)]}' { "log": [ { "commit": "623898671c8eb05639e746e6d84cffa281616438", "tree": "476176839bb7ab1c5ff435d5993abba18e5070f6", "parents": [ "57400d305201e1025ea0c20c851173146271bd1b", "d48756228ee9161ac8836b346589a43fabdc9f3c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:59:39 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:59:39 2016 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n \"A set of fixes, one for NVMe from Keith, and a set for nvme-{rdma,t,f}\n from the usual suspects, fixing actual problems that would be a shame\n to release 4.9 with\"\n\n* \u0027for-linus\u0027 of git://git.kernel.dk/linux-block:\n nvme/pci: Don\u0027t free queues on error\n nvmet-rdma: drain the queue-pair just before freeing it\n nvme-rdma: stop and free io queues on connect failure\n nvmet-rdma: don\u0027t forget to delete a queue from the list of connection failed\n nvmet: Don\u0027t queue fatal error work if csts.cfs is set\n nvme-rdma: reject non-connect commands before the queue is live\n nvmet-rdma: Fix possible NULL deref when handling rdma cm events\n" }, { "commit": "57400d305201e1025ea0c20c851173146271bd1b", "tree": "16c025ab63f43f314f300735723f51ce9d96bbae", "parents": [ "bec1b089ab287d5df160205f5949114e5a3d3162", "5c6b2aaf9316fd0983c0c999d920306ddc65bd2d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:53:02 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:53:02 2016 -0800" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma\n\nPull rmda fixes from Doug Ledford.\n \"First round of -rc fixes.\n\n Due to various issues, I\u0027ve been away and couldn\u0027t send a pull request\n for about three weeks. There were a number of -rc patches that built\n up in the meantime (some where there already from the early -rc\n stages). Obviously, there were way too many to send now, so I tried to\n pare the list down to the more important patches for the -rc cycle.\n\n Most of the code has had plenty of soak time at the various vendor\u0027s\n testing setups, so I doubt there will be another -rc pull request this\n cycle. I also tried to limit the patches to those with smaller\n footprints, so even though a shortlog is longer than I would like, the\n actual diffstat is mostly very small with the exception of just three\n files that had more changes, and a couple files with pure removals.\n\n Summary:\n - Misc Intel hfi1 fixes\n - Misc Mellanox mlx4, mlx5, and rxe fixes\n - A couple cxgb4 fixes\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (34 commits)\n iw_cxgb4: invalidate the mr when posting a read_w_inv wr\n iw_cxgb4: set *bad_wr for post_send/post_recv errors\n IB/rxe: Update qp state for user query\n IB/rxe: Clear queue buffer when modifying QP to reset\n IB/rxe: Fix handling of erroneous WR\n IB/rxe: Fix kernel panic in UDP tunnel with GRO and RX checksum\n IB/mlx4: Fix create CQ error flow\n IB/mlx4: Check gid_index return value\n IB/mlx5: Fix NULL pointer dereference on debug print\n IB/mlx5: Fix fatal error dispatching\n IB/mlx5: Resolve soft lock on massive reg MRs\n IB/mlx5: Use cache line size to select CQE stride\n IB/mlx5: Validate requested RQT size\n IB/mlx5: Fix memory leak in query device\n IB/core: Avoid unsigned int overflow in sg_alloc_table\n IB/core: Add missing check for addr_resolve callback return value\n IB/core: Set routable RoCE gid type for ipv4/ipv6 networks\n IB/cm: Mark stale CM id\u0027s whenever the mad agent was unregistered\n IB/uverbs: Fix leak of XRC target QPs\n IB/hfi1: Remove incorrect IS_ERR check\n ...\n" }, { "commit": "bec1b089ab287d5df160205f5949114e5a3d3162", "tree": "15b0f849a45ee661e65bc3a07c13d129f4de0221", "parents": [ "d46bc34da9bcdab815c4972ad0d433af8eb98c8a", "680bb946a1ae04fe0ff369a4965f76b48c07dc54" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:49:30 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:49:30 2016 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull vfs fixes from Al Viro:\n \"A couple of regression fixes\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n fix iov_iter_advance() for ITER_PIPE\n xattr: Fix setting security xattrs on sockfs\n" }, { "commit": "d46bc34da9bcdab815c4972ad0d433af8eb98c8a", "tree": "f50ae5a5b8feb8d664c5be0fb215d39df8d7a318", "parents": [ "5d1904204c99596b50a700f092fe49d78edba400", "19ff7fcc76e6911a955742b40f85ba1030ccba5e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:45:57 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 13:45:57 2016 -0800" }, "message": "Merge tag \u0027for-linus-4.9-rc5-ofs-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hubcap/linux\n\nPull orangefs fix from Mike Marshall:\n \"orangefs: add .owner to debugfs file_operations\n\n Without \".owner \u003d THIS_MODULE\" it is possible to crash the kernel by\n unloading the Orangefs module while someone is reading debugfs files\"\n\n* tag \u0027for-linus-4.9-rc5-ofs-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hubcap/linux:\n orangefs: add .owner to debugfs file_operations\n" }, { "commit": "5d1904204c99596b50a700f092fe49d78edba400", "tree": "c51b0321e4dd99246d4c61bcb1d7e38fa47aec08", "parents": [ "961b708e95181041f403251f660bc70be3ff6ba3" ], "author": { "name": "Aaron Lu", "email": "aaron.lu@intel.com", "time": "Thu Nov 10 17:16:33 2016 +0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Nov 17 09:46:56 2016 -0800" }, "message": "mremap: fix race between mremap() and page cleanning\n\nPrior to 3.15, there was a race between zap_pte_range() and\npage_mkclean() where writes to a page could be lost. Dave Hansen\ndiscovered by inspection that there is a similar race between\nmove_ptes() and page_mkclean().\n\nWe\u0027ve been able to reproduce the issue by enlarging the race window with\na msleep(), but have not been able to hit it without modifying the code.\nSo, we think it\u0027s a real issue, but is difficult or impossible to hit in\npractice.\n\nThe zap_pte_range() issue is fixed by commit 1cf35d47712d(\"mm: split\n\u0027tlb_flush_mmu()\u0027 into tlb flushing and memory freeing parts\"). And\nthis patch is to fix the race between page_mkclean() and mremap().\n\nHere is one possible way to hit the race: suppose a process mmapped a\nfile with READ | WRITE and SHARED, it has two threads and they are bound\nto 2 different CPUs, e.g. CPU1 and CPU2. mmap returned X, then thread\n1 did a write to addr X so that CPU1 now has a writable TLB for addr X\non it. Thread 2 starts mremaping from addr X to Y while thread 1\ncleaned the page and then did another write to the old addr X again.\nThe 2nd write from thread 1 could succeed but the value will get lost.\n\n thread 1 thread 2\n (bound to CPU1) (bound to CPU2)\n\n 1: write 1 to addr X to get a\n writeable TLB on this CPU\n\n 2: mremap starts\n\n 3: move_ptes emptied PTE for addr X\n and setup new PTE for addr Y and\n then dropped PTL for X and Y\n\n 4: page laundering for N by doing\n fadvise FADV_DONTNEED. When done,\n pageframe N is deemed clean.\n\n 5: *write 2 to addr X\n\n 6: tlb flush for addr X\n\n 7: munmap (Y, pagesize) to make the\n page unmapped\n\n 8: fadvise with FADV_DONTNEED again\n to kick the page off the pagecache\n\n 9: pread the page from file to verify\n the value. If 1 is there, it means\n we have lost the written 2.\n\n *the write may or may not cause segmentation fault, it depends on\n if the TLB is still on the CPU.\n\nPlease note that this is only one specific way of how the race could\noccur, it didn\u0027t mean that the race could only occur in exact the above\nconfig, e.g. more than 2 threads could be involved and fadvise() could\nbe done in another thread, etc.\n\nFor anonymous pages, they could race between mremap() and page reclaim:\nTHP: a huge PMD is moved by mremap to a new huge PMD, then the new huge\nPMD gets unmapped/splitted/pagedout before the flush tlb happened for\nthe old huge PMD in move_page_tables() and we could still write data to\nit. The normal anonymous page has similar situation.\n\nTo fix this, check for any dirty PTE in move_ptes()/move_huge_pmd() and\nif any, did the flush before dropping the PTL. If we did the flush for\nevery move_ptes()/move_huge_pmd() call then we do not need to do the\nflush in move_pages_tables() for the whole range. But if we didn\u0027t, we\nstill need to do the whole range flush.\n\nAlternatively, we can track which part of the range is flushed in\nmove_ptes()/move_huge_pmd() and which didn\u0027t to avoid flushing the whole\nrange in move_page_tables(). But that would require multiple tlb\nflushes for the different sub-ranges and should be less efficient than\nthe single whole range flush.\n\nKBuild test on my Sandybridge desktop doesn\u0027t show any noticeable change.\nv4.9-rc4:\n real 5m14.048s\n user 32m19.800s\n sys 4m50.320s\n\nWith this commit:\n real 5m13.888s\n user 32m19.330s\n sys 4m51.200s\n\nReported-by: Dave Hansen \u003cdave.hansen@intel.com\u003e\nSigned-off-by: Aaron Lu \u003caaron.lu@intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "680bb946a1ae04fe0ff369a4965f76b48c07dc54", "tree": "be2ca455a6186b86a08e3ab1c22d49748cc6ed75", "parents": [ "4a59015372840a6fc35d7fd40638a9d5dc3ec958" ], "author": { "name": "Abhi Das", "email": "adas@redhat.com", "time": "Wed Nov 16 21:44:23 2016 -0600" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Nov 17 00:00:48 2016 -0500" }, "message": "fix iov_iter_advance() for ITER_PIPE\n\niov_iter_advance() needs to decrement iter-\u003ecount by the number of\nbytes we\u0027d moved beyond. Normal flavours do that, but ITER_PIPE\ndoesn\u0027t and ITER_PIPE generic_file_read_iter() for O_DIRECT files\nends up with a bogus fallback to page cache read, resulting in incorrect\nvalues for file offset and bytes read.\n\nSigned-off-by: Abhi Das \u003cadas@redhat.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "4a59015372840a6fc35d7fd40638a9d5dc3ec958", "tree": "5116875c49182649c9230d0fee6e9b53c58abf3e", "parents": [ "961b708e95181041f403251f660bc70be3ff6ba3" ], "author": { "name": "Andreas Gruenbacher", "email": "agruenba@redhat.com", "time": "Sun Nov 13 21:23:34 2016 +0100" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Nov 17 00:00:23 2016 -0500" }, "message": "xattr: Fix setting security xattrs on sockfs\n\nThe IOP_XATTR flag is set on sockfs because sockfs supports getting the\n\"system.sockprotoname\" xattr. Since commit 6c6ef9f2, this flag is checked for\nsetxattr support as well. This is wrong on sockfs because security xattr\nsupport there is supposed to be provided by security_inode_setsecurity. The\nsmack security module relies on socket labels (xattrs).\n\nFix this by adding a security xattr handler on sockfs that returns\n-EAGAIN, and by checking for -EAGAIN in setxattr.\n\nWe cannot simply check for -EOPNOTSUPP in setxattr because there are\nfilesystems that neither have direct security xattr support nor support\nvia security_inode_setsecurity. A more proper fix might be to move the\ncall to security_inode_setsecurity into sockfs, but it\u0027s not clear to me\nif that is safe: we would end up calling security_inode_post_setxattr after\nthat as well.\n\nSigned-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "961b708e95181041f403251f660bc70be3ff6ba3", "tree": "e2e3a5d17e25c527c4eabd4359d50cf38030339c", "parents": [ "5fd0f1cae3cced7d3518d22afb4fc7192a0b8fa1", "29ed197333bdb1ccda1790bd2418f3a835de86fd" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 17:24:21 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 17:24:21 2016 -0800" }, "message": "Merge tag \u0027drm-fixes-for-v4.9-rc6\u0027 of git://people.freedesktop.org/~airlied/linux\n\nPull drm fixes fr9om Dave Airlie:\n \"Fixes for amdgpu, and a bunch of arm drivers.\n\n There seems to be an uptick in the ARM drivers sending things for\n fixes which is good, so I\u0027ve decided to dequeue a bit early, more\n stuff may arrive before the weekend.\n\n This contains mediatek, arcpgu, sunxi, fsl-dcu display controller\n fixes along with 3 amdgpu fixes, one for a fencing issue with\n secondary GPUs\"\n\n* tag \u0027drm-fixes-for-v4.9-rc6\u0027 of git://people.freedesktop.org/~airlied/linux:\n drm/amdgpu:fix vpost_needed routine\n drm/amdgpu/powerplay: drop a redundant NULL check\n drm/amdgpu: Attach exclusive fence to prime exported bo\u0027s. (v5)\n drm/arcpgu: Accommodate adv7511 switch to DRM bridge\n drm/fsl-dcu: disable planes before disabling CRTC\n drm/fsl-dcu: update all registers on flush\n drm/fsl-dcu: do not update when modifying irq registers\n drm/sun4i: Propagate error to the caller\n drm/sun4i: Fix error handling\n drm/mediatek: modify the factor to make the pll_rate set in the 1G-2G range\n drm/mediatek: enhance the HDMI driving current\n drm/mediatek: do mtk_hdmi_send_infoframe after HDMI clock enable\n drm/mediatek: clear IRQ status before enable OVL interrupt\n drm/mediatek: set vblank_disable_allowed to true\n drm/mediatek: fix a typo of OD_CFG to OD_RELAYMODE\n drm/sun4i: rgb: Remove the bridge enable/disable functions\n drm/sun4i: rgb: Enable panel after controller\n" }, { "commit": "5c6b2aaf9316fd0983c0c999d920306ddc65bd2d", "tree": "f1f6f98a80c3b67904bd00dc8329f46eef32abb7", "parents": [ "4ff522ea47944ffd3d4d27023ace8bc6a722c834" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Thu Nov 03 12:09:38 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:10:36 2016 -0500" }, "message": "iw_cxgb4: invalidate the mr when posting a read_w_inv wr\n\nAlso, rearrange things a bit to have a common c4iw_invalidate_mr()\nfunction used everywhere that we need to invalidate.