) { + pr_warn("unable to allocate queue for srq\n"); + return -ENOMEM; + } + + srq->rq.queue = q; + + err = do_mmap_info(rxe, udata, false, context, q->buf, + q->buf_size, &q->ip); + if (err) + return err; + + if (udata && udata->outlen >= sizeof(struct mminfo) + sizeof(u32)) { + if (copy_to_user(udata->outbuf + sizeof(struct mminfo), + &srq->srq_num, sizeof(u32))) + return -EFAULT; + } + return 0; +} + +int rxe_srq_from_attr(struct rxe_dev *rxe, struct rxe_srq *srq, + struct ib_srq_attr *attr, enum ib_srq_attr_mask mask, + struct ib_udata *udata) +{ + int err; + struct rxe_queue *q = srq->rq.queue; + struct mminfo mi = { .offset = 1, .size = 0}; + + if (mask & IB_SRQ_MAX_WR) { + /* Check that we can write the mminfo struct to user space */ + if (udata && udata->inlen >= sizeof(__u64)) { + __u64 mi_addr; + + /* Get address of user space mminfo struct */ + err = ib_copy_from_udata(&mi_addr, udata, + sizeof(mi_addr)); + if (err) + goto err1; + + udata->outbuf = (void __user *)(unsigned long)mi_addr; + udata->outlen = sizeof(mi); + + if (!access_ok(VERIFY_WRITE, + (void __user *)udata->outbuf, + udata->outlen)) { + err = -EFAULT; + goto err1; + } + } + + err = rxe_queue_resize(q, (unsigned int *)&attr->max_wr, + rcv_wqe_size(srq->rq.max_sge), + srq->rq.queue->ip ? + srq->rq.queue->ip->context : + NULL, + udata, &srq->rq.producer_lock, + &srq->rq.consumer_lock); + if (err) + goto err2; + } + + if (mask & IB_SRQ_LIMIT) + srq->limit = attr->srq_limit; + + return 0; + +err2: + rxe_queue_cleanup(q); + srq->rq.queue = NULL; +err1: + return err; +} diff --git a/drivers/infiniband/sw/rxe/rxe_sysfs.c b/drivers/infiniband/sw/rxe/rxe_sysfs.c new file mode 100644 index 000000000000..cf8e77800046 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_sysfs.c @@ -0,0 +1,157 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_net.h" + +/* Copy argument and remove trailing CR. Return the new length. */ +static int sanitize_arg(const char *val, char *intf, int intf_len) +{ + int len; + + if (!val) + return 0; + + /* Remove newline. */ + for (len = 0; len < intf_len - 1 && val[len] && val[len] != '\n'; len++) + intf[len] = val[len]; + intf[len] = 0; + + if (len == 0 || (val[len] != 0 && val[len] != '\n')) + return 0; + + return len; +} + +static void rxe_set_port_state(struct net_device *ndev) +{ + struct rxe_dev *rxe = net_to_rxe(ndev); + bool is_up = netif_running(ndev) && netif_carrier_ok(ndev); + + if (!rxe) + goto out; + + if (is_up) + rxe_port_up(rxe); + else + rxe_port_down(rxe); /* down for unknown state */ +out: + return; +} + +static int rxe_param_set_add(const char *val, const struct kernel_param *kp) +{ + int len; + int err = 0; + char intf[32]; + struct net_device *ndev = NULL; + struct rxe_dev *rxe; + + len = sanitize_arg(val, intf, sizeof(intf)); + if (!len) { + pr_err("rxe: add: invalid interface name\n"); + err = -EINVAL; + goto err; + } + + ndev = dev_get_by_name(&init_net, intf); + if (!ndev) { + pr_err("interface %s not found\n", intf); + err = -EINVAL; + goto err; + } + + if (net_to_rxe(ndev)) { + pr_err("rxe: already configured on %s\n", intf); + err = -EINVAL; + goto err; + } + + rxe = rxe_net_add(ndev); + if (!rxe) { + pr_err("rxe: failed to add %s\n", intf); + err = -EINVAL; + goto err; + } + + rxe_set_port_state(ndev); + pr_info("rxe: added %s to %s\n", rxe->ib_dev.name, intf); +err: + if (ndev) + dev_put(ndev); + return err; +} + +static int rxe_param_set_remove(const char *val, const struct kernel_param *kp) +{ + int len; + char intf[32]; + struct rxe_dev *rxe; + + len = sanitize_arg(val, intf, sizeof(intf)); + if (!len) { + pr_err("rxe: add: invalid interface name\n"); + return -EINVAL; + } + + if (strncmp("all", intf, len) == 0) { + pr_info("rxe_sys: remove all"); + rxe_remove_all(); + return 0; + } + + rxe = get_rxe_by_name(intf); + + if (!rxe) { + pr_err("rxe: not configured on %s\n", intf); + return -EINVAL; + } + + list_del(&rxe->list); + rxe_remove(rxe); + + return 0; +} + +static const struct kernel_param_ops rxe_add_ops = { + .set = rxe_param_set_add, +}; + +static const struct kernel_param_ops rxe_remove_ops = { + .set = rxe_param_set_remove, +}; + +module_param_cb(add, &rxe_add_ops, NULL, 0200); +MODULE_PARM_DESC(add, "Create RXE device over network interface"); +module_param_cb(remove, &rxe_remove_ops, NULL, 0200); +MODULE_PARM_DESC(remove, "Remove RXE device over network interface"); diff --git a/drivers/infiniband/sw/rxe/rxe_task.c b/drivers/infiniband/sw/rxe/rxe_task.c new file mode 100644 index 000000000000..1e19bf828a6e --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_task.c @@ -0,0 +1,154 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include + +#include "rxe_task.h" + +int __rxe_do_task(struct rxe_task *task) + +{ + int ret; + + while ((ret = task->func(task->arg)) == 0) + ; + + task->ret = ret; + + return ret; +} + +/* + * this locking is due to a potential race where + * a second caller finds the task already running + * but looks just after the last call to func + */ +void rxe_do_task(unsigned long data) +{ + int cont; + int ret; + unsigned long flags; + struct rxe_task *task = (struct rxe_task *)data; + + spin_lock_irqsave(&task->state_lock, flags); + switch (task->state) { + case TASK_STATE_START: + task->state = TASK_STATE_BUSY; + spin_unlock_irqrestore(&task->state_lock, flags); + break; + + case TASK_STATE_BUSY: + task->state = TASK_STATE_ARMED; + /* fall through to */ + case TASK_STATE_ARMED: + spin_unlock_irqrestore(&task->state_lock, flags); + return; + + default: + spin_unlock_irqrestore(&task->state_lock, flags); + pr_warn("bad state = %d in rxe_do_task\n", task->state); + return; + } + + do { + cont = 0; + ret = task->func(task->arg); + + spin_lock_irqsave(&task->state_lock, flags); + switch (task->state) { + case TASK_STATE_BUSY: + if (ret) + task->state = TASK_STATE_START; + else + cont = 1; + break; + + /* soneone tried to run the task since the last time we called + * func, so we will call one more time regardless of the + * return value + */ + case TASK_STATE_ARMED: + task->state = TASK_STATE_BUSY; + cont = 1; + break; + + default: + pr_warn("bad state = %d in rxe_do_task\n", + task->state); + } + spin_unlock_irqrestore(&task->state_lock, flags); + } while (cont); + + task->ret = ret; +} + +int rxe_init_task(void *obj, struct rxe_task *task, + void *arg, int (*func)(void *), char *name) +{ + task->obj = obj; + task->arg = arg; + task->func = func; + snprintf(task->name, sizeof(task->name), "%s", name); + + tasklet_init(&task->tasklet, rxe_do_task, (unsigned long)task); + + task->state = TASK_STATE_START; + spin_lock_init(&task->state_lock); + + return 0; +} + +void rxe_cleanup_task(struct rxe_task *task) +{ + tasklet_kill(&task->tasklet); +} + +void rxe_run_task(struct rxe_task *task, int sched) +{ + if (sched) + tasklet_schedule(&task->tasklet); + else + rxe_do_task((unsigned long)task); +} + +void rxe_disable_task(struct rxe_task *task) +{ + tasklet_disable(&task->tasklet); +} + +void rxe_enable_task(struct rxe_task *task) +{ + tasklet_enable(&task->tasklet); +} diff --git a/drivers/infiniband/sw/rxe/rxe_task.h b/drivers/infiniband/sw/rxe/rxe_task.h new file mode 100644 index 000000000000..d14aa6daed05 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_task.h @@ -0,0 +1,95 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_TASK_H +#define RXE_TASK_H + +enum { + TASK_STATE_START = 0, + TASK_STATE_BUSY = 1, + TASK_STATE_ARMED = 2, +}; + +/* + * data structure to describe a 'task' which is a short + * function that returns 0 as long as it needs to be + * called again. + */ +struct rxe_task { + void *obj; + struct tasklet_struct tasklet; + int state; + spinlock_t state_lock; /* spinlock for task state */ + void *arg; + int (*func)(void *arg); + int ret; + char name[16]; +}; + +/* + * init rxe_task structure + * arg => parameter to pass to fcn + * fcn => function to call until it returns != 0 + */ +int rxe_init_task(void *obj, struct rxe_task *task, + void *arg, int (*func)(void *), char *name); + +/* cleanup task */ +void rxe_cleanup_task(struct rxe_task *task); + +/* + * raw call to func in loop without any checking + * can call when tasklets are disabled + */ +int __rxe_do_task(struct rxe_task *task); + +/* + * common function called by any of the main tasklets + * If there is any chance that there is additional + * work to do someone must reschedule the task before + * leaving + */ +void rxe_do_task(unsigned long data); + +/* run a task, else schedule it to run as a tasklet, The decision + * to run or schedule tasklet is based on the parameter sched. + */ +void rxe_run_task(struct rxe_task *task, int sched); + +/* keep a task from scheduling */ +void rxe_disable_task(struct rxe_task *task); + +/* allow task to run */ +void rxe_enable_task(struct rxe_task *task); + +#endif /* RXE_TASK_H */ diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c new file mode 100644 index 000000000000..4552be960c6a --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -0,0 +1,1330 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "rxe.h" +#include "rxe_loc.h" +#include "rxe_queue.h" + +static int rxe_query_device(struct ib_device *dev, + struct ib_device_attr *attr, + struct ib_udata *uhw) +{ + struct rxe_dev *rxe = to_rdev(dev); + + if (uhw->inlen || uhw->outlen) + return -EINVAL; + + *attr = rxe->attr; + return 0; +} + +static void rxe_eth_speed_to_ib_speed(int speed, u8 *active_speed, + u8 *active_width) +{ + if (speed <= 1000) { + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_SDR; + } else if (speed <= 10000) { + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_FDR10; + } else if (speed <= 20000) { + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_DDR; + } else if (speed <= 30000) { + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_QDR; + } else if (speed <= 40000) { + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_FDR10; + } else { + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_EDR; + } +} + +static int rxe_query_port(struct ib_device *dev, + u8 port_num, struct ib_port_attr *attr) +{ + struct rxe_dev *rxe = to_rdev(dev); + struct rxe_port *port; + u32 speed; + + if (unlikely(port_num != 1)) { + pr_warn("invalid port_number %d\n", port_num); + goto err1; + } + + port = &rxe->port; + + *attr = port->attr; + + mutex_lock(&rxe->usdev_lock); + if (rxe->ndev->ethtool_ops->get_link_ksettings) { + struct ethtool_link_ksettings ks; + + rxe->ndev->ethtool_ops->get_link_ksettings(rxe->ndev, &ks); + speed = ks.base.speed; + } else if (rxe->ndev->ethtool_ops->get_settings) { + struct ethtool_cmd cmd; + + rxe->ndev->ethtool_ops->get_settings(rxe->ndev, &cmd); + speed = cmd.speed; + } else { + pr_warn("%s speed is unknown, defaulting to 1000\n", rxe->ndev->name); + speed = 1000; + } + rxe_eth_speed_to_ib_speed(speed, &attr->active_speed, &attr->active_width); + mutex_unlock(&rxe->usdev_lock); + + return 0; + +err1: + return -EINVAL; +} + +static int rxe_query_gid(struct ib_device *device, + u8 port_num, int index, union ib_gid *gid) +{ + int ret; + + if (index > RXE_PORT_GID_TBL_LEN) + return -EINVAL; + + ret = ib_get_cached_gid(device, port_num, index, gid, NULL); + if (ret == -EAGAIN) { + memcpy(gid, &zgid, sizeof(*gid)); + return 0; + } + + return ret; +} + +static int rxe_add_gid(struct ib_device *device, u8 port_num, unsigned int + index, const union ib_gid *gid, + const struct ib_gid_attr *attr, void **context) +{ + if (index >= RXE_PORT_GID_TBL_LEN) + return -EINVAL; + return 0; +} + +static int rxe_del_gid(struct ib_device *device, u8 port_num, unsigned int + index, void **context) +{ + if (index >= RXE_PORT_GID_TBL_LEN) + return -EINVAL; + return 0; +} + +static struct net_device *rxe_get_netdev(struct ib_device *device, + u8 port_num) +{ + struct rxe_dev *rxe = to_rdev(device); + + if (rxe->ndev) { + dev_hold(rxe->ndev); + return rxe->ndev; + } + + return NULL; +} + +static int rxe_query_pkey(struct ib_device *device, + u8 port_num, u16 index, u16 *pkey) +{ + struct rxe_dev *rxe = to_rdev(device); + struct rxe_port *port; + + if (unlikely(port_num != 1)) { + dev_warn(device->dma_device, "invalid port_num = %d\n", + port_num); + goto err1; + } + + port = &rxe->port; + + if (unlikely(index >= port->attr.pkey_tbl_len)) { + dev_warn(device->dma_device, "invalid index = %d\n", + index); + goto err1; + } + + *pkey = port->pkey_tbl[index]; + return 0; + +err1: + return -EINVAL; +} + +static int rxe_modify_device(struct ib_device *dev, + int mask, struct ib_device_modify *attr) +{ + struct rxe_dev *rxe = to_rdev(dev); + + if (mask & IB_DEVICE_MODIFY_SYS_IMAGE_GUID) + rxe->attr.sys_image_guid = cpu_to_be64(attr->sys_image_guid); + + if (mask & IB_DEVICE_MODIFY_NODE_DESC) { + memcpy(rxe->ib_dev.node_desc, + attr->node_desc, sizeof(rxe->ib_dev.node_desc)); + } + + return 0; +} + +static int rxe_modify_port(struct ib_device *dev, + u8 port_num, int mask, struct ib_port_modify *attr) +{ + struct rxe_dev *rxe = to_rdev(dev); + struct rxe_port *port; + + if (unlikely(port_num != 1)) { + pr_warn("invalid port_num = %d\n", port_num); + goto err1; + } + + port = &rxe->port; + + port->attr.port_cap_flags |= attr->set_port_cap_mask; + port->attr.port_cap_flags &= ~attr->clr_port_cap_mask; + + if (mask & IB_PORT_RESET_QKEY_CNTR) + port->attr.qkey_viol_cntr = 0; + + return 0; + +err1: + return -EINVAL; +} + +static enum rdma_link_layer rxe_get_link_layer(struct ib_device *dev, + u8 port_num) +{ + struct rxe_dev *rxe = to_rdev(dev); + + return rxe->ifc_ops->link_layer(rxe, port_num); +} + +static struct ib_ucontext *rxe_alloc_ucontext(struct ib_device *dev, + struct ib_udata *udata) +{ + struct rxe_dev *rxe = to_rdev(dev); + struct rxe_ucontext *uc; + + uc = rxe_alloc(&rxe->uc_pool); + return uc ? &uc->ibuc : ERR_PTR(-ENOMEM); +} + +static int rxe_dealloc_ucontext(struct ib_ucontext *ibuc) +{ + struct rxe_ucontext *uc = to_ruc(ibuc); + + rxe_drop_ref(uc); + return 0; +} + +static int rxe_port_immutable(struct ib_device *dev, u8 port_num, + struct ib_port_immutable *immutable) +{ + int err; + struct ib_port_attr attr; + + err = rxe_query_port(dev, port_num, &attr); + if (err) + return err; + + immutable->pkey_tbl_len = attr.pkey_tbl_len; + immutable->gid_tbl_len = attr.gid_tbl_len; + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; + immutable->max_mad_size = IB_MGMT_MAD_SIZE; + + return 0; +} + +static struct ib_pd *rxe_alloc_pd(struct ib_device *dev, + struct ib_ucontext *context, + struct ib_udata *udata) +{ + struct rxe_dev *rxe = to_rdev(dev); + struct rxe_pd *pd; + + pd = rxe_alloc(&rxe->pd_pool); + return pd ? &pd->ibpd : ERR_PTR(-ENOMEM); +} + +static int rxe_dealloc_pd(struct ib_pd *ibpd) +{ + struct rxe_pd *pd = to_rpd(ibpd); + + rxe_drop_ref(pd); + return 0; +} + +static int rxe_init_av(struct rxe_dev *rxe, struct ib_ah_attr *attr, + struct rxe_av *av) +{ + int err; + union ib_gid sgid; + struct ib_gid_attr sgid_attr; + + err = ib_get_cached_gid(&rxe->ib_dev, attr->port_num, + attr->grh.sgid_index, &sgid, + &sgid_attr); + if (err) { + pr_err("Failed to query sgid. err = %d\n", err); + return err; + } + + err = rxe_av_from_attr(rxe, attr->port_num, av, attr); + if (!err) + err = rxe_av_fill_ip_info(rxe, av, attr, &sgid_attr, &sgid); + + if (sgid_attr.ndev) + dev_put(sgid_attr.ndev); + return err; +} + +static struct ib_ah *rxe_create_ah(struct ib_pd *ibpd, struct ib_ah_attr *attr) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibpd->device); + struct rxe_pd *pd = to_rpd(ibpd); + struct rxe_ah *ah; + + err = rxe_av_chk_attr(rxe, attr); + if (err) + goto err1; + + ah = rxe_alloc(&rxe->ah_pool); + if (!ah) { + err = -ENOMEM; + goto err1; + } + + rxe_add_ref(pd); + ah->pd = pd; + + err = rxe_init_av(rxe, attr, &ah->av); + if (err) + goto err2; + + return &ah->ibah; + +err2: + rxe_drop_ref(pd); + rxe_drop_ref(ah); +err1: + return ERR_PTR(err); +} + +static int rxe_modify_ah(struct ib_ah *ibah, struct ib_ah_attr *attr) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibah->device); + struct rxe_ah *ah = to_rah(ibah); + + err = rxe_av_chk_attr(rxe, attr); + if (err) + return err; + + err = rxe_init_av(rxe, attr, &ah->av); + if (err) + return err; + + return 0; +} + +static int rxe_query_ah(struct ib_ah *ibah, struct ib_ah_attr *attr) +{ + struct rxe_dev *rxe = to_rdev(ibah->device); + struct rxe_ah *ah = to_rah(ibah); + + rxe_av_to_attr(rxe, &ah->av, attr); + return 0; +} + +static int rxe_destroy_ah(struct ib_ah *ibah) +{ + struct rxe_ah *ah = to_rah(ibah); + + rxe_drop_ref(ah->pd); + rxe_drop_ref(ah); + return 0; +} + +static int post_one_recv(struct rxe_rq *rq, struct ib_recv_wr *ibwr) +{ + int err; + int i; + u32 length; + struct rxe_recv_wqe *recv_wqe; + int num_sge = ibwr->num_sge; + + if (unlikely(queue_full(rq->queue))) { + err = -ENOMEM; + goto err1; + } + + if (unlikely(num_sge > rq->max_sge)) { + err = -EINVAL; + goto err1; + } + + length = 0; + for (i = 0; i < num_sge; i++) + length += ibwr->sg_list[i].length; + + recv_wqe = producer_addr(rq->queue); + recv_wqe->wr_id = ibwr->wr_id; + recv_wqe->num_sge = num_sge; + + memcpy(recv_wqe->dma.sge, ibwr->sg_list, + num_sge * sizeof(struct ib_sge)); + + recv_wqe->dma.length = length; + recv_wqe->dma.resid = length; + recv_wqe->dma.num_sge = num_sge; + recv_wqe->dma.cur_sge = 0; + recv_wqe->dma.sge_offset = 0; + + /* make sure all changes to the work queue are written before we + * update the producer pointer + */ + smp_wmb(); + + advance_producer(rq->queue); + return 0; + +err1: + return err; +} + +static struct ib_srq *rxe_create_srq(struct ib_pd *ibpd, + struct ib_srq_init_attr *init, + struct ib_udata *udata) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibpd->device); + struct rxe_pd *pd = to_rpd(ibpd); + struct rxe_srq *srq; + struct ib_ucontext *context = udata ? ibpd->uobject->context : NULL; + + err = rxe_srq_chk_attr(rxe, NULL, &init->attr, IB_SRQ_INIT_MASK); + if (err) + goto err1; + + srq = rxe_alloc(&rxe->srq_pool); + if (!srq) { + err = -ENOMEM; + goto err1; + } + + rxe_add_index(srq); + rxe_add_ref(pd); + srq->pd = pd; + + err = rxe_srq_from_init(rxe, srq, init, context, udata); + if (err) + goto err2; + + return &srq->ibsrq; + +err2: + rxe_drop_ref(pd); + rxe_drop_index(srq); + rxe_drop_ref(srq); +err1: + return ERR_PTR(err); +} + +static int rxe_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr, + enum ib_srq_attr_mask mask, + struct ib_udata *udata) +{ + int err; + struct rxe_srq *srq = to_rsrq(ibsrq); + struct rxe_dev *rxe = to_rdev(ibsrq->device); + + err = rxe_srq_chk_attr(rxe, srq, attr, mask); + if (err) + goto err1; + + err = rxe_srq_from_attr(rxe, srq, attr, mask, udata); + if (err) + goto err1; + + return 0; + +err1: + return err; +} + +static int rxe_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr) +{ + struct rxe_srq *srq = to_rsrq(ibsrq); + + if (srq->error) + return -EINVAL; + + attr->max_wr = srq->rq.queue->buf->index_mask; + attr->max_sge = srq->rq.max_sge; + attr->srq_limit = srq->limit; + return 0; +} + +static int rxe_destroy_srq(struct ib_srq *ibsrq) +{ + struct rxe_srq *srq = to_rsrq(ibsrq); + + if (srq->rq.queue) + rxe_queue_cleanup(srq->rq.queue); + + rxe_drop_ref(srq->pd); + rxe_drop_index(srq); + rxe_drop_ref(srq); + + return 0; +} + +static int rxe_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, + struct ib_recv_wr **bad_wr) +{ + int err = 0; + unsigned long flags; + struct rxe_srq *srq = to_rsrq(ibsrq); + + spin_lock_irqsave(&srq->rq.producer_lock, flags); + + while (wr) { + err = post_one_recv(&srq->rq, wr); + if (unlikely(err)) + break; + wr = wr->next; + } + + spin_unlock_irqrestore(&srq->rq.producer_lock, flags); + + if (err) + *bad_wr = wr; + + return err; +} + +static struct ib_qp *rxe_create_qp(struct ib_pd *ibpd, + struct ib_qp_init_attr *init, + struct ib_udata *udata) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibpd->device); + struct rxe_pd *pd = to_rpd(ibpd); + struct rxe_qp *qp; + + err = rxe_qp_chk_init(rxe, init); + if (err) + goto err1; + + qp = rxe_alloc(&rxe->qp_pool); + if (!qp) { + err = -ENOMEM; + goto err1; + } + + if (udata) { + if (udata->inlen) { + err = -EINVAL; + goto err1; + } + qp->is_user = 1; + } + + rxe_add_index(qp); + + err = rxe_qp_from_init(rxe, qp, pd, init, udata, ibpd); + if (err) + goto err2; + + return &qp->ibqp; + +err2: + rxe_drop_index(qp); + rxe_drop_ref(qp); +err1: + return ERR_PTR(err); +} + +static int rxe_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + int mask, struct ib_udata *udata) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibqp->device); + struct rxe_qp *qp = to_rqp(ibqp); + + err = rxe_qp_chk_attr(rxe, qp, attr, mask); + if (err) + goto err1; + + err = rxe_qp_from_attr(qp, attr, mask, udata); + if (err) + goto err1; + + return 0; + +err1: + return err; +} + +static int rxe_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + int mask, struct ib_qp_init_attr *init) +{ + struct rxe_qp *qp = to_rqp(ibqp); + + rxe_qp_to_init(qp, init); + rxe_qp_to_attr(qp, attr, mask); + + return 0; +} + +static int rxe_destroy_qp(struct ib_qp *ibqp) +{ + struct rxe_qp *qp = to_rqp(ibqp); + + rxe_qp_destroy(qp); + rxe_drop_index(qp); + rxe_drop_ref(qp); + return 0; +} + +static int validate_send_wr(struct rxe_qp *qp, struct ib_send_wr *ibwr, + unsigned int mask, unsigned int length) +{ + int num_sge = ibwr->num_sge; + struct rxe_sq *sq = &qp->sq; + + if (unlikely(num_sge > sq->max_sge)) + goto err1; + + if (unlikely(mask & WR_ATOMIC_MASK)) { + if (length < 8) + goto err1; + + if (atomic_wr(ibwr)->remote_addr & 0x7) + goto err1; + } + + if (unlikely((ibwr->send_flags & IB_SEND_INLINE) && + (length > sq->max_inline))) + goto err1; + + return 0; + +err1: + return -EINVAL; +} + +static void init_send_wr(struct rxe_qp *qp, struct rxe_send_wr *wr, + struct ib_send_wr *ibwr) +{ + wr->wr_id = ibwr->wr_id; + wr->num_sge = ibwr->num_sge; + wr->opcode = ibwr->opcode; + wr->send_flags = ibwr->send_flags; + + if (qp_type(qp) == IB_QPT_UD || + qp_type(qp) == IB_QPT_SMI || + qp_type(qp) == IB_QPT_GSI) { + wr->wr.ud.remote_qpn = ud_wr(ibwr)->remote_qpn; + wr->wr.ud.remote_qkey = ud_wr(ibwr)->remote_qkey; + if (qp_type(qp) == IB_QPT_GSI) + wr->wr.ud.pkey_index = ud_wr(ibwr)->pkey_index; + if (wr->opcode == IB_WR_SEND_WITH_IMM) + wr->ex.imm_data = ibwr->ex.imm_data; + } else { + switch (wr->opcode) { + case IB_WR_RDMA_WRITE_WITH_IMM: + wr->ex.imm_data = ibwr->ex.imm_data; + case IB_WR_RDMA_READ: + case IB_WR_RDMA_WRITE: + wr->wr.rdma.remote_addr = rdma_wr(ibwr)->remote_addr; + wr->wr.rdma.rkey = rdma_wr(ibwr)->rkey; + break; + case IB_WR_SEND_WITH_IMM: + wr->ex.imm_data = ibwr->ex.imm_data; + break; + case IB_WR_SEND_WITH_INV: + wr->ex.invalidate_rkey = ibwr->ex.invalidate_rkey; + break; + case IB_WR_ATOMIC_CMP_AND_SWP: + case IB_WR_ATOMIC_FETCH_AND_ADD: + wr->wr.atomic.remote_addr = + atomic_wr(ibwr)->remote_addr; + wr->wr.atomic.compare_add = + atomic_wr(ibwr)->compare_add; + wr->wr.atomic.swap = atomic_wr(ibwr)->swap; + wr->wr.atomic.rkey = atomic_wr(ibwr)->rkey; + break; + case IB_WR_LOCAL_INV: + wr->ex.invalidate_rkey = ibwr->ex.invalidate_rkey; + break; + case IB_WR_REG_MR: + wr->wr.reg.mr = reg_wr(ibwr)->mr; + wr->wr.reg.key = reg_wr(ibwr)->key; + wr->wr.reg.access = reg_wr(ibwr)->access; + break; + default: + break; + } + } +} + +static int init_send_wqe(struct rxe_qp *qp, struct ib_send_wr *ibwr, + unsigned int mask, unsigned int length, + struct rxe_send_wqe *wqe) +{ + int num_sge = ibwr->num_sge; + struct ib_sge *sge; + int i; + u8 *p; + + init_send_wr(qp, &wqe->wr, ibwr); + + if (qp_type(qp) == IB_QPT_UD || + qp_type(qp) == IB_QPT_SMI || + qp_type(qp) == IB_QPT_GSI) + memcpy(&wqe->av, &to_rah(ud_wr(ibwr)->ah)->av, sizeof(wqe->av)); + + if (unlikely(ibwr->send_flags & IB_SEND_INLINE)) { + p = wqe->dma.inline_data; + + sge = ibwr->sg_list; + for (i = 0; i < num_sge; i++, sge++) { + if (qp->is_user && copy_from_user(p, (__user void *) + (uintptr_t)sge->addr, sge->length)) + return -EFAULT; + + else if (!qp->is_user) + memcpy(p, (void *)(uintptr_t)sge->addr, + sge->length); + + p += sge->length; + } + } else if (mask & WR_REG_MASK) { + wqe->mask = mask; + wqe->state = wqe_state_posted; + return 0; + } else + memcpy(wqe->dma.sge, ibwr->sg_list, + num_sge * sizeof(struct ib_sge)); + + wqe->iova = (mask & WR_ATOMIC_MASK) ? + atomic_wr(ibwr)->remote_addr : + rdma_wr(ibwr)->remote_addr; + wqe->mask = mask; + wqe->dma.length = length; + wqe->dma.resid = length; + wqe->dma.num_sge = num_sge; + wqe->dma.cur_sge = 0; + wqe->dma.sge_offset = 0; + wqe->state = wqe_state_posted; + wqe->ssn = atomic_add_return(1, &qp->ssn); + + return 0; +} + +static int post_one_send(struct rxe_qp *qp, struct ib_send_wr *ibwr, + unsigned mask, u32 length) +{ + int err; + struct rxe_sq *sq = &qp->sq; + struct rxe_send_wqe *send_wqe; + unsigned long flags; + + err = validate_send_wr(qp, ibwr, mask, length); + if (err) + return err; + + spin_lock_irqsave(&qp->sq.sq_lock, flags); + + if (unlikely(queue_full(sq->queue))) { + err = -ENOMEM; + goto err1; + } + + send_wqe = producer_addr(sq->queue); + + err = init_send_wqe(qp, ibwr, mask, length, send_wqe); + if (unlikely(err)) + goto err1; + + /* + * make sure all changes to the work queue are + * written before we update the producer pointer + */ + smp_wmb(); + + advance_producer(sq->queue); + spin_unlock_irqrestore(&qp->sq.sq_lock, flags); + + return 0; + +err1: + spin_unlock_irqrestore(&qp->sq.sq_lock, flags); + return err; +} + +static int rxe_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, + struct ib_send_wr **bad_wr) +{ + int err = 0; + struct rxe_qp *qp = to_rqp(ibqp); + unsigned int mask; + unsigned int length = 0; + int i; + int must_sched; + + if (unlikely(!qp->valid)) { + *bad_wr = wr; + return -EINVAL; + } + + if (unlikely(qp->req.state < QP_STATE_READY)) { + *bad_wr = wr; + return -EINVAL; + } + + while (wr) { + mask = wr_opcode_mask(wr->opcode, qp); + if (unlikely(!mask)) { + err = -EINVAL; + *bad_wr = wr; + break; + } + + if (unlikely((wr->send_flags & IB_SEND_INLINE) && + !