summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2016-04-18netfilter: connlabels: change nf_connlabels_get bit arg to 'highest used'Florian Westphal5-8/+11
nf_connlabel_set() takes the bit number that we would like to set. nf_connlabels_get() however took the number of bits that we want to support. So e.g. nf_connlabels_get(32) support bits 0 to 31, but not 32. This changes nf_connlabels_get() to take the highest bit that we want to set. Callers then don't have to cope with a potential integer wrap when using nf_connlabels_get(bit + 1) anymore. Current callers are fine, this change is only to make folloup nft ct label set support simpler. Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
2016-04-18netfilter: labels: don't emit ct event if labels were not changedFlorian Westphal1-6/+10
make the replace function only send a ctnetlink event if the contents of the new set is different. Otherwise 'ct label set ct label | bar' will cause netlink event storm since we "replace" labels for each packet. Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
2016-04-18netfilter: connlabels: move helpers to xt_connlabelFlorian Westphal3-20/+12
Currently labels can only be set either by iptables connlabel match or via ctnetlink. Before adding nftables set support, clean up the clabel core and move helpers that nft will not need after all to the xtables module. Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
2016-04-18tun: don't require serialization lock on txPaolo Abeni1-1/+1
The current tun_net_xmit() implementation don't need any external lock since it relies on rcu protection for the tun data structure and on socket queue lock for skb queuing. This patch set the NETIF_F_LLTX feature bit in the tun device, so that on xmit, in absence of qdisc, no serialization lock is acquired by the caller. The user space can remove the default tun qdisc with: tc qdisc replace dev <tun device name> root noqueue Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18udp: fix if statement in SIOCINQ ioctlDan Carpenter1-6/+0
We deleted a line of code and accidentally made the "return put_user()" part of the if statement when it's supposed to be unconditional. Fixes: 9f9a45beaa96 ('udp: do not expect udp headers on ioctl SIOCINQ') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18Merge branch 'for-linus' of ↵Linus Torvalds6-4/+12
git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux Pull s390 fixes from Martin Schwidefsky: "A couple of bug fixes" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: s390: add CPU_BIG_ENDIAN config option s390/spinlock: avoid yield to non existent cpu s390/dcssblk: fix possible deadlock in remove vs. per-device attributes s390/seccomp: include generic seccomp header file s390/pci: add extra padding to function measurement block s390/scm_blk: fix deadlock for requests != REQ_TYPE_FS
2016-04-18rtnetlink: rtnl_fill_stats: avoid an unnecssary stats copyRoopa Prabhu1-15/+8
This patch passes netlink attr data ptr directly to dev_get_stats thus elimiating a stats copy. Suggested-by: David Miller <davem@davemloft.net> Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18omapfb: panel-sharp-ls037v7dw01: fix check of gpio_to_desc() return valueVladimir Zapolskiy1-8/+4
The change fixes a check of gpio_to_desc() return value, the function returns either a valid pointer to struct gpio_desc or NULL, this makes IS_ERR() check invalid and may lead to a NULL pointer dereference in runtime. Signed-off-by: Vladimir Zapolskiy <vz@mleia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2016-04-18video: ARM CLCD: runtime check for VersatileLinus Walleij1-7/+8
The current compile-time check for inversed IENB/CNTL does not work in multiplatform boots: as soon as versatile is included in the build, the IENB/CNTL is switched and breaks graphics. Convert this to a runtime switch. Cc: stable@vger.kernel.org Cc: Rob Herring <robh@kernel.org> Cc: Russell King <linux@arm.linux.org.uk> Fixes: a29da136de34 ("ARM: versatile: convert to multi-platform") Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
2016-04-18Linux 4.6-rc4Linus Torvalds1-1/+1
2016-04-18Merge branch 'dsa-mv88e6xxx-switch-factorization'David S. Miller10-305/+247
Vivien Didelot says: ==================== net: dsa: mv88e6xxx: factorize switch info This patchset factorizes the mv88e6xxx code by sharing a new extendable info structure to store static data such as switch family, product number, number of ports, number of databases and the name. The next step is to add a "flags" bitmap member to the info structure in order to simplify the shared code with a feature-based logic instead of checking their family/ID. This is a step forward having a single mv88e6xxx driver supporting many similar devices, like any usual Linux driver. Changes v3 -> v4: - constify probed name in DSA - rebase patchset above conflicting commit 48ace4e Changes v2 -> v3: - update commit messages and add Andrew's tags - keep the info lookup code in a separated function - split the single switch ID reading in probe in a new commit Changes v1 -> v2: - define PORT_SWITCH_ID_PROD_NUM_* values - use plain struct mv88e6xxx_info - remove non used yet ps->rev ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: remove switch ID from psVivien Didelot2-33/+0
