)]}' { "log": [ { "commit": "4214de33eca9331b4ec147e1b112d1a3e0876584", "tree": "2d020fe8d6e981ec0eb8eb7ee059e68ed3357dee", "parents": [ "734a7dc997ced8c8294fe42c5345c34f112a7e5f" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Jun 12 17:01:52 2015 -0500" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Oct 09 03:51:51 2015 -0500" }, "message": "ipv6: Pass struct net into nf_ct_frag6_gather\n\nThe function nf_ct_frag6_gather is called on both the input and the\noutput paths of the networking stack. In particular ipv6_defrag which\ncalls nf_ct_frag6_gather is called from both the the PRE_ROUTING chain\non input and the LOCAL_OUT chain on output.\n\nThe addition of a net parameter makes it explicit which network\nnamespace the packets are being reassembled in, and removes the need\nfor nf_ct_frag6_gather to guess.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n" }, { "commit": "734a7dc997ced8c8294fe42c5345c34f112a7e5f", "tree": "ddb182a5b4cd79cff47441514bd217b0ac3253ee", "parents": [ "7984613e8ad8493c3ab69a41f35eccc3ccc4e528" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Jun 12 16:54:29 2015 -0500" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Oct 09 02:50:49 2015 -0500" }, "message": "ipv4: Pass struct net into ip_defrag and ip_check_defrag\n\nThe function ip_defrag is called on both the input and the output\npaths of the networking stack. In particular conntrack when it is\ntracking outbound packets from the local machine calls ip_defrag.\n\nSo add a struct net parameter and stop making ip_defrag guess which\nnetwork namespace it needs to defragment packets in.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n" }, { "commit": "7984613e8ad8493c3ab69a41f35eccc3ccc4e528", "tree": "eb94065fba447614a69c095f9aa3c9d781ebf53d", "parents": [ "df718423250c000ca4323a767cedc2f3219b685c" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Tue Sep 15 21:39:09 2015 -0500" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Fri Oct 09 02:49:20 2015 -0500" }, "message": "ipv4: Only compute net once in ip_call_ra_chain\n\nip_call_ra_chain is called early in the forwarding chain from\nip_forward and ip_mr_input, which makes skb-\u003edev the correct\nexpression to get the input network device and dev_net(skb-\u003edev) a\ncorrect expression for the network namespace the packet is being\nprocessed in.\n\nCompute the network namespace and store it in a variable to make the\ncode clearer.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n" }, { "commit": "df718423250c000ca4323a767cedc2f3219b685c", "tree": "3afa13e8acbdf49a9f8c1a7993065c3836997a01", "parents": [ "28335a7445202a3d118145a07d9138e9881ebe18", "3ad0040573b0c00f88488bc31958acd07a55ee2e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:44 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:44 2015 -0700" }, "message": "Merge branch \u0027bpf_random32\u0027\n\nDaniel Borkmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nBPF/random32 updates\n\nBPF update to split the prandom state apart, and to move the\n*once helpers to the core. For details, please see individual\npatches. Given the changes and since it\u0027s in the tree for\nquite some time, net-next is a better choice in our opinion.\n\nv1 -\u003e v2:\n - Make DO_ONCE() type-safe, remove the kvec helper. Credits\n go to Alexei Starovoitov for the __VA_ARGS__ hint, thanks!\n - Add a comment to the DO_ONCE() helper as suggested by Alexei.\n - Rework prandom_init_once() helper to the new API.\n - Keep Alexei\u0027s Acked-by on the last patch.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3ad0040573b0c00f88488bc31958acd07a55ee2e", "tree": "3afa13e8acbdf49a9f8c1a7993065c3836997a01", "parents": [ "897ece56e714a2cc64e6914cb89a362d7021b36e" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Thu Oct 08 01:20:39 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:39 2015 -0700" }, "message": "bpf: split state from prandom_u32() and consolidate {c, e}BPF prngs\n\nWhile recently arguing on a seccomp discussion that raw prandom_u32()\naccess shouldn\u0027t be exposed to unpriviledged user space, I forgot the\nfact that SKF_AD_RANDOM extension actually already does it for some time\nin cBPF via commit 4cd3675ebf74 (\"filter: added BPF random opcode\").\n\nSince prandom_u32() is being used in a lot of critical networking code,\nlets be more conservative and split their states. Furthermore, consolidate\neBPF and cBPF prandom handlers to use the new internal PRNG. For eBPF,\nbpf_get_prandom_u32() was only accessible for priviledged users, but\nshould that change one day, we also don\u0027t want to leak raw sequences\nthrough things like eBPF maps.\n\nOne thought was also to have own per bpf_prog states, but due to ABI\nreasons this is not easily possible, i.e. the program code currently\ncannot access bpf_prog itself, and copying the rnd_state to/from the\nstack scratch space whenever a program uses the prng seems not really\nworth the trouble and seems too hacky. If needed, taus113 could in such\ncases be implemented within eBPF using a map entry to keep the state\nspace, or get_random_bytes() could become a second helper in cases where\nperformance would not be critical.\n\nBoth sides can trigger a one-time late init via prandom_init_once() on\nthe shared state. Performance-wise, there should even be a tiny gain\nas bpf_user_rnd_u32() saves one function call. The PRNG needs to live\ninside the BPF core since kernels could have a NET-less config as well.\n\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nAcked-by: Alexei Starovoitov \u003cast@plumgrid.com\u003e\nCc: Chema Gonzalez \u003cchema@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "897ece56e714a2cc64e6914cb89a362d7021b36e", "tree": "c965cddb6a87f0bc6d96cceaa36efe42af8158da", "parents": [ "0dd50d1b0c003ab4f17597fe1198bb57a2fadc06" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Thu Oct 08 01:20:38 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:38 2015 -0700" }, "message": "random32: add prandom_init_once helper for own rngs\n\nAdd a prandom_init_once() facility that works on the rnd_state, so that\nusers that are keeping their own state independent from prandom_u32() can\ninitialize their taus113 per cpu states.\n\nThe motivation here is similar to net_get_random_once(): initialize the\nstate as late as possible in the hope that enough entropy has been\ncollected for the seeding. prandom_init_once() makes use of the recently\nintroduced prandom_seed_full_state() helper and is generic enough so that\nit could also be used on fast-paths due to the DO_ONCE().\n\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0dd50d1b0c003ab4f17597fe1198bb57a2fadc06", "tree": "e60fd71c1309bbad3c3d213555731b47b9967741", "parents": [ "c90aeb948222a7b3d3391d232ec4f50fd8322ad3" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Thu Oct 08 01:20:37 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:37 2015 -0700" }, "message": "random32: add prandom_seed_full_state helper\n\nFactor out the full reseed handling code that populates the state\nthrough get_random_bytes() and runs prandom_warmup(). The resulting\nprandom_seed_full_state() will be used later on in more than the\ncurrent __prandom_reseed() user. Fix also two minor whitespace\nissues along the way.\n\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c90aeb948222a7b3d3391d232ec4f50fd8322ad3", "tree": "c4625885c7705076fb157b6394fa3969b44bd9f3", "parents": [ "46234253b9363894a254844a6550b4cc5f3edfe8" ], "author": { "name": "Hannes Frederic Sowa", "email": "hannes@stressinduktion.org", "time": "Thu Oct 08 01:20:36 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:36 2015 -0700" }, "message": "once: make helper generic for calling functions once\n\nMake the get_random_once() helper generic enough, so that functions\nin general would only be called once, where one user of this is then\nnet_get_random_once().\n\nThe only implementation specific call is to get_random_bytes(), all\nthe rest of this *_once() facility would be duplicated among different\nsubsystems otherwise. The new DO_ONCE() helper will be used by prandom()\nlater on, but might also be useful for other scenarios/subsystems as\nwell where a one-time initialization in often-called, possibly fast\npath code could occur.\n\nSigned-off-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "46234253b9363894a254844a6550b4cc5f3edfe8", "tree": "ab3b795f7ff1765d2f7dbb318b5fae08945a0bdf", "parents": [ "28335a7445202a3d118145a07d9138e9881ebe18" ], "author": { "name": "Hannes Frederic Sowa", "email": "hannes@stressinduktion.org", "time": "Thu Oct 08 01:20:35 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:26:35 2015 -0700" }, "message": "net: move net_get_random_once to lib\n\nThere\u0027s no good reason why users outside of networking should not\nbe using this facility, f.e. for initializing their seeds.\n\nTherefore, make it accessible from there as get_random_once().\n\nSigned-off-by: Hannes Frederic Sowa \u003channes@stressinduktion.org\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "28335a7445202a3d118145a07d9138e9881ebe18", "tree": "74e8f59a3c75022504fa5e73643af85c8582c58a", "parents": [ "cfc81b50387086c3a1ca6d2be3c46561edfbc3b5" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Wed Oct 07 08:40:13 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:18:47 2015 -0700" }, "message": "net: Do not drop to make_route if oif is l3mdev\n\nCommit deaa0a6a930 (\"net: Lookup actual route when oif is VRF device\")\nexposed a bug in __ip_route_output_key_hash for VRF devices: on FIB lookup\nfailure if the oif is specified the current logic drops to make_route on\nthe assumption that the route tables are wrong. For VRF/L3 master devices\nthis leads to wrong dst entries and route lookups. For example:\n $ ip route ls table vrf-red\n unreachable default\n broadcast 10.2.1.0 dev eth1 proto kernel scope link src 10.2.1.2\n 10.2.1.0/24 dev eth1 proto kernel scope link src 10.2.1.2\n local 10.2.1.2 dev eth1 proto kernel scope host src 10.2.1.2\n broadcast 10.2.1.255 dev eth1 proto kernel scope link src 10.2.1.2\n\n $ ip route get oif vrf-red 1.1.1.1\n 1.1.1.1 dev vrf-red src 10.0.0.2\n cache\n\nWith this patch:\n $ ip route get oif vrf-red 1.1.1.1\n RTNETLINK answers: No route to host\n\nwhich is the correct response based on the default route\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cfc81b50387086c3a1ca6d2be3c46561edfbc3b5", "tree": "4990bcfc30708d064da3d3898aef295ce3041f62", "parents": [ "dfdd7230c5a2d9d675b4e7d6e111a8ead13cfb11" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Wed Oct 07 10:16:09 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 05:03:08 2015 -0700" }, "message": "bpf, skb_do_redirect: clear sender_cpu before xmit\n\nSimilar to commit c29390c6dfee (\"xps: must clear sender_cpu before\nforwarding\"), we also need to clear the skb-\u003esender_cpu when moving\nfrom RX to TX via skb_do_redirect() due to the shared location of\nnapi_id (used on RX) and sender_cpu (used on TX).