\n\nFixes: 49b53a93a64a (\"iw_cxgb4: add fast-path for small REG_MR operations\")\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "4ff522ea47944ffd3d4d27023ace8bc6a722c834", "tree": "8f3a55cb119b4a0fc58485eb0368375a38aad56e", "parents": [ "6fa1f2f0aa6191193704b9ff10e5a2cafe540738" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Tue Oct 18 14:04:39 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:10:36 2016 -0500" }, "message": "iw_cxgb4: set *bad_wr for post_send/post_recv errors\n\nThere are a few cases in c4iw_post_send() and c4iw_post_receive()\nwhere *bad_wr is not set when an error is returned. This can\ncause a crash if the application tries to use bad_wr.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "6fa1f2f0aa6191193704b9ff10e5a2cafe540738", "tree": "86398ad896bd0bfbf173b86f0d01210e47848537", "parents": [ "2b16056f845207967a32497f41cf92b57849f934", "6d931308f55faaef3f30bd0346c47f99528b229d" ], "author": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:05:10 2016 -0500" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:05:10 2016 -0500" }, "message": "Merge branches \u0027hfi1\u0027 and \u0027mlx\u0027 into k.o/for-4.9-rc\n" }, { "commit": "6d931308f55faaef3f30bd0346c47f99528b229d", "tree": "455fb5bd1eeadf3572275aabcc57e47e4c1933c2", "parents": [ "aa75b07b478a774b1432e2df1be5cd8ae834de0f" ], "author": { "name": "Yonatan Cohen", "email": "yonatanc@mellanox.com", "time": "Wed Nov 16 10:39:18 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/rxe: Update qp state for user query\n\nThe method rxe_qp_error() transitions QP to error state\nand make sure the QP is drained. It did not though update\nthe QP state for user\u0027s query.\n\nThis patch fixes this.\n\nFixes: 8700e3e7c485 (\"Soft RoCE driver\")\nSigned-off-by: Yonatan Cohen \u003cyonatanc@mellanox.com\u003e\nReviewed-by: Moni Shoua \u003cmonis@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "aa75b07b478a774b1432e2df1be5cd8ae834de0f", "tree": "61258e0d529054999d30e085d1609ed4f4c5e0ba", "parents": [ "002e062e13db10973adb8302f231e48b477c7ccf" ], "author": { "name": "Yonatan Cohen", "email": "yonatanc@mellanox.com", "time": "Wed Nov 16 10:39:17 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/rxe: Clear queue buffer when modifying QP to reset\n\nRXE resets the send-q only once in rxe_qp_init_req() when\nQP is created, but when the QP is reused after QP reset, the send-q\nholds previous garbage data.\n\nThis garbage data wrongly fails CQEs that otherwise\nshould have completed successfully.\n\nFixes: 8700e3e7c485 (\"Soft RoCE driver\")\nSigned-off-by: Yonatan Cohen \u003cyonatanc@mellanox.com\u003e\nReviewed-by: Moni Shoua \u003cmonis@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "002e062e13db10973adb8302f231e48b477c7ccf", "tree": "d8e35b3e448a1008581af2db04de0aa8c87a86d7", "parents": [ "1454ca3a97e147bb91e98b087446c39cf6692a48" ], "author": { "name": "Yonatan Cohen", "email": "yonatanc@mellanox.com", "time": "Wed Nov 16 10:39:15 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/rxe: Fix handling of erroneous WR\n\nTo correctly handle a erroneous WR this fix does the following\n1. Make sure the bad WQE causes a user completion event.\n2. Call rxe_completer to handle the erred WQE.\n\nBefore the fix, when rxe_requester found a bad WQE, it changed its\nstatus to IB_WC_LOC_PROT_ERR and exit with 0 for non RC QPs.\n\nIf this was the 1st WQE then there would be no ACK to invoke the\ncompleter and this bad WQE would be stuck in the QP\u0027s send-q.\n\nOn top of that the requester exiting with 0 caused rxe_do_task to\nendlessly invoke rxe_requester, resulting in a soft-lockup attached\nbelow.\n\nIn case the WQE was not the 1st and rxe_completer did get a chance to\nhandle the bad WQE, it did not cause a complete event since the WQE\u0027s\nIB_SEND_SIGNALED flag was not set.\n\nSetting WQE status to IB_SEND_SIGNALED is subject to IBA spec\nversion 1.2.1, section 10.7.3.1 Signaled Completions.\n\nNMI watchdog: BUG: soft lockup - CPU#7 stuck for 22s!\n[\u003cffffffffa0590145\u003e] ? rxe_pool_get_index+0x35/0xb0 [rdma_rxe]\n[\u003cffffffffa05952ec\u003e] lookup_mem+0x3c/0xc0 [rdma_rxe]\n[\u003cffffffffa0595534\u003e] copy_data+0x1c4/0x230 [rdma_rxe]\n[\u003cffffffffa058c180\u003e] rxe_requester+0x9d0/0x1100 [rdma_rxe]\n[\u003cffffffff8158e98a\u003e] ? kfree_skbmem+0x5a/0x60\n[\u003cffffffffa05962c9\u003e] rxe_do_task+0x89/0xf0 [rdma_rxe]\n[\u003cffffffffa05963e2\u003e] rxe_run_task+0x12/0x30 [rdma_rxe]\n[\u003cffffffffa059110a\u003e] rxe_post_send+0x41a/0x550 [rdma_rxe]\n[\u003cffffffff811ef922\u003e] ? __kmalloc+0x182/0x200\n[\u003cffffffff816ba512\u003e] ? down_read+0x12/0x40\n[\u003cffffffffa054bd32\u003e] ib_uverbs_post_send+0x532/0x540 [ib_uverbs]\n[\u003cffffffff815f8722\u003e] ? tcp_sendmsg+0x402/0xb80\n[\u003cffffffffa05453dc\u003e] ib_uverbs_write+0x18c/0x3f0 [ib_uverbs]\n[\u003cffffffff81623c2e\u003e] ? inet_recvmsg+0x7e/0xb0\n[\u003cffffffff8158764d\u003e] ? sock_recvmsg+0x3d/0x50\n[\u003cffffffff81215b87\u003e] __vfs_write+0x37/0x140\n[\u003cffffffff81216892\u003e] vfs_write+0xb2/0x1b0\n[\u003cffffffff81217ce5\u003e] SyS_write+0x55/0xc0\n[\u003cffffffff816bc672\u003e] entry_SYSCALL_64_fastpath+0x1a/0xa\n\nFixes: 8700e3e7c485 (\"Soft RoCE driver\")\nSigned-off-by: Yonatan Cohen \u003cyonatanc@mellanox.com\u003e\nReviewed-by: Moni Shoua \u003cmonis@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "1454ca3a97e147bb91e98b087446c39cf6692a48", "tree": "dc6d36e49d54258a9a7d54428539a130271fb1ab", "parents": [ "593ff73bcfdc79f79a8a0df55504f75ad3e5d1a9" ], "author": { "name": "Yonatan Cohen", "email": "yonatanc@mellanox.com", "time": "Wed Nov 16 10:39:14 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/rxe: Fix kernel panic in UDP tunnel with GRO and RX checksum\n\nMissing initialization of udp_tunnel_sock_cfg causes to following\nkernel panic, while kernel tries to execute gro_receive().\n\nWhile being there, we converted udp_port_cfg to use the same\ninitialization scheme as udp_tunnel_sock_cfg.\n\n------------[ cut here ]------------\nkernel tried to execute NX-protected page - exploit attempt? (uid: 0)\nBUG: unable to handle kernel paging request at ffffffffa0588c50\nIP: [\u003cffffffffa0588c50\u003e] __this_module+0x50/0xffffffffffff8400 [ib_rxe]\nPGD 1c09067 PUD 1c0a063 PMD bb394067 PTE 80000000ad5e8163\nOops: 0011 [#1] SMP\nModules linked in: ib_rxe ip6_udp_tunnel udp_tunnel\nCPU: 5 PID: 0 Comm: swapper/5 Not tainted 4.7.0-rc3+ #2\nHardware name: Red Hat KVM, BIOS Bochs 01/01/2011\ntask: ffff880235e4e680 ti: ffff880235e68000 task.ti: ffff880235e68000\nRIP: 0010:[\u003cffffffffa0588c50\u003e]\n[\u003cffffffffa0588c50\u003e] __this_module+0x50/0xffffffffffff8400 [ib_rxe]\nRSP: 0018:ffff880237343c80 EFLAGS: 00010282\nRAX: 00000000dffe482d RBX: ffff8800ae330900 RCX: 000000002001b712\nRDX: ffff8800ae330900 RSI: ffff8800ae102578 RDI: ffff880235589c00\nRBP: ffff880237343cb0 R08: 0000000000000000 R09: 0000000000000000\nR10: 0000000000000000 R11: 0000000000000000 R12: ffff8800ae33e262\nR13: ffff880235589c00 R14: 0000000000000014 R15: ffff8800ae102578\nFS: 0000000000000000(0000) GS:ffff880237340000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: ffffffffa0588c50 CR3: 0000000001c06000 CR4: 00000000000006e0\nStack:\nffffffff8160860e ffff8800ae330900 ffff8800ae102578 0000000000000014\n000000000000004e ffff8800ae102578 ffff880237343ce0 ffffffff816088fb\n0000000000000000 ffff8800ae330900 0000000000000000 00000000ffad0000\nCall Trace:\n\u003cIRQ\u003e\n[\u003cffffffff8160860e\u003e] ? udp_gro_receive+0xde/0x130\n[\u003cffffffff816088fb\u003e] udp4_gro_receive+0x10b/0x2d0\n[\u003cffffffff81611373\u003e] inet_gro_receive+0x1d3/0x270\n[\u003cffffffff81594e29\u003e] dev_gro_receive+0x269/0x3b0\n[\u003cffffffff81595188\u003e] napi_gro_receive+0x38/0x120\n[\u003cffffffffa011caee\u003e] mlx5e_handle_rx_cqe+0x27e/0x340 [mlx5_core]\n[\u003cffffffffa011d076\u003e] mlx5e_poll_rx_cq+0x66/0x6d0 [mlx5_core]\n[\u003cffffffffa011d7ae\u003e] mlx5e_napi_poll+0x8e/0x400 [mlx5_core]\n[\u003cffffffff815949a0\u003e] net_rx_action+0x160/0x380\n[\u003cffffffff816a9197\u003e] __do_softirq+0xd7/0x2c5\n[\u003cffffffff81085c35\u003e] irq_exit+0xf5/0x100\n[\u003cffffffff816a8f16\u003e] do_IRQ+0x56/0xd0\n[\u003cffffffff816a6dcc\u003e] common_interrupt+0x8c/0x8c\n\u003cEOI\u003e\n[\u003cffffffff81061f96\u003e] ? native_safe_halt+0x6/0x10\n[\u003cffffffff81037ade\u003e] default_idle+0x1e/0xd0\n[\u003cffffffff8103828f\u003e] arch_cpu_idle+0xf/0x20\n[\u003cffffffff810c37dc\u003e] default_idle_call+0x3c/0x50\n[\u003cffffffff810c3b13\u003e] cpu_startup_entry+0x323/0x3c0\n[\u003cffffffff81050d8c\u003e] start_secondary+0x15c/0x1a0\nRIP [\u003cffffffffa0588c50\u003e] __this_module+0x50/0xffffffffffff8400 [ib_rxe]\nRSP \u003cffff880237343c80\u003e\nCR2: ffffffffa0588c50\n---[ end trace 489ee31fa7614ac5 ]---\nKernel panic - not syncing: Fatal exception in interrupt\nKernel Offset: disabled\n---[ end Kernel panic - not syncing: Fatal exception in interrupt\n------------[ cut here ]------------\n\nFixes: 8700e3e7c485 (\"Soft RoCE driver\")\nSigned-off-by: Yonatan Cohen \u003cyonatanc@mellanox.com\u003e\nReviewed-by: Moni Shoua \u003cmonis@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "593ff73bcfdc79f79a8a0df55504f75ad3e5d1a9", "tree": "ee53147d4228608b6c83ba2d9fde51916da8af2c", "parents": [ "37995116fecfce2b61ee3da6e73b3e394c6818f9" ], "author": { "name": "Matan Barak", "email": "matanb@mellanox.com", "time": "Thu Nov 10 11:30:55 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx4: Fix create CQ error flow\n\nCurrently, if ib_copy_to_udata fails, the CQ\nwon\u0027t be deleted from the radix tree and the HW (HW2SW).\n\nFixes: 225c7b1feef1 (\u0027IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters\u0027)\nSigned-off-by: Matan Barak \u003cmatanb@mellanox.com\u003e\nSigned-off-by: Daniel Jurgens \u003cdanielj@mellanox.com\u003e\nReviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "37995116fecfce2b61ee3da6e73b3e394c6818f9", "tree": "53f4cec46b40bd9e9ea13bbe2c18dd8f40b15b1e", "parents": [ "a1ab8402d15d2305d2315d96ec3294bfdf16587e" ], "author": { "name": "Daniel Jurgens", "email": "danielj@mellanox.com", "time": "Thu Nov 10 11:30:54 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx4: Check gid_index return value\n\nCheck the returned GID index value and return an error if it is invalid.\n\nFixes: 5070cd2239bd (\u0027IB/mlx4: Replace mechanism for RoCE GID management\u0027)\nSigned-off-by: Daniel Jurgens \u003cdanielj@mellanox.com\u003e\nReviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nReviewed-by: Yuval Shaia \u003cyuval.shaia@oracle.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "a1ab8402d15d2305d2315d96ec3294bfdf16587e", "tree": "de32b4419cf2121d2a3d66d6667ef37f65d54d95", "parents": [ "dbaaff2a2caa03d472b5cc53a3fbfd415c97dc26" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.com", "time": "Thu Oct 27 16:36:46 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx5: Fix NULL pointer dereference on debug print\n\nFor XRC QP CQs may not exist. Check before attempting dereference.\n\nFixes: e126ba97dba9 (\u0027mlx5: Add driver for Mellanox Connect-IB adapters\u0027)\nSigned-off-by: Eli Cohen \u003celi@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nReviewed-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "dbaaff2a2caa03d472b5cc53a3fbfd415c97dc26", "tree": "c48f2bee323daccd92b2331cd5f538b7c4e8e0f1", "parents": [ "6bc1a656ab9f57f0112823b4a36930c9a29d1f89" ], "author": { "name": "Eli Cohen", "email": "eli@mellanox.com", "time": "Thu Oct 27 16:36:44 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx5: Fix fatal error dispatching\n\nWhen an internal error condition is detected, make sure to set the\ndevice inactive after dispatching the event so ULPs can get a\nnotification of this event.\n\nFixes: e126ba97dba9 (\u0027mlx5: Add driver for Mellanox Connect-IB adapters\u0027)\nSigned-off-by: Eli Cohen \u003celi@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nReviewed-by: Mohamad Haj Yahia \u003cmohamad@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "6bc1a656ab9f57f0112823b4a36930c9a29d1f89", "tree": "6e76546703284ab7d3e1a25c389b31511ff930f8", "parents": [ "16b0e0695a73b68d8ca40288c8f9614ef208917b" ], "author": { "name": "Moshe Lazer", "email": "moshel@mellanox.com", "time": "Thu Oct 27 16:36:42 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx5: Resolve soft lock on massive reg MRs\n\nWhen calling reg_mr of large MRs (e.g. 4GB) from multiple processes\nand MR caches can\u0027t supply the required amount of MRs the slow-path\nof MR allocation may be used. In this case we need to serialize the\nslow-path between the processes to avoid soft lock.