ps->id is not needed anymore, so remove it as well as the related defined values. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: add number of db to infoVivien Didelot6-18/+19
Add the number of databases to the info structure. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: add number of ports to infoVivien Didelot6-56/+38
Drop the ps->num_ports variable in favor of a new member of the info structure. This removes the need to assign it at setup time. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: add family to infoVivien Didelot6-63/+38
Add an mv88e6xxx_family enum to the info structure for better family indentification. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: add switch infoVivien Didelot6-32/+93
Add a new switch info structure which is meant to store switch models static information, such as product number, name, number of ports, number of databases, etc. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: read switch ID in probeVivien Didelot1-27/+30
Read the switch ID only once, at probe time, to avoid multiple read accesses and MII bus checking. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: drop revision probingVivien Didelot5-42/+1
There is no point in having a special case for the revision when probing a switch model. The code gets cluttered with unnecessary defines, and leads to errors when code such as mv88e6131_setup compares PORT_SWITCH_ID_6131_B2 to ps->id which masks the revision. Drop every revision definition, and lookup only the product number. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: mv88e6xxx: drop double ds assignmentVivien Didelot4-8/+0
Every driver assigns ps->ds even though it gets assigned in the shared mv88e6xxx_setup_common function. Kill redundancy. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18net: dsa: constify probed nameVivien Didelot10-37/+39
Change the dsa_switch_driver.probe function to return a const char *. Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-18macsec: fix crypto Kconfig dependencyArnd Bergmann1-0/+1
The new MACsec driver uses the AES crypto algorithm, but can be configured even if CONFIG_CRYPTO is disabled, leading to a build error: warning: (MAC80211 && MACSEC) selects CRYPTO_GCM which has unmet direct dependencies (CRYPTO) warning: (BT && CEPH_LIB && INET && MAC802154 && MAC80211 && BLK_DEV_RBD && MACSEC && AIRO_CS && LIBIPW && HOSTAP && USB_WUSB && RTLLIB_CRYPTO_CCMP && FS_ENCRYPTION && EXT4_ENCRYPTION && CEPH_FS && BIG_KEYS && ENCRYPTED_KEYS) selects CRYPTO_AES which has unmet direct dependencies (CRYPTO) crypto/built-in.o: In function `gcm_enc_copy_hash': aes_generic.c:(.text+0x2b8): undefined reference to `crypto_xor' aes_generic.c:(.text+0x2dc): undefined reference to `scatterwalk_map_and_copy' This adds an explicit 'select CRYPTO' statement the way that other drivers handle it. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17Merge tag 'dm-4.6-fix-2' of ↵Linus Torvalds1-2/+2
git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm Pull device mapper fix from Mike Snitzer: "Fix for earlier 4.6-rc4 stable@ commit that introduced improper use of write lock in cmd_read_lock() -- due to cut-n-paste gone awry (and sparse didn't catch it)" * tag 'dm-4.6-fix-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: dm cache metadata: fix cmd_read_lock() acquiring write lock
2016-04-17dm cache metadata: fix cmd_read_lock() acquiring write lockAhmed Samy1-2/+2
Commit 9567366fefdd ("dm cache metadata: fix READ_LOCK macros and cleanup WRITE_LOCK macros") uses down_write() instead of down_read() in cmd_read_lock(), yet up_read() is used to release the lock in READ_UNLOCK(). Fix it. Fixes: 9567366fefdd ("dm cache metadata: fix READ_LOCK macros and cleanup WRITE_LOCK macros") Cc: stable@vger.kernel.org Signed-off-by: Ahmed Samy <f.fallen45@gmail.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com>
2016-04-17Merge tag 'char-misc-4.6-rc4' of ↵Linus Torvalds2-5/+9
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc Pull char/misc fixes from Greg KH: "Here are some small char/misc driver fixes for 4.6-rc4. Full details are in the shortlog, nothing major here. These have all been in linux-next for a while with no reported issues" * tag 'char-misc-4.6-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: lkdtm: do not leak free page on kmalloc failure lkdtm: fix memory leak of base lkdtm: fix memory leak of val extcon: palmas: Drop stray IRQF_EARLY_RESUME flag
2016-04-17Merge tag 'driver-core-4.6-rc4' of ↵Linus Torvalds2-13/+14
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core Pull misc fixes from Greg KH: "Here are three small fixes for 4.6-rc4. Two fix up some lz4 issues with big endian systems, and the remaining one resolves a minor debugfs issue that was reported. All have been in linux-next with no reported issues" * tag 'driver-core-4.6-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: lib: lz4: cleanup unaligned access efficiency detection lib: lz4: fixed zram with lz4 on big endian machines debugfs: Make automount point inodes permanently empty