\n\nFixes: 27b29f63058d (\"bpf: add bpf_redirect() helper\")\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@plumgrid.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "dfdd7230c5a2d9d675b4e7d6e111a8ead13cfb11", "tree": "cc0754c9308eb28faa0022819240d2434af0c841", "parents": [ "d70e53262f5cfc0b88a211ef9cbd1ca5256f4a6b" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Tue Oct 06 23:53:57 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:54:15 2015 -0700" }, "message": "net: hns: fix 32-bit build warning\n\nThe recently added hns driver causes a build warning in ARM\nallmodconfig builds:\n\ndrivers/net/ethernet/hisilicon/hns/hnae.c: In function \u0027handles_show\u0027:\ndrivers/net/ethernet/hisilicon/hns/hnae.c:452:13: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]\n j, (u64)h-\u003eqs[i]-\u003eio_base);\n ^\n\nThis removes the pointless cast and prints the pointer address using\nthe \"%p\" format string in all three locations.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d70e53262f5cfc0b88a211ef9cbd1ca5256f4a6b", "tree": "f617ca5dba505ee3b84d5bde7b1a7ad3cfcf6892", "parents": [ "494f8eb9b616590c78bb1974aed835a842f012b6" ], "author": { "name": "Jon Ringle", "email": "jringle@gridpoint.com", "time": "Tue Oct 06 16:37:46 2015 -0400" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:49:55 2015 -0700" }, "message": "net: Microchip encx24j600 driver\n\nThis ethernet driver supports the Micorchip enc424j600/626j600 Ethernet\ncontroller over a SPI bus interface. This driver makes use of the regmap API to\noptimize access to registers by caching registers where possible.\n\nDatasheet:\nhttp://ww1.microchip.com/downloads/en/DeviceDoc/39935b.pdf\n\nSigned-off-by: Jon Ringle \u003cjringle@gridpoint.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "494f8eb9b616590c78bb1974aed835a842f012b6", "tree": "cd3753e0b39d3072dc87e55a53d5f37b2dd39477", "parents": [ "91d2f14bc38336177c2d65e0c2cbf6447133cee1", "9200c27a1cc7ac719e4c43411e5a170144ba0ac5" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:46:03 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:46:03 2015 -0700" }, "message": "Merge branch \u0027broadcom-iproc\u0027\n\nArun Parameswaran says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nAdd support for Broadcom\u0027s iProc MDIO and Cygnus Ethernet PHY\n\nThis patchset adds support for the iProc MDIO interface and the\nBroadcom Cygnus SoC\u0027s internal Ethernet PHY.\n\nThe internal Ethernet PHY(s) in the Cygnus SoC\u0027s are accessed\nvia the MDIO interface found in most of the iProc based chips.\n\nThe patch also consolidates the common API\u0027s used by the\nBroadcom phys to a common library. Existing Broadcom phy\ndrivers have been modified to use the common library API\u0027s.\n\nThis patch series is based on Linux v4.3-rc1 and is avaliable in:\nhttps://github.com/Broadcom/cygnus-linux/tree/cygnus-net-phy-mdio-v3\n\nThe Ethernet driver for the iProc family will be submitted soon,\nas will the device tree configurations for the different iProc\nfamily SoCs.\n\nChanges from v2:\n- Modified drivers/net/phy/Kconfig to modify the BCM_CYGNUS_PHY\n driver to \u0027depends on MDIO_BCM_IPROC\u0027 instead of \u0027select\u0027.\n- Added github branch to the cover letter\n\nChanges from v1:\n- Updated device tree documentation for the iProc MDIO driver\n based on Florian\u0027s feedback.\n- Moved the core register defines from the Cygnus PHY driver to\n \u0027include/linux/brcmphy.h\u0027 based on Florian\u0027s feedback.\n- Created a new patch/commit to modify the bcm7xxx phy driver\n to use the new core register defines.\n- Modified the Kconfig entry for the Broadcom PHY library to\n \u0027tristate\u0027 instead of \u0027bool\u0027\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9200c27a1cc7ac719e4c43411e5a170144ba0ac5", "tree": "cd3753e0b39d3072dc87e55a53d5f37b2dd39477", "parents": [ "8e185d6997bb67068f0ca8f062a50caa2608cf1b" ], "author": { "name": "Arun Parameswaran", "email": "arunp@broadcom.com", "time": "Tue Oct 06 12:25:50 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:45:53 2015 -0700" }, "message": "net: phy: bcm7xxx: Modified to use global core register defines\n\nModified the bcm7xxx phy driver to remove local core register\ndefines and use the common ones from \"include/linux/brcmphy.h\"\n\nSigned-off-by: Arun Parameswaran \u003carunp@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8e185d6997bb67068f0ca8f062a50caa2608cf1b", "tree": "b2d42099f3777dc770023566cfd8d1dc3f1cc761", "parents": [ "a1cba5613edf50c2a213fa90c30aa10500b241b7" ], "author": { "name": "Arun Parameswaran", "email": "arunp@broadcom.com", "time": "Tue Oct 06 12:25:49 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:45:52 2015 -0700" }, "message": "net: phy: Broadcom Cygnus internal Etherent PHY driver\n\nAdd support for the Broadcom Cygnus SoCs internal PHY\u0027s.\nThe PHYs are 1000M/100M/10M capable with support for \u0027EEE\u0027\nand \u0027APD\u0027 (Auto Power Down).\n\nThis driver supports the following Broadcom Cygnus SoCs:\n - BCM583XX (BCM58300, BCM58302, BCM58303, BCM58305)\n - BCM113XX (BCM11300, BCM11320, BCM11350, BCM11360)\n\nThe PHY\u0027s on these SoC\u0027s require some workarounds for\nstable operation, both during configuration time and\nduring suspend/resume. This driver handles the\napplication of the workarounds.\n\nSigned-off-by: Arun Parameswaran \u003carunp@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a1cba5613edf50c2a213fa90c30aa10500b241b7", "tree": "b015ddfca52bdcb0766dbc2a6abbaf5a3417c851", "parents": [ "ddc24ae1fd6ae2638365b0b824750b5b2ef13ca5" ], "author": { "name": "Arun Parameswaran", "email": "arunp@broadcom.com", "time": "Tue Oct 06 12:25:48 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:45:46 2015 -0700" }, "message": "net: phy: Add Broadcom phy library for common interfaces\n\nThis patch adds the Broadcom phy library to consolidate common\ninterfaces shared by Broadcom phy\u0027s.\n\nMoved the common interfaces to the \u0027bcm-phy-lib.c\u0027 and updated\nthe Broadcom PHY drivers to use the new APIs.\n\nSigned-off-by: Arun Parameswaran \u003carunp@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ddc24ae1fd6ae2638365b0b824750b5b2ef13ca5", "tree": "1525c486cbf183cf234f4c95c0c29143901b64ca", "parents": [ "bb257c3813613ce11329568c21c200e95afe6fc1" ], "author": { "name": "Arun Parameswaran", "email": "arunp@broadcom.com", "time": "Tue Oct 06 12:25:47 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:44:46 2015 -0700" }, "message": "net: phy: Broadcom iProc MDIO bus driver\n\nThis patch adds support for the Broadcom iProc MDIO bus interface.\nThe MDIO interface can be found in the Broadcom iProc family Soc\u0027s.\n\nThe MDIO bus is accessed using a combination of command and data\nregisters. This MDIO driver provides access to the Etherent GPHY\u0027s\nconnected to the MDIO bus.\n\nSigned-off-by: Arun Parameswaran \u003carunp@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bb257c3813613ce11329568c21c200e95afe6fc1", "tree": "f6507cb98810492426a58ede895ccae55b64379f", "parents": [ "91d2f14bc38336177c2d65e0c2cbf6447133cee1" ], "author": { "name": "Arun Parameswaran", "email": "arunp@broadcom.com", "time": "Tue Oct 06 12:25:46 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:44:44 2015 -0700" }, "message": "dt-bindings: net: Broadcom iProc MDIO bus driver device tree binding\n\nAdd device tree binding documentation for the Broadcom iProc MDIO\nbus driver.\n\nSigned-off-by: Arun Parameswaran \u003carunp@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "91d2f14bc38336177c2d65e0c2cbf6447133cee1", "tree": "9f27eb7f95fb3a839a48465f5e49f7a9bcef1912", "parents": [ "6b92d0c4a6eefd32da7ecaf6873cb7e7e7602cfa", "06766513232d1619ac84e87b1d839d3fcc23a540" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:38:37 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:38:37 2015 -0700" }, "message": "Merge branch \u0027net/rds/4.3-v3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ssantosh/linux\n\nSantosh Shilimkar says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nRDS: connection scalability and performance improvements\n\n[v4]\nRe-sending the same patches from v3 again since my repost of\npatch 05/14 from v3 was whitespace damaged.\n\n[v3]\nUpdated patch \"[PATCH v2 05/14] RDS: defer the over_batch work to\nsend worker\" as per David Miller\u0027s comment [4] to avoid the magic\nvalue usage. Patch now makes use of already available but unused\nsend_batch_count module parameter. Rest of the patches are same as\nearlier version v2 [3]\n\n[v2]:\nDropped \"[PATCH 05/15] RDS: increase size of hash-table to 8K\" from\nearlier version [1]. I plan to address the hash table scalability using\nre-sizable hash tables as suggested by David Laight and David Miller [2]\n\nThis series addresses RDS connection bottlenecks on massive workloads and\nimprove the RDMA performance almost by 3X. RDS TCP also gets a small gain\nof about 12%.\n\nRDS is being used in massive systems with high scalability where several\nhundred thousand end points and tens of thousands of local processes\nare operating in tens of thousand sockets. Being RC(reliable connection),\nsocket bind and release happens very often and any inefficiencies in\nbind hash look ups hurts the overall system performance. RDS bin hash-table\nuses global spin-lock which is the biggest bottleneck. To make matter worst,\nit uses rcu inside global lock for hash buckets.\nThis is being addressed by simply using per bucket rw lock which makes the\nlocking simple and very efficient. The hash table size is still an issue and\nI plan to address it by using re-sizable hash tables as suggested on the list.