\n\nFixes: e126ba97dba9 (\u0027mlx5: Add driver for Mellanox Connect-IB adapters\u0027)\nSigned-off-by: Moshe Lazer \u003cmoshel@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nReviewed-by: Eli Cohen \u003celi@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "16b0e0695a73b68d8ca40288c8f9614ef208917b", "tree": "33cc55e4a4a7ac46c66cabfdf8be6379b60b39bd", "parents": [ "efd7f40082a0dfd112eb87ff2124467a5739216f" ], "author": { "name": "Daniel Jurgens", "email": "danielj@mellanox.com", "time": "Thu Oct 27 16:36:41 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx5: Use cache line size to select CQE stride\n\nWhen creating kernel CQs use 128B CQE stride if the\ncache line size is 128B, 64B otherwise. This prevents\nmultiple CQEs from residing in a 128B cache line,\nwhich can cause retries when there are concurrent\nread and writes in one cache line.\n\nTested with IPoIB on PPC64, saw ~5% throughput\nimprovement.\n\nFixes: e126ba97dba9 (\u0027mlx5: Add driver for Mellanox Connect-IB adapters\u0027)\nSigned-off-by: Daniel Jurgens \u003cdanielj@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "efd7f40082a0dfd112eb87ff2124467a5739216f", "tree": "d7de053538cf744c5eb85853a02705d020551f91", "parents": [ "90be7c8ab72853ff9fc407f01518a898df1f3045" ], "author": { "name": "Maor Gottlieb", "email": "maorg@mellanox.com", "time": "Thu Oct 27 16:36:40 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx5: Validate requested RQT size\n\nValidate that the requested size of RQT is supported by firmware.\n\nFixes: c5f9092936fe (\u0027IB/mlx5: Add Receive Work Queue Indirection table operations\u0027)\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nReviewed-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "90be7c8ab72853ff9fc407f01518a898df1f3045", "tree": "e1c741e4146a24b3601cb87280cbf3b8c3b8c3fd", "parents": [ "3c7ba5760ab8eedec01159b267bb9bfcffe522ac" ], "author": { "name": "Majd Dibbiny", "email": "majd@mellanox.com", "time": "Thu Oct 27 16:36:39 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/mlx5: Fix memory leak in query device\n\nWe need to free dev-\u003eport when we fail to enable RoCE or\ninitialize node data.\n\nFixes: 0837e86a7a34 (\u0027IB/mlx5: Add per port counters\u0027)\nSigned-off-by: Majd Dibbiny \u003cmajd@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nReviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "3c7ba5760ab8eedec01159b267bb9bfcffe522ac", "tree": "36e6c1ab64e36f899718b69ce3b6b989c2c13bf1", "parents": [ "61c3702863be9e9f1ef12ed5a5b17bae6cdfac0b" ], "author": { "name": "Mark Bloch", "email": "markb@mellanox.com", "time": "Thu Oct 27 16:36:31 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/core: Avoid unsigned int overflow in sg_alloc_table\n\nsg_alloc_table gets unsigned int as parameter while the driver\nreturns it as size_t. Check npages isn\u0027t greater than maximum\nunsigned int.\n\nFixes: eeb8461e36c9 (\"IB: Refactor umem to use linear SG table\")\nSigned-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "61c3702863be9e9f1ef12ed5a5b17bae6cdfac0b", "tree": "fe6aa837a5642a25c3cfa070bcc638e043f40104", "parents": [ "aeb76df46d1158d5f7f3d30f993a1bb6ee9c67a0" ], "author": { "name": "Mark Bloch", "email": "markb@mellanox.com", "time": "Thu Oct 27 16:36:29 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/core: Add missing check for addr_resolve callback return value\n\nWhen calling rdma_resolve_ip inside rdma_addr_find_l2_eth_by_grh,\nthe return status of the request was ignored in the callback function\ncausing a successful return and an empty dmac.\n\nSigned-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Alex Vesker \u003cvalex@mellanox.com\u003e\nReviewed-by: Or Gerlitz \u003cogerlitz@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "aeb76df46d1158d5f7f3d30f993a1bb6ee9c67a0", "tree": "3f27e2ae43ab694602bc31756b9de1fd7788868d", "parents": [ "9db0ff53cb9b43ed75bacd42a89c1a0ab048b2b0" ], "author": { "name": "Leon Romanovsky", "email": "leon@kernel.org", "time": "Mon Oct 31 07:50:56 2016 +0200" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/core: Set routable RoCE gid type for ipv4/ipv6 networks\n\nOn Thu, Oct 27, 2016 at 04:36:28PM +0300, Leon Romanovsky wrote:\n\u003e From: Mark Bloch \u003cmarkb@mellanox.com\u003e\n\u003e\n\u003e If the underlying netowrk type is ipv4 or ipv6 and the device supports\n\u003e routable RoCE, prefer it so the traffic could cross subnets.\n\u003e\n\u003e Signed-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n\u003e Signed-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\n\u003e Signed-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\n\u003e ---\n\nHi Doug,\n\nPlease take the following v1 of this patch where I fixed spelling error\nfrom \"netowrk\" to be \"network\".\n\nThanks.\n\n\u003eFrom 09f96ba3e9b4442cfb44dca04c6726e55525c9c3 Mon Sep 17 00:00:00 2001\nFrom: Mark Bloch \u003cmarkb@mellanox.com\u003e\nDate: Sun, 11 Sep 2016 06:25:10 +0000\nSubject: [PATCH rdma-rc v1 3/6] IB/core: Set routable RoCE gid type for ipv4/ipv6\n networks\n\nIf the underlying network type is ipv4 or ipv6 and the device supports\nroutable RoCE, prefer it so the traffic could cross subnets.\n\nSigned-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "9db0ff53cb9b43ed75bacd42a89c1a0ab048b2b0", "tree": "36bd48fc3624afcea95a09e9b9404bba18aa1181", "parents": [ "5b810a242c28e1d8d64d718cebe75b79d86a0b2d" ], "author": { "name": "Mark Bloch", "email": "markb@mellanox.com", "time": "Thu Oct 27 16:36:27 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/cm: Mark stale CM id\u0027s whenever the mad agent was unregistered\n\nWhen there is a CM id object that has port assigned to it, it means that\nthe cm-id asked for the specific port that it should go by it, but if\nthat port was removed (hot-unplug event) the cm-id was not updated.\nIn order to fix that the port keeps a list of all the cm-id\u0027s that are\nplanning to go by it, whenever the port is removed it marks all of them\nas invalid.\n\nThis commit fixes a kernel panic which happens when running traffic between\nguests and we force reboot a guest mid traffic, it triggers a kernel panic:\n\n Call Trace:\n [\u003cffffffff815271fa\u003e] ? panic+0xa7/0x16f\n [\u003cffffffff8152b534\u003e] ? oops_end+0xe4/0x100\n [\u003cffffffff8104a00b\u003e] ? no_context+0xfb/0x260\n [\u003cffffffff81084db2\u003e] ? del_timer_sync+0x22/0x30\n [\u003cffffffff8104a295\u003e] ? __bad_area_nosemaphore+0x125/0x1e0\n [\u003cffffffff81084240\u003e] ? process_timeout+0x0/0x10\n [\u003cffffffff8104a363\u003e] ? bad_area_nosemaphore+0x13/0x20\n [\u003cffffffff8104aabf\u003e] ? __do_page_fault+0x31f/0x480\n [\u003cffffffff81065df0\u003e] ? default_wake_function+0x0/0x20\n [\u003cffffffffa0752675\u003e] ? free_msg+0x55/0x70 [mlx5_core]\n [\u003cffffffffa0753434\u003e] ? cmd_exec+0x124/0x840 [mlx5_core]\n [\u003cffffffff8105a924\u003e] ? find_busiest_group+0x244/0x9f0\n [\u003cffffffff8152d45e\u003e] ? do_page_fault+0x3e/0xa0\n [\u003cffffffff8152a815\u003e] ? page_fault+0x25/0x30\n [\u003cffffffffa024da25\u003e] ? cm_alloc_msg+0x35/0xc0 [ib_cm]\n [\u003cffffffffa024e821\u003e] ? ib_send_cm_dreq+0xb1/0x1e0 [ib_cm]\n [\u003cffffffffa024f836\u003e] ? cm_destroy_id+0x176/0x320 [ib_cm]\n [\u003cffffffffa024fb00\u003e] ? ib_destroy_cm_id+0x10/0x20 [ib_cm]\n [\u003cffffffffa034f527\u003e] ? ipoib_cm_free_rx_reap_list+0xa7/0x110 [ib_ipoib]\n [\u003cffffffffa034f590\u003e] ? ipoib_cm_rx_reap+0x0/0x20 [ib_ipoib]\n [\u003cffffffffa034f5a5\u003e] ? ipoib_cm_rx_reap+0x15/0x20 [ib_ipoib]\n [\u003cffffffff81094d20\u003e] ? worker_thread+0x170/0x2a0\n [\u003cffffffff8109b2a0\u003e] ? autoremove_wake_function+0x0/0x40\n [\u003cffffffff81094bb0\u003e] ? worker_thread+0x0/0x2a0\n [\u003cffffffff8109aef6\u003e] ? kthread+0x96/0xa0\n [\u003cffffffff8100c20a\u003e] ? child_rip+0xa/0x20\n [\u003cffffffff8109ae60\u003e] ? kthread+0x0/0xa0\n [\u003cffffffff8100c200\u003e] ? child_rip+0x0/0x20\n\nFixes: a977049dacde (\"[PATCH] IB: Add the kernel CM implementation\")\nSigned-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\nSigned-off-by: Erez Shitrit \u003cerezsh@mellanox.com\u003e\nReviewed-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "5b810a242c28e1d8d64d718cebe75b79d86a0b2d", "tree": "f767f7da2c8d3c27866a813a6facdfcffa853f0c", "parents": [ "a909d3e636995ba7c349e2ca5dbb528154d4ac30" ], "author": { "name": "Tariq Toukan", "email": "tariqt@mellanox.com", "time": "Thu Oct 27 16:36:26 2016 +0300" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Wed Nov 16 20:03:44 2016 -0500" }, "message": "IB/uverbs: Fix leak of XRC target QPs\n\nThe real QP is destroyed in case of the ref count reaches zero, but\nfor XRC target QPs this call was missed and caused to QP leaks.\n\nLet\u0027s call to destroy for all flows.\n\nFixes: 0e0ec7e0638e (\u0027RDMA/core: Export ib_open_qp() to share XRC...\u0027)\nSigned-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nSigned-off-by: Noa Osherovich \u003cnoaos@mellanox.com\u003e\nSigned-off-by: Leon Romanovsky \u003cleon@kernel.org\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "5fd0f1cae3cced7d3518d22afb4fc7192a0b8fa1", "tree": "8b875f609809154b22b96ef8072d80934395e648", "parents": [ "984573abf8d09bace3cf8cda224bacb75b4c61d2", "709fb1f961ea5c287107c3f903e81c9529224c8b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 16:39:01 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 16:39:01 2016 -0800" }, "message": "Merge tag \u0027xtensa-20161116\u0027 of git://github.com/jcmvbkbc/linux-xtensa\n\nPull Xtensa fixes from Max Filippov:\n\n - fix register dumps, stack dumps and stack traces that got torn due to\n recent printk changes\n\n - wire up pkey_{mprotect,alloc,free} syscalls\n\n* tag \u0027xtensa-20161116\u0027 of git://github.com/jcmvbkbc/linux-xtensa:\n xtensa: wire up new pkey_{mprotect,alloc,free} syscalls\n xtensa: clean up printk usage for boot/crash logging\n" }, { "commit": "29ed197333bdb1ccda1790bd2418f3a835de86fd", "tree": "8f68d7745a19b6a92ff5b7c065d2d20209989ae0", "parents": [ "51a4c38a5511c0027c54d330f7dd2239f6c95b82", "1da2c326e43b0834105993d13610647337bbad67" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 17 09:45:27 2016 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 17 09:45:27 2016 +1000" }, "message": "Merge branch \u0027drm-fixes-4.9\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\nJust a few bug fixes for 4.9. The big one is Mario\u0027s prime fencing fix.\n\n* \u0027drm-fixes-4.9\u0027 of git://people.freedesktop.org/~agd5f/linux:\n drm/amdgpu:fix vpost_needed routine\n drm/amdgpu/powerplay: drop a redundant NULL check\n drm/amdgpu: Attach exclusive fence to prime exported bo\u0027s. (v5)\n" }, { "commit": "51a4c38a5511c0027c54d330f7dd2239f6c95b82", "tree": "042e58dea131b78fa59adb5192001c0e60fda05c", "parents": [ "94ea29b116652b8b08934493fae68a6b83e2bc45", "0d2200794f0a2c1ebb3b6613842914d8ce4b67f9" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 17 09:44:52 2016 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Nov 17 09:44:52 2016 +1000" }, "message": "Merge branch \u0027mediatek-drm-fixes-2016-11-11\u0027 of https://github.com/ckhu-mediatek/linux.git-tags into drm-fixes\n\nThis branch include one patch to fix a typo, two patches to disable\nvblank interrupt, and three patches to support HDMI 4K resolution.\n\n* \u0027mediatek-drm-fixes-2016-11-11\u0027 of https://github.com/ckhu-mediatek/linux.git-tags:\n drm/mediatek: modify the factor to make the pll_rate set in the 1G-2G range\n drm/mediatek: enhance the HDMI driving current\n drm/mediatek: do mtk_hdmi_send_infoframe after HDMI clock enable\n drm/mediatek: clear IRQ status before enable OVL interrupt\n drm/mediatek: set vblank_disable_allowed to true\n drm/mediatek: fix a typo of OD_CFG to OD_RELAYMODE\n" }, { "commit": "d48756228ee9161ac8836b346589a43fabdc9f3c", "tree": "9f1c735aa5b75345e71baca7a3bcbe19c8edc568", "parents": [ "959401aa2b8cc6422809762a6ba4b2635154e9a6" ], "author": { "name": "Keith Busch", "email": "keith.busch@intel.com", "time": "Tue Nov 15 15:56:26 2016 -0500" }, "committer": { "name": "Jens Axboe", "email": "axboe@fb.com", "time": "Wed Nov 16 12:39:57 2016 -0700" }, "message": "nvme/pci: Don\u0027t free queues on error\n\nThe nvme_remove function tears down all allocated resources in the correct\norder, so no need to free queues on error during initialization. This\nfixes possible use-after-free errors when queues are still associated\nwith a blk-mq hctx.