)]}' { "log": [ { "commit": "f40ddce88593482919761f74910f42f4b84c004b", "tree": "54016ed97001939c1662de9846545346ba8ceb2d", "parents": [ "28a17338738839494752c5da3e12c889a55219db" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 14:32:24 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 14:32:24 2021 -0800" }, "message": "Linux 5.11\n" }, { "commit": "28a17338738839494752c5da3e12c889a55219db", "tree": "e4726925112176ebe9c2f0aeac77d688aeb287a3", "parents": [ "ab30c7f9c3ca2599f5ab3e4d29ae56453c8668e5", "dbeb02a0bc41b9e9b9c05e460890351efecf1352" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 11:50:31 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 11:50:31 2021 -0800" }, "message": "Merge branch \u0027for-rc8-5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/pavel/linux-leds\n\nPull LED fix from Pavel Machek:\n \"One-liner fixing a build problem\"\n\n* \u0027for-rc8-5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/pavel/linux-leds:\n leds: rt8515: add V4L2_FLASH_LED_CLASS dependency\n" }, { "commit": "ab30c7f9c3ca2599f5ab3e4d29ae56453c8668e5", "tree": "140907f0bac189baa499e417af4659e5b1179815", "parents": [ "c553021498640050340924bd62fa580e253467ac", "fe968c41ac4f4ec9ffe3c4cf16b72285f5e9674f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 11:36:32 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 11:36:32 2021 -0800" }, "message": "Merge tag \u0027kbuild-fixes-v5.11-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild\n\nPull Kbuild fixes from Masahiro Yamada:\n\n - Fix CONFIG_TRIM_UNUSED_KSYMS build for ppc64\n\n - Use pkg-config for scripts/sign-file.c CFLAGS\n\n* tag \u0027kbuild-fixes-v5.11-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:\n scripts: set proper OpenSSL include dir also for sign-file\n sparc: remove wrong comment from arch/sparc/include/asm/Kbuild\n kbuild: fix CONFIG_TRIM_UNUSED_KSYMS build for ppc64\n" }, { "commit": "c553021498640050340924bd62fa580e253467ac", "tree": "77351d0100c901ec41e2cdcd3e8e058b6645e738", "parents": [ "358feceebbf68f33c44c6650d14455389e65282d", "70245f86c109e0eafb92ea9653184c0e44b4b35c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 11:10:55 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 14 11:10:55 2021 -0800" }, "message": "Merge tag \u0027x86_urgent_for_v5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Borislav Petkov:\n \"I kinda knew while typing \u0027I hope this is the last batch of x86/urgent\n updates\u0027 last week, Murphy was reading too and uttered \u0027Hold my\n beer!\u0027.\n\n So here\u0027s more fixes... Thanks Murphy.\n\n Anyway, three more x86/urgent fixes for 5.11 final. We should be\n finally ready (famous last words). :-)\n\n - An SGX use after free fix\n\n - A fix for the fix to disable CET instrumentation generation for\n kernel code. We forgot 32-bit, which we seem to do very often\n nowadays\n\n - A Xen PV fix to irqdomain init ordering\"\n\n* tag \u0027x86_urgent_for_v5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/pci: Create PCI/MSI irqdomain after x86_init.pci.arch_init()\n x86/build: Disable CET instrumentation in the kernel for 32-bit too\n x86/sgx: Maintain encl-\u003erefcount for each encl-\u003emm_list entry\n" }, { "commit": "dbeb02a0bc41b9e9b9c05e460890351efecf1352", "tree": "bd5d0ca6df22385d6a9a576f69e1e252b4b5692d", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Feb 04 16:39:44 2021 +0100" }, "committer": { "name": "Pavel Machek", "email": "pavel@ucw.cz", "time": "Sun Feb 14 18:01:41 2021 +0100" }, "message": "leds: rt8515: add V4L2_FLASH_LED_CLASS dependency\n\nThe leds-rt8515 driver can optionall use the v4l2 flash led class,\nbut it causes a link error when that class is in a loadable module\nand the rt8515 driver itself is built-in:\n\nld.lld: error: undefined symbol: v4l2_flash_init\n\u003e\u003e\u003e referenced by leds-rt8515.c\n\u003e\u003e\u003e leds/flash/leds-rt8515.o:(rt8515_probe) in archive\ndrivers/built-in.a\n\nAdding \u0027depends on V4L2_FLASH_LED_CLASS\u0027 in Kconfig would avoid that,\nbut it would make it impossible to use the driver without the\nv4l2 support.\n\nAdd the same dependency that the other users of this class have\ninstead, which just prevents the broken configuration.\n\nFixes: e1c6edcbea13 (\"leds: rt8515: Add Richtek RT8515 LED driver\")\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nReviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n" }, { "commit": "fe968c41ac4f4ec9ffe3c4cf16b72285f5e9674f", "tree": "e54f1b4810f52cc8b113ec894583a4e577239003", "parents": [ "27dad89bab2d71a8c4b13b497054db33d264b281" ], "author": { "name": "Rolf Eike Beer", "email": "eb@emlix.com", "time": "Fri Feb 12 08:22:27 2021 +0100" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Mon Feb 15 01:54:11 2021 +0900" }, "message": "scripts: set proper OpenSSL include dir also for sign-file\n\nFixes: 2cea4a7a1885 (\"scripts: use pkg-config to locate libcrypto\")\nSigned-off-by: Rolf Eike Beer \u003ceb@emlix.com\u003e\nCc: stable@vger.kernel.org # 5.6.x\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\n" }, { "commit": "27dad89bab2d71a8c4b13b497054db33d264b281", "tree": "30f1320c6a925f058afa5e8b5009685dff1aeb93", "parents": [ "29500f15b54b63ad0ea60b58e85144262bd24df2" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Jan 28 09:51:03 2021 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Mon Feb 15 01:52:56 2021 +0900" }, "message": "sparc: remove wrong comment from arch/sparc/include/asm/Kbuild\n\nThese are NOT exported to userspace.\n\nThe headers listed in arch/sparc/include/uapi/asm/Kbuild are exported.\n\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\n" }, { "commit": "358feceebbf68f33c44c6650d14455389e65282d", "tree": "1ac4d2c666ccacbfda260ea186d8844f883cfee2", "parents": [ "0001ec9b1418f01a6dd44a83a1caa4b4f3d11f29", "245090ab2636c0869527ce563afbfb8aff29e825" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 14:25:22 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 14:25:22 2021 -0800" }, "message": "Merge tag \u0027clk-fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux\n\nPull clk fix from Stephen Boyd:\n \"One small fix for the Allwinner clk driver so that display clks figure\n out the correct rate to use.\n\n This fixes displays running 4k@60Hz and some other resolutions that\n haven\u0027t been exercised and fully understood until now\"\n\n* tag \u0027clk-fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:\n clk: sunxi-ng: mp: fix parent rate change flag check\n" }, { "commit": "0001ec9b1418f01a6dd44a83a1caa4b4f3d11f29", "tree": "8466054cca03144c36e683a4ebe05039a5c86a37", "parents": [ "ac30d8ce28d61c05ac3a8b1452e889371136f3af", "f852c596f2ee6f0eb364ea8f28f89da6da0ae7b5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 14:14:47 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 14:14:47 2021 -0800" }, "message": "Merge tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi\n\nPull SCSI fix from James Bottomley:\n \"One fix for scsi_debug that fixes a memory leak on module removal\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: scsi_debug: Fix a memory leak\n" }, { "commit": "ac30d8ce28d61c05ac3a8b1452e889371136f3af", "tree": "3d6b932ffafc4a34c214b52c3a7a7542d4e55dd6", "parents": [ "25cbda46779418e5ae435707675a312c99a16dff", "74bdd45c85d02f695a1cd1c3dccf8b3960a86d8f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 12:25:42 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 12:25:42 2021 -0800" }, "message": "Merge branch \u0027for-5.11-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup\n\nPull cgroup fixes from Tejun Heo:\n \"Two cgroup fixes:\n\n - fix a NULL deref when trying to poll PSI in the root cgroup\n\n - fix confusing controller parsing corner case when mounting cgroup\n v1 hierarchies\n\n And doc / maintainer file updates\"\n\n* \u0027for-5.11-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:\n cgroup: update PSI file description in docs\n cgroup: fix psi monitor for root cgroup\n MAINTAINERS: Update my email address\n MAINTAINERS: Remove stale URLs for cpuset\n cgroup-v1: add disabled controller check in cgroup1_parse_param()\n" }, { "commit": "25cbda46779418e5ae435707675a312c99a16dff", "tree": "1dbe252ac15ec4cbc5f897b1b9c5bfdab33791dc", "parents": [ "8cc8e6aaf27db47985f9e1c24ac2f7393390971e", "ade9679c159d5bbe14fb7e59e97daf6062872e2b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 12:04:18 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 12:04:18 2021 -0800" }, "message": "Merge branch \u0027akpm\u0027 (patches from Andrew)\n\nMerge fixes from Andrew Morton:\n \"6 patches.\n\n Subsystems affected by this patch series: mm/pagemap, scripts,\n MAINTAINERS, and h8300\"\n\n* emailed patches from Andrew Morton \u003cakpm@linux-foundation.org\u003e:\n h8300: fix PREEMPTION build, TI_PRE_COUNT undefined\n MAINTAINERS: add Andrey Konovalov to KASAN reviewers\n MAINTAINERS: update Andrey Konovalov\u0027s email address\n MAINTAINERS: update KASAN file list\n scripts/recordmcount.pl: support big endian for ARCH sh\n m68k: make __pfn_to_phys() and __phys_to_pfn() available for !MMU\n" }, { "commit": "8cc8e6aaf27db47985f9e1c24ac2f7393390971e", "tree": "e0af0ad633b5dc9ced805f4fa0f9467449226b98", "parents": [ "e42ee56fe59759023cb252fabb3d6f279fe8cec8", "3d6a3d3a2a7a3a60a824e7c04e95fd50dec57812" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:59:10 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:59:10 2021 -0800" }, "message": "Merge branch \u0027i2c/for-current\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux\n\nPull i2c fix from Wolfram Sang:\n \"One more I2C driver bugfix\"\n\n* \u0027i2c/for-current\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:\n i2c: stm32f7: fix configuration of the digital filter\n" }, { "commit": "e42ee56fe59759023cb252fabb3d6f279fe8cec8", "tree": "c91dce029c041b89c294824a2a22232097d0fe90", "parents": [ "7989807dc0c95c4b787698eba10905a5baac42e8", "83c68bbcb6ac2dbbcaf12e2281a29a9f73b97d0f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:55:29 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:55:29 2021 -0800" }, "message": "Merge tag \u0027for-5.11-rc7-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux\n\nPull btrfs fix from David Sterba:\n \"A regression fix caused by a refactoring in 5.11.\n\n A corrupted superblock wouldn\u0027t be detected by checksum verification\n due to wrongly placed initialization of the checksum length, thus\n making memcmp always work\"\n\n* tag \u0027for-5.11-rc7-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:\n btrfs: initialize fs_info::csum_size earlier in open_ctree\n" }, { "commit": "ade9679c159d5bbe14fb7e59e97daf6062872e2b", "tree": "f5af58d2de55c5ee57b6fa476543d22c0a45abfd", "parents": [ "fee92a765fbcc9412680228f8fdc425ab9c8151e" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Fri Feb 12 20:52:54 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:42:40 2021 -0800" }, "message": "h8300: fix PREEMPTION build, TI_PRE_COUNT undefined\n\nFix a build error for undefined \u0027TI_PRE_COUNT\u0027 by adding it to\nasm-offsets.c.\n\n h8300-linux-ld: arch/h8300/kernel/entry.o: in function `resume_kernel\u0027: (.text+0x29a): undefined reference to `TI_PRE_COUNT\u0027\n\nLink: https://lkml.kernel.org/r/20210212021650.22740-1-rdunlap@infradead.org\nFixes: df2078b8daa7 (\"h8300: Low level entry\")\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nCc: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fee92a765fbcc9412680228f8fdc425ab9c8151e", "tree": "90a88e3f58e8ec60007f7d181519263d634de836", "parents": [ "872fad10f8340207a2a6896a1311c5c7b50d2e11" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Fri Feb 12 20:52:50 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:42:40 2021 -0800" }, "message": "MAINTAINERS: add Andrey Konovalov to KASAN reviewers\n\nAdd my personal email address to KASAN reviewers list.\n\nLink: https://lkml.kernel.org/r/c1ce89a7aae0e2d6852249c280b1eb59aeac30c0.1613150186.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nCc: Alexander Potapenko \u003cglider@google.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Marco Elver \u003celver@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "872fad10f8340207a2a6896a1311c5c7b50d2e11", "tree": "a60e109b5178cb1e7bc720a99213da7268595468", "parents": [ "30d320f089d8e1ad369375e02cf62fbe5da18e85" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Fri Feb 12 20:52:47 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:42:40 2021 -0800" }, "message": "MAINTAINERS: update Andrey Konovalov\u0027s email address\n\nUse my personal email address.\n\nLink: https://lkml.kernel.org/r/b0ec98dabbc12336c162788f5ccde97045a0d65e.1613150186.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nCc: Alexander Potapenko \u003cglider@google.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Marco Elver \u003celver@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "30d320f089d8e1ad369375e02cf62fbe5da18e85", "tree": "0d0ecec7bcfd0755140610b62c772d4655bca0c0", "parents": [ "93ca696376dd3d44b9e5eae835ffbc84772023ec" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Fri Feb 12 20:52:44 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:42:40 2021 -0800" }, "message": "MAINTAINERS: update KASAN file list\n\nAccount for the following files:\n\n - lib/Kconfig.kasan\n\n - lib/test_kasan_module.c\n\n - arch/arm64/include/asm/mte-kasan.h\n\nLink: https://lkml.kernel.org/r/7f9771d97b34d396bfdc4e288ad93486bb865a06.1613150186.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nCc: Alexander Potapenko \u003cglider@google.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Marco Elver \u003celver@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "93ca696376dd3d44b9e5eae835ffbc84772023ec", "tree": "f7b4b6db712818d922cd33f63e392be650d05d70", "parents": [ "3c62cfdd10c44221050b4b94cfdf8707d31fea53" ], "author": { "name": "Rong Chen", "email": "rong.a.chen@intel.com", "time": "Fri Feb 12 20:52:41 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:42:40 2021 -0800" }, "message": "scripts/recordmcount.pl: support big endian for ARCH sh\n\nThe kernel test robot reported the following issue:\n\n CC [M] drivers/soc/litex/litex_soc_ctrl.o\n sh4-linux-objcopy: Unable to change endianness of input file(s)\n sh4-linux-ld: cannot find drivers/soc/litex/.tmp_gl_litex_soc_ctrl.o: No such file or directory\n sh4-linux-objcopy: \u0027drivers/soc/litex/.tmp_mx_litex_soc_ctrl.o\u0027: No such file\n\nThe problem is that the format of input file is elf32-shbig-linux, but\nsh4-linux-objcopy wants to output a file which format is elf32-sh-linux:\n\n $ sh4-linux-objdump -d drivers/soc/litex/litex_soc_ctrl.o | grep format\n drivers/soc/litex/litex_soc_ctrl.o: file format elf32-shbig-linux\n\nLink: https://lkml.kernel.org/r/20210210150435.2171567-1-rong.a.chen@intel.com\nLink: https://lore.kernel.org/linux-mm/202101261118.GbbYSlHu-lkp@intel.com\nSigned-off-by: Rong Chen \u003crong.a.chen@intel.com\u003e\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nCc: Yoshinori Sato \u003cysato@users.osdn.me\u003e\nCc: Rich Felker \u003cdalias@libc.