)]}' { "log": [ { "commit": "2226fb57a908330c7e2b83d363d450f2000de837", "tree": "e5fc494bea8650e31ce046be6bdc924d54d4c187", "parents": [ "15a555a4b8be26683d77df8d5dbe8ac83f5ec3a6", "a111ef6b082270f6cbeea5556caf1cbb0143b812" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:35:23 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:35:23 2019 -0700" }, "message": "Merge tag \u0027drm-fixes-2019-08-09\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Usual fixes roundup. Nothing too crazy or serious, one non-released\n ioctl is removed in the amdkfd driver.\n\n core:\n - mode parser strncpy fix\n\n i915:\n - GLK DSI escape clock setting\n - HDCP memleak fix\n\n tegra:\n - one gpiod/of regression fix\n\n amdgpu:\n - fix VCN to handle the latest navi10 firmware\n - fix for fan control on navi10\n - properly handle SMU metrics table on navi10\n - fix a resume regression on Stoney\n - kfd revert a GWS ioctl\n\n vmwgfx:\n - memory leak fix\n\n rockchip:\n - suspend fix\"\n\n* tag \u0027drm-fixes-2019-08-09\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/vmwgfx: fix memory leak when too many retries have occurred\n Revert \"drm/amdkfd: New IOCTL to allocate queue GWS\"\n Revert \"drm/amdgpu: fix transform feedback GDS hang on gfx10 (v2)\"\n drm/amdgpu: pin the csb buffer on hw init for gfx v8\n drm/rockchip: Suspend DP late\n drm/i915: Fix wrong escape clock divisor init for GLK\n drm/i915: fix possible memory leak in intel_hdcp_auth_downstream()\n drm/modes: Fix unterminated strncpy\n drm/amd/powerplay: correct navi10 vcn powergate\n drm/amd/powerplay: honor hw limit on fetching metrics data for navi10\n drm/amd/powerplay: Allow changing of fan_control in smu_v11_0\n drm/amd/amdgpu/vcn_v2_0: Move VCN 2.0 specific dec ring test to vcn_v2_0\n drm/amd/amdgpu/vcn_v2_0: Mark RB commands as KMD commands\n drm/tegra: Fix gpiod_get_from_of_node() regression\n" }, { "commit": "15a555a4b8be26683d77df8d5dbe8ac83f5ec3a6", "tree": "f977510e24ad0c14e3543c1477576025315fe87e", "parents": [ "cb7ef4bc927233304aeeeb891c1cfc5ad6f87975", "30e235389faadb9e3d918887b1f126155d7d761d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:31:40 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:31:40 2019 -0700" }, "message": "Merge tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 fix from Catalin Marinas:\n \"Fix bad_pte warning caused by pte_mkdevmap() not setting PTE_SPECIAL\"\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n arm64: mm: add missing PTE_SPECIAL in pte_mkdevmap on arm64\n" }, { "commit": "cb7ef4bc927233304aeeeb891c1cfc5ad6f87975", "tree": "0c9ff9c3c072df8ceeb614d5ba855c4e1db5aeef", "parents": [ "50e73a4a41598f9a785a986d25b731d3968dedb1", "404861e15b5fa7edbab22400f9174c1a21fde731" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:30:00 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:30:00 2019 -0700" }, "message": "Merge tag \u0027s390-5.3-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 fixes from Vasily Gorbik:\n\n - Map vdso also for statically linked binaries like all other\n architectures.\n\n - Fix no .bss usage compile-time check to account common objects with\n the help of binutils size tool. Top level Makefile change acked-by\n Masahiro.\n\n - A fix to make perf happy with _etext symbol type.\n\n - Fix dump_pagetables which is broken since p*d_offset implementation\n change to comply with mm/gup.c expectations.\n\n - Revert memory sharing for diag calls in protected virtualization,\n since this is not required after all.\n\n - Couple of other minor code cleanups.\n\n* tag \u0027s390-5.3-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n s390/vdso: map vdso also for statically linked binaries\n s390/build: use size command to perform empty .bss check\n kbuild: add OBJSIZE variable for the size tool\n s390: put _stext and _etext into .text section\n s390/head64: cleanup unused labels\n s390/unwind: remove stack recursion warning\n s390/setup: adjust start_code of init_mm to _text\n s390/mm: fix dump_pagetables top level page table walking\n s390/protvirt: avoid memory sharing for diag 308 set/store\n" }, { "commit": "50e73a4a41598f9a785a986d25b731d3968dedb1", "tree": "c915250d4d2637aaf9e1eef585d9e6cf15842599", "parents": [ "461d2815ac9ec1b3f66fc243cc39bd2374facafa", "20621fedb2a696e4dc60bc1c5de37cf21976abcb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:28:18 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:28:18 2019 -0700" }, "message": "Merge tag \u0027for-linus-20190809\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n\n - Revert of a bcache patch that caused an oops for some (Coly)\n\n - ata rb532 unused warning fix (Gustavo)\n\n - AoE kernel crash fix (He)\n\n - Error handling fixup for blkdev_get() (Jan)\n\n - libata read/write translation and SFF PIO fix (me)\n\n - Use after free and error handling fix for O_DIRECT fragments. There\u0027s\n still a nowait + sync oddity in there, we\u0027ll nail that start next\n week. If all else fails, I\u0027ll queue a revert of the NOWAIT change.\n (me)\n\n - Loop GFP_KERNEL -\u003e GFP_NOIO deadlock fix (Mikulas)\n\n - Two BFQ regression fixes that caused crashes (Paolo)\n\n* tag \u0027for-linus-20190809\u0027 of git://git.kernel.dk/linux-block:\n bcache: Revert \"bcache: use sysfs_match_string() instead of __sysfs_match_string()\"\n loop: set PF_MEMALLOC_NOIO for the worker thread\n bdev: Fixup error handling in blkdev_get()\n block, bfq: handle NULL return value by bfq_init_rq()\n block, bfq: move update of waker and woken list to queue freeing\n block, bfq: reset last_completed_rq_bfqq if the pointed queue is freed\n block: aoe: Fix kernel crash due to atomic sleep when exiting\n libata: add SG safety checks in SFF pio transfers\n libata: have ata_scsi_rw_xlat() fail invalid passthrough requests\n block: fix O_DIRECT error handling for bio fragments\n ata: rb532_cf: Fix unused variable warning in rb532_pata_driver_probe\n" }, { "commit": "461d2815ac9ec1b3f66fc243cc39bd2374facafa", "tree": "975865d5ecd5387c0b1b7f3ed726698754d2d1a4", "parents": [ "ec4c99ad7bd23dd39ffb1381136cefa4bb632e31", "b803974a86039913d5280add083d730b2b9ed8ec" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:26:47 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:26:47 2019 -0700" }, "message": "Merge tag \u0027mmc-v5.3-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc\n\nPull MMC fixes from Ulf Hansson:\n\n - cavium: Fix DMA support\n\n - sdhci-sprd: Fix soft reset when runtime resuming\"\n\n* tag \u0027mmc-v5.3-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:\n mmc: cavium: Add the missing dma unmap when the dma has finished.\n mmc: cavium: Set the correct dma max segment size for mmc_host\n mmc: sdhci-sprd: Fix the incorrect soft reset operation when runtime resuming\n" }, { "commit": "ec4c99ad7bd23dd39ffb1381136cefa4bb632e31", "tree": "250db1e674cd0d18e899d894e8357fb2b40f8b23", "parents": [ "c808f2ad8fcddb856ce42901cd5b4f06f9da9184", "6a7553e8d84d5322d883cb83bb9888c49a0f04e0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:24:49 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:24:49 2019 -0700" }, "message": "Merge tag \u0027fbdev-v5.3-rc4\u0027 of git://github.com/bzolnier/linux\n\nPull fbdev fix from Bartlomiej Zolnierkiewicz:\n \"fbdev patches will now go to upstream through drm-misc tree for\n improved maintainership and better integration testing so update\n MAINTAINERS file accordingly\"\n\n* tag \u0027fbdev-v5.3-rc4\u0027 of git://github.com/bzolnier/linux:\n MAINTAINERS: handle fbdev changes through drm-misc tree\n" }, { "commit": "c808f2ad8fcddb856ce42901cd5b4f06f9da9184", "tree": "2201c88c000a3d52851d241501cdc06c2a487ddb", "parents": [ "cb42f06c9f418ec2ee56667d01e9ba965a4c61c9", "6cf9481b440da6d6d86bd8e4c99a8b553b9d1271" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:23:23 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:23:23 2019 -0700" }, "message": "Merge tag \u0027pwm/for-5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm\n\nPull pwm fix from Thierry Reding:\n \"A single fix for a backlight brightness regression introduced in\n this merge window\"\n\n* tag \u0027pwm/for-5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm:\n pwm: Fallback to the static lookup-list when acpi_pwm_get fails\n" }, { "commit": "cb42f06c9f418ec2ee56667d01e9ba965a4c61c9", "tree": "da6a0b3ff41dc4ee67358cc0faf4139a2ab71dc5", "parents": [ "6c02b6300fec062405ed0215e8ed1ed2b690da55", "1be3c1fae6c1e1f5bb982b255d2034034454527a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:21:27 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:21:27 2019 -0700" }, "message": "Merge tag \u0027sound-5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound\n\nPull sound fixes from Takashi Iwai:\n \"Lots of small fixes at this time since we\u0027ve received the ASoC fix\n batch now.\n\n - Some coverage in ASoC core mostly for minor issues like NULL checks\n for DPCM and proper error handling in DAI instantiation\n\n - A collection of small device-specific changes in various ASoC codec\n and platform drivers\n\n - OF-tree refcount fixes in a few ASoC drivers\n\n - Fixes of memory leaks in the error paths of various ASoC / ALSA\n drivers\n\n - A workaround for a long-standing issue on AMD HD-audio device\n\n - Updates of MAINTAINERS, mail addresses, file permission fixups\"\n\n* tag \u0027sound-5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (38 commits)\n ALSA: firewire: fix a memory leak bug\n sound: fix a memory leak bug\n ALSA: hda - Workaround for crackled sound on AMD controller (1022:1457)\n ALSA: hiface: fix multiple memory leak bugs\n ALSA: hda - Don\u0027t override global PCM hw info flag\n ALSA: usb-audio: fix a memory leak bug\n ASoC: max98373: Remove executable bits\n ASoC: amd: acp3x: use dma address for acp3x dma driver\n ASoC: amd: acp3x: use dma_ops of parent device for acp3x dma driver\n ASoC: max98373: add 88200 and 96000 sampling rate support\n ASoC: sun4i-i2s: Incorrect SR and WSS computation\n MAINTAINERS: Update Intel ASoC drivers maintainers\n ASoC: ti: davinci-mcasp: Correct slot_width posed constraint\n ASoC: rockchip: Fix mono capture\n ASoC: Intel: Fix some acpi vs apci typo in somme comments\n ASoC: ti: davinci-mcasp: Fix clk PDIR handling for i2s master mode\n ASoC: Fail card instantiation if DAI format setup fails\n ASoC: SOF: Intel: hda: remove misleading error trace from IRQ thread\n ASoC: qcom: apq8016_sbc: Fix oops with multiple DAI links\n ASoC: dapm: fix a memory leak bug\n ...\n" }, { "commit": "6c02b6300fec062405ed0215e8ed1ed2b690da55", "tree": "21f2b8c5b2b58943fce16bdd8a3e1d852ae37204", "parents": [ "0acf5676dc0ffe0683543a20d5ecbd112af5b8ee", "92f5b0313e37e2b37aaf8f0bb75b6c50eafb5808" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:19:53 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:19:53 2019 -0700" }, "message": "Merge tag \u0027media/v5.3-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media\n\nPull media fix from Mauro Carvalho Chehab:\n \"A fix at the vivid CEC support\"\n\n* tag \u0027media/v5.3-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:\n media: vivid: fix missing cec adapter name\n" }, { "commit": "0acf5676dc0ffe0683543a20d5ecbd112af5b8ee", "tree": "402a558747c4124b922c175592185778f29024f3", "parents": [ "e577dc152e232c78e5774e4c9b5486a04561920b", "0617bdede5114a0002298b12cd0ca2b0cfd0395d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:18:36 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:18:36 2019 -0700" }, "message": "Merge tag \u0027pm-5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fix from Rafael Wysocki:\n \"Revert a recent PCI power management change that caused problems to\n occur on multiple systems (Mika Westerberg)\"\n\n* tag \u0027pm-5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n Revert \"PCI: Add missing link delays required by the PCIe spec\"\n" }, { "commit": "e577dc152e232c78e5774e4c9b5486a04561920b", "tree": "800bdb8d7a539b7ccf5af67a4d36b54754094e92", "parents": [ "b678c568c561cf7e4ed40f4bcc3e85d2b50310a2", "e2664ecbb2f26225ac6646876f2899558ffb2604" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:17:05 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Aug 09 09:17:05 2019 -0700" }, "message": "Merge branch \u0027linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6\n\nPull crypto fixes from Herbert Xu:\n \"Fix a number of bugs in the ccp driver\"\n\n* \u0027linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:\n crypto: ccp - Ignore tag length when decrypting GCM ciphertext\n crypto: ccp - Add support for valid authsize values less than 16\n crypto: ccp - Fix oops by properly managing allocated structures\n" }, { "commit": "6a7553e8d84d5322d883cb83bb9888c49a0f04e0", "tree": "3bf8f36202f9a690b05c23fb7977a0df65e1f634", "parents": [ "e21a712a9685488f5ce80495b37b9fdbe96c230d" ], "author": { "name": "Bartlomiej Zolnierkiewicz", "email": "b.zolnierkie@samsung.com", "time": "Fri Aug 09 15:46:40 2019 +0200" }, "committer": { "name": "Bartlomiej Zolnierkiewicz", "email": "b.zolnierkie@samsung.com", "time": "Fri Aug 09 15:46:40 2019 +0200" }, "message": "MAINTAINERS: handle fbdev changes through drm-misc tree\n\nfbdev patches will now go to upstream through drm-misc tree (IOW\nstarting with v5.4 merge window fbdev changes will be included in\nDRM pull request) for improved maintainership and better integration\ntesting. Update MAINTAINERS file accordingly.\n\nAcked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nSigned-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n" }, { "commit": "20621fedb2a696e4dc60bc1c5de37cf21976abcb", "tree": "cb4f4c96be2e013659fabe7a6193a9ee4b7d53a6", "parents": [ "d0a255e795ab976481565f6ac178314b34fbf891" ], "author": { "name": "Coly Li", "email": "colyli@suse.de", "time": "Fri Aug 09 14:14:05 2019 +0800" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Aug 09 07:37:33 2019 -0600" }, "message": "bcache: Revert \"bcache: use sysfs_match_string() instead of __sysfs_match_string()\"\n\nThis reverts commit 89e0341af082dbc170019f908846f4a424efc86b.\n\nIn drivers/md/bcache/sysfs.c:bch_snprint_string_list(), NULL pointer at\nthe end of list is necessary. Remove the NULL from last element of each\nlists will cause the following panic,\n\n[ 4340.455652] bcache: register_cache() registered cache device nvme0n1\n[ 4340.464603] bcache: register_bdev() registered backing device sdk\n[ 4421.587335] bcache: bch_cached_dev_run() cached dev sdk is running already\n[ 4421.587348] bcache: bch_cached_dev_attach() Caching sdk as bcache0 on set 354e1d46-d99f-4d8b-870b-078b80dc88a6\n[ 5139.247950] general protection fault: 0000 [#1] SMP NOPTI\n[ 5139.247970] CPU: 9 PID: 5896 Comm: cat Not tainted 4.12.14-95.29-default #1 SLE12-SP4\n[ 5139.247988] Hardware name: HPE ProLiant DL380 Gen10/ProLiant DL380 Gen10, BIOS U30 04/18/2019\n[ 5139.248006] task: ffff888fb25c0b00 task.stack: ffff9bbacc704000\n[ 5139.248021] RIP: 0010:string+0x21/0x70\n[ 5139.248030] RSP: 0018:ffff9bbacc707bf0 EFLAGS: 00010286\n[ 5139.248043] RAX: ffffffffa7e432e3 RBX: ffff8881c20da02a RCX: ffff0a00ffffff04\n[ 5139.248058] RDX: 3f00656863616362 RSI: ffff8881c20db000 RDI: ffffffffffffffff\n[ 5139.248075] RBP: ffff8881c20db000 R08: 0000000000000000 R09: ffff8881c20da02a\n[ 5139.248090] R10: 0000000000000004 R11: 0000000000000000 R12: ffff9bbacc707c48\n[ 5139.248104] R13: 0000000000000fd6 R14: ffffffffc0665855 R15: ffffffffc0665855\n[ 5139.248119] FS: 00007faf253b8700(0000) GS:ffff88903f840000(0000) knlGS:0000000000000000\n[ 5139.248137] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[ 5139.248149] CR2: 00007faf25395008 CR3: 0000000f72150006 CR4: 00000000007606e0\n[ 5139.248164] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n[ 5139.248179] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n[ 5139.248193] PKRU: 55555554\n[ 5139.248200] Call Trace:\n[ 5139.248210] vsnprintf+0x1fb/0x510\n[ 5139.248221] snprintf+0x39/0x40\n[ 5139.248238] bch_snprint_string_list.constprop.15+0x5b/0x90 [bcache]\n[ 5139.248256] __bch_cached_dev_show+0x44d/0x5f0 [bcache]\n[ 5139.248270] ? __alloc_pages_nodemask+0xb2/0x210\n[ 5139.248284] bch_cached_dev_show+0x2c/0x50 [bcache]\n[ 5139.248297] sysfs_kf_seq_show+0xbb/0x190\n[ 5139.248308] seq_read+0xfc/0x3c0\n[ 5139.248317] __vfs_read+0x26/0x140\n[ 5139.248327] vfs_read+0x87/0x130\n[ 5139.248336] SyS_read+0x42/0x90\n[ 5139.248346] do_syscall_64+0x74/0x160\n[ 5139.248358] entry_SYSCALL_64_after_hwframe+0x3d/0xa2\n[ 5139.248370] RIP: 0033:0x7faf24eea370\n[ 5139.248379] RSP: 002b:00007fff82d03f38 EFLAGS: 00000246 ORIG_RAX: 0000000000000000\n[ 5139.248395] RAX: ffffffffffffffda RBX: 0000000000020000 RCX: 00007faf24eea370\n[ 5139.248411] RDX: 0000000000020000 RSI: 00007faf25396000 RDI: 0000000000000003\n[ 5139.248426] RBP: 00007faf25396000 R08: 00000000ffffffff R09: 0000000000000000\n[ 5139.248441] R10: 000000007c9d4d41 R11: 0000000000000246 R12: 00007faf25396000\n[ 5139.248456] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000fff\n[ 5139.248892] Code: ff ff ff 0f 1f 80 00 00 00 00 49 89 f9 48 89 cf 48 c7 c0 e3 32 e4 a7 48 c1 ff 30 48 81 fa ff 0f 00 00 48 0f 46 d0 48 85 ff 74 45 \u003c44\u003e 0f b6 02 48 8d 42 01 45 84 c0 74 38 48 01 fa 4c 89 cf eb 0e\n\nThe simplest way to fix is to revert commit 89e0341af082 (\"bcache: use\nsysfs_match_string() instead of __sysfs_match_string()\").\n\nThis bug was introduced in Linux v5.2, so this fix only applies to\nLinux v5.2 is enough for stable tree maintainer.