\n\nFor RDS RDMA improvement, the completion handling is revamped so that we\ncan do batch completions. Both send and receive completion handlers are\nsplit logically to achieve the same. RDS 8K messages being one of the\nkey usecase, mr pool is adapted to have the 8K mrs along with default 1M\nmrs. And while doing this, few fixes and couple of bottlenecks seen with\nrds_sendmsg() are addressed.\n\nSeries applies against 4.3-rc1 as well net-next. Its tested on Oracle\nhardware with IB fabric for both bcopy as well as RDMA mode. RDS TCP is\ntested with iXGB NIC. Like last time, iWARP transport is untested with\nthese changes. The patchset is also available at below git repo:\n\ngit://git.kernel.org/pub/scm/linux/kernel/git/ssantosh/linux.git net/rds/4.3-v3\n\nAs a side note, the IB HCA driver I used for testing misses at least 3\nimportant patches in upstream to see the full blown IB performance and\nam hoping to get that in mainline with help of them.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6b92d0c4a6eefd32da7ecaf6873cb7e7e7602cfa", "tree": "d6d15cbeb675b52ef22890a2228b5301b1396c45", "parents": [ "e28383dd28afafefb0e445354c88f7b1e99aae9d", "ede2059dbaf9c6557a49d466c8c7778343b208ff" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:13 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:13 2015 -0700" }, "message": "Merge branch \u0027pass_net_through_output_path\u0027\n\nEric W. Biederman says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: Pass net through the output path v2\n\nThis is the next installment of my work to pass struct net through the\noutput path so the code does not need to guess how to figure out which\nnetwork namespace it is in, and ultimately routes can have output\ndevices in another network namespace.\n\nThe first patch in this series is a fix for a bug that came in when sk\nwas passed through the functions in the output path, and as such is\nprobably a candidate for net. At the same time my later patches depend\non it so sending the fix separately would be confusing.\n\nThe second patch in this series is another fix that for an issue that\ncame in when sk was passed through the output path. I don\u0027t think it\nneeds a backport as I don\u0027t think anyone uses the path where the code\nwas incorrect.\n\nThe rest of the patchset focuses on the path from xxx_local_out to\ndst_output and in the end succeeds in passing sock_net(sk) from the\nsocket a packet locally originates on to the dst-\u003eoutput function.\n\nGiven the size reduction in the code I think this counts as a cleanup as\nmuch as feature work.\n\nThere remain a number of helper functions (like ip option processing) to\ntake care of before the network stack can support destination devices in\nother network namespaces but with this set of changes the backbone of\nthe work is done.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ede2059dbaf9c6557a49d466c8c7778343b208ff", "tree": "d6d15cbeb675b52ef22890a2228b5301b1396c45", "parents": [ "33224b16ffccb49cf798317670389e0bfba0024c" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:47 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:03 2015 -0700" }, "message": "dst: Pass net into dst-\u003eoutput\n\nThe network namespace is already passed into dst_output pass it into\ndst-\u003eoutput lwt-\u003eoutput and friends.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "33224b16ffccb49cf798317670389e0bfba0024c", "tree": "25a11f38cf0da4331c558a06d4f3a5ccaabf12fa", "parents": [ "cf91a99daa4651d0c1f52b8c3d813fd44b43cada" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:46 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:02 2015 -0700" }, "message": "ipv4, ipv6: Pass net into ip_local_out and ip6_local_out\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cf91a99daa4651d0c1f52b8c3d813fd44b43cada", "tree": "211da00c404e813c85c8574cf09e91b2003e8202", "parents": [ "57c4bf859cad9d6c4f73d8c98a95e00f156301e0" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:45 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:02 2015 -0700" }, "message": "ipv4, ipv6: Pass net into __ip_local_out and __ip6_local_out\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "57c4bf859cad9d6c4f73d8c98a95e00f156301e0", "tree": "7fc1b076a979efaf3516e678bc0db71541eab927", "parents": [ "a7093fefa543afad2bf292bba163800e81a4ba33" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:44 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:01 2015 -0700" }, "message": "ipvlan: Cache net in ipvlan_process_v4_outbound and ipvlan_process_v6_outbound\n\nCompute net once in ipvlan_process_v4_outbound and\nipvlan_process_v6_outbound and store it in a variable so that net does\nnot need to be recomputed next time it is used.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a7093fefa543afad2bf292bba163800e81a4ba33", "tree": "2a56bea439ecfc5c566cd7dd57ceb3e776eda3f7", "parents": [ "77589ce0f84dd99cc946fd71fe6fb44dd8220d0a" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:43 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:27:00 2015 -0700" }, "message": "ppp: Cache net in pptp_xmit\n\nCompute net and store it in a variable in pptp_xmit, so that the value\ncan be reused the next time it is needed.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "77589ce0f84dd99cc946fd71fe6fb44dd8220d0a", "tree": "27c8a96e43cebfbe89c5cc8e3961c412137b8b36", "parents": [ "f859b0f662493e4f53d462f5759e3c4302933077" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:42 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:59 2015 -0700" }, "message": "ipv4: Cache net in ip_build_and_send_pkt and ip_queue_xmit\n\nCompute net and store it in a variable in the functions\nip_build_and_send_pkt and ip_queue_xmit so that it does not need to be\nrecomputed next time it is needed.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f859b0f662493e4f53d462f5759e3c4302933077", "tree": "6846a844be6dcfa407d4efcd5b3393d1f203c66a", "parents": [ "792883303cdb3a7edd16017d7aba53926189ef41" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:41 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:59 2015 -0700" }, "message": "ipv4: Cache net in iptunnel_xmit\n\nStore net in a variable in ip_tunnel_xmit so it does not need\nto be recomputed when it is used again.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "792883303cdb3a7edd16017d7aba53926189ef41", "tree": "6f55c849c97bb97fed30b4ca92c709cd4e5fe96f", "parents": [ "9f8955cc468ddb7d08a0e614a45f9a82c4019b00" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:40 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:58 2015 -0700" }, "message": "ipv6: Merge ip6_local_out and ip6_local_out_sk\n\nStop hidding the sk parameter with an inline helper function and make\nall of the callers pass it, so that it is clear what the function is\ndoing.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9f8955cc468ddb7d08a0e614a45f9a82c4019b00", "tree": "52c4d3eeb3a0744d92c8bffe87c6bc9615fe3a7e", "parents": [ "e2cb77db089796f163092326ca25512845df7a3a" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:39 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:58 2015 -0700" }, "message": "ipv6: Merge __ip6_local_out and __ip6_local_out_sk\n\nOnly __ip6_local_out_sk has callers so rename __ip6_local_out_sk\n__ip6_local_out and remove the previous __ip6_local_out.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e2cb77db089796f163092326ca25512845df7a3a", "tree": "d3ebd206afc735e0a6b3cb5ffe959774fd5d1e56", "parents": [ "b92dacd45698e120104ff81066ceb534916090d9" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:38 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:57 2015 -0700" }, "message": "ipv4: Merge ip_local_out and ip_local_out_sk\n\nIt is confusing and silly hiding a parameter so modify all of\nthe callers to pass in the appropriate socket or skb-\u003esk if\nno socket is known.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b92dacd45698e120104ff81066ceb534916090d9", "tree": "d317a02ae93e19a7ed471079eb5c510d9c566575", "parents": [ "4ebdfba73c09d8568d891bae87c40fad43dd7f41" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:37 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:57 2015 -0700" }, "message": "ipv4: Merge __ip_local_out and __ip_local_out_sk\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4ebdfba73c09d8568d891bae87c40fad43dd7f41", "tree": "f90cbfbed25de48ec9972d9b294cdcd856f05a12", "parents": [ "13206b6bff3b15b724926a222406476bf2c23c40" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:36 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:55 2015 -0700" }, "message": "dst: Pass a sk into .local_out\n\nFor consistency with the other similar methods in the kernel pass a\nstruct sock into the dst_ops .local_out method.\n\nSimplifying the socket passing case is needed a prequel to passing a\nstruct net reference into .local_out.\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "13206b6bff3b15b724926a222406476bf2c23c40", "tree": "4d433aa937254f44d86411cd33355072b56925ab", "parents": [ "3f5312ae620c79e877a6aa0e8894c6603a090b4e" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:35 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:54 2015 -0700" }, "message": "net: Pass net into dst_output and remove dst_output_okfn\n\nReplace dst_output_okfn with dst_output\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3f5312ae620c79e877a6aa0e8894c6603a090b4e", "tree": "2b539f61bd4a25b3019aad218f34e068ead42f1f", "parents": [ "850dcc4d4dd7d5da5c1b2a780c5e649c3b649545" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:34 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:53 2015 -0700" }, "message": "xfrm: Only compute net once in xfrm_policy_queue_process\n\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "850dcc4d4dd7d5da5c1b2a780c5e649c3b649545", "tree": "623983d4d59ea7de05c200006e112e2640d4fb2c", "parents": [ "fd2874b3bbe832e90ac480971a7a8bd736b629b9" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:33 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:52 2015 -0700" }, "message": "ipv4: Fix ip_queue_xmit to pass sk into ip_local_out_sk\n\nAfter a packet has been encapsulated by a tunnel we should use the\ntunnel sockets local multicast loopback flag to control if the\nencapsulated packet should be locally loopback back.\n\nPass sk into ip_local_out_sk so that in the rare case we are dealing\nwith a tunneled packet whose tunnel destination address is a multicast\naddress the kernel properly decides to loopback this packet.\n\nIn practice I don\u0027t think this matters as ip_queue_xmit is used by\ntcp, l2tp and sctp none of which I am aware of uses ip level\nmulticasting as they are all point to point communications protocols.