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3c62cfdd10c44221050b4b94cfdf8707d31fea53", "tree": "3bdc03f07e81dc0444d27c5f5924c2122f8039ae", "parents": [ "dcc0b49040c70ad827a7f3d58a21b01fdb14e749" ], "author": { "name": "Mike Rapoport", "email": "rppt@linux.ibm.com", "time": "Fri Feb 12 20:52:38 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 13 11:42:40 2021 -0800" }, "message": "m68k: make __pfn_to_phys() and __phys_to_pfn() available for !MMU\n\nRecent changes that obsoleted DISCONTIGMEM on m68k switched the MMU\nvariant to use generic definitions of __pfn_to_phys() and __phys_to_pfn(),\nbut missed the !MMU variant which caused a build failure:\n\n drivers/media/common/videobuf2/videobuf2-dma-contig.c: In function \u0027vb2_dc_get_userptr\u0027:\n drivers/media/common/videobuf2/videobuf2-dma-contig.c:509:5: error: implicit declaration of function \u0027__pfn_to_phys\u0027 [-Werror\u003dimplicit-function-declaration]\n 509 | __pfn_to_phys(nums[0]), size, buf-\u003edma_dir, 0);\n | ^~~~~~~~~~~~~\n cc1: some warnings being treated as errors\n\nEnable __pfn_to_phys() and __phys_to_pfn() on !MMU builds.\n\nLink: https://lkml.kernel.org/r/20210211232202.GS299309@linux.ibm.com\nFixes: 4bfc848e0981 (\"m68k/mm: enable use of generic memory_model.h for !DISCONTIGMEM\")\nSigned-off-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nCc: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nCc: Greg Ungerer \u003cgerg@linux-m68k.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "7989807dc0c95c4b787698eba10905a5baac42e8", "tree": "68ae5507ac9838c17ce16d5d414f899c76d08029", "parents": [ "c6d8570e4d642a0c0bfbe7362ffa1b1433c72db1", "a738c93fb1c17e386a09304b517b1c6b2a6a5a8b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 14:45:39 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 14:45:39 2021 -0800" }, "message": "Merge tag \u00275.11-rc7-smb3-github\u0027 of git://github.com/smfrench/smb3-kernel\n\nPull cifs fixes from Steve French:\n \"Four small smb3 fixes to the new mount API (including a particularly\n important one for DFS links).\n\n These were found in testing this week of additional DFS scenarios, and\n a user testing of an apache container problem\"\n\n* tag \u00275.11-rc7-smb3-github\u0027 of git://github.com/smfrench/smb3-kernel:\n cifs: Set CIFS_MOUNT_USE_PREFIX_PATH flag on setting cifs_sb-\u003eprepath.\n cifs: In the new mount api we get the full devname as source\u003d\n cifs: do not disable noperm if multiuser mount option is not provided\n cifs: fix dfs-links\n" }, { "commit": "c6d8570e4d642a0c0bfbe7362ffa1b1433c72db1", "tree": "9644823f7df64364f458d06a373e40c9f3e89ed7", "parents": [ "a81bfdf8bf5396824d7d139560180854cb599b06", "92c75f7594d5060a4cb240f0e987a802f8486b11" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:48:02 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:48:02 2021 -0800" }, "message": "Merge tag \u0027io_uring-5.11-2021-02-12\u0027 of git://git.kernel.dk/linux-block\n\nPull io_uring fix from Jens Axboe:\n \"Revert of a patch from this release that caused a regression\"\n\n* tag \u0027io_uring-5.11-2021-02-12\u0027 of git://git.kernel.dk/linux-block:\n Revert \"io_uring: don\u0027t take fs for recvmsg/sendmsg\"\n" }, { "commit": "a81bfdf8bf5396824d7d139560180854cb599b06", "tree": "34a6e491fe9a0257474d078ca7b26bc75955c75d", "parents": [ "e77a6817d413589be35461d0cd5a431a6794b3b9", "551c81853d6d3ff016269d62612e7cd0a53104ab" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:29:06 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:29:06 2021 -0800" }, "message": "Merge tag \u0027drm-fixes-2021-02-12\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Regular fixes for final, there is a ttm regression fix, dp-mst fix,\n one amdgpu revert, two i915 fixes, and some misc fixes for sun4i,\n xlnx, and vc4.\n\n All pretty quiet and don\u0027t think we have any known outstanding\n regressions.\n\n ttm:\n - page pool regression fix.\n\n dp_mst:\n - don\u0027t report un-attached ports as connected\n\n amdgpu:\n - blank screen fix\n\n i915:\n - ensure Type-C FIA is powered when initializing\n - fix overlay frontbuffer tracking\n\n sun4i:\n - tcon1 sync polarity fix\n - always set HDMI clock rate\n - fix H6 HDMI PHY config\n - fix H6 max frequency\n\n vc4:\n - fix buffer overflow\n\n xlnx:\n - fix memory leak\"\n\n* tag \u0027drm-fixes-2021-02-12\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/ttm: make sure pool pages are cleared\n drm/sun4i: dw-hdmi: Fix max. frequency for H6\n drm/sun4i: Fix H6 HDMI PHY configuration\n drm/sun4i: dw-hdmi: always set clock rate\n drm/sun4i: tcon: set sync polarity for tcon1 channel\n drm/i915: Fix overlay frontbuffer tracking\n Revert \"drm/amd/display: Update NV1x SR latency values\"\n drm/i915/tgl+: Make sure TypeC FIA is powered up when initializing it\n drm/dp_mst: Don\u0027t report ports connected if nothing is attached to them\n drm/xlnx: fix kmemleak by sending vblank_event in atomic_disable\n drm/vc4: hvs: Fix buffer overflow with the dlist handling\n" }, { "commit": "e77a6817d413589be35461d0cd5a431a6794b3b9", "tree": "7c7b9a465f358d465fe51b5dd911148781ea22e6", "parents": [ "2dbbaae5f7b3855697e2decc5de79c7574403254", "b220c049d5196dd94d992dd2dc8cba1a5e6123bf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:16:17 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:16:17 2021 -0800" }, "message": "Merge tag \u0027trace-v5.11-rc7-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fix from Steven Rostedt:\n \"Fix buffer overflow in trace event filter.\n\n It was reported that if an trace event was larger than a page and was\n filtered, that it caused memory corruption. The reason is that\n filtered events first go into a buffer to test the filter before being\n written into the ring buffer. Unfortunately, this write did not check\n the size\"\n\n* tag \u0027trace-v5.11-rc7-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n tracing: Check length before giving out the filter buffer\n" }, { "commit": "2dbbaae5f7b3855697e2decc5de79c7574403254", "tree": "70ad2de7956c2a35def9e42da4d6b29280902dd1", "parents": [ "f9516259804123c416c4a9de40bc2e14d686478e", "c4295ab0b485b8bc50d2264bcae2acd06f25caaf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:12:58 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:12:58 2021 -0800" }, "message": "Merge tag \u0027for-linus-5.11-rc8-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip\n\nPull xen fix from Juergen Gross:\n \"A single fix for an issue introduced this development cycle: when\n running as a Xen guest on Arm systems the kernel will hang during\n boot\"\n\n* tag \u0027for-linus-5.11-rc8-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:\n arm/xen: Don\u0027t probe xenbus as part of an early initcall\n" }, { "commit": "f9516259804123c416c4a9de40bc2e14d686478e", "tree": "ab7b516e76e88d43463aef327f4325112a68b1f5", "parents": [ "93908500b8da0423c9f0511130c8ab86d59576a0", "3da3cc1b5f47115b16b5ffeeb4bf09ec331b0164" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:07:29 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:07:29 2021 -0800" }, "message": "Merge tag \u0027riscv-for-linus-5.11-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V fix from Palmer Dabbelt:\n \"A single fix this week: the removal of the GPIO reset method for the\n Ethernet phy on the HiFive Unleashed.\n\n This returns to relying on the bootloader\u0027s phy reset sequence, which\n we\u0027ll have to continue doing until we can sort out how to get the\n Linux phy driver to perform the special reset dance required for this\n phy\"\n\n* tag \u0027riscv-for-linus-5.11-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n Revert \"dts: phy: add GPIO number and active state used for phy reset\"\n" }, { "commit": "93908500b8da0423c9f0511130c8ab86d59576a0", "tree": "b67a591dfefc5463fd83f1725d4c4b37a66862e9", "parents": [ "dcc0b49040c70ad827a7f3d58a21b01fdb14e749", "68d54ceeec0e5fee4fb8048e6a04c193f32525ca" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:03:30 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 12 11:03:30 2021 -0800" }, "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 PTRACE_PEEKMTETAGS access to an mmapped region before the first\n write\"\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n arm64: mte: Allow PTRACE_PEEKMTETAGS access to the zero page\n" }, { "commit": "68d54ceeec0e5fee4fb8048e6a04c193f32525ca", "tree": "b13558953da871a0dfb7ac1aebbc285fba9317c5", "parents": [ "22cd5edb2d9c6d68b6ac0fc9584104d88710fa57" ], "author": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Wed Feb 10 18:03:16 2021 +0000" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Fri Feb 12 16:08:31 2021 +0000" }, "message": "arm64: mte: Allow PTRACE_PEEKMTETAGS access to the zero page\n\nThe ptrace(PTRACE_PEEKMTETAGS) implementation checks whether the user\npage has valid tags (mapped with PROT_MTE) by testing the PG_mte_tagged\npage flag. If this bit is cleared, ptrace(PTRACE_PEEKMTETAGS) returns\n-EIO.\n\nA newly created (PROT_MTE) mapping points to the zero page which had its\ntags zeroed during cpu_enable_mte(). If there were no prior writes to\nthis mapping, ptrace(PTRACE_PEEKMTETAGS) fails with -EIO since the zero\npage does not have the PG_mte_tagged flag set.\n\nSet PG_mte_tagged on the zero page when its tags are cleared during\nboot. In addition, to avoid ptrace(PTRACE_PEEKMTETAGS) succeeding on\n!PROT_MTE mappings pointing to the zero page, change the\n__access_remote_tags() check to (vm_flags \u0026 VM_MTE) instead of\nPG_mte_tagged.\n\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nFixes: 34bfeea4a9e9 (\"arm64: mte: Clear the tags when a page is mapped in user-space with PROT_MTE\")\nCc: \u003cstable@vger.kernel.org\u003e # 5.10.x\nCc: Will Deacon \u003cwill@kernel.org\u003e\nReported-by: Luis Machado \u003cluis.machado@linaro.org\u003e\nTested-by: Luis Machado \u003cluis.machado@linaro.org\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nLink: https://lore.kernel.org/r/20210210180316.23654-1-catalin.marinas@arm.com\n" }, { "commit": "83c68bbcb6ac2dbbcaf12e2281a29a9f73b97d0f", "tree": "1f096452c2e5b57c033eeeb41018983b8dfd3a1e", "parents": [ "9ad6d91f056b99dbe59a262810cb342519ea8d39" ], "author": { "name": "Su Yue", "email": "l@damenly.su", "time": "Thu Feb 11 16:38:28 2021 +0800" }, "committer": { "name": "David Sterba", "email": "dsterba@suse.com", "time": "Fri Feb 12 14:48:24 2021 +0100" }, "message": "btrfs: initialize fs_info::csum_size earlier in open_ctree\n\nUser reported that btrfs-progs misc-tests/028-superblock-recover fails:\n\n [TEST/misc] 028-superblock-recover\n unexpected success: mounted fs with corrupted superblock\n test failed for case 028-superblock-recover\n\nThe test case expects that a broken image with bad superblock will be\nrejected to be mounted. However, the test image just passed csum check\nof superblock and was successfully mounted.\n\nCommit 55fc29bed8dd (\"btrfs: use cached value of fs_info::csum_size\neverywhere\") replaces all calls to btrfs_super_csum_size by\nfs_info::csum_size. The calls include the place where fs_info-\u003ecsum_size\nis not initialized. So btrfs_check_super_csum() passes because memcmp()\nwith len 0 always returns 0.\n\nFix it by caching csum size in btrfs_fs_info::csum_size once we know the\ncsum type in superblock is valid in open_ctree().\n\nLink: https://github.com/kdave/btrfs-progs/issues/250\nFixes: 55fc29bed8dd (\"btrfs: use cached value of fs_info::csum_size everywhere\")\nSigned-off-by: Su Yue \u003cl@damenly.su\u003e\nReviewed-by: David Sterba \u003cdsterba@suse.com\u003e\nSigned-off-by: David Sterba \u003cdsterba@suse.com\u003e\n" }, { "commit": "3d6a3d3a2a7a3a60a824e7c04e95fd50dec57812", "tree": "b98d17c378e8ba7c57c27e7c22cca394070e6ec4", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Alain Volmat", "email": "alain.volmat@foss.st.com", "time": "Fri Feb 05 09:51:40 2021 +0100" }, "committer": { "name": "Wolfram Sang", "email": "wsa@kernel.org", "time": "Fri Feb 12 11:36:40 2021 +0100" }, "message": "i2c: stm32f7: fix configuration of the digital filter\n\nThe digital filter related computation are present in the driver\nhowever the programming of the filter within the IP is missing.\nThe maximum value for the DNF is wrong and should be 15 instead of 16.\n\nFixes: aeb068c57214 (\"i2c: i2c-stm32f7: add driver\")\n\nSigned-off-by: Alain Volmat \u003calain.volmat@foss.st.com\u003e\nSigned-off-by: Pierre-Yves MORDRET \u003cpierre-yves.mordret@foss.st.com\u003e\nSigned-off-by: Wolfram Sang \u003cwsa@kernel.org\u003e\n" }, { "commit": "551c81853d6d3ff016269d62612e7cd0a53104ab", "tree": "7637dde23d0c1283c34f7177ba34262f4c25c4d7", "parents": [ "0594bc74be82d0ce50e90de29475b331c1e83382", "811ee9dff58072742644da2c07641728f5e078e4" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 12 13:38:31 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 12 13:38:51 2021 +1000" }, "message": "Merge branch \u0027drm-misc-fixes\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\n(I\u0027ve pulled from a non-tag to get the ttm regression fix)\ndrm-misc-fixes-2021-02-10:\n * dp_mst: Don\u0027t report un-attached ports as connected\n * sun4i: tcon1 sync polarity fix; Always set HDMI clock rate; Fix\n H6 HDMI PHY config; Fix H6 max frequency\n * vc4: Fix buffer overflow\n * xlnx: Fix memory leak\n * ttm: page pool regression fix.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/YCPo6g3gDxD3P//h@linux-uq9g\n" }, { "commit": "245090ab2636c0869527ce563afbfb8aff29e825", "tree": "a6b227722db9fd92b18dd99be4328211d31b67a4", "parents": [ "c361c5a6c559d1e0a2717abe9162a71aa602954f" ], "author": { "name": "Jernej Skrabec", "email": "jernej.skrabec@siol.net", "time": "Tue Feb 09 18:58:56 2021 +0100" }, "committer": { "name": "Stephen Boyd", "email": "sboyd@kernel.org", "time": "Thu Feb 11 19:02:24 2021 -0800" }, "message": "clk: sunxi-ng: mp: fix parent rate change flag check\n\nCLK_SET_RATE_PARENT flag is checked on parent clock instead of current\none. Fix that.