From d02f51cbcf12b09ab945873e35046045875eed9a Mon Sep 17 00:00:00 2001 From: Daniel Axtens Date: Sat, 3 Mar 2018 03:03:46 +0100 Subject: bpf: fix bpf_skb_adjust_net/bpf_skb_proto_xlat to deal with gso sctp skbs SCTP GSO skbs have a gso_size of GSO_BY_FRAGS, so any sort of unconditionally mangling of that will result in nonsense value and would corrupt the skb later on. Therefore, i) add two helpers skb_increase_gso_size() and skb_decrease_gso_size() that would throw a one time warning and bail out for such skbs and ii) refuse and return early with an error in those BPF helpers that are affected. We do need to bail out as early as possible from there before any changes on the skb have been performed. Fixes: 6578171a7ff0 ("bpf: add bpf_skb_change_proto helper") Co-authored-by: Daniel Borkmann Signed-off-by: Daniel Axtens Cc: Marcelo Ricardo Leitner Acked-by: Alexei Starovoitov Signed-off-by: Alexei Starovoitov --- include/linux/skbuff.h | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index c1e66bdcf583..8c67c33f40c9 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4038,6 +4038,12 @@ static inline bool skb_is_gso_v6(const struct sk_buff *skb) return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6; } +/* Note: Should be called only if skb_is_gso(skb) is true */ +static inline bool skb_is_gso_sctp(const struct sk_buff *skb) +{ + return skb_shinfo(skb)->gso_type & SKB_GSO_SCTP; +} + static inline void skb_gso_reset(struct sk_buff *skb) { skb_shinfo(skb)->gso_size = 0; @@ -4045,6 +4051,22 @@ static inline void skb_gso_reset(struct sk_buff *skb) skb_shinfo(skb)->gso_type = 0; } +static inline void skb_increase_gso_size(struct skb_shared_info *shinfo, + u16 increment) +{ + if (WARN_ON_ONCE(shinfo->gso_size == GSO_BY_FRAGS)) + return; + shinfo->gso_size += increment; +} + +static inline void skb_decrease_gso_size(struct skb_shared_info *shinfo, + u16 decrement) +{ + if (WARN_ON_ONCE(shinfo->gso_size == GSO_BY_FRAGS)) + return; + shinfo->gso_size -= decrement; +} + void __skb_warn_lro_forwarding(const struct sk_buff *skb); static inline bool skb_warn_if_lro(const struct sk_buff *skb) -- cgit v1.2.3 From d3dcf8eb615537526bd42ff27a081d46d337816e Mon Sep 17 00:00:00 2001 From: Paul Blakey Date: Sun, 4 Mar 2018 17:29:48 +0200 Subject: rhashtable: Fix rhlist duplicates insertion When inserting duplicate objects (those with the same key), current rhlist implementation messes up the chain pointers by updating the bucket pointer instead of prev next pointer to the newly inserted node. This causes missing elements on removal and travesal. Fix that by properly updating pprev pointer to point to the correct rhash_head next pointer. Issue: 1241076 Change-Id: I86b2c140bcb4aeb10b70a72a267ff590bb2b17e7 Fixes: ca26893f05e8 ('rhashtable: Add rhlist interface') Signed-off-by: Paul Blakey Acked-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/rhashtable.h | 4 +++- lib/rhashtable.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h index c9df2527e0cd..668a21f04b09 100644 --- a/include/linux/rhashtable.h +++ b/include/linux/rhashtable.h @@ -766,8 +766,10 @@ slow_path: if (!key || (params.obj_cmpfn ? params.obj_cmpfn(&arg, rht_obj(ht, head)) : - rhashtable_compare(&arg, rht_obj(ht, head)))) + rhashtable_compare(&arg, rht_obj(ht, head)))) { + pprev = &head->next; continue; + } data = rht_obj(ht, head); diff --git a/lib/rhashtable.c b/lib/rhashtable.c index 3825c30aaa36..47de025b6245 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c @@ -506,8 +506,10 @@ static void *rhashtable_lookup_one(struct rhashtable *ht, if (!key || (ht->p.obj_cmpfn ? ht->p.obj_cmpfn(&arg, rht_obj(ht, head)) : - rhashtable_compare(&arg, rht_obj(ht, head)))) + rhashtable_compare(&arg, rht_obj(ht, head)))) { + pprev = &head->next; continue; + } if (!ht->rhlist) return rht_obj(ht, head); -- cgit v1.2.3 From 2695578b896aea472b2c0dcbe9d92daa71738484 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 5 Mar 2018 11:41:13 -0800 Subject: net: usbnet: fix potential deadlock on 32bit hosts Marek reported a LOCKDEP issue occurring on 32bit host, that we tracked down to the fact that usbnet could either run from soft or hard irqs. This patch adds u64_stats_update_begin_irqsave() and u64_stats_update_end_irqrestore() helpers to solve this case. [ 17.768040] ================================ [ 17.772239] WARNING: inconsistent lock state [ 17.776511] 4.16.0-rc3-next-20180227-00007-g876c53a7493c #453 Not tainted [ 17.783329] -------------------------------- [ 17.787580] inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage. [ 17.793607] swapper/0/0 [HC0[0]:SC1[1]:HE1:SE0] takes: [ 17.798751] (&syncp->seq#5){?.-.}, at: [<9b22e5f0>] asix_rx_fixup_internal+0x188/0x288 [ 17.806790] {IN-HARDIRQ-W} state was registered at: [ 17.811677] tx_complete+0x100/0x208 [ 17.815319] __usb_hcd_giveback_urb+0x60/0xf0 [ 17.819770] xhci_giveback_urb_in_irq+0xa8/0x240 [ 17.824469] xhci_td_cleanup+0xf4/0x16c [ 17.828367] xhci_irq+0xe74/0x2240 [ 17.831827] usb_hcd_irq+0x24/0x38 [ 17.835343] __handle_irq_event_percpu+0x98/0x510 [ 17.840111] handle_irq_event_percpu+0x1c/0x58 [ 17.844623] handle_irq_event+0x38/0x5c [ 17.848519] handle_fasteoi_irq+0xa4/0x138 [ 17.852681] generic_handle_irq+0x18/0x28 [ 17.856760] __handle_domain_irq+0x6c/0xe4 [ 17.860941] gic_handle_irq+0x54/0xa0 [ 17.864666] __irq_svc+0x70/0xb0 [ 17.867964] arch_cpu_idle+0x20/0x3c [ 17.871578] arch_cpu_idle+0x20/0x3c [ 17.875190] do_idle+0x144/0x218 [ 17.878468] cpu_startup_entry+0x18/0x1c [ 17.882454] start_kernel+0x394/0x400 [ 17.886177] irq event stamp: 161912 [ 17.889616] hardirqs last enabled at (161912): [<7bedfacf>] __netdev_alloc_skb+0xcc/0x140 [ 17.897893] hardirqs last disabled at (161911): [] __netdev_alloc_skb+0x94/0x140 [ 17.904903] exynos5-hsi2c 12ca0000.i2c: tx timeout [ 17.906116] softirqs last enabled at (161904): [<387102ff>] irq_enter+0x78/0x80 [ 17.906123] softirqs last disabled at (161905): [] irq_exit+0x134/0x158 [ 17.925722]. [ 17.925722] other info that might help us debug this: [ 17.933435] Possible unsafe locking scenario: [ 17.933435]. [ 17.940331] CPU0 [ 17.942488] ---- [ 17.944894] lock(&syncp->seq#5); [ 17.948274] [ 17.950847] lock(&syncp->seq#5); [ 17.954386]. [ 17.954386] *** DEADLOCK *** [ 17.954386]. [ 17.962422] no locks held by swapper/0/0. Fixes: c8b5d129ee29 ("net: usbnet: support 64bit stats") Signed-off-by: Eric Dumazet Reported-by: Marek Szyprowski Cc: Greg Ungerer Signed-off-by: David S. Miller --- drivers/net/usb/usbnet.c | 10 ++++++---- include/linux/u64_stats_sync.h | 22 ++++++++++++++++++++++ 2 files changed, 28 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 8a22ff67b026..d9eea8cfe6cb 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -315,6 +315,7 @@ static void __usbnet_status_stop_force(struct usbnet *dev) void usbnet_skb_return (struct usbnet *dev, struct sk_buff *skb) { struct pcpu_sw_netstats *stats64 = this_cpu_ptr(dev->stats64); + unsigned long