\n\nFixes: 89e0341af082 (\"bcache: use sysfs_match_string() instead of __sysfs_match_string()\")\nCc: stable@vger.kernel.org\nCc: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nReported-by: Peifeng Lin \u003cpflin@suse.com\u003e\nAcked-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nSigned-off-by: Coly Li \u003ccolyli@suse.de\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "404861e15b5fa7edbab22400f9174c1a21fde731", "tree": "bb5aedd91edcb90d3d1099bf3b5f1c647d35cdfd", "parents": [ "739bacbf7aa2c44bb25d9ad5f7d5b256082c5e66" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Wed Aug 07 14:31:59 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Fri Aug 09 11:03:28 2019 +0200" }, "message": "s390/vdso: map vdso also for statically linked binaries\n\ns390 does not map the vdso for statically linked binaries, assuming\nthat this doesn\u0027t make sense. See commit fc5243d98ac2 (\"[S390]\narch_setup_additional_pages arguments\").\n\nHowever with glibc commit d665367f596d (\"linux: Enable vDSO for static\nlinking as default (BZ#19767)\") and commit 5e855c895401 (\"s390: Enable\nVDSO for static linking\") the vdso is also used for statically linked\nbinaries - if the kernel would make it available.\n\nTherefore map the vdso always, just like all other architectures.\n\nReported-by: Stefan Liebler \u003cstli@linux.ibm.com\u003e\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "a111ef6b082270f6cbeea5556caf1cbb0143b812", "tree": "23d6dbf282b4696f1be16539d8e203fbc93d88d1", "parents": [ "359f1dcbae12b691fa40aae912fab4c129cb7561", "73a0ff0b30af79bf0303d557eb82f1d1945bb6ee" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Aug 09 15:46:09 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Aug 09 15:46:10 2019 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2019-08-08\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\ndrm/i915 fixes for v5.3-rc4:\n- Fix GLK DSI escape clock setting\n- Fix a memleak on HDCP revoked Ksv error path\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Jani Nikula \u003cjani.nikula@intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/87pnlghz79.fsf@intel.com\n" }, { "commit": "359f1dcbae12b691fa40aae912fab4c129cb7561", "tree": "7f5d5eab96fc5d49cb7b9fd3ed0d333a41c5439c", "parents": [ "6ca847a947d9da958047dba570d4727069a4861b", "f7ccbed656f78212593ca965d9a8f34bf24e0aab" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Aug 09 15:45:02 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Aug 09 15:45:09 2019 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2019-08-08\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\ndrm-misc-fixes for v5.3-rc4:\n- Suspend fix for rockchip\n- Fix unterminated strncpy cmdline mode parser\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/ace294a6-6bb2-d9b1-695d-3260e1d60831@linux.intel.com\n" }, { "commit": "6ca847a947d9da958047dba570d4727069a4861b", "tree": "2d1c63d57abcf7ae63b81a4a5b08024cda989993", "parents": [ "f536579c148249505b388d525ac1866e080fd66a", "6b7c3b86f0b63134b2ab56508921a0853ffa687a" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Aug 09 15:40:06 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Aug 09 15:41:33 2019 +1000" }, "message": "Merge branch \u0027vmwgfx-fixes-5.3\u0027 of git://people.freedesktop.org/~thomash/linux into drm-fixes\n\nOne single memory leak fix.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Thomas Hellstrom \"VMware\" \u003cthomas@shipmail.org\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20190808094615.31040-1-thomas@shipmail.org\n" }, { "commit": "b678c568c561cf7e4ed40f4bcc3e85d2b50310a2", "tree": "0e3e8943553f94a9ce9315129789f8eb8e103710", "parents": [ "518a1c2f091290219783d5624284c140544a261e", "67e7b52d44e3d539dfbfcd866c3d3d69da23a909" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Aug 08 14:47:19 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Aug 08 14:47:19 2019 -0700" }, "message": "Merge tag \u0027nfs-for-5.3-2\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs\n\nPull NFS client fixes from Trond Myklebust:\n \"Highlights include:\n\n Stable fixes:\n\n - NFSv4: Ensure we check the return value of update_open_stateid() so\n we correctly track active open state.\n\n - NFSv4: Fix for delegation state recovery to ensure we recover all\n open modes that are active.\n\n - NFSv4: Fix an Oops in nfs4_do_setattr\n\n Fixes:\n\n - NFS: Fix regression whereby fscache errors are appearing on \u0027nofsc\u0027\n mounts\n\n - NFSv4: Fix a potential sleep while atomic in nfs4_do_reclaim()\n\n - NFSv4: Fix a credential refcount leak in nfs41_check_delegation_stateid\n\n - pNFS: Report errors from the call to nfs4_select_rw_stateid()\n\n - NFSv4: Various other delegation and open stateid recovery fixes\n\n - NFSv4: Fix state recovery behaviour when server connection times\n out\"\n\n* tag \u0027nfs-for-5.3-2\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs:\n NFSv4: Ensure state recovery handles ETIMEDOUT correctly\n NFS: Fix regression whereby fscache errors are appearing on \u0027nofsc\u0027 mounts\n NFSv4: Fix an Oops in nfs4_do_setattr\n NFSv4: Fix a potential sleep while atomic in nfs4_do_reclaim()\n NFSv4: Check the return value of update_open_stateid()\n NFSv4.1: Only reap expired delegations\n NFSv4.1: Fix open stateid recovery\n NFSv4: Report the error from nfs4_select_rw_stateid()\n NFSv4: When recovering state fails with EAGAIN, retry the same recovery\n NFSv4: Print an error in the syslog when state is marked as irrecoverable\n NFSv4: Fix delegation state recovery\n NFSv4: Fix a credential refcount leak in nfs41_check_delegation_stateid\n" }, { "commit": "30e235389faadb9e3d918887b1f126155d7d761d", "tree": "00e6b4e2fe87f7edd0f2ed98acf1b802d721ae7f", "parents": [ "e21a712a9685488f5ce80495b37b9fdbe96c230d" ], "author": { "name": "Jia He", "email": "justin.he@arm.com", "time": "Wed Aug 07 12:58:51 2019 +0800" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Thu Aug 08 18:38:20 2019 +0100" }, "message": "arm64: mm: add missing PTE_SPECIAL in pte_mkdevmap on arm64\n\nWithout this patch, the MAP_SYNC test case will cause a print_bad_pte\nwarning on arm64 as follows:\n\n[ 25.542693] BUG: Bad page map in process mapdax333 pte:2e8000448800f53 pmd:41ff5f003\n[ 25.546360] page:ffff7e0010220000 refcount:1 mapcount:-1 mapping:ffff8003e29c7440 index:0x0\n[ 25.550281] ext4_dax_aops\n[ 25.550282] name:\"__aaabbbcccddd__\"\n[ 25.551553] flags: 0x3ffff0000001002(referenced|reserved)\n[ 25.555802] raw: 03ffff0000001002 ffff8003dfffa908 0000000000000000 ffff8003e29c7440\n[ 25.559446] raw: 0000000000000000 0000000000000000 00000001fffffffe 0000000000000000\n[ 25.563075] page dumped because: bad pte\n[ 25.564938] addr:0000ffffbe05b000 vm_flags:208000fb anon_vma:0000000000000000 mapping:ffff8003e29c7440 index:0\n[ 25.574272] file:__aaabbbcccddd__ fault:ext4_dax_fault mmmmap:ext4_file_mmap readpage:0x0\n[ 25.578799] CPU: 1 PID: 1180 Comm: mapdax333 Not tainted 5.2.0+ #21\n[ 25.581702] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015\n[ 25.585624] Call trace:\n[ 25.587008] dump_backtrace+0x0/0x178\n[ 25.588799] show_stack+0x24/0x30\n[ 25.590328] dump_stack+0xa8/0xcc\n[ 25.591901] print_bad_pte+0x18c/0x218\n[ 25.593628] unmap_page_range+0x778/0xc00\n[ 25.595506] unmap_single_vma+0x94/0xe8\n[ 25.597304] unmap_vmas+0x90/0x108\n[ 25.598901] unmap_region+0xc0/0x128\n[ 25.600566] __do_munmap+0x284/0x3f0\n[ 25.602245] __vm_munmap+0x78/0xe0\n[ 25.603820] __arm64_sys_munmap+0x34/0x48\n[ 25.605709] el0_svc_common.constprop.0+0x78/0x168\n[ 25.607956] el0_svc_handler+0x34/0x90\n[ 25.609698] el0_svc+0x8/0xc\n[...]\n\nThe root cause is in _vm_normal_page, without the PTE_SPECIAL bit,\nthe return value will be incorrectly set to pfn_to_page(pfn) instead\nof NULL. Besides, this patch also rewrite the pmd_mkdevmap to avoid\nsetting PTE_SPECIAL for pmd\n\nThe MAP_SYNC test case is as follows(Provided by Yibo Cai)\n$#include \u003cstdio.h\u003e\n$#include \u003cstring.h\u003e\n$#include \u003cunistd.h\u003e\n$#include \u003csys/file.h\u003e\n$#include \u003csys/mman.h\u003e\n\n$#ifndef MAP_SYNC\n$#define MAP_SYNC 0x80000\n$#endif\n\n/* mount -o dax /dev/pmem0 /mnt */\n$#define F \"/mnt/__aaabbbcccddd__\"\n\nint main(void)\n{\n int fd;\n char buf[4096];\n void *addr;\n\n if ((fd \u003d open(F, O_CREAT|O_TRUNC|O_RDWR, 0644)) \u003c 0) {\n perror(\"open1\");\n return 1;\n }\n\n if (write(fd, buf, 4096) !\u003d 4096) {\n perror(\"lseek\");\n return 1;\n }\n\n addr \u003d mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED|MAP_SYNC, fd, 0);\n if (addr \u003d\u003d MAP_FAILED) {\n perror(\"mmap\");\n printf(\"did you mount with \u0027-o dax\u0027?\\n\");\n return 1;\n }\n\n memset(addr, 0x55, 4096);\n\n if (munmap(addr, 4096) \u003d\u003d -1) {\n perror(\"munmap\");\n return 1;\n }\n\n close(fd);\n\n return 0;\n}\n\nFixes: 73b20c84d42d (\"arm64: mm: implement pte_devmap support\")\nReported-by: Yibo Cai \u003cYibo.Cai@arm.com\u003e\nAcked-by: Will Deacon \u003cwill@kernel.org\u003e\nAcked-by: Robin Murphy \u003cRobin.Murphy@arm.com\u003e\nSigned-off-by: Jia He \u003cjustin.he@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n" }, { "commit": "518a1c2f091290219783d5624284c140544a261e", "tree": "bb0c4b58b67666406b7297ff6d297e2288751212", "parents": [ "ecb095bff5d4b8711a81968625b3b4a235d3e477", "ee9d66182392695535cc9fccfcb40c16f72de2a9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Aug 08 09:57:50 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Aug 08 09:57:50 2019 -0700" }, "message": "Merge tag \u00275.3-rc3-smb3-fixes\u0027 of git://git.samba.org/sfrench/cifs-2.6\n\nPull cifs fixes from Steve French:\n \"Six small SMB3 fixes, two for stable\"\n\n* tag \u00275.3-rc3-smb3-fixes\u0027 of git://git.samba.org/sfrench/cifs-2.6:\n SMB3: Kernel oops mounting a encryptData share with CONFIG_DEBUG_VIRTUAL\n smb3: update TODO list of missing features\n smb3: send CAP_DFS capability during session setup\n SMB3: Fix potential memory leak when processing compound chain\n SMB3: Fix deadlock in validate negotiate hits reconnect\n cifs: fix rmmod regression in cifs.ko caused by force_sig changes\n" }, { "commit": "d0a255e795ab976481565f6ac178314b34fbf891", "tree": "2b8892d852f271573516310f24f6007f9ddded69", "parents": [ "e91455bad5cff40a8c232f2204a5104127e3fec2" ], "author": { "name": "Mikulas Patocka", "email": "mpatocka@redhat.com", "time": "Thu Aug 08 11:17:01 2019 -0400" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Aug 08 10:12:21 2019 -0600" }, "message": "loop: set PF_MEMALLOC_NOIO for the worker thread\n\nA deadlock with this stacktrace was observed.\n\nThe loop thread does a GFP_KERNEL allocation, it calls into dm-bufio\nshrinker and the shrinker depends on I/O completion in the dm-bufio\nsubsystem.\n\nIn order to fix the deadlock (and other similar ones), we set the flag\nPF_MEMALLOC_NOIO at loop thread entry.\n\nPID: 474 TASK: ffff8813e11f4600 CPU: 10 COMMAND: \"kswapd0\"\n #0 [ffff8813dedfb938] __schedule at ffffffff8173f405\n #1 [ffff8813dedfb990] schedule at ffffffff8173fa27\n #2 [ffff8813dedfb9b0] schedule_timeout at ffffffff81742fec\n #3 [ffff8813dedfba60] io_schedule_timeout at ffffffff8173f186\n #4 [ffff8813dedfbaa0] bit_wait_io at ffffffff8174034f\n #5 [ffff8813dedfbac0] __wait_on_bit at ffffffff8173fec8\n #6 [ffff8813dedfbb10] out_of_line_wait_on_bit at ffffffff8173ff81\n #7 [ffff8813dedfbb90] __make_buffer_clean at ffffffffa038736f [dm_bufio]\n #8 [ffff8813dedfbbb0] __try_evict_buffer at ffffffffa0387bb8 [dm_bufio]\n #9 [ffff8813dedfbbd0] dm_bufio_shrink_scan at ffffffffa0387cc3 [dm_bufio]\n #10 [ffff8813dedfbc40] shrink_slab at ffffffff811a87ce\n #11 [ffff8813dedfbd30] shrink_zone at ffffffff811ad778\n #12 [ffff8813dedfbdc0] kswapd at ffffffff811ae92f\n #13 [ffff8813dedfbec0] kthread at ffffffff810a8428\n #14 [ffff8813dedfbf50] ret_from_fork at ffffffff81745242\n\n PID: 14127 TASK: ffff881455749c00 CPU: 11 COMMAND: \"loop1\"\n #0 [ffff88272f5af228] __schedule at ffffffff8173f405\n #1 [ffff88272f5af280] schedule at ffffffff8173fa27\n #2 [ffff88272f5af2a0] schedule_preempt_disabled at ffffffff8173fd5e\n #3 [ffff88272f5af2b0] __mutex_lock_slowpath at ffffffff81741fb5\n #4 [ffff88272f5af330] mutex_lock at ffffffff81742133\n #5 [ffff88272f5af350] dm_bufio_shrink_count at ffffffffa03865f9 [dm_bufio]\n #6 [ffff88272f5af380] shrink_slab at ffffffff811a86bd\n #7 [ffff88272f5af470] shrink_zone at ffffffff811ad778\n #8 [ffff88272f5af500] do_try_to_free_pages at ffffffff811adb34\n #9 [ffff88272f5af590] try_to_free_pages at ffffffff811adef8\n #10 [ffff88272f5af610] __alloc_pages_nodemask at ffffffff811a09c3\n #11 [ffff88272f5af710] alloc_pages_current at ffffffff811e8b71\n #12 [ffff88272f5af760] new_slab at ffffffff811f4523\n #13 [ffff88272f5af7b0] __slab_alloc at ffffffff8173a1b5\n #14 [ffff88272f5af880] kmem_cache_alloc at ffffffff811f484b\n #15 [ffff88272f5af8d0] do_blockdev_direct_IO at ffffffff812535b3\n #16 [ffff88272f5afb00] __blockdev_direct_IO at ffffffff81255dc3\n #17 [ffff88272f5afb30] xfs_vm_direct_IO at ffffffffa01fe3fc [xfs]\n #18 [ffff88272f5afb90] generic_file_read_iter at ffffffff81198994\n #19 [ffff88272f5afc50] __dta_xfs_file_read_iter_2398 at ffffffffa020c970 [xfs]\n #20 [ffff88272f5afcc0] lo_rw_aio at ffffffffa0377042 [loop]\n #21 [ffff88272f5afd70] loop_queue_work at ffffffffa0377c3b [loop]\n #22 [ffff88272f5afe60] kthread_worker_fn at ffffffff810a8a0c\n #23 [ffff88272f5afec0] kthread at ffffffff810a8428\n #24 [ffff88272f5aff50] ret_from_fork at ffffffff81745242\n\nSigned-off-by: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "e91455bad5cff40a8c232f2204a5104127e3fec2", "tree": "bb50daf2d6637e3d9e5dbab4fc613075a19af297", "parents": [ "fd03177c33b287c6541f4048f1d67b7b45a1abc9" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Wed Aug 07 11:36:47 2019 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Aug 08 07:37:03 2019 -0600" }, "message": "bdev: Fixup error handling in blkdev_get()\n\nCommit 89e524c04fa9 (\"loop: Fix mount(2) failure due to race with\nLOOP_SET_FD\") converted blkdev_get() to use the new helpers for\nfinishing claiming of a block device. However the conversion botched the\nerror handling in blkdev_get() and thus the bdev has been marked as held\neven in case __blkdev_get() returned error. This led to occasional\nwarnings with block/001 test from blktests like:\n\nkernel: WARNING: CPU: 5 PID: 907 at fs/block_dev.c:1899 __blkdev_put+0x396/0x3a0\n\nCorrect the error handling.\n\nCC: stable@vger.kernel.org\nFixes: 89e524c04fa9 (\"loop: Fix mount(2) failure due to race with LOOP_SET_FD\")\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "fd03177c33b287c6541f4048f1d67b7b45a1abc9", "tree": "1e95e71d6ba6bd86620b9d1267b89e463ed4df2c", "parents": [ "3f758e844aa9800eb660d60ee10226fa802594d4" ], "author": { "name": "Paolo Valente", "email": "paolo.valente@linaro.org", "time": "Wed Aug 07 19:21:11 2019 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Aug 08 07:31:50 2019 -0600" }, "message": "block, bfq: handle NULL return value by bfq_init_rq()\n\nAs reported in [1], the call bfq_init_rq(rq) may return NULL in case\nof OOM (in particular, if rq-\u003eelv.icq is NULL because memory\nallocation failed in failed in ioc_create_icq()).\n\nThis commit handles this circumstance.\n\n[1] https://lkml.org/lkml/2019/7/22/824\n\nCc: Hsin-Yi Wang \u003chsinyi@google.com\u003e\nCc: Nicolas Boichat \u003cdrinkcat@chromium.org\u003e\nCc: Doug Anderson \u003cdianders@chromium.org\u003e\nReported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\nReported-by: Hsin-Yi Wang \u003chsinyi@google.com\u003e\nReviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\nSigned-off-by: Paolo Valente \u003cpaolo.valente@linaro.org\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "3f758e844aa9800eb660d60ee10226fa802594d4", "tree": "25fa18af9d966e3f30ad409b51190e61a205fad3", "parents": [ "08d383a74948b43eb6e96c86153e63cbf276f1fa" ], "author": { "name": "Paolo Valente", "email": "paolo.valente@linaro.org", "time": "Wed Aug 07 16:17:54 2019 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Aug 08 07:30:52 2019 -0600" }, "message": "block, bfq: move update of waker and woken list to queue freeing\n\nSince commit 13a857a4c4e8 (\"block, bfq: detect wakers and\nunconditionally inject their I/O\"), every bfq_queue has a pointer to a\nwaker bfq_queue and a list of the bfq_queues it may wake. In this\nrespect, when a bfq_queue, say Q, remains with no I/O source attached\nto it, Q cannot be woken by any other bfq_queue, and cannot wake any\nother bfq_queue. Then Q must be removed from the woken list of its\npossible waker bfq_queue, and all bfq_queues in the woken list of Q\nmust stop having a waker bfq_queue.\n\nQ remains with no I/O source in two cases: when the last process\nassociated with Q exits or when such a process gets associated with a\ndifferent bfq_queue. Unfortunately, commit 13a857a4c4e8 (\"block, bfq:\ndetect wakers and unconditionally inject their I/O\") performed the\nabove updates only in the first case.\n\nThis commit fixes this bug by moving these updates to when Q gets\nfreed. This is a simple and safe way to handle all cases, as both the\nabove events, process exit and re-association, lead to Q being freed\nsoon, and because dangling references would come out only after Q gets\nfreed (if no update were performed).