\n\nFixes: 3f790433c3cb (\"clk: sunxi-ng: Adjust MP clock parent rate when allowed\")\nReviewed-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\nTested-by: Andre Heider \u003ca.heider@gmail.com\u003e\nSigned-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nLink: https://lore.kernel.org/r/20210209175900.7092-2-jernej.skrabec@siol.net\nAcked-by: Maxime Ripard \u003cmripard@kernel.org\u003e\nSigned-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n" }, { "commit": "0594bc74be82d0ce50e90de29475b331c1e83382", "tree": "39ae546ab2206347ebad440a004a95c87ad3bb83", "parents": [ "d92d3d8f4cc6a9aa8b8f8c229d4b9c43d6eded40", "5feba0e905c495a217aea9db4ea91093d8fe5dde" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 12 10:16:58 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 12 10:16:59 2021 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2021-02-11\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\ndrm/i915 fixes for v5.11 final:\n- Ensure Type-C FIA is powered when initializing\n- Fix overlay frontbuffer tracking\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/87r1lnc78t.fsf@intel.com\n" }, { "commit": "d92d3d8f4cc6a9aa8b8f8c229d4b9c43d6eded40", "tree": "8355eb1c9c037d7392b2fd7218bba55f1ae7352b", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3", "cf050f96e0970a557601953ed7269d07a7885078" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 12 09:51:15 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 12 09:51:16 2021 +1000" }, "message": "Merge tag \u0027amd-drm-fixes-5.11-2021-02-10\u0027 of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes\n\namd-drm-fixes-5.11-2021-02-10:\n\namdgpu:\n- Blank screen fix\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexdeucher@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210210223508.4428-1-alexander.deucher@amd.com\n" }, { "commit": "dcc0b49040c70ad827a7f3d58a21b01fdb14e749", "tree": "4f8024e086e9f7bfeb8eeab052481a51593b90ac", "parents": [ "c05263df6c1ab82277cc3b2778b183b469a71a60", "8c511eff1827239f24ded212b1bcda7ca5b16203" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 11 15:41:07 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 11 15:41:07 2021 -0800" }, "message": "Merge tag \u0027powerpc-5.11-8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull powerpc fix from Michael Ellerman:\n \"One fix for a regression seen in io_uring, introduced by our support\n for KUAP (Kernel User Access Prevention) with the Hash MMU.\n\n Thanks to Aneesh Kumar K.V, and Zorro Lang\"\n\n* tag \u0027powerpc-5.11-8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n powerpc/kuap: Allow kernel thread to access userspace after kthread_use_mm\n" }, { "commit": "b220c049d5196dd94d992dd2dc8cba1a5e6123bf", "tree": "33d5cd03fa6e48c31266ae3104c0f2f1bc4ca60c", "parents": [ "256cfdd6fdf70c6fcf0f7c8ddb0ebd73ce8f3bc9" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Feb 10 11:53:22 2021 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Feb 11 14:23:37 2021 -0500" }, "message": "tracing: Check length before giving out the filter buffer\n\nWhen filters are used by trace events, a page is allocated on each CPU and\nused to copy the trace event fields to this page before writing to the ring\nbuffer. The reason to use the filter and not write directly into the ring\nbuffer is because a filter may discard the event and there\u0027s more overhead\non discarding from the ring buffer than the extra copy.\n\nThe problem here is that there is no check against the size being allocated\nwhen using this page. If an event asks for more than a page size while being\nfiltered, it will get only a page, leading to the caller writing more that\nwhat was allocated.\n\nCheck the length of the request, and if it is more than PAGE_SIZE minus the\nheader default back to allocating from the ring buffer directly. The ring\nbuffer may reject the event if its too big anyway, but it wont overflow.\n\nLink: https://lore.kernel.org/ath10k/1612839593-2308-1-git-send-email-wgong@codeaurora.org/\n\nCc: stable@vger.kernel.org\nFixes: 0fc1b09ff1ff4 (\"tracing: Use temp buffer when filtering events\")\nReported-by: Wen Gong \u003cwgong@codeaurora.org\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "c05263df6c1ab82277cc3b2778b183b469a71a60", "tree": "cdc210f8e2edd9d0be19e25657f2265d234747dd", "parents": [ "291009f656e8eaebbdfd3a8d99f6b190a9ce9deb", "28dc10eb77a2db7681b08e3b109764bbe469e347" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 11 11:21:08 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 11 11:21:08 2021 -0800" }, "message": "Merge tag \u0027gpio-fixes-for-v5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux\n\nPull gpio fixes from Bartosz Golaszewski:\n \"This is hopefully the last batch of fixes for this release cycle. We\n have a minor fix for a Kconfig regression as well as fixes for older\n bugs in gpio-ep93xx:\n\n - don\u0027t build gpio-mxs unconditionally with COMPILE_TEST enabled\n\n - fix two problems with interrupt handling in gpio-ep93xx\"\n\n* tag \u0027gpio-fixes-for-v5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:\n gpio: ep93xx: Fix single irqchip with multi gpiochips\n gpio: ep93xx: fix BUG_ON port F usage\n gpio: mxs: GPIO_MXS should not default to y unconditionally\n" }, { "commit": "29500f15b54b63ad0ea60b58e85144262bd24df2", "tree": "07fa6b1a005b09e3c66b1fbbbbc72246907ed33e", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 11 15:14:16 2021 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Fri Feb 12 03:02:21 2021 +0900" }, "message": "kbuild: fix CONFIG_TRIM_UNUSED_KSYMS build for ppc64\n\nStephen Rothwell reported a build error on ppc64 when\nCONFIG_TRIM_UNUSED_KSYMS is enabled.\n\nJessica Yu pointed out the cause of the error with the reference to the\nppc64 ELF ABI:\n \"Symbol names with a dot (.) prefix are reserved for holding entry\n point addresses. The value of a symbol named \".FN\", if it exists,\n is the entry point of the function \"FN\".\n\nAs it turned out, CONFIG_TRIM_UNUSED_KSYMS has never worked for ppc64,\nbut this issue has been unnoticed until recently because this option\ndepends on !UNUSED_SYMBOLS hence is disabled by all{mod,yes}config.\n(Then, it was uncovered by another patch removing UNUSED_SYMBOLS.)\n\nRemoving the dot prefix in scripts/gen_autoksyms.sh fixes the issue.\nPlease note it must be done before \u0027sort -u\u0027 because modules have\nboth ._mcount and _mcount undefined when CONFIG_FUNCTION_TRACER\u003dy.\n\nLink: https://lore.kernel.org/lkml/20210209210843.3af66662@canb.auug.org.au/\nReported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\nTested-by: Jessica Yu \u003cjeyu@kernel.org\u003e\n" }, { "commit": "a738c93fb1c17e386a09304b517b1c6b2a6a5a8b", "tree": "b852c26ae5873c4888bc11739ac4b80833227a93", "parents": [ "af1a3d2ba9543e99d78914d8fb88b61d0531d9a1" ], "author": { "name": "Shyam Prasad N", "email": "sprasad@microsoft.com", "time": "Thu Feb 11 03:26:54 2021 -0800" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Thu Feb 11 11:08:32 2021 -0600" }, "message": "cifs: Set CIFS_MOUNT_USE_PREFIX_PATH flag on setting cifs_sb-\u003eprepath.\n\nWhile debugging another issue today, Steve and I noticed that if a\nsubdir for a file share is already mounted on the client, any new\nmount of any other subdir (or the file share root) of the same share\nresults in sharing the cifs superblock, which e.g. can result in\nincorrect device name.\n\nWhile setting prefix path for the root of a cifs_sb,\nCIFS_MOUNT_USE_PREFIX_PATH flag should also be set.\nWithout it, prepath is not even considered in some places,\nand output of \"mount\" and various /proc/\u003c\u003e/*mount* related\noptions can be missing part of the device name.\n\nSigned-off-by: Shyam Prasad N \u003csprasad@microsoft.com\u003e\nReviewed-by: Aurelien Aptel \u003caaptel@suse.com\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "af1a3d2ba9543e99d78914d8fb88b61d0531d9a1", "tree": "60cccec8ae313f3d88142f0d49b85df25ee54038", "parents": [ "a0f85e38a3c23a334d9ea111cfa4fcfa4b3f9b64" ], "author": { "name": "Ronnie Sahlberg", "email": "lsahlber@redhat.com", "time": "Thu Feb 11 16:06:16 2021 +1000" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Thu Feb 11 10:58:08 2021 -0600" }, "message": "cifs: In the new mount api we get the full devname as source\u003d\n\nso we no longer need to handle or parse the UNC\u003d and prefixpath\u003d\noptions that mount.cifs are generating.\n\nThis also fixes a bug in the mount command option where the devname\nwould be truncated into just //server/share because we were looking\nat the truncated UNC value and not the full path.\n\nI.e. in the mount command output the devive //server/share/path\nwould show up as just //server/share\n\nSigned-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nReviewed-by: Shyam Prasad N \u003cnspmangalore@gmail.com\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "811ee9dff58072742644da2c07641728f5e078e4", "tree": "0289f4b7a313e23113b5729bcbcf792afd5ab5a2", "parents": [ "1926a0508d8947cf081280d85ff035300dc71da7" ], "author": { "name": "Christian König", "email": "christian.koenig@amd.com", "time": "Wed Feb 10 14:24:27 2021 +0100" }, "committer": { "name": "Christian König", "email": "christian.koenig@amd.com", "time": "Thu Feb 11 09:35:19 2021 +0100" }, "message": "drm/ttm: make sure pool pages are cleared\n\nThe old implementation wasn\u0027t consistend on this.\n\nBut it looks like we depend on this so better bring it back.\n\nSigned-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\nReported-and-tested-by: Mike Galbraith \u003cefault@gmx.de\u003e\nFixes: d099fc8f540a (\"drm/ttm: new TT backend allocation pool v3\")\nReviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210210160549.1462-1-christian.koenig@amd.com\n" }, { "commit": "c4295ab0b485b8bc50d2264bcae2acd06f25caaf", "tree": "e0bfcd215aeb0bdb33d0e20f58351ed9f61f39fe", "parents": [ "2e92493637a09547734f92c62a2471f6f0cb9a2c" ], "author": { "name": "Julien Grall", "email": "jgrall@amazon.com", "time": "Wed Feb 10 17:06:54 2021 +0000" }, "committer": { "name": "Juergen Gross", "email": "jgross@suse.com", "time": "Thu Feb 11 07:49:37 2021 +0100" }, "message": "arm/xen: Don\u0027t probe xenbus as part of an early initcall\n\nAfter Commit 3499ba8198cad (\"xen: Fix event channel callback via\nINTX/GSI\"), xenbus_probe() will be called too early on Arm. This will\nrecent to a guest hang during boot.\n\nIf the hang wasn\u0027t there, we would have ended up to call\nxenbus_probe() twice (the second time is in xenbus_probe_initcall()).\n\nWe don\u0027t need to initialize xenbus_probe() early for Arm guest.\nTherefore, the call in xen_guest_init() is now removed.\n\nAfter this change, there is no more external caller for xenbus_probe().\nSo the function is turned to a static one. Interestingly there were two\nprototypes for it.\n\nCc: stable@vger.kernel.org\nFixes: 3499ba8198cad (\"xen: Fix event channel callback via INTX/GSI\")\nReported-by: Ian Jackson \u003ciwj@xenproject.org\u003e\nSigned-off-by: Julien Grall \u003cjgrall@amazon.com\u003e\nReviewed-by: David Woodhouse \u003cdwmw@amazon.co.uk\u003e\nReviewed-by: Stefano Stabellini \u003csstabellini@kernel.org\u003e\nLink: https://lore.kernel.org/r/20210210170654.5377-1-julien@xen.org\nSigned-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n" }, { "commit": "3da3cc1b5f47115b16b5ffeeb4bf09ec331b0164", "tree": "774874a736a4dcd6ebdd99fd9df322f1afbbf8e3", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Thu Feb 04 19:41:12 2021 -0800" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Wed Feb 10 16:06:14 2021 -0800" }, "message": "Revert \"dts: phy: add GPIO number and active state used for phy reset\"\n\nVSC8541 phys need a special reset sequence, which the driver doesn\u0027t\ncurrentlny support. As a result enabling the reset via GPIO essentially\nguarnteees that the device won\u0027t work correctly. We\u0027ve been relying on\nbootloaders to reset the device for years, with this revert we\u0027ll go\nback to doing so until we can sort out how to get the reset sequence\ninto the kernel.\n\nThis reverts commit a0fa9d727043da2238432471e85de0bdb8a8df65.\n\nFixes: a0fa9d727043 (\"dts: phy: add GPIO number and active state used for phy reset\")\nCc: stable@vger.kernel.org\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "70245f86c109e0eafb92ea9653184c0e44b4b35c", "tree": "3342f2a2b3165f132b2ca0c7e71fd06fa7137101", "parents": [ "256b92af784d5043eeb7d559b6d5963dcc2ecb10" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Wed Feb 10 16:27:41 2021 +0100" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Wed Feb 10 22:06:47 2021 +0100" }, "message": "x86/pci: Create PCI/MSI irqdomain after x86_init.pci.arch_init()\n\nInvoking x86_init.irqs.create_pci_msi_domain() before\nx86_init.pci.arch_init() breaks XEN PV.\n\nThe XEN_PV specific pci.arch_init() function overrides the default\ncreate_pci_msi_domain() which is obviously too late.\n\nAs a consequence the XEN PV PCI/MSI allocation goes through the native\npath which runs out of vectors and causes malfunction.\n\nInvoke it after x86_init.pci.arch_init().\n\nFixes: 6b15ffa07dc3 (\"x86/irq: Initialize PCI/MSI domain at PCI init time\")\nReported-by: Juergen Gross \u003cjgross@suse.com\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nTested-by: Juergen Gross \u003cjgross@suse.com\u003e\nCc: stable@vger.kernel.org\nLink: https://lore.kernel.org/r/87pn18djte.fsf@nanos.tec.linutronix.de\n" }, { "commit": "291009f656e8eaebbdfd3a8d99f6b190a9ce9deb", "tree": "b0163945b7f1985345716fc0d4b077e0cbe2928d", "parents": [ "a3961497bd9c7ca94212922a46729a9410568eb8", "d11a1d08a082a7dc0ada423d2b2e26e9b6f2525c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 12:03:35 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 12:03:35 2021 -0800" }, "message": "Merge tag \u0027pm-5.11-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fixes from Rafael Wysocki:\n \"Address a performance regression related to scale-invariance on x86\n that may prevent turbo CPU frequencies from being used in certain\n workloads on systems using acpi-cpufreq as the CPU performance scaling\n driver and schedutil as the scaling governor\"\n\n* tag \u0027pm-5.11-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n cpufreq: ACPI: Update arch scale-invariance max perf ratio if CPPC is not there\n cpufreq: ACPI: Extend frequency tables to cover boost frequencies\n" }, { "commit": "a3961497bd9c7ca94212922a46729a9410568eb8", "tree": "bd5900e19f810b353a596930070b950f5bf4f441", "parents": [ "708c2e41814209e5dde27c61ad032f4c1ed3624b", "fe0af09074bfeb46a35357e67635eefe33cdfc49" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:58:21 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:58:21 2021 -0800" }, "message": "Merge tag \u0027acpi-5.11-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI fix from Rafael Wysocki:\n \"Revert a problematic ACPICA commit that changed the code to attempt to\n update memory regions which may be read-only on some systems (Ard\n Biesheuvel)\"\n\n* tag \u0027acpi-5.11-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n Revert \"ACPICA: Interpreter: fix memory leak by using existing buffer\"\n" }, { "commit": "708c2e41814209e5dde27c61ad032f4c1ed3624b", "tree": "9c901110d631faedb196145234fd3d46d374dd12", "parents": [ "6016bf19b3854b6e70ba9278a7ca0fce75278d3a", "b6c14d7a83802046f7098e9bae78fbde23affa74" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:51:25 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:51:25 2021 -0800" }, "message": "Merge tag \u0027dmaengine-fix2-5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine\n\nPull dmaengine fixes from Vinod Koul:\n \"Some late fixes for dmaengine:\n\n Core:\n - fix channel device_node deletion\n\n Driver fixes:\n - dw: revert of runtime pm enabling\n - idxd: device state fix, interrupt completion and list corruption\n - ti: resource leak\n\n* tag \u0027dmaengine-fix2-5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine:\n dmaengine dw: Revert \"dmaengine: dw: Enable runtime PM\"\n dmaengine: idxd: check device state before issue command\n dmaengine: ti: k3-udma: Fix a resource leak in an error handling path\n dmaengine: move channel device_node deletion to driver\n dmaengine: idxd: fix misc interrupt completion\n dmaengine: idxd: Fix list corruption in description completion\n" }, { "commit": "92c75f7594d5060a4cb240f0e987a802f8486b11", "tree": "3c21436557ed92c0dbaf8d03d772f527c0817c85", "parents": [ "aec18a57edad562d620f7d19016de1fc0cc2208c" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 10 12:37:58 2021 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 10 12:37:58 2021 -0700" }, "message": "Revert \"io_uring: don\u0027t take fs for recvmsg/sendmsg\"\n\nThis reverts commit 10cad2c40dcb04bb46b2bf399e00ca5ea93d36b0.