flags; int status; if (test_bit(EVENT_RX_PAUSED, &dev->flags)) { @@ -326,10 +327,10 @@ void usbnet_skb_return (struct usbnet *dev, struct sk_buff *skb) if (skb->protocol == 0) skb->protocol = eth_type_trans (skb, dev->net); - u64_stats_update_begin(&stats64->syncp); + flags = u64_stats_update_begin_irqsave(&stats64->syncp); stats64->rx_packets++; stats64->rx_bytes += skb->len; - u64_stats_update_end(&stats64->syncp); + u64_stats_update_end_irqrestore(&stats64->syncp, flags); netif_dbg(dev, rx_status, dev->net, "< rx, len %zu, type 0x%x\n", skb->len + sizeof (struct ethhdr), skb->protocol); @@ -1248,11 +1249,12 @@ static void tx_complete (struct urb *urb) if (urb->status == 0) { struct pcpu_sw_netstats *stats64 = this_cpu_ptr(dev->stats64); + unsigned long flags; - u64_stats_update_begin(&stats64->syncp); + flags = u64_stats_update_begin_irqsave(&stats64->syncp); stats64->tx_packets += entry->packets; stats64->tx_bytes += entry->length; - u64_stats_update_end(&stats64->syncp); + u64_stats_update_end_irqrestore(&stats64->syncp, flags); } else { dev->net->stats.tx_errors++; diff --git a/include/linux/u64_stats_sync.h b/include/linux/u64_stats_sync.h index 5bdbd9f49395..07ee0f84a46c 100644 --- a/include/linux/u64_stats_sync.h +++ b/include/linux/u64_stats_sync.h @@ -90,6 +90,28 @@ static inline void u64_stats_update_end(struct u64_stats_sync *syncp) #endif } +static inline unsigned long +u64_stats_update_begin_irqsave(struct u64_stats_sync *syncp) +{ + unsigned long flags = 0; + +#if BITS_PER_LONG==32 && defined(CONFIG_SMP) + local_irq_save(flags); + write_seqcount_begin(&syncp->seq); +#endif + return flags; +} + +static inline void +u64_stats_update_end_irqrestore(struct u64_stats_sync *syncp, + unsigned long flags) +{ +#if BITS_PER_LONG==32 && defined(CONFIG_SMP) + write_seqcount_end(&syncp->seq); + local_irq_restore(flags); +#endif +} + static inline void u64_stats_update_begin_raw(struct u64_stats_sync *syncp) { #if BITS_PER_LONG==32 && defined(CONFIG_SMP) -- cgit v1.2.3 From 3a4030761ea88ff439030ca98e3094b9900e96b7 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 9 Mar 2018 14:50:34 +0800 Subject: vhost_net: examine pointer types during un-producing After commit fc72d1d54dd9 ("tuntap: XDP transmission"), we can actually queueing XDP pointers in the pointer ring, so we should examine the pointer type before freeing the pointer. Fixes: fc72d1d54dd9 ("tuntap: XDP transmission") Reported-by: Michael S. Tsirkin Acked-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/tun.c | 3 ++- drivers/vhost/net.c | 2 +- include/linux/if_tun.h | 4 ++++ 3 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 7433bb2e4451..28cfa642e39a 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -655,7 +655,7 @@ static struct tun_struct *tun_enable_queue(struct tun_file *tfile) return tun; } -static void tun_ptr_free(void *ptr) +void tun_ptr_free(void *ptr) { if (!ptr) return; @@ -667,6 +667,7 @@ static void tun_ptr_free(void *ptr) __skb_array_destroy_skb(ptr); } } +EXPORT_SYMBOL_GPL(tun_ptr_free); static void tun_queue_purge(struct tun_file *tfile) { diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index efb93063fda1..8139bc70ad7d 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -170,7 +170,7 @@ static void vhost_net_buf_unproduce(struct vhost_net_virtqueue *nvq) if (nvq->rx_ring && !vhost_net_buf_is_empty(rxq)) { ptr_ring_unconsume(nvq->rx_ring, rxq->queue + rxq->head, vhost_net_buf_get_size(rxq), - __skb_array_destroy_skb); + tun_ptr_free); rxq->head = rxq->tail = 0; } } diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h index c5b0a75a7812..fd00170b494f 100644 --- a/include/linux/if_tun.h +++ b/include/linux/if_tun.h @@ -25,6 +25,7 @@ struct ptr_ring *tun_get_tx_ring(struct file *file); bool tun_is_xdp_buff(void *ptr); void *tun_xdp_to_ptr(void *ptr); void *tun_ptr_to_xdp(void *ptr); +void tun_ptr_free(void *ptr); #else #include #include @@ -50,5 +51,8 @@ static inline void *tun_ptr_to_xdp(void *ptr) { return NULL; } +static inline void tun_ptr_free(void *ptr) +{ +} #endif /* CONFIG_TUN */ #endif /* __IF_TUN_H */ -- cgit v1.2.3 From b1d0a5d0cba4597c0394997b2d5fced3e3841b4e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 10 Mar 2018 01:15:45 +0100 Subject: netfilter: x_tables: add and use xt_check_proc_name recent and hashlimit both create /proc files, but only check that name is 0 terminated. This can trigger WARN() from procfs when name is "" or "/". Add helper for this and then use it for both. Cc: Eric Dumazet Reported-by: Eric Dumazet Reported-by: Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/x_tables.h | 2 ++ net/netfilter/x_tables.c | 30 ++++++++++++++++++++++++++++++ net/netfilter/xt_hashlimit.c | 16 ++++++++++------ net/netfilter/xt_recent.c | 6 +++--- 4 files changed, 45 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 1313b35c3ab7..14529511c4b8 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -285,6 +285,8 @@ unsigned int *xt_alloc_entry_offsets(unsigned int size); bool xt_find_jump_offset(const unsigned int *offsets, unsigned int target, unsigned int size); +int xt_check_proc_name(const char *name, unsigned int size); + int xt_check_match(struct xt_mtchk_param *, unsigned int size, u_int8_t proto, bool inv_proto); int xt_check_target(struct xt_tgchk_param *, unsigned int size, u_int8_t proto, diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index fa1655aff8d3..4aa01c90e9d1 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -423,6 +423,36 @@ textify_hooks(char *buf, size_t size, unsigned int mask, uint8_t nfproto) return buf; } +/** + * xt_check_proc_name - check that name is suitable for /proc file creation + * + * @name: file name candidate + * @size: length of buffer + * + * some x_tables modules wish to create a file in /proc. + * This function makes sure that the name is suitable for this + * purpose, it checks that name is NUL terminated and isn't a 'special' + * name, like "..". + * + * returns negative number on error or 0 if name is useable. + */ +int xt_check_proc_name(const char *name, unsigned int size) +{ + if (name[0] == '\0') + return -EINVAL; + + if (strnlen(name, size) == size) + return -ENAMETOOLONG; + + if (strcmp(name, ".") == 0 || + strcmp(name, "..") == 0 || + strchr(name, '/')) + return -EINVAL; + + return 0; +} +EXPORT_SYMBOL(xt_check_proc_name); + int xt_check_match(struct xt_mtchk_param *par, unsigned int size, u_int8_t proto, bool inv_proto) { diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index 66f5aca62a08..3360f13dc208 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -917,8 +917,9 @@ static int hashlimit_mt_check_v1(const struct xt_mtchk_param *par) struct hashlimit_cfg3 cfg = {}; int ret; - if (info->name[sizeof(info->name) - 1] != '\0') - return -EINVAL; + ret = xt_check_proc_name(info->name, sizeof(info->name)); + if (ret) + return ret; ret = cfg_copy(&cfg, (void *)&info->cfg, 1); @@ -935,8 +936,9 @@ static int