\n\nFixes: 13a857a4c4e8 (\"block, bfq: detect wakers and unconditionally inject their I/O\")\nReported-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nTested-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nSigned-off-by: Paolo Valente \u003cpaolo.valente@linaro.org\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "08d383a74948b43eb6e96c86153e63cbf276f1fa", "tree": "7b893061ad7cee76de8e1b868b0c23ed45200ffc", "parents": [ "430380b4637aec646996b4aef67ad417593923b2" ], "author": { "name": "Paolo Valente", "email": "paolo.valente@linaro.org", "time": "Wed Aug 07 16:17:53 2019 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Aug 08 07:30:50 2019 -0600" }, "message": "block, bfq: reset last_completed_rq_bfqq if the pointed queue is freed\n\nSince commit 13a857a4c4e8 (\"block, bfq: detect wakers and\nunconditionally inject their I/O\"), BFQ stores, in a per-device\npointer last_completed_rq_bfqq, the last bfq_queue that had an I/O\nrequest completed. If some bfq_queue receives new I/O right after the\nlast request of last_completed_rq_bfqq has been completed, then\nlast_completed_rq_bfqq may be a waker bfq_queue.\n\nBut if the bfq_queue last_completed_rq_bfqq points to is freed, then\nlast_completed_rq_bfqq becomes a dangling reference. This commit\nresets last_completed_rq_bfqq if the pointed bfq_queue is freed.\n\nFixes: 13a857a4c4e8 (\"block, bfq: detect wakers and unconditionally inject their I/O\")\nReported-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nTested-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nSigned-off-by: Paolo Valente \u003cpaolo.valente@linaro.org\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "430380b4637aec646996b4aef67ad417593923b2", "tree": "296c33e5686cea56f73366525ca7528df2b55fe2", "parents": [ "752ead44491e8c91e14d7079625c5916b30921c5" ], "author": { "name": "He Zhe", "email": "zhe.he@windriver.com", "time": "Thu Aug 08 11:09:54 2019 +0800" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Aug 08 07:29:02 2019 -0600" }, "message": "block: aoe: Fix kernel crash due to atomic sleep when exiting\n\nSince commit 3582dd291788 (\"aoe: convert aoeblk to blk-mq\"), aoedev_downdev\nhas had the possibility of sleeping and causing the following crash.\n\nBUG: scheduling while atomic: rmmod/2242/0x00000003\nModules linked in: aoe\nPreemption disabled at:\n[\u003cffffffffc01d95e5\u003e] flush+0x95/0x4a0 [aoe]\nCPU: 7 PID: 2242 Comm: rmmod Tainted: G I 5.2.3 #1\nHardware name: Intel Corporation S5520HC/S5520HC, BIOS S5500.86B.01.10.0025.030220091519 03/02/2009\nCall Trace:\n dump_stack+0x4f/0x6a\n ? flush+0x95/0x4a0 [aoe]\n __schedule_bug.cold+0x44/0x54\n __schedule+0x44f/0x680\n schedule+0x44/0xd0\n blk_mq_freeze_queue_wait+0x46/0xb0\n ? wait_woken+0x80/0x80\n blk_mq_freeze_queue+0x1b/0x20\n aoedev_downdev+0x111/0x160 [aoe]\n flush+0xff/0x4a0 [aoe]\n aoedev_exit+0x23/0x30 [aoe]\n aoe_exit+0x35/0x948 [aoe]\n __se_sys_delete_module+0x183/0x210\n __x64_sys_delete_module+0x16/0x20\n do_syscall_64+0x4d/0x130\n entry_SYSCALL_64_after_hwframe+0x44/0xa9\nRIP: 0033:0x7f24e0043b07\nCode: 73 01 c3 48 8b 0d 89 73 0b 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f\n1f 84 00 00 00 00 00 0f 1f 44 00 00 b8 b0 00 00 00 0f 05 \u003c48\u003e 3d 01 f0 ff\nff 73 01 c3 48 8b 0d 59 73 0b 00 f7 d8 64 89 01 48\nRSP: 002b:00007ffe18f7f1e8 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0\nRAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f24e0043b07\nRDX: 000000000000000a RSI: 0000000000000800 RDI: 0000555c3ecf87c8\nRBP: 00007ffe18f7f1f0 R08: 0000000000000000 R09: 0000000000000000\nR10: 00007f24e00b4ac0 R11: 0000000000000206 R12: 00007ffe18f7f238\nR13: 00007ffe18f7f410 R14: 00007ffe18f80e73 R15: 0000555c3ecf8760\n\nThis patch, handling in the same way of pass two, unlocks the locks and\nrestart pass one after aoedev_downdev is done.\n\nFixes: 3582dd291788 (\"aoe: convert aoeblk to blk-mq\")\nSigned-off-by: He Zhe \u003czhe.he@windriver.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "739bacbf7aa2c44bb25d9ad5f7d5b256082c5e66", "tree": "b3217c13a4fc437a62764a9a1027b6da4c17f58a", "parents": [ "7bac98707f65b93bf994ef4e99b1eb9e7dbb9c32" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Mon Jan 21 13:54:53 2019 +0100" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Thu Aug 08 14:36:53 2019 +0200" }, "message": "s390/build: use size command to perform empty .bss check\n\nCurrently empty .bss checks performed do not pay attention to \"common\nobjects\" in object files which end up in .bss section eventually.\n\nThe \"size\" tool is a part of binutils and since version 2.18 provides\n\"--common\" command line option, which allows to account \"common objects\"\nsizes in .bss section size. Utilize \"size --common\" to perform accurate\ncheck that .bss section is unused. Besides that the size tool handles\nobject files without .bss section gracefully and doesn\u0027t require\nadditional objdump run.\n\nThe linux kernel requires binutils 2.20 since 4.13.\n\nKbuild exports OBJSIZE to reference the right size tool.\n\nLink: http://lkml.kernel.org/r/patch-2.thread-2257a1.git-2257a1c53d4a.your-ad-here.call-01565088755-ext-5120@work.hours\nReported-and-tested-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nAcked-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "7bac98707f65b93bf994ef4e99b1eb9e7dbb9c32", "tree": "72fcf82b7d0cbebafb960d2b4da906e843453b94", "parents": [ "24350fdadbdec780406a1ef988e6cd3875e374a8" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Mon Jan 21 13:54:39 2019 +0100" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Thu Aug 08 14:36:51 2019 +0200" }, "message": "kbuild: add OBJSIZE variable for the size tool\n\nDefine and export OBJSIZE variable for \"size\" tool from binutils to be\nused in architecture specific Makefiles (naming the variable just \"SIZE\"\nwould be too risky). In particular this tool is useful to perform checks\nthat early boot code is not using bss section (which might have not been\nzeroed yet or intersects with initrd or other files boot loader might\nhave put right after the linux kernel).\n\nLink: http://lkml.kernel.org/r/patch-1.thread-2257a1.git-188f5a3d81d5.your-ad-here.call-01565088755-ext-5120@work.hours\nAcked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "6cf9481b440da6d6d86bd8e4c99a8b553b9d1271", "tree": "b329dee49270d5708aaeee9b86c272a774f2f89d", "parents": [ "5f9e832c137075045d15cd6899ab0505cfb2ca4b" ], "author": { "name": "Hans de Goede", "email": "hdegoede@redhat.com", "time": "Tue Jul 30 17:48:48 2019 +0200" }, "committer": { "name": "Thierry Reding", "email": "thierry.reding@gmail.com", "time": "Thu Aug 08 13:17:38 2019 +0200" }, "message": "pwm: Fallback to the static lookup-list when acpi_pwm_get fails\n\nCommit 4a6ef8e37c4d (\"pwm: Add support referencing PWMs from ACPI\")\nmade pwm_get unconditionally return the acpi_pwm_get return value if\nthe device passed to pwm_get has an ACPI fwnode.\n\nBut even if the passed in device has an ACPI fwnode, it does not\nnecessarily have the necessary ACPI package defining its pwm bindings,\nespecially since the binding / API of this ACPI package has only been\nintroduced very recently.\n\nUp until now X86/ACPI devices which use a separate pwm controller for\ncontrolling their LCD screen\u0027s backlight brightness have been relying\non the static lookup-list to get their pwm.\n\npwm_get unconditionally returning the acpi_pwm_get return value breaks\nthis, breaking backlight control on these devices.\n\nThis commit fixes this by making pwm_get fall back to the static\nlookup-list if acpi_pwm_get returns -ENOENT.\n\nBugLink: https://bugs.freedesktop.org/show_bug.cgi?id\u003d96571\nReported-by: youling257@gmail.com\nFixes: 4a6ef8e37c4d (\"pwm: Add support referencing PWMs from ACPI\")\nCc: Nikolaus Voss \u003cnikolaus.voss@loewensteinmedical.de\u003e\nSigned-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nReviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nAcked-by: Nikolaus Voss \u003cnikolaus.voss@loewensteinmedical.de\u003e\nSigned-off-by: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n" }, { "commit": "6b7c3b86f0b63134b2ab56508921a0853ffa687a", "tree": "2d1c63d57abcf7ae63b81a4a5b08024cda989993", "parents": [ "f536579c148249505b388d525ac1866e080fd66a" ], "author": { "name": "Colin Ian King", "email": "colin.king@canonical.com", "time": "Mon Jun 24 09:39:59 2019 -0700" }, "committer": { "name": "Thomas Hellstrom", "email": "thellstrom@vmware.com", "time": "Thu Aug 08 11:22:54 2019 +0200" }, "message": "drm/vmwgfx: fix memory leak when too many retries have occurred\n\nCurrently when too many retries have occurred there is a memory\nleak on the allocation for reply on the error return path. Fix\nthis by kfree\u0027ing reply before returning.\n\nAddresses-Coverity: (\"Resource leak\")\nFixes: a9cd9c044aa9 (\"drm/vmwgfx: Add a check to handle host message failure\")\nSigned-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\nReviewed-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\nSigned-off-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\nSigned-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n" }, { "commit": "1be3c1fae6c1e1f5bb982b255d2034034454527a", "tree": "6804594da24bb8a8aa3ffe48362da900d45e8bdb", "parents": [ "c7cd7c748a3250ca33509f9235efab9c803aca09" ], "author": { "name": "Wenwen Wang", "email": "wenwen@cs.uga.edu", "time": "Thu Aug 08 00:50:58 2019 -0500" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Thu Aug 08 11:12:26 2019 +0200" }, "message": "ALSA: firewire: fix a memory leak bug\n\nIn iso_packets_buffer_init(), \u0027b-\u003epackets\u0027 is allocated through\nkmalloc_array(). Then, the aligned packet size is checked. If it is\nlarger than PAGE_SIZE, -EINVAL will be returned to indicate the error.\nHowever, the allocated \u0027b-\u003epackets\u0027 is not deallocated on this path,\nleading to a memory leak.\n\nTo fix the above issue, free \u0027b-\u003epackets\u0027 before returning the error code.\n\nFixes: 31ef9134eb52 (\"ALSA: add LaCie FireWire Speakers/Griffin FireWave Surround driver\")\nSigned-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nReviewed-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nCc: \u003cstable@vger.kernel.org\u003e # v2.6.39+\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" }, { "commit": "c7cd7c748a3250ca33509f9235efab9c803aca09", "tree": "159ab4582286ab702dd0db9c84904a968ab325df", "parents": [ "c02f77d32d2c45cfb1b2bb99eabd8a78f5ecc7db" ], "author": { "name": "Wenwen Wang", "email": "wenwen@cs.uga.edu", "time": "Thu Aug 08 00:15:21 2019 -0500" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Thu Aug 08 08:18:32 2019 +0200" }, "message": "sound: fix a memory leak bug\n\nIn sound_insert_unit(), the controlling structure \u0027s\u0027 is allocated through\nkmalloc(). Then it is added to the sound driver list by invoking\n__sound_insert_unit(). Later on, if __register_chrdev() fails, \u0027s\u0027 is\nremoved from the list through __sound_remove_unit(). If \u0027index\u0027 is not less\nthan 0, -EBUSY is returned to indicate the error. However, \u0027s\u0027 is not\ndeallocated on this execution path, leading to a memory leak bug.\n\nTo fix the above issue, free \u0027s\u0027 before -EBUSY is returned.\n\nSigned-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" }, { "commit": "f536579c148249505b388d525ac1866e080fd66a", "tree": "88048b9899e4ff280ac25bd3bb830d270130acdb", "parents": [ "abffedaaa7702d1194934169fe65553804a2b064", "4b3e30ed3ec7864e798403a63ff2e96bd0c19ab0" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Aug 08 13:25:49 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Aug 08 13:25:50 2019 +1000" }, "message": "Merge tag \u0027drm-fixes-5.3-2019-08-07\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\ndrm-fixes-5.3-2019-08-07:\n\namdgpu:\n- Fixes VCN to handle the latest navi10 firmware\n- Fixes for fan control on navi10\n- Properly handle SMU metrics table on navi10\n- Fix a resume regression on Stoney\n\namdkfd:\n- Revert new GWS ioctl. It\u0027s not ready.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexdeucher@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20190807184221.3323-1-alexander.deucher@amd.com\n" }, { "commit": "abffedaaa7702d1194934169fe65553804a2b064", "tree": "4499a12397867d1bb71dc78eab6e31c91028f5fe", "parents": [ "e21a712a9685488f5ce80495b37b9fdbe96c230d", "2a6fc3cb5cb68597f1072bfeef28d2ca02310220" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Aug 08 13:21:46 2019 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Aug 08 13:25:28 2019 +1000" }, "message": "Merge tag \u0027drm/tegra/for-5.3-rc4\u0027 of git://anongit.freedesktop.org/tegra/linux into drm-fixes\n\ndrm/tegra: Fixes for v5.3-rc4\n\nThis contains a single fix for a regression introduced by a combination\nof a GPIO and a drm/tegra patch merged in v5.3-rc1.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Thierry Reding \u003cthierry.reding@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20190807140634.29166-1-thierry.reding@gmail.com\n" }, { "commit": "ecb095bff5d4b8711a81968625b3b4a235d3e477", "tree": "93076ca82d7a48e359d20085711f8a53ac1cc830", "parents": [ "33920f1ec5bf47c5c0a1d2113989bdd9dfb3fae9", "a95a4f3f2702b55a89393bf0f1b2b3d79e0f7da2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 07 15:06:31 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Aug 07 15:06:31 2019 -0700" }, "message": "Merge tag \u0027hwmon-for-v5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging\n\nPull hwmon fixes from Guenter Roeck:\n \"Fixes to lm75 and nct7802 drivers\n\n In the lm75 driver, fix TMP75B chip description to ensure correct\n initialization. In the nct7802 driver, fix in4 presence detection\"\n\n* tag \u0027hwmon-for-v5.3-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:\n hwmon: (lm75) Fixup tmp75b clr_mask\n hwmon: (nct7802) Fix wrong detection of in4 presence\n" }, { "commit": "a95a4f3f2702b55a89393bf0f1b2b3d79e0f7da2", "tree": "015b2e5ac38396edee6023302ea974331ffb64fc", "parents": [ "38ada2f406a9b81fb1249c5c9227fa657e7d5671" ], "author": { "name": "Iker Perez del Palomar Sustatxa", "email": "iker.perez@codethink.co.uk", "time": "Thu Aug 01 08:53:24 2019 +0100" }, "committer": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Wed Aug 07 14:50:49 2019 -0700" }, "message": "hwmon: (lm75) Fixup tmp75b clr_mask\n\nThe configuration register of the tmp75b sensor is 16bit long, however\nthe first byte is reserved, so there is not no need to take care of it.\n\nBecause the order of the bytes is little endian and it is only necessary\nto write one byte, the desired bits must be shifted into a 8 bit range.\n\nFixes: 39abe9d88b30 (\"hwmon: (lm75) Add support for TMP75B\")\nCc: stable@vger.kernel.org\nSigned-off-by: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\nLink: https://lore.kernel.org/r/20190801075324.4638-1-iker.perez@codethink.co.uk\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n" }, { "commit": "38ada2f406a9b81fb1249c5c9227fa657e7d5671", "tree": "a9a690d2a889a7b1a7033fbcf9c95da0c3711815", "parents": [ "609488bc979f99f805f34e9a32c1e3b71179d10b" ], "author": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Fri Jul 26 08:00:49 2019 -0700" }, "committer": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Wed Aug 07 14:49:46 2019 -0700" }, "message": "hwmon: (nct7802) Fix wrong detection of in4 presence\n\nThe code to detect if in4 is present is wrong; if in4 is not present,\nthe in4_input sysfs attribute is still present.\n\nIn detail:\n\n- Ihen RTD3_MD\u003d11 (VSEN3 present), everything is as expected (no bug).\n- If we have RTD3_MD!\u003d11 (no VSEN3), we unexpectedly have a in4_input\n file under /sys and the \"sensors\" command displays in4_input.\n But as expected, we have no in4_min, in4_max, in4_alarm, in4_beep.\n\nFix is_visible function to detect and report in4_input visibility\nas expected.\n\nReported-by: Gilles Buloz \u003cGilles.Buloz@kontron.com\u003e\nCc: Gilles Buloz \u003cGilles.Buloz@kontron.com\u003e\nCc: stable@vger.kernel.org\nFixes: 3434f37835804 (\"hwmon: Driver for Nuvoton NCT7802Y\")\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n" }, { "commit": "752ead44491e8c91e14d7079625c5916b30921c5", "tree": "35e8ce8c5b731a922a53b6d45cc48804f0e5af11", "parents": [ "2d7271501720038381d45fb3dcbe4831228fc8cc" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Aug 07 12:23:57 2019 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Aug 07 12:23:57 2019 -0600" }, "message": "libata: add SG safety checks in SFF pio transfers\n\nAbort processing of a command if we run out of mapped data in the\nSG list. This should never happen, but a previous bug caused it to\nbe possible. Play it safe and attempt to abort nicely if we don\u0027t\nhave more SG segments left.\n\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "2d7271501720038381d45fb3dcbe4831228fc8cc", "tree": "6430450545ca1000e4dac80c2828cbbcd745abef", "parents": [ "e15c2ffa1091c4f72370f01af4de8f9dddeb17a6" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Aug 07 12:20:52 2019 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Aug 07 12:20:52 2019 -0600" }, "message": "libata: have ata_scsi_rw_xlat() fail invalid passthrough requests\n\nFor passthrough requests, libata-scsi takes what the user passes in\nas gospel. This can be problematic if the user fills in the CDB\nincorrectly. One example of that is in request sizes. For read/write\ncommands, the CDB contains fields describing the transfer length of\nthe request. These should match with the SG_IO header fields, but\nlibata-scsi currently does no validation of that.\n\nCheck that the number of blocks in the CDB for passthrough requests\nmatches what was mapped into the request. If the CDB asks for more\ndata then the validated SG_IO header fields, error it.