\n\nPetr reports that with this commit in place, io_uring fails the chroot\ntest (CVE-202-29373). We do need to retain -\u003efs for send/recvmsg, so\nrevert this commit.\n\nReported-by: Petr Vorel \u003cpvorel@suse.cz\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "6016bf19b3854b6e70ba9278a7ca0fce75278d3a", "tree": "0e1951f6db284df503eccac94df34c89b5db396f", "parents": [ "4b16b656b1ce04868a31af65c846cf97823d32c5", "b8776f14a47046796fe078c4a2e691f58e00ae06" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:33:39 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:33:39 2021 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n \"Another pile of networing fixes:\n\n 1) ath9k build error fix from Arnd Bergmann\n\n 2) dma memory leak fix in mediatec driver from Lorenzo Bianconi.\n\n 3) bpf int3 kprobe fix from Alexei Starovoitov.\n\n 4) bpf stackmap integer overflow fix from Bui Quang Minh.\n\n 5) Add usb device ids for Cinterion MV31 to qmi_qwwan driver, from\n Christoph Schemmel.\n\n 6) Don\u0027t update deleted entry in xt_recent netfilter module, from\n Jazsef Kadlecsik.\n\n 7) Use after free in nftables, fix from Pablo Neira Ayuso.\n\n 8) Header checksum fix in flowtable from Sven Auhagen.\n\n 9) Validate user controlled length in qrtr code, from Sabyrzhan\n Tasbolatov.\n\n 10) Fix race in xen/netback, from Juergen Gross,\n\n 11) New device ID in cxgb4, from Raju Rangoju.\n\n 12) Fix ring locking in rxrpc release call, from David Howells.\n\n 13) Don\u0027t return LAPB error codes from x25_open(), from Xie He.\n\n 14) Missing error returns in gsi_channel_setup() from Alex Elder.\n\n 15) Get skb_copy_and_csum_datagram working properly with odd segment\n sizes, from Willem de Bruijn.\n\n 16) Missing RFS/RSS table init in enetc driver, from Vladimir Oltean.\n\n 17) Do teardown on probe failure in DSA, from Vladimir Oltean.\n\n 18) Fix compilation failures of txtimestamp selftest, from Vadim\n Fedorenko.\n\n 19) Limit rx per-napi gro queue size to fix latency regression, from\n Eric Dumazet.\n\n 20) dpaa_eth xdp fixes from Camelia Groza.\n\n 21) Missing txq mode update when switching CBS off, in stmmac driver,\n from Mohammad Athari Bin Ismail.\n\n 22) Failover pending logic fix in ibmvnic driver, from Sukadev\n Bhattiprolu.\n\n 23) Null deref fix in vmw_vsock, from Norbert Slusarek.\n\n 24) Missing verdict update in xdp paths of ena driver, from Shay\n Agroskin.\n\n 25) seq_file iteration fix in sctp from Neil Brown.\n\n 26) bpf 32-bit src register truncation fix on div/mod, from Daniel\n Borkmann.\n\n 27) Fix jmp32 pruning in bpf verifier, from Daniel Borkmann.\n\n 28) Fix locking in vsock_shutdown(), from Stefano Garzarella.\n\n 29) Various missing index bound checks in hns3 driver, from Yufeng Mo.\n\n 30) Flush ports on .phylink_mac_link_down() in dsa felix driver, from\n Vladimir Oltean.\n\n 31) Don\u0027t mix up stp and mrp port states in bridge layer, from Horatiu\n Vultur.\n\n 32) Fix locking during netif_tx_disable(), from Edwin Peer\"\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (45 commits)\n bpf: Fix 32 bit src register truncation on div/mod\n bpf: Fix verifier jmp32 pruning decision logic\n bpf: Fix verifier jsgt branch analysis on max bound\n vsock: fix locking in vsock_shutdown()\n net: hns3: add a check for index in hclge_get_rss_key()\n net: hns3: add a check for tqp_index in hclge_get_ring_chain_from_mbx()\n net: hns3: add a check for queue_id in hclge_reset_vf_queue()\n net: dsa: felix: implement port flushing on .phylink_mac_link_down\n switchdev: mrp: Remove SWITCHDEV_ATTR_ID_MRP_PORT_STAT\n bridge: mrp: Fix the usage of br_mrp_port_switchdev_set_state\n net: watchdog: hold device global xmit lock during tx disable\n netfilter: nftables: relax check for stateful expressions in set definition\n netfilter: conntrack: skip identical origin tuple in same zone only\n vsock/virtio: update credit only if socket is not closed\n net: fix iteration for sctp transport seq_files\n net: ena: Update XDP verdict upon failure\n net/vmw_vsock: improve locking in vsock_connect_timeout()\n net/vmw_vsock: fix NULL pointer dereference\n ibmvnic: Clear failover_pending if unable to schedule\n net: stmmac: set TxQ mode back to DCB after disabling CBS\n ...\n" }, { "commit": "4b16b656b1ce04868a31af65c846cf97823d32c5", "tree": "e33fa759205db84d23035e945830c1e8c4556285", "parents": [ "ef7d0b599938450c54a8dc0aa4b954d73d9a9370", "a35d8f016e0b68634035217d06d1c53863456b50" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:22:41 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:22:41 2021 -0800" }, "message": "Merge branch \u0027akpm\u0027 (patches from Andrew)\n\nMerge misc fixes from Andrew Morton:\n \"14 patches.\n\n Subsystems affected by this patch series: mm (kasan, mremap, tmpfs,\n selftests, memcg, and slub), MAINTAINERS, squashfs, nilfs2, and\n firmware\"\n\n* emailed patches from Andrew Morton \u003cakpm@linux-foundation.org\u003e:\n nilfs2: make splice write available again\n mm, slub: better heuristic for number of cpus when calculating slab order\n Revert \"mm: memcontrol: avoid workload stalls when lowering memory.high\"\n MAINTAINERS: update Andrey Ryabinin\u0027s email address\n selftests/vm: rename file run_vmtests to run_vmtests.sh\n tmpfs: disallow CONFIG_TMPFS_INODE64 on alpha\n tmpfs: disallow CONFIG_TMPFS_INODE64 on s390\n mm/mremap: fix BUILD_BUG_ON() error in get_extent\n firmware_loader: align .builtin_fw to 8\n kasan: fix stack traces dependency for HW_TAGS\n squashfs: add more sanity checks in xattr id lookup\n squashfs: add more sanity checks in inode lookup\n squashfs: add more sanity checks in id lookup\n squashfs: avoid out of bounds writes in decompressors\n" }, { "commit": "a35d8f016e0b68634035217d06d1c53863456b50", "tree": "480e16f85f9f96d920af45b2a9bff3997bd7872f", "parents": [ "3286222fc609dea27bd16ac02c55d3f1c3190063" ], "author": { "name": "Joachim Henke", "email": "joachim.henke@t-systems.com", "time": "Tue Feb 09 13:42:36 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:19:58 2021 -0800" }, "message": "nilfs2: make splice write available again\n\nSince 5.10, splice() or sendfile() to NILFS2 return EINVAL. This was\ncaused by commit 36e2c7421f02 (\"fs: don\u0027t allow splice read/write\nwithout explicit ops\").\n\nThis patch initializes the splice_write field in file_operations, like\nmost file systems do, to restore the functionality.\n\nLink: https://lkml.kernel.org/r/1612784101-14353-1-git-send-email-konishi.ryusuke@gmail.com\nSigned-off-by: Joachim Henke \u003cjoachim.henke@t-systems.com\u003e\nSigned-off-by: Ryusuke Konishi \u003ckonishi.ryusuke@gmail.com\u003e\nTested-by: Ryusuke Konishi \u003ckonishi.ryusuke@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\t[5.10+]\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3286222fc609dea27bd16ac02c55d3f1c3190063", "tree": "5306451a2a2d8b8da01d1829b44a9fd4689720a7", "parents": [ "e82553c10b0899994153f9bf0af333c0a1550fd7" ], "author": { "name": "Vlastimil Babka", "email": "vbabka@suse.cz", "time": "Tue Feb 09 13:42:32 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 10 11:19:27 2021 -0800" }, "message": "mm, slub: better heuristic for number of cpus when calculating slab order\n\nWhen creating a new kmem cache, SLUB determines how large the slab pages\nwill based on number of inputs, including the number of CPUs in the\nsystem. Larger slab pages mean that more objects can be allocated/free\nfrom per-cpu slabs before accessing shared structures, but also\npotentially more memory can be wasted due to low slab usage and\nfragmentation. The rough idea of using number of CPUs is that larger\nsystems will be more likely to benefit from reduced contention, and also\nshould have enough memory to spare.\n\nNumber of CPUs used to be determined as nr_cpu_ids, which is number of\npossible cpus, but on some systems many will never be onlined, thus\ncommit 045ab8c9487b (\"mm/slub: let number of online CPUs determine the\nslub page order\") changed it to nr_online_cpus(). However, for kmem\ncaches created early before CPUs are onlined, this may lead to\npermamently low slab page sizes.\n\nVincent reports a regression [1] of hackbench on arm64 systems:\n\n \"I\u0027m facing significant performances regression on a large arm64\n server system (224 CPUs). Regressions is also present on small arm64\n system (8 CPUs) but in a far smaller order of magnitude\n\n On 224 CPUs system : 9 iterations of hackbench -l 16000 -g 16\n v5.11-rc4 : 9.135sec (+/- 0.45%)\n v5.11-rc4 + revert this patch: 3.173sec (+/- 0.48%)\n v5.10: 3.136sec (+/- 0.40%)\"\n\nMel reports a regression [2] of hackbench on x86_64, with lockstat suggesting\npage allocator contention:\n\n \"i.e. the patch incurs a 7% to 32% performance penalty. This bisected\n cleanly yesterday when I was looking for the regression and then\n found the thread.\n\n Numerous caches change size. For example, kmalloc-512 goes from\n order-0 (vanilla) to order-2 with the revert.\n\n So mostly this is down to the number of times SLUB calls into the\n page allocator which only caches order-0 pages on a per-cpu basis\"\n\nClearly num_online_cpus() doesn\u0027t work too early in bootup. We could\nchange the order dynamically in a memory hotplug callback, but runtime\norder changing for existing kmem caches has been already shown as\ndangerous, and removed in 32a6f409b693 (\"mm, slub: remove runtime\nallocation order changes\").\n\nIt could be resurrected in a safe manner with some effort, but to fix\nthe regression we need something simpler.\n\nWe could use num_present_cpus() that should be the number of physically\npresent CPUs even before they are onlined. That would work for PowerPC\n[3], which triggered the original commit, but that still doesn\u0027t work on\narm64 [4] as explained in [5].\n\nSo this patch tries to determine the best available value without\nspecific arch knowledge.\n\n - num_present_cpus() if the number is larger than 1, as that means the\n arch is likely setting it properly\n\n - nr_cpu_ids otherwise\n\nThis should fix the reported regressions while also keeping the effect\nof 045ab8c9487b for PowerPC systems. It\u0027s possible there are\nconfigurations where num_present_cpus() is 1 during boot while\nnr_cpu_ids is at the same time bloated, so these (if they exist) would\nkeep the large orders based on nr_cpu_ids as was before 045ab8c9487b.\n\n[1] https://lore.kernel.org/linux-mm/CAKfTPtA_JgMf_+zdFbcb_V9rM7JBWNPjAz9irgwFj7Rou\u003dxzZg@mail.gmail.com/\n[2] https://lore.kernel.org/linux-mm/20210128134512.GF3592@techsingularity.net/\n[3] https://lore.kernel.org/linux-mm/20210123051607.GC2587010@in.ibm.com/\n[4] https://lore.kernel.org/linux-mm/CAKfTPtAjyVmS5VYvU6DBxg4-JEo5bdmWbngf-03YsY18cmWv_g@mail.gmail.com/\n[5] https://lore.kernel.org/linux-mm/20210126230305.GD30941@willie-the-truck/\n\nLink: https://lkml.kernel.org/r/20210208134108.22286-1-vbabka@suse.cz\nFixes: 045ab8c9487b (\"mm/slub: let number of online CPUs determine the slub page order\")\nSigned-off-by: Vlastimil Babka \u003cvbabka@suse.cz\u003e\nReported-by: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\nReported-by: Mel Gorman \u003cmgorman@techsingularity.net\u003e\nTested-by: Mel Gorman \u003cmgorman@techsingularity.net\u003e\nTested-by: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\nCc: Bharata B Rao \u003cbharata@linux.ibm.com\u003e\nCc: Christoph Lameter \u003ccl@linux.com\u003e\nCc: Roman Gushchin \u003cguro@fb.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Joonsoo Kim \u003ciamjoonsoo.kim@lge.com\u003e\nCc: Jann Horn \u003cjannh@google.com\u003e\nCc: Michal Hocko \u003cmhocko@kernel.org\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "28dc10eb77a2db7681b08e3b109764bbe469e347", "tree": "4e96e21d92eba2d6ff1b0b67afdb14de98a35648", "parents": [ "8b81a7ab8055d01d827ef66374b126eeac3bd108" ], "author": { "name": "Nikita Shubin", "email": "nikita.shubin@maquefel.me", "time": "Tue Feb 09 16:31:05 2021 +0300" }, "committer": { "name": "Bartosz Golaszewski", "email": "bgolaszewski@baylibre.com", "time": "Wed Feb 10 14:47:27 2021 +0100" }, "message": "gpio: ep93xx: Fix single irqchip with multi gpiochips\n\nFixes the following warnings which results in interrupts disabled on\nport B/F:\n\ngpio gpiochip1: (B): detected irqchip that is shared with multiple gpiochips: please fix the driver.\ngpio gpiochip5: (F): detected irqchip that is shared with multiple gpiochips: please fix the driver.\n\n- added separate irqchip for each interrupt capable gpiochip\n- provided unique names for each irqchip\n\nFixes: d2b091961510 (\"gpio: ep93xx: Pass irqchip when adding gpiochip\")\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Nikita Shubin \u003cnikita.shubin@maquefel.me\u003e\nTested-by: Alexander Sverdlin \u003calexander.sverdlin@gmail.com\u003e\nSigned-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n" }, { "commit": "8b81a7ab8055d01d827ef66374b126eeac3bd108", "tree": "a056453b1e8f35f76b504f6f43b8a970438b8289", "parents": [ "97c6e28d388a5000d780d2a63c32f422827f5aa3" ], "author": { "name": "Nikita Shubin", "email": "nikita.shubin@maquefel.me", "time": "Tue Feb 09 16:31:04 2021 +0300" }, "committer": { "name": "Bartosz Golaszewski", "email": "bgolaszewski@baylibre.com", "time": "Wed Feb 10 14:47:16 2021 +0100" }, "message": "gpio: ep93xx: fix BUG_ON port F usage\n\nTwo index spaces and ep93xx_gpio_port are confusing.