2016-04-17Merge tag 'usb-4.6-rc4' of ↵Linus Torvalds15-13/+94
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb Pull USB driver fixes from Greg KH: "Here are some small USB fixes for 4.6-rc4. Mostly xhci fixes for reported issues, a UAS bug that has hit a number of people, including stable tree users, and a few other minor things. All have been in linux-next for a while with no reported issues" * tag 'usb-4.6-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: usb: hcd: out of bounds access in for_each_companion USB: uas: Add a new NO_REPORT_LUNS quirk USB: uas: Limit qdepth at the scsi-host level doc: usb: Fix typo in gadget_multi documentation usb: host: xhci-plat: Make enum xhci_plat_type start at a non zero value xhci: fix 10 second timeout on removal of PCI hotpluggable xhci controllers usb: xhci: fix wild pointers in xhci_mem_cleanup usb: host: xhci-plat: fix cannot work if R-Car Gen2/3 run on above 4GB phys usb: host: xhci: add a new quirk XHCI_NO_64BIT_SUPPORT xhci: resume USB 3 roothub first usb: xhci: applying XHCI_PME_STUCK_QUIRK to Intel BXT B0 host cdc-acm: fix crash if flushed with nothing buffered
2016-04-17Merge branch 'nfp-next'David S. Miller4-48/+176
Jakub Kicinski says ==================== nfp: cleanups and improvements Main purpose of this set is to get rid of doing potentially long mdelay()s but it also contains some trivial changes I've accumulated. First two patches fix harmless copy-paste errors, next two clean up the documentation and remove unused defines. Patch 5 clarifies the interpretation of RX descriptor fields. Patch 6, by far the biggest, adds ability to perform FW reconfig asynchronously thanks to which we can stop using mdelay(). ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17nfp: add async reconfiguration mechanismJakub Kicinski2-27/+157
Some callers of nfp_net_reconfig() are in atomic context so we used to busy wait for commands to complete. In worst case scenario that means locking up a core for up to 5 seconds when a command times out. Lets add a timer-based mechanism of asynchronously checking whether reconfiguration completed successfully for atomic callers to use. Non-atomic callers can now just sleep. The approach taken is quite simple because (1) synchronous reconfigurations always happen under RTNL (or before device is registered); (2) we can coalesce pending reconfigs. There is no need for request queues, timer which eventually takes a look at reconfiguration result to report errors is good enough. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17nfp: remove buggy RX buffer length validationJakub Kicinski1-12/+14
Meaning of data_len and meta_len RX WB descriptor fields is slightly confusing. Add a comment with a diagram clarifying the layout. Also remove the buffer length validation: (a) it's imprecise for static rx-offsets; (b) if firmware is buggy enough to DMA past the end of the buffer WARN_ON_ONCE() doesn't seem like a strong enough response. skb_put() will do the checking for us anyway. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17nfp: remove unused suspicious mask definesJakub Kicinski1-4/+0
NFP_NET_RXR_MASK sounds like a mask which could be used on NFP_NET_CFG_RXRS_ENABLE register but its value is quite strange. In fact there are no users of this define so let's just remove it. Same for TX rings. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17nfp: correct names of constants in commentsJakub Kicinski1-3/+3
Documentation in comments lacks CFG in some names. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17nfp: remove unnecessary staticJakub Kicinski1-1/+1
There is no reason for those local variables to be static. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17nfp: check the right pointer for errorsJakub Kicinski1-1/+1
Correct checking error condition on wrong pointer - copy/paste mistake most likely. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17net: bcmgenet: device stats are unsigned longEric Dumazet1-1/+5
On 64bit kernels, device stats are 64bit wide, not 32bit. Fixes: 1c1008c793fa4 ("net: bcmgenet: add main driver file") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17Merge branch 'IFF_NO_QUEUE-followups'David S. Miller2-3/+2
Phil Sutter says: ==================== Minor IFF_NO_QUEUE conversion follow-up The following series converts two further drivers away from setting 'tx_queue_len = 0' to adding IFF_NO_QUEUE to priv_flags instead. The first one, rtl8188eu in staging didn't exist back when all drivers were converted. The second one, openvswitch seems to have slipped through my grep'ing back then, no idea why. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17openvswitch: Convert to using IFF_NO_QUEUEPhil Sutter1-2/+1
Cc: Pravin Shelar <pshelar@nicira.com> Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17staging: rtl8188eu: Convert to using IFF_NO_QUEUEPhil Sutter1-1/+1