(mask & WR_INLINE_MASK))) { + err = -EINVAL; + *bad_wr = wr; + break; + } + + length = 0; + for (i = 0; i < wr->num_sge; i++) + length += wr->sg_list[i].length; + + err = post_one_send(qp, wr, mask, length); + + if (err) { + *bad_wr = wr; + break; + } + wr = wr->next; + } + + /* + * Must sched in case of GSI QP because ib_send_mad() hold irq lock, + * and the requester call ip_local_out_sk() that takes spin_lock_bh. + */ + must_sched = (qp_type(qp) == IB_QPT_GSI) || + (queue_count(qp->sq.queue) > 1); + + rxe_run_task(&qp->req.task, must_sched); + + return err; +} + +static int rxe_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, + struct ib_recv_wr **bad_wr) +{ + int err = 0; + struct rxe_qp *qp = to_rqp(ibqp); + struct rxe_rq *rq = &qp->rq; + unsigned long flags; + + if (unlikely((qp_state(qp) < IB_QPS_INIT) || !qp->valid)) { + *bad_wr = wr; + err = -EINVAL; + goto err1; + } + + if (unlikely(qp->srq)) { + *bad_wr = wr; + err = -EINVAL; + goto err1; + } + + spin_lock_irqsave(&rq->producer_lock, flags); + + while (wr) { + err = post_one_recv(rq, wr); + if (unlikely(err)) { + *bad_wr = wr; + break; + } + wr = wr->next; + } + + spin_unlock_irqrestore(&rq->producer_lock, flags); + +err1: + return err; +} + +static struct ib_cq *rxe_create_cq(struct ib_device *dev, + const struct ib_cq_init_attr *attr, + struct ib_ucontext *context, + struct ib_udata *udata) +{ + int err; + struct rxe_dev *rxe = to_rdev(dev); + struct rxe_cq *cq; + + if (attr->flags) + return ERR_PTR(-EINVAL); + + err = rxe_cq_chk_attr(rxe, NULL, attr->cqe, attr->comp_vector, udata); + if (err) + goto err1; + + cq = rxe_alloc(&rxe->cq_pool); + if (!cq) { + err = -ENOMEM; + goto err1; + } + + err = rxe_cq_from_init(rxe, cq, attr->cqe, attr->comp_vector, + context, udata); + if (err) + goto err2; + + return &cq->ibcq; + +err2: + rxe_drop_ref(cq); +err1: + return ERR_PTR(err); +} + +static int rxe_destroy_cq(struct ib_cq *ibcq) +{ + struct rxe_cq *cq = to_rcq(ibcq); + + rxe_drop_ref(cq); + return 0; +} + +static int rxe_resize_cq(struct ib_cq *ibcq, int cqe, struct ib_udata *udata) +{ + int err; + struct rxe_cq *cq = to_rcq(ibcq); + struct rxe_dev *rxe = to_rdev(ibcq->device); + + err = rxe_cq_chk_attr(rxe, cq, cqe, 0, udata); + if (err) + goto err1; + + err = rxe_cq_resize_queue(cq, cqe, udata); + if (err) + goto err1; + + return 0; + +err1: + return err; +} + +static int rxe_poll_cq(struct ib_cq *ibcq, int num_entries, struct ib_wc *wc) +{ + int i; + struct rxe_cq *cq = to_rcq(ibcq); + struct rxe_cqe *cqe; + unsigned long flags; + + spin_lock_irqsave(&cq->cq_lock, flags); + for (i = 0; i < num_entries; i++) { + cqe = queue_head(cq->queue); + if (!cqe) + break; + + memcpy(wc++, &cqe->ibwc, sizeof(*wc)); + advance_consumer(cq->queue); + } + spin_unlock_irqrestore(&cq->cq_lock, flags); + + return i; +} + +static int rxe_peek_cq(struct ib_cq *ibcq, int wc_cnt) +{ + struct rxe_cq *cq = to_rcq(ibcq); + int count = queue_count(cq->queue); + + return (count > wc_cnt) ? wc_cnt : count; +} + +static int rxe_req_notify_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags flags) +{ + struct rxe_cq *cq = to_rcq(ibcq); + + if (cq->notify != IB_CQ_NEXT_COMP) + cq->notify = flags & IB_CQ_SOLICITED_MASK; + + return 0; +} + +static struct ib_mr *rxe_get_dma_mr(struct ib_pd *ibpd, int access) +{ + struct rxe_dev *rxe = to_rdev(ibpd->device); + struct rxe_pd *pd = to_rpd(ibpd); + struct rxe_mem *mr; + int err; + + mr = rxe_alloc(&rxe->mr_pool); + if (!mr) { + err = -ENOMEM; + goto err1; + } + + rxe_add_index(mr); + + rxe_add_ref(pd); + + err = rxe_mem_init_dma(rxe, pd, access, mr); + if (err) + goto err2; + + return &mr->ibmr; + +err2: + rxe_drop_ref(pd); + rxe_drop_index(mr); + rxe_drop_ref(mr); +err1: + return ERR_PTR(err); +} + +static struct ib_mr *rxe_reg_user_mr(struct ib_pd *ibpd, + u64 start, + u64 length, + u64 iova, + int access, struct ib_udata *udata) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibpd->device); + struct rxe_pd *pd = to_rpd(ibpd); + struct rxe_mem *mr; + + mr = rxe_alloc(&rxe->mr_pool); + if (!mr) { + err = -ENOMEM; + goto err2; + } + + rxe_add_index(mr); + + rxe_add_ref(pd); + + err = rxe_mem_init_user(rxe, pd, start, length, iova, + access, udata, mr); + if (err) + goto err3; + + return &mr->ibmr; + +err3: + rxe_drop_ref(pd); + rxe_drop_index(mr); + rxe_drop_ref(mr); +err2: + return ERR_PTR(err); +} + +static int rxe_dereg_mr(struct ib_mr *ibmr) +{ + struct rxe_mem *mr = to_rmr(ibmr); + + mr->state = RXE_MEM_STATE_ZOMBIE; + rxe_drop_ref(mr->pd); + rxe_drop_index(mr); + rxe_drop_ref(mr); + return 0; +} + +static struct ib_mr *rxe_alloc_mr(struct ib_pd *ibpd, + enum ib_mr_type mr_type, + u32 max_num_sg) +{ + struct rxe_dev *rxe = to_rdev(ibpd->device); + struct rxe_pd *pd = to_rpd(ibpd); + struct rxe_mem *mr; + int err; + + if (mr_type != IB_MR_TYPE_MEM_REG) + return ERR_PTR(-EINVAL); + + mr = rxe_alloc(&rxe->mr_pool); + if (!mr) { + err = -ENOMEM; + goto err1; + } + + rxe_add_index(mr); + + rxe_add_ref(pd); + + err = rxe_mem_init_fast(rxe, pd, max_num_sg, mr); + if (err) + goto err2; + + return &mr->ibmr; + +err2: + rxe_drop_ref(pd); + rxe_drop_index(mr); + rxe_drop_ref(mr); +err1: + return ERR_PTR(err); +} + +static int rxe_set_page(struct ib_mr *ibmr, u64 addr) +{ + struct rxe_mem *mr = to_rmr(ibmr); + struct rxe_map *map; + struct rxe_phys_buf *buf; + + if (unlikely(mr->nbuf == mr->num_buf)) + return -ENOMEM; + + map = mr->map[mr->nbuf / RXE_BUF_PER_MAP]; + buf = &map->buf[mr->nbuf % RXE_BUF_PER_MAP]; + + buf->addr = addr; + buf->size = ibmr->page_size; + mr->nbuf++; + + return 0; +} + +static int rxe_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents, + unsigned int *sg_offset) +{ + struct rxe_mem *mr = to_rmr(ibmr); + int n; + + mr->nbuf = 0; + + n = ib_sg_to_pages(ibmr, sg, sg_nents, sg_offset, rxe_set_page); + + mr->va = ibmr->iova; + mr->iova = ibmr->iova; + mr->length = ibmr->length; + mr->page_shift = ilog2(ibmr->page_size); + mr->page_mask = ibmr->page_size - 1; + mr->offset = mr->iova & mr->page_mask; + + return n; +} + +static int rxe_attach_mcast(struct ib_qp *ibqp, union ib_gid *mgid, u16 mlid) +{ + int err; + struct rxe_dev *rxe = to_rdev(ibqp->device); + struct rxe_qp *qp = to_rqp(ibqp); + struct rxe_mc_grp *grp; + + /* takes a ref on grp if successful */ + err = rxe_mcast_get_grp(rxe, mgid, &grp); + if (err) + return err; + + err = rxe_mcast_add_grp_elem(rxe, qp, grp); + + rxe_drop_ref(grp); + return err; +} + +static int rxe_detach_mcast(struct ib_qp *ibqp, union ib_gid *mgid, u16 mlid) +{ + struct rxe_dev *rxe = to_rdev(ibqp->device); + struct rxe_qp *qp = to_rqp(ibqp); + + return rxe_mcast_drop_grp_elem(rxe, qp, mgid); +} + +static ssize_t rxe_show_parent(struct device *device, + struct device_attribute *attr, char *buf) +{ + struct rxe_dev *rxe = container_of(device, struct rxe_dev, + ib_dev.dev); + char *name; + + name = rxe->ifc_ops->parent_name(rxe, 1); + return snprintf(buf, 16, "%s\n", name); +} + +static DEVICE_ATTR(parent, S_IRUGO, rxe_show_parent, NULL); + +static struct device_attribute *rxe_dev_attributes[] = { + &dev_attr_parent, +}; + +int rxe_register_device(struct rxe_dev *rxe) +{ + int err; + int i; + struct ib_device *dev = &rxe->ib_dev; + + strlcpy(dev->name, "rxe%d", IB_DEVICE_NAME_MAX); + strlcpy(dev->node_desc, "rxe", sizeof(dev->node_desc)); + + dev->owner = THIS_MODULE; + dev->node_type = RDMA_NODE_IB_CA; + dev->phys_port_cnt = 1; + dev->num_comp_vectors = RXE_NUM_COMP_VECTORS; + dev->dma_device = rxe->ifc_ops->dma_device(rxe); + dev->local_dma_lkey = 0; + dev->node_guid = rxe->ifc_ops->node_guid(rxe); + dev->dma_ops = &rxe_dma_mapping_ops; + + dev->uverbs_abi_ver = RXE_UVERBS_ABI_VERSION; + dev->uverbs_cmd_mask = BIT_ULL(IB_USER_VERBS_CMD_GET_CONTEXT) + | BIT_ULL(IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL) + | BIT_ULL(IB_USER_VERBS_CMD_QUERY_DEVICE) + | BIT_ULL(IB_USER_VERBS_CMD_QUERY_PORT) + | BIT_ULL(IB_USER_VERBS_CMD_ALLOC_PD) + | BIT_ULL(IB_USER_VERBS_CMD_DEALLOC_PD) + | BIT_ULL(IB_USER_VERBS_CMD_CREATE_SRQ) + | BIT_ULL(IB_USER_VERBS_CMD_MODIFY_SRQ) + | BIT_ULL(IB_USER_VERBS_CMD_QUERY_SRQ) + | BIT_ULL(IB_USER_VERBS_CMD_DESTROY_SRQ) + | BIT_ULL(IB_USER_VERBS_CMD_POST_SRQ_RECV) + | BIT_ULL(IB_USER_VERBS_CMD_CREATE_QP) + | BIT_ULL(IB_USER_VERBS_CMD_MODIFY_QP) + | BIT_ULL(IB_USER_VERBS_CMD_QUERY_QP) + | BIT_ULL(IB_USER_VERBS_CMD_DESTROY_QP) + | BIT_ULL(IB_USER_VERBS_CMD_POST_SEND) + | BIT_ULL(IB_USER_VERBS_CMD_POST_RECV) + | BIT_ULL(IB_USER_VERBS_CMD_CREATE_CQ) + | BIT_ULL(IB_USER_VERBS_CMD_RESIZE_CQ) + | BIT_ULL(IB_USER_VERBS_CMD_DESTROY_CQ) + | BIT_ULL(IB_USER_VERBS_CMD_POLL_CQ) + | BIT_ULL(IB_USER_VERBS_CMD_PEEK_CQ) + | BIT_ULL(IB_USER_VERBS_CMD_REQ_NOTIFY_CQ) + | BIT_ULL(IB_USER_VERBS_CMD_REG_MR) + | BIT_ULL(IB_USER_VERBS_CMD_DEREG_MR) + | BIT_ULL(IB_USER_VERBS_CMD_CREATE_AH) + | BIT_ULL(IB_USER_VERBS_CMD_MODIFY_AH) + | BIT_ULL(IB_USER_VERBS_CMD_QUERY_AH) + | BIT_ULL(IB_USER_VERBS_CMD_DESTROY_AH) + | BIT_ULL(IB_USER_VERBS_CMD_ATTACH_MCAST) + | BIT_ULL(IB_USER_VERBS_CMD_DETACH_MCAST) + ; + + dev->query_device = rxe_query_device; + dev->modify_device = rxe_modify_device; + dev->query_port = rxe_query_port; + dev->modify_port = rxe_modify_port; + dev->get_link_layer = rxe_get_link_layer; + dev->query_gid = rxe_query_gid; + dev->get_netdev = rxe_get_netdev; + dev->add_gid = rxe_add_gid; + dev->del_gid = rxe_del_gid; + dev->query_pkey = rxe_query_pkey; + dev->alloc_ucontext = rxe_alloc_ucontext; + dev->dealloc_ucontext = rxe_dealloc_ucontext; + dev->mmap = rxe_mmap; + dev->get_port_immutable = rxe_port_immutable; + dev->alloc_pd = rxe_alloc_pd; + dev->dealloc_pd = rxe_dealloc_pd; + dev->create_ah = rxe_create_ah; + dev->modify_ah = rxe_modify_ah; + dev->query_ah = rxe_query_ah; + dev->destroy_ah = rxe_destroy_ah; + dev->create_srq = rxe_create_srq; + dev->modify_srq = rxe_modify_srq; + dev->query_srq = rxe_query_srq; + dev->destroy_srq = rxe_destroy_srq; + dev->post_srq_recv = rxe_post_srq_recv; + dev->create_qp = rxe_create_qp; + dev->modify_qp = rxe_modify_qp; + dev->query_qp = rxe_query_qp; + dev->destroy_qp = rxe_destroy_qp; + dev->post_send = rxe_post_send; + dev->post_recv = rxe_post_recv; + dev->create_cq = rxe_create_cq; + dev->destroy_cq = rxe_destroy_cq; + dev->resize_cq = rxe_resize_cq; + dev->poll_cq = rxe_poll_cq; + dev->peek_cq = rxe_peek_cq; + dev->req_notify_cq = rxe_req_notify_cq; + dev->get_dma_mr = rxe_get_dma_mr; + dev->reg_user_mr = rxe_reg_user_mr; + dev->dereg_mr = rxe_dereg_mr; + dev->alloc_mr = rxe_alloc_mr; + dev->map_mr_sg = rxe_map_mr_sg; + dev->attach_mcast = rxe_attach_mcast; + dev->detach_mcast = rxe_detach_mcast; + + err = ib_register_device(dev, NULL); + if (err) { + pr_warn("rxe_register_device failed, err = %d\n", err); + goto err1; + } + + for (i = 0; i < ARRAY_SIZE(rxe_dev_attributes); ++i) { + err = device_create_file(&dev->dev, rxe_dev_attributes[i]); + if (err) { + pr_warn("device_create_file failed, i = %d, err = %d\n", + i, err); + goto err2; + } + } + + return 0; + +err2: + ib_unregister_device(dev); +err1: + return err; +} + +int rxe_unregister_device(struct rxe_dev *rxe) +{ + int i; + struct ib_device *dev = &rxe->ib_dev; + + for (i = 0; i < ARRAY_SIZE(rxe_dev_attributes); ++i) + device_remove_file(&dev->dev, rxe_dev_attributes[i]); + + ib_unregister_device(dev); + + return 0; +} diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.h b/drivers/infiniband/sw/rxe/rxe_verbs.h new file mode 100644 index 000000000000..cac1d52a08f0 --- /dev/null +++ b/drivers/infiniband/sw/rxe/rxe_verbs.h @@ -0,0 +1,480 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * Copyright (c) 2015 System Fabric Works, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RXE_VERBS_H +#define RXE_VERBS_H + +#include +#include +#include "rxe_pool.h" +#include "rxe_task.h" + +static inline int pkey_match(u16 key1, u16 key2) +{ + return (((key1 & 0x7fff) != 0) && + ((key1 & 0x7fff) == (key2 & 0x7fff)) && + ((key1 & 0x8000) || (key2 & 0x8000))) ? 1 : 0; +} + +/* Return >0 if psn_a > psn_b + * 0 if psn_a == psn_b + * <0 if psn_a < psn_b + */ +static inline int psn_compare(u32 psn_a, u32 psn_b) +{ + s32 diff; + + diff = (psn_a - psn_b) << 8; + return diff; +} + +struct rxe_ucontext { + struct rxe_pool_entry pelem; + struct ib_ucontext ibuc; +}; + +struct rxe_pd { + struct rxe_pool_entry pelem; + struct ib_pd ibpd; +}; + +struct rxe_ah { + struct rxe_pool_entry pelem; + struct ib_ah ibah; + struct rxe_pd *pd; + struct rxe_av av; +}; + +struct rxe_cqe { + union { + struct ib_wc ibwc; + struct ib_uverbs_wc uibwc; + }; +}; + +struct rxe_cq { + struct rxe_pool_entry pelem; + struct ib_cq ibcq; + struct rxe_queue *queue; + spinlock_t cq_lock; + u8 notify; + int is_user; + struct tasklet_struct comp_task; +}; + +enum wqe_state { + wqe_state_posted, + wqe_state_processing, + wqe_state_pending, + wqe_state_done, + wqe_state_error, +}; + +struct rxe_sq { + int max_wr; + int max_sge; + int max_inline; + spinlock_t sq_lock; /* guard queue */ + struct rxe_queue *queue; +}; + +struct rxe_rq { + int max_wr; + int max_sge; + spinlock_t producer_lock; /* guard queue producer */ + spinlock_t consumer_lock; /* guard queue consumer */ + struct rxe_queue *queue; +}; + +struct rxe_srq { + struct rxe_pool_entry pelem; + struct ib_srq ibsrq; + struct rxe_pd *pd; + struct rxe_rq rq; + u32 srq_num; + + int limit; + int error; +}; + +enum rxe_qp_state { + QP_STATE_RESET, + QP_STATE_INIT, + QP_STATE_READY, + QP_STATE_DRAIN, /* req only */ + QP_STATE_DRAINED, /* req only */ + QP_STATE_ERROR +}; + +extern char *rxe_qp_state_name[]; + +struct rxe_req_info { + enum rxe_qp_state state; + int wqe_index; + u32 psn; + int opcode; + atomic_t rd_atomic; + int wait_fence; + int need_rd_atomic; + int wait_psn; + int need_retry; + int noack_pkts; + struct rxe_task task; +}; + +struct rxe_comp_info { + u32 psn; + int opcode; + int timeout; + int timeout_retry; + u32 retry_cnt; + u32 rnr_retry; + struct rxe_task task; +}; + +enum rdatm_res_state { + rdatm_res_state_next, + rdatm_res_state_new, + rdatm_res_state_replay, +}; + +struct resp_res { + int type; + u32 first_psn; + u32 last_psn; + u32 cur_psn; + enum rdatm_res_state state; + + union { + struct { + struct sk_buff *skb; + } atomic; + struct { + struct rxe_mem *mr; + u64 va_org; + u32 rkey; + u32 length; + u64 va; + u32 resid; + } read; + }; +}; + +struct rxe_resp_info { + enum rxe_qp_state state; + u32 msn; + u32 psn; + int opcode; + int drop_msg; + int goto_error; + int sent_psn_nak; + enum ib_wc_status status; + u8 aeth_syndrome; + + /* Receive only */ + struct rxe_recv_wqe *wqe; + + /* RDMA read / atomic only */ + u64 va; + struct rxe_mem *mr; + u32 resid; + u32 rkey; + u64 atomic_orig; + + /* SRQ only */ + struct { + struct rxe_recv_wqe wqe; + struct ib_sge sge[RXE_MAX_SGE]; + } srq_wqe; + + /* Responder resources. It's a circular list where the oldest + * resource is dropped first. + */ + struct resp_res *resources; + unsigned int res_head; + unsigned int res_tail; + struct resp_res *res; + struct rxe_task task; +}; + +struct rxe_qp { + struct rxe_pool_entry pelem; + struct ib_qp ibqp; + struct ib_qp_attr attr; + unsigned int valid; + unsigned int mtu; + int is_user; + + struct rxe_pd *pd; + struct rxe_srq *srq; + struct rxe_cq *scq; + struct rxe_cq *rcq; + + enum ib_sig_type sq_sig_type; + + struct rxe_sq sq; + struct rxe_rq rq; + + struct socket *sk; + + struct rxe_av pri_av; + struct rxe_av alt_av; + + /* list of mcast groups qp has joined (for cleanup) */ + struct list_head grp_list; + spinlock_t grp_lock; /* guard grp_list */ + + struct sk_buff_head req_pkts; + struct sk_buff_head resp_pkts; + struct sk_buff_head send_pkts; + + struct rxe_req_info req; + struct rxe_comp_info comp; + struct rxe_resp_info resp; + + atomic_t ssn; + atomic_t skb_out; + int need_req_skb; + + /* Timer for retranmitting packet when ACKs have been lost. RC + * only. The requester sets it when it is not already + * started. The responder resets it whenever an ack is + * received. + */ + struct timer_list retrans_timer; + u64 qp_timeout_jiffies; + + /* Timer for handling RNR NAKS. */ + struct timer_list rnr_nak_timer; + + spinlock_t state_lock; /* guard requester and completer */ +}; + +enum rxe_mem_state { + RXE_MEM_STATE_ZOMBIE, + RXE_MEM_STATE_INVALID, + RXE_MEM_STATE_FREE, + RXE_MEM_STATE_VALID, +}; + +enum rxe_mem_type { + RXE_MEM_TYPE_NONE, + RXE_MEM_TYPE_DMA, + RXE_MEM_TYPE_MR, + RXE_MEM_TYPE_FMR, + RXE_MEM_TYPE_MW, +}; + +#define RXE_BUF_PER_MAP (PAGE_SIZE / sizeof(struct rxe_phys_buf)) + +struct rxe_phys_buf { + u64 addr; + u64 size; +}; + +struct rxe_map { + struct rxe_phys_buf buf[RXE_BUF_PER_MAP]; +}; + +struct rxe_mem { + struct rxe_pool_entry pelem; + union { + struct ib_mr ibmr; + struct ib_mw ibmw; + }; + + struct rxe_pd *pd; + struct ib_umem *umem; + + u32 lkey; + u32 rkey; + + enum rxe_mem_state state; + enum rxe_mem_type type; + u64 va; + u64 iova; + size_t length; + u32 offset; + int access; + + int page_shift; + int page_mask; + int map_shift; + int map_mask; + + u32 num_buf; + u32 nbuf; + + u32 max_buf; + u32 num_map; + + struct rxe_map **map; +}; + +struct rxe_mc_grp { + struct rxe_pool_entry pelem; + spinlock_t mcg_lock; /* guard group */ + struct rxe_dev *rxe; + struct list_head qp_list; + union ib_gid mgid; + int num_qp; + u32 qkey; + u16 pkey; +}; + +struct rxe_mc_elem { + struct rxe_pool_entry pelem; + struct list_head qp_list; + struct list_head grp_list; + struct rxe_qp *qp; + struct rxe_mc_grp *grp; +}; + +struct rxe_port { + struct ib_port_attr attr; + u16 *pkey_tbl; + __be64 port_guid; + __be64 subnet_prefix; + spinlock_t port_lock; /* guard port */ + unsigned int mtu_cap; + /* special QPs */ + u32 qp_smi_index; + u32 qp_gsi_index; +}; + +/* callbacks from rdma_rxe to network interface layer */ +struct rxe_ifc_ops { + void (*release)(struct rxe_dev *rxe); + __be64 (*node_guid)(struct rxe_dev *rxe); + __be64 (*port_guid)(struct rxe_dev *rxe); + struct device *(*dma_device)(struct rxe_dev *rxe); + int (*mcast_add)(struct rxe_dev *rxe, union ib_gid *mgid); + int (*mcast_delete)(struct rxe_dev *rxe, union ib_gid *mgid); + int (*prepare)(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb, u32 *crc); + int (*send)(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, + struct sk_buff *skb); + int (*loopback)(struct sk_buff *skb); + struct sk_buff *(*init_packet)(struct rxe_dev *rxe, struct rxe_av *av, + int paylen, struct rxe_pkt_info *pkt); + char *(*parent_name)(struct rxe_dev *rxe, unsigned int port_num); + enum rdma_link_layer (*link_layer)(struct rxe_dev *rxe, + unsigned int port_num); +}; + +struct rxe_dev { + struct ib_device ib_dev; + struct ib_device_attr attr; + int max_ucontext; + int max_inline_data; + struct kref ref_cnt; + struct mutex usdev_lock; + + struct rxe_ifc_ops *ifc_ops; + + struct net_device *ndev; + + int xmit_errors; + + struct rxe_pool uc_pool; + struct rxe_pool pd_pool; + struct rxe_pool ah_pool; + struct rxe_pool srq_pool; + struct rxe_pool qp_pool; + struct rxe_pool cq_pool; + struct rxe_pool mr_pool; + struct rxe_pool mw_pool; + struct rxe_pool mc_grp_pool; + struct rxe_pool mc_elem_pool; + + spinlock_t pending_lock; /* guard pending_mmaps */ + struct list_head pending_mmaps; + + spinlock_t mmap_offset_lock; /* guard mmap_offset */ + int mmap_offset; + + struct rxe_port port; + struct list_head list; +}; + +static inline struct rxe_dev *to_rdev(struct ib_device *dev) +{ + return dev ? container_of(dev, struct rxe_dev, ib_dev) : NULL; +} + +static inline struct rxe_ucontext *to_ruc(struct ib_ucontext *uc) +{ + return uc ? container_of(uc, struct rxe_ucontext, ibuc) : NULL; +} + +static inline struct rxe_pd *to_rpd(struct ib_pd *pd) +{ + return pd ? container_of(pd, struct rxe_pd, ibpd) : NULL; +} + +static inline struct rxe_ah *to_rah(struct ib_ah *ah) +{ + return ah ? container_of(ah, struct rxe_ah, ibah) : NULL; +} + +static inline struct rxe_srq *to_rsrq(struct ib_srq *srq) +{ + return srq ? container_of(srq, struct rxe_srq, ibsrq) : NULL; +} + +static inline struct rxe_qp *to_rqp(struct ib_qp *qp) +{ + return qp ? container_of(qp, struct rxe_qp, ibqp) : NULL; +} + +static inline struct rxe_cq *to_rcq(struct ib_cq *cq) +{ + return cq ? container_of(cq, struct rxe_cq, ibcq) : NULL; +} + +static inline struct rxe_mem *to_rmr(struct ib_mr *mr) +{ + return mr ? container_of(mr, struct rxe_mem, ibmr) : NULL; +} + +static inline struct rxe_mem *to_rmw(struct ib_mw *mw) +{ + return mw ? container_of(mw, struct rxe_mem, ibmw) : NULL; +} + +int rxe_register_device(struct rxe_dev *rxe); +int rxe_unregister_device(struct rxe_dev *rxe); + +void rxe_mc_cleanup(void *arg); + +#endif /* RXE_VERBS_H */ diff --git a/include/uapi/rdma/Kbuild b/include/uapi/rdma/Kbuild index 231901b08f6c..4edb0f2b4f9f 100644 --- a/include/uapi/rdma/Kbuild +++ b/include/uapi/rdma/Kbuild @@ -6,3 +6,4 @@ header-y += ib_user_verbs.h header-y += rdma_netlink.h header-y += rdma_user_cm.h header-y += hfi/ +header-y += rdma_user_rxe.h diff --git a/include/uapi/rdma/rdma_user_rxe.h b/include/uapi/rdma/rdma_user_rxe.h new file mode 100644 index 000000000000..1de99cfdaf7d --- /dev/null +++ b/include/uapi/rdma/rdma_user_rxe.h @@ -0,0 +1,144 @@ +/* + * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef RDMA_USER_RXE_H +#define RDMA_USER_RXE_H + +#include + +union rxe_gid { + __u8 raw[16]; + struct { + __be64 subnet_prefix; + __be64 interface_id; + } global; +}; + +struct rxe_global_route { + union rxe_gid dgid; + __u32 flow_label; + __u8 sgid_index; + __u8 hop_limit; + __u8 traffic_class; +}; + +struct rxe_av { + __u8 port_num; + __u8 network_type; + struct rxe_global_route grh; + union { + struct sockaddr _sockaddr; + struct sockaddr_in _sockaddr_in; + struct sockaddr_in6 _sockaddr_in6; + } sgid_addr, dgid_addr; +}; + +struct rxe_send_wr { + __u64 wr_id; + __u32 num_sge; + __u32 opcode; + __u32 send_flags; + union { + __be32 imm_data; + __u32 invalidate_rkey; + } ex; + union { + struct { + __u64 remote_addr; + __u32 rkey; + } rdma; + struct { + __u64 remote_addr; + __u64 compare_add; + __u64 swap; + __u32 rkey; + } atomic; + struct { + __u32 remote_qpn; + __u32 remote_qkey; + __u16 pkey_index; + } ud; + struct { + struct ib_mr *mr; + __u32 key; + int access; + } reg; + } wr; +}; + +struct rxe_sge { + __u64 addr; + __u32 length; + __u32 lkey; +}; + +struct mminfo { + __u64 offset; + __u32 size; + __u32 pad; +}; + +struct rxe_dma_info { + __u32 length; + __u32 resid; + __u32 cur_sge; + __u32 num_sge; + __u32 sge_offset; + union { + __u8 inline_data[0]; + struct rxe_sge sge[0]; + }; +}; + +struct rxe_send_wqe { + struct rxe_send_wr wr; + struct rxe_av av; + __u32 status; + __u32 state; + __u64 iova; + __u32 mask; + __u32 first_psn; + __u32 last_psn; + __u32 ack_length; + __u32 ssn; + __u32 has_rd_atomic; + struct rxe_dma_info dma; +}; + +struct rxe_recv_wqe { + __u64 wr_id; + __u32 num_sge; + __u32 padding; + struct rxe_dma_info dma; +}; + +#endif /* RDMA_USER_RXE_H */ -- cgit v1.2.3 From 2c86943c20e375b0fe562af0626f2e5461d8d203 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 24 Jul 2016 19:53:19 +0200 Subject: netfilter: nf_tables: s/MFT_REG32_01/NFT_REG32_01 MFT_REG32_01 is a typo, rename this to NFT_REG32_01. Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 01751faccaf8..c674ba2563b7 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -24,7 +24,7 @@ enum nft_registers { __NFT_REG_MAX, NFT_REG32_00 = 8, - MFT_REG32_01, + NFT_REG32_01, NFT_REG32_02, NFT_REG32_03, NFT_REG32_04, -- cgit v1.2.3 From dca3f53c02e325bb19dfc05b1571b9a706226fea Mon Sep 17 00:00:00 2001 From: Phil Sutter Date: Thu, 4 Aug 2016 12:11:55 +0200 Subject: sctp: Export struct sctp_info to userspace This is required to correctly interpret INET_DIAG_INFO messages exported by sctp_diag module. Signed-off-by: Phil Sutter Signed-off-by: David S. Miller --- include/linux/sctp.h | 64 ----------------------------------------------- include/uapi/linux/sctp.h | 64 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 64 insertions(+), 64 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index de1f64318fc4..fcb4c3646173 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -705,70 +705,6 @@ typedef struct sctp_auth_chunk { sctp_authhdr_t auth_hdr; } __packed sctp_auth_chunk_t; -struct sctp_info { - __u32 sctpi_tag; - __u32 sctpi_state; - __u32 sctpi_rwnd; - __u16 sctpi_unackdata; - __u16 sctpi_penddata; - __u16 sctpi_instrms; - __u16 sctpi_outstrms; - __u32 sctpi_fragmentation_point; - __u32 sctpi_inqueue; - __u32 sctpi_outqueue; - __u32 sctpi_overall_error; - __u32 sctpi_max_burst; - __u32 sctpi_maxseg; - __u32 sctpi_peer_rwnd; - __u32 sctpi_peer_tag; - __u8 sctpi_peer_capable; - __u8 sctpi_peer_sack; - __u16 __reserved1; - - /* assoc status info */ - __u64 sctpi_isacks; - __u64 sctpi_osacks; - __u64 sctpi_opackets; - __u64 sctpi_ipackets; - __u64 sctpi_rtxchunks; - __u64 sctpi_outofseqtsns; - __u64 sctpi_idupchunks; - __u64 sctpi_gapcnt; - __u64 sctpi_ouodchunks; - __u64 sctpi_iuodchunks; - __u64 sctpi_oodchunks; - __u64 sctpi_iodchunks; - __u64 sctpi_octrlchunks; - __u64 sctpi_ictrlchunks; - - /* primary transport info */ - struct sockaddr_storage sctpi_p_address; - __s32 sctpi_p_state; - __u32 sctpi_p_cwnd; - __u32 sctpi_p_srtt; - __u32 sctpi_p_rto; - __u32 sctpi_p_hbinterval; - __u32 sctpi_p_pathmaxrxt; - __u32 sctpi_p_sackdelay; - __u32 sctpi_p_sackfreq; - __u32 sctpi_p_ssthresh; - __u32 sctpi_p_partial_bytes_acked; - __u32 sctpi_p_flight_size; - __u16 sctpi_p_error; - __u16 __reserved2; - - /* sctp sock info */ - __u32 sctpi_s_autoclose; - __u32 sctpi_s_adaptation_ind; - __u32 sctpi_s_pd_point; - __u8 sctpi_s_nodelay; - __u8 sctpi_s_disable_fragments; - __u8 sctpi_s_v4mapped; - __u8 sctpi_s_frag_interleave; - __u32 sctpi_s_type; - __u32 __reserved3; -}; - struct sctp_infox { struct sctp_info *sctpinfo; struct sctp_association *asoc; diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index d304f4c9792c..a406adcc0793 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -944,4 +944,68 @@ struct sctp_default_prinfo { __u16 pr_policy; }; +struct sctp_info { + __u32 sctpi_tag; + __u32 sctpi_state; + __u32 sctpi_rwnd; + __u16 sctpi_unackdata; + __u16 sctpi_penddata; + __u16 sctpi_instrms; + __u16 sctpi_outstrms; + __u32 sctpi_fragmentation_point; + __u32 sctpi_inqueue; + __u32 sctpi_outqueue; + __u32 sctpi_overall_error; + __u32 sctpi_max_burst; + __u32 sctpi_maxseg; + __u32 sctpi_peer_rwnd; + __u32 sctpi_peer_tag; + __u8 sctpi_peer_capable; + __u8 sctpi_peer_sack; + __u16 __reserved1; + + /* assoc status info */ + __u64 sctpi_isacks; + __u64 sctpi_osacks; + __u64 sctpi_opackets; + __u64 sctpi_ipackets; + __u64 sctpi_rtxchunks; + __u64 sctpi_outofseqtsns; + __u64 sctpi_idupchunks; + __u64 sctpi_gapcnt; + __u64 sctpi_ouodchunks; + __u64 sctpi_iuodchunks; + __u64 sctpi_oodchunks; + __u64 sctpi_iodchunks; + __u64 sctpi_octrlchunks; + __u64 sctpi_ictrlchunks; + + /* primary transport info */ + struct sockaddr_storage sctpi_p_address; + __s32 sctpi_p_state; + __u32 sctpi_p_cwnd; + __u32 sctpi_p_srtt; + __u32 sctpi_p_rto; + __u32 sctpi_p_hbinterval; + __u32 sctpi_p_pathmaxrxt; + __u32 sctpi_p_sackdelay; + __u32 sctpi_p_sackfreq; + __u32 sctpi_p_ssthresh; + __u32 sctpi_p_partial_bytes_acked; + __u32 sctpi_p_flight_size; + __u16 sctpi_p_error; + __u16 __reserved2; + + /* sctp sock info */ + __u32 sctpi_s_autoclose; + __u32 sctpi_s_adaptation_ind; + __u32 sctpi_s_pd_point; + __u8 sctpi_s_nodelay; + __u8 sctpi_s_disable_fragments; + __u8 sctpi_s_v4mapped; + __u8 sctpi_s_frag_interleave; + __u32 sctpi_s_type; + __u32 __reserved3; +}; + #endif /* _UAPI_SCTP_H */ -- cgit v1.2.3 From cbd74e1bc8129efb9908f130a8a6e60fd95d2106 Mon Sep 17 00:00:00 2001 From: Philippe Bergheaud Date: Fri, 5 Aug 2016 14:02:00 +0200 Subject: cxl: Use fixed width predefined types in data structure. This patch fixes a regression introduced by commit b810253bd934 ("cxl: Add mechanism for delivering AFU driver specific events"). It changes the type u8 to __u8 in the uapi header cxl.h, because the former is a kernel internal type, and may not be defined in userland build environments, in particular when cross-compiling libcxl on x86_64 linux machines (RHEL6.7 and Ubuntu 16.04). This patch also changes the size of the field data_size, and makes it constant, to support 32-bit userland applications running on big-endian ppc64 kernels transparently. mpe: This is an ABI change, however the ABI was only added during the 4.8 merge window so has never been part of a released kernel - therefore we give ourselves permission to change it. Fixes: b810253bd934 ("cxl: Add mechanism for delivering AFU driver specific events") Signed-off-by: Philippe Bergheaud Reviewed-by: Frederic Barrat Signed-off-by: Michael Ellerman --- include/uapi/misc/cxl.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/misc/cxl.h b/include/uapi/misc/cxl.h index cbae529b7ce0..180d526a55c3 100644 --- a/include/uapi/misc/cxl.h +++ b/include/uapi/misc/cxl.h @@ -136,8 +136,8 @@ struct cxl_event_afu_driver_reserved { * * Of course the contents will be ABI, but that's up the AFU driver. */ - size_t data_size; - u8 data[]; + __u32 data_size; + __u8 data[]; }; struct cxl_event { -- cgit v1.2.3 From 28ad55578b8a76390d966b09da8c7fa3644f5140 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Fri, 5 Aug 2016 13:52:09 +0100 Subject: virtio-vsock: fix include guard typo Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin --- include/uapi/linux/virtio_vsock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/virtio_vsock.h b/include/uapi/linux/virtio_vsock.h index 6b011c19b50f..1d57ed3d84d2 100644 --- a/include/uapi/linux/virtio_vsock.h +++ b/include/uapi/linux/virtio_vsock.h @@ -32,7 +32,7 @@ */ #ifndef _UAPI_LINUX_VIRTIO_VSOCK_H -#define _UAPI_LINUX_VIRTIO_VOSCK_H +#define _UAPI_LINUX_VIRTIO_VSOCK_H #include #include -- cgit v1.2.3 From 747ea55e4f78fd980350c39570a986b8c1c3e4aa Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 12 Aug 2016 22:17:17 +0200 Subject: bpf: fix bpf_skb_in_cgroup helper naming While hashing out BPF's current_task_under_cgroup helper bits, it came to discussion that the skb_in_cgroup helper name was suboptimally chosen. Tejun says: So, I think in_cgroup should mean that the object is in that particular cgroup while under_cgroup in the subhierarchy of that cgroup. Let's rename the other subhierarchy test to under too. I think that'd be a lot less confusing going forward. [...] It's more intuitive and gives us the room to implement the real "in" test if ever necessary in the future. Since this touches uapi bits, we need to change this as long as v4.8 is not yet officially released. Thus, change the helper enum and rename related bits. Fixes: 4a482f34afcc ("cgroup: bpf: Add bpf_skb_in_cgroup_proto") Reference: http://patchwork.ozlabs.org/patch/658500/ Suggested-by: Sargun Dhillon Suggested-by: Tejun Heo Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov --- include/uapi/linux/bpf.h | 4 ++-- kernel/bpf/verifier.c | 4 ++-- net/core/filter.c | 10 +++++----- samples/bpf/bpf_helpers.h | 4 ++-- samples/bpf/test_cgrp2_tc_kern.c | 2 +- 5 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index da218fec6056..9e5fc168c8a3 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -339,7 +339,7 @@ enum bpf_func_id { BPF_FUNC_skb_change_type, /** - * bpf_skb_in_cgroup(skb, map, index) - Check cgroup2 membership of skb + * bpf_skb_under_cgroup(skb, map, index) - Check cgroup2 membership of skb * @skb: pointer to skb * @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type * @index: index of the cgroup in the bpf_map @@ -348,7 +348,7 @@ enum bpf_func_id { * == 1 skb succeeded the cgroup2 descendant test * < 0 error */ - BPF_FUNC_skb_in_cgroup, + BPF_FUNC_skb_under_cgroup, /** * bpf_get_hash_recalc(skb) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 7094c69ac199..daea765d72e6 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1053,7 +1053,7 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) goto error; break; case BPF_MAP_TYPE_CGROUP_ARRAY: - if (func_id != BPF_FUNC_skb_in_cgroup) + if (func_id != BPF_FUNC_skb_under_cgroup) goto error; break; default: @@ -1075,7 +1075,7 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) if (map->map_type != BPF_MAP_TYPE_STACK_TRACE) goto error; break; - case BPF_FUNC_skb_in_cgroup: + case BPF_FUNC_skb_under_cgroup: if (map->map_type != BPF_MAP_TYPE_CGROUP_ARRAY) goto error; break; diff --git a/net/core/filter.c b/net/core/filter.c index b5add4ef0d1d..bd9bf2e5fafa 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2317,7 +2317,7 @@ bpf_get_skb_set_tunnel_proto(enum bpf_func_id which) } #ifdef CONFIG_SOCK_CGROUP_DATA -static u64 bpf_skb_in_cgroup(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +static u64 bpf_skb_under_cgroup(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) { struct sk_buff *skb = (struct sk_buff *)(long)r1; struct bpf_map *map = (struct bpf_map *)(long)r2; @@ -2340,8 +2340,8 @@ static u64 bpf_skb_in_cgroup(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) return cgroup_is_descendant(sock_cgroup_ptr(&sk->sk_cgrp_data), cgrp); } -static const struct bpf_func_proto bpf_skb_in_cgroup_proto = { - .func = bpf_skb_in_cgroup, +static const struct bpf_func_proto bpf_skb_under_cgroup_proto = { + .func = bpf_skb_under_cgroup, .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_PTR_TO_CTX, @@ -2421,8 +2421,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) case BPF_FUNC_get_smp_processor_id: return &bpf_get_smp_processor_id_proto; #ifdef CONFIG_SOCK_CGROUP_DATA - case BPF_FUNC_skb_in_cgroup: - return &bpf_skb_in_cgroup_proto; + case BPF_FUNC_skb_under_cgroup: + return &bpf_skb_under_cgroup_proto; #endif default: return sk_filter_func_proto(func_id); diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h index 217c8d507f2e..7927a090fa0d 100644 --- a/samples/bpf/bpf_helpers.h +++ b/samples/bpf/bpf_helpers.h @@ -72,8 +72,8 @@ static int (*bpf_l3_csum_replace)(void *ctx, int off, int from, int to, int flag (void *) BPF_FUNC_l3_csum_replace; static int (*bpf_l4_csum_replace)(void *ctx, int off, int from, int to, int flags) = (void *) BPF_FUNC_l4_csum_replace; -static int (*bpf_skb_in_cgroup)(void *ctx, void *map, int index) = - (void *) BPF_FUNC_skb_in_cgroup; +static int (*bpf_skb_under_cgroup)(void *ctx, void *map, int index) = + (void *) BPF_FUNC_skb_under_cgroup; #if defined(__x86_64__) diff --git a/samples/bpf/test_cgrp2_tc_kern.c b/samples/bpf/test_cgrp2_tc_kern.c index 2732c37c8d5b..10ff73404e3a 100644 --- a/samples/bpf/test_cgrp2_tc_kern.c +++ b/samples/bpf/test_cgrp2_tc_kern.c @@ -57,7 +57,7 @@ int handle_egress(struct __sk_buff *skb) bpf_trace_printk(dont_care_msg, sizeof(dont_care_msg), eth->h_proto, ip6h->nexthdr); return TC_ACT_OK; - } else if (bpf_skb_in_cgroup(skb, &test_cgrp2_array_pin, 0) != 1) { + } else if (bpf_skb_under_cgroup(skb, &test_cgrp2_array_pin, 0) != 1) { bpf_trace_printk(pass_msg, sizeof(pass_msg)); return TC_ACT_OK; } else { -- cgit v1.2.3 From b47b0cc73032d3c8225b5ea9a077941632f16d91 Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:32:38 +0200 Subject: include/uapi/linux/if_pppox.h: include linux/if.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compilation error: error: ‘IFNAMSIZ’ undeclared here (not in a function) Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/if_pppox.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_pppox.h b/include/uapi/linux/if_pppox.h index e128769331b5..473c3c45e3d8 100644 --- a/include/uapi/linux/if_pppox.h +++ b/include/uapi/linux/if_pppox.h @@ -21,6 +21,7 @@ #include #include +#include #include #include -- cgit v1.2.3 From 1fe8e0f074c77aa41aaa579345a9e675acbebfa9 Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:32:39 +0200 Subject: include/uapi/linux/if_tunnel.h: include linux/if.h, linux/ip.h and linux/in6.