\n\nInstead add a separate struct to store necessary data and remove\nep93xx_gpio_port.\n\n- add struct to store IRQ related data for each IRQ capable chip\n- replace offset array with defined offsets\n- add IRQ registers offset for each IRQ capable chip into\n ep93xx_gpio_banks\n\n------------[ cut here ]------------\nkernel BUG at drivers/gpio/gpio-ep93xx.c:64!\n---[ end trace 3f6544e133e9f5ae ]---\n\nFixes: fd935fc421e74 (\"gpio: ep93xx: Do not pingpong irq numbers\")\nCc: \u003cstable@vger.kernel.org\u003e\nReviewed-by: Alexander Sverdlin \u003calexander.sverdlin@gmail.com\u003e\nTested-by: Alexander Sverdlin \u003calexander.sverdlin@gmail.com\u003e\nSigned-off-by: Nikita Shubin \u003cnikita.shubin@maquefel.me\u003e\nSigned-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n" }, { "commit": "97c6e28d388a5000d780d2a63c32f422827f5aa3", "tree": "a35228293136909bde1434880cd6878050bd2318", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Geert Uytterhoeven", "email": "geert+renesas@glider.be", "time": "Mon Feb 08 15:51:53 2021 +0100" }, "committer": { "name": "Bartosz Golaszewski", "email": "bgolaszewski@baylibre.com", "time": "Wed Feb 10 14:25:59 2021 +0100" }, "message": "gpio: mxs: GPIO_MXS should not default to y unconditionally\n\nMerely enabling CONFIG_COMPILE_TEST should not enable additional code.\nTo fix this, restrict the automatic enabling of GPIO_MXS to ARCH_MXS,\nand ask the user in case of compile-testing.\n\nFixes: 6876ca311bfca5d7 (\"gpio: mxs: add COMPILE_TEST support for GPIO_MXS\")\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nSigned-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n" }, { "commit": "1926a0508d8947cf081280d85ff035300dc71da7", "tree": "58c47c9726f49024dc12ddac8d7e728c524227ff", "parents": [ "6a155216c48f2f65c8dcb02c4c27549c170d24a9" ], "author": { "name": "Jernej Skrabec", "email": "jernej.skrabec@siol.net", "time": "Tue Feb 09 18:59:00 2021 +0100" }, "committer": { "name": "Maxime Ripard", "email": "maxime@cerno.tech", "time": "Wed Feb 10 11:20:38 2021 +0100" }, "message": "drm/sun4i: dw-hdmi: Fix max. frequency for H6\n\nIt turns out that reasoning for lowering max. supported frequency is\nwrong. Scrambling works just fine. Several now fixed bugs prevented\nproper functioning, even with rates lower than 340 MHz. Issues were just\nmore pronounced with higher frequencies.\n\nFix that by allowing max. supported frequency in HW and fix the comment.\n\nFixes: cd9063757a22 (\"drm/sun4i: DW HDMI: Lower max. supported rate for H6\")\nReviewed-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\nTested-by: Andre Heider \u003ca.heider@gmail.com\u003e\nSigned-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nSigned-off-by: Maxime Ripard \u003cmaxime@cerno.tech\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210209175900.7092-6-jernej.skrabec@siol.net\n" }, { "commit": "6a155216c48f2f65c8dcb02c4c27549c170d24a9", "tree": "ef4522891f2bf21d851436266254ca538a3c7004", "parents": [ "36b53581fe0dc2e25b67de4e58920307f22d195a" ], "author": { "name": "Jernej Skrabec", "email": "jernej.skrabec@siol.net", "time": "Tue Feb 09 18:58:59 2021 +0100" }, "committer": { "name": "Maxime Ripard", "email": "maxime@cerno.tech", "time": "Wed Feb 10 11:20:13 2021 +0100" }, "message": "drm/sun4i: Fix H6 HDMI PHY configuration\n\nAs it turns out, vendor HDMI PHY driver for H6 has a pretty big table\nof predefined values for various pixel clocks. However, most of them are\nnot useful/tested because they come from reference driver code. Vendor\nPHY driver is concerned with only few of those, namely 27 MHz, 74.25\nMHz, 148.5 MHz, 297 MHz and 594 MHz. These are all frequencies for\nstandard CEA modes.\n\nFix sun50i_h6_cur_ctr and sun50i_h6_phy_config with the values only for\naforementioned frequencies.\n\nTable sun50i_h6_mpll_cfg doesn\u0027t need to be changed because values are\nactually frequency dependent and not so much SoC dependent. See i.MX6\ndocumentation for explanation of those values for similar PHY.\n\nFixes: c71c9b2fee17 (\"drm/sun4i: Add support for Synopsys HDMI PHY\")\nTested-by: Andre Heider \u003ca.heider@gmail.com\u003e\nSigned-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nSigned-off-by: Maxime Ripard \u003cmaxime@cerno.tech\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210209175900.7092-5-jernej.skrabec@siol.net\n" }, { "commit": "36b53581fe0dc2e25b67de4e58920307f22d195a", "tree": "271423ed36fdba8685ed752422da69226a454a4a", "parents": [ "50791f5d7b6a14b388f46c8885f71d1b98216d1d" ], "author": { "name": "Jernej Skrabec", "email": "jernej.skrabec@siol.net", "time": "Tue Feb 09 18:58:58 2021 +0100" }, "committer": { "name": "Maxime Ripard", "email": "maxime@cerno.tech", "time": "Wed Feb 10 11:20:01 2021 +0100" }, "message": "drm/sun4i: dw-hdmi: always set clock rate\n\nAs expected, HDMI controller clock should always match pixel clock. In\nthe past, changing HDMI controller rate would seemingly worsen\nsituation. However, that was the result of other bugs which are now\nfixed.\n\nFix that by removing set_rate quirk and always set clock rate.\n\nFixes: 40bb9d3147b2 (\"drm/sun4i: Add support for H6 DW HDMI controller\")\nReviewed-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\nTested-by: Andre Heider \u003ca.heider@gmail.com\u003e\nSigned-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nSigned-off-by: Maxime Ripard \u003cmaxime@cerno.tech\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210209175900.7092-4-jernej.skrabec@siol.net\n" }, { "commit": "50791f5d7b6a14b388f46c8885f71d1b98216d1d", "tree": "84da1bd23227a2b8295a8b69148fa3b40f717d07", "parents": [ "873e5bb9fbd99e4a26c448b5c7af942a6d7aa60d" ], "author": { "name": "Jernej Skrabec", "email": "jernej.skrabec@siol.net", "time": "Tue Feb 09 18:58:57 2021 +0100" }, "committer": { "name": "Maxime Ripard", "email": "maxime@cerno.tech", "time": "Wed Feb 10 11:19:56 2021 +0100" }, "message": "drm/sun4i: tcon: set sync polarity for tcon1 channel\n\nChannel 1 has polarity bits for vsync and hsync signals but driver never\nsets them. It turns out that with pre-HDMI2 controllers seemingly there\nis no issue if polarity is not set. However, with HDMI2 controllers\n(H6) there often comes to de-synchronization due to phase shift. This\ncauses flickering screen. It\u0027s safe to assume that similar issues might\nhappen also with pre-HDMI2 controllers.\n\nSolve issue with setting vsync and hsync polarity. Note that display\nstacks with tcon top have polarity bits actually in tcon0 polarity\nregister.\n\nFixes: 9026e0d122ac (\"drm: Add Allwinner A10 Display Engine support\")\nReviewed-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\nTested-by: Andre Heider \u003ca.heider@gmail.com\u003e\nSigned-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nSigned-off-by: Maxime Ripard \u003cmaxime@cerno.tech\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210209175900.7092-3-jernej.skrabec@siol.net\n" }, { "commit": "5feba0e905c495a217aea9db4ea91093d8fe5dde", "tree": "d60a0d34ab21569d35f4dd451f3b3be401178174", "parents": [ "2f51312bebb77962a518b4c6de777dd378b6110a" ], "author": { "name": "Ville Syrjälä", "email": "ville.syrjala@linux.intel.com", "time": "Tue Feb 09 04:19:17 2021 +0200" }, "committer": { "name": "Jani Nikula", "email": "jani.nikula@intel.com", "time": "Wed Feb 10 11:03:56 2021 +0200" }, "message": "drm/i915: Fix overlay frontbuffer tracking\n\nWe don\u0027t have a persistent fb holding a reference to the frontbuffer\nobject, so every time we do the get+put we throw the frontbuffer object\nimmediately away. And so the next time around we get a pristine\nfrontbuffer object with bits\u003d\u003d0 even for the old vma. This confuses\nthe frontbuffer tracking code which understandably expects the old\nfrontbuffer to have the overlay\u0027s bit set.\n\nFix this by hanging on to the frontbuffer reference until the next\nflip. And just to make this a bit more clear let\u0027s track the frontbuffer\nexplicitly instead of just grabbing it via the old vma.\n\nCc: stable@vger.kernel.org\nCc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nCloses: https://gitlab.freedesktop.org/drm/intel/-/issues/1136\nSigned-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210209021918.16234-2-ville.syrjala@linux.intel.com\nFixes: 8e7cb1799b4f (\"drm/i915: Extract intel_frontbuffer active tracking\")\nReviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n(cherry picked from commit 553c23bdb4775130f333f07a51b047276bc53f79)\nSigned-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n" }, { "commit": "cf050f96e0970a557601953ed7269d07a7885078", "tree": "8355eb1c9c037d7392b2fd7218bba55f1ae7352b", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Feb 03 14:03:50 2021 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue Feb 09 23:23:18 2021 -0500" }, "message": "Revert \"drm/amd/display: Update NV1x SR latency values\"\n\nThis reverts commit 4a3dea8932d3b1199680d2056dd91d31d94d70b7.\n\nThis causes blank screens for some users.\n\nBug: https://gitlab.freedesktop.org/drm/amd/-/issues/1482\nCc: Alvin Lee \u003calvin.lee2@amd.com\u003e\nCc: Jun Lei \u003cJun.Lei@amd.com\u003e\nCc: Rodrigo Siqueira \u003cRodrigo.Siqueira@amd.com\u003e\nReviewed-by: Rodrigo Siqueira \u003cRodrigo.Siqueira@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "b8776f14a47046796fe078c4a2e691f58e00ae06", "tree": "4dba5bc10eaf44a6d97223f0ec8a3972d9f8a772", "parents": [ "450bbc3395185963b133177a9ddae305dae7c5fa", "e88b2c6e5a4d9ce30d75391e4d950da74bb2bd90" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 18:55:17 2021 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 18:55:17 2021 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf\n\nDaniel Borkmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\npull-request: bpf 2021-02-10\n\nThe following pull-request contains BPF updates for your *net* tree.\n\nWe\u0027ve added 5 non-merge commits during the last 8 day(s) which contain\na total of 3 files changed, 22 insertions(+), 21 deletions(-).\n\nThe main changes are:\n\n1) Fix missed execution of kprobes BPF progs when kprobe is firing via\n int3, from Alexei Starovoitov.\n\n2) Fix potential integer overflow in map max_entries for stackmap on\n 32 bit archs, from Bui Quang Minh.\n\n3) Fix a verifier pruning and a insn rewrite issue related to 32 bit ops,\n from Daniel Borkmann.\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\nc# Please enter a commit message to explain why this merge is necessary,\n" }, { "commit": "a0f85e38a3c23a334d9ea111cfa4fcfa4b3f9b64", "tree": "51dc7e355c709a5c26c7161b1e3811c879f4f749", "parents": [ "abd4af47d3fb64157133bd35de25d39a21a31122" ], "author": { "name": "Ronnie Sahlberg", "email": "lsahlber@redhat.com", "time": "Wed Feb 10 11:55:47 2021 +1000" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Tue Feb 09 20:47:05 2021 -0600" }, "message": "cifs: do not disable noperm if multiuser mount option is not provided\n\nFixes small regression in implementation of new mount API.\n\nSigned-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nReported-by: Hyunchul Lee \u003chyc.lee@gmail.com\u003e\nTested-by: Hyunchul Lee \u003chyc.lee@gmail.com\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "e82553c10b0899994153f9bf0af333c0a1550fd7", "tree": "cea923c6d152eff6bbd9a178f0f19e17df6eba9f", "parents": [ "a0c2eb0a4387322ebc629c01f5adb2d957c343fe" ], "author": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Tue Feb 09 13:42:28 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "Revert \"mm: memcontrol: avoid workload stalls when lowering memory.high\"\n\nThis reverts commit 536d3bf261a2fc3b05b3e91e7eef7383443015cf, as it can\ncause writers to memory.high to get stuck in the kernel forever,\nperforming page reclaim and consuming excessive amounts of CPU cycles.\n\nBefore the patch, a write to memory.high would first put the new limit\nin place for the workload, and then reclaim the requested delta. After\nthe patch, the kernel tries to reclaim the delta before putting the new\nlimit into place, in order to not overwhelm the workload with a sudden,\nlarge excess over the limit. However, if reclaim is actively racing\nwith new allocations from the uncurbed workload, it can keep the write()\nworking inside the kernel indefinitely.\n\nThis is causing problems in Facebook production. A privileged\nsystem-level daemon that adjusts memory.high for various workloads\nrunning on a host can get unexpectedly stuck in the kernel and\nessentially turn into a sort of involuntary kswapd for one of the\nworkloads. We\u0027ve observed that daemon busy-spin in a write() for\nminutes at a time, neglecting its other duties on the system, and\nexpending privileged system resources on behalf of a workload.\n\nTo remedy this, we have first considered changing the reclaim logic to\nbreak out after a couple of loops - whether the workload has converged\nto the new limit or not - and bound the write() call this way. However,\nthe root cause that inspired the sequence change in the first place has\nbeen fixed through other means, and so a revert back to the proven\nlimit-setting sequence, also used by memory.max, is preferable.\n\nThe sequence was changed to avoid extreme latencies in the workload when\nthe limit was lowered: the sudden, large excess created by the limit\nlowering would erroneously trigger the penalty sleeping code that is\nmeant to throttle excessive growth from below. Allocating threads could\nend up sleeping long after the write() had already reclaimed the delta\nfor which they were being punished.\n\nHowever, erroneous throttling also caused problems in other scenarios at\naround the same time. This resulted in commit b3ff92916af3 (\"mm, memcg:\nreclaim more aggressively before high allocator throttling\"), included\nin the same release as the offending commit. When allocating threads\nnow encounter large excess caused by a racing write() to memory.high,\ninstead of entering punitive sleeps, they will simply be tasked with\nhelping reclaim down the excess, and will be held no longer than it\ntakes to accomplish that. This is in line with regular limit\nenforcement - i.e. if the workload allocates up against or over an\notherwise unchanged limit from below.\n\nWith the patch breaking userspace, and the root cause addressed by other\nmeans already, revert it again.