\nLet\u0027s fix this before someone uses ip_queue_xmit for a tunnel protocol\nthat does use multicast.\n\nFixes: aad88724c9d5 (\"ipv4: add a sock pointer to dst-\u003eoutput() path.\")\nFixes: b0270e91014d (\"ipv4: add a sock pointer to ip_queue_xmit()\")\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fd2874b3bbe832e90ac480971a7a8bd736b629b9", "tree": "33640db1c1474a00defca1b810fc2208de415309", "parents": [ "e28383dd28afafefb0e445354c88f7b1e99aae9d" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Wed Oct 07 16:48:32 2015 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:26:52 2015 -0700" }, "message": "ipv4: Fix ip_local_out_sk by passing the sk into __ip_local_out_sk\n\nIn the rare case where sk !\u003d skb-\u003esk ip_local_out_sk arranges\nto call dst-\u003eoutput differently if the skb is queued or not.\nThis is a bug.\n\nFix this bug by passing the sk parameter of ip_local_out_sk through\nfrom ip_local_out_sk to __ip_local_out_sk (skipping __ip_local_out).\n\nFixes: 7026b1ddb6b8 (\"netfilter: Pass socket pointer down through okfn().\")\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e28383dd28afafefb0e445354c88f7b1e99aae9d", "tree": "8ac36d95a3a70c81c1173eb458a65aa8bf069de4", "parents": [ "61d0372028d98046176a8a36e40b78b5bbac5512", "7fd0ac66c2de5767e234fea6436d28dce210da80" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:21:09 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:21:09 2015 -0700" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue\n\nJeff Kirsher says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nIntel Wired LAN Driver Updates 2015-10-07\n\nThis series contains updates to i40e and i40evf only.\n\nPaul updates i40e to simply increase the amount of time we wait for a\nreset to complete since we have seen in some rare occasions the reset\ncan take longer to complete.\n\nShannon updates the driver to turn on Wake-on-LAN by default if it is\nenabled in the hardware config to begin with, rather than always disable\nit and wait for the user to expressly turn it on. Added new device id\u0027s\nand support for future devices. Fixed a possible type compare problem\nbetween a size and possible negative number. Also fixed a shift value\nthat was wrong, which ended up with a bad bitmask. Did general house\ncleaning of the driver to cleanup several low lying fruit in the\ndriver. Fixed an issue where new unicast address\u0027s would be added to\nthe VSI list and then immediately removed and would never actually\nmake it down to the hardware. Resolved the issue by removing the\nseparation from unicast and multicast in the search for filters to be\ndeleted.\n\nMitch fixes an issue where the hardware would continue to access the\nmemory formerly used by the rings for a VF which have been removed,\ncausing memory corruption or DMAR errors. To relieve this condition,\nexplicitly stop all rings associated with each VF before releasing its\nresources. Also fixed a panic if the driver is unable to enable MSI-X\nor its unable to acquire enough vectors, so propagate interrupt\nallocation failure information to the calling function. Cleaned up\nopcode that is not required.\n\nCarolyn extends the size of the test available for the interrupt names\nso that all the descriptive data available for the Flow Director\ninterrupts is not truncated.\n\nCatherine fixes an issue where there was a possibility of speed getting\nset to 0 if advertised is set to 0 (which is the case when autoneg is\ndisabled).\n\nJesse fixes the checksum on big endian machines, so added code to swap\nit correctly. Also fixed a bug in the return from get_link_status()\nwhere only true or false was being returned, but false could mean\nmultiple things. So allow the caller to get all the return values\nin the call chain bubbled back to the source so that the reason for\nthe failure does not get lost.\n\nAnjali adds statistics to keep track of how many times we ask the stack\nto linearize the SKB because the hardware cannot handle SKBs with more\nthan 8 frags per segment/single packet.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "61d0372028d98046176a8a36e40b78b5bbac5512", "tree": "1ebd9c640207328e5282a6b3d61c8fc39771e742", "parents": [ "acb4a6bfc80ddeea4c44074dd630f916259e909e", "77792b11409c9270d98e604b4314b85ce886ac7d" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:01:28 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Oct 08 04:01:28 2015 -0700" }, "message": "Merge tag \u0027regmap-offload-update-bits\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap\n\nregmap: Allow buses to provide a custom update_bits() operation\n\nSome buses provide a native _update_bits() operation which for uncached\nregisters is faster than doing a read/modify/write cycle as it is a\nsingle bus transaction. Add support for implementing this to regmap.\n" }, { "commit": "7fd0ac66c2de5767e234fea6436d28dce210da80", "tree": "7cb0454193509a8c82230bf3c4efb633764b470f", "parents": [ "313ed2d52052ad8dab9a6182cde55a832873cbea" ], "author": { "name": "Mitch Williams", "email": "mitch.a.williams@intel.com", "time": "Thu Aug 27 11:42:32 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:57:26 2015 -0700" }, "message": "i40e/i40evf: remove unused opcode\n\nThis opcode is not required. VFs that program RSS through the firmware\ndo it by interacting directly with the firmware, and do not need to use\nthe virtual channel for this functionality.\n\nChange-ID: Iaf17d2600e28ff1b6be8653f2fe9df1facd23b0e\nSigned-off-by: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "313ed2d52052ad8dab9a6182cde55a832873cbea", "tree": "eac13db25fa59215a0d10adca694c3e35c39f4ac", "parents": [ "2642f02528b6e9c1e904f6e3ec902d8a2deb7af5" ], "author": { "name": "Mitch Williams", "email": "mitch.a.williams@intel.com", "time": "Thu Aug 27 11:42:31 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:43:45 2015 -0700" }, "message": "i40evf: propagate interrupt allocation failure\n\nLower level functions are properly reporting errors, and higher-level\nfunctions are correctly responding to errors, but the errors aren\u0027t\nactually getting through. Typically, the middle-manager function seems\nto want to shield its boss from any bad news.\n\nThis change fixes a panic if the driver is unable to enable MSI-X or is\nunable to acquire enough vectors.\n\nChange-ID: Ifd5787ce92519a5d97e4b465902db930d97b71a1\nSigned-off-by: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "2642f02528b6e9c1e904f6e3ec902d8a2deb7af5", "tree": "a2087bc3e18a7b83df4880850c9144cb3be53a90", "parents": [ "2fc3d7152ae9562c15c30ed4a766ba05a3db8200" ], "author": { "name": "Neerav Parikh", "email": "neerav.parikh@intel.com", "time": "Thu Aug 27 11:42:30 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:39:30 2015 -0700" }, "message": "i40e: Additional checks for CEE APP priority validity\n\nThe firmware has added additional status information to allow software\nto determine if the APP priority for FCoE/iSCSI/FIP is valid or not in\nCEE DCBX mode.\n\nThis patch adds to support those additional checks and will only add\napplications to the software table that have oper and sync bits set\nwithout any error.\n\nChange-ID: I0a76c52427dadf97d4dba4538a3068d05e4eb56b\nSigned-off-by: Neerav Parikh \u003cneerav.parikh@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "2fc3d7152ae9562c15c30ed4a766ba05a3db8200", "tree": "add557469920af138a3961a20207c0e3a0c34d2e", "parents": [ "2f41f3358672dfda67c1e254f1e823d98e6a3099" ], "author": { "name": "Anjali Singhai Jain", "email": "anjali.singhai@intel.com", "time": "Thu Aug 27 11:42:29 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:35:45 2015 -0700" }, "message": "i40e/i40evf: Add a stat to keep track of linearization count\n\nKeep track of how many times we ask the stack to linearize the\nskb because the HW cannot handle skbs with more than 8 frags per\nsegment/single packet.\n\nChange-ID: If455452060963a769bbe6112cba952e79e944b52\nSigned-off-by: Anjali Singhai Jain \u003canjali.singhai@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "2f41f3358672dfda67c1e254f1e823d98e6a3099", "tree": "e5e26f525af249ec5dab704eb3ae83aba9fa288f", "parents": [ "a72a5abcb37beac163704efba6a3d33ebca4d90a" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:20 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:26:04 2015 -0700" }, "message": "i40e/i40evf: fix unicast mac address add\n\nWhen using something like \"ip maddr add ...\" to add another unicast mac\naddress to the netdev, the mac address comes into the set_rx_mode handler\nin the multicast list whether it is a unicast or multicast address.\nThis was confusing the code when it was trying to search for addresses\nthat needed to be deleted from the VSI, because it was looking for the\nVSI unicast address in the netdev unicast list. The result was that a\nnew unicast address would get added to the VSI list and then immediately\nremoved, and would never actually make it down into the hardware.\n\nThis patch removes the separation from unicast and multicast in the search\nfor filters to be deleted. It also simplifies the logic a little with a\njump to the bottom of the loop when an address is found. Now it doesn\u0027t\nmatter which netdev list the address is hiding in, we\u0027ll check them all.\n\nChange-ID: Ie3685a92427ae7d2212bf948919ce295bc7a874c\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "a72a5abcb37beac163704efba6a3d33ebca4d90a", "tree": "7922bf1bc1bcffb8ffc96353f684f295eb2d1b65", "parents": [ "dd38c583aea07c508caeaf979c64e267584ee067" ], "author": { "name": "Jesse Brandeburg", "email": "jesse.brandeburg@intel.com", "time": "Wed Aug 26 15:14:19 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:23:09 2015 -0700" }, "message": "i40e: fix bug in return from get_link_status and avoid spurious link messages\n\nPreviously, the driver could call this function and have only true/false\nreturned, but false could mean multiple things like failure to read\nor link was down. This change allows the caller to get all return values\nin the call chain bubbled back to the source, which keeps information about\nfailures from being lost.\n\nAlso, in some unlikely scenarios, the firmware can become slow to respond\nto admin queue (AQ) queries for link state. Should the AQ time out,\nthe driver can detect the state and avoid a link change when there\nmay have been none.