\n\nReported-by: Scott Bauer \u003cscott.bauer@intel.com\u003e\nTested-by: Scott Bauer \u003cscott.bauer@intel.com\u003e\nSigned-off-by: Keith Busch \u003ckeith.busch@intel.com\u003e\nReviewed-by: Sagi Grimberg \u003csagi@grimbeg.me\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Jens Axboe \u003caxboe@fb.com\u003e\n" }, { "commit": "984573abf8d09bace3cf8cda224bacb75b4c61d2", "tree": "3f3ca88a40facc0552fc204a154c9a8f1fc4394b", "parents": [ "116fc01f2ed7578e70ea85c67f6507ae50a5932e", "59c3b76cc61d1d676f965c192cc7969aa5cb2744" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 09:20:10 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 09:20:10 2016 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse\n\nPull fuse fixes from Miklos Szeredi:\n \"A regression fix and bug fix bound for stable\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:\n fuse: fix fuse_write_end() if zero bytes were copied\n fuse: fix root dentry initialization\n" }, { "commit": "116fc01f2ed7578e70ea85c67f6507ae50a5932e", "tree": "d77c7bb464e458b88e420ffa0b7ea84d36744ef1", "parents": [ "81bcfe5e48f9b8c42cf547f1c74c7f60c44c34c8", "722f191080de641f023feaa7d5648caf377844f5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 09:09:00 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Nov 16 09:09:00 2016 -0800" }, "message": "Merge tag \u0027mfd-fixes-4.9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd\n\nPull MFD fixes from Lee Jones:\n - Fix PCI properties in intel-lpss-pci\n - Fix Resetting issue during suspend in intel-lpss-pci\n - Seperate IRQs for USBC device and CHRG in intel_soc_pmic_bxtwc\n - Add timeout to fix Resetting issue in stmpe\n - Ensure we \u0027put\u0027 reference to device when done in mfd-core\n\n* tag \u0027mfd-fixes-4.9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd:\n mfd: core: Fix device reference leak in mfd_clone_cell\n mfd: stmpe: Fix RESET regression on STMPE2401\n mfd: intel_soc_pmic_bxtwc: Fix usbc interrupt\n mfd: intel-lpss: Do not put device in reset state on suspend\n mfd: lpss: Fix Intel Kaby Lake PCH-H properties\n" }, { "commit": "19ff7fcc76e6911a955742b40f85ba1030ccba5e", "tree": "8e844e7bf2bb21b0d21e0d769e72b6243f7ef2e9", "parents": [ "dc0336214eb07ee9de2a41dd4c81c744ffa419ac" ], "author": { "name": "Mike Marshall", "email": "hubcap@omnibond.com", "time": "Wed Nov 16 11:52:19 2016 -0500" }, "committer": { "name": "Mike Marshall", "email": "hubcap@omnibond.com", "time": "Wed Nov 16 11:52:19 2016 -0500" }, "message": "orangefs: add .owner to debugfs file_operations\n\nWithout \".owner \u003d THIS_MODULE\" it is possible to crash the kernel\nby unloading the Orangefs module while someone is reading debugfs\nfiles.\n\nSigned-off-by: Mike Marshall \u003chubcap@omnibond.com\u003e\n" }, { "commit": "722f191080de641f023feaa7d5648caf377844f5", "tree": "85e0ffd8bbfb5148ca2bb6b59704e079a8a8497d", "parents": [ "f40584200bc4af7aa4399635b9ac213c62a13ae7" ], "author": { "name": "Johan Hovold", "email": "johan@kernel.org", "time": "Tue Nov 01 11:38:18 2016 +0100" }, "committer": { "name": "Lee Jones", "email": "lee.jones@linaro.org", "time": "Wed Nov 16 09:50:41 2016 +0000" }, "message": "mfd: core: Fix device reference leak in mfd_clone_cell\n\nMake sure to drop the reference taken by bus_find_device_by_name()\nbefore returning from mfd_clone_cell().\n\nFixes: a9bbba996302 (\"mfd: add platform_device sharing support for mfd\")\nSigned-off-by: Johan Hovold \u003cjohan@kernel.org\u003e\nSigned-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n" }, { "commit": "f40584200bc4af7aa4399635b9ac213c62a13ae7", "tree": "b165c82493622e97536f4bd5cae953a6b050ab1e", "parents": [ "9600702082b29fd3f8a6d744df74ad4c48d4a432" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Tue Nov 01 10:22:53 2016 +0100" }, "committer": { "name": "Lee Jones", "email": "lee.jones@linaro.org", "time": "Wed Nov 16 09:50:33 2016 +0000" }, "message": "mfd: stmpe: Fix RESET regression on STMPE2401\n\nSince commit c4dd1ba355aae2bc3d1213da6c66c53e3c31e028\n(\"mfd: stmpe: Add reset support for all STMPE variant\")\nwe\u0027re resetting the STMPE expanders before use.\n\nThis caused a regression on the STMP2401 on the Nomadik\nNHK8815:\n\nstmpe-i2c 0-0043: stmpe2401 detected, chip id: 0x101\nnmk-i2c 101f8000.i2c0: write to slave 0x43 timed out\nnmk-i2c 101f8000.i2c0: no ack received after address transmission\nstmpe-i2c 0-0044: stmpe2401 detected, chip id: 0x101\nnmk-i2c 101f8000.i2c0: write to slave 0x44 timed out\nnmk-i2c 101f8000.i2c0: no ack received after address transmission\n\nIt turns out that we start to poll for the reset bit to\ngo low again too quickly: the STMPE2401 is not yet online and\nready to be asked for the status of the RESET bit.\n\nBy introducing a 10ms delay before starting to hammer\nthe register for information, we get back to normal:\n\nstmpe-i2c 0-0043: stmpe2401 detected, chip id: 0x101\nstmpe-i2c 0-0044: stmpe2401 detected, chip id: 0x101\n\nCc: stable@vger.kernel.org\nCc: Amelie Delaunay \u003camelie.delaunay@st.com\u003e\nFixes: c4dd1ba355aa (\"mfd: stmpe: Add reset support for all STMPE variant\")\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nAcked-by: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\nSigned-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n" }, { "commit": "9600702082b29fd3f8a6d744df74ad4c48d4a432", "tree": "43d893f59255ff116d2c3abb2efd9e6c4aa4b8c9", "parents": [ "274e43edcda6f709aa67e436b3123e45a6270923" ], "author": { "name": "Heikki Krogerus", "email": "heikki.krogerus@linux.intel.com", "time": "Mon Oct 17 10:32:13 2016 +0300" }, "committer": { "name": "Lee Jones", "email": "lee.jones@linaro.org", "time": "Wed Nov 16 09:50:29 2016 +0000" }, "message": "mfd: intel_soc_pmic_bxtwc: Fix usbc interrupt\n\nThe wcove USB Type-C driver is currently being flooded with\ninterrupts that are not targeted to it. The reason for that\nis because all CHRG first level interrupts are mapped to it.\nThis fixes the issue by introducing separate irq for the\nusbc device, and mapping only USB Type-C PHY interrupts to\nit.\n\nFixes: 9c6235c86332 (\"mfd: intel_soc_pmic_bxtwc: Add bxt_wcove_usbc device\")\nSigned-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nSigned-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n" }, { "commit": "274e43edcda6f709aa67e436b3123e45a6270923", "tree": "4c6a3e9c64fe5b6f7d03eebf522846e772866274", "parents": [ "2c8c34167c987e463d62a55384fcec7fa8d03a54" ], "author": { "name": "Azhar Shaikh", "email": "azhar.shaikh@intel.com", "time": "Wed Oct 12 10:12:20 2016 -0700" }, "committer": { "name": "Lee Jones", "email": "lee.jones@linaro.org", "time": "Wed Nov 16 09:50:25 2016 +0000" }, "message": "mfd: intel-lpss: Do not put device in reset state on suspend\n\nCommit 41a3da2b8e163 (\"mfd: intel-lpss: Save register context on\nsuspend\") saved the register context while going to suspend and\nalso put the device in reset state.\n\nDue to the resetting of device, system cannot enter S3/S0ix\nstates when no_console_suspend flag is enabled. The system\nand serial console both hang. The resetting of device is not\nneeded while going to suspend. Hence remove this code.\n\nCc: stable@vger.kernel.org\nFixes: 41a3da2b8e163 (\"mfd: intel-lpss: Save register context on suspend\")\nSigned-off-by: Azhar Shaikh \u003cazhar.shaikh@intel.com\u003e\nAcked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nReviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nSigned-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n" }, { "commit": "2c8c34167c987e463d62a55384fcec7fa8d03a54", "tree": "3a7585e0eba25ea9b87bb675eaeeb6919b2e539d", "parents": [ "1001354ca34179f3db924eb66672442a173147dc" ], "author": { "name": "Jarkko Nikula", "email": "jarkko.nikula@linux.intel.com", "time": "Thu Sep 29 12:59:39 2016 +0300" }, "committer": { "name": "Lee Jones", "email": "lee.jones@linaro.org", "time": "Wed Nov 16 09:50:18 2016 +0000" }, "message": "mfd: lpss: Fix Intel Kaby Lake PCH-H properties\n\nThere are a few issues on Intel Kaby Lake PCH-H properties added by\ncommit a6a576b78e09 (\"mfd: lpss: Add Intel Kaby Lake PCH-H PCI IDs\"):\n\n- Input clock of I2C controller on Intel Kaby Lake PCH-H is 120 MHz not\n 133 MHz. This was probably copy-paste error from Intel Broxton I2C\n properties.\n- There is no default I2C SDA hold time specified which is used when\n ACPI doesn\u0027t provide it. I got information from Windows driver team\n that Kaby Lake PCH-H can use the same configuration than Intel\n Sunrisepoint PCH.\n- Common HS-UART properties are not used.\n\nFix these by reusing the Sunrisepoint properties on Kaby Lake PCH-H.\n\nFixes: a6a576b78e09 (\"mfd: lpss: Add Intel Kaby Lake PCH-H PCI IDs\")\nReported-by: Xiang A Wang \u003cxiang.a.wang@intel.com\u003e\nSigned-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\nAcked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nSigned-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n" }, { "commit": "94ea29b116652b8b08934493fae68a6b83e2bc45", "tree": "871fa934bfab9fca8a09b0c9c2114ba09f2b6bb8", "parents": [ "e2384535a623ed51151d1bbed5263b9f8a87a74d", "4db069a2bf990e278ea57ff615dcaa89b85376bd" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Nov 16 09:41:08 2016 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Wed Nov 16 09:41:08 2016 +1000" }, "message": "Merge tag \u0027sunxi-drm-fixes-for-4.9\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/mripard/linux into drm-fixes\n\nsun4i-drm fixes for 4.9\n\nA few patches to fix our error handling and our panel / bridge calls.\n\n* tag \u0027sunxi-drm-fixes-for-4.9\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/mripard/linux:\n drm/sun4i: Propagate error to the caller\n drm/sun4i: Fix error handling\n drm/sun4i: rgb: Remove the bridge enable/disable functions\n drm/sun4i: rgb: Enable panel after controller\n" }, { "commit": "2b16056f845207967a32497f41cf92b57849f934", "tree": "9b36971b747cedaab6df7673b28a18ad907866ee", "parents": [ "09a7908b1ba616eed349d49058ee909907ee0885" ], "author": { "name": "Dennis Dalessandro", "email": "dennis.dalessandro@intel.com", "time": "Tue Oct 25 13:12:46 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:18:57 2016 -0500" }, "message": "IB/hfi1: Remove incorrect IS_ERR check\n\nRemove IS_ERR check from caching code as the function being called does\nnot actually return error pointers.\n\nFixes: f19bd643dbde: \"IB/hfi1: Prevent NULL pointer deferences in caching code\"\nReported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nReviewed-by: Dean Luick \u003cdean.luick@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "09a7908b1ba616eed349d49058ee909907ee0885", "tree": "93fea89fa543d9c63c1160ee7c34c7c179c444ae", "parents": [ "f2d8a0b367e735ab157222ce74a5f2481216c878" ], "author": { "name": "Jianxin Xiong", "email": "jianxin.xiong@intel.com", "time": "Tue Oct 25 13:12:40 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:18:57 2016 -0500" }, "message": "IB/hfi1: Prevent hardware counter names from being cut off\n\nIncrease the size of the buffer that is used to construct per-VL\nand per-SDMA counter names.\n\nReviewed-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Jianxin Xiong \u003cjianxin.xiong@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "f2d8a0b367e735ab157222ce74a5f2481216c878", "tree": "6c5ddd69f786cdf672944fe175dced45e2e6486e", "parents": [ "505efe3e46d5eaab726295cd023fb86d5b789d00" ], "author": { "name": "Dasaratharaman Chandramouli", "email": "dasaratharaman.chandramouli@intel.com", "time": "Tue Oct 25 13:12:23 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:46 2016 -0500" }, "message": "IB/hfi1: Fix ECN processing in prescan_rxq\n\nWhen processing ECN via the prescan_rxq path, some fields in the packet\nstructure are passed uninitialized. This can potentially\ncause NULL pointer exceptions during ECN handling.\n\nReviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\nReviewed-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Dasaratharaman Chandramouli \u003cdasaratharaman.chandramouli@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "505efe3e46d5eaab726295cd023fb86d5b789d00", "tree": "d9fa0bec1b5c74324b3db29c0dce6d835d446755", "parents": [ "11501ab9df687c6f0852719a5165e16cd3eb3c10" ], "author": { "name": "Jakub Pawlak", "email": "jakub.pawlak@intel.com", "time": "Tue Oct 25 13:12:17 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:45 2016 -0500" }, "message": "IB/hfi1: Fix status error code for unsupported packets\n\nSet the status code BAD_L2 when unsupported type of packet\nis received and dropped.\n\nReviewed-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Jakub Pawlak \u003cjakub.pawlak@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "11501ab9df687c6f0852719a5165e16cd3eb3c10", "tree": "71b8b40f3815c9148000d9afc2e21d8a396e1367", "parents": [ "458ed666fe14a54dfb6690a1a7f541782d1342c9" ], "author": { "name": "Krzysztof Blaszkowski", "email": "krzysztof.blaszkowski@intel.com", "time": "Tue Oct 25 13:12:11 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:45 2016 -0500" }, "message": "IB/hfi1: Relocate rcvhdrcnt module parameter check.\n\nValidate the rcvhdrcnt module parameter in a single function at module\nload time. This allows proper error reporting.