\n\nLink: https://lkml.kernel.org/r/20210122184341.292461-1-hannes@cmpxchg.org\nFixes: 536d3bf261a2 (\"mm: memcontrol: avoid workload stalls when lowering memory.high\")\nSigned-off-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nReported-by: Tejun Heo \u003ctj@kernel.org\u003e\nAcked-by: Chris Down \u003cchris@chrisdown.name\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: Roman Gushchin \u003cguro@fb.com\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Michal Koutný \u003cmkoutny@suse.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\t[5.8+]\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a0c2eb0a4387322ebc629c01f5adb2d957c343fe", "tree": "10b288c2ac17ad4ca65f2ca9e5c3f64744c70dd7", "parents": [ "d52db800846f66d98a4e14c39cf88a06bcd9985f" ], "author": { "name": "Andrey Ryabinin", "email": "ryabinin.a.a@gmail.com", "time": "Tue Feb 09 13:42:24 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "MAINTAINERS: update Andrey Ryabinin\u0027s email address\n\nUpdate my email, @virtuozzo.com will stop working shortly.\n\nLink: https://lkml.kernel.org/r/20210204223904.3824-1-ryabinin.a.a@gmail.com\nSigned-off-by: Andrey Ryabinin \u003cryabinin.a.a@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d52db800846f66d98a4e14c39cf88a06bcd9985f", "tree": "51dfb1c0740e31eff7eff474c4991b55be600215", "parents": [ "ad69c389ec110ea54f8b0c0884b255340ef1c736" ], "author": { "name": "Rong Chen", "email": "rong.a.chen@intel.com", "time": "Tue Feb 09 13:42:21 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "selftests/vm: rename file run_vmtests to run_vmtests.sh\n\nCommit c2aa8afc36fa has renamed run_vmtests in Makefile, but the file\nstill uses the old name.\n\nThe kernel test robot reported the following issue:\n\n # selftests: vm: run_vmtests.sh\n # Warning: file run_vmtests.sh is missing!\n not ok 1 selftests: vm: run_vmtests.sh\n\nLink: https://lkml.kernel.org/r/20210205085507.1479894-1-rong.a.chen@intel.com\nFixes: c2aa8afc36fa (selftests/vm: rename run_vmtests --\u003e run_vmtests.sh)\nSigned-off-by: Rong Chen \u003crong.a.chen@intel.com\u003e\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nReviewed-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ad69c389ec110ea54f8b0c0884b255340ef1c736", "tree": "1db493abac355579651b8f0b701467b4e42e8895", "parents": [ "b85a7a8bb5736998b8a681937a9749b350c17988" ], "author": { "name": "Seth Forshee", "email": "seth.forshee@canonical.com", "time": "Tue Feb 09 13:42:17 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "tmpfs: disallow CONFIG_TMPFS_INODE64 on alpha\n\nAs with s390, alpha is a 64-bit architecture with a 32-bit ino_t. With\nCONFIG_TMPFS_INODE64\u003dy tmpfs mounts will get 64-bit inode numbers and\ndisplay \"inode64\" in the mount options, whereas passing \"inode64\" in the\nmount options will fail. This leads to erroneous behaviours such as\nthis:\n\n # mkdir mnt\n # mount -t tmpfs nodev mnt\n # mount -o remount,rw mnt\n mount: /home/ubuntu/mnt: mount point not mounted or bad option.\n\nPrevent CONFIG_TMPFS_INODE64 from being selected on alpha.\n\nLink: https://lkml.kernel.org/r/20210208215726.608197-1-seth.forshee@canonical.com\nFixes: ea3271f7196c (\"tmpfs: support 64-bit inums per-sb\")\nSigned-off-by: Seth Forshee \u003cseth.forshee@canonical.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Chris Down \u003cchris@chrisdown.name\u003e\nCc: Amir Goldstein \u003camir73il@gmail.com\u003e\nCc: Richard Henderson \u003crth@twiddle.net\u003e\nCc: Ivan Kokshaysky \u003cink@jurassic.park.msu.ru\u003e\nCc: Matt Turner \u003cmattst88@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\t[5.9+]\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b85a7a8bb5736998b8a681937a9749b350c17988", "tree": "4f74c12e21f16caafb680a083c9bf4147c1c6a30", "parents": [ "a30a29091b5a6d4c64b5fc77040720a65e2dd4e6" ], "author": { "name": "Seth Forshee", "email": "seth.forshee@canonical.com", "time": "Tue Feb 09 13:42:14 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "tmpfs: disallow CONFIG_TMPFS_INODE64 on s390\n\nCurrently there is an assumption in tmpfs that 64-bit architectures also\nhave a 64-bit ino_t. This is not true on s390 which has a 32-bit ino_t.\nWith CONFIG_TMPFS_INODE64\u003dy tmpfs mounts will get 64-bit inode numbers\nand display \"inode64\" in the mount options, but passing the \"inode64\"\nmount option will fail. This leads to the following behavior:\n\n # mkdir mnt\n # mount -t tmpfs nodev mnt\n # mount -o remount,rw mnt\n mount: /home/ubuntu/mnt: mount point not mounted or bad option.\n\nAs mount sees \"inode64\" in the mount options and thus passes it in the\noptions for the remount.\n\nSo prevent CONFIG_TMPFS_INODE64 from being selected on s390.\n\nLink: https://lkml.kernel.org/r/20210205230620.518245-1-seth.forshee@canonical.com\nFixes: ea3271f7196c (\"tmpfs: support 64-bit inums per-sb\")\nSigned-off-by: Seth Forshee \u003cseth.forshee@canonical.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Chris Down \u003cchris@chrisdown.name\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: Amir Goldstein \u003camir73il@gmail.com\u003e\nCc: Heiko Carstens \u003chca@linux.ibm.com\u003e\nCc: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nCc: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\t[5.9+]\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "a30a29091b5a6d4c64b5fc77040720a65e2dd4e6", "tree": "ac222c5119601c8be28468638b3531365791d268", "parents": [ "793f49a87aae24e5bcf92ad98d764153fc936570" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Tue Feb 09 13:42:10 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "mm/mremap: fix BUILD_BUG_ON() error in get_extent\n\nclang can\u0027t evaluate this function argument at compile time when the\nfunction is not inlined, which leads to a link time failure:\n\n ld.lld: error: undefined symbol: __compiletime_assert_414\n \u003e\u003e\u003e referenced by mremap.c\n \u003e\u003e\u003e mremap.o:(get_extent) in archive mm/built-in.a\n\nMark the function as __always_inline to avoid it.\n\nLink: https://lkml.kernel.org/r/20201230154104.522605-1-arnd@kernel.org\nFixes: 9ad9718bfa41 (\"mm/mremap: calculate extent in one place\")\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nTested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nReviewed-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nTested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\nCc: Kirill A. Shutemov\" \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Wei Yang \u003crichard.weiyang@linux.alibaba.com\u003e\nCc: Vlastimil Babka \u003cvbabka@suse.cz\u003e\nCc: Dmitry Safonov \u003c0x7f454c46@gmail.com\u003e\nCc: Brian Geffon \u003cbgeffon@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "793f49a87aae24e5bcf92ad98d764153fc936570", "tree": "28325640c8ffa574400c7cc3abf1290a0f2f35d9", "parents": [ "1cc4cdb521f9689183474bc89eefc451ac44fa1c" ], "author": { "name": "Fangrui Song", "email": "maskray@google.com", "time": "Tue Feb 09 13:42:07 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "firmware_loader: align .builtin_fw to 8\n\narm64 references the start address of .builtin_fw (__start_builtin_fw)\nwith a pair of R_AARCH64_ADR_PREL_PG_HI21/R_AARCH64_LDST64_ABS_LO12_NC\nrelocations. The compiler is allowed to emit the\nR_AARCH64_LDST64_ABS_LO12_NC relocation because struct builtin_fw in\ninclude/linux/firmware.h is 8-byte aligned.\n\nThe R_AARCH64_LDST64_ABS_LO12_NC relocation requires the address to be a\nmultiple of 8, which may not be the case if .builtin_fw is empty.\nUnconditionally align .builtin_fw to fix the linker error. 32-bit\narchitectures could use ALIGN(4) but that would add unnecessary\ncomplexity, so just use ALIGN(8).\n\nLink: https://lkml.kernel.org/r/20201208054646.2913063-1-maskray@google.com\nLink: https://github.com/ClangBuiltLinux/linux/issues/1204\nFixes: 5658c76 (\"firmware: allow firmware files to be built into kernel image\")\nSigned-off-by: Fangrui Song \u003cmaskray@google.com\u003e\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nAcked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nReviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nTested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nTested-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nAcked-by: Nathan Chancellor \u003cnathan@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1cc4cdb521f9689183474bc89eefc451ac44fa1c", "tree": "1b6fad4426474239dc6a16aceaf00f516e6193d7", "parents": [ "506220d2ba21791314af569211ffd8870b8208fa" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Tue Feb 09 13:42:03 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "kasan: fix stack traces dependency for HW_TAGS\n\nCurrently, whether the alloc/free stack traces collection is enabled by\ndefault for hardware tag-based KASAN depends on CONFIG_DEBUG_KERNEL.\nThe intention for this dependency was to only enable collection on slow\ndebug kernels due to a significant perf and memory impact.\n\nAs it turns out, CONFIG_DEBUG_KERNEL is not considered a debug option\nand is enabled on many productions kernels including Android and Ubuntu.\nAs the result, this dependency is pointless and only complicates the\ncode and documentation.\n\nHaving stack traces collection disabled by default would make the\nhardware mode work differently to to the software ones, which is\nconfusing.\n\nThis change removes the dependency and enables stack traces collection\nby default.\n\nLooking into the future, this default might makes sense for production\nkernels, assuming we implement a fast stack trace collection approach.\n\nLink: https://lkml.kernel.org/r/6678d77ceffb71f1cff2cf61560e2ffe7bb6bfe9.1612808820.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Marco Elver \u003celver@google.com\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Alexander Potapenko \u003cglider@google.com\u003e\nCc: Will Deacon \u003cwill.deacon@arm.com\u003e\nCc: Andrey Ryabinin \u003caryabinin@virtuozzo.com\u003e\nCc: Peter Collingbourne \u003cpcc@google.com\u003e\nCc: Evgenii Stepanov \u003ceugenis@google.com\u003e\nCc: Branislav Rankov \u003cBranislav.Rankov@arm.com\u003e\nCc: Kevin Brodsky \u003ckevin.brodsky@arm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "506220d2ba21791314af569211ffd8870b8208fa", "tree": "b627cb280569c5ed0e293dd2ba59ed198ed967f5", "parents": [ "eabac19e40c095543def79cb6ffeb3a8588aaff4" ], "author": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Tue Feb 09 13:42:00 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "squashfs: add more sanity checks in xattr id lookup\n\nSysbot has reported a warning where a kmalloc() attempt exceeds the\nmaximum limit. This has been identified as corruption of the xattr_ids\ncount when reading the xattr id lookup table.\n\nThis patch adds a number of additional sanity checks to detect this\ncorruption and others.\n\n1. It checks for a corrupted xattr index read from the inode. This could\n be because the metadata block is uncompressed, or because the\n \"compression\" bit has been corrupted (turning a compressed block\n into an uncompressed block). This would cause an out of bounds read.\n\n2. It checks against corruption of the xattr_ids count. This can either\n lead to the above kmalloc failure, or a smaller than expected\n table to be read.\n\n3. It checks the contents of the index table for corruption.\n\n[phillip@squashfs.org.uk: fix checkpatch issue]\n Link: https://lkml.kernel.org/r/270245655.754655.1612770082682@webmail.123-reg.co.uk\n\nLink: https://lkml.kernel.org/r/20210204130249.4495-5-phillip@squashfs.org.uk\nSigned-off-by: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\nReported-by: syzbot+2ccea6339d368360800d@syzkaller.appspotmail.com\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "eabac19e40c095543def79cb6ffeb3a8588aaff4", "tree": "6fbdbedde99af1b08b12344d640a8a5d758914a9", "parents": [ "f37aa4c7366e23f91b81d00bafd6a7ab54e4a381" ], "author": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Tue Feb 09 13:41:56 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "squashfs: add more sanity checks in inode lookup\n\nSysbot has reported an \"slab-out-of-bounds read\" error which has been\nidentified as being caused by a corrupted \"ino_num\" value read from the\ninode. This could be because the metadata block is uncompressed, or\nbecause the \"compression\" bit has been corrupted (turning a compressed\nblock into an uncompressed block).\n\nThis patch adds additional sanity checks to detect this, and the\nfollowing corruption.\n\n1. It checks against corruption of the inodes count. This can either\n lead to a larger table to be read, or a smaller than expected\n table to be read.\n\n In the case of a too large inodes count, this would often have been\n trapped by the existing sanity checks, but this patch introduces\n a more exact check, which can identify too small values.\n\n2. It checks the contents of the index table for corruption.\n\n[phillip@squashfs.org.uk: fix checkpatch issue]\n Link: https://lkml.kernel.org/r/527909353.754618.1612769948607@webmail.123-reg.co.uk\n\nLink: https://lkml.kernel.org/r/20210204130249.4495-4-phillip@squashfs.org.uk\nSigned-off-by: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\nReported-by: syzbot+04419e3ff19d2970ea28@syzkaller.appspotmail.com\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "f37aa4c7366e23f91b81d00bafd6a7ab54e4a381", "tree": "53aadb0ec27a249dca24aa315e945d1c82ca671e", "parents": [ "e812cbbbbbb15adbbbee176baa1e8bda53059bf0" ], "author": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Tue Feb 09 13:41:53 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "squashfs: add more sanity checks in id lookup\n\nSysbot has reported a number of \"slab-out-of-bounds reads\" and\n\"use-after-free read\" errors which has been identified as being caused\nby a corrupted index value read from the inode. This could be because\nthe metadata block is uncompressed, or because the \"compression\" bit has\nbeen corrupted (turning a compressed block into an uncompressed block).\n\nThis patch adds additional sanity checks to detect this, and the\nfollowing corruption.\n\n1. It checks against corruption of the ids count. This can either\n lead to a larger table to be read, or a smaller than expected\n table to be read.\n\n In the case of a too large ids count, this would often have been\n trapped by the existing sanity checks, but this patch introduces\n a more exact check, which can identify too small values.\n\n2. It checks the contents of the index table for corruption.