\n\nChange-ID: Ib2ac38407b7880750fb891b392fa77457fe6c21c\nSigned-off-by: Jesse Brandeburg \u003cjesse.brandeburg@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "dd38c583aea07c508caeaf979c64e267584ee067", "tree": "ff05cbb666f985b9f43c31c9e5ff9fb31397c9f5", "parents": [ "fb43201f150300fa2c0215f00418eda9143cdab1" ], "author": { "name": "Jesse Brandeburg", "email": "jesse.brandeburg@intel.com", "time": "Wed Aug 26 15:14:18 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:18:53 2015 -0700" }, "message": "i40e: add little endian conversion for checksum\n\nThe checksum is not correct on big endian machines so add code to swap it\ncorrectly.\n\nChange-ID: Ic92b886d172a2cbe49f5d7eee1bc78e447023c7b\nSigned-off-by: Jesse Brandeburg \u003cjesse.brandeburg@intel.com\u003e\nSigned-off-by: Paul M Stillwell Jr \u003cpaul.m.stillwell.jr@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "fb43201f150300fa2c0215f00418eda9143cdab1", "tree": "3236c4fd6931d313c4511709fccaf393e285d7ec", "parents": [ "0002e1189b594d462c7a93417f5ac8bbc22bd768" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:17 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:11:24 2015 -0700" }, "message": "i40e/i40evf: give up the __func__\n\nDuring early development, we added the function name to all of the error\nstrings to make debugging simpler. Now that we\u0027ve released the driver,\nour users should have more comprehensible error messages. So tear the\nroof off and give up the __func__. Ow.\n\nChange-ID: I7e1766252c7a032b9af6520da6aff536bdfd533c\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nSigned-off-by: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "0002e1189b594d462c7a93417f5ac8bbc22bd768", "tree": "572506715ae324e8d9cd92b9410a03052ac3711d", "parents": [ "fba52e21e5cad1347255d0d3600f7e4df6e69435" ], "author": { "name": "Catherine Sullivan", "email": "catherine.sullivan@intel.com", "time": "Wed Aug 26 15:14:16 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:08:30 2015 -0700" }, "message": "i40e: Never let speed get set to 0 in get_settings\n\nIn ethtool, there is a possibility of speed getting set to 0\nif advertise is set to 0 (which it is when autoneg is disabled).\nWe never want this to happen as the firmware will actually attempt\nto set the speed to 0 sending link down, so add an extra check\nto make sure this doesn\u0027t happen.\n\nChange-ID: I62e0eeee2cbf043d8e6f5c9c9f0b92794e877f01\nSigned-off-by: Catherine Sullivan \u003ccatherine.sullivan@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "fba52e21e5cad1347255d0d3600f7e4df6e69435", "tree": "93def42f4b62e7ae1dcbe8cbd6071e6036574cf1", "parents": [ "9e1c26e35030aa56057a69514acb17da5e81eb16" ], "author": { "name": "Carolyn Wyborny", "email": "carolyn.wyborny@intel.com", "time": "Wed Aug 26 15:14:15 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 14:00:00 2015 -0700" }, "message": "i40e: Fix for truncated interrupt name\n\nThis patch extends the size of the text available for the interrupt names.\nWithout this patch, all the descriptive data available for the Flow\nDirector interrupts is truncated.\n\nChange-ID: I2ac458f23ac3b4ea8f1edf73edc283b1d3704c7f\nSigned-off-by: Carolyn Wyborny \u003ccarolyn.wyborny@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "9e1c26e35030aa56057a69514acb17da5e81eb16", "tree": "7543830abedabd6d9ad513cc93fb6ed8a3b69220", "parents": [ "21364bcfc7810894114ab18f5335745e374cc2ea" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:14 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:59:02 2015 -0700" }, "message": "i40e/i40evf: assure clean asq status report\n\nThere was a possibility where the asq_last_status could get through without\nupdate and thus report a previous error. I don\u0027t think we\u0027ve actually seen\nthis happen, but this patch will help make sure it doesn\u0027t.\n\nChange-ID: I9e33927052a5ee6ea21f80b66d4c4b76c2760b17\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nSigned-off-by: Christopher Pau \u003cchristopher.pau@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "21364bcfc7810894114ab18f5335745e374cc2ea", "tree": "2a0554540003a7a449a3b0e10013c0b833348dbf", "parents": [ "725821f340184fcdd35dbd369c5058eec8a30fcb" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:13 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:55:22 2015 -0700" }, "message": "i40e: make i40e_init_pf_fcoe to void\n\ni40e_init_pf_fcoe() didn\u0027t return anything except 0, it prints enough\nerror info already, and no driver logic depends on the return value,\nso this can be void.\n\nChange-ID: Ie6afad849857d87a7064c42c3cce14c74c2f29d8\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "725821f340184fcdd35dbd369c5058eec8a30fcb", "tree": "e908e91462dcecffab2218b12a3c3f88f2bfd6b6", "parents": [ "df08fd4dbcc95cbc5a35745b6a77a9df0dc5776f" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:12 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:48:07 2015 -0700" }, "message": "i40e: fix bad CEE status shift value\n\nFix a shift value that was wrong, ending up with a bad bitmask. Also add\na blank line between two sets of #defines for better readability.\n\nChange-ID: I3e41fa2a2ab904d3a4e6cbf13972ab0036a10601\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "df08fd4dbcc95cbc5a35745b6a77a9df0dc5776f", "tree": "2a3147419146e937e89ff9797e3d1c36eae63444", "parents": [ "bc5166b908aaa126d524b84c767323b75c17bdee" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:11 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:38:11 2015 -0700" }, "message": "i40e/i40evf: fix a potential type compare issue\n\nRework an if expression to assure there is no type compare problem between\na size and a possible negative number.\n\nChange-ID: I4921fcc96abfcf69490efce020a9e4007f251c99\nReported-by: Helin Zhang \u003chelin.zhang@intel.com\u003e\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "bc5166b908aaa126d524b84c767323b75c17bdee", "tree": "0f9002fe302cd5d86f6603f42c172de1415af4b7", "parents": [ "0325fca7577c5157da2cdf064fc3e8d38242e47c" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:10 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:34:09 2015 -0700" }, "message": "i40e/i40evf: add driver support for new device ids\n\nEarly addition of new a device id.\n\nChange-ID: I61a8c8556fdf4f5714be4e4089689e374f30293c\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "0325fca7577c5157da2cdf064fc3e8d38242e47c", "tree": "cc2ac5e07000b95d72226d9026f554ff4750b325", "parents": [ "1d5109d18757d8d6450905275bf362819781368c" ], "author": { "name": "Mitch Williams", "email": "mitch.a.williams@intel.com", "time": "Wed Aug 26 15:14:09 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:31:20 2015 -0700" }, "message": "i40e: stop VF rings\n\nExplicitly stop the rings belonging to each VF when disabling SR-IOV.\nEven though the VFs were gone, and the associated VSIs were removed, the\nrings were not stopped, and in some circumstances the hardware would\ncontinue to access the memory formerly used by the rings, causing memory\ncorruption or DMAR errors, both of which would lead to general malaise\nof the kernel.\n\nTo relieve this condition, explicitly stop all the rings associated with\neach VF before releasing its resources.\n\nChange-ID: I78c05d562c66e7b594b7e48d67860f49b3e5b6ec\nSigned-off-by: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "1d5109d18757d8d6450905275bf362819781368c", "tree": "6c80774983956a3d777e4f9bed28348eefca2727", "parents": [ "22e05bd6f7949d3e638c9f4697a2198a19fff312" ], "author": { "name": "Shannon Nelson", "email": "shannon.nelson@intel.com", "time": "Wed Aug 26 15:14:08 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:29:14 2015 -0700" }, "message": "i40e: enable WoL operation if config bit show WoL capable\n\nThe driver was disabling Wake-on-LAN by default and waiting for the user\nto expressly turn it on. This patch has the driver turning on WoL from\nthe start if enabled in the hardware config, which matches the behavior\nof our other drivers.\n\nChange-ID: I43faedb907f8ba4d1a61b72a7c86072b97af12b1\nSigned-off-by: Shannon Nelson \u003cshannon.nelson@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "22e05bd6f7949d3e638c9f4697a2198a19fff312", "tree": "38b32ebd4c2a7ef746e79fcd345a109d837358a8", "parents": [ "acb4a6bfc80ddeea4c44074dd630f916259e909e" ], "author": { "name": "Paul M Stillwell Jr", "email": "paul.m.stillwell.jr@intel.com", "time": "Wed Aug 26 15:14:07 2015 -0400" }, "committer": { "name": "Jeff Kirsher", "email": "jeffrey.t.kirsher@intel.com", "time": "Wed Oct 07 13:10:10 2015 -0700" }, "message": "i40e: Increase the amount of time we wait for reset to be done\n\nIn some rare cases the reset can take longer to complete so increase the\namount of time we wait.\n\nChange-ID: Ib5628ec54b526a811ee33d1214fe763226406671\nSigned-off-by: Paul M Stillwell Jr \u003cpaul.m.stillwell.jr@intel.com\u003e\nTested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\nSigned-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n" }, { "commit": "acb4a6bfc80ddeea4c44074dd630f916259e909e", "tree": "c7745772d9dac017daf2754ed3e46be9149d7659", "parents": [ "686a562449af96a0e8c18c6f1b87b47ff8c36de8" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue Oct 06 14:49:58 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 05:08:58 2015 -0700" }, "message": "tcp: ensure prior synack rtx behavior with small backlogs\n\nSome applications use a listen() backlog of 1.\n\nPrior kernels were silently enforcing a qlen_log of 4, so that we were\nsending up to /proc/sys/net/ipv4/tcp_synack_retries SYNACK messages.\n\nFixes: ef547f2ac16b (\"tcp: remove max_qlen_log\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "686a562449af96a0e8c18c6f1b87b47ff8c36de8", "tree": "48308fd9e59fb9947c50abd4d69f8977ec334e4d", "parents": [ "729ecbc77a1fe763de409be0307c61c1fa32cc06" ], "author": { "name": "Yuvaraja Mariappan", "email": "ymariappan@gmail.com", "time": "Tue Oct 06 10:53:29 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 05:01:04 2015 -0700" }, "message": "net: ipv4: tcp.c Fixed an assignment coding style issue\n\nFixed an assignment coding style issue\n\nSigned-off-by: Yuvaraja Mariappan \u003cymariappan@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "729ecbc77a1fe763de409be0307c61c1fa32cc06", "tree": "b12a51139bdbad49ab42340b44c28d57e9bed029", "parents": [ "68e4bd2778fb48a98cf9db8e218ef2f3a817bafc", "ec6674c6382819542446b727a2bf54f3dd879ca2" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:52:14 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:52:14 2015 -0700" }, "message": "Merge branch \u0027s390-net\u0027\n\nUrsula Braun says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\ns390: qeth patches for net-next\n\nhere are some s390 related patches for net-next. The qeth patches\nare performance optimizations in the driver. The qdio patch corrects\na warning condition.