\n\nReported-by: Krishna Ram Prakash R \u003ckrp@gtux.in\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "e15c2ffa1091c4f72370f01af4de8f9dddeb17a6", "tree": "e597c9173fe365533d3990b66a3b794c2f1b9b2d", "parents": [ "db341a049ec7e87053c91008cb452d0bfa6dde72" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Tue Aug 06 13:34:31 2019 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Aug 07 12:19:43 2019 -0600" }, "message": "block: fix O_DIRECT error handling for bio fragments\n\n0eb6ddfb865c tried to fix this up, but introduced a use-after-free\nof dio. Additionally, we still had an issue with error handling,\nas reported by Darrick:\n\n\"I noticed a regression in xfs/747 (an unreleased xfstest for the\nxfs_scrub media scanning feature) on 5.3-rc3. I\u0027ll condense that down\nto a simpler reproducer:\n\nerror-test: 0 209 linear 8:48 0\nerror-test: 209 1 error\nerror-test: 210 6446894 linear 8:48 210\n\nBasically we have a ~3G /dev/sdd and we set up device mapper to fail IO\nfor sector 209 and to pass the io to the scsi device everywhere else.\n\nOn 5.3-rc3, performing a directio pread of this range with a \u003c 1M buffer\n(in other words, a request for fewer than MAX_BIO_PAGES bytes) yields\nEIO like you\u0027d expect:\n\npread64(3, 0x7f880e1c7000, 1048576, 0) \u003d -1 EIO (Input/output error)\npread: Input/output error\n+++ exited with 0 +++\n\nBut doing it with a larger buffer succeeds(!):\n\npread64(3, \"XFSB\\0\\0\\20\\0\\0\\0\\0\\0\\0\\fL\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\\0\"..., 1146880, 0) \u003d 1146880\nread 1146880/1146880 bytes at offset 0\n1 MiB, 1 ops; 0.0009 sec (1.124 GiB/sec and 1052.6316 ops/sec)\n+++ exited with 0 +++\n\n(Note that the part of the buffer corresponding to the dm-error area is\nuninitialized)\n\nOn 5.3-rc2, both commands would fail with EIO like you\u0027d expect. The\nonly change between rc2 and rc3 is commit 0eb6ddfb865c (\"block: Fix\n__blkdev_direct_IO() for bio fragments\").\n\nAFAICT we end up in __blkdev_direct_IO with a 1120K buffer, which gets\nsplit into two bios: one for the first BIO_MAX_PAGES worth of data (1MB)\nand a second one for the 96k after that.\"\n\nFix this by noting that it\u0027s always safe to dereference dio if we get\nBLK_QC_T_EAGAIN returned, as end_io hasn\u0027t been run for that case. So\nwe can safely increment the dio size before calling submit_bio(), and\nthen decrement it on failure (not that it really matters, as the bio\nand dio are going away).\n\nFor error handling, return to the original method of just using \u0027ret\u0027\nfor tracking the error, and the size tracking in dio-\u003esize.\n\nFixes: 0eb6ddfb865c (\"block: Fix __blkdev_direct_IO() for bio fragments\")\nFixes: 6a43074e2f46 (\"block: properly handle IOCB_NOWAIT for async O_DIRECT IO\")\nReported-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "67e7b52d44e3d539dfbfcd866c3d3d69da23a909", "tree": "26281529036a4e28c8c5438884dee0af60288992", "parents": [ "dea1bb35c5f35e0577cfc61f79261d80b8715221" ], "author": { "name": "Trond Myklebust", "email": "trond.myklebust@hammerspace.com", "time": "Wed Aug 07 07:31:27 2019 -0400" }, "committer": { "name": "Trond Myklebust", "email": "trond.myklebust@hammerspace.com", "time": "Wed Aug 07 12:55:11 2019 -0400" }, "message": "NFSv4: Ensure state recovery handles ETIMEDOUT correctly\n\nEnsure that the state recovery code handles ETIMEDOUT correctly,\nand also that we set RPC_TASK_TIMEOUT when recovering open state.\n\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n" }, { "commit": "4b3e30ed3ec7864e798403a63ff2e96bd0c19ab0", "tree": "091041e6db9ce25d2baca2ac4875e58c8a4651a2", "parents": [ "d9dfe768b3f30faa8340cbf34196668714780c3c" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Aug 07 00:23:07 2019 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Aug 07 10:21:38 2019 -0500" }, "message": "Revert \"drm/amdkfd: New IOCTL to allocate queue GWS\"\n\nThis reverts commit 1a058c3376765ee31d65e28cbbb9d4ff15120056.\n\nThis interface is still in too much flux. Revert until\nit\u0027s sorted out.\n\nAcked-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "c02f77d32d2c45cfb1b2bb99eabd8a78f5ecc7db", "tree": "583760d003a852b0607fb20f0ce6d6967ca8a3f8", "parents": [ "3d92aa45fbfd7319e3a19f4ec59fd32b3862b723" ], "author": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 17:31:48 2019 +0200" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Wed Aug 07 13:47:46 2019 +0200" }, "message": "ALSA: hda - Workaround for crackled sound on AMD controller (1022:1457)\n\nA long-time problem on the recent AMD chip (X370, X470, B450, etc with\nPCI ID 1022:1457) with Realtek codecs is the crackled or distorted\nsound for capture streams, as well as occasional playback hiccups.\nAfter lengthy debugging sessions, the workarounds we\u0027ve found are like\nthe following:\n\n- Set up the proper driver caps for this controller, similar as the\n other AMD controller.\n\n- Correct the DMA position reporting with the fixed FIFO size, which\n is similar like as workaround used for VIA chip set.\n\n- Even after the position correction, PulseAudio still shows\n mysterious stalls of playback streams when a capture is triggered in\n timer-scheduled mode. Since we have no clear way to eliminate the\n stall, pass the BATCH PCM flag for PA to suppress the tsched mode as\n a temporary workaround.\n\nThis patch implements the workarounds. For the driver caps, it\ndefines a new preset, AXZ_DCAPS_PRESET_AMD_SB. It enables the FIFO-\ncorrected position reporting (corresponding to the new position_fix\u003d6)\nand enforces the SNDRV_PCM_INFO_BATCH flag.\n\nNote that the current implementation is merely a workaround.\nHopefully we\u0027ll find a better alternative in future, especially about\nremoving the BATCH flag hack again.\n\nBugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d195303\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" }, { "commit": "0617bdede5114a0002298b12cd0ca2b0cfd0395d", "tree": "11a76457d1711325d562793da8ffebe48d65c7bb", "parents": [ "e21a712a9685488f5ce80495b37b9fdbe96c230d" ], "author": { "name": "Mika Westerberg", "email": "mika.westerberg@linux.intel.com", "time": "Wed Aug 07 13:57:18 2019 +0300" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Aug 07 13:06:42 2019 +0200" }, "message": "Revert \"PCI: Add missing link delays required by the PCIe spec\"\n\nCommit c2bf1fc212f7 (\"PCI: Add missing link delays required by the PCIe\nspec\") turned out causing issues with some systems either by making them\nunresponsive or slowing down runtime and system wide resume of PCIe\ndevices. While root cause for the unresponsiveness is still under\ninvestigation given the amount of issues reported better to revert it\nfor now.\n\nBugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204413\nLink: https://lore.kernel.org/linux-pci/SL2P216MB01878BBCD75F21D882AEEA2880C60@SL2P216MB0187.KORP216.PROD.OUTLOOK.COM/\nLink: https://lore.kernel.org/linux-pci/2857501d-c167-547d-c57d-d5d24ea1f1dc@molgen.mpg.de/\nReported-by: Matthias Andree \u003cmatthias.andree@gmx.de\u003e\nReported-by: Paul Menzel \u003cpmenzel@molgen.mpg.de\u003e\nReported-by: Nicholas Johnson \u003cnicholas.johnson-opensource@outlook.com.au\u003e\nSigned-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "3d92aa45fbfd7319e3a19f4ec59fd32b3862b723", "tree": "a185bad2ee4923bb7ec865bd78d4fa032ade10f3", "parents": [ "c1c6c877b0c79fd7e05c931435aa42211eaeebaf" ], "author": { "name": "Wenwen Wang", "email": "wenwen@cs.uga.edu", "time": "Wed Aug 07 04:08:51 2019 -0500" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Wed Aug 07 12:20:00 2019 +0200" }, "message": "ALSA: hiface: fix multiple memory leak bugs\n\nIn hiface_pcm_init(), \u0027rt\u0027 is firstly allocated through kzalloc(). Later\non, hiface_pcm_init_urb() is invoked to initialize \u0027rt-\u003eout_urbs[i]\u0027. In\nhiface_pcm_init_urb(), \u0027rt-\u003eout_urbs[i].buffer\u0027 is allocated through\nkzalloc(). However, if hiface_pcm_init_urb() fails, both \u0027rt\u0027 and\n\u0027rt-\u003eout_urbs[i].buffer\u0027 are not deallocated, leading to memory leak bugs.\nAlso, \u0027rt-\u003eout_urbs[i].buffer\u0027 is not deallocated if snd_pcm_new() fails.\n\nTo fix the above issues, free \u0027rt\u0027 and \u0027rt-\u003eout_urbs[i].buffer\u0027.\n\nFixes: a91c3fb2f842 (\"Add M2Tech hiFace USB-SPDIF driver\")\nSigned-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" }, { "commit": "d9dfe768b3f30faa8340cbf34196668714780c3c", "tree": "1d0c7088930a4495c372d906df221255ea061d4a", "parents": [ "72cda9bb5e219aea0f2f62f56ae05198c59022a7" ], "author": { "name": "Marek Olšák", "email": "marek.olsak@amd.com", "time": "Fri Aug 02 17:44:06 2019 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Aug 06 23:28:41 2019 -0500" }, "message": "Revert \"drm/amdgpu: fix transform feedback GDS hang on gfx10 (v2)\"\n\nThis reverts commit 9ed2c993d723129f85101e51b2ccc36ef5400a67.\n\nSET_CONFIG_REG writes to memory if register shadowing is enabled,\ncausing a VM fault.\n\nNGG streamout is unstable anyway, so all UMDs should use legacy\nstreamout. I think Mesa is the only driver using NGG streamout.\n\nSigned-off-by: Marek Olšák \u003cmarek.olsak@amd.com\u003e\nReviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "33920f1ec5bf47c5c0a1d2113989bdd9dfb3fae9", "tree": "dade12a1bfba22613acd3fc63cb3ce62110e779a", "parents": [ "76d7961ff4ee02cc70365600a52fb59ca544dc7c", "feac1d680233a48603213d52230f92222462a1c8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 17:11:59 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 17:11:59 2019 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n \"Yeah I should have sent a pull request last week, so there is a lot\n more here than usual:\n\n 1) Fix memory leak in ebtables compat code, from Wenwen Wang.\n\n 2) Several kTLS bug fixes from Jakub Kicinski (circular close on\n disconnect etc.)\n\n 3) Force slave speed check on link state recovery in bonding 802.3ad\n mode, from Thomas Falcon.\n\n 4) Clear RX descriptor bits before assigning buffers to them in\n stmmac, from Jose Abreu.\n\n 5) Several missing of_node_put() calls, mostly wrt. for_each_*() OF\n loops, from Nishka Dasgupta.\n\n 6) Double kfree_skb() in peak_usb can driver, from Stephane Grosjean.\n\n 7) Need to hold sock across skb-\u003edestructor invocation, from Cong\n Wang.\n\n 8) IP header length needs to be validated in ipip tunnel xmit, from\n Haishuang Yan.\n\n 9) Use after free in ip6 tunnel driver, also from Haishuang Yan.\n\n 10) Do not use MSI interrupts on r8169 chips before RTL8168d, from\n Heiner Kallweit.\n\n 11) Upon bridge device init failure, we need to delete the local fdb.\n From Nikolay Aleksandrov.\n\n 12) Handle erros from of_get_mac_address() properly in stmmac, from\n Martin Blumenstingl.\n\n 13) Handle concurrent rename vs. dump in netfilter ipset, from Jozsef\n Kadlecsik.\n\n 14) Setting NETIF_F_LLTX on mac80211 causes complete breakage with\n some devices, so revert. From Johannes Berg.\n\n 15) Fix deadlock in rxrpc, from David Howells.\n\n 16) Fix Kconfig deps of enetc driver, we must have PHYLIB. From Yue\n Haibing.\n\n 17) Fix mvpp2 crash on module removal, from Matteo Croce.\n\n 18) Fix race in genphy_update_link, from Heiner Kallweit.\n\n 19) bpf_xdp_adjust_head() stopped working with generic XDP when we\n fixes generic XDP to support stacked devices properly, fix from\n Jesper Dangaard Brouer.\n\n 20) Unbalanced RCU locking in rt6_update_exception_stamp_rt(), from\n David Ahern.\n\n 21) Several memory leaks in new sja1105 driver, from Vladimir Oltean\"\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (214 commits)\n net: dsa: sja1105: Fix memory leak on meta state machine error path\n net: dsa: sja1105: Fix memory leak on meta state machine normal path\n net: dsa: sja1105: Really fix panic on unregistering PTP clock\n net: dsa: sja1105: Use the LOCKEDS bit for SJA1105 E/T as well\n net: dsa: sja1105: Fix broken learning with vlan_filtering disabled\n net: dsa: qca8k: Add of_node_put() in qca8k_setup_mdio_bus()\n net: sched: sample: allow accessing psample_group with rtnl\n net: sched: police: allow accessing police-\u003eparams with rtnl\n net: hisilicon: Fix dma_map_single failed on arm64\n net: hisilicon: fix hip04-xmit never return TX_BUSY\n net: hisilicon: make hip04_tx_reclaim non-reentrant\n tc-testing: updated vlan action tests with batch create/delete\n net sched: update vlan action for batched events operations\n net: stmmac: tc: Do not return a fragment entry\n net: stmmac: Fix issues when number of Queues \u003e\u003d 4\n net: stmmac: xgmac: Fix XGMAC selftests\n be2net: disable bh with spin_lock in be_process_mcc\n net: cxgb3_main: Fix a resource leak in a error path in \u0027init_one()\u0027\n net: ethernet: sun4i-emac: Support phy-handle property for finding PHYs\n net: bridge: move default pvid init/deinit to NETDEV_REGISTER/UNREGISTER\n ...\n" }, { "commit": "feac1d680233a48603213d52230f92222462a1c8", "tree": "69c0a85f3fd18fec875563b529f9db0bcc4b99d3", "parents": [ "f26e0cca14c9494c863d8fa6825b10bd12dc9eaa", "93fa8587b25356382a39f1ca3a81d6c1b42ac731" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "message": "Merge branch \u0027sja1105-fixes\u0027\n\nVladimir Oltean says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nFixes for SJA1105 DSA: FDBs, Learning and PTP\n\nThis is an assortment of functional fixes for the sja1105 switch driver\ntargeted for the \"net\" tree (although they apply on net-next just as\nwell).\n\nPatch 1/5 (\"net: dsa: sja1105: Fix broken learning with vlan_filtering\ndisabled\") repairs a breakage introduced in the early development stages\nof the driver: support for traffic from the CPU has broken \"normal\"\nframe forwarding (based on DMAC) - there is connectivity through the\nswitch only because all frames are flooded.\nI debated whether this patch qualifies as a fix, since it puts the\nswitch into a mode it has never operated in before (aka SVL). But\n\"normal\" forwarding did use to work before the \"Traffic support for\nSJA1105 DSA driver\" patchset, and arguably this patch should have been\npart of that.\nAlso, it would be strange for this feature to be broken in the 5.2 LTS.\n\nPatch 2/5 (\"net: dsa: sja1105: Use the LOCKEDS bit for SJA1105 E/T as\nwell\") is a simplification of a previous FDB-related patch that is\ncurrently in the 5.3 rc\u0027s.\n\nPatches 3/5 - 5/5 fix various crashes found while running linuxptp over the\nswitch ports for extended periods of time, or in conjunction with other\nerror conditions. The fixed-up commits were all introduced in 5.2.\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": "93fa8587b25356382a39f1ca3a81d6c1b42ac731", "tree": "69c0a85f3fd18fec875563b529f9db0bcc4b99d3", "parents": [ "f163fed2764e66511fb5c489bf87e532ad7606fb" ], "author": { "name": "Vladimir Oltean", "email": "olteanv@gmail.com", "time": "Mon Aug 05 01:38:48 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "message": "net: dsa: sja1105: Fix memory leak on meta state machine error path\n\nWhen RX timestamping is enabled and two link-local (non-meta) frames are\nreceived in a row, this constitutes an error.\n\nThe tagger is always caching the last link-local frame, in an attempt to\nmerge it with the meta follow-up frame when that arrives. To recover\nfrom the above error condition, the initial cached link-local frame is\ndropped and the second frame in a row is cached (in expectance of the\nsecond meta frame).\n\nHowever, when dropping the initial link-local frame, its backing memory\nwas being leaked.\n\nFixes: f3097be21bf1 (\"net: dsa: sja1105: Add a state machine for RX timestamping\")\nSigned-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f163fed2764e66511fb5c489bf87e532ad7606fb", "tree": "9d3a460259bbdfcb8ed6a52af57ccb330fb73a5c", "parents": [ "6cb0abbdf90c180e1310976c47399f57477e0e53" ], "author": { "name": "Vladimir Oltean", "email": "olteanv@gmail.com", "time": "Mon Aug 05 01:38:47 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "message": "net: dsa: sja1105: Fix memory leak on meta state machine normal path\n\nAfter a meta frame is received, it is associated with the cached\nsp-\u003edata-\u003estampable_skb from the DSA tagger private structure.\n\nCached means its refcount is incremented with skb_get() in order for\ndsa_switch_rcv() to not free it when the tagger .rcv returns NULL.\n\nThe mistake is that skb_unref() is not the correct function to use. It\nwill correctly decrement the refcount (which will go back to zero) but\nthe skb memory will not be freed. That is the job of kfree_skb(), which\nalso calls skb_unref().\n\nBut it turns out that freeing the cached stampable_skb is in fact not\nnecessary. It is still a perfectly valid skb, and now it is even\nannotated with the partial RX timestamp. So remove the skb_copy()\naltogether and simply pass the stampable_skb with a refcount of 1\n(incremented by us, decremented by dsa_switch_rcv) up the stack.\n\nFixes: f3097be21bf1 (\"net: dsa: sja1105: Add a state machine for RX timestamping\")\nSigned-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6cb0abbdf90c180e1310976c47399f57477e0e53", "tree": "b85bb0da5c7eb10794feeb1fe9ed74da472ba986", "parents": [ "4b7da3d808f91cdad3e34059cd68ba3dfe4c3695" ], "author": { "name": "Vladimir Oltean", "email": "olteanv@gmail.com", "time": "Mon Aug 05 01:38:46 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "message": "net: dsa: sja1105: Really fix panic on unregistering PTP clock\n\nThe IS_ERR_OR_NULL(priv-\u003eclock) check inside\nsja1105_ptp_clock_unregister() is preventing cancel_delayed_work_sync\nfrom actually being run.