hashlimit_mt_check_v2(const struct xt_mtchk_param *par) struct hashlimit_cfg3 cfg = {}; int ret; - if (info->name[sizeof(info->name) - 1] != '\0') - return -EINVAL; + ret = xt_check_proc_name(info->name, sizeof(info->name)); + if (ret) + return ret; ret = cfg_copy(&cfg, (void *)&info->cfg, 2); @@ -950,9 +952,11 @@ static int hashlimit_mt_check_v2(const struct xt_mtchk_param *par) static int hashlimit_mt_check(const struct xt_mtchk_param *par) { struct xt_hashlimit_mtinfo3 *info = par->matchinfo; + int ret; - if (info->name[sizeof(info->name) - 1] != '\0') - return -EINVAL; + ret = xt_check_proc_name(info->name, sizeof(info->name)); + if (ret) + return ret; return hashlimit_mt_check_common(par, &info->hinfo, &info->cfg, info->name, 3); diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 6d232d18faff..81ee1d6543b2 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -361,9 +361,9 @@ static int recent_mt_check(const struct xt_mtchk_param *par, info->hit_count, XT_RECENT_MAX_NSTAMPS - 1); return -EINVAL; } - if (info->name[0] == '\0' || - strnlen(info->name, XT_RECENT_NAME_LEN) == XT_RECENT_NAME_LEN) - return -EINVAL; + ret = xt_check_proc_name(info->name, sizeof(info->name)); + if (ret) + return ret; if (ip_pkt_list_tot && info->hit_count < ip_pkt_list_tot) nstamp_mask = roundup_pow_of_two(ip_pkt_list_tot) - 1; -- cgit v1.2.3 From a2c054a896b8ac794ddcfc7c92e2dc7ec4ed4ed5 Mon Sep 17 00:00:00 2001 From: Brad Mouring Date: Thu, 8 Mar 2018 16:23:03 -0600 Subject: net: phy: Tell caller result of phy_change() In 664fcf123a30e (net: phy: Threaded interrupts allow some simplification) the phy_interrupt system was changed to use a traditional threaded interrupt scheme instead of a workqueue approach. With this change, the phy status check moved into phy_change, which did not report back to the caller whether or not the interrupt was handled. This means that, in the case of a shared phy interrupt, only the first phydev's interrupt registers are checked (since phy_interrupt() would always return IRQ_HANDLED). This leads to interrupt storms when it is a secondary device that's actually the interrupt source. Signed-off-by: Brad Mouring Signed-off-by: David S. Miller --- drivers/net/phy/phy.c | 145 +++++++++++++++++++++++++------------------------- include/linux/phy.h | 1 - 2 files changed, 72 insertions(+), 74 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index a6f924fee584..9aabfa1a455a 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -617,6 +617,77 @@ static void phy_error(struct phy_device *phydev) phy_trigger_machine(phydev, false); } +/** + * phy_disable_interrupts - Disable the PHY interrupts from the PHY side + * @phydev: target phy_device struct + */ +static int phy_disable_interrupts(struct phy_device *phydev) +{ + int err; + + /* Disable PHY interrupts */ + err = phy_config_interrupt(phydev, PHY_INTERRUPT_DISABLED); + if (err) + goto phy_err; + + /* Clear the interrupt */ + err = phy_clear_interrupt(phydev); + if (err) + goto phy_err; + + return 0; + +phy_err: + phy_error(phydev); + + return err; +} + +/** + * phy_change - Called by the phy_interrupt to handle PHY changes + * @phydev: phy_device struct that interrupted + */ +static irqreturn_t phy_change(struct phy_device *phydev) +{ + if (phy_interrupt_is_valid(phydev)) { + if (phydev->drv->did_interrupt && + !phydev->drv->did_interrupt(phydev)) + return IRQ_NONE; + + if (phydev->state == PHY_HALTED) + if (phy_disable_interrupts(phydev)) + goto phy_err; + } + + mutex_lock(&phydev->lock); + if ((PHY_RUNNING == phydev->state) || (PHY_NOLINK == phydev->state)) + phydev->state = PHY_CHANGELINK; + mutex_unlock(&phydev->lock); + + /* reschedule state queue work to run as soon as possible */ + phy_trigger_machine(phydev, true); + + if (phy_interrupt_is_valid(phydev) && phy_clear_interrupt(phydev)) + goto phy_err; + return IRQ_HANDLED; + +phy_err: + phy_error(phydev); + return IRQ_NONE; +} + +/** + * phy_change_work - Scheduled by the phy_mac_interrupt to handle PHY changes + * @work: work_struct that describes the work to be done + */ +void phy_change_work(struct work_struct *work) +{ + struct phy_device *phydev = + container_of(work, struct phy_device, phy_queue); + + phy_change(phydev); +} + /** * phy_interrupt - PHY interrupt handler * @irq: interrupt line @@ -632,9 +703,7 @@ static irqreturn_t phy_interrupt(int irq, void *phy_dat) if (PHY_HALTED == phydev->state) return IRQ_NONE; /* It can't be ours. */ - phy_change(phydev); - - return IRQ_HANDLED; + return phy_change(phydev); } /** @@ -651,32 +720,6 @@ static int phy_enable_interrupts(struct phy_device *phydev) return phy_config_interrupt(phydev, PHY_INTERRUPT_ENABLED); } -/** - * phy_disable_interrupts - Disable the PHY interrupts from the PHY side - * @phydev: target phy_device struct - */ -static int phy_disable_interrupts(struct phy_device *phydev) -{ - int err; - - /* Disable PHY interrupts */ - err = phy_config_interrupt(phydev, PHY_INTERRUPT_DISABLED); - if (err) - goto phy_err; - - /* Clear the interrupt */ - err = phy_clear_interrupt(phydev); - if (err) - goto phy_err; - - return 0; - -phy_err: - phy_error(phydev); - - return err; -} - /** * phy_start_interrupts - request and enable interrupts for a PHY device * @phydev: target phy_device struct @@ -719,50 +762,6 @@ int phy_stop_interrupts(struct phy_device *phydev) } EXPORT_SYMBOL(phy_stop_interrupts); -/** - * phy_change - Called by the phy_interrupt to handle PHY changes - * @phydev: phy_device struct that interrupted - */ -void phy_change(struct phy_device *phydev) -{ - if (phy_interrupt_is_valid(phydev)) { - if (phydev->drv->did_interrupt && - !phydev->drv->did_interrupt(phydev)) - return; - - if (phydev->state == PHY_HALTED) - if (phy_disable_interrupts(phydev)) - goto phy_err; - } - - mutex_lock(&phydev->lock); - if ((PHY_RUNNING == phydev->state) || (PHY_NOLINK == phydev->state)) - phydev->state = PHY_CHANGELINK; - mutex_unlock(&phydev->lock); - - /* reschedule state queue work to run as soon as possible */ - phy_trigger_machine(phydev, true); - - if (phy_interrupt_is_valid(phydev) && phy_clear_interrupt(phydev)) - goto phy_err; - return; - -phy_err: - phy_error(phydev); -} - -/** - * phy_change_work - Scheduled by the phy_mac_interrupt to handle PHY changes - * @work: work_struct that describes the work to be done - */ -void phy_change_work(struct work_struct *work) -{ - struct phy_device *phydev = - container_of(work, struct phy_device, phy_queue); - - phy_change(phydev); -} - /** * phy_stop - Bring down the PHY link, and stop checking the status * @phydev: target phy_device struct diff --git a/include/linux/phy.h b/include/linux/phy.h index d7069539f351..b260fb336b25 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -1012,7 +1012,6 @@ int phy_driver_register(struct phy_driver *new_driver, struct module *owner); int phy_drivers_register(struct phy_driver *new_driver, int n, struct module *owner); void phy_state_machine(struct