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ec6674c6382819542446b727a2bf54f3dd879ca2", "tree": "b12a51139bdbad49ab42340b44c28d57e9bed029", "parents": [ "fe5c802882f2eaaf5475b072f16066038754782a" ], "author": { "name": "Eugene Crosser", "email": "Eugene.Crosser@ru.ibm.com", "time": "Tue Oct 06 15:12:29 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:52:06 2015 -0700" }, "message": "s390/qdio: fix WARN_ON_ONCE condition\n\nIf HiperSockets Completion Queueing is enabled, qdio always\nissues a warning, since the condition is always met.\nThis patch fixes the condition in WARN_ON_ONCE that was always\ntrue.\n\nSigned-off-by: Eugene Crosser \u003cEugene.Crosser@ru.ibm.com\u003e\nSigned-off-by: Ursula Braun \u003cursula.braun@de.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fe5c802882f2eaaf5475b072f16066038754782a", "tree": "2f16f8c86917ab8d242100ac09720de75da6a4bc", "parents": [ "9abfa8cb201b2167872654b92f567f8bff6016f4" ], "author": { "name": "Lakhvich Dmitriy", "email": "ldmitriy@ru.ibm.com", "time": "Tue Oct 06 15:12:28 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:52:04 2015 -0700" }, "message": "s390/qeth: optimize MAC handling in rx_mode callback\n\nIn layer2 mode of the qeth driver, MAC address lists\nfrom struct net_device require mapping to the OSA-card.\nThe existing implementation is inefficient for lists with\nmore than several MAC addresses, since for every\nndo_set_rx_mode callback it removes all MAC addresses first,\nand then registers the current MAC address list.\nThis patch changes implementation of ndo_set_rx_mode callback\nin qeth, only performing hardware registration/removal for\nnew/deleted addresses. To shorten lookup of MAC addresses\nregistered addresses are kept in a hashtable instead of a\nlinear list.\n\nSigned-off-by: Lakhvich Dmitriy \u003cldmitriy@ru.ibm.com\u003e\nSigned-off-by: Ursula Braun \u003cursula.braun@de.ibm.com\u003e\nReviewed-by: Eugene Crosser \u003cEugene.Crosser@ru.ibm.com\u003e\nReviewed-by: Thomas Richter \u003ctmricht@de.ibm.com\u003e\nTested-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9abfa8cb201b2167872654b92f567f8bff6016f4", "tree": "2e212d132e6728378f0e723dbbf21bb7c713e2db", "parents": [ "68e4bd2778fb48a98cf9db8e218ef2f3a817bafc" ], "author": { "name": "Thomas Richter", "email": "tmricht@linux.vnet.ibm.com", "time": "Tue Oct 06 15:12:27 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:52:04 2015 -0700" }, "message": "s390/qeth: switch to napi_gro_receive\n\nAdd support for GRO (generic receive offload) in the layer 2\npart of device driver qeth. This results in a performance\nimprovement when GRO and RX is turned on.\n\nSigned-off-by: Thomas Richter \u003ctmricht@linux.vnet.ibm.com\u003e\nSigned-off-by: Ursula Braun \u003cursula.braun@de.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "68e4bd2778fb48a98cf9db8e218ef2f3a817bafc", "tree": "287195fb2c3dab3b9bffe7e4ec89ba80df80b612", "parents": [ "deaa0a6a930edc79081268bf23b196d0340499af", "5d6ae479ab7ddf77bb22bdf739268581453ff886" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:39 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:39 2015 -0700" }, "message": "Merge branch \u0027bridge-netlink-port-attrs\u0027\n\nNikolay Aleksandrov says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nbridge: netlink: complete port attribute support\n\nThis is the second set that completes the bridge port\u0027s netlink support and\nmakes everything from sysfs available via netlink. I\u0027ve used sysfs as a\nguide of what and how to set again. I\u0027ve tested setting/getting every\noption and also this time tested enabling KASAN. Again there\u0027re a few long\nline warnings about the ifla attribute names in br_port_info_size() but\nas the previous set - it\u0027s good to know what\u0027s been accounted for.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5d6ae479ab7ddf77bb22bdf739268581453ff886", "tree": "287195fb2c3dab3b9bffe7e4ec89ba80df80b612", "parents": [ "9b0c6e4deb3df91bf0aea8158ea77dc58c9d90b6" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:12:02 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:34 2015 -0700" }, "message": "bridge: netlink: add support for port\u0027s multicast_router attribute\n\nAdd IFLA_BRPORT_MULTICAST_ROUTER to allow setting/getting port\u0027s\nmulticast_router via netlink.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9b0c6e4deb3df91bf0aea8158ea77dc58c9d90b6", "tree": "2fb9c31e971436462f94c2b0960d0c2392c7d6b6", "parents": [ "61c0a9a83e0b12c712cd686172446aba8ea48685" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:12:01 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:32 2015 -0700" }, "message": "bridge: netlink: allow to flush port\u0027s fdb\n\nAdd IFLA_BRPORT_FLUSH to allow flushing port\u0027s fdb similar to sysfs\u0027s\nflush.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "61c0a9a83e0b12c712cd686172446aba8ea48685", "tree": "e66a99d2c1c624799fce4c2d24d097f7d8254945", "parents": [ "e08e838ac5707cb1f1294e0d53b31997a0367b99" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:12:00 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:31 2015 -0700" }, "message": "bridge: netlink: export port\u0027s timer values\n\nAdd the following attributes in order to export port\u0027s timer values:\nIFLA_BRPORT_MESSAGE_AGE_TIMER, IFLA_BRPORT_FORWARD_DELAY_TIMER and\nIFLA_BRPORT_HOLD_TIMER.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e08e838ac5707cb1f1294e0d53b31997a0367b99", "tree": "c290b5298fa6306d56401a34afdd1119a02a2adf", "parents": [ "42d452c4b5e7bf0e3024fa9512ec462f70545ae5" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:11:59 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:30 2015 -0700" }, "message": "bridge: netlink: export port\u0027s topology_change_ack and config_pending\n\nAdd IFLA_BRPORT_TOPOLOGY_CHANGE_ACK and IFLA_BRPORT_CONFIG_PENDING to\nallow getting port\u0027s topology_change_ack and config_pending respectively\nvia netlink.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "42d452c4b5e7bf0e3024fa9512ec462f70545ae5", "tree": "6bd7eabec0576557bc2ac6b3a434c974a2a23f34", "parents": [ "96f94e7f4a216282a24819968184c881e6343692" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:11:58 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:29 2015 -0700" }, "message": "bridge: netlink: export port\u0027s id and number\n\nAdd IFLA_BRPORT_(ID|NO) to allow getting port\u0027s port_id and port_no\nrespectively via netlink.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "96f94e7f4a216282a24819968184c881e6343692", "tree": "404163f5a0e0fb4c79954059025f4b322868be89", "parents": [ "80df9a2692edf7afffda9282e716e7b1df198e07" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:11:57 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:29 2015 -0700" }, "message": "bridge: netlink: export port\u0027s designated cost and port\n\nAdd IFLA_BRPORT_DESIGNATED_(COST|PORT) to allow getting the port\u0027s\ndesignated cost and port respectively via netlink.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "80df9a2692edf7afffda9282e716e7b1df198e07", "tree": "549b960d23c5f51050afbf257644c42e51fd16e7", "parents": [ "4ebc7660ab4559cad10b6595e05f70562bb26dc5" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:11:56 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:28 2015 -0700" }, "message": "bridge: netlink: export port\u0027s bridge id\n\nAdd IFLA_BRPORT_BRIDGE_ID to allow getting the designated bridge id via\nnetlink.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4ebc7660ab4559cad10b6595e05f70562bb26dc5", "tree": "d71e83da8746bc9a5a49a68e634ec2352d3e7ef3", "parents": [ "deaa0a6a930edc79081268bf23b196d0340499af" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Tue Oct 06 14:11:55 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:49:27 2015 -0700" }, "message": "bridge: netlink: export port\u0027s root id\n\nAdd IFLA_BRPORT_ROOT_ID to allow getting the designated root id via\nnetlink.\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "deaa0a6a930edc79081268bf23b196d0340499af", "tree": "f545f2206ff791f60b301a70ae0b779e08545e8a", "parents": [ "2579c98f0d04075ab8b041de7d5396393d2417f1" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 10:49:04 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:31:16 2015 -0700" }, "message": "net: Lookup actual route when oif is VRF device\n\nIf the user specifies a VRF device in a get route query the custom route\npointing to the VRF device is returned:\n\n $ ip route ls table vrf-red\n unreachable default\n broadcast 10.2.1.0 dev eth1 proto kernel scope link src 10.2.1.2\n 10.2.1.0/24 dev eth1 proto kernel scope link src 10.2.1.2\n local 10.2.1.2 dev eth1 proto kernel scope host src 10.2.1.2\n broadcast 10.2.1.255 dev eth1 proto kernel scope link src 10.2.1.2\n\n $ ip route get oif vrf-red 10.2.1.40\n 10.2.1.40 dev vrf-red\n cache\n\nAdd the flags to skip the custom route and go directly to the FIB. With\nthis patch the actual route is returned:\n\n $ ip route get oif vrf-red 10.2.1.40\n 10.2.1.40 dev eth1 src 10.2.1.2\n cache\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2579c98f0d04075ab8b041de7d5396393d2417f1", "tree": "094fb6486f5ede529cd772223d2fa0334a28f852", "parents": [ "390a4bee5c2ade628565dd21fb5d8656a58f7804", "5edfcee5ed73eb9537987c4ddb6bf062b6943b73" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:29:18 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:29:18 2015 -0700" }, "message": "Merge tag \u0027mac80211-next-for-davem-2015-10-05\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next\n\nJohannes Berg says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nFor the current cycle, we have the following right now:\n * many internal fixes, API improvements, cleanups, etc.