\n\nAdditionally, sja1105_ptp_clock_unregister() does not actually get run,\nwhen placed in sja1105_remove(). The DSA switch gets torn down, but the\nsja1105 module does not get unregistered. So sja1105_ptp_clock_unregister\nneeds to be moved to sja1105_teardown, to be symmetrical with\nsja1105_ptp_clock_register which is called from the DSA sja1105_setup.\n\nIt is strange to fix a \"fixes\" patch, but the probe failure can only be\nseen when the attached PHY does not respond to MDIO (issue which I can\u0027t\npinpoint the reason to) and it goes away after I power-cycle the board.\nThis time the patch was validated on a failing board, and the kernel\npanic from the fixed commit\u0027s message can no longer be seen.\n\nFixes: 29dd908d355f (\"net: dsa: sja1105: Cancel PTP delayed work on unregister\")\nSigned-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4b7da3d808f91cdad3e34059cd68ba3dfe4c3695", "tree": "280665825ed865f8e84500da867202beb3e189ef", "parents": [ "6d7c7d948a2e9f87b4e7726dee94c59300e1786b" ], "author": { "name": "Vladimir Oltean", "email": "olteanv@gmail.com", "time": "Mon Aug 05 01:38:45 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "message": "net: dsa: sja1105: Use the LOCKEDS bit for SJA1105 E/T as well\n\nIt looks like the FDB dump taken from first-generation switches also\ncontains information on whether entries are static or not. So use that\ninstead of searching through the driver\u0027s tables.\n\nFixes: d763778224ea (\"net: dsa: sja1105: Implement is_static for FDB entries on E/T\")\nSigned-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6d7c7d948a2e9f87b4e7726dee94c59300e1786b", "tree": "9ebbe5d9c4c88143dd95cb74cc12f9916763a833", "parents": [ "f26e0cca14c9494c863d8fa6825b10bd12dc9eaa" ], "author": { "name": "Vladimir Oltean", "email": "olteanv@gmail.com", "time": "Mon Aug 05 01:38:44 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:37:02 2019 -0700" }, "message": "net: dsa: sja1105: Fix broken learning with vlan_filtering disabled\n\nWhen put under a bridge with vlan_filtering 0, the SJA1105 ports will\nflood all traffic as if learning was broken. This is because learning\ninterferes with the rx_vid\u0027s configured by dsa_8021q as unique pvid\u0027s.\n\nSo learning technically still *does* work, it\u0027s just that the learnt\nentries never get matched due to their unique VLAN ID.\n\nThe setting that saves the day is Shared VLAN Learning, which on this\nswitch family works exactly as desired: VLAN tagging still works\n(untagged traffic gets the correct pvid) and FDB entries are still\npopulated with the correct contents including VID. Also, a frame cannot\nviolate the forwarding domain restrictions enforced by its classified\nVLAN. It is just that the VID is ignored when looking up the FDB for\ntaking a forwarding decision (selecting the egress port).\n\nThis patch activates SVL, and the result is that frames with a learnt\nDMAC are no longer flooded in the scenario described above.\n\nNow exactly *because* SVL works as desired, we have to revisit some\nearlier patches:\n\n- It is no longer necessary to manipulate the VID of the \u0027bridge fdb\n {add,del}\u0027 command when vlan_filtering is off. This is because now,\n SVL is enabled for that case, so the actual VID does not matter*.\n\n- It is still desirable to hide dsa_8021q VID\u0027s in the FDB dump\n callback. But right now the dump callback should no longer hide\n duplicates (one per each front panel port\u0027s pvid, plus one for the\n VLAN that the CPU port is going to tag a TX frame with), because there\n shouldn\u0027t be any (the switch will match a single FDB entry no matter\n its VID anyway).\n\n* Not really... It\u0027s no longer necessary to transform a \u0027bridge fdb add\u0027\n into 5 fdb add operations, but the user might still add a fdb entry with\n any vid, and all of them would appear as duplicates in \u0027bridge fdb\n show\u0027. So force a \u0027bridge fdb add\u0027 to insert the VID of 0**, so that we\n can prune the duplicates at insertion time.\n\n** The VID of 0 is better than 1 because it is always guaranteed to be\n in the ports\u0027 hardware filter. DSA also avoids putting the VID inside\n the netlink response message towards the bridge driver when we return\n this particular VID, which makes it suitable for FDB entries learnt\n with vlan_filtering off.\n\nFixes: 227d07a07ef1 (\"net: dsa: sja1105: Add support for traffic through standalone ports\")\nSigned-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\nSigned-off-by: Georg Waibel \u003cgeorg.waibel@sensor-technik.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f26e0cca14c9494c863d8fa6825b10bd12dc9eaa", "tree": "eec432e5613dcf4d48a61c067f716bee990cd67e", "parents": [ "443bfb4acb83a6f0b7d9b11ac32f17c67f14e995" ], "author": { "name": "Nishka Dasgupta", "email": "nishkadg.linux@gmail.com", "time": "Sun Aug 04 21:00:18 2019 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:35:13 2019 -0700" }, "message": "net: dsa: qca8k: Add of_node_put() in qca8k_setup_mdio_bus()\n\nEach iteration of for_each_available_child_of_node() puts the previous\nnode, but in the case of a return from the middle of the loop, there\nis no put, thus causing a memory leak. Hence add an of_node_put() before\nthe return.\nAdditionally, the local variable ports in the function\nqca8k_setup_mdio_bus() takes the return value of of_get_child_by_name(),\nwhich gets a node but does not put it. If the function returns without\nputting ports, it may cause a memory leak. Hence put ports before the\nmid-loop return statement, and also outside the loop after its last usage\nin this function.\nIssues found with Coccinelle.\n\nSigned-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "443bfb4acb83a6f0b7d9b11ac32f17c67f14e995", "tree": "a070b8dedee983ee82415fe636b0ca7d435c4d77", "parents": [ "2b0dfc1753332d81d8e6654d682d8740aed4dec9", "67cbf7dedd03a63ca2fbd9df2049eabba7a37edf" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:15:39 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:15:39 2019 -0700" }, "message": "Merge branch \u0027flow_offload-action-fixes\u0027\n\nVlad Buslov says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\naction fixes for flow_offload infra compatibility\n\nFix rcu warnings due to usage of action helpers that expect rcu read lock\nprotection from rtnl-protected context of flow_offload infra.\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": "67cbf7dedd03a63ca2fbd9df2049eabba7a37edf", "tree": "a070b8dedee983ee82415fe636b0ca7d435c4d77", "parents": [ "c4bd48699beb92d6bb99d6139d1e9737cca73480" ], "author": { "name": "Vlad Buslov", "email": "vladbu@mellanox.com", "time": "Sat Aug 03 16:36:19 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:15:39 2019 -0700" }, "message": "net: sched: sample: allow accessing psample_group with rtnl\n\nRecently implemented support for sample action in flow_offload infra leads\nto following rcu usage warning:\n\n[ 1938.234856] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ 1938.234858] WARNING: suspicious RCU usage\n[ 1938.234863] 5.3.0-rc1+ #574 Not tainted\n[ 1938.234866] -----------------------------\n[ 1938.234869] include/net/tc_act/tc_sample.h:47 suspicious rcu_dereference_check() usage!\n[ 1938.234872]\n other info that might help us debug this:\n\n[ 1938.234875]\n rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n[ 1938.234879] 1 lock held by tc/19540:\n[ 1938.234881] #0: 00000000b03cb918 (rtnl_mutex){+.+.}, at: tc_new_tfilter+0x47c/0x970\n[ 1938.234900]\n stack backtrace:\n[ 1938.234905] CPU: 2 PID: 19540 Comm: tc Not tainted 5.3.0-rc1+ #574\n[ 1938.234908] Hardware name: Supermicro SYS-2028TP-DECR/X10DRT-P, BIOS 2.0b 03/30/2017\n[ 1938.234911] Call Trace:\n[ 1938.234922] dump_stack+0x85/0xc0\n[ 1938.234930] tc_setup_flow_action+0xed5/0x2040\n[ 1938.234944] fl_hw_replace_filter+0x11f/0x2e0 [cls_flower]\n[ 1938.234965] fl_change+0xd24/0x1b30 [cls_flower]\n[ 1938.234990] tc_new_tfilter+0x3e0/0x970\n[ 1938.235021] ? tc_del_tfilter+0x720/0x720\n[ 1938.235028] rtnetlink_rcv_msg+0x389/0x4b0\n[ 1938.235038] ? netlink_deliver_tap+0x95/0x400\n[ 1938.235044] ? rtnl_dellink+0x2d0/0x2d0\n[ 1938.235053] netlink_rcv_skb+0x49/0x110\n[ 1938.235063] netlink_unicast+0x171/0x200\n[ 1938.235073] netlink_sendmsg+0x224/0x3f0\n[ 1938.235091] sock_sendmsg+0x5e/0x60\n[ 1938.235097] ___sys_sendmsg+0x2ae/0x330\n[ 1938.235111] ? __handle_mm_fault+0x12cd/0x19e0\n[ 1938.235125] ? __handle_mm_fault+0x12cd/0x19e0\n[ 1938.235138] ? find_held_lock+0x2b/0x80\n[ 1938.235147] ? do_user_addr_fault+0x22d/0x490\n[ 1938.235160] __sys_sendmsg+0x59/0xa0\n[ 1938.235178] do_syscall_64+0x5c/0xb0\n[ 1938.235187] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n[ 1938.235192] RIP: 0033:0x7ff9a4d597b8\n[ 1938.235197] Code: 89 02 48 c7 c0 ff ff ff ff eb bb 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 8d 05 65 8f 0c 00 8b 00 85 c0 75 17 b8 2e 00 00 00 0f 05 \u003c48\u003e 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83\n ec 28 89 54\n[ 1938.235200] RSP: 002b:00007ffcfe381c48 EFLAGS: 00000246 ORIG_RAX: 000000000000002e\n[ 1938.235205] RAX: ffffffffffffffda RBX: 000000005d4497f9 RCX: 00007ff9a4d597b8\n[ 1938.235208] RDX: 0000000000000000 RSI: 00007ffcfe381cb0 RDI: 0000000000000003\n[ 1938.235211] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000006\n[ 1938.235214] R10: 0000000000404ec2 R11: 0000000000000246 R12: 0000000000000001\n[ 1938.235217] R13: 0000000000480640 R14: 0000000000000012 R15: 0000000000000001\n\nChange tcf_sample_psample_group() helper to allow using it from both rtnl\nand rcu protected contexts.\n\nFixes: a7a7be6087b0 (\"net/sched: add sample action to the hardware intermediate representation\")\nSigned-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nReviewed-by: Pieter Jansen van Vuuren \u003cpieter.jansenvanvuuren@netronome.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c4bd48699beb92d6bb99d6139d1e9737cca73480", "tree": "0b1b097b96b1029a45a992dede7e13087e438fd3", "parents": [ "2b0dfc1753332d81d8e6654d682d8740aed4dec9" ], "author": { "name": "Vlad Buslov", "email": "vladbu@mellanox.com", "time": "Sat Aug 03 16:36:18 2019 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:15:39 2019 -0700" }, "message": "net: sched: police: allow accessing police-\u003eparams with rtnl\n\nRecently implemented support for police action in flow_offload infra leads\nto following rcu usage warning:\n\n[ 1925.881092] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ 1925.881094] WARNING: suspicious RCU usage\n[ 1925.881098] 5.3.0-rc1+ #574 Not tainted\n[ 1925.881100] -----------------------------\n[ 1925.881104] include/net/tc_act/tc_police.h:57 suspicious rcu_dereference_check() usage!\n[ 1925.881106]\n other info that might help us debug this:\n\n[ 1925.881109]\n rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n[ 1925.881112] 1 lock held by tc/18591:\n[ 1925.881115] #0: 00000000b03cb918 (rtnl_mutex){+.+.}, at: tc_new_tfilter+0x47c/0x970\n[ 1925.881124]\n stack backtrace:\n[ 1925.881127] CPU: 2 PID: 18591 Comm: tc Not tainted 5.3.0-rc1+ #574\n[ 1925.881130] Hardware name: Supermicro SYS-2028TP-DECR/X10DRT-P, BIOS 2.0b 03/30/2017\n[ 1925.881132] Call Trace:\n[ 1925.881138] dump_stack+0x85/0xc0\n[ 1925.881145] tc_setup_flow_action+0x1771/0x2040\n[ 1925.881155] fl_hw_replace_filter+0x11f/0x2e0 [cls_flower]\n[ 1925.881175] fl_change+0xd24/0x1b30 [cls_flower]\n[ 1925.881200] tc_new_tfilter+0x3e0/0x970\n[ 1925.881231] ? tc_del_tfilter+0x720/0x720\n[ 1925.881243] rtnetlink_rcv_msg+0x389/0x4b0\n[ 1925.881250] ? netlink_deliver_tap+0x95/0x400\n[ 1925.881257] ? rtnl_dellink+0x2d0/0x2d0\n[ 1925.881264] netlink_rcv_skb+0x49/0x110\n[ 1925.881275] netlink_unicast+0x171/0x200\n[ 1925.881284] netlink_sendmsg+0x224/0x3f0\n[ 1925.881299] sock_sendmsg+0x5e/0x60\n[ 1925.881305] ___sys_sendmsg+0x2ae/0x330\n[ 1925.881309] ? task_work_add+0x43/0x50\n[ 1925.881314] ? fput_many+0x45/0x80\n[ 1925.881329] ? __lock_acquire+0x248/0x1930\n[ 1925.881342] ? find_held_lock+0x2b/0x80\n[ 1925.881347] ? task_work_run+0x7b/0xd0\n[ 1925.881359] __sys_sendmsg+0x59/0xa0\n[ 1925.881375] do_syscall_64+0x5c/0xb0\n[ 1925.881381] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n[ 1925.881384] RIP: 0033:0x7feb245047b8\n[ 1925.881388] Code: 89 02 48 c7 c0 ff ff ff ff eb bb 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 8d 05 65 8f 0c 00 8b 00 85 c0 75 17 b8 2e 00 00 00 0f 05 \u003c48\u003e 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83\n ec 28 89 54\n[ 1925.881391] RSP: 002b:00007ffc2d2a5788 EFLAGS: 00000246 ORIG_RAX: 000000000000002e\n[ 1925.881395] RAX: ffffffffffffffda RBX: 000000005d4497ed RCX: 00007feb245047b8\n[ 1925.881398] RDX: 0000000000000000 RSI: 00007ffc2d2a57f0 RDI: 0000000000000003\n[ 1925.881400] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000006\n[ 1925.881403] R10: 0000000000404ec2 R11: 0000000000000246 R12: 0000000000000001\n[ 1925.881406] R13: 0000000000480640 R14: 0000000000000012 R15: 0000000000000001\n\nChange tcf_police_rate_bytes_ps() and tcf_police_tcfp_burst() helpers to\nallow using them from both rtnl and rcu protected contexts.\n\nFixes: 8c8cfc6ed274 (\"net/sched: add police action to the hardware intermediate representation\")\nSigned-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nReviewed-by: Pieter Jansen van Vuuren \u003cpieter.jansenvanvuuren@netronome.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2b0dfc1753332d81d8e6654d682d8740aed4dec9", "tree": "21714b958e1c9786895f5dc70821c45997efb467", "parents": [ "5b0bce247cc2423defedf289f8b84e22385426ee", "96a50c0d907ac8f5c3d6b051031a19eb8a2b53e3" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:14:01 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:14:01 2019 -0700" }, "message": "Merge branch \u0027hisilicon-fixes\u0027\n\nJiangfeng Xiao says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: hisilicon: Fix a few problems with hip04_eth\n\nDuring the use of the hip04_eth driver,\nseveral problems were found,\nwhich solved the hip04_tx_reclaim reentry problem,\nfixed the problem that hip04_mac_start_xmit never\nreturns NETDEV_TX_BUSY\nand the dma_map_single failed on the arm64 platform.\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": "96a50c0d907ac8f5c3d6b051031a19eb8a2b53e3", "tree": "21714b958e1c9786895f5dc70821c45997efb467", "parents": [ "f2243b82785942be519016067ee6c55a063bbfe2" ], "author": { "name": "Jiangfeng Xiao", "email": "xiaojiangfeng@huawei.com", "time": "Sat Aug 03 20:31:41 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:14:01 2019 -0700" }, "message": "net: hisilicon: Fix dma_map_single failed on arm64\n\nOn the arm64 platform, executing \"ifconfig eth0 up\" will fail,\nreturning \"ifconfig: SIOCSIFFLAGS: Input/output error.\"\n\nndev-\u003edev is not initialized, dma_map_single-\u003eget_dma_ops-\u003e\ndummy_dma_ops-\u003e__dummy_map_page will return DMA_ERROR_CODE\ndirectly, so when we use dma_map_single, the first parameter\nis to use the device of platform_device.\n\nSigned-off-by: Jiangfeng Xiao \u003cxiaojiangfeng@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f2243b82785942be519016067ee6c55a063bbfe2", "tree": "46c815f3b8cea0148dfd1caab9a6a47550c38d87", "parents": [ "1a2c070ae805910a853b4a14818481ed2e17c727" ], "author": { "name": "Jiangfeng Xiao", "email": "xiaojiangfeng@huawei.com", "time": "Sat Aug 03 20:31:40 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:14:01 2019 -0700" }, "message": "net: hisilicon: fix hip04-xmit never return TX_BUSY\n\nTX_DESC_NUM is 256, in tx_count, the maximum value of\nmod(TX_DESC_NUM - 1) is 254, the variable \"count\" in\nthe hip04_mac_start_xmit function is never equal to\n(TX_DESC_NUM - 1), so hip04_mac_start_xmit never\nreturn NETDEV_TX_BUSY.\n\ntx_count is modified to mod(TX_DESC_NUM) so that\nthe maximum value of tx_count can reach\n(TX_DESC_NUM - 1), then hip04_mac_start_xmit can reurn\nNETDEV_TX_BUSY.\n\nSigned-off-by: Jiangfeng Xiao \u003cxiaojiangfeng@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1a2c070ae805910a853b4a14818481ed2e17c727", "tree": "b2d1a9d138ec4cecb29bddb42d3a7e98fa3cf2b6", "parents": [ "5b0bce247cc2423defedf289f8b84e22385426ee" ], "author": { "name": "Jiangfeng Xiao", "email": "xiaojiangfeng@huawei.com", "time": "Sat Aug 03 20:31:39 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:14:01 2019 -0700" }, "message": "net: hisilicon: make hip04_tx_reclaim non-reentrant\n\nIf hip04_tx_reclaim is interrupted while it is running\nand then __napi_schedule continues to execute\nhip04_rx_poll-\u003ehip04_tx_reclaim, reentrancy occurs\nand oops is generated. So you need to mask the interrupt\nduring the hip04_tx_reclaim run.