\n\nLink: https://lkml.kernel.org/r/20210204130249.4495-3-phillip@squashfs.org.uk\nSigned-off-by: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\nReported-by: syzbot+b06d57ba83f604522af2@syzkaller.appspotmail.com\nReported-by: syzbot+c021ba012da41ee9807c@syzkaller.appspotmail.com\nReported-by: syzbot+5024636e8b5fd19f0f19@syzkaller.appspotmail.com\nReported-by: syzbot+bcbc661df46657d0fa4f@syzkaller.appspotmail.com\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e812cbbbbbb15adbbbee176baa1e8bda53059bf0", "tree": "ecce401b92a2351b67ef46e6dfc8d4ffa8c70875", "parents": [ "e0756cfc7d7cd08c98a53b6009c091a3f6a50be6" ], "author": { "name": "Phillip Lougher", "email": "phillip@squashfs.org.uk", "time": "Tue Feb 09 13:41:50 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:26:44 2021 -0800" }, "message": "squashfs: avoid out of bounds writes in decompressors\n\nPatch series \"Squashfs: fix BIO migration regression and add sanity checks\".\n\nPatch [1/4] fixes a regression introduced by the \"migrate from\nll_rw_block usage to BIO\" patch, which has produced a number of\nSysbot/Syzkaller reports.\n\nPatches [2/4], [3/4], and [4/4] fix a number of filesystem corruption\nissues which have produced Sysbot reports in the id, inode and xattr\nlookup code.\n\nEach patch has been tested against the Sysbot reproducers using the\ngiven kernel configuration. They have the appropriate \"Reported-by:\"\nlines added.\n\nAdditionally, all of the reproducer filesystems are indirectly fixed by\npatch [4/4] due to the fact they all have xattr corruption which is now\ndetected there.\n\nAdditional testing with other configurations and architectures (32bit,\nbig endian), and normal filesystems has also been done to trap any\ninadvertent regressions caused by the additional sanity checks.\n\nThis patch (of 4):\n\nThis is a regression introduced by the patch \"migrate from ll_rw_block\nusage to BIO\".\n\nSysbot/Syskaller has reported a number of \"out of bounds writes\" and\n\"unable to handle kernel paging request in squashfs_decompress\" errors\nwhich have been identified as a regression introduced by the above\npatch.\n\nSpecifically, the patch removed the following sanity check\n\n if (length \u003c 0 || length \u003e output-\u003elength ||\n\t\t(index + length) \u003e msblk-\u003ebytes_used)\n\nThis check did two things:\n\n1. It ensured any reads were not beyond the end of the filesystem\n\n2. It ensured that the \"length\" field read from the filesystem\n was within the expected maximum length. Without this any\n corrupted values can over-run allocated buffers.\n\nLink: https://lkml.kernel.org/r/20210204130249.4495-1-phillip@squashfs.org.uk\nLink: https://lkml.kernel.org/r/20210204130249.4495-2-phillip@squashfs.org.uk\nFixes: 93e72b3c612adc (\"squashfs: migrate from ll_rw_block usage to BIO\")\nReported-by: syzbot+6fba78f99b9afd4b5634@syzkaller.appspotmail.com\nSigned-off-by: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\nCc: Philippe Liard \u003cpliard@google.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ef7d0b599938450c54a8dc0aa4b954d73d9a9370", "tree": "9a5e4176997c56fc3f82809c1ad79140309df811", "parents": [ "e0756cfc7d7cd08c98a53b6009c091a3f6a50be6", "291b5c9870fc546376d69cf792b7885cd0c9c1b3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:19:56 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Feb 09 17:19:56 2021 -0800" }, "message": "Merge tag \u0027i3c/fixes-for-5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux\n\nPull i3c fix from Alexandre Belloni:\n \"A single build warning fix\"\n\n* tag \u0027i3c/fixes-for-5.11\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux:\n i3c/master/mipi-i3c-hci: Fix position of __maybe_unused in i3c_hci_of_match\n" }, { "commit": "e88b2c6e5a4d9ce30d75391e4d950da74bb2bd90", "tree": "df00cea6815c0e36093bef3b8169694a38cdd81c", "parents": [ "fd675184fc7abfd1e1c52d23e8e900676b5a1c1a" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Tue Feb 09 18:46:10 2021 +0000" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Wed Feb 10 01:32:40 2021 +0100" }, "message": "bpf: Fix 32 bit src register truncation on div/mod\n\nWhile reviewing a different fix, John and I noticed an oddity in one of the\nBPF program dumps that stood out, for example:\n\n # bpftool p d x i 13\n 0: (b7) r0 \u003d 808464450\n 1: (b4) w4 \u003d 808464432\n 2: (bc) w0 \u003d w0\n 3: (15) if r0 \u003d\u003d 0x0 goto pc+1\n 4: (9c) w4 %\u003d w0\n [...]\n\nIn line 2 we noticed that the mov32 would 32 bit truncate the original src\nregister for the div/mod operation. While for the two operations the dst\nregister is typically marked unknown e.g. from adjust_scalar_min_max_vals()\nthe src register is not, and thus verifier keeps tracking original bounds,\nsimplified:\n\n 0: R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R10\u003dfp0\n 0: (b7) r0 \u003d -1\n 1: R0_w\u003dinvP-1 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R10\u003dfp0\n 1: (b7) r1 \u003d -1\n 2: R0_w\u003dinvP-1 R1_w\u003dinvP-1 R10\u003dfp0\n 2: (3c) w0 /\u003d w1\n 3: R0_w\u003dinvP(id\u003d0,umax_value\u003d4294967295,var_off\u003d(0x0; 0xffffffff)) R1_w\u003dinvP-1 R10\u003dfp0\n 3: (77) r1 \u003e\u003e\u003d 32\n 4: R0_w\u003dinvP(id\u003d0,umax_value\u003d4294967295,var_off\u003d(0x0; 0xffffffff)) R1_w\u003dinvP4294967295 R10\u003dfp0\n 4: (bf) r0 \u003d r1\n 5: R0_w\u003dinvP4294967295 R1_w\u003dinvP4294967295 R10\u003dfp0\n 5: (95) exit\n processed 6 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0\n\nRuntime result of r0 at exit is 0 instead of expected -1. Remove the\nverifier mov32 src rewrite in div/mod and replace it with a jmp32 test\ninstead. After the fix, we result in the following code generation when\nhaving dividend r1 and divisor r6:\n\n div, 64 bit: div, 32 bit:\n\n 0: (b7) r6 \u003d 8 0: (b7) r6 \u003d 8\n 1: (b7) r1 \u003d 8 1: (b7) r1 \u003d 8\n 2: (55) if r6 !\u003d 0x0 goto pc+2 2: (56) if w6 !\u003d 0x0 goto pc+2\n 3: (ac) w1 ^\u003d w1 3: (ac) w1 ^\u003d w1\n 4: (05) goto pc+1 4: (05) goto pc+1\n 5: (3f) r1 /\u003d r6 5: (3c) w1 /\u003d w6\n 6: (b7) r0 \u003d 0 6: (b7) r0 \u003d 0\n 7: (95) exit 7: (95) exit\n\n mod, 64 bit: mod, 32 bit:\n\n 0: (b7) r6 \u003d 8 0: (b7) r6 \u003d 8\n 1: (b7) r1 \u003d 8 1: (b7) r1 \u003d 8\n 2: (15) if r6 \u003d\u003d 0x0 goto pc+1 2: (16) if w6 \u003d\u003d 0x0 goto pc+1\n 3: (9f) r1 %\u003d r6 3: (9c) w1 %\u003d w6\n 4: (b7) r0 \u003d 0 4: (b7) r0 \u003d 0\n 5: (95) exit 5: (95) exit\n\nx86 in particular can throw a \u0027divide error\u0027 exception for div\ninstruction not only for divisor being zero, but also for the case\nwhen the quotient is too large for the designated register. For the\nedx:eax and rdx:rax dividend pair it is not an issue in x86 BPF JIT\nsince we always zero edx (rdx). Hence really the only protection\nneeded is against divisor being zero.\n\nFixes: 68fda450a7df (\"bpf: fix 32-bit divide by zero\")\nCo-developed-by: John Fastabend \u003cjohn.fastabend@gmail.com\u003e\nSigned-off-by: John Fastabend \u003cjohn.fastabend@gmail.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n" }, { "commit": "fd675184fc7abfd1e1c52d23e8e900676b5a1c1a", "tree": "9a12e915fb6ec9d04f32c78d5ba8c656c5128678", "parents": [ "ee114dd64c0071500345439fc79dd5e0f9d106ed" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 05 20:48:21 2021 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Wed Feb 10 01:31:46 2021 +0100" }, "message": "bpf: Fix verifier jmp32 pruning decision logic\n\nAnatoly has been fuzzing with kBdysch harness and reported a hang in\none of the outcomes:\n\n func#0 @0\n 0: R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R10\u003dfp0\n 0: (b7) r0 \u003d 808464450\n 1: R0_w\u003dinvP808464450 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R10\u003dfp0\n 1: (b4) w4 \u003d 808464432\n 2: R0_w\u003dinvP808464450 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP808464432 R10\u003dfp0\n 2: (9c) w4 %\u003d w0\n 3: R0_w\u003dinvP808464450 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0,umax_value\u003d4294967295,var_off\u003d(0x0; 0xffffffff)) R10\u003dfp0\n 3: (66) if w4 s\u003e 0x30303030 goto pc+0\n R0_w\u003dinvP808464450 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0,umax_value\u003d4294967295,var_off\u003d(0x0; 0xffffffff),s32_max_value\u003d808464432) R10\u003dfp0\n 4: R0_w\u003dinvP808464450 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0,umax_value\u003d4294967295,var_off\u003d(0x0; 0xffffffff),s32_max_value\u003d808464432) R10\u003dfp0\n 4: (7f) r0 \u003e\u003e\u003d r0\n 5: R0_w\u003dinvP(id\u003d0) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0,umax_value\u003d4294967295,var_off\u003d(0x0; 0xffffffff),s32_max_value\u003d808464432) R10\u003dfp0\n 5: (9c) w4 %\u003d w0\n 6: R0_w\u003dinvP(id\u003d0) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 6: (66) if w0 s\u003e 0x3030 goto pc+0\n R0_w\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 7: R0\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4\u003dinvP(id\u003d0) R10\u003dfp0\n 7: (d6) if w0 s\u003c\u003d 0x303030 goto pc+1\n 9: R0\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4\u003dinvP(id\u003d0) R10\u003dfp0\n 9: (95) exit\n propagating r0\n\n from 6 to 7: safe\n 4: R0_w\u003dinvP808464450 R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0,umin_value\u003d808464433,umax_value\u003d2147483647,var_off\u003d(0x0; 0x7fffffff)) R10\u003dfp0\n 4: (7f) r0 \u003e\u003e\u003d r0\n 5: R0_w\u003dinvP(id\u003d0) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0,umin_value\u003d808464433,umax_value\u003d2147483647,var_off\u003d(0x0; 0x7fffffff)) R10\u003dfp0\n 5: (9c) w4 %\u003d w0\n 6: R0_w\u003dinvP(id\u003d0) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 6: (66) if w0 s\u003e 0x3030 goto pc+0\n R0_w\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n propagating r0\n 7: safe\n propagating r0\n\n from 6 to 7: safe\n processed 15 insns (limit 1000000) max_states_per_insn 0 total_states 1 peak_states 1 mark_read 1\n\nThe underlying program was xlated as follows:\n\n # bpftool p d x i 10\n 0: (b7) r0 \u003d 808464450\n 1: (b4) w4 \u003d 808464432\n 2: (bc) w0 \u003d w0\n 3: (15) if r0 \u003d\u003d 0x0 goto pc+1\n 4: (9c) w4 %\u003d w0\n 5: (66) if w4 s\u003e 0x30303030 goto pc+0\n 6: (7f) r0 \u003e\u003e\u003d r0\n 7: (bc) w0 \u003d w0\n 8: (15) if r0 \u003d\u003d 0x0 goto pc+1\n 9: (9c) w4 %\u003d w0\n 10: (66) if w0 s\u003e 0x3030 goto pc+0\n 11: (d6) if w0 s\u003c\u003d 0x303030 goto pc+1\n 12: (05) goto pc-1\n 13: (95) exit\n\nThe verifier rewrote original instructions it recognized as dead code with\n\u0027goto pc-1\u0027, but reality differs from verifier simulation in that we are\nactually able to trigger a hang due to hitting the \u0027goto pc-1\u0027 instructions.\n\nTaking a closer look at the verifier analysis, the reason is that it misjudges\nits pruning decision at the first \u0027from 6 to 7: safe\u0027 occasion. What happens\nis that while both old/cur registers are marked as precise, they get misjudged\nfor the jmp32 case as range_within() yields true, meaning that the prior\nverification path with a wider register bound could be verified successfully\nand therefore the current path with a narrower register bound is deemed safe\nas well whereas in reality it\u0027s not. R0 old/cur path\u0027s bounds compare as\nfollows:\n\n old: smin_value\u003d0x8000000000000000,smax_value\u003d0x7fffffffffffffff,umin_value\u003d0x0,umax_value\u003d0xffffffffffffffff,var_off\u003d(0x0; 0xffffffffffffffff)\n cur: smin_value\u003d0x8000000000000000,smax_value\u003d0x7fffffff7fffffff,umin_value\u003d0x0,umax_value\u003d0xffffffff7fffffff,var_off\u003d(0x0; 0xffffffff7fffffff)\n\n old: s32_min_value\u003d0x80000000,s32_max_value\u003d0x00003030,u32_min_value\u003d0x00000000,u32_max_value\u003d0xffffffff\n cur: s32_min_value\u003d0x00003031,s32_max_value\u003d0x7fffffff,u32_min_value\u003d0x00003031,u32_max_value\u003d0x7fffffff\n\nThe 64 bit bounds generally look okay and while the information that got\npropagated from 32 to 64 bit looks correct as well, it\u0027s not precise enough\nfor judging a conditional jmp32. Given the latter only operates on subregisters\nwe also need to take these into account as well for a range_within() probe\nin order to be able to prune paths. Extending the range_within() constraint\nto both bounds will be able to tell us that the old signed 32 bit bounds are\nnot wider than the cur signed 32 bit bounds.\n\nWith the fix in place, the program will now verify the \u0027goto\u0027 branch case as\nit should have been:\n\n [...]\n 6: R0_w\u003dinvP(id\u003d0) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 6: (66) if w0 s\u003e 0x3030 goto pc+0\n R0_w\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 7: R0\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4\u003dinvP(id\u003d0) R10\u003dfp0\n 7: (d6) if w0 s\u003c\u003d 0x303030 goto pc+1\n 9: R0\u003dinvP(id\u003d0,s32_max_value\u003d12336) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4\u003dinvP(id\u003d0) R10\u003dfp0\n 9: (95) exit\n\n 7: R0_w\u003dinvP(id\u003d0,smax_value\u003d9223372034707292159,umax_value\u003d18446744071562067967,var_off\u003d(0x0; 0xffffffff7fffffff),s32_min_value\u003d12337,u32_min_value\u003d12337,u32_max_value\u003d2147483647) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 7: (d6) if w0 s\u003c\u003d 0x303030 goto pc+1\n R0_w\u003dinvP(id\u003d0,smax_value\u003d9223372034707292159,umax_value\u003d18446744071562067967,var_off\u003d(0x0; 0xffffffff7fffffff),s32_min_value\u003d3158065,u32_min_value\u003d3158065,u32_max_value\u003d2147483647) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 8: R0_w\u003dinvP(id\u003d0,smax_value\u003d9223372034707292159,umax_value\u003d18446744071562067967,var_off\u003d(0x0; 0xffffffff7fffffff),s32_min_value\u003d3158065,u32_min_value\u003d3158065,u32_max_value\u003d2147483647) R1\u003dctx(id\u003d0,off\u003d0,imm\u003d0) R4_w\u003dinvP(id\u003d0) R10\u003dfp0\n 8: (30) r0 \u003d *(u8 *)skb[808464432]\n BPF_LD_[ABS|IND] uses reserved fields\n processed 11 insns (limit 1000000) max_states_per_insn 1 total_states 1 peak_states 1 mark_read 1\n\nThe bug is quite subtle in the sense that when verifier would determine that\na given branch is dead code, it would (here: wrongly) remove these instructions\nfrom the program and hard-wire the taken branch for privileged programs instead\nof the \u0027goto pc-1\u0027 rewrites which will cause hard to debug problems.