\n\nReviewed-by: Dean Luick \u003cdean.luick@intel.com\u003e\nSigned-off-by: Krzysztof Blaszkowski \u003ckrzysztof.blaszkowski@intel.com\u003e\nSigned-off-by: Tymoteusz Kielan \u003ctymoteusz.kielan@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "458ed666fe14a54dfb6690a1a7f541782d1342c9", "tree": "4b962cbefd213f1859fc531a4d446dac2eb8fbca", "parents": [ "f0f98f74c91c68502e97e0d5526aa4e81b40b28a" ], "author": { "name": "Ira Weiny", "email": "ira.weiny@intel.com", "time": "Mon Oct 17 04:20:09 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:44 2016 -0500" }, "message": "IB/hfi1: Fix rnr_timer addition\n\nThe new s_rnr_timeout was not properly being set and the code was\nincorrectly setting a different timer.\n\nFound by code inspection.\n\nCc: \u003cstable@vger.kernel.org\u003e # 4.7.x\nFixes: 08279d5c9424 (\"staging/rdma/hfi1: use new RNR timer\")\nReviewed-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nSigned-off-by: Ira Weiny \u003cira.weiny@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "f0f98f74c91c68502e97e0d5526aa4e81b40b28a", "tree": "d6aba371905e29a14e9d8a0f8f5f0f93b351c94e", "parents": [ "26ea2544ddbe8855cb251e41ff3641c61655a15f" ], "author": { "name": "Easwar Hariharan", "email": "easwar.hariharan@intel.com", "time": "Mon Oct 17 04:20:04 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:44 2016 -0500" }, "message": "IB/hfi1: Delete unused lock\n\nThe lock is an unused vestige from qib. Remove it.\n\nReviewed-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nSigned-off-by: Easwar Hariharan \u003ceaswar.hariharan@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "26ea2544ddbe8855cb251e41ff3641c61655a15f", "tree": "0e5630c042083e8ec1b6313827aed6437abfac09", "parents": [ "eacc830f95c0d8c5cbbda1bdba2ddc8f14bc248d" ], "author": { "name": "Easwar Hariharan", "email": "easwar.hariharan@intel.com", "time": "Mon Oct 17 04:19:58 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:43 2016 -0500" }, "message": "IB/hfi1: Clean up unused argument\n\nhfi1_pcie_ddinit takes the PCI device id as an argument but never\nuses it. Clean it up.\n\nReviewed-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Easwar Hariharan \u003ceaswar.hariharan@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "eacc830f95c0d8c5cbbda1bdba2ddc8f14bc248d", "tree": "77914703d334030d7b85134413e152fb9a1487a8", "parents": [ "4dfe7cceb2bfd98783b4966d7c881a7552932d31" ], "author": { "name": "Dennis Dalessandro", "email": "dennis.dalessandro@intel.com", "time": "Mon Oct 17 04:19:52 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:43 2016 -0500" }, "message": "IB/hfi1: Remove leftover snoop references\n\nA few snoop related variables were missed in the snoop/capture removal\nto get out of staging. Go back and clean those up too.\n\nReviewed-by: Dean Luick \u003cdean.luick@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "4dfe7cceb2bfd98783b4966d7c881a7552932d31", "tree": "18994f0e4d2c7c5da4a0346b92054aea96c33d21", "parents": [ "83fb4af6800deb4f3d19b297df6148cda5c016de" ], "author": { "name": "Jianxin Xiong", "email": "jianxin.xiong@intel.com", "time": "Mon Oct 17 04:19:41 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:42 2016 -0500" }, "message": "IB/hfi1: Fix a potential memory leak in hfi1_create_ctxts()\n\nIn the function hfi1_create_ctxts the array \"dd-\u003ercd\" is allocated and\nthen populated with allocated resources in a loop. Previously, if\nerror happened during the loop, only resource allocated in the current\niteration would be freed. The array itself would then be freed, leaving\nthe resources that were allocated in previous iterations and referenced\nby the array elements in limbo.\n\nThis patch makes sure all allocated resources are freed before freeing\nthe array \"dd-\u003ercd\". Also the resource allocation now takes account of\nthe numa node the device is attached to.\n\nReviewed-by: Tadeusz Struk \u003ctadeusz.struk@intel.com\u003e\nSigned-off-by: Jianxin Xiong \u003cjianxin.xiong@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "83fb4af6800deb4f3d19b297df6148cda5c016de", "tree": "9dc126613bded5a86bde548c8d0498b649090fb7", "parents": [ "acd7c8fe14938a315f0ac1b92a92375f7226c2fd" ], "author": { "name": "Krzysztof Blaszkowski", "email": "krzysztof.blaszkowski@intel.com", "time": "Mon Oct 17 04:19:24 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:42 2016 -0500" }, "message": "IB/hfi1: Return ENODEV for unsupported PCI device ids.\n\nClean up device type checking.\n\nReviewed-by: Dean Luick \u003cdean.luick@intel.com\u003e\nSigned-off-by: Krzysztof Blaszkowski \u003ckrzysztof.blaszkowski@intel.com\u003e\nSigned-off-by: Tymoteusz Kielan \u003ctymoteusz.kielan@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "acd7c8fe14938a315f0ac1b92a92375f7226c2fd", "tree": "a7c672b8310d578ed82ea0f9f353ee8826abcfd9", "parents": [ "d9ac4555fb2bcd6b794aaa0b39acad81111d9f42" ], "author": { "name": "Tadeusz Struk", "email": "tadeusz.struk@intel.com", "time": "Tue Oct 25 08:57:55 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:41 2016 -0500" }, "message": "IB/hfi1: Fix an Oops on pci device force remove\n\nThis patch fixes an Oops on device unbind, when the device is used\nby a PSM user process. PSM processes access device resources which\nare freed on device removal. Similar protection exists in uverbs\nin ib_core for Verbs clients, but PSM doesn\u0027t use ib_uverbs hence\na separate protection is required for PSM clients.\n\nCc: Jason Gunthorpe \u003cjgunthorpe@obsidianresearch.com\u003e\nReviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\nReviewed-by: Dean Luick \u003cdean.luick@intel.com\u003e\nReviewed-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Tadeusz Struk \u003ctadeusz.struk@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "d9ac4555fb2bcd6b794aaa0b39acad81111d9f42", "tree": "6f0db0c542352bcba065d2ba3281603d775d12ca", "parents": [ "39eb2795f19233330bc14a8450b4042d784b15a7" ], "author": { "name": "Jakub Pawlak", "email": "jakub.pawlak@intel.com", "time": "Mon Oct 10 06:14:56 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:41 2016 -0500" }, "message": "IB/hfi1: Fix integrity check flags default values\n\nPrevent setting up integrity check flags when module is loaded\nwith NO_INTEGRITY capability.\n\nReviewed-by: Dean Luick \u003cdean.luick@intel.com\u003e\nSigned-off-by: Jakub Pawlak \u003cjakub.pawlak@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "39eb2795f19233330bc14a8450b4042d784b15a7", "tree": "d81925b90e30a96bf3d50a240e8d1be53d261b51", "parents": [ "e1fafdcbe0e3e769c6a83317dd845bc99b4fe61d" ], "author": { "name": "Tadeusz Struk", "email": "tadeusz.struk@intel.com", "time": "Mon Oct 10 06:14:50 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:40 2016 -0500" }, "message": "IB/hfi1: Remove redundant sysfs irq affinity entry\n\nThe IRQ affinity entry is not needed after the irq notifier patch has been\nadded to the hfi1 driver.\nThe irq affinity settings for SDMA engine should be set using the standard\n/proc/irq/\u003cN\u003e/ interface.\n\nReviewed-by: Jianxin Xiong \u003cjianxin.xiong@intel.com\u003e\nSigned-off-by: Tadeusz Struk \u003ctadeusz.struk@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "e1fafdcbe0e3e769c6a83317dd845bc99b4fe61d", "tree": "63e6e64282e79d02feafcd956e2573bdea7d5208", "parents": [ "a909d3e636995ba7c349e2ca5dbb528154d4ac30" ], "author": { "name": "Dennis Dalessandro", "email": "dennis.dalessandro@intel.com", "time": "Mon Oct 10 06:14:45 2016 -0700" }, "committer": { "name": "Doug Ledford", "email": "dledford@redhat.com", "time": "Tue Nov 15 16:16:40 2016 -0500" }, "message": "IB/rdmavt: rdmavt can handle non aligned page maps\n\nThe initial code for rdmavt carried with it a restriction that was a\nvestige from the qib driver, that to dma map a page it had to be less\nthan a page size. This is not the case on modern hardware, both qib and\nhfi1 will be just fine with unaligned map requests.\n\nThis fixes a 4.8 regression where by an IPoIB transfer of \u003e PAGE_SIZE\nwill hang because the dma map page call always fails. This was\nintroduced after commit 5faba5469522 (\"IB/ipoib: Report SG feature\nregardless of HW UD CSUM capability\") added the capability to use SG by\ndefault. Rather than override this, the HW supports it, so allow SG.\n\nCc: Stable \u003cstable@vger.kernel.org\u003e # 4.8\nReviewed-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nSigned-off-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nSigned-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n" }, { "commit": "1da2c326e43b0834105993d13610647337bbad67", "tree": "01f7da15ac792681f80132f40bdafba2f87b34d0", "parents": [ "cb434658a8ff151c221a9ac1d44fb6788100cd0d" ], "author": { "name": "Monk Liu", "email": "Monk.Liu@amd.com", "time": "Fri Nov 11 11:24:29 2016 +0800" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Nov 15 14:06:07 2016 -0500" }, "message": "drm/amdgpu:fix vpost_needed routine\n\n1,cleanup description/comments\n2,for FIJI \u0026 passthrough, force post when smc fw version below 22.15\n3,for other cases, follow regular rules\n\nSigned-off-by: Monk Liu \u003cMonk.Liu@amd.com\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "cb434658a8ff151c221a9ac1d44fb6788100cd0d", "tree": "12ab3d7cb5599140364811ba8e682504038fbebf", "parents": [ "8e94a46c1770884166b31adc99eba7da65a446a7" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Nov 15 11:39:08 2016 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Nov 15 14:05:52 2016 -0500" }, "message": "drm/amdgpu/powerplay: drop a redundant NULL check\n\nLeft over from an earlier rev of the patch.\n\nAcked-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\nCc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nCc: Colin King \u003ccolin.king@canonical.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "81bcfe5e48f9b8c42cf547f1c74c7f60c44c34c8", "tree": "55f6ce31dbb95c967fa19c508ccebeb02a16a1d9", "parents": [ "80fc2f7fd8dece7bfa9b5b5dafb8705bbcefa6b4", "546fece4eae871f033925ccf0ff2b740725ae915" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 15 08:49:13 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 15 08:49:13 2016 -0800" }, "message": "Merge tag \u0027trace-v4.9-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fixes from Steven Rostedt:\n \"Alexei discovered a race condition in modules failing to load that can\n cause a ftrace check to trigger and disable ftrace.\n\n This is because of the way modules are registered to ftrace. Their\n functions are loaded in the ftrace function tables but set to\n \"disabled\" since they are still in the process of being loaded by the\n module. After the module is finished, it calls back into the ftrace\n infrastructure to enable it.\n\n Looking deeper into the locations that access all the functions in the\n table, I found more locations that should ignore the disabled ones\"\n\n* tag \u0027trace-v4.9-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n ftrace: Add more checks for FTRACE_FL_DISABLED in processing ip records\n ftrace: Ignore FTRACE_FL_DISABLED while walking dyn_ftrace records\n" }, { "commit": "80fc2f7fd8dece7bfa9b5b5dafb8705bbcefa6b4", "tree": "76e23158f4acb7a63924704fdb609e54274ea8a3", "parents": [ "e76d21c40bd6c67fd4e2c1540d77e113df962b4d", "cfbd950d5e6e649c6c1a88925feada64f890c894" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 15 08:28:59 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Nov 15 08:28:59 2016 -0800" }, "message": "Merge tag \u0027fbdev-fixes-4.9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tomba/linux\n\nPull fbdev fix from Tomi Valkeinen:\n \"Fix CLCD regression on Vexpress\"\n\n* tag \u0027fbdev-fixes-4.9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tomba/linux:\n video: ARM CLCD: fix Vexpress regression\n" }, { "commit": "959401aa2b8cc6422809762a6ba4b2635154e9a6", "tree": "59b3ca71970c1f3e7a96f3e913fa8d3d51898ec4", "parents": [ "e76d21c40bd6c67fd4e2c1540d77e113df962b4d", "14c862dbb0a0e0a9baec20480d441e32cb54b2b9" ], "author": { "name": "Jens Axboe", "email": "axboe@fb.com", "time": "Tue Nov 15 07:51:41 2016 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@fb.com", "time": "Tue Nov 15 07:51:41 2016 -0700" }, "message": "Merge branch \u0027nvmf-4.9-rc\u0027 of git://git.infradead.org/nvme-fabrics into for-linus\n\nSagi writes:\n\nThese are the relevant fixes for rc6\n- fix possible crash in nvmet-rdma cm_handler from Bart\n- fix possible memory leak in nvmet-rdma for connection failures\n- fix possible use-after-free conditions in nvmet-rdma\n- fix possible IO errors during reconnect stage from Christoph\n- fix possible memory leak in nvme-rdma during IO queues connect\n failures from Steve\n" }, { "commit": "59c3b76cc61d1d676f965c192cc7969aa5cb2744", "tree": "0958090c3fe9bd7d265cb46b02f04a9e4b2169bf", "parents": [ "0ce267ff95a0302cf6fb2a552833abbfb7861a43" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Thu Aug 18 09:10:44 2016 +0200" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Tue Nov 15 12:34:21 2016 +0100" }, "message": "fuse: fix fuse_write_end() if zero bytes were copied\n\nIf pos is at the beginning of a page and copied is zero then page is not\nzeroed but is marked uptodate.\n\nFix by skipping everything except unlock/put of page if zero bytes were\ncopied.