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compilation errors like: error: field ‘iph’ has incomplete type error: field ‘prefix’ has incomplete type Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/if_tunnel.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index 1046f5515174..777b6cdb1b7b 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -2,6 +2,9 @@ #define _UAPI_IF_TUNNEL_H_ #include +#include +#include +#include #include -- cgit v1.2.3 From 05ee5de7451796cf9a8aeb2f05a57790d4fd2336 Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:32:42 +0200 Subject: include/uapi/linux/if_pppol2tp.h: include linux/in.h and linux/in6.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compilation errors like: error: field ‘addr’ has incomplete type struct sockaddr_in addr; /* IP address and port to send to */ ^ error: field ‘addr’ has incomplete type struct sockaddr_in6 addr; /* IP address and port to send to */ Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/if_pppol2tp.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_pppol2tp.h b/include/uapi/linux/if_pppol2tp.h index 163e8adac2d6..4bd1f55d6377 100644 --- a/include/uapi/linux/if_pppol2tp.h +++ b/include/uapi/linux/if_pppol2tp.h @@ -16,7 +16,8 @@ #define _UAPI__LINUX_IF_PPPOL2TP_H #include - +#include +#include /* Structure used to connect() the socket to a particular tunnel UDP * socket over IPv4. -- cgit v1.2.3 From eafe92114308acf14e45c6c3d154a5dad5523d1a Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:32:43 +0200 Subject: include/uapi/linux/if_pppox.h: include linux/in.h and linux/in6.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compilation errors: error: field ‘addr’ has incomplete type struct sockaddr_in addr; /* IP address and port to send to */ error: field ‘addr’ has incomplete type struct sockaddr_in6 addr; /* IP address and port to send to */ Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/if_pppox.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_pppox.h b/include/uapi/linux/if_pppox.h index 473c3c45e3d8..d37bbb17a007 100644 --- a/include/uapi/linux/if_pppox.h +++ b/include/uapi/linux/if_pppox.h @@ -24,6 +24,8 @@ #include #include #include +#include +#include /* For user-space programs to pick up these definitions * which they wouldn't get otherwise without defining __KERNEL__ -- cgit v1.2.3 From e6571aa5cb65ff52a87843652d0d8120a48aae7c Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:32:55 +0200 Subject: include/uapi/linux/openvswitch.h: use __u32 from linux/types.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compiler error: error: unknown type name ‘uint32_t’ Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/openvswitch.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index d95a3018f6a1..645499a634f7 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -632,8 +632,8 @@ enum ovs_hash_alg { * @hash_basis: basis used for computing hash. */ struct ovs_action_hash { - uint32_t hash_alg; /* One of ovs_hash_alg. */ - uint32_t hash_basis; + __u32 hash_alg; /* One of ovs_hash_alg. */ + __u32 hash_basis; }; /** -- cgit v1.2.3 From cf00713a655d3019be7faa184402f16c43a0fed3 Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:32:58 +0200 Subject: include/uapi/linux/atm_zatm.h: include linux/time.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes userspace compile error: error: field ‘real’ has incomplete type struct timeval real; /* real (wall-clock) time */ Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/atm_zatm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/atm_zatm.h b/include/uapi/linux/atm_zatm.h index 9c9c6ad55f14..5cd4d4d2dd1d 100644 --- a/include/uapi/linux/atm_zatm.h +++ b/include/uapi/linux/atm_zatm.h @@ -14,6 +14,7 @@ #include #include +#include #define ZATM_GETPOOL _IOW('a',ATMIOC_SARPRV+1,struct atmif_sioc) /* get pool statistics */ -- cgit v1.2.3 From a1d1f65ff5ac27276a585b41a619d30995bb92fe Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:33:19 +0200 Subject: include/uapi/linux/openvswitch.h: use __u32 from linux/types.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Kernel uapi header are supposed to use them. Fixes userspace compile error: linux/openvswitch.h:583:2: error: unknown type name ‘uint32_t’ Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/openvswitch.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index 645499a634f7..54c3b4f4aceb 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -583,7 +583,7 @@ enum ovs_userspace_attr { #define OVS_USERSPACE_ATTR_MAX (__OVS_USERSPACE_ATTR_MAX - 1) struct ovs_action_trunc { - uint32_t max_len; /* Max packet size in bytes. */ + __u32 max_len; /* Max packet size in bytes. */ }; /** -- cgit v1.2.3 From 53dc65d4d33c422d086c9d9ad8c03ab400ffc0a1 Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Mon, 22 Aug 2016 20:33:21 +0200 Subject: include/uapi/linux/ipx.h: fix conflicting defitions with glibc netipx/ipx.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes these compiler warnings via libc-compat.h when glibc netipx/ipx.h is included before linux/ipx.h: ./linux/ipx.h:9:8: error: redefinition of ‘struct sockaddr_ipx’ ./linux/ipx.h:26:8: error: redefinition of ‘struct ipx_route_definition’ ./linux/ipx.h:32:8: error: redefinition of ‘struct ipx_interface_definition’ ./linux/ipx.h:49:8: error: redefinition of ‘struct ipx_config_data’ ./linux/ipx.h:58:8: error: redefinition of ‘struct ipx_route_def’ Signed-off-by: Mikko Rapeli Signed-off-by: David S. Miller --- include/uapi/linux/ipx.h | 13 ++++++++++++- include/uapi/linux/libc-compat.h | 26 ++++++++++++++++++++++++++ 2 files changed, 38 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/ipx.h b/include/uapi/linux/ipx.h index 3d48014cdd71..30f031db12f6 100644 --- a/include/uapi/linux/ipx.h +++ b/include/uapi/linux/ipx.h @@ -1,11 +1,13 @@ #ifndef _IPX_H_ #define _IPX_H_ +#include /* for compatibility with glibc netipx/ipx.h */ #include #include #include #define IPX_NODE_LEN 6 #define IPX_MTU 576 +#if __UAPI_DEF_SOCKADDR_IPX struct sockaddr_ipx { __kernel_sa_family_t sipx_family; __be16 sipx_port; @@ -14,6 +16,7 @@ struct sockaddr_ipx { __u8 sipx_type; unsigned char sipx_zero; /* 16 byte fill */ }; +#endif /* __UAPI_DEF_SOCKADDR_IPX */ /* * So we can fit the extra info for SIOCSIFADDR into the address nicely @@ -23,12 +26,15 @@ struct sockaddr_ipx { #define IPX_DLTITF 0 #define IPX_CRTITF 1 +#if __UAPI_DEF_IPX_ROUTE_DEFINITION struct ipx_route_definition { __be32 ipx_network; __be32 ipx_router_network; unsigned char ipx_router_node[IPX_NODE_LEN]; }; +#endif /* __UAPI_DEF_IPX_ROUTE_DEFINITION */ +#if __UAPI_DEF_IPX_INTERFACE_DEFINITION struct ipx_interface_definition { __be32 ipx_network; unsigned char ipx_device[16]; @@ -45,16 +51,20 @@ struct ipx_interface_definition { #define IPX_INTERNAL 2 unsigned char ipx_node[IPX_NODE_LEN]; }; - +#endif /* __UAPI_DEF_IPX_INTERFACE_DEFINITION */ + +#if __UAPI_DEF_IPX_CONFIG_DATA struct ipx_config_data { unsigned char ipxcfg_auto_select_primary; unsigned char ipxcfg_auto_create_interfaces; }; +#endif /* __UAPI_DEF_IPX_CONFIG_DATA */ /* * OLD Route Definition for backward compatibility. */ +#if __UAPI_DEF_IPX_ROUTE_DEF struct ipx_route_def { __be32 ipx_network; __be32 ipx_router_network; @@ -67,6 +77,7 @@ struct ipx_route_def { #define IPX_RT_BLUEBOOK 2 #define IPX_RT_ROUTED 1 }; +#endif /* __UAPI_DEF_IPX_ROUTE_DEF */ #define SIOCAIPXITFCRT (SIOCPROTOPRIVATE) #define SIOCAIPXPRISLT (SIOCPROTOPRIVATE + 1) diff --git a/include/uapi/linux/libc-compat.h b/include/uapi/linux/libc-compat.h index e4f048ee7043..44b8a6bd5fe1 100644 --- a/include/uapi/linux/libc-compat.h +++ b/include/uapi/linux/libc-compat.h @@ -139,6 +139,25 @@ #endif /* _NETINET_IN_H */ +/* Coordinate with glibc netipx/ipx.h header. */ +#if defined(__NETIPX_IPX_H) + +#define __UAPI_DEF_SOCKADDR_IPX 0 +#define __UAPI_DEF_IPX_ROUTE_DEFINITION 0 +#define __UAPI_DEF_IPX_INTERFACE_DEFINITION 0 +#define __UAPI_DEF_IPX_CONFIG_DATA 0 +#define __UAPI_DEF_IPX_ROUTE_DEF 0 + +#else /* defined(__NETIPX_IPX_H) */ + +#define __UAPI_DEF_SOCKADDR_IPX 1 +#define __UAPI_DEF_IPX_ROUTE_DEFINITION 1 +#define __UAPI_DEF_IPX_INTERFACE_DEFINITION 1 +#define __UAPI_DEF_IPX_CONFIG_DATA 1 +#define __UAPI_DEF_IPX_ROUTE_DEF 1 + +#endif /* defined(__NETIPX_IPX_H) */ + /* Definitions for xattr.h */ #if defined(_SYS_XATTR_H) #define __UAPI_DEF_XATTR 0 @@ -179,6 +198,13 @@ #define __UAPI_DEF_IN6_PKTINFO 1 #define __UAPI_DEF_IP6_MTUINFO 1 +/* Definitions for ipx.h */ +#define __UAPI_DEF_SOCKADDR_IPX 1 +#define __UAPI_DEF_IPX_ROUTE_DEFINITION 1 +#define __UAPI_DEF_IPX_INTERFACE_DEFINITION 1 +#define __UAPI_DEF_IPX_CONFIG_DATA 1 +#define __UAPI_DEF_IPX_ROUTE_DEF 1 + /* Definitions for xattr.h */ #define __UAPI_DEF_XATTR 1 -- cgit v1.2.3