\n * full AP client state tracking in cfg80211/mac80211 from Ayala\n * VHT support (in mac80211) for mesh\n * some A-MSDU in A-MPDU support from Emmanuel\n * show current TX power to userspace (from Rafał)\n * support for netlink dump in vendor commands (myself)\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "390a4bee5c2ade628565dd21fb5d8656a58f7804", "tree": "237b14ce30a495aa106b8e4710202f058dcd2834", "parents": [ "4148987a5111b0c8062bd78f39a67c361f621a39", "bb191c3e874650ae8f701885f3dd5f8ea8989b19" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:51 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:51 2015 -0700" }, "message": "Merge branch \u0027l3mdev_saddr_op\u0027\n\nDavid Ahern says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: Add saddr op to l3mdev and vrf\n\nFirst 2 patches are re-sends of patches that got lost in the ethosphere\nTuesday; they were part of the first round of l3mdev conversions.\nNext 3 handle the source address lookup for raw and datagram sockets\nbound to a VRF device.\n\nThe conversion to the get_saddr op also fixes locally originated TCP\npackets showing up at the VRF device. The use of the FLOWI_FLAG_L3MDEV_SRC\nflag in ip_route_connect_init was causing locally generated packets\nto skip the VRF device.\n\nv2\n- rebased to top of net-next per device delete fix and hash based\n multipath patches\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bb191c3e874650ae8f701885f3dd5f8ea8989b19", "tree": "237b14ce30a495aa106b8e4710202f058dcd2834", "parents": [ "8cbb512c923d5f695ff6265b2b741b1718e3b444" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 08:51:27 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:46 2015 -0700" }, "message": "net: Add l3mdev saddr lookup to raw_sendmsg\n\nping originated on box through a VRF device is showing up in tcpdump\nwithout a source address:\n $ tcpdump -n -i vrf-blue\n 08:58:33.311303 IP 0.0.0.0 \u003e 10.2.2.254: ICMP echo request, id 2834, seq 1, length 64\n 08:58:33.311562 IP 10.2.2.254 \u003e 10.2.2.2: ICMP echo reply, id 2834, seq 1, length 64\n\nAdd the call to l3mdev_get_saddr to raw_sendmsg.\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8cbb512c923d5f695ff6265b2b741b1718e3b444", "tree": "194de9f9c5ec859456dc8274136fcda881dc8408", "parents": [ "3ce58d84358c7b477811b5100152fad848f936fc" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 08:51:26 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:44 2015 -0700" }, "message": "net: Add source address lookup op for VRF\n\nAdd operation to l3mdev to lookup source address for a given flow.\nAdd support for the operation to VRF driver and convert existing\nIPv4 hooks to use the new lookup.\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3ce58d84358c7b477811b5100152fad848f936fc", "tree": "be2bfdfeddad87ab8f9374e91063213982e24ffa", "parents": [ "fee6d4c777a125e56de9370db3b2bf359bf958d6" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 08:51:25 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:44 2015 -0700" }, "message": "net: Refactor path selection in __ip_route_output_key_hash\n\nVRF device needs the same path selection following lookup to set source\naddress. Rather than duplicating code, move existing code into a\nfunction that is exported to modules.\n\nCode move only; no functional change.\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fee6d4c777a125e56de9370db3b2bf359bf958d6", "tree": "19bf245321e6d1265e34f550fd479d83ffc69aed", "parents": [ "6e2895a8e3824eb5611c97a015a3b6d678b4503e" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 08:51:24 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:43 2015 -0700" }, "message": "net: Add netif_is_l3_slave\n\nIPv6 addrconf keys off of IFF_SLAVE so can not use it for L3 slave.\nAdd a new private flag and add netif_is_l3_slave function for checking\nit.\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6e2895a8e3824eb5611c97a015a3b6d678b4503e", "tree": "b63a4f49f7d386d993027a14c2898877666dc638", "parents": [ "4148987a5111b0c8062bd78f39a67c361f621a39" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 08:51:23 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:27:42 2015 -0700" }, "message": "net: Rename FLOWI_FLAG_VRFSRC to FLOWI_FLAG_L3MDEV_SRC\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4148987a5111b0c8062bd78f39a67c361f621a39", "tree": "63f4d56dcabe89d032976c97d29ce1aeaffd1c54", "parents": [ "3e905b80b92146d8dd0bf2da3ade0944a3026597" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Mon Oct 05 08:32:51 2015 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:24:00 2015 -0700" }, "message": "net: Fix vti use case with oif in dst lookups for IPv6\n\nIt occurred to me yesterday that 741a11d9e4103 (\"net: ipv6: Add\nRT6_LOOKUP_F_IFACE flag if oif is set\") means that xfrm6_dst_lookup\nneeds the FLOWI_FLAG_SKIP_NH_OIF flag set. This latest commit causes\nthe oif to be considered in lookups which is known to break vti. This\nexplains why 58189ca7b274 did not the IPv6 change at the time it was\nsubmitted.\n\nFixes: 42a7b32b73d6 (\"xfrm: Add oif to dst lookups\")\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3e905b80b92146d8dd0bf2da3ade0944a3026597", "tree": "20eada1f69c35dabaa1ae50dee96455897fc772d", "parents": [ "70963d245ed11412658b89fe308b6188b62259c7" ], "author": { "name": "Claudiu Manoil", "email": "claudiu.manoil@freescale.com", "time": "Mon Oct 05 17:19:59 2015 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:19:45 2015 -0700" }, "message": "gianfar: Add WAKE_UCAST and \"wake-on-filer\" support\n\nThis enables eTSEC\u0027s filer (Rx parser) and the FGPI Rx\ninterrupt (Filer General Purpose Interrupt) as a wakeup\nsource event.\n\nUpon entering suspend state, the eTSEC filer is given\na rule to match incoming L2 unicast packets. A packet\nmatching the rule will be enqueued in the Rx ring and\na FGPI Rx interrupt will be asserted by the filer to\nwakeup the system. Other packet types will be dropped.\nOn resume the filer table is restored to the content\nbefore entering suspend state.\nThe set of rules from gfar_filer_config_wol() could be\nextended to implement other WoL capabilities as well.\n\nThe \"fsl,wake-on-filer\" DT binding enables this capability\non certain platforms that feature the necessary power\nmanagement infrastructure, targeting mainly printing and\nimaging applications.\n(refer to Power Management section of the SoC Ref Man)\n\nCc: Li Yang \u003cleoli@freescale.com\u003e\nCc: Zhao Chenhui \u003cchenhui.zhao@freescale.com\u003e\n\nSigned-off-by: Claudiu Manoil \u003cclaudiu.manoil@freescale.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "70963d245ed11412658b89fe308b6188b62259c7", "tree": "9b0f6342c510add92475491c8198a623b6cc2085", "parents": [ "66cebb86ac77c6dff0751efa382551cab24075cd" ], "author": { "name": "Claudiu Manoil", "email": "claudiu.manoil@freescale.com", "time": "Mon Oct 05 17:19:58 2015 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:19:43 2015 -0700" }, "message": "powerpc: dts: p1022si: Add fsl,wake-on-filer for eTSEC\n\nEnable the \"wake-on-filer\" (aka. wake on user defined packet)\nwake on lan capability for the eTSEC ethernet nodes.\n\nCc: Li Yang \u003cleoli@freescale.com\u003e\nCc: Zhao Chenhui \u003cchenhui.zhao@freescale.com\u003e\n\nSigned-off-by: Claudiu Manoil \u003cclaudiu.manoil@freescale.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "66cebb86ac77c6dff0751efa382551cab24075cd", "tree": "e719505bfd716e2bcd5504c50977de3c21c5bf55", "parents": [ "04a8250b40394f34a54439c62c883322b3107bd2" ], "author": { "name": "Claudiu Manoil", "email": "claudiu.manoil@freescale.com", "time": "Mon Oct 05 17:19:57 2015 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:19:42 2015 -0700" }, "message": "doc: dt: net: Add fsl,wake-on-filer for eTSEC\n\nAdd the \"fsl,wake-on-filer\" property for eTSEC nodes to\nindicate that the system has the power management\ninfrastructure needed to be able to wake up the system\nvia FGPI (filer, aka. h/w rx parser) interrupt.\n\nCc: Li Yang \u003cleoli@freescale.com\u003e\nCc: Zhao Chenhui \u003cchenhui.zhao@freescale.com\u003e\n\nSigned-off-by: Claudiu Manoil \u003cclaudiu.manoil@freescale.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "04a8250b40394f34a54439c62c883322b3107bd2", "tree": "6c15c6b572dd2656ddd6401549b1cfbb4ecf1931", "parents": [ "4917a1548ff41e53d863d6845b4da1884e4282b4", "6b26ba3a7d952e611dcde1f3f77ce63bcc70540a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:18:04 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:18:04 2015 -0700" }, "message": "Merge branch \u0027ovs-ipv6-tunnel\u0027\n\nJiri Benc says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nopenvswitch: add IPv6 tunneling support\n\nThis builds on the previous work that added IPv6 support to lwtunnels and\nadds IPv6 tunneling support to ovs.\n\nTo use IPv6 tunneling, there needs to be a metadata based tunnel net_device\ncreated and added to the ovs bridge. Currently, only vxlan is supported by\nthe kernel, with geneve to follow shortly. There\u0027s no need nor intent to add\na support for this into the vport-vxlan (etc.) compat layer.\n\nv3: dropped the last two patches added in v2.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6b26ba3a7d952e611dcde1f3f77ce63bcc70540a", "tree": "6c15c6b572dd2656ddd6401549b1cfbb4ecf1931", "parents": [ "00a93babd06aaad31d23384cda576ede0f586a8c" ], "author": { "name": "Jiri Benc", "email": "jbenc@redhat.com", "time": "Mon Oct 05 13:09:47 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:18:00 2015 -0700" }, "message": "openvswitch: netlink attributes for IPv6 tunneling\n\nAdd netlink attributes for IPv6 tunnel addresses. This enables IPv6 support\nfor tunnels.\n\nSigned-off-by: Jiri Benc \u003cjbenc@redhat.com\u003e\nAcked-by: Pravin B Shelar \u003cpshelar@nicira.com\u003e\nAcked-by: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "00a93babd06aaad31d23384cda576ede0f586a8c", "tree": "2fd12c2f928eae294ae0892e99562ac1f2d9997e", "parents": [ "4917a1548ff41e53d863d6845b4da1884e4282b4" ], "author": { "name": "Jiri Benc", "email": "jbenc@redhat.com", "time": "Mon Oct 05 13:09:46 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:17:59 2015 -0700" }, "message": "openvswitch: add tunnel protocol to sw_flow_key\n\nStore tunnel protocol (AF_INET or AF_INET6) in sw_flow_key. This field now\nalso acts as an indicator whether the flow contains tunnel data (this was\npreviously indicated by tun_key.u.ipv4.dst being set but with IPv6 addresses\nin an union with IPv4 ones this won\u0027t work anymore).\n\nThe new field was added to a hole in sw_flow_key.\n\nSigned-off-by: Jiri Benc \u003cjbenc@redhat.com\u003e\nAcked-by: Pravin B Shelar \u003cpshelar@nicira.com\u003e\nAcked-by: Thomas Graf \u003ctgraf@suug.