\n\nThe kernel oops exception stack is as follows:\n\nUnable to handle kernel NULL pointer dereference\nat virtual address 00000050\npgd \u003d c0003000\n[00000050] *pgd\u003d80000000a04003, *pmd\u003d00000000\nInternal error: Oops: 206 [#1] SMP ARM\nModules linked in: hip04_eth mtdblock mtd_blkdevs mtd\nohci_platform ehci_platform ohci_hcd ehci_hcd\nvfat fat sd_mod usb_storage scsi_mod usbcore usb_common\nCPU: 0 PID: 0 Comm: swapper/0 Tainted: G O 4.4.185 #1\nHardware name: Hisilicon A15\ntask: c0a250e0 task.stack: c0a00000\nPC is at hip04_tx_reclaim+0xe0/0x17c [hip04_eth]\nLR is at hip04_tx_reclaim+0x30/0x17c [hip04_eth]\npc : [\u003cbf30c3a4\u003e] lr : [\u003cbf30c2f4\u003e] psr: 600e0313\nsp : c0a01d88 ip : 00000000 fp : c0601f9c\nr10: 00000000 r9 : c3482380 r8 : 00000001\nr7 : 00000000 r6 : 000000e1 r5 : c3482000 r4 : 0000000c\nr3 : f2209800 r2 : 00000000 r1 : 00000000 r0 : 00000000\nFlags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel\nControl: 32c5387d Table: 03d28c80 DAC: 55555555\nProcess swapper/0 (pid: 0, stack limit \u003d 0xc0a00190)\nStack: (0xc0a01d88 to 0xc0a02000)\n[\u003cbf30c3a4\u003e] (hip04_tx_reclaim [hip04_eth]) from [\u003cbf30d2e0\u003e]\n (hip04_rx_poll+0x88/0x368 [hip04_eth])\n[\u003cbf30d2e0\u003e] (hip04_rx_poll [hip04_eth]) from [\u003cc04c2d9c\u003e] (net_rx_action+0x114/0x34c)\n[\u003cc04c2d9c\u003e] (net_rx_action) from [\u003cc021eed8\u003e] (__do_softirq+0x218/0x318)\n[\u003cc021eed8\u003e] (__do_softirq) from [\u003cc021f284\u003e] (irq_exit+0x88/0xac)\n[\u003cc021f284\u003e] (irq_exit) from [\u003cc0240090\u003e] (msa_irq_exit+0x11c/0x1d4)\n[\u003cc0240090\u003e] (msa_irq_exit) from [\u003cc02677e0\u003e] (__handle_domain_irq+0x110/0x148)\n[\u003cc02677e0\u003e] (__handle_domain_irq) from [\u003cc0201588\u003e] (gic_handle_irq+0xd4/0x118)\n[\u003cc0201588\u003e] (gic_handle_irq) from [\u003cc0551700\u003e] (__irq_svc+0x40/0x58)\nException stack(0xc0a01f30 to 0xc0a01f78)\n1f20: c0ae8b40 00000000 00000000 00000000\n1f40: 00000002 ffffe000 c0601f9c 00000000 ffffffff c0a2257c c0a22440 c0831a38\n1f60: c0a01ec4 c0a01f80 c0203714 c0203718 600e0213 ffffffff\n[\u003cc0551700\u003e] (__irq_svc) from [\u003cc0203718\u003e] (arch_cpu_idle+0x20/0x3c)\n[\u003cc0203718\u003e] (arch_cpu_idle) from [\u003cc025bfd8\u003e] (cpu_startup_entry+0x244/0x29c)\n[\u003cc025bfd8\u003e] (cpu_startup_entry) from [\u003cc054b0d8\u003e] (rest_init+0xc8/0x10c)\n[\u003cc054b0d8\u003e] (rest_init) from [\u003cc0800c58\u003e] (start_kernel+0x468/0x514)\nCode: a40599e5 016086e2 018088e2 7660efe6 (503090e5)\n---[ end trace 1db21d6d09c49d74 ]---\nKernel panic - not syncing: Fatal exception in interrupt\nCPU3: stopping\nCPU: 3 PID: 0 Comm: swapper/3 Tainted: G D O 4.4.185 #1\n\nSigned-off-by: Jiangfeng Xiao \u003cxiaojiangfeng@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5b0bce247cc2423defedf289f8b84e22385426ee", "tree": "36172fae20323f312c3c5a5b10ec2fa310908516", "parents": [ "3abd24a1831d0507e15702184c6319c253145a6c", "8571deb013812f35260b2b7152a522eacfa9ccf9" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:05:40 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:06:26 2019 -0700" }, "message": "Merge branch \u0027Fix-batched-event-generation-for-vlan-action\u0027\n\nRoman Mashak says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nFix batched event generation for vlan action\n\nWhen adding or deleting a batch of entries, the kernel sends up to\nTCA_ACT_MAX_PRIO (defined to 32 in kernel) entries in an event to user\nspace. However it does not consider that the action sizes may vary and\nrequire different skb sizes.\n\nFor example, consider the following script adding 32 entries with all\nsupported vlan parameters (in order to maximize netlink messages size):\n\n% cat tc-batch.sh\nTC\u003d\"sudo /mnt/iproute2.git/tc/tc\"\n\n$TC actions flush action vlan\nfor i in `seq 1 $1`;\ndo\n cmd\u003d\"action vlan push protocol 802.1q id 4094 priority 7 pipe \\\n index $i cookie aabbccddeeff112233445566778800a1 \"\n args\u003d$args$cmd\ndone\n$TC actions add $args\n%\n% ./tc-batch.sh 32\nError: Failed to fill netlink attributes while adding TC action.\nWe have an error talking to the kernel\n%\n\npatch 1 adds callback in tc_action_ops of vlan action, which calculates\nthe action size, and passes size to tcf_add_notify()/tcf_del_notify().\n\npatch 2 updates the TDC test suite with relevant vlan test cases.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nAcked-by: Jamal Hadi Salim \u003cjhs@mojatatu.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8571deb013812f35260b2b7152a522eacfa9ccf9", "tree": "36172fae20323f312c3c5a5b10ec2fa310908516", "parents": [ "b35475c5491a14c8ce7a5046ef7bcda8a860581a" ], "author": { "name": "Roman Mashak", "email": "mrv@mojatatu.com", "time": "Fri Aug 02 15:16:47 2019 -0400" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:05:40 2019 -0700" }, "message": "tc-testing: updated vlan action tests with batch create/delete\n\nUpdate TDC tests with cases varifying ability of TC to install or delete\nbatches of vlan actions.\n\nSigned-off-by: Roman Mashak \u003cmrv@mojatatu.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b35475c5491a14c8ce7a5046ef7bcda8a860581a", "tree": "65514b916e8211821c410283d235677a4770daa5", "parents": [ "3abd24a1831d0507e15702184c6319c253145a6c" ], "author": { "name": "Roman Mashak", "email": "mrv@mojatatu.com", "time": "Fri Aug 02 15:16:46 2019 -0400" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 14:05:39 2019 -0700" }, "message": "net sched: update vlan action for batched events operations\n\nAdd get_fill_size() routine used to calculate the action size\nwhen building a batch of events.\n\nFixes: c7e2b9689 (\"sched: introduce vlan action\")\nSigned-off-by: Roman Mashak \u003cmrv@mojatatu.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "76d7961ff4ee02cc70365600a52fb59ca544dc7c", "tree": "4fc4b1ff1628616a665b025ac1aa07c0b555a22d", "parents": [ "f4eb1423e43376bec578c5696635b074c8bd2035", "74034a09267c1f48d5ce7ae4c4a317fac7d43418" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 14:01:08 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 14:01:08 2019 -0700" }, "message": "Merge tag \u0027mips_fixes_5.3_1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux\n\nPull MIPS fixes from Paul Burton:\n \"A few MIPS fixes for 5.3:\n\n - Various switch fall through annotations to fixup warnings \u0026 errors\n resulting from -Wimplicit-fallthrough.\n\n - A fix for systems (at least jazz) using an i8253 PIT as clocksource\n when it\u0027s not suitably configured.\n\n - Set struct cacheinfo\u0027s cpu_map_populated field to true, indicating\n that we filled in cache info detected from cop0 registers \u0026\n avoiding complaints about that info being (intentionally) missing\n in devicetree\"\n\n* tag \u0027mips_fixes_5.3_1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:\n MIPS: BCM63XX: Mark expected switch fall-through\n MIPS: OProfile: Mark expected switch fall-throughs\n MIPS: Annotate fall-through in Cavium Octeon code\n MIPS: Annotate fall-through in kvm/emulate.c\n mips: fix cacheinfo\n MIPS: kernel: only use i8253 clocksource with periodic clockevent\n" }, { "commit": "72cda9bb5e219aea0f2f62f56ae05198c59022a7", "tree": "e4087457295a20e12d96abceb14b56d247336db3", "parents": [ "b887011803582c8539cd9f963035eef91373a169" ], "author": { "name": "Likun Gao", "email": "Likun.Gao@amd.com", "time": "Fri Aug 02 15:18:57 2019 +0800" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Aug 06 15:04:40 2019 -0500" }, "message": "drm/amdgpu: pin the csb buffer on hw init for gfx v8\n\nWithout this pin, the csb buffer will be filled with inconsistent\ndata after S3 resume. And that will causes gfx hang on gfxoff\nexit since this csb will be executed then.\n\nSigned-off-by: Likun Gao \u003cLikun.Gao@amd.com\u003e\nTested-by: Paul Gover \u003cpmw.gover@yahoo.co.uk\u003e\nReviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\nReviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "3abd24a1831d0507e15702184c6319c253145a6c", "tree": "cd242c0b14b929aef7c683330e09fd64bc552c89", "parents": [ "0574f2ed30609fd1720489a9dcb50a805958c796", "4a6a1385a4db5f42258a40fcd497cbfd22075968" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 12:26:11 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 12:26:11 2019 -0700" }, "message": "Merge branch \u0027stmmac-fixes\u0027\n\nJose Abreu says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: stmmac: Fixes for -net\n\nCouple of fixes for -net. More info in commit log.\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": "4a6a1385a4db5f42258a40fcd497cbfd22075968", "tree": "cd242c0b14b929aef7c683330e09fd64bc552c89", "parents": [ "e8df7e8c233a18d2704e37ecff47583b494789d3" ], "author": { "name": "Jose Abreu", "email": "Jose.Abreu@synopsys.com", "time": "Tue Aug 06 15:16:18 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 12:26:11 2019 -0700" }, "message": "net: stmmac: tc: Do not return a fragment entry\n\nDo not try to return a fragment entry from TC list. Otherwise we may not\nclean properly allocated entries.\n\nSigned-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e8df7e8c233a18d2704e37ecff47583b494789d3", "tree": "e623ad933f709489efdcc5a6c6d740849acf6824", "parents": [ "0efedbf11f07adee555e0c4ba9c6eb58760aa94f" ], "author": { "name": "Jose Abreu", "email": "Jose.Abreu@synopsys.com", "time": "Tue Aug 06 15:16:17 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 12:26:11 2019 -0700" }, "message": "net: stmmac: Fix issues when number of Queues \u003e\u003d 4\n\nWhen queues \u003e\u003d 4 we use different registers but we were not subtracting\nthe offset of 4. Fix this.\n\nFound out by Coverity.\n\nSigned-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0efedbf11f07adee555e0c4ba9c6eb58760aa94f", "tree": "1f7e66b28d783adf02c4a6c881dadeb618c1115f", "parents": [ "0574f2ed30609fd1720489a9dcb50a805958c796" ], "author": { "name": "Jose Abreu", "email": "Jose.Abreu@synopsys.com", "time": "Tue Aug 06 15:16:16 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 12:26:11 2019 -0700" }, "message": "net: stmmac: xgmac: Fix XGMAC selftests\n\nFixup the XGMAC selftests by correctly finishing the implementation of\nset_filter callback.\n\nResult:\n$ ethtool -t enp4s0\nThe test result is PASS\nThe test extra info:\n 1. MAC Loopback \t 0\n 2. PHY Loopback \t -95\n 3. MMC Counters \t -95\n 4. EEE \t -95\n 5. Hash Filter MC \t 0\n 6. Perfect Filter UC \t 0\n 7. MC Filter \t 0\n 8. UC Filter \t 0\n 9. Flow Control \t 0\n\nSigned-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0574f2ed30609fd1720489a9dcb50a805958c796", "tree": "53a68c7f7aa01b27e13c970beeabf008a96e90a5", "parents": [ "d0d006a43e9a7a796f6f178839c92fcc222c564d", "1f66072503316134873060b24b7895dbbcccf00e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 11:48:48 2019 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 11:48:48 2019 -0700" }, "message": "Merge tag \u0027wireless-drivers-for-davem-2019-08-06\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers\n\nKalle Valo says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nwireless-drivers fixes for 5.3\n\nSecond set of fixes for 5.3. Lots of iwlwifi fixes have accumulated\nwhich consists most of patches in this pull request. Only most notable\niwlwifi fixes are listed below.\n\nmwifiex\n\n* fix a regression related to WPA1 networks since v5.3-rc1\n\niwlwifi\n\n* fix use-after-free issues\n\n* fix DMA mapping API usage errors\n\n* fix frame drop occurring due to reorder buffer handling in\n RSS in certain conditions\n\n* fix rate scale locking issues\n\n* disable TX A-MSDU on older NICs as it causes problems and was\n never supposed to be supported\n\n* new PCI IDs\n\n* GEO_TX_POWER_LIMIT API issue that many people were hitting\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": "f4eb1423e43376bec578c5696635b074c8bd2035", "tree": "ac0b55118421f1548fd44ce108355fa8ab23b392", "parents": [ "4368c4bc9d36821690d6bb2e743d5a075b6ddb55", "e0f6974a54d3f7f1b5fdf5a593bd43ce9206ec04" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 11:47:23 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 11:47:23 2019 -0700" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid\n\nPull HID fixes from Jiri Kosina:\n\n - functional regression fix for some of the Logitech unifying devices,\n from Hans de Goede\n\n - race condition fix in hid-sony for bug severely affecting\n Valve/Android deployments, from Roderick Colenbrander\n\n - several fixes for issues found by syzbot/kasan, from Oliver Neukum\n and Hillf Danton\n\n - functional regression fix for Wacom Cintiq device, from Aaron\n Armstrong Skomra\n\n - a few other assorted device-specific quirks\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:\n HID: sony: Fix race condition between rumble and device remove.\n HID: hiddev: do cleanup in failure of opening a device\n HID: hiddev: avoid opening a disconnected device\n HID: input: fix a4tech horizontal wheel custom usage\n HID: Add quirk for HP X1200 PIXART OEM mouse\n HID: holtek: test for sanity of intfdata\n HID: wacom: fix bit shift for Cintiq Companion 2\n HID: quirks: Set the INCREMENT_USAGE_ON_DUPLICATE quirk on Saitek X52\n HID: logitech-dj: Really fix return value of logi_dj_recv_query_hidpp_devices\n HID: Add 044f:b320 ThrustMaster, Inc. 2 in 1 DT\n HID: logitech-dj: add the Powerplay receiver\n HID: logitech-hidpp: add USB PID for a few more supported mice\n HID: logitech-dj: rename \"gaming\" receiver to \"lightspeed\"\n" }, { "commit": "d0d006a43e9a7a796f6f178839c92fcc222c564d", "tree": "f44e29fb12bad719be315c5dbc14ef8d3015448b", "parents": [ "debea2cd3193ac868289e8893c3a719c265b0612" ], "author": { "name": "Denis Kirjanov", "email": "kda@linux-powerpc.org", "time": "Tue Aug 06 12:51:11 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 11:47:19 2019 -0700" }, "message": "be2net: disable bh with spin_lock in be_process_mcc\n\nbe_process_mcc() is invoked in 3 different places and\nalways with BHs disabled except the be_poll function\nbut since it\u0027s invoked from softirq with BHs\ndisabled it won\u0027t hurt.\n\nv1-\u003ev2: added explanation to the patch\nv2-\u003ev3: add a missing call from be_cmds.c\n\nSigned-off-by: Denis Kirjanov \u003ckda@linux-powerpc.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "debea2cd3193ac868289e8893c3a719c265b0612", "tree": "beb750f62b8f14506b9f6e11f474b8f4831c148f", "parents": [ "5c4e2e1af345426f63410a50e2a678673574aa02" ], "author": { "name": "Christophe JAILLET", "email": "christophe.jaillet@wanadoo.fr", "time": "Tue Aug 06 10:55:12 2019 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 11:34:23 2019 -0700" }, "message": "net: cxgb3_main: Fix a resource leak in a error path in \u0027init_one()\u0027\n\nA call to \u0027kfree_skb()\u0027 is missing in the error handling path of\n\u0027init_one()\u0027.\nThis is already present in \u0027remove_one()\u0027 but is missing here.\n\nSigned-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5c4e2e1af345426f63410a50e2a678673574aa02", "tree": "ef580c4faf8b86e65ebf3828ec0d16803d570d70", "parents": [ "f7813d5c637e43209ffdc20a9b25a13fcbb86eef" ], "author": { "name": "Chen-Yu Tsai", "email": "wens@csie.org", "time": "Tue Aug 06 15:35:39 2019 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Aug 06 11:29:16 2019 -0700" }, "message": "net: ethernet: sun4i-emac: Support phy-handle property for finding PHYs\n\nThe sun4i-emac uses the \"phy\" property to find the PHY it\u0027s supposed to\nuse. This property was deprecated in favor of \"phy-handle\" in commit\n8c5b09447625 (\"dt-bindings: net: sun4i-emac: Convert the binding to a\nschemas\").\n\nAdd support for this new property name, and fall back to the old one in\ncase the device tree hasn\u0027t been updated.\n\nSigned-off-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4368c4bc9d36821690d6bb2e743d5a075b6ddb55", "tree": "830266f667f9315e072704282f640464dd80996b", "parents": [ "0eb0ce0a78e1f57082bca6cbdea6fd04feedb876", "4c92057661a3412f547ede95715641d7ee16ddac" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 11:22:22 2019 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Aug 06 11:22:22 2019 -0700" }, "message": "Merge branch \u0027x86/grand-schemozzle\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull pti updates from Thomas Gleixner:\n \"The performance deterioration departement is not proud at all to\n present yet another set of speculation fences to mitigate the next\n chapter in the \u0027what could possibly go wrong\u0027 story.\n\n The new vulnerability belongs to the Spectre class and affects GS\n based data accesses and has therefore been dubbed \u0027Grand Schemozzle\u0027\n for secret communication purposes. It\u0027s officially listed as\n CVE-2019-1125.\n\n Conditional branches in the entry paths which contain a SWAPGS\n instruction (interrupts and exceptions) can be mis-speculated which\n results in speculative accesses with a wrong GS base.\n\n This can happen on entry from user mode through a mis-speculated\n branch which takes the entry from kernel mode path and therefore does\n not execute the SWAPGS instruction. The following speculative accesses\n are done with user GS base.\n\n On entry from kernel mode the mis-speculated branch executes the\n SWAPGS instruction in the entry from user mode path which has the same\n effect that the following GS based accesses are done with user GS\n base.\n\n If there is a disclosure gadget available in these code paths the\n mis-speculated data access can be leaked through the usual side\n channels.\n\n The entry from user mode issue affects all CPUs which have speculative\n execution. The entry from kernel mode issue affects only Intel CPUs\n which can speculate through SWAPGS. On CPUs from other vendors SWAPGS\n has semantics which prevent that.\n\n SMAP migitates both problems but only when the CPU is not affected by\n the Meltdown vulnerability.\n\n The mitigation is to issue LFENCE instructions in the entry from\n kernel mode path for all affected CPUs and on the affected Intel CPUs\n also in the entry from user mode path unless PTI is enabled because\n the CR3 write is serializing.