\n\nReported-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nFixes: 6b12c1b37e55 (\"fuse: Implement write_begin/write_end callbacks\")\nCc: \u003cstable@vger.kernel.org\u003e # v3.15+\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "e76d21c40bd6c67fd4e2c1540d77e113df962b4d", "tree": "26412781fb3f51719cb32d7b031660a7139c1a3b", "parents": [ "d4b9532367c77a35849d6cce491a3ca474c519cf", "ac571de999e14b87890cb960ad6f03fbdde6abc8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 14:15:53 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 14:15:53 2016 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net\n\nPull networking fixes from David Miller:\n\n 1) Fix off by one wrt. indexing when dumping /proc/net/route entries,\n from Alexander Duyck.\n\n 2) Fix lockdep splats in iwlwifi, from Johannes Berg.\n\n 3) Cure panic when inserting certain netfilter rules when NFT_SET_HASH\n is disabled, from Liping Zhang.\n\n 4) Memory leak when nft_expr_clone() fails, also from Liping Zhang.\n\n 5) Disable UFO when path will apply IPSEC tranformations, from Jakub\n Sitnicki.\n\n 6) Don\u0027t bogusly double cwnd in dctcp module, from Florian Westphal.\n\n 7) skb_checksum_help() should never actually use the value \"0\" for the\n resulting checksum, that has a special meaning, use CSUM_MANGLED_0\n instead. From Eric Dumazet.\n\n 8) Per-tx/rx queue statistic strings are wrong in qed driver, fix from\n Yuval MIntz.\n\n 9) Fix SCTP reference counting of associations and transports in\n sctp_diag. From Xin Long.\n\n10) When we hit ip6tunnel_xmit() we could have come from an ipv4 path in\n a previous layer or similar, so explicitly clear the ipv6 control\n block in the skb. From Eli Cooper.\n\n11) Fix bogus sleeping inside of inet_wait_for_connect(), from WANG\n Cong.\n\n12) Correct deivce ID of T6 adapter in cxgb4 driver, from Hariprasad\n Shenai.\n\n13) Fix potential access past the end of the skb page frag array in\n tcp_sendmsg(). From Eric Dumazet.\n\n14) \u0027skb\u0027 can legitimately be NULL in inet{,6}_exact_dif_match(). Fix\n from David Ahern.\n\n15) Don\u0027t return an error in tcp_sendmsg() if we wronte any bytes\n successfully, from Eric Dumazet.\n\n16) Extraneous unlocks in netlink_diag_dump(), we removed the locking\n but forgot to purge these unlock calls. From Eric Dumazet.\n\n17) Fix memory leak in error path of __genl_register_family(). We leak\n the attrbuf, from WANG Cong.\n\n18) cgroupstats netlink policy table is mis-sized, from WANG Cong.\n\n19) Several XDP bug fixes in mlx5, from Saeed Mahameed.\n\n20) Fix several device refcount leaks in network drivers, from Johan\n Hovold.\n\n21) icmp6_send() should use skb dst device not skb-\u003edev to determine L3\n routing domain. From David Ahern.\n\n22) ip_vs_genl_family sets maxattr incorrectly, from WANG Cong.\n\n23) We leak new macvlan port in some cases of maclan_common_netlink()\n errors. Fix from Gao Feng.\n\n24) Similar to the icmp6_send() fix, icmp_route_lookup() should\n determine L3 routing domain using skb_dst(skb)-\u003edev not skb-\u003edev.\n Also from David Ahern.\n\n25) Several fixes for route offloading and FIB notification handling in\n mlxsw driver, from Jiri Pirko.\n\n26) Properly cap __skb_flow_dissect()\u0027s return value, from Eric Dumazet.\n\n27) Fix long standing regression in ipv4 redirect handling, wrt.\n validating the new neighbour\u0027s reachability. From Stephen Suryaputra\n Lin.\n\n28) If sk_filter() trims the packet excessively, handle it reasonably in\n tcp input instead of exploding. From Eric Dumazet.\n\n29) Fix handling of napi hash state when copying channels in sfc driver,\n from Bert Kenward.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (121 commits)\n mlxsw: spectrum_router: Flush FIB tables during fini\n net: stmmac: Fix lack of link transition for fixed PHYs\n sctp: change sk state only when it has assocs in sctp_shutdown\n bnx2: Wait for in-flight DMA to complete at probe stage\n Revert \"bnx2: Reset device during driver initialization\"\n ps3_gelic: fix spelling mistake in debug message\n net: ethernet: ixp4xx_eth: fix spelling mistake in debug message\n ibmvnic: Fix size of debugfs name buffer\n ibmvnic: Unmap ibmvnic_statistics structure\n sfc: clear napi_hash state when copying channels\n mlxsw: spectrum_router: Correctly dump neighbour activity\n mlxsw: spectrum: Fix refcount bug on span entries\n bnxt_en: Fix VF virtual link state.\n bnxt_en: Fix ring arithmetic in bnxt_setup_tc().\n Revert \"include/uapi/linux/atm_zatm.h: include linux/time.h\"\n tcp: take care of truncations done by sk_filter()\n ipv4: use new_gw for redirect neigh lookup\n r8152: Fix error path in open function\n net: bpqether.h: remove if_ether.h guard\n net: __skb_flow_dissect() must cap its return value\n ...\n" }, { "commit": "d4b9532367c77a35849d6cce491a3ca474c519cf", "tree": "f28d93cdd9c62a4368e71647ae1d0578432b9312", "parents": [ "ac38126b2b9533b9e660d25308091bf4bb395726", "e123386bc31bbf467dc558f2f919de0b8b4ba58c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 14:07:13 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 14:07:13 2016 -0800" }, "message": "Merge branch \u0027stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile\n\nPull arch/tile bugfix from Chris Metcalf:\n \"This just fixes an incompatibility with tile __ro_after_init\"\n\n* \u0027stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:\n tile: handle __ro_after_init like parisc does\n" }, { "commit": "ac38126b2b9533b9e660d25308091bf4bb395726", "tree": "9f596b79fdd648d6627b6a403c7b9954312edb87", "parents": [ "ee2bd216e1fa9fa980e6ac702e5973d157c40c48", "efce21fc43e00a76aee7b0a1eda73730ed2d5d3a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 14:00:29 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 14:00:29 2016 -0800" }, "message": "Merge tag \u0027rtc-4.9-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux\n\nPull RTC fixes from Alexandre Belloni:\n \"Here are a few driver fixes for 4.9. It has been calm for a while so I\n don\u0027t expect more for this cycle.\n\n Drivers:\n - asm9260: fix module autoload\n - cmos: fix crashes\n - omap: fix clock handling\"\n\n* tag \u0027rtc-4.9-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux:\n rtc: omap: prevent disabling of clock/module during suspend\n rtc: omap: Fix selecting external osc\n rtc: cmos: Don\u0027t enable interrupts in the middle of the interrupt handler\n rtc: cmos: remove all __exit_p annotations\n rtc: asm9260: fix module autoload\n" }, { "commit": "e123386bc31bbf467dc558f2f919de0b8b4ba58c", "tree": "547970f8b4d4738328445bacc2431753b3faf6e0", "parents": [ "bc33b0ca11e3df467777a4fa7639ba488c9d4911" ], "author": { "name": "Chris Metcalf", "email": "cmetcalf@mellanox.com", "time": "Mon Nov 07 14:32:02 2016 -0500" }, "committer": { "name": "Chris Metcalf", "email": "cmetcalf@mellanox.com", "time": "Mon Nov 14 16:46:41 2016 -0500" }, "message": "tile: handle __ro_after_init like parisc does\n\nThe tile architecture already marks RO_DATA as read-only in\nthe kernel, so grouping RO_AFTER_INIT_DATA with RO_DATA, as is\ndone by default, means the kernel faults in init when it tries\nto write to RO_AFTER_INIT_DATA. For now, just arrange that\n__ro_after_init is handled like __write_once, i.e. __read_mostly.\n\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: Chris Metcalf \u003ccmetcalf@mellanox.com\u003e\n" }, { "commit": "ac571de999e14b87890cb960ad6f03fbdde6abc8", "tree": "85afb4689162f0be63b92611c418b2bccdd90d22", "parents": [ "c51e424dc79e1428afc4d697cdb6a07f7af70cbf" ], "author": { "name": "Ido Schimmel", "email": "idosch@mellanox.com", "time": "Mon Nov 14 11:26:32 2016 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:45:16 2016 -0500" }, "message": "mlxsw: spectrum_router: Flush FIB tables during fini\n\nSince commit b45f64d16d45 (\"mlxsw: spectrum_router: Use FIB notifications\ninstead of switchdev calls\") we reflect to the device the entire FIB\ntable and not only FIBs that point to netdevs created by the driver.\n\nDuring module removal, FIBs of the second type are removed following\nNETDEV_UNREGISTER events sent. The other FIBs are still present in both\nthe driver\u0027s cache and the device\u0027s table.\n\nFix this by iterating over all the FIB tables in the device and flush\nthem. There\u0027s no need to take locks, as we\u0027re the only writer.\n\nFixes: b45f64d16d45 (\"mlxsw: spectrum_router: Use FIB notifications instead of switchdev calls\")\nSigned-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nSigned-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c51e424dc79e1428afc4d697cdb6a07f7af70cbf", "tree": "134649e585c87c481cb11cce69461befdef732eb", "parents": [ "5bf35ddfee052d44f39ebaa395d87101c8918405" ], "author": { "name": "Florian Fainelli", "email": "f.fainelli@gmail.com", "time": "Sun Nov 13 17:50:35 2016 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:39:15 2016 -0500" }, "message": "net: stmmac: Fix lack of link transition for fixed PHYs\n\nCommit 52f95bbfcf72 (\"stmmac: fix adjust link call in case of a switch\nis attached\") added some logic to avoid polling the fixed PHY and\ntherefore invoking the adjust_link callback more than once, since this\nis a fixed PHY and link events won\u0027t be generated.\n\nThis works fine the first time, because we start with phydev-\u003eirq \u003d\nPHY_POLL, so we call adjust_link, then we set phydev-\u003eirq \u003d\nPHY_IGNORE_INTERRUPT and we stop polling the PHY.\n\nNow, if we called ndo_close(), which calls both phy_stop() and does an\nexplicit netif_carrier_off(), we end up with a link down. Upon calling\nndo_open() again, despite starting the PHY state machine, we have\nPHY_IGNORE_INTERRUPT set, and we generate no link event at all, so the\nlink is permanently down.\n\nFixes: 52f95bbfcf72 (\"stmmac: fix adjust link call in case of a switch is attached\")\nSigned-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nAcked-by: Giuseppe Cavallaro \u003cpeppe.cavallaro@st.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "546fece4eae871f033925ccf0ff2b740725ae915", "tree": "04608cf612ed5f5173992abc25c554ed5169e9f7", "parents": [ "977c1f9c8c022d0173181766b34a0db3705265a4" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Nov 14 16:31:49 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Nov 14 16:31:49 2016 -0500" }, "message": "ftrace: Add more checks for FTRACE_FL_DISABLED in processing ip records\n\nWhen a module is first loaded and its function ip records are added to the\nftrace list of functions to modify, they are set to DISABLED, as their text\nis still in a read only state. When the module is fully loaded, and can be\nupdated, the flag is cleared, and if their\u0027s any functions that should be\ntracing them, it is updated at that moment.\n\nBut there\u0027s several locations that do record accounting and should ignore\nrecords that are marked as disabled, or they can cause issues.\n\nAlexei already fixed one location, but others need to be addressed.\n\nCc: stable@vger.kernel.org\nFixes: b7ffffbb46f2 \"ftrace: Add infrastructure for delayed enabling of module functions\"\nReported-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "977c1f9c8c022d0173181766b34a0db3705265a4", "tree": "8b38040b883d49ced50f39d17e35f7f8d6c2a126", "parents": [ "a25f0944ba9b1d8a6813fd6f1a86f1bd59ac25a6" ], "author": { "name": "Alexei Starovoitov", "email": "ast@fb.com", "time": "Mon Nov 07 15:14:20 2016 -0800" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Nov 14 16:31:41 2016 -0500" }, "message": "ftrace: Ignore FTRACE_FL_DISABLED while walking dyn_ftrace records\n\nftrace_shutdown() checks for sanity of ftrace records\nand if dyn_ftrace-\u003eflags is not zero, it will warn.\nIt can happen that \u0027flags\u0027 are set to FTRACE_FL_DISABLED at this point,\nsince some module was loaded, but before ftrace_module_enable()\ncleared the flags for this module.\n\nIn other words the module.c is doing:\nftrace_module_init(mod); // calls ftrace_update_code() that sets flags\u003dFTRACE_FL_DISABLED\n... // here ftrace_shutdown() is called that warns, since\nerr \u003d prepare_coming_module(mod); // didn\u0027t have a chance to clear FTRACE_FL_DISABLED\n\nFix it by ignoring disabled records.\nIt\u0027s similar to what __ftrace_hash_rec_update() is already doing.\n\nLink: http://lkml.kernel.org/r/1478560460-3818619-1-git-send-email-ast@fb.com\n\nCc: stable@vger.kernel.org\nFixes: b7ffffbb46f2 \"ftrace: Add infrastructure for delayed enabling of module functions\"\nSigned-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "5bf35ddfee052d44f39ebaa395d87101c8918405", "tree": "0482e404f2e9307e27224e6ffc660c59d34325b9", "parents": [ "193f5122878c5d4d30dbfc9128affe58bdd18206" ], "author": { "name": "Xin Long", "email": "lucien.xin@gmail.com", "time": "Sun Nov 13 21:44:37 2016 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:22:33 2016 -0500" }, "message": "sctp: change sk state only when it has assocs in sctp_shutdown\n\nNow when users shutdown a sock with SEND_SHUTDOWN in sctp, even if\nthis sock has no connection (assoc), sk state would be changed to\nSCTP_SS_CLOSING, which is not as we expect.