work_struct *work); -void phy_change(struct phy_device *phydev); void phy_change_work(struct work_struct *work); void phy_mac_interrupt(struct phy_device *phydev); void phy_start_machine(struct phy_device *phydev); -- cgit v1.2.3 From bf2ae2e4bf9360e07c0cdfa166bcdc0afd92f4ce Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 10 Mar 2018 18:57:50 +0800 Subject: sock_diag: request _diag module only when the family or proto has been registered Now when using 'ss' in iproute, kernel would try to load all _diag modules, which also causes corresponding family and proto modules to be loaded as well due to module dependencies. Like after running 'ss', sctp, dccp, af_packet (if it works as a module) would be loaded. For example: $ lsmod|grep sctp $ ss $ lsmod|grep sctp sctp_diag 16384 0 sctp 323584 5 sctp_diag inet_diag 24576 4 raw_diag,tcp_diag,sctp_diag,udp_diag libcrc32c 16384 3 nf_conntrack,nf_nat,sctp As these family and proto modules are loaded unintentionally, it could cause some problems, like: - Some debug tools use 'ss' to collect the socket info, which loads all those diag and family and protocol modules. It's noisy for identifying issues. - Users usually expect to drop sctp init packet silently when they have no sense of sctp protocol instead of sending abort back. - It wastes resources (especially with multiple netns), and SCTP module can't be unloaded once it's loaded. ... In short, it's really inappropriate to have these family and proto modules loaded unexpectedly when just doing debugging with inet_diag. This patch is to introduce sock_load_diag_module() where it loads the _diag module only when it's corresponding family or proto has been already registered. Note that we can't just load _diag module without the family or proto loaded, as some symbols used in _diag module are from the family or proto module. v1->v2: - move inet proto check to inet_diag to avoid a compiling err. v2->v3: - define sock_load_diag_module in sock.c and export one symbol only. - improve the changelog. Reported-by: Sabrina Dubroca Acked-by: Marcelo Ricardo Leitner Acked-by: Phil Sutter Acked-by: Sabrina Dubroca Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/linux/net.h | 1 + include/net/sock.h | 1 + net/core/sock.c | 21 +++++++++++++++++++++ net/core/sock_diag.c | 12 ++++-------- net/ipv4/inet_diag.c | 3 +-- net/socket.c | 5 +++++ 6 files changed, 33 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/net.h b/include/linux/net.h index 91216b16feb7..2a0391eea05c 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -222,6 +222,7 @@ enum { int sock_wake_async(struct socket_wq *sk_wq, int how, int band); int sock_register(const struct net_proto_family *fam); void sock_unregister(int family); +bool sock_is_registered(int family); int __sock_create(struct net *net, int family, int type, int proto, struct socket **res, int kern); int sock_create(int family, int type, int proto, struct socket **res); diff --git a/include/net/sock.h b/include/net/sock.h index 169c92afcafa..ae23f3b389ca 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1137,6 +1137,7 @@ struct proto { int proto_register(struct proto *prot, int alloc_slab); void proto_unregister(struct proto *prot); +int sock_load_diag_module(int family, int protocol); #ifdef SOCK_REFCNT_DEBUG static inline void sk_refcnt_debug_inc(struct sock *sk) diff --git a/net/core/sock.c b/net/core/sock.c index c501499a04fe..85b0b64e7f9d 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3261,6 +3261,27 @@ void proto_unregister(struct proto *prot) } EXPORT_SYMBOL(proto_unregister); +int sock_load_diag_module(int family, int protocol) +{ + if (!protocol) { + if (!sock_is_registered(family)) + return -ENOENT; + + return request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, + NETLINK_SOCK_DIAG, family); + } + +#ifdef CONFIG_INET + if (family == AF_INET && + !rcu_access_pointer(inet_protos[protocol])) + return -ENOENT; +#endif + + return request_module("net-pf-%d-proto-%d-type-%d-%d", PF_NETLINK, + NETLINK_SOCK_DIAG, family, protocol); +} +EXPORT_SYMBOL(sock_load_diag_module); + #ifdef CONFIG_PROC_FS static void *proto_seq_start(struct seq_file *seq, loff_t *pos) __acquires(proto_list_mutex) diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c index 146b50e30659..c37b5be7c5e4 100644 --- a/net/core/sock_diag.c +++ b/net/core/sock_diag.c @@ -220,8 +220,7 @@ static int __sock_diag_cmd(struct sk_buff *skb, struct nlmsghdr *nlh) return -EINVAL; if (sock_diag_handlers[req->sdiag_family] == NULL) - request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, - NETLINK_SOCK_DIAG, req->sdiag_family); + sock_load_diag_module(req->sdiag_family, 0); mutex_lock(&sock_diag_table_mutex); hndl = sock_diag_handlers[req->sdiag_family]; @@ -247,8 +246,7 @@ static int sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, case TCPDIAG_GETSOCK: case DCCPDIAG_GETSOCK: if (inet_rcv_compat == NULL) - request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, - NETLINK_SOCK_DIAG, AF_INET); + sock_load_diag_module(AF_INET, 0); mutex_lock(&sock_diag_table_mutex); if (inet_rcv_compat != NULL) @@ -281,14 +279,12 @@ static int sock_diag_bind(struct net *net, int group) case SKNLGRP_INET_TCP_DESTROY: case SKNLGRP_INET_UDP_DESTROY: if (!sock_diag_handlers[AF_INET]) - request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, - NETLINK_SOCK_DIAG, AF_INET); + sock_load_diag_module(AF_INET, 0); break; case SKNLGRP_INET6_TCP_DESTROY: case SKNLGRP_INET6_UDP_DESTROY: if (!sock_diag_handlers[AF_INET6]) - request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, - NETLINK_SOCK_DIAG, AF_INET6); + sock_load_diag_module(AF_INET6, 0); break; } return 0; diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index a383f299ce24..4e5bc4b2f14e 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -53,8 +53,7 @@ static DEFINE_MUTEX(inet_diag_table_mutex); static const struct inet_diag_handler *inet_diag_lock_handler(int proto) { if (!inet_diag_table[proto]) - request_module("net-pf-%d-proto-%d-type-%d-%d", PF_NETLINK, - NETLINK_SOCK_DIAG, AF_INET, proto); + sock_load_diag_module(AF_INET, proto); mutex_lock(&inet_diag_table_mutex); if (!inet_diag_table[proto]) diff --git a/net/socket.c b/net/socket.c index a93c99b518ca..08847c3b8c39 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2587,6 +2587,11 @@ void sock_unregister(int family) } EXPORT_SYMBOL(sock_unregister); +bool sock_is_registered(int family) +{ + return family < NPROTO && rcu_access_pointer(net_families[family]); +} + static int __init sock_init(void) { int err; -- cgit v1.2.3 From c2b37f76485f073f020e60b5954b6dc4e55f693c Mon Sep 17 00:00:00 2001 From: Boris Pismenny Date: Thu, 8 Mar 2018 15:51:41 +0200 Subject: IB/mlx5: Fix integer overflows in mlx5_ib_create_srq This patch validates user provided input to prevent integer overflow due to integer manipulation in the mlx5_ib_create_srq function. Cc: syzkaller Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters") Signed-off-by: Boris Pismenny Signed-off-by: Leon Romanovsky Signed-off-by: Doug Ledford --- drivers/infiniband/hw/mlx5/srq.c | 15 +++++++++------ include/linux/mlx5/driver.h | 4 ++-- 2 files changed, 11 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c index 6d5fadad9090..3c7522d025f2 100644 --- a/drivers/infiniband/hw/mlx5/srq.c +++ b/drivers/infiniband/hw/mlx5/srq.c @@ -241,8 +241,8 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd, { struct mlx5_ib_dev *dev = to_mdev(pd->device); struct mlx5_ib_srq *srq; - int desc_size; - int buf_size; + size_t desc_size; + size_t buf_size; int err; struct mlx5_srq_attr in = {0}; __u32 max_srq_wqes = 1 << MLX5_CAP_GEN(dev->mdev, log_max_srq_sz); @@ -266,15 +266,18 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd, desc_size = sizeof(struct mlx5_wqe_srq_next_seg) + srq->msrq.max_gs * sizeof(struct mlx5_wqe_data_seg); + if (desc_size == 0 || srq->msrq.max_gs > desc_size) + return ERR_PTR(-EINVAL); desc_size = roundup_pow_of_two(desc_size); - desc_size = max_t(int, 32, desc_size); + desc_size = max_t(size_t, 32, desc_size); + if (desc_size < sizeof(struct mlx5_wqe_srq_next_seg)) + return ERR_PTR(-EINVAL); srq->msrq.max_avail_gather = (desc_size - sizeof(struct mlx5_wqe_srq_next_seg)) / sizeof(struct mlx5_wqe_data_seg); srq->msrq.wqe_shift = ilog2(desc_size); buf_size = srq->msrq.max * desc_size; - mlx5_ib_dbg(dev, "desc_size 0x%x, req wr 0x%x, srq size 0x%x, max_gs 0x%x, max_avail_gather 0x%x\n", - desc_size, init_attr->attr.max_wr, srq->msrq.max, srq->msrq.max_gs, - srq->msrq.max_avail_gather); + if (buf_size < desc_size) + return ERR_PTR(-EINVAL); in.type = init_attr->srq_type; if (pd->uobject) diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 6ed79a8a8318..9d3a03364e6e 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -453,8 +453,8 @@ struct mlx5_core_srq { struct mlx5_core_rsc_common common; /* must be first */ u32 srqn; int max; - int max_gs; - int max_avail_gather; + size_t max_gs; + size_t max_avail_gather; int wqe_shift; void (*event) (struct mlx5_core_srq *, enum mlx5_event); -- cgit v1.2.3 From 6417250d3f894e66a68ba1cd93676143f2376a6f Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 6 Mar 2018 19:34:42 -0800 Subject: workqueue: remove unused cancel_work() Found this by accident. There are no usages of bare cancel_work() in current kernel source. Signed-off-by: Stephen Hemminger Signed-off-by: Tejun Heo --- include/linux/workqueue.h | 1 - kernel/workqueue.c | 8 -------- 2 files changed, 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index bc0cda180c8b..0c3301421c57 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -456,7 +456,6 @@ extern int schedule_on_each_cpu(work_func_t func); int execute_in_process_context(work_func_t fn, struct execute_work *); extern bool flush_work(struct work_struct *work); -extern bool cancel_work(struct work_struct *work); extern bool cancel_work_sync(struct work_struct *work); extern bool flush_delayed_work(struct delayed_work *dwork); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index ccd1080dd6e7..6ec6ba65127b 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3018,14 +3018,6 @@ static bool __cancel_work(struct work_struct *work, bool is_dwork) return ret; } -/* - * See cancel_delayed_work() - */ -bool cancel_work(struct work_struct *work) -{ - return __cancel_work(work, false); -} - /** * cancel_delayed_work - cancel a delayed work * @dwork: delayed_work to cancel -- cgit v1.2.3 From 4dcb31d4649df36297296b819437709f5407059c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 14 Mar 2018 09:04:16 -0700 Subject: net: use skb_to_full_sk() in skb_update_prio() Andrei Vagin reported a KASAN: slab-out-of-bounds error in skb_update_prio() Since SYNACK might be attached to a request socket, we need to get back to the listener socket. Since this listener is manipulated without locks, add const qualifiers to sock_cgroup_prioidx() so that the const can also be used in skb_update_prio() Also add the const qualifier to sock_cgroup_classid() for consistency. Fixes: ca6fb0651883 ("tcp: attach SYNACK messages to request sockets instead of listener") Signed-off-by: Eric Dumazet Reported-by: Andrei Vagin Signed-off-by: David S. Miller --- include/linux/cgroup-defs.h | 4 ++-- net/core/dev.c | 22 +++++++++++++++------- 2 files changed, 17 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 9f242b876fde..f8e76d01a5ad 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -755,13 +755,13 @@ struct sock_cgroup_data { * updaters and return part of the previous pointer as the prioidx or * classid. Such races are short-lived and the result isn't critical. */ -static inline u16 sock_cgroup_prioidx(struct sock_cgroup_data *skcd) +static inline u16 sock_cgroup_prioidx(const struct sock_cgroup_data *skcd) { /* fallback to 1 which is always the ID of the root cgroup */ return (skcd->is_data & 1) ? skcd->prioidx : 1; } -static inline u32 sock_cgroup_classid(struct sock_cgroup_data *skcd) +static inline u32 sock_cgroup_classid(const struct sock_cgroup_data *skcd) { /* fallback to 0 which is the unconfigured default classid */ return (skcd->is_data & 1) ? skcd->classid : 0; diff --git a/net/core/dev.c b/net/core/dev.c index 2cedf520cb28..12be20535714 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3278,15 +3278,23 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, #if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) static void skb_update_prio(struct sk_buff *skb) { - struct netprio_map *map = rcu_dereference_bh(skb->dev->priomap); + const struct netprio_map *map; + const struct sock *sk; + unsigned int prioidx; - if (!skb->priority && skb->sk && map) { - unsigned int prioidx = - sock_cgroup_prioidx(&skb->sk->sk_cgrp_data); + if (skb->priority) + return; + map = rcu_dereference_bh(skb->dev->priomap); + if (!map) + return; + sk = skb_to_full_sk(skb); + if (!sk) + return; - if (prioidx < map->priomap_len) - skb->priority = map->priomap[prioidx]; - } + prioidx = sock_cgroup_prioidx(&sk->sk_cgrp_data); + + if (prioidx < map->priomap_len) + skb->priority = map->priomap[prioidx]; } #else #define skb_update_prio(skb) -- cgit v1.2.3 From cbe7128c4b92e2004984f477fd38dfa81662f02e Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Tue, 13 Mar 2018 14:51:28 +0900 Subject: vlan: Fix out of order vlan headers with reorder header off With reorder header off, received packets are untagged in skb_vlan_untag() called from within __netif_receive_skb_core(), and later the tag will be inserted back in vlan_do_receive(). This caused out of order vlan headers when we create a vlan device on top of another vlan device, because vlan_do_receive() inserts a tag as the outermost vlan tag. E.g. the outer tag is first removed in skb_vlan_untag() and inserted back in vlan_do_receive(), then the inner tag is next removed and inserted back as the outermost tag. This patch fixes the behaviour by inserting