\n\n The fences are as usual enabled conditionally and can be completely\n disabled on the kernel command line. The Spectre V1 documentation is\n updated accordingly.\n\n A big \"Thank You!\" goes to Josh for doing the heavy lifting for this\n round of hardware misfeature \u0027repair\u0027. Of course also \"Thank You!\" to\n everybody else who contributed in one way or the other\"\n\n* \u0027x86/grand-schemozzle\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n Documentation: Add swapgs description to the Spectre v1 documentation\n x86/speculation/swapgs: Exclude ATOMs from speculation through SWAPGS\n x86/entry/64: Use JMP instead of JMPQ\n x86/speculation: Enable Spectre v1 swapgs mitigations\n x86/speculation: Prepare entry code for Spectre v1 swapgs mitigations\n" }, { "commit": "b803974a86039913d5280add083d730b2b9ed8ec", "tree": "53ca0e2396fa7a72254f646c2491e9a5e5f28aee", "parents": [ "fa25eba6993b3750f417baabba169afaba076178" ], "author": { "name": "Kevin Hao", "email": "haokexin@gmail.com", "time": "Fri Jul 26 10:30:49 2019 +0800" }, "committer": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Tue Aug 06 18:59:14 2019 +0200" }, "message": "mmc: cavium: Add the missing dma unmap when the dma has finished.\n\nThis fixes the below calltrace when the CONFIG_DMA_API_DEBUG is enabled.\n DMA-API: thunderx_mmc 0000:01:01.4: cpu touching an active dma mapped cacheline [cln\u003d0x000000002fdf9800]\n WARNING: CPU: 21 PID: 1 at kernel/dma/debug.c:596 debug_dma_assert_idle+0x1f8/0x270\n Modules linked in:\n CPU: 21 PID: 1 Comm: init Not tainted 5.3.0-rc1-next-20190725-yocto-standard+ #64\n Hardware name: Marvell OcteonTX CN96XX board (DT)\n pstate: 80400009 (Nzcv daif +PAN -UAO)\n pc : debug_dma_assert_idle+0x1f8/0x270\n lr : debug_dma_assert_idle+0x1f8/0x270\n sp : ffff0000113cfc10\n x29: ffff0000113cfc10 x28: 0000ffff8c880000\n x27: ffff800bc72a0000 x26: ffff000010ff8000\n x25: ffff000010ff8940 x24: ffff000010ff8968\n x23: 0000000000000000 x22: ffff000010e83700\n x21: ffff000010ea2000 x20: ffff000010e835c8\n x19: ffff800bc2c73300 x18: ffffffffffffffff\n x17: 0000000000000000 x16: 0000000000000000\n x15: ffff000010e835c8 x14: 6d20616d64206576\n x13: 69746361206e6120 x12: 676e696863756f74\n x11: 20757063203a342e x10: 31303a31303a3030\n x9 : 303020636d6d5f78 x8 : 3230303030303030\n x7 : 00000000000002fd x6 : ffff000010fd57d0\n x5 : 0000000000000000 x4 : ffff0000106c5210\n x3 : 00000000ffffffff x2 : 0000800bee9c0000\n x1 : 57d5843f4aa62800 x0 : 0000000000000000\n Call trace:\n debug_dma_assert_idle+0x1f8/0x270\n wp_page_copy+0xb0/0x688\n do_wp_page+0xa8/0x5b8\n __handle_mm_fault+0x600/0xd00\n handle_mm_fault+0x118/0x1e8\n do_page_fault+0x200/0x500\n do_mem_abort+0x50/0xb0\n el0_da+0x20/0x24\n ---[ end trace a005534bd23e109f ]---\n DMA-API: Mapped at:\n debug_dma_map_sg+0x94/0x350\n cvm_mmc_request+0x3c4/0x988\n __mmc_start_request+0x9c/0x1f8\n mmc_start_request+0x7c/0xb0\n mmc_blk_mq_issue_rq+0x5c4/0x7b8\n\nSigned-off-by: Kevin Hao \u003chaokexin@gmail.com\u003e\nFixes: ba3869ff32e4 (\"mmc: cavium: Add core MMC driver for Cavium SOCs\")\nCc: stable@vger.kernel.org\nSigned-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n" }, { "commit": "fa25eba6993b3750f417baabba169afaba076178", "tree": "749c7cb4ecc3e1f1b64ab3761051e459eb7793c4", "parents": [ "c6303c5d52d5ec3e5bce2e6a5480fa2a1baa45e6" ], "author": { "name": "Kevin Hao", "email": "haokexin@gmail.com", "time": "Fri Jul 26 10:30:48 2019 +0800" }, "committer": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Tue Aug 06 18:59:14 2019 +0200" }, "message": "mmc: cavium: Set the correct dma max segment size for mmc_host\n\nWe have set the mmc_host.max_seg_size to 8M, but the dma max segment\nsize of PCI device is set to 64K by default in function pci_device_add().\nThe mmc_host.max_seg_size is used to set the max segment size of\nthe blk queue. Then this mismatch will trigger a calltrace like below\nwhen a bigger than 64K segment request arrives at mmc dev. So we should\nconsider the limitation of the cvm_mmc_host when setting the\nmmc_host.max_seg_size.\n DMA-API: thunderx_mmc 0000:01:01.4: mapping sg segment longer than device claims to support [len\u003d131072] [max\u003d65536]\n WARNING: CPU: 6 PID: 238 at kernel/dma/debug.c:1221 debug_dma_map_sg+0x2b8/0x350\n Modules linked in:\n CPU: 6 PID: 238 Comm: kworker/6:1H Not tainted 5.3.0-rc1-next-20190724-yocto-standard+ #62\n Hardware name: Marvell OcteonTX CN96XX board (DT)\n Workqueue: kblockd blk_mq_run_work_fn\n pstate: 80c00009 (Nzcv daif +PAN +UAO)\n pc : debug_dma_map_sg+0x2b8/0x350\n lr : debug_dma_map_sg+0x2b8/0x350\n sp : ffff00001770f9e0\n x29: ffff00001770f9e0 x28: ffffffff00000000\n x27: 00000000ffffffff x26: ffff800bc2c73180\n x25: ffff000010e83700 x24: 0000000000000002\n x23: 0000000000000001 x22: 0000000000000001\n x21: 0000000000000000 x20: ffff800bc48ba0b0\n x19: ffff800bc97e8c00 x18: ffffffffffffffff\n x17: 0000000000000000 x16: 0000000000000000\n x15: ffff000010e835c8 x14: 6874207265676e6f\n x13: 6c20746e656d6765 x12: 7320677320676e69\n x11: 7070616d203a342e x10: 31303a31303a3030\n x9 : 303020636d6d5f78 x8 : 35363d78616d5b20\n x7 : 00000000000002fd x6 : ffff000010fd57dc\n x5 : 0000000000000000 x4 : ffff0000106c61f0\n x3 : 00000000ffffffff x2 : 0000800bee060000\n x1 : 7010678df3041a00 x0 : 0000000000000000\n Call trace:\n debug_dma_map_sg+0x2b8/0x350\n cvm_mmc_request+0x3c4/0x988\n __mmc_start_request+0x9c/0x1f8\n mmc_start_request+0x7c/0xb0\n mmc_blk_mq_issue_rq+0x5c4/0x7b8\n mmc_mq_queue_rq+0x11c/0x278\n blk_mq_dispatch_rq_list+0xb0/0x568\n blk_mq_do_dispatch_sched+0x6c/0x108\n blk_mq_sched_dispatch_requests+0x110/0x1b8\n __blk_mq_run_hw_queue+0xb0/0x118\n blk_mq_run_work_fn+0x28/0x38\n process_one_work+0x210/0x490\n worker_thread+0x48/0x458\n kthread+0x130/0x138\n ret_from_fork+0x10/0x1c\n\nSigned-off-by: Kevin Hao \u003chaokexin@gmail.com\u003e\nFixes: ba3869ff32e4 (\"mmc: cavium: Add core MMC driver for Cavium SOCs\")\nCc: stable@vger.kernel.org\nSigned-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n" }, { "commit": "c6303c5d52d5ec3e5bce2e6a5480fa2a1baa45e6", "tree": "979f6cfebb12a4732d460920a9f93a986143cfc5", "parents": [ "e21a712a9685488f5ce80495b37b9fdbe96c230d" ], "author": { "name": "Baolin Wang", "email": "baolin.wang@linaro.org", "time": "Thu Jul 25 11:14:22 2019 +0800" }, "committer": { "name": "Ulf Hansson", "email": "ulf.hansson@linaro.org", "time": "Tue Aug 06 18:59:14 2019 +0200" }, "message": "mmc: sdhci-sprd: Fix the incorrect soft reset operation when runtime resuming\n\nThe SD host controller specification defines 3 types software reset:\nsoftware reset for data line, software reset for command line and software\nreset for all. Software reset for all means this reset affects the entire\nHost controller except for the card detection circuit.\n\nIn sdhci_runtime_resume_host() we always do a software \"reset for all\",\nwhich causes the Spreadtrum variant controller to work abnormally after\nresuming. To fix the problem, let\u0027s do a software reset for the data and\nthe command part, rather than \"for all\".\n\nHowever, as sdhci_runtime_resume() is a common sdhci function and we don\u0027t\nwant to change the behaviour for other variants, let\u0027s introduce a new\nin-parameter for it. This enables the caller to decide if a \"reset for all\"\nshall be done or not.\n\nSigned-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nFixes: fb8bd90f83c4 (\"mmc: sdhci-sprd: Add Spreadtrum\u0027s initial host controller\")\nCc: stable@vger.kernel.org\nSigned-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n" }, { "commit": "db341a049ec7e87053c91008cb452d0bfa6dde72", "tree": "42e619cf1a9dc9673f6248f765fef668a594d037", "parents": [ "41995342b40c418a47603e1321256d2c4a2ed0fb" ], "author": { "name": "Gustavo A. R. Silva", "email": "gustavo@embeddedor.com", "time": "Tue Aug 06 03:08:08 2019 -0500" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Tue Aug 06 07:44:59 2019 -0600" }, "message": "ata: rb532_cf: Fix unused variable warning in rb532_pata_driver_probe\n\nFix the following warning (Building: rb532_defconfig mips):\n\ndrivers/ata/pata_rb532_cf.c: In function ‘rb532_pata_driver_remove’:\ndrivers/ata/pata_rb532_cf.c:161:24: warning: unused variable ‘info’ [-Wunused-variable]\n struct rb532_cf_info *info \u003d ah-\u003eprivate_data;\n ^~~~\n\nFixes: cd56f35e52d9 (\"ata: rb532_cf: Convert to use GPIO descriptors\")\nSigned-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "c1c6c877b0c79fd7e05c931435aa42211eaeebaf", "tree": "da042ea3718ff43df8175b7a991beb775d592886", "parents": [ "a67060201b746a308b1674f66bf289c9faef6d09" ], "author": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 14:03:56 2019 +0200" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 14:06:58 2019 +0200" }, "message": "ALSA: hda - Don\u0027t override global PCM hw info flag\n\nThe commit bfcba288b97f (\"ALSA - hda: Add support for link audio time\nreporting\") introduced the conditional PCM hw info setup, but it\noverwrites the global azx_pcm_hw object. This will cause a problem if\nany other HD-audio controller, as it\u0027ll inherit the same bit flag\nalthough another controller doesn\u0027t support that feature.\n\nFix the bug by setting the PCM hw info flag locally.\n\nFixes: bfcba288b97f (\"ALSA - hda: Add support for link audio time reporting\")\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" }, { "commit": "24350fdadbdec780406a1ef988e6cd3875e374a8", "tree": "75fc24d0714d6bd3d3bb30f1d6849d35764d66c4", "parents": [ "b9f23b7376c21f5bf7f5e2b7dfcb82cc9ce92dfc" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Mon Aug 05 14:25:16 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Tue Aug 06 13:58:35 2019 +0200" }, "message": "s390: put _stext and _etext into .text section\n\nPerf relies on _etext and _stext symbols being one of \u0027t\u0027, \u0027T\u0027, \u0027v\u0027 or\n\u0027V\u0027. Put them into .text section to guarantee that.\n\nAlso moves padding to page boundary inside .text which has an effect that\n.text section is now padded with nops rather than 0\u0027s, which apparently\nhas been the initial intention for specifying 0x0700 fill expression.\n\nReported-by: Thomas Richter \u003ctmricht@linux.ibm.com\u003e\nTested-by: Thomas Richter \u003ctmricht@linux.ibm.com\u003e\nSuggested-by: Andreas Krebbel \u003ckrebbel@linux.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "b9f23b7376c21f5bf7f5e2b7dfcb82cc9ce92dfc", "tree": "a330d7e4788061876a42818c086c3bcc94fb052b", "parents": [ "fd0c7435d7bfd46161ef43b185f55d759022238c" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Mon Aug 05 16:36:56 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Tue Aug 06 13:58:35 2019 +0200" }, "message": "s390/head64: cleanup unused labels\n\nCleanup labels in head64 some of which are not being used since git\nrecorded history.\n\nAcked-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "fd0c7435d7bfd46161ef43b185f55d759022238c", "tree": "8dde6bcd8efc34f712a621f1d7435a66663b854e", "parents": [ "218ddd5acfeae9e19a2fa97c284bf3c2e9de8ebd" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Mon Aug 05 15:07:19 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Tue Aug 06 13:58:35 2019 +0200" }, "message": "s390/unwind: remove stack recursion warning\n\nRemove pointless stack recursion on stack type ... warning, which\nonly confuses people. There is no way to make backchain unwinder 100%\nreliable. When a task is interrupted in-between stack frame allocation\nand backchain write instructions new stack frame backchain pointer is\nleft uninitialized (there are also sometimes additional instruction\nin-between stack frame allocation and backchain write instructions due\nto gcc shrink-wrapping). In attempt to unwind such stack the unwinder\nwould still try to use that invalid backchain value and perform all kind\nof sanity checks on it to make sure we are not pointed out of stack. In\nsome cases that invalid backchain value would be 0 and we would falsely\ntreat next stackframe as pt_regs and again gprs[15] in those pt_regs\nmight happen to point at some address within the task\u0027s stack.\n\nAcked-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "218ddd5acfeae9e19a2fa97c284bf3c2e9de8ebd", "tree": "066d640c9289dd5e91cb7858ce32511d9d681e38", "parents": [ "8024b5a9fc2bed9a00f0bdba60b443fa3cc4bb5d" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Fri Aug 02 17:44:29 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Tue Aug 06 13:58:34 2019 +0200" }, "message": "s390/setup: adjust start_code of init_mm to _text\n\nAfter some investigation it doesn\u0027t look like init_mm fields\nstart_code/end_code are used anywhere besides potentially in dump_mm for\ndebugging purposes. Originally the value of 0 for start_code reflected\nthe presence of lowcore and early boot code. But with kaslr in place\nstart_code/end_code range should not span over unoccupied by the code\nsegment memory. So, adjust init_mm start_code to point at the beginning\nof the code segment like other architectures do it.\n\nAcked-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "8024b5a9fc2bed9a00f0bdba60b443fa3cc4bb5d", "tree": "f50d0805dd78b45781b6c0757f6938f566799341", "parents": [ "a287a49e672d9762bb85de117b477bdf3ef20bd5" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Fri Jul 26 08:23:20 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Tue Aug 06 13:58:34 2019 +0200" }, "message": "s390/mm: fix dump_pagetables top level page table walking\n\nSince commit d1874a0c2805 (\"s390/mm: make the pxd_offset functions more\nrobust\") behaviour of p4d_offset, pud_offset and pmd_offset has been\nchanged so that they cannot be used to iterate through top level page\ntable, because the index for the top level page table is now calculated\nin pgd_offset. To avoid dumping the very first region/segment top level\ntable entry 2048 times simply iterate entry pointer like it is already\ndone in other page walking cases.\n\nFixes: d1874a0c2805 (\"s390/mm: make the pxd_offset functions more robust\")\nReported-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\nReviewed-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "a287a49e672d9762bb85de117b477bdf3ef20bd5", "tree": "4cb54baa96ff3a09561b7e021034a0f51c5d2ea8", "parents": [ "e21a712a9685488f5ce80495b37b9fdbe96c230d" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Fri Aug 02 13:27:22 2019 +0200" }, "committer": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Tue Aug 06 13:58:34 2019 +0200" }, "message": "s390/protvirt: avoid memory sharing for diag 308 set/store\n\nThis reverts commit db9492cef45e (\"s390/protvirt: add memory sharing for\ndiag 308 set/store\") which due to ultravisor implementation change is\nnot needed after all.\n\nFixes: db9492cef45e (\"s390/protvirt: add memory sharing for diag 308 set/store\")\nReviewed-by: Janosch Frank \u003cfrankja@linux.ibm.com\u003e\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n" }, { "commit": "e0f6974a54d3f7f1b5fdf5a593bd43ce9206ec04", "tree": "2e217dec39f7f52d1cbfb53f883c683ce740f47c", "parents": [ "6d4472d7bec39917b54e4e80245784ea5d60ce49" ], "author": { "name": "Roderick Colenbrander", "email": "roderick@gaikai.com", "time": "Fri Aug 02 15:50:19 2019 -0700" }, "committer": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Tue Aug 06 12:46:15 2019 +0200" }, "message": "HID: sony: Fix race condition between rumble and device remove.\n\nValve reported a kernel crash on Ubuntu 18.04 when disconnecting a DS4\ngamepad while rumble is enabled. This issue is reproducible with a\nfrequency of 1 in 3 times in the game Borderlands 2 when using an\nautomatic weapon, which triggers many rumble operations.\n\nWe found the issue to be a race condition between sony_remove and the\nfinal device destruction by the HID / input system. The problem was\nthat sony_remove didn\u0027t clean some of its work_item state in\n\"struct sony_sc\". After sony_remove work, the corresponding evdev\nnode was around for sufficient time for applications to still queue\nrumble work after \"sony_remove\".\n\nOn pre-4.19 kernels the race condition caused a kernel crash due to a\nNULL-pointer dereference as \"sc-\u003eoutput_report_dmabuf\" got freed during\nsony_remove. On newer kernels this crash doesn\u0027t happen due the buffer\nnow being allocated using devm_kzalloc. However we can still queue work,\nwhile the driver is an undefined state.\n\nThis patch fixes the described problem, by guarding the work_item\n\"state_worker\" with an initialized variable, which we are setting back\nto 0 on cleanup.\n\nSigned-off-by: Roderick Colenbrander \u003croderick.colenbrander@sony.com\u003e\nCC: stable@vger.kernel.org\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "a67060201b746a308b1674f66bf289c9faef6d09", "tree": "d87075ec063df5a0a5fd98c92a708bbfa7cb2b36", "parents": [ "d1d03306659d24cba3f594bbbb11a22ef32f45fe" ], "author": { "name": "Wenwen Wang", "email": "wenwen@cs.uga.edu", "time": "Tue Aug 06 03:00:27 2019 -0400" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 12:44:44 2019 +0200" }, "message": "ALSA: usb-audio: fix a memory leak bug\n\nIn snd_usb_get_audioformat_uac3(), a structure for channel maps \u0027chmap\u0027 is\nallocated through kzalloc() before the execution goto \u0027found_clock\u0027.\nHowever, this structure is not deallocated if the memory allocation for\n\u0027pd\u0027 fails, leading to a memory leak bug.\n\nTo fix the above issue, free \u0027fp-\u003echmap\u0027 before returning NULL.