\n\nBesides, after that if users try to listen on this sock, kernel\ncould even panic when it dereference sctp_sk(sk)-\u003ebind_hash in\nsctp_inet_listen, as bind_hash is null when sock has no assoc.\n\nThis patch is to move sk state change after checking sk assocs\nis not empty, and also merge these two if() conditions and reduce\nindent level.\n\nFixes: d46e416c11c8 (\"sctp: sctp should change socket state when shutdown is received\")\nReported-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nTested-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nSigned-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nAcked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "193f5122878c5d4d30dbfc9128affe58bdd18206", "tree": "31491c40f00c8b5a9f0fe2d7fc2a1bbcf878578f", "parents": [ "7020637bdf59589a403e01aca128bef643404317", "6df77862f63f389df3b1ad879738e04440d7385d" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:20:54 2016 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:20:54 2016 -0500" }, "message": "Merge branch \u0027bnx2-kdump-fix\u0027\n\nBaoquan He says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nbnx2: Wait for in-flight DMA to complete at probe stage\n\nThis is v2 post.\n\nIn commit 3e1be7a (\"bnx2: Reset device during driver initialization\"),\nfirmware requesting code was moved from open stage to probe stage.\nThe reason is in kdump kernel hardware iommu need device be reset in\ndriver probe stage, otherwise those in-flight DMA from 1st kernel\nwill continue going and look up into the newly created io-page tables.\nHowever bnx2 chip resetting involves firmware requesting issue, that\nneed be done in open stage.\n\nMichale Chan suggested we can just wait for the old in-flight DMA to\ncomplete at probe stage, then though without device resetting, we\ndon\u0027t need to worry the old in-flight DMA could continue looking up\nthe newly created io-page tables.\n\nv1-\u003ev2:\n Michael suggested to wait for the in-flight DMA to complete at probe\n stage. So give up the old method of trying to reset chip at probe\n stage, take the new way accordingly.\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": "6df77862f63f389df3b1ad879738e04440d7385d", "tree": "31491c40f00c8b5a9f0fe2d7fc2a1bbcf878578f", "parents": [ "5d0d4b91bf627f14f95167b738d524156c9d440b" ], "author": { "name": "Baoquan He", "email": "bhe@redhat.com", "time": "Sun Nov 13 13:01:33 2016 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:20:53 2016 -0500" }, "message": "bnx2: Wait for in-flight DMA to complete at probe stage\n\nIn-flight DMA from 1st kernel could continue going in kdump kernel.\nNew io-page table has been created before bnx2 does reset at open stage.\nWe have to wait for the in-flight DMA to complete to avoid it look up\ninto the newly created io-page table at probe stage.\n\nSuggested-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nSigned-off-by: Baoquan He \u003cbhe@redhat.com\u003e\nAcked-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5d0d4b91bf627f14f95167b738d524156c9d440b", "tree": "d9bc8fe39ce2644dfa12820b90d8566a7f959c24", "parents": [ "7020637bdf59589a403e01aca128bef643404317" ], "author": { "name": "Baoquan He", "email": "bhe@redhat.com", "time": "Sun Nov 13 13:01:32 2016 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 16:20:53 2016 -0500" }, "message": "Revert \"bnx2: Reset device during driver initialization\"\n\nThis reverts commit 3e1be7ad2d38c6bd6aeef96df9bd0a7822f4e51c.\n\nWhen people build bnx2 driver into kernel, it will fail to detect\nand load firmware because firmware is contained in initramfs and\ninitramfs has not been uncompressed yet during do_initcalls. So\nrevert commit 3e1be7a and work out a new way in the later patch.\n\nSigned-off-by: Baoquan He \u003cbhe@redhat.com\u003e\nAcked-by: Paul Menzel \u003cpmenzel@molgen.mpg.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "709fb1f961ea5c287107c3f903e81c9529224c8b", "tree": "cd52512c7403df49ddbba7edf9087b7e4c430a7f", "parents": [ "d4eccafcaf339de77ec562e96e6b223d447f924a" ], "author": { "name": "Max Filippov", "email": "jcmvbkbc@gmail.com", "time": "Mon Nov 14 12:31:49 2016 -0800" }, "committer": { "name": "Max Filippov", "email": "jcmvbkbc@gmail.com", "time": "Mon Nov 14 12:31:49 2016 -0800" }, "message": "xtensa: wire up new pkey_{mprotect,alloc,free} syscalls\n\nSigned-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n" }, { "commit": "7020637bdf59589a403e01aca128bef643404317", "tree": "ab32923b7c0c789b43ae33beb04ecaf56c8a9c08", "parents": [ "7774d46b2037b98d3f7e414bffb1d53082dc139b" ], "author": { "name": "Colin Ian King", "email": "colin.king@canonical.com", "time": "Sat Nov 12 17:20:30 2016 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Nov 14 13:38:57 2016 -0500" }, "message": "ps3_gelic: fix spelling mistake in debug message\n\nTrivial fix to spelling mistake \"unmached\" to \"unmatched\" in\ndebug message.\n\nSigned-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ee2bd216e1fa9fa980e6ac702e5973d157c40c48", "tree": "e125c1155d5c602a816856f38a6a9b6665714be3", "parents": [ "f5c9f9c72395c3291c2e35c905dedae2b98475a4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 09:46:08 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 09:46:08 2016 -0800" }, "message": "ASoC: lpass-platform: fix uninitialized variable\n\nIn commit 022d00ee0b55 (\"ASoC: lpass-platform: Fix broken pcm data\nusage\") the stream specific information initialization was broken, with\nthe dma channel information not being initialized if there was no\nalloc_dma_channel() helper function.\n\nBefore that, the DMA channel number was implicitly initialized to zero\nbecause the backing store was allocated with devm_kzalloc(). When the\ninit code was rewritten, that implicit initialization was lost, and gcc\nrightfully complains about an uninitialized variable being used.\n\nCc: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\nCc: Mark Brown \u003cbroonie@kernel.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f5c9f9c72395c3291c2e35c905dedae2b98475a4", "tree": "d9ce28fbeb0a3c4baf5d1fcc75e6e0584e18fcc9", "parents": [ "b15efc38626f20f3fc8b831b826b50740d90dab9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 09:31:52 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 09:31:52 2016 -0800" }, "message": "Revert \"printk: make reading the kernel log flush pending lines\"\n\nThis reverts commit bfd8d3f23b51018388be0411ccbc2d56277fe294.\n\nIt turns out that this flushes things much too aggressiverly, and causes\nlines to break up when the system logger races with new continuation\nlines being printed.\n\nThere\u0027s a pending patch to make printk() flushing much more\nstraightforward, but it\u0027s too invasive for 4.9, so in the meantime let\u0027s\njust not make the system message logging flush continuation lines.\nThey\u0027ll be flushed by the final newline anyway.\n\nSuggested-by: Petr Mladek \u003cpmladek@suse.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b15efc38626f20f3fc8b831b826b50740d90dab9", "tree": "9e666cc321bf71b9845b69c5334e1f20c43ca152", "parents": [ "8528d662489a980305c7b0d1d41aca3e140c1e3d" ], "author": { "name": "Mauro Carvalho Chehab", "email": "mchehab@infradead.org", "time": "Mon Nov 14 11:14:37 2016 -0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:43:13 2016 -0800" }, "message": "gp8psk-fe: add missing MODULE_foo() macros\n\nThis file was converted to a separate module at commit 7a0786c19d65\n(\"gp8psk: Fix DVB frontend attach\"), because the DVB attach routines\nrequire it to work. However, I forgot to copy the MODULE_foo() macros\nfrom the original module, causing this warning:\n\n WARNING: modpost: missing MODULE_LICENSE() in drivers/media/dvb-frontends/gp8psk-fe.o\n\nReported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nFixes: 7a0786c19d65 (\"gp8psk: Fix DVB frontend attach\")\nSigned-off-by: Mauro Carvalho Chehab \u003cmchehab@s-opensource.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8528d662489a980305c7b0d1d41aca3e140c1e3d", "tree": "0161b8390b141ba6560fb9309b5efe7f9eafcd16", "parents": [ "5d69561b7b32b89380b5390b62e5a0c50bfe6568", "d49597fd3bc7d9534de55e9256767f073be1b33a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:39:56 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:39:56 2016 -0800" }, "message": "Merge branch \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Ingo Molnar:\n \"Misc fixes:\n\n - fix an Intel/MID boot crash/hang bug\n\n - fix a cache topology mis-parsing bug on certain AMD CPUs\n\n - fix a virtualization firmware bug by adding a check+quirk\n workaround on the kernel side\"\n\n* \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/cpu: Deal with broken firmware (VMWare/XEN)\n x86/cpu/AMD: Fix cpu_llc_id for AMD Fam17h systems\n x86/platform/intel-mid: Retrofit pci_platform_pm_ops -\u003eget_state hook\n" }, { "commit": "5d69561b7b32b89380b5390b62e5a0c50bfe6568", "tree": "7862fbe44321ffb8c39c49800587f9a622311c62", "parents": [ "5ad62a9e5cf82e31ac8fc302e4090cbd0bbc1615", "7ee7e87dfb158e79019ea1d5ea1b0e6f2bc93ee4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:34:56 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:34:56 2016 -0800" }, "message": "Merge branch \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq fix from Ingo Molnar:\n \"This fixes a genirq regression that resulted in the Intel/Broxton\n pinctrl/GPIO driver (and possibly others) spewing warnings\"\n\n* \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n genirq: Use irq type from irqdata instead of irqdesc\n" }, { "commit": "5ad62a9e5cf82e31ac8fc302e4090cbd0bbc1615", "tree": "a3adf6532e572b2b0cf66cbb44807f33e8808b53", "parents": [ "53381e2e34f741350bd00f592d2b6c3710ba5177", "ce75632cc4012f1832bd56efd97c2ba75ca964bb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:30:06 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:30:06 2016 -0800" }, "message": "Merge branch \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf fixes from Ingo Molnar:\n \"An uncore PMU driver hardware enablement change for Intel SkyLake\n uncore PMUs (Skylake Y, U, H and S platforms), plus a number of\n tooling fixes for the histogram handling/displaying code\"\n\n* \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n perf/x86/intel/uncore: Add more Intel uncore IMC PCI IDs for SkyLake\n perf hists: Fix column length on --hierarchy\n perf hists browser: Fix column indentation on --hierarchy\n perf hists browser: Show folded sign properly on --hierarchy\n perf hists browser: Fix indentation of folded sign on --hierarchy\n perf hist browser: Fix hierarchy column counts\n" }, { "commit": "53381e2e34f741350bd00f592d2b6c3710ba5177", "tree": "338e5093e343aaea9da9018ce71a5814c799437d", "parents": [ "28ddafa5905310889defc485dbefe0a484edb2b6", "f6697df36bdf0bf7fce984605c2918d4a7b4269f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:26:24 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:26:24 2016 -0800" }, "message": "Merge branch \u0027efi-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull EFI fixes from Ingo Molnar:\n \"A boot crash fix and a build warning fix\"\n\n* \u0027efi-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/efi: Prevent mixed mode boot corruption with CONFIG_VMAP_STACK\u003dy\n x86/efi: Fix EFI memmap pointer size warning\n" }, { "commit": "28ddafa5905310889defc485dbefe0a484edb2b6", "tree": "c4c576acfae821524f886b7f12841f9168ffa6ff", "parents": [ "a25f0944ba9b1d8a6813fd6f1a86f1bd59ac25a6", "819baf885953b588b63bef28e5598daf9ed4ddf9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:14:49 2016 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Nov 14 08:14:49 2016 -0800" }, "message": "Merge tag \u0027ntb-4.9\u0027 of git://github.com/jonmason/ntb\n\nPull NTB fixes from Jon Mason:\n \"NTB bug fixes for ntb_hw_intel, ntb_perf, and ntb_pingpong.\n\n Also, a fixup to use jiffies in schedule_timeout_* call instead of a\n constant\"\n\n* tag \u0027ntb-4.9\u0027 of git://github.com/jonmason/ntb:\n ntb_perf: potential info leak in debugfs\n ntb: ntb_hw_intel: init peer_addr in struct intel_ntb_dev\n ntb: make DMA_OUT_RESOURCE_TO HZ independent\n ntb_transport: make DMA_OUT_RESOURCE_TO HZ independent\n NTB: ntb_hw_intel: Fix typo in module parameter descriptions\n ntb_pingpong: Fix db_init parameter description\n" }, { "commit": "14c862dbb0a0e0a9baec20480d441e32cb54b2b9", "tree": "18cc13ffb25ab02c05f50a09f2b4d63fff77144a", "parents": [ "c8dbc37cd81d4705fce51123f5d81ea3267a5b88" ], "author": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Sun Nov 06 11:03:59 2016 +0200" }, "committer": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Mon Nov 14 02:08:53 2016 +0200" }, "message": "nvmet-rdma: drain the queue-pair just before freeing it\n\ndraining the qp right after disconnect might not suffice because\nthe nvmet sq is not fully drained (in nvmet_sq_destroy) and we might\nsee completions after the drain. Instead, drain right before the\nqp destroy which comes after the sq destruction and we can be sure\nthat no posts come after the drain.