the inner tag at the right position. Signed-off-by: Toshiaki Makita Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 66 ++++++++++++++++++++++++++++++++++++++++--------- net/8021q/vlan_core.c | 4 +-- 2 files changed, 57 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 5e6a2d4dc366..c4a1cff9c768 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -300,30 +300,34 @@ static inline bool vlan_hw_offload_capable(netdev_features_t features, } /** - * __vlan_insert_tag - regular VLAN tag inserting + * __vlan_insert_inner_tag - inner VLAN tag inserting * @skb: skbuff to tag * @vlan_proto: VLAN encapsulation protocol * @vlan_tci: VLAN TCI to insert + * @mac_len: MAC header length including outer vlan headers * - * Inserts the VLAN tag into @skb as part of the payload + * Inserts the VLAN tag into @skb as part of the payload at offset mac_len * Returns error if skb_cow_head failes. * * Does not change skb->protocol so this function can be used during receive. */ -static inline int __vlan_insert_tag(struct sk_buff *skb, - __be16 vlan_proto, u16 vlan_tci) +static inline int __vlan_insert_inner_tag(struct sk_buff *skb, + __be16 vlan_proto, u16 vlan_tci, + unsigned int mac_len) { struct vlan_ethhdr *veth; if (skb_cow_head(skb, VLAN_HLEN) < 0) return -ENOMEM; - veth = skb_push(skb, VLAN_HLEN); + skb_push(skb, VLAN_HLEN); - /* Move the mac addresses to the beginning of the new header. */ - memmove(skb->data, skb->data + VLAN_HLEN, 2 * ETH_ALEN); + /* Move the mac header sans proto to the beginning of the new header. */ + memmove(skb->data, skb->data + VLAN_HLEN, mac_len - ETH_TLEN); skb->mac_header -= VLAN_HLEN; + veth = (struct vlan_ethhdr *)(skb->data + mac_len - ETH_HLEN); + /* first, the ethernet type */ veth->h_vlan_proto = vlan_proto; @@ -334,12 +338,30 @@ static inline int __vlan_insert_tag(struct sk_buff *skb, } /** - * vlan_insert_tag - regular VLAN tag inserting + * __vlan_insert_tag - regular VLAN tag inserting * @skb: skbuff to tag * @vlan_proto: VLAN encapsulation protocol * @vlan_tci: VLAN TCI to insert * * Inserts the VLAN tag into @skb as part of the payload + * Returns error if skb_cow_head failes. + * + * Does not change skb->protocol so this function can be used during receive. + */ +static inline int __vlan_insert_tag(struct sk_buff *skb, + __be16 vlan_proto, u16 vlan_tci) +{ + return __vlan_insert_inner_tag(skb, vlan_proto, vlan_tci, ETH_HLEN); +} + +/** + * vlan_insert_inner_tag - inner VLAN tag inserting + * @skb: skbuff to tag + * @vlan_proto: VLAN encapsulation protocol + * @vlan_tci: VLAN TCI to insert + * @mac_len: MAC header length including outer vlan headers + * + * Inserts the VLAN tag into @skb as part of the payload at offset mac_len * Returns a VLAN tagged skb. If a new skb is created, @skb is freed. * * Following the skb_unshare() example, in case of error, the calling function @@ -347,12 +369,14 @@ static inline int __vlan_insert_tag(struct sk_buff *skb, * * Does not change skb->protocol so this function can be used during receive. */ -static inline struct sk_buff *vlan_insert_tag(struct sk_buff *skb, - __be16 vlan_proto, u16 vlan_tci) +static inline struct sk_buff *vlan_insert_inner_tag(struct sk_buff *skb, + __be16 vlan_proto, + u16 vlan_tci, + unsigned int mac_len) { int err; - err = __vlan_insert_tag(skb, vlan_proto, vlan_tci); + err = __vlan_insert_inner_tag(skb, vlan_proto, vlan_tci, mac_len); if (err) { dev_kfree_skb_any(skb); return NULL; @@ -360,6 +384,26 @@ static inline struct sk_buff *vlan_insert_tag(struct sk_buff *skb, return skb; } +/** + * vlan_insert_tag - regular VLAN tag inserting + * @skb: skbuff to tag + * @vlan_proto: VLAN encapsulation protocol + * @vlan_tci: VLAN TCI to insert + * + * Inserts the VLAN tag into @skb as part of the payload + * Returns a VLAN tagged skb. If a new skb is created, @skb is freed. + * + * Following the skb_unshare() example, in case of error, the calling function + * doesn't have to worry about freeing the original skb. + * + * Does not change skb->protocol so this function can be used during receive. + */ +static inline struct sk_buff *vlan_insert_tag(struct sk_buff *skb, + __be16 vlan_proto, u16 vlan_tci) +{ + return vlan_insert_inner_tag(skb, vlan_proto, vlan_tci, ETH_HLEN); +} + /** * vlan_insert_tag_set_proto - regular VLAN tag inserting * @skb: skbuff to tag diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index 64aa9f755e1d..45c9bf5ff3a0 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -48,8 +48,8 @@ bool vlan_do_receive(struct sk_buff **skbp) * original position later */ skb_push(skb, offset); - skb = *skbp = vlan_insert_tag(skb, skb->vlan_proto, - skb->vlan_tci); + skb = *skbp = vlan_insert_inner_tag(skb, skb->vlan_proto, + skb->vlan_tci, skb->mac_len); if (!skb) return false; skb_pull(skb, offset + VLAN_HLEN); -- cgit v1.2.3 From b3a5d111994450909158929560906f2c1c6c1d85 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 14 Mar 2018 12:45:12 -0700 Subject: percpu_ref: Update doc to dissuade users from depending on internal RCU grace periods percpu_ref internally uses sched-RCU to implement the percpu -> atomic mode switching and the documentation suggested that this could be depended upon. This doesn't seem like a good idea. * percpu_ref uses sched-RCU which has different grace periods regular RCU. Users may combine percpu_ref with regular RCU usage and incorrectly believe that regular RCU grace periods are performed by percpu_ref. This can lead to, for example, use-after-free due to premature freeing. * percpu_ref has a grace period when switching from percpu to atomic mode. It doesn't have one between the last put and release. This distinction is subtle and can lead to surprising bugs. * percpu_ref allows starting in and switching to atomic mode manually for debugging and other purposes. This means that there may not be any grace periods from kill to release. This patch makes it clear that the grace periods are percpu_ref's internal implementation detail and can't be depended upon by the users. Signed-off-by: Tejun Heo Cc: Kent Overstreet Cc: Linus Torvalds Signed-off-by: Tejun Heo --- include/linux/percpu-refcount.h | 18 ++++++++++++------ lib/percpu-refcount.c | 2 ++ 2 files changed, 14 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/percpu-refcount.h b/include/linux/percpu-refcount.h index 864d167a1073..009cdf3d65b6 100644 --- a/include/linux/percpu-refcount.h +++ b/include/linux/percpu-refcount.h @@ -30,10 +30,14 @@ * calls io_destroy() or the process exits. * * In the aio code, kill_ioctx() is called when we wish to destroy a kioctx; it - * calls percpu_ref_kill(), then hlist_del_rcu() and synchronize_rcu() to remove - * the kioctx from the proccess's list of kioctxs - after that, there can't be - * any new users of the kioctx (from lookup_ioctx()) and it's then safe to drop - * the initial ref with percpu_ref_put(). + * removes the kioctx from the proccess's table of kioctxs and kills