Cc: Jakub Sitnicki <jsitnicki@gmail.com> Signed-off-by: Phil Sutter <phil@nwl.cc> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17Merge branch 'fjes-next'David S. Miller3-30/+141
Taku Izumi says: ==================== FUJITSU Extended Socket driver version 1.1 This patchsets update FUJITSU Extended Socket network driver into version 1.1. This mainly includes some improvements and minor bugfix. v1->v2: - Remove ioctl and debugfs facility according to David comment ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17fjes: Update fjes driver version : 1.1Taku Izumi1-1/+1
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17fjes: Introduce spinlock for rx_statusTaku Izumi3-9/+72
This patch introduces spinlock of rx_status for proper excusive control. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17fjes: Enhance changing MTU related workTaku Izumi3-11/+58
This patch enhances the fjes_change_mtu() method by introducing new flag named FJES_RX_MTU_CHANGING_DONE in rx_status. At the same time, default MTU value is changed into 65510 bytes. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17fjes: fix bitwise check bug in fjes_raise_intr_rxdata_taskTaku Izumi1-1/+2
In fjes_raise_intr_rxdata_task(), there's a bug of bitwise check because of missing "& FJES_RX_POLL_WORK". This patch fixes this bug. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17fjes: fix incorrect statistics information in fjes_xmit_frame()Taku Izumi1-7/+7
There are bugs of acounting statistics in fjes_xmit_frame(). Accounting self stats is wrong. accounting stats of other EPs to be transmitted is right. This patch fixes this bug. Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17fjes: optimize timeout valueTaku Izumi1-3/+3
This patch optimizes the following timeout value. - FJES_DEVICE_RESET_TIMEOUT - FJES_COMMAND_REQ_TIMEOUT - FJES_COMMAND_REQ_BUFF_TIMEOUT Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17stmmac: socfpga: remove extra call to socfpga_dwmac_setupDinh Nguyen1-6/+0
In the socfpga_dwmac_probe function, we have a call to socfpga_dwmac_setup, which is already called from socfpga_dwmac_init later in the probe function. Remove this extra call to socfpga_dwmac_setup. Also we should not be calling socfpga_dwmac_setup() directly without wrapping it around the proper reset assert/deasserts. That is because the socfpga_dwmac_setup() is setting up PHY modes in the system manager, and it is requires the EMAC's to be in reset during the PHY setup. Reported-by: Matthew Gerlach <mgerlach@opensource.altera.com> Signed-off-by: Dinh Nguyen <dinguyen@opensource.altera.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17dsa: mv88e6xxx: Kill the REG_READ and REG_WRITE macrosAndrew Lunn6-123/+224
These macros hide a ds variable and a return statement on error, which can lead to locking issues. Kill them off. Signed-off-by: Andrew Lunn <andrew@lunn.ch> Tested-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17netdev_features: Add NETIF_F_TSO_MANGLEID to NETIF_F_ALL_TSOAlexander Duyck1-1/+2
I realized that when I added NETIF_F_TSO_MANGLEID as a TSO type I forgot to add it to NETIF_F_ALL_TSO. This patch corrects that so the flag will be included correctly. The result should be minor as it was only used by a few drivers and in a few specific cases such as when NETIF_F_SG was not supported on a device so the TSO flags were cleared. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17Merge branch 'ipv6-gre-offloads'David S. Miller12-98/+116
Alexander Duyck says: ==================== Add support for offloads with IPv6 GRE tunnels This patch series enables the use of segmentation and checksum offloads with IPv6 based GRE tunnels. In order to enable this series I had to make a change to iptunnel_handle_offloads so that it would no longer free the skb. This was necessary as there were multiple paths in the IPv6 GRE code that required the skb to still be present so it could be freed. As it turned out I believe this actually fixes a bug that was present in FOU/GUE based tunnels anyway. Below is a quick breakdown of the performance gains seen with a simple netperf test passing traffic through a ip6gretap tunnel and then an i40e interface: Throughput Throughput Local Local Result Units CPU Service Tag Util Demand % 3544.93 10^6bits/s 6.30 4.656 "before" 13081.75 10^6bits/s 3.75 0.752 "after" ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17ip6gre: Add support for GSOAlexander Duyck1-23/+33
This patch adds code borrowed from bits and pieces of other protocols to the IPv6 GRE path so that we can support GSO over IPv6 based GRE tunnels. By adding this support we are able to significantly improve the throughput for GRE tunnels as we are able to make use of GSO. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-17GRE: Add support for GRO/GSO of IPv6 GRE trafficAlexander Duyck1-1/+13
Since GRE doesn't really care about L3 protocol we can support IPv4 and IPv6 using the same offloads. With that being the case we can add a call to register the offloads for IPv6 as a part of our GRE offload initialization. Signed-off-by: Alexander Duyck <aduyck@mirantis.com> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-28 02:33:53 +0300