\n\nFixes: 7edf3b5e6a45 (\"ALSA: usb-audio: AudioStreaming Power Domain parsing\")\nSigned-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" }, { "commit": "6d4472d7bec39917b54e4e80245784ea5d60ce49", "tree": "56698233d5c1d5b4434003f3ff5849a3db94b6c3", "parents": [ "9c09b214f30e3c11f9b0b03f89442df03643794d" ], "author": { "name": "Hillf Danton", "email": "hdanton@sina.com", "time": "Tue Aug 06 16:40:15 2019 +0800" }, "committer": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Tue Aug 06 12:42:22 2019 +0200" }, "message": "HID: hiddev: do cleanup in failure of opening a device\n\nUndo what we did for opening before releasing the memory slice.\n\nReported-by: syzbot \u003csyzbot+62a1e04fd3ec2abf099e@syzkaller.appspotmail.com\u003e\nCc: Andrey Konovalov \u003candreyknvl@google.com\u003e\nSigned-off-by: Hillf Danton \u003chdanton@sina.com\u003e\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "9c09b214f30e3c11f9b0b03f89442df03643794d", "tree": "9390906fa75d5dde80001e8778a61fb3b1c0247a", "parents": [ "1c703b53e5bfb5c2205c30f0fb157ce271fd42fb" ], "author": { "name": "Hillf Danton", "email": "hdanton@sina.com", "time": "Tue Aug 06 16:38:58 2019 +0800" }, "committer": { "name": "Jiri Kosina", "email": "jkosina@suse.cz", "time": "Tue Aug 06 12:42:21 2019 +0200" }, "message": "HID: hiddev: avoid opening a disconnected device\n\nsyzbot found the following crash on:\n\nHEAD commit: e96407b4 usb-fuzzer: main usb gadget fuzzer driver\ngit tree: https://github.com/google/kasan.git usb-fuzzer\nconsole output: https://syzkaller.appspot.com/x/log.txt?x\u003d147ac20c600000\nkernel config: https://syzkaller.appspot.com/x/.config?x\u003d792eb47789f57810\ndashboard link: https://syzkaller.appspot.com/bug?extid\u003d62a1e04fd3ec2abf099e\ncompiler: gcc (GCC) 9.0.0 20181231 (experimental)\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nBUG: KASAN: use-after-free in __lock_acquire+0x302a/0x3b50\nkernel/locking/lockdep.c:3753\nRead of size 8 at addr ffff8881cf591a08 by task syz-executor.1/26260\n\nCPU: 1 PID: 26260 Comm: syz-executor.1 Not tainted 5.3.0-rc2+ #24\nHardware name: Google Google Compute Engine/Google Compute Engine, BIOS\nGoogle 01/01/2011\nCall Trace:\n __dump_stack lib/dump_stack.c:77 [inline]\n dump_stack+0xca/0x13e lib/dump_stack.c:113\n print_address_description+0x6a/0x32c mm/kasan/report.c:351\n __kasan_report.cold+0x1a/0x33 mm/kasan/report.c:482\n kasan_report+0xe/0x12 mm/kasan/common.c:612\n __lock_acquire+0x302a/0x3b50 kernel/locking/lockdep.c:3753\n lock_acquire+0x127/0x320 kernel/locking/lockdep.c:4412\n __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]\n _raw_spin_lock_irqsave+0x32/0x50 kernel/locking/spinlock.c:159\n hiddev_release+0x82/0x520 drivers/hid/usbhid/hiddev.c:221\n __fput+0x2d7/0x840 fs/file_table.c:280\n task_work_run+0x13f/0x1c0 kernel/task_work.c:113\n exit_task_work include/linux/task_work.h:22 [inline]\n do_exit+0x8ef/0x2c50 kernel/exit.c:878\n do_group_exit+0x125/0x340 kernel/exit.c:982\n get_signal+0x466/0x23d0 kernel/signal.c:2728\n do_signal+0x88/0x14e0 arch/x86/kernel/signal.c:815\n exit_to_usermode_loop+0x1a2/0x200 arch/x86/entry/common.c:159\n prepare_exit_to_usermode arch/x86/entry/common.c:194 [inline]\n syscall_return_slowpath arch/x86/entry/common.c:274 [inline]\n do_syscall_64+0x45f/0x580 arch/x86/entry/common.c:299\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\nRIP: 0033:0x459829\nCode: fd b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7\n48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01 f0 ff\nff 0f 83 cb b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00\nRSP: 002b:00007f75b2a6ccf8 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca\nRAX: fffffffffffffe00 RBX: 000000000075c078 RCX: 0000000000459829\nRDX: 0000000000000000 RSI: 0000000000000080 RDI: 000000000075c078\nRBP: 000000000075c070 R08: 0000000000000000 R09: 0000000000000000\nR10: 0000000000000000 R11: 0000000000000246 R12: 000000000075c07c\nR13: 00007ffcdfe1023f R14: 00007f75b2a6d9c0 R15: 000000000075c07c\n\nAllocated by task 104:\n save_stack+0x1b/0x80 mm/kasan/common.c:69\n set_track mm/kasan/common.c:77 [inline]\n __kasan_kmalloc mm/kasan/common.c:487 [inline]\n __kasan_kmalloc.constprop.0+0xbf/0xd0 mm/kasan/common.c:460\n kmalloc include/linux/slab.h:552 [inline]\n kzalloc include/linux/slab.h:748 [inline]\n hiddev_connect+0x242/0x5b0 drivers/hid/usbhid/hiddev.c:900\n hid_connect+0x239/0xbb0 drivers/hid/hid-core.c:1882\n hid_hw_start drivers/hid/hid-core.c:1981 [inline]\n hid_hw_start+0xa2/0x130 drivers/hid/hid-core.c:1972\n appleir_probe+0x13e/0x1a0 drivers/hid/hid-appleir.c:308\n hid_device_probe+0x2be/0x3f0 drivers/hid/hid-core.c:2209\n really_probe+0x281/0x650 drivers/base/dd.c:548\n driver_probe_device+0x101/0x1b0 drivers/base/dd.c:709\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:816\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:882\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2114\n hid_add_device+0x33c/0x990 drivers/hid/hid-core.c:2365\n usbhid_probe+0xa81/0xfa0 drivers/hid/usbhid/hid-core.c:1386\n usb_probe_interface+0x305/0x7a0 drivers/usb/core/driver.c:361\n really_probe+0x281/0x650 drivers/base/dd.c:548\n driver_probe_device+0x101/0x1b0 drivers/base/dd.c:709\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:816\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:882\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2114\n usb_set_configuration+0xdf6/0x1670 drivers/usb/core/message.c:2023\n generic_probe+0x9d/0xd5 drivers/usb/core/generic.c:210\n usb_probe_device+0x99/0x100 drivers/usb/core/driver.c:266\n really_probe+0x281/0x650 drivers/base/dd.c:548\n driver_probe_device+0x101/0x1b0 drivers/base/dd.c:709\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:816\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:882\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2114\n usb_new_device.cold+0x6a4/0xe79 drivers/usb/core/hub.c:2536\n hub_port_connect drivers/usb/core/hub.c:5098 [inline]\n hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]\n port_event drivers/usb/core/hub.c:5359 [inline]\n hub_event+0x1b5c/0x3640 drivers/usb/core/hub.c:5441\n process_one_work+0x92b/0x1530 kernel/workqueue.c:2269\n worker_thread+0x96/0xe20 kernel/workqueue.c:2415\n kthread+0x318/0x420 kernel/kthread.c:255\n ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352\n\nFreed by task 104:\n save_stack+0x1b/0x80 mm/kasan/common.c:69\n set_track mm/kasan/common.c:77 [inline]\n __kasan_slab_free+0x130/0x180 mm/kasan/common.c:449\n slab_free_hook mm/slub.c:1423 [inline]\n slab_free_freelist_hook mm/slub.c:1470 [inline]\n slab_free mm/slub.c:3012 [inline]\n kfree+0xe4/0x2f0 mm/slub.c:3953\n hiddev_connect.cold+0x45/0x5c drivers/hid/usbhid/hiddev.c:914\n hid_connect+0x239/0xbb0 drivers/hid/hid-core.c:1882\n hid_hw_start drivers/hid/hid-core.c:1981 [inline]\n hid_hw_start+0xa2/0x130 drivers/hid/hid-core.c:1972\n appleir_probe+0x13e/0x1a0 drivers/hid/hid-appleir.c:308\n hid_device_probe+0x2be/0x3f0 drivers/hid/hid-core.c:2209\n really_probe+0x281/0x650 drivers/base/dd.c:548\n driver_probe_device+0x101/0x1b0 drivers/base/dd.c:709\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:816\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:882\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2114\n hid_add_device+0x33c/0x990 drivers/hid/hid-core.c:2365\n usbhid_probe+0xa81/0xfa0 drivers/hid/usbhid/hid-core.c:1386\n usb_probe_interface+0x305/0x7a0 drivers/usb/core/driver.c:361\n really_probe+0x281/0x650 drivers/base/dd.c:548\n driver_probe_device+0x101/0x1b0 drivers/base/dd.c:709\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:816\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:882\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2114\n usb_set_configuration+0xdf6/0x1670 drivers/usb/core/message.c:2023\n generic_probe+0x9d/0xd5 drivers/usb/core/generic.c:210\n usb_probe_device+0x99/0x100 drivers/usb/core/driver.c:266\n really_probe+0x281/0x650 drivers/base/dd.c:548\n driver_probe_device+0x101/0x1b0 drivers/base/dd.c:709\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:816\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:882\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2114\n usb_new_device.cold+0x6a4/0xe79 drivers/usb/core/hub.c:2536\n hub_port_connect drivers/usb/core/hub.c:5098 [inline]\n hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]\n port_event drivers/usb/core/hub.c:5359 [inline]\n hub_event+0x1b5c/0x3640 drivers/usb/core/hub.c:5441\n process_one_work+0x92b/0x1530 kernel/workqueue.c:2269\n worker_thread+0x96/0xe20 kernel/workqueue.c:2415\n kthread+0x318/0x420 kernel/kthread.c:255\n ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352\n\nThe buggy address belongs to the object at ffff8881cf591900\n which belongs to the cache kmalloc-512 of size 512\nThe buggy address is located 264 bytes inside of\n 512-byte region [ffff8881cf591900, ffff8881cf591b00)\nThe buggy address belongs to the page:\npage:ffffea00073d6400 refcount:1 mapcount:0 mapping:ffff8881da002500\nindex:0x0 compound_mapcount: 0\nflags: 0x200000000010200(slab|head)\nraw: 0200000000010200 0000000000000000 0000000100000001 ffff8881da002500\nraw: 0000000000000000 00000000000c000c 00000001ffffffff 0000000000000000\npage dumped because: kasan: bad access detected\n\nMemory state around the buggy address:\n ffff8881cf591900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n ffff8881cf591980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n\u003e ffff8881cf591a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n ^\n ffff8881cf591a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n ffff8881cf591b00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nIn order to avoid opening a disconnected device, we need to check exist\nagain after acquiring the existance lock, and bail out if necessary.\n\nReported-by: syzbot \u003csyzbot+62a1e04fd3ec2abf099e@syzkaller.appspotmail.com\u003e\nCc: Andrey Konovalov \u003candreyknvl@google.com\u003e\nSigned-off-by: Hillf Danton \u003chdanton@sina.com\u003e\nSigned-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n" }, { "commit": "d1d03306659d24cba3f594bbbb11a22ef32f45fe", "tree": "bdfc4dc03c79f88638ca5f67ef560e9de1c96d01", "parents": [ "c2913b140f901a705809f8b6c83998cf8147f848", "7c0767643f3b6b0dd2cda923ae37a18590d431cf" ], "author": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 12:28:28 2019 +0200" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 12:28:28 2019 +0200" }, "message": "Merge tag \u0027asoc-fix-v5.3-rc3-2\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus\n\nASoC: Fixes for v5.3\n\nIncremental fix removing executable bits added in a prior patch\naccidentally.\n" }, { "commit": "c2913b140f901a705809f8b6c83998cf8147f848", "tree": "53686878548921b83e8d9db437c4437140561980", "parents": [ "5d78e1c2b7f4be00bbe62141603a631dc7812f35", "30c21734d853dae99d05a5295a59b7e26ccd5135" ], "author": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 12:28:08 2019 +0200" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Aug 06 12:28:08 2019 +0200" }, "message": "Merge tag \u0027asoc-fix-v5.3-rc3\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus\n\nASoC: Fixes for v5.3\n\nA relatively large batch of mostly unremarkable fixes here, a couple of\nsmall core fixes for fairly obscure issues, more comment/email updates\nwith no code impact than usual and a bunch of small driver fixes.\n\nThe support for new sample rates in the max98373 driver is a fix for the\nfact that the driver declared support for those rates but would in fact\nreturn an error if these rates were selected.\n" }, { "commit": "7c0767643f3b6b0dd2cda923ae37a18590d431cf", "tree": "6307fa15c95e7af6a52d11ff95462d0773bf3a1d", "parents": [ "30c21734d853dae99d05a5295a59b7e26ccd5135" ], "author": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Tue Aug 06 11:15:06 2019 +0100" }, "committer": { "name": "Mark Brown", "email": "broonie@kernel.org", "time": "Tue Aug 06 11:15:06 2019 +0100" }, "message": "ASoC: max98373: Remove executable bits\n\nSigned-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n" }, { "commit": "ee9d66182392695535cc9fccfcb40c16f72de2a9", "tree": "d350eb332727522f430901b46d8e4d3cf32d51a8", "parents": [ "108624eaff6f17f1966e4a94dc7bc5df2c5115ee" ], "author": { "name": "Sebastien Tisserant", "email": "stisserant@wallix.com", "time": "Thu Aug 01 12:06:08 2019 -0500" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Mon Aug 05 22:50:38 2019 -0500" }, "message": "SMB3: Kernel oops mounting a encryptData share with CONFIG_DEBUG_VIRTUAL\n\nFix kernel oops when mounting a encryptData CIFS share with\nCONFIG_DEBUG_VIRTUAL\n\nSigned-off-by: Sebastien Tisserant \u003cstisserant@wallix.com\u003e\nReviewed-by: Pavel Shilovsky \u003cpshilov@microsoft.com\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "108624eaff6f17f1966e4a94dc7bc5df2c5115ee", "tree": "6f7a42118e38e6d83d890a5f9ed78306a9d78a5b", "parents": [ "8d33096a460d5b9bd13300f01615df5bb454db10" ], "author": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Wed Jul 31 16:41:03 2019 -0500" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Mon Aug 05 22:50:38 2019 -0500" }, "message": "smb3: update TODO list of missing features\n\nminor cleanup of documentation, updating to more current status.\n\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "8d33096a460d5b9bd13300f01615df5bb454db10", "tree": "ce78d625cb14e0f9ab67af127a80f69dac726f40", "parents": [ "3edeb4a4146dc3b54d6fa71b7ee0585cb52ebfdf" ], "author": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Thu Jul 25 18:13:10 2019 -0500" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Mon Aug 05 22:50:38 2019 -0500" }, "message": "smb3: send CAP_DFS capability during session setup\n\nWe had a report of a server which did not do a DFS referral\nbecause the session setup Capabilities field was set to 0\n(unlike negotiate protocol where we set CAP_DFS). Better to\nsend it session setup in the capabilities as well (this also\nmore closely matches Windows client behavior).\n\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\nReviewed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nReviewed-by: Pavel Shilovsky \u003cpshilov@microsoft.com\u003e\nCC: Stable \u003cstable@vger.kernel.org\u003e\n" }, { "commit": "3edeb4a4146dc3b54d6fa71b7ee0585cb52ebfdf", "tree": "6c4487f7274644d38f128e31e37bba38f9880085", "parents": [ "e99c63e4d86d3a94818693147b469fa70de6f945" ], "author": { "name": "Pavel Shilovsky", "email": "pshilov@microsoft.com", "time": "Mon Jul 22 11:38:22 2019 -0700" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Mon Aug 05 22:50:13 2019 -0500" }, "message": "SMB3: Fix potential memory leak when processing compound chain\n\nWhen a reconnect happens in the middle of processing a compound chain\nthe code leaks a buffer from the memory pool. Fix this by properly\nchecking for a return code and freeing buffers in case of error.\n\nAlso maintain a buf variable to be equal to either smallbuf or bigbuf\ndepending on a response buffer size while parsing a chain and when\nreturning to the caller.\n\nSigned-off-by: Pavel Shilovsky \u003cpshilov@microsoft.com\u003e\nReviewed-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "e99c63e4d86d3a94818693147b469fa70de6f945", "tree": "2328d27ed358c8b4c1cf0af0788d30382a7f6779", "parents": [ "247bc9470b1eeefc7b58cdf2c39f2866ba651509" ], "author": { "name": "Pavel Shilovsky", "email": "pshilov@microsoft.com", "time": "Mon Jul 22 11:34:59 2019 -0700" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Mon Aug 05 22:49:54 2019 -0500" }, "message": "SMB3: Fix deadlock in validate negotiate hits reconnect\n\nCurrently we skip SMB2_TREE_CONNECT command when checking during\nreconnect because Tree Connect happens when establishing\nan SMB session. For SMB 3.0 protocol version the code also calls\nvalidate negotiate which results in SMB2_IOCL command being sent\nover the wire. This may deadlock on trying to acquire a mutex when\nchecking for reconnect. Fix this by skipping SMB2_IOCL command\nwhen doing the reconnect check.\n\nSigned-off-by: Pavel Shilovsky \u003cpshilov@microsoft.com\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\nReviewed-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nCC: Stable \u003cstable@vger.kernel.org\u003e\n" }, { "commit": "74034a09267c1f48d5ce7ae4c4a317fac7d43418", "tree": "b09d30a0d31fabbdc967e28e99e3b61d5c672d7a", "parents": [ "2d291e6cc183608e1f6eb8580268df7c68bdf7c3" ], "author": { "name": "Gustavo A. R. Silva", "email": "gustavo@embeddedor.com", "time": "Mon Aug 05 13:55:33 2019 -0500" }, "committer": { "name": "Paul Burton", "email": "paul.burton@mips.com", "time": "Mon Aug 05 18:25:30 2019 -0700" }, "message": "MIPS: BCM63XX: Mark expected switch fall-through\n\nMark switch cases where we are expecting to fall through.\n\nThis patch fixes the following warning (Building: bcm63xx_defconfig mips):\n\narch/mips/pci/ops-bcm63xx.c: In function ‘bcm63xx_pcie_can_access’:\narch/mips/pci/ops-bcm63xx.c:474:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (PCI_SLOT(devfn) \u003d\u003d 0)\n ^\narch/mips/pci/ops-bcm63xx.c:477:2: note: here\n default:\n ^~~~~~~\n\nSigned-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nSigned-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: James Hogan \u003cjhogan@kernel.org\u003e\nCc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nCc: bcm-kernel-feedback-list@broadcom.com\nCc: linux-mips@vger.kernel.org\nCc: linux-arm-kernel@lists.infradead.org\nCc: linux-kernel@vger.kernel.org\n" } ], "next": "f7813d5c637e43209ffdc20a9b25a13fcbb86eef" }