percpu_ref. + * After that, there can't be any new users of the kioctx (from lookup_ioctx()) + * and it's then safe to drop the initial ref with percpu_ref_put(). + * + * Note that the free path, free_ioctx(), needs to go through explicit call_rcu() + * to synchronize with RCU protected lookup_ioctx(). percpu_ref operations don't + * imply RCU grace periods of any kind and if a user wants to combine percpu_ref + * with RCU protection, it must be done explicitly. * * Code that does a two stage shutdown like this often needs some kind of * explicit synchronization to ensure the initial refcount can only be dropped @@ -113,8 +117,10 @@ void percpu_ref_reinit(struct percpu_ref *ref); * Must be used to drop the initial ref on a percpu refcount; must be called * precisely once before shutdown. * - * Puts @ref in non percpu mode, then does a call_rcu() before gathering up the - * percpu counters and dropping the initial ref. + * Switches @ref into atomic mode before gathering up the percpu counters + * and dropping the initial ref. + * + * There are no implied RCU grace periods between kill and release. */ static inline void percpu_ref_kill(struct percpu_ref *ref) { diff --git a/lib/percpu-refcount.c b/lib/percpu-refcount.c index 30e7dd88148b..9f96fa7bc000 100644 --- a/lib/percpu-refcount.c +++ b/lib/percpu-refcount.c @@ -322,6 +322,8 @@ EXPORT_SYMBOL_GPL(percpu_ref_switch_to_percpu); * This function normally doesn't block and can be called from any context * but it may block if @confirm_kill is specified and @ref is in the * process of switching to atomic mode by percpu_ref_switch_to_atomic(). + * + * There are no implied RCU grace periods between kill and release. */ void percpu_ref_kill_and_confirm(struct percpu_ref *ref, percpu_ref_func_t *confirm_kill) -- cgit v1.2.3 From 578ae447e7e5d78c90ac40a06406c1741f79ba96 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Mon, 19 Mar 2018 13:18:57 -0500 Subject: jump_label: Disable jump labels in __exit code With the following commit: 333522447063 ("jump_label: Explicitly disable jump labels in __init code") ... we explicitly disabled jump labels in __init code, so they could be detected and not warned about in the following commit: dc1dd184c2f0 ("jump_label: Warn on failed jump_label patching attempt") In-kernel __exit code has the same issue. It's never used, so it's freed along with the rest of initmem. But jump label entries in __exit code aren't explicitly disabled, so we get the following warning when enabling pr_debug() in __exit code: can't patch jump_label at dmi_sysfs_exit+0x0/0x2d WARNING: CPU: 0 PID: 22572 at kernel/jump_label.c:376 __jump_label_update+0x9d/0xb0 Fix the warning by disabling all jump labels in initmem (which includes both __init and __exit code). Reported-and-tested-by: Li Wang Signed-off-by: Josh Poimboeuf Cc: Borislav Petkov Cc: Jason Baron Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Fixes: dc1dd184c2f0 ("jump_label: Warn on failed jump_label patching attempt") Link: http://lkml.kernel.org/r/7121e6e595374f06616c505b6e690e275c0054d1.1521483452.git.jpoimboe@redhat.com Signed-off-by: Ingo Molnar --- include/linux/jump_label.h | 4 ++-- init/main.c | 2 +- kernel/jump_label.c | 7 ++++--- 3 files changed, 7 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 2168cc6b8b30..b46b541c67c4 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -151,7 +151,7 @@ extern struct jump_entry __start___jump_table[]; extern struct jump_entry __stop___jump_table[]; extern void jump_label_init(void); -extern void jump_label_invalidate_init(void); +extern void jump_label_invalidate_initmem(void); extern void jump_label_lock(void); extern void jump_label_unlock(void); extern void arch_jump_label_transform(struct jump_entry *entry, @@ -199,7 +199,7 @@ static __always_inline void jump_label_init(void) static_key_initialized = true; } -static inline void jump_label_invalidate_init(void) {} +static inline void jump_label_invalidate_initmem(void) {} static __always_inline bool static_key_false(struct static_key *key) { diff --git a/init/main.c b/init/main.c index 969eaf140ef0..21efbf6ace93 100644 --- a/init/main.c +++ b/init/main.c @@ -1001,7 +1001,7 @@ static int __ref kernel_init(void *unused) /* need to finish all async __init code before freeing the memory */ async_synchronize_full(); ftrace_free_init_mem(); - jump_label_invalidate_init(); + jump_label_invalidate_initmem(); free_initmem(); mark_readonly(); system_state = SYSTEM_RUNNING; diff --git a/kernel/jump_label.c b/kernel/jump_label.c index e7214093dcd1..01ebdf1f9f40 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -16,6 +16,7 @@ #include #include #include +#include #ifdef HAVE_JUMP_LABEL @@ -421,15 +422,15 @@ void __init jump_label_init(void) cpus_read_unlock(); } -/* Disable any jump label entries in __init code */ -void __init jump_label_invalidate_init(void) +/* Disable any jump label entries in __init/__exit code */ +void __init jump_label_invalidate_initmem(void) { struct jump_entry *iter_start = __start___jump_table; struct jump_entry *iter_stop = __stop___jump_table; struct jump_entry *iter; for (iter = iter_start; iter < iter_stop; iter++) { - if (init_kernel_text(iter->code)) + if (init_section_contains((void *)(unsigned long)iter->code, 1)) iter->code = 0; } } -- cgit v1.2.3 From 6b00c35138b404be98b85f4a703be594cbed501c Mon Sep 17 00:00:00 2001 From: Jagdish Gediya Date: Thu, 22 Mar 2018 01:08:10 +0530 Subject: mtd: nand: fsl_ifc: Read ECCSTAT0 and ECCSTAT1 registers for IFC 2.0 Due to missing information in Hardware manual, current implementation doesn't read ECCSTAT0 and ECCSTAT1 registers for IFC 2.0. Add support to read ECCSTAT0 and ECCSTAT1 registers during ecccheck for IFC 2.0. Fixes: 656441478ed5 ("mtd: nand: ifc: Fix location of eccstat registers for IFC V1.0") Cc: stable@vger.kernel.org # v3.18+ Signed-off-by: Jagdish Gediya Reviewed-by: Prabhakar Kushwaha Signed-off-by: Boris Brezillon --- drivers/mtd/nand/fsl_ifc_nand.c | 6 +----- include/linux/fsl_ifc.h | 6 +----- 2 files changed, 2 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c index fae01b04abc7..5a9c2f0020c2 100644 --- a/drivers/mtd/nand/fsl_ifc_nand.c +++ b/drivers/mtd/nand/fsl_ifc_nand.c @@ -227,11 +227,7 @@ static void fsl_ifc_run_command(struct mtd_info *mtd) int sector_end = sector_start + chip->ecc.steps - 1; __be32 *eccstat_regs; - if (ctrl->version >= FSL_IFC_VERSION_2_0_0) - eccstat_regs = ifc->ifc_nand.v2_nand_eccstat; - else - eccstat_regs = ifc->ifc_nand.v1_nand_eccstat; - + eccstat_regs = ifc->ifc_nand.nand_eccstat; eccstat = ifc_in32(&eccstat_regs[sector_start / 4]); for (i = sector_start; i <= sector_end; i++) { diff --git a/include/linux/fsl_ifc.h b/include/linux/fsl_ifc.h index c332f0a45607..3fdfede2f0f3 100644 --- a/include/linux/fsl_ifc.h +++ b/include/linux/fsl_ifc.h @@ -734,11 +734,7 @@ struct fsl_ifc_nand { u32 res19[0x10]; __be32 nand_fsr; u32 res20; - /* The V1 nand_eccstat is