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4917a1548ff41e53d863d6845b4da1884e4282b4", "tree": "437b1b791b33585da556e1d52214be34159a8d7c", "parents": [ "16660f0bd942cec203eaf4de0e2ac1695bd9d32d" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Mon Oct 05 12:11:21 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:15:57 2015 -0700" }, "message": "bridge: netlink: make br_fill_info\u0027s frame size smaller\n\nWhen KASAN is enabled the frame size grows \u003e 2048 bytes and we get a\nwarning, so make it smaller.\nnet/bridge/br_netlink.c: In function \u0027br_fill_info\u0027:\n\u003e\u003e net/bridge/br_netlink.c:1110:1: warning: the frame size of 2160 bytes\n\u003e\u003e is larger than 2048 bytes [-Wframe-larger-than\u003d]\n\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "16660f0bd942cec203eaf4de0e2ac1695bd9d32d", "tree": "219d0bcfe5dee72bc73fdd5099b3117ac47b29f9", "parents": [ "e892406f00e1f93d81fdb350e1bcb5cb0e5dabff" ], "author": { "name": "David Ahern", "email": "dsa@cumulusnetworks.com", "time": "Sat Oct 03 11:43:46 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 04:12:02 2015 -0700" }, "message": "net: Add support for filtering neigh dump by device index\n\nAdd support for filtering neighbor dumps by device by adding the\nNDA_IFINDEX attribute to the dump request.\n\nSigned-off-by: David Ahern \u003cdsa@cumulusnetworks.com\u003e\nReviewed-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e892406f00e1f93d81fdb350e1bcb5cb0e5dabff", "tree": "3492cf7c5184453bac6bd4957d0095f51a2cd44b", "parents": [ "d25b8e74291fec2dbf3fe3df7f20289eeaa9d28f", "8552d85442dd814d733f67592627ea55f7cbfb01" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 03:01:53 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 03:01:53 2015 -0700" }, "message": "Merge branch \u0027master\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue\n\nJeff Kirsher says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nIntel Wired LAN Driver Updates 2015-10-03\n\nThis series contains updates to i40e and i40evf, some of which are to\nresolve more Red Hat bugzilla issues.\n\nJiang Liu updates the i40e and i40evf drivers to use numa_mem_id()\ninstead of numa_node_id() to get the nearest node with memory which\nbetter supports memoryless nodes.\n\nAnjali fixes an issue from Dan Carpenter \u003cdan.carpenter@oracle.com\u003e,\nto resolve a memory leak in X722 RSS configuration path, where we should\nfree the memory allocated before exiting.\n\nShannon modifies the drivers to ensure we have the spinlocks before we\nclear the ARQ and ASQ management registers. In addition, we widen the\nlocked portion insert a sanity check to ensure we are working with safe\nregister values.\n\nMitch fixes an issue where under certain circumstances, we can get an\nextra VF_RESOURCES message from the PF driver at runtime. When this\noccurs, we need to parse it because our VSI may have changed and that\nwill affect the relationship with the PF driver. But this parsing also\nblows away our current MAC address, so resolve the issue by restoring\nthe current MAC address from the netdev struct after we parse the\nresource message.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d25b8e74291fec2dbf3fe3df7f20289eeaa9d28f", "tree": "c20a5e5778954ba7cbe12ae2dc24c5fea3f10806", "parents": [ "4bac50bace0377138fed2ac3627c1ad470ea1eca" ], "author": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sat Oct 03 18:09:07 2015 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 02:58:49 2015 -0700" }, "message": "net: dsa: better error reporting\n\nAdd additional error reporting to the generic DSA code, so it\u0027s easier\nto debug when things go wrong. This was useful when initially bringing\nup 88e6176 on a new board.\n\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nReviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nAcked-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4bac50bace0377138fed2ac3627c1ad470ea1eca", "tree": "ab1ea5bfd6009271431cd35b75c3dfd2804f27be", "parents": [ "21c4c073f14509d685ed219aa3c76362a7bfa0ac" ], "author": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sat Oct 03 18:09:01 2015 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Oct 07 02:58:47 2015 -0700" }, "message": "net: dsa: mv88e6xxx: remove link polling\n\nThe link status is polled by the generic phy layer, there\u0027s no need to\nduplicate that polling with additional polling. This additional polling\nadds additional MDIO traffic, and races with the generic phy layer,\nresulting in missing or duplicated link status messages.\n\nTested-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "77792b11409c9270d98e604b4314b85ce886ac7d", "tree": "13f3a518de69bbbccbd410b26ff24f0379c28dfd", "parents": [ "6ff33f3902c3b1c5d0db6b1e2c70b6d76fba357f" ], "author": { "name": "Jon Ringle", "email": "jringle@gridpoint.com", "time": "Thu Oct 01 12:38:07 2015 -0400" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Tue Oct 06 16:12:34 2015 +0100" }, "message": "regmap: Allow installing custom reg_update_bits function\n\nThis commit allows installing a custom reg_update_bits function for cases where\nthe hardware provides a mechanism to set or clear register bits without a\nread/modify/write cycle. Such is the case with the Microchip ENCX24J600.\n\nIf a custom reg_update_bits function is provided, it will only be used against\nvolatile registers.\n\nSigned-off-by: Jon Ringle \u003cjringle@gridpoint.com\u003e\nSigned-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n" }, { "commit": "21c4c073f14509d685ed219aa3c76362a7bfa0ac", "tree": "46e100b34be7598be3ac8e4f8793f04597828819", "parents": [ "6a27a6c3be95d382cd158c5705c1840be291f28f" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 06 06:25:43 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 06 06:25:43 2015 -0700" }, "message": "Revert \"regmap: Allow installing custom reg_update_bits function\"\n\nThis reverts commit 7741c373cf3ea1f5383fa97fb7a640a429d3dd7c.\n" }, { "commit": "6a27a6c3be95d382cd158c5705c1840be291f28f", "tree": "baf1695a6d2e53779588cb25d0fb14b708c629b0", "parents": [ "c664bc6d946d253077f3a5d5ca6769492de59e04" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 06 06:25:36 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 06 06:25:36 2015 -0700" }, "message": "Revert \"net: Microchip encx24j600 driver\"\n\nThis reverts commit 04fbfce7a222327b97ca165294ef19f0faa45960.\n" }, { "commit": "c664bc6d946d253077f3a5d5ca6769492de59e04", "tree": "bbf7f1dd03fa2957ac64929ab7996a5f027ecfcf", "parents": [ "0a837fe4724713ef701e47d6bfab98a5efaff3eb" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 06 06:25:29 2015 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 06 06:25:29 2015 -0700" }, "message": "Revert \"net: encx24j600_exit() can be static\"\n\nThis reverts commit 9886ce2b9d4e5a8bb3d78d0f7eff3c0f1ed58d67.\n" }, { "commit": "0a837fe4724713ef701e47d6bfab98a5efaff3eb", "tree": "8732e87e93a53f5e54e728b72fc7280903f5fbac", "parents": [ "0cdf5640e4f6940bdbbefee4bb0adb7dffb185ec" ], "author": { "name": "Peter Nørlund", "email": "pch@ordbogen.com", "time": "Tue Oct 06 07:24:47 2015 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Oct 05 23:48:09 2015 -0700" }, "message": "ipv4: Fix compilation errors in fib_rebalance\n\nThis fixes\n\nnet/built-in.o: In function `fib_rebalance\u0027:\nfib_semantics.c:(.text+0x9df14): undefined reference to `__divdi3\u0027\n\nand\n\nnet/built-in.o: In function `fib_rebalance\u0027:\nnet/ipv4/fib_semantics.c:572: undefined reference to `__aeabi_ldivmod\u0027\n\nFixes: 0e884c78ee19 (\"ipv4: L3 hash-based multipath\")\n\nSigned-off-by: Peter Nørlund \u003cpch@ordbogen.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "06766513232d1619ac84e87b1d839d3fcc23a540", "tree": "2ba73da74159985e20e1dd164c143c0cefd0ee7f", "parents": [ "41a4e9646229801624e38f7a1cc53033a0affdb1" ], "author": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@oracle.com", "time": "Thu Sep 10 21:20:57 2015 -0700" }, "committer": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@oracle.com", "time": "Mon Oct 05 11:19:02 2015 -0700" }, "message": "RDS: IB: split mr pool to improve 8K messages performance\n\n8K message sizes are pretty important usecase for RDS current\nworkloads so we make provison to have 8K mrs available from the pool.\nBased on number of SG\u0027s in the RDS message, we pick a pool to use.\n\nAlso to make sure that we don\u0027t under utlise mrs when say 8k messages\nare dominating which could lead to 8k pull being exhausted, we fall-back\nto 1m pool till 8k pool recovers for use.\n\nThis helps to at least push ~55 kB/s bidirectional data which\nis a nice improvement.\n\nSigned-off-by: Santosh Shilimkar \u003cssantosh@kernel.org\u003e\nSigned-off-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n" }, { "commit": "41a4e9646229801624e38f7a1cc53033a0affdb1", "tree": "fca8012e32d2215b5202274842a4fe15dad20806", "parents": [ "67161e250a28de5cdafa99a3b659b1e2e269fd7e" ], "author": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@oracle.com", "time": "Sat Sep 19 13:06:08 2015 -0400" }, "committer": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@oracle.com", "time": "Mon Oct 05 11:19:02 2015 -0700" }, "message": "RDS: IB: use max_mr from HCA caps than max_fmr\n\nAll HCA drivers seems to popullate max_mr caps and few of\nthem do both max_mr and max_fmr.\n\nHence update RDS code to make use of max_mr.\n\nSigned-off-by: Santosh Shilimkar \u003cssantosh@kernel.org\u003e\nSigned-off-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n" }, { "commit": "67161e250a28de5cdafa99a3b659b1e2e269fd7e", "tree": "0a08be128ef6eeb7b2a5a3fa41a1bf1aa6ba0236", "parents": [ "26139dc1dbf79fd1ae1e2766a1f66b0728bd67b3" ], "author": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@oracle.com", "time": "Sat Sep 19 17:21:22 2015 -0400" }, "committer": { "name": "Santosh Shilimkar", "email": "santosh.shilimkar@oracle.com", "time": "Mon Oct 05 11:19:02 2015 -0700" }, "message": "RDS: IB: mark rds_ib_fmr_wq static\n\nFix below warning by marking rds_ib_fmr_wq static\n\nnet/rds/ib_rdma.c:87:25: warning: symbol \u0027rds_ib_fmr_wq\u0027 was not declared. Should it be static?\n\nSigned-off-by: Santosh Shilimkar \u003cssantosh@kernel.org\u003e\nSigned-off-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n" } ], "next": "26139dc1dbf79fd1ae1e2766a1f66b0728bd67b3" }