\n\nTested-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n" }, { "commit": "c8dbc37cd81d4705fce51123f5d81ea3267a5b88", "tree": "930d2f3e3778b18d28f8cde75bf256111de120cb", "parents": [ "766dbb179d41d6337fed2b3ca00caa5845d298ce" ], "author": { "name": "Steve Wise", "email": "swise@opengridcomputing.com", "time": "Tue Nov 08 09:16:02 2016 -0800" }, "committer": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Mon Nov 14 02:08:53 2016 +0200" }, "message": "nvme-rdma: stop and free io queues on connect failure\n\nWhile testing nvme-rdma with the spdk nvmf target over iw_cxgb4, I\nconfigured the target (mistakenly) to generate an error creating the\nNVMF IO queues. This resulted a \"Invalid SQE Parameter\" error sent back\nto the host on the first IO queue connect:\n\n[ 9610.928182] nvme nvme1: queue_size 128 \u003e ctrl maxcmd 120, clamping down\n[ 9610.938745] nvme nvme1: creating 32 I/O queues.\n\nSo nvmf_connect_io_queue() returns an error to\nnvmf_connect_io_queue() / nvmf_connect_io_queues(), and that\nis returned to nvme_rdma_create_io_queues(). In the error path,\nnvmf_rdma_create_io_queues() frees the queue tagset memory _before_\nstopping and freeing the IB queues, which causes yet another\ntouch-after-free crash due to SQ CQEs being flushed after the ib_cqe\nstructs pointed-to by the flushed WRs have been freed (since they are\npart of the nvme_rdma_request struct).\n\nThe fix is to stop and free the queues in nvmf_connect_io_queues()\nif there is an error connecting any of the queues.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n" }, { "commit": "766dbb179d41d6337fed2b3ca00caa5845d298ce", "tree": "eeebfbc27fd93915d79c47d686f0d442ea72823c", "parents": [ "8242ddac1bfcf6eb8873b4d0a4e7a172c2b5b625" ], "author": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Sun Nov 06 11:09:49 2016 +0200" }, "committer": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Mon Nov 14 02:08:52 2016 +0200" }, "message": "nvmet-rdma: don\u0027t forget to delete a queue from the list of connection failed\n\nIn case we accepted a queue connection and it failed, we might not\nremove the queue from the list until we unload and clean it up.\nWe should delete it from the queue list on the relevant handler.\n\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n" }, { "commit": "8242ddac1bfcf6eb8873b4d0a4e7a172c2b5b625", "tree": "05c0b3fae51a96304bee12b75697ad29d723a5a0", "parents": [ "553cd9ef82edd811948782a8f73ae73c4bfeedd3" ], "author": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Sun Nov 06 11:03:30 2016 +0200" }, "committer": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Mon Nov 14 02:08:51 2016 +0200" }, "message": "nvmet: Don\u0027t queue fatal error work if csts.cfs is set\n\nIn the transport, in case of an interal queue error like\nerror completion in rdma we trigger a fatal error. However,\nmultiple queues in the same controller can serr error completions\nand we don\u0027t want to trigger fatal error work more than once.\n\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n" }, { "commit": "553cd9ef82edd811948782a8f73ae73c4bfeedd3", "tree": "5209ca1a85e34a8e0fb9605e24759a50ce903b89", "parents": [ "fa14a0acea1ffe67913ba384a2897130a36dfe03" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Nov 02 08:49:18 2016 -0600" }, "committer": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Mon Nov 14 02:08:51 2016 +0200" }, "message": "nvme-rdma: reject non-connect commands before the queue is live\n\nIf we reconncect we might have command queue up that get resent as soon\nas the queue is restarted. But until the connect command succeeded we\ncan\u0027t send other command. Add a new flag that marks a queue as live when\nconnect finishes, and delay any non-connect command until the queue is\nlive based on it.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nReported-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nTested-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\n[sagig: fixes admin queue LIVE setting]\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n" }, { "commit": "fa14a0acea1ffe67913ba384a2897130a36dfe03", "tree": "ea33c1c7340d3f77d19820172d2906cf73b38c23", "parents": [ "a25f0944ba9b1d8a6813fd6f1a86f1bd59ac25a6" ], "author": { "name": "Bart Van Assche", "email": "bart.vanassche@sandisk.com", "time": "Tue Nov 01 18:36:46 2016 +0200" }, "committer": { "name": "Sagi Grimberg", "email": "sagi@grimberg.me", "time": "Mon Nov 14 02:08:50 2016 +0200" }, "message": "nvmet-rdma: Fix possible NULL deref when handling rdma cm events\n\nWhen we initiate queue teardown sequence we call rdma_destroy_qp\nwhich clears cm_id-\u003eqp, afterwards we call rdma_destroy_id, but\nwe might see a rdma_cm event in between with a cleared cm_id-\u003eqp\nso watch out for that and silently ignore the event because this\nmeans that the queue teardown sequence is in progress.\n\nSigned-off-by: Bart Van Assche \u003cbart.vanassche@sandisk.com\u003e\nSigned-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n" }, { "commit": "819baf885953b588b63bef28e5598daf9ed4ddf9", "tree": "c4c576acfae821524f886b7f12841f9168ffa6ff", "parents": [ "25ea9f2bf5f76082da919f2a91ea8d920932c1da" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Fri Oct 14 10:34:18 2016 +0300" }, "committer": { "name": "Jon Mason", "email": "jdmason@kudzu.us", "time": "Sun Nov 13 16:48:30 2016 -0500" }, "message": "ntb_perf: potential info leak in debugfs\n\nThis is a static checker warning, not something I\u0027m desperately\nconcerned about. But snprintf() returns the number of bytes that\nwould have been copied if there were space. We really care about the\nnumber of bytes that actually were copied so we should use scnprintf()\ninstead.\n\nIt probably won\u0027t overrun, and in that case we may as well just use\nsprintf() but these sorts of things make static checkers and code\nreviewers happier.\n\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nAcked-by: Dave Jiang \u003cdave.jiang@intel.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@kudzu.us\u003e\n" }, { "commit": "25ea9f2bf5f76082da919f2a91ea8d920932c1da", "tree": "e011001648928005b421e63ed8587d06b4facf5a", "parents": [ "cdc08982a5f334cecc15d802464588115512cc36" ], "author": { "name": "Dave Jiang", "email": "dave.jiang@intel.com", "time": "Thu Oct 27 11:06:44 2016 -0700" }, "committer": { "name": "Jon Mason", "email": "jdmason@kudzu.us", "time": "Sun Nov 13 16:48:29 2016 -0500" }, "message": "ntb: ntb_hw_intel: init peer_addr in struct intel_ntb_dev\n\nThe peer_addr member of intel_ntb_dev is not set, therefore when\nacquiring ntb_peer_db and ntb_peer_spad we only get the offset rather\nthan the actual physical address. Adding fix to correct that.\n\nSigned-off-by: Dave Jiang \u003cdave.jiang@intel.com\u003e\nAcked-by: Allen Hubbe \u003cAllen.Hubbe@emc.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@kudzu.us\u003e\n" }, { "commit": "cdc08982a5f334cecc15d802464588115512cc36", "tree": "e3ace761f166c89620d7849e0c0ce6c0be9f2c3a", "parents": [ "c0a88032ef8e6814d4dd84551e5f333c1de639b3" ], "author": { "name": "Nicholas Mc Guire", "email": "hofrat@osadl.org", "time": "Mon Aug 22 18:51:36 2016 +0200" }, "committer": { "name": "Jon Mason", "email": "jdmason@kudzu.us", "time": "Sun Nov 13 16:48:29 2016 -0500" }, "message": "ntb: make DMA_OUT_RESOURCE_TO HZ independent\n\nschedule_timeout_* takes a timeout in jiffies but the code currently is\npassing in a constant which makes this timeout HZ dependent, so pass it\nthrough msecs_to_jiffies() to fix this up.\n\nSigned-off-by: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\nAcked-by: Dave Jiang \u003cdave.jiang@intel.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@kudzu.us\u003e\n" }, { "commit": "c0a88032ef8e6814d4dd84551e5f333c1de639b3", "tree": "0c639ad65368e7f4f647b2bbcb7a1334a139ec98", "parents": [ "49b89de41f8d97eb13a60c1865ed61fbebed0d15" ], "author": { "name": "Nicholas Mc Guire", "email": "hofrat@osadl.org", "time": "Mon Aug 22 18:51:35 2016 +0200" }, "committer": { "name": "Jon Mason", "email": "jdmason@kudzu.us", "time": "Sun Nov 13 16:48:29 2016 -0500" }, "message": "ntb_transport: make DMA_OUT_RESOURCE_TO HZ independent\n\nschedule_timeout_* takes a timeout in jiffies but the code currently is\npassing in a constant which makes this timeout HZ dependent, so pass it\nthrough msecs_to_jiffies() to fix this up.\n\nSigned-off-by: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\nSigned-off-by: Jon Mason \u003cjdmason@kudzu.us\u003e\n" }, { "commit": "49b89de41f8d97eb13a60c1865ed61fbebed0d15", "tree": "305e80cf55b0ddecd15ff2e5494ae706139412ad", "parents": [ "cedecbc5e0f39d2987b8e1004908e90459a82e78" ], "author": { "name": "Wei Yongjun", "email": "weiyj.lk@gmail.com", "time": "Mon Aug 08 09:48:42 2016 +0000" }, "committer": { "name": "Jon Mason", "email": "jdmason@kudzu.us", "time": "Sun Nov 13 16:48:29 2016 -0500" }, "message": "NTB: ntb_hw_intel: Fix typo in module parameter descriptions\n\nFix typo in module parameter descriptions.\n\nSigned-off-by: Wei Yongjun \u003cweiyj.lk@gmail.com\u003e\nAcked-by: Allen Hubbe \u003cAllen.Hubbe@emc.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@kudzu.us\u003e\n" }, { "commit": "cedecbc5e0f39d2987b8e1004908e90459a82e78", "tree": "14d148dc5c3a94dcbe35283006b75343da4eb7aa", "parents": [ "a25f0944ba9b1d8a6813fd6f1a86f1bd59ac25a6" ], "author": { "name": "Wei Yongjun", "email": "weiyj.lk@gmail.com", "time": "Mon Aug 08 09:48:00 2016 +0000" }, "committer": { "name": "Jon Mason", "email": "jdmason@kudzu.us", "time": "Sun Nov 13 16:48:29 2016 -0500" }, "message": "ntb_pingpong: Fix db_init parameter description\n\nFix \u0027db_init\u0027 parameter description.\n\nSigned-off-by: Wei Yongjun \u003cweiyj.lk@gmail.com\u003e\nAcked-by: Allen Hubbe \u003cAllen.Hubbe@emc.com\u003e\nSigned-off-by: Jon Mason \u003cjdmason@kudzu.us\u003e\n" }, { "commit": "7774d46b2037b98d3f7e414bffb1d53082dc139b", "tree": "d3d42b0f17f58c28dc7f23578d9b9cf5422af93d", "parents": [ "e1fac0adf0f9b2c1eb49e658e6ed070a744bbaef" ], "author": { "name": "Colin Ian King", "email": "colin.king@canonical.com", "time": "Sat Nov 12 17:44:06 2016 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 13 13:48:28 2016 -0500" }, "message": "net: ethernet: ixp4xx_eth: fix spelling mistake in debug message\n\nTrivial fix to spelling mistake \"successed\" to \"succeeded\"\nin debug message. Also unwrap multi-line literal string.\n\nSigned-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e1fac0adf0f9b2c1eb49e658e6ed070a744bbaef", "tree": "3531e8c93fea65833e16f85a4f09236faa88b23a", "parents": [ "b7f193da17fb18b752bef77ce52eb49723299bd8" ], "author": { "name": "Thomas Falcon", "email": "tlfalcon@linux.vnet.ibm.com", "time": "Fri Nov 11 11:00:46 2016 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 13 13:42:35 2016 -0500" }, "message": "ibmvnic: Fix size of debugfs name buffer\n\nThis mistake was causing debugfs directory creation\nfailures when multiple ibmvnic devices were probed.\n\nSigned-off-by: Thomas Falcon \u003ctlfalcon@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b7f193da17fb18b752bef77ce52eb49723299bd8", "tree": "8fe901b4522ed025637d9735b91dd52f47861470", "parents": [ "46d054f8f540612f09987a53154aa39ae15f2e4c" ], "author": { "name": "Thomas Falcon", "email": "tlfalcon@linux.vnet.ibm.com", "time": "Fri Nov 11 11:00:45 2016 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 13 13:42:35 2016 -0500" }, "message": "ibmvnic: Unmap ibmvnic_statistics structure\n\nThis structure was mapped but never subsequently unmapped.\n\nSigned-off-by: Thomas Falcon \u003ctlfalcon@linux.vnet.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "46d054f8f540612f09987a53154aa39ae15f2e4c", "tree": "e062d97ded0146e016ee064c02c2bfb7a3f56e3d", "parents": [ "9e37aaa39dccad05c3d8883f0790a804e3ec4810" ], "author": { "name": "Bert Kenward", "email": "bkenward@solarflare.com", "time": "Fri Nov 11 15:56:51 2016 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Nov 13 13:41:42 2016 -0500" }, "message": "sfc: clear napi_hash state when copying channels\n\nefx_copy_channel() doesn\u0027t correctly clear the napi_hash related state.\nThis means that when napi_hash_add is called for that channel nothing is\ndone, and we are left with a copy of the napi_hash_node from the old\nchannel. When we later call napi_hash_del() on this channel we have a\nstale napi_hash_node.\n\nCorruption is only seen when there are multiple entries in one of the\nnapi_hash lists. This is made more likely by having a very large number\nof channels. Testing was carried out with 512 channels - 32 channels on\neach of 16 ports.\n\nThis failure typically appears as protection faults within napi_by_id()\nor napi_hash_add(). efx_copy_channel() is only used when tx or rx ring\nsizes are changed (ethtool -G).\n\nFixes: 36763266bbe8 (\"sfc: Add support for busy polling\")\nSigned-off-by: Bert Kenward \u003cbkenward@solarflare.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" } ], "next": "a25f0944ba9b1d8a6813fd6f1a86f1bd59ac25a6" }