actually 4 words that overlaps the - * V2 nand_eccstat. - */ - __be32 v1_nand_eccstat[2]; - __be32 v2_nand_eccstat[6]; + __be32 nand_eccstat[8]; u32 res21[0x1c]; __be32 nanndcr; u32 res22[0x2]; -- cgit v1.2.3 From 5df7af85ecd88e8b5f1f31d6456c3cf38a8bbdda Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Tue, 20 Mar 2018 09:44:52 +0800 Subject: net: phy: Add general dummy stubs for MMD register access For some phy devices, even though they don't support the MMD extended register access, it does have some side effect if we are trying to read/write the MMD registers via indirect method. So introduce general dummy stubs for MMD register access which these devices can use to avoid such side effect. Fixes: b6b5e8a69118 ("gianfar: Disable EEE autoneg by default") Signed-off-by: Kevin Hao Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 17 +++++++++++++++++ include/linux/phy.h | 4 ++++ 2 files changed, 21 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index fe16f5894092..74664a6c0cdc 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -1673,6 +1673,23 @@ int genphy_config_init(struct phy_device *phydev) } EXPORT_SYMBOL(genphy_config_init); +/* This is used for the phy device which doesn't support the MMD extended + * register access, but it does have side effect when we are trying to access + * the MMD register via indirect method. + */ +int genphy_read_mmd_unsupported(struct phy_device *phdev, int devad, u16 regnum) +{ + return -EOPNOTSUPP; +} +EXPORT_SYMBOL(genphy_read_mmd_unsupported); + +int genphy_write_mmd_unsupported(struct phy_device *phdev, int devnum, + u16 regnum, u16 val) +{ + return -EOPNOTSUPP; +} +EXPORT_SYMBOL(genphy_write_mmd_unsupported); + int genphy_suspend(struct phy_device *phydev) { return phy_set_bits(phydev, MII_BMCR, BMCR_PDOWN); diff --git a/include/linux/phy.h b/include/linux/phy.h index b260fb336b25..7c4c2379e010 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -984,6 +984,10 @@ static inline int genphy_no_soft_reset(struct phy_device *phydev) { return 0; } +int genphy_read_mmd_unsupported(struct phy_device *phdev, int devad, + u16 regnum); +int genphy_write_mmd_unsupported(struct phy_device *phdev, int devnum, + u16 regnum, u16 val); /* Clause 45 PHY */ int genphy_c45_restart_aneg(struct phy_device *phydev); -- cgit v1.2.3 From f59f1caf72ba00d519c793c3deb32cd3be32edc2 Mon Sep 17 00:00:00 2001 From: Daniel Vacek Date: Thu, 22 Mar 2018 16:17:38 -0700 Subject: Revert "mm: page_alloc: skip over regions of invalid pfns where possible" This reverts commit b92df1de5d28 ("mm: page_alloc: skip over regions of invalid pfns where possible"). The commit is meant to be a boot init speed up skipping the loop in memmap_init_zone() for invalid pfns. But given some specific memory mapping on x86_64 (or more generally theoretically anywhere but on arm with CONFIG_HAVE_ARCH_PFN_VALID) the implementation also skips valid pfns which is plain wrong and causes 'kernel BUG at mm/page_alloc.c:1389!' crash> log | grep -e BUG -e RIP -e Call.Trace -e move_freepages_block -e rmqueue -e freelist -A1 kernel BUG at mm/page_alloc.c:1389! invalid opcode: 0000 [#1] SMP -- RIP: 0010: move_freepages+0x15e/0x160 -- Call Trace: move_freepages_block+0x73/0x80 __rmqueue+0x263/0x460 get_page_from_freelist+0x7e1/0x9e0 __alloc_pages_nodemask+0x176/0x420 -- crash> page_init_bug -v | grep RAM 1000 - 9bfff System RAM (620.00 KiB) 100000 - 430bffff System RAM ( 1.05 GiB = 1071.75 MiB = 1097472.00 KiB) 4b0c8000 - 4bf9cfff System RAM ( 14.83 MiB = 15188.00 KiB) 4bfac000 - 646b1fff System RAM (391.02 MiB = 400408.00 KiB) 7b788000 - 7b7fffff System RAM (480.00 KiB) 100000000 - 67fffffff System RAM ( 22.00 GiB) crash> page_init_bug | head -6 7b788000 - 7b7fffff System RAM (480.00 KiB) 1fffff00000000 0 1 DMA32 4096 1048575 505736 505344 505855 0 0 0 DMA 1 4095 1fffff00000400 0 1 DMA32 4096 1048575 BUG, zones differ! crash> kmem -p 77fff000 78000000 7b5ff000 7b600000 7b787000 7b788000 PAGE PHYSICAL MAPPING INDEX CNT FLAGS ffffea0001e00000 78000000 0 0 0 0 ffffea0001ed7fc0 7b5ff000 0 0 0 0 ffffea0001ed8000 7b600000 0 0 0 0 <<<< ffffea0001ede1c0 7b787000 0 0 0 0 ffffea0001ede200 7b788000 0 0 1 1fffff00000000 Link: http://lkml.kernel.org/r/20180316143855.29838-1-neelx@redhat.com Fixes: b92df1de5d28 ("mm: page_alloc: skip over regions of invalid pfns where possible") Signed-off-by: Daniel Vacek Acked-by: Ard Biesheuvel Acked-by: Michal Hocko Reviewed-by: Andrew Morton Cc: Vlastimil Babka Cc: Mel Gorman Cc: Pavel Tatashin Cc: Paul Burton Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memblock.h | 1 - mm/memblock.c | 28 ---------------------------- mm/page_alloc.c | 11 +---------- 3 files changed, 1 insertion(+), 39 deletions(-) (limited to 'include/linux') diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 8be5077efb5f..f92ea7783652 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -187,7 +187,6 @@ int memblock_search_pfn_nid(unsigned long pfn, unsigned long *start_pfn, unsigned long *end_pfn); void __next_mem_pfn_range(int *idx, int nid, unsigned long *out_start_pfn, unsigned long *out_end_pfn, int *out_nid); -unsigned long memblock_next_valid_pfn(unsigned long pfn, unsigned long max_pfn); /** * for_each_mem_pfn_range - early memory pfn range iterator diff --git a/mm/memblock.c b/mm/memblock.c index b6ba6b7adadc..48376bd33274 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -1101,34 +1101,6 @@ void __init_memblock __next_mem_pfn_range(int *idx, int nid, *out_nid = r->nid; } -unsigned long __init_memblock memblock_next_valid_pfn(unsigned long pfn, - unsigned long max_pfn) -{ - struct memblock_type *type = &memblock.memory; - unsigned int right = type->cnt; - unsigned int mid, left = 0; - phys_addr_t addr = PFN_PHYS(++pfn); - - do { - mid = (right + left) / 2; - - if (addr < type->regions[mid].base) - right = mid; - else if (addr >= (type->regions[mid].base + - type->regions[mid].size)) - left = mid + 1; - else { - /* addr is within the region, so pfn is valid */ - return pfn; - } - } while (left < right); - - if (right == type->cnt) - return -1UL; - else - return PHYS_PFN(type->regions[right].base); -} - /** * memblock_set_node - set node ID on memblock regions * @base: base of area to set node ID for diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 010dee0f089e..1741dd23e7c1 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5356,17 +5356,8 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, if (context != MEMMAP_EARLY) goto not_early; - if (!early_pfn_valid(pfn)) { -#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP - /* - * Skip to the pfn preceding the next valid one (or - * end_pfn), such that we hit a valid pfn (or end_pfn) - * on our next iteration of the loop. - */ - pfn = memblock_next_valid_pfn(pfn, end_pfn) - 1; -#endif + if (!early_pfn_valid(pfn)) continue; - } if (!early_pfn_in_nid(pfn, nid)) continue; if (!update_defer_init(pgdat, pfn, end_pfn, &nr_initialised)) -- cgit v1.2.3