\n\nFixes: 3f50f132d840 (\"bpf: Verifier, do explicit ALU32 bounds tracking\")\nReported-by: Anatoly Trosinenko \u003canatoly.trosinenko@gmail.com\u003e\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nReviewed-by: John Fastabend \u003cjohn.fastabend@gmail.com\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n" }, { "commit": "ee114dd64c0071500345439fc79dd5e0f9d106ed", "tree": "32a6a5f19f508c60da0e9eaef0d419c26bdf2063", "parents": [ "6183f4d3a0a2ad230511987c6c362ca43ec0055f" ], "author": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Fri Feb 05 17:20:14 2021 +0100" }, "committer": { "name": "Daniel Borkmann", "email": "daniel@iogearbox.net", "time": "Wed Feb 10 01:31:45 2021 +0100" }, "message": "bpf: Fix verifier jsgt branch analysis on max bound\n\nFix incorrect is_branch{32,64}_taken() analysis for the jsgt case. The return\ncode for both will tell the caller whether a given conditional jump is taken\nor not, e.g. 1 means branch will be taken [for the involved registers] and the\ngoto target will be executed, 0 means branch will not be taken and instead we\nfall-through to the next insn, and last but not least a -1 denotes that it is\nnot known at verification time whether a branch will be taken or not. Now while\nthe jsgt has the branch-taken case correct with reg-\u003es32_min_value \u003e sval, the\nbranch-not-taken case is off-by-one when testing for reg-\u003es32_max_value \u003c sval\nsince the branch will also be taken for reg-\u003es32_max_value \u003d\u003d sval. The jgt\nbranch analysis, for example, gets this right.\n\nFixes: 3f50f132d840 (\"bpf: Verifier, do explicit ALU32 bounds tracking\")\nFixes: 4f7b3e82589e (\"bpf: improve verifier branch analysis\")\nSigned-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nReviewed-by: John Fastabend \u003cjohn.fastabend@gmail.com\u003e\nAcked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n" }, { "commit": "450bbc3395185963b133177a9ddae305dae7c5fa", "tree": "6a7f648b967531b188336fb8bc0be1af3e4f3034", "parents": [ "1c5fae9c9a092574398a17facc31c533791ef232", "664899e85c1312e51d2761e7f8b2f25d053e8489" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:55:59 2021 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:55:59 2021 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf\n\nPablo Neira Ayuso says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nNetfilter fixes for net\n\nThe following patchset contains Netfilter fixes for net:\n\n1) nf_conntrack_tuple_taken() needs to recheck zone for\n NAT clash resolution, from Florian Westphal.\n\n2) Restore support for stateful expressions when set definition\n specifies no stateful expressions.\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": "1c5fae9c9a092574398a17facc31c533791ef232", "tree": "338dfb4c5f7f0dc46ddfb3aa6f5372c63d1faaa2", "parents": [ "49c2547b82c6da8875d375e3544354e8bd7cf082" ], "author": { "name": "Stefano Garzarella", "email": "sgarzare@redhat.com", "time": "Tue Feb 09 09:52:19 2021 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:31:22 2021 -0800" }, "message": "vsock: fix locking in vsock_shutdown()\n\nIn vsock_shutdown() we touched some socket fields without holding the\nsocket lock, such as \u0027state\u0027 and \u0027sk_flags\u0027.\n\nAlso, after the introduction of multi-transport, we are accessing\n\u0027vsk-\u003etransport\u0027 in vsock_send_shutdown() without holding the lock\nand this call can be made while the connection is in progress, so\nthe transport can change in the meantime.\n\nTo avoid issues, we hold the socket lock when we enter in\nvsock_shutdown() and release it when we leave.\n\nAmong the transports that implement the \u0027shutdown\u0027 callback, only\nhyperv_transport acquired the lock. Since the caller now holds it,\nwe no longer take it.\n\nFixes: d021c344051a (\"VSOCK: Introduce VM Sockets\")\nSigned-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "49c2547b82c6da8875d375e3544354e8bd7cf082", "tree": "514fff9bbd472bb30699e30b8e2cd9d08115f1ca", "parents": [ "eb4733d7cffc547e08fe5a216e4f03663bb71108", "532cfc0df1e4d68e74522ef4a0dcbf6ebbe68287" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:20:43 2021 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:20:43 2021 -0800" }, "message": "Merge branch \u0027hns3-fixes\u0027\n\nHuazhong Tan says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: hns3: fixes for -net\n\nThe parameters sent from vf may be unreliable. If these\nparameters are used directly, memory overwriting may occur.\n\nSo this series adds some checks for this case.\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": "532cfc0df1e4d68e74522ef4a0dcbf6ebbe68287", "tree": "514fff9bbd472bb30699e30b8e2cd9d08115f1ca", "parents": [ "326334aad024a60f46dc5e7dbe1efe32da3ca66f" ], "author": { "name": "Yufeng Mo", "email": "moyufeng@huawei.com", "time": "Tue Feb 09 17:03:07 2021 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:20:43 2021 -0800" }, "message": "net: hns3: add a check for index in hclge_get_rss_key()\n\nThe index is received from vf, if use it directly,\nan out-of-bound issue may be caused, so add a check for\nthis index before using it in hclge_get_rss_key().\n\nFixes: a638b1d8cc87 (\"net: hns3: fix get VF RSS issue\")\nSigned-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nSigned-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "326334aad024a60f46dc5e7dbe1efe32da3ca66f", "tree": "696ae6870a96d1d1edb32ac240e227ed498157e9", "parents": [ "67a69f84cab60484f02eb8cbc7a76edffbb28a25" ], "author": { "name": "Yufeng Mo", "email": "moyufeng@huawei.com", "time": "Tue Feb 09 17:03:06 2021 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:20:43 2021 -0800" }, "message": "net: hns3: add a check for tqp_index in hclge_get_ring_chain_from_mbx()\n\nThe tqp_index is received from vf, if use it directly,\nan out-of-bound issue may be caused, so add a check for\nthis tqp_index before using it in hclge_get_ring_chain_from_mbx().\n\nFixes: 84e095d64ed9 (\"net: hns3: Change PF to add ring-vect binding \u0026 resetQ to mailbox\")\nSigned-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nSigned-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "67a69f84cab60484f02eb8cbc7a76edffbb28a25", "tree": "932d1e7399d5b3dc7199c95c3b4b5a36a0620d53", "parents": [ "eb4733d7cffc547e08fe5a216e4f03663bb71108" ], "author": { "name": "Yufeng Mo", "email": "moyufeng@huawei.com", "time": "Tue Feb 09 17:03:05 2021 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 15:20:43 2021 -0800" }, "message": "net: hns3: add a check for queue_id in hclge_reset_vf_queue()\n\nThe queue_id is received from vf, if use it directly,\nan out-of-bound issue may be caused, so add a check for\nthis queue_id before using it in hclge_reset_vf_queue().\n\nFixes: 1a426f8b40fc (\"net: hns3: fix the VF queue reset flow error\")\nSigned-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nSigned-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eb4733d7cffc547e08fe5a216e4f03663bb71108", "tree": "38bb8742810e1f9e6fb9da320e559a8b393a6e45", "parents": [ "6bbc088d6ebfd5a4284641dbe8413ebab0dfeb8c" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Mon Feb 08 19:36:27 2021 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Feb 09 11:41:11 2021 -0800" }, "message": "net: dsa: felix: implement port flushing on .phylink_mac_link_down\n\nThere are several issues which may be seen when the link goes down while\nforwarding traffic, all of which can be attributed to the fact that the\nport flushing procedure from the reference manual was not closely\nfollowed.\n\nWith flow control enabled on both the ingress port and the egress port,\nit may happen when a link goes down that Ethernet packets are in flight.\nIn flow control mode, frames are held back and not dropped. When there\nis enough traffic in flight (example: iperf3 TCP), then the ingress port\nmight enter congestion and never exit that state. This is a problem,\nbecause it is the egress port\u0027s link that went down, and that has caused\nthe inability of the ingress port to send packets to any other port.\nThis is solved by flushing the egress port\u0027s queues when it goes down.\n\nThere is also a problem when performing stream splitting for\nIEEE 802.1CB traffic (not yet upstream, but a sort of multicast,\nbasically). There, if one port from the destination ports mask goes\ndown, splitting the stream towards the other destinations will no longer\nbe performed. This can be traced down to this line:\n\n\tocelot_port_writel(ocelot_port, 0, DEV_MAC_ENA_CFG);\n\nwhich should have been instead, as per the reference manual:\n\n\tocelot_port_rmwl(ocelot_port, 0, DEV_MAC_ENA_CFG_RX_ENA,\n\t\t\t DEV_MAC_ENA_CFG);\n\nBasically only DEV_MAC_ENA_CFG_RX_ENA should be disabled, but not\nDEV_MAC_ENA_CFG_TX_ENA - I don\u0027t have further insight into why that is\nthe case, but apparently multicasting to several ports will cause issues\nif at least one of them doesn\u0027t have DEV_MAC_ENA_CFG_TX_ENA set.\n\nI am not sure what the state of the Ocelot VSC7514 driver is, but\nprobably not as bad as Felix/Seville, since VSC7514 uses phylib and has\nthe following in ocelot_adjust_link:\n\n\tif (!phydev-\u003elink)\n\t\treturn;\n\ntherefore the port is not really put down when the link is lost, unlike\nthe DSA drivers which use .phylink_mac_link_down for that.\n\nNonetheless, I put ocelot_port_flush() in the common ocelot.c because it\nneeds to access some registers from drivers/net/ethernet/mscc/ocelot_rew.h\nwhich are not exported in include/soc/mscc/ and a bugfix patch should\nprobably not move headers around.\n\nFixes: bdeced75b13f (\"net: dsa: felix: Add PCS operations for PHYLINK\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "2f51312bebb77962a518b4c6de777dd378b6110a", "tree": "accaaef3a2cfca6ea7bb7ffbdc06e4633e7451a0", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Imre Deak", "email": "imre.deak@intel.com", "time": "Mon Feb 08 17:43:03 2021 +0200" }, "committer": { "name": "Jani Nikula", "email": "jani.nikula@intel.com", "time": "Tue Feb 09 20:27:31 2021 +0200" }, "message": "drm/i915/tgl+: Make sure TypeC FIA is powered up when initializing it\n\nThe TypeC FIA can be powered down if the TC-COLD power state is allowed,\nso block the TC-COLD state when initializing the FIA.\n\nNote that this isn\u0027t needed on ICL where the FIA is never modular and\nwhich has no generic way to block TC-COLD (except for platforms with a\nlegacy TypeC port and on those too only via these legacy ports, not via\na DP-alt/TBT port).\n\nCc: \u003cstable@vger.kernel.org\u003e # v5.10+\nCc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nReported-by: Paul Menzel \u003cpmenzel@molgen.mpg.de\u003e\nCloses: https://gitlab.freedesktop.org/drm/intel/-/issues/3027\nSigned-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210208154303.6839-1-imre.deak@intel.com\nReviewed-by: Jos� Roberto de Souza \u003cjose.souza@intel.com\u003e\n(cherry picked from commit f48993e5d26b079e8c80fff002499a213dbdb1b4)\nSigned-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n" }, { "commit": "abd4af47d3fb64157133bd35de25d39a21a31122", "tree": "dcedc5a5f0f81e6ab36997a85c810233641035f5", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Ronnie Sahlberg", "email": "lsahlber@redhat.com", "time": "Mon Feb 08 16:48:31 2021 +1000" }, "committer": { "name": "Steve French", "email": "stfrench@microsoft.com", "time": "Tue Feb 09 10:59:52 2021 -0600" }, "message": "cifs: fix dfs-links\n\nThis fixes a regression following dfs links that was introduced in the\npatch series for the new mount api.\n\nSigned-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nReviewed-by: Paulo Alcantara (SUSE) \u003cpc@cjr.nz\u003e\nSigned-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n" }, { "commit": "256b92af784d5043eeb7d559b6d5963dcc2ecb10", "tree": "16799aeb136741b9a47c2565d0f058954618cbb7", "parents": [ "2ade0d60939bcd54197c133b03b460fe62a4ec47" ], "author": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Mon Feb 08 16:43:30 2021 +0100" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Tue Feb 09 11:23:47 2021 +0100" }, "message": "x86/build: Disable CET instrumentation in the kernel for 32-bit too\n\nCommit\n\n 20bf2b378729 (\"x86/build: Disable CET instrumentation in the kernel\")\n\ndisabled CET instrumentation which gets added by default by the Ubuntu\ngcc9 and 10 by default, but did that only for 64-bit builds. It would\nstill fail when building a 32-bit target. So disable CET for all x86\nbuilds.\n\nFixes: 20bf2b378729 (\"x86/build: Disable CET instrumentation in the kernel\")\nReported-by: AC \u003cachirvasub@gmail.com\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nAcked-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nTested-by: AC \u003cachirvasub@gmail.com\u003e\nLink: https://lkml.kernel.org/r/YCCIgMHkzh/xT4ex@arch-chirva.localdomain\n" }, { "commit": "f852c596f2ee6f0eb364ea8f28f89da6da0ae7b5", "tree": "40e43dc8d0da1cb49aacc2258a1d076974789291", "parents": [ "8c65830ae1629b03e5d65e9aafae7e2cf5f8b743" ], "author": { "name": "Maurizio Lombardi", "email": "mlombard@redhat.com", "time": "Mon Feb 08 12:17:34 2021 +0100" }, "committer": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Mon Feb 08 21:51:25 2021 -0500" }, "message": "scsi: scsi_debug: Fix a memory leak\n\nThe sdebug_q_arr pointer must be freed when the module is unloaded.\n\n$ cat /sys/kernel/debug/kmemleak\nunreferenced object 0xffff888e1cfb0000 (size 4096):\n comm \"modprobe\", pid 165555, jiffies 4325987516 (age 685.194s)\n hex dump (first 32 bytes):\n 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n backtrace:\n [\u003c00000000458f4f5d\u003e] 0xffffffffc06702d9\n [\u003c000000003edc4b1f\u003e] do_one_initcall+0xe9/0x57d\n [\u003c00000000da7d518c\u003e] do_init_module+0x1d1/0x6f0\n [\u003c000000009a6a9248\u003e] load_module+0x36bd/0x4f50\n [\u003c00000000ddb0c3ce\u003e] __do_sys_init_module+0x1db/0x260\n [\u003c000000009532db57\u003e] do_syscall_64+0xa5/0x420\n [\u003c000000002916b13d\u003e] entry_SYSCALL_64_after_hwframe+0x6a/0xdf\n\nFixes: 87c715dcde63 (\"scsi: scsi_debug: Add per_host_store option\")\nLink: https://lore.kernel.org/r/20210208111734.34034-1-mlombard@redhat.com\nAcked-by: Douglas Gilbert \u003cdgilbert@interlog.com\u003e\nSigned-off-by: Maurizio Lombardi \u003cmlombard@redhat.com\u003e\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n" }, { "commit": "6bbc088d6ebfd5a4284641dbe8413ebab0dfeb8c", "tree": "3e51cc659f77932742a2c749a2d9da441806cfb2", "parents": [ "3aa6bce9af0e25b735c9c1263739a5639a336ae8", "059d2a1004981dce19f0127dabc1b4ec927d202a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 08 16:20:58 2021 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 08 16:20:58 2021 -0800" }, "message": "Merge branch \u0027bridge-mrp\u0027\n\nHoratiu Vultur says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nbridge: mrp: Fix br_mrp_port_switchdev_set_state\n\nBased on the discussion here[1], there was a problem with the function\nbr_mrp_port_switchdev_set_state. The problem was that it was called\nboth with BR_STATE* and BR_MRP_PORT_STATE* types. This patch series\nfixes this issue and removes SWITCHDEV_ATTR_ID_MRP_PORT_STAT because\nis not used anymore.\n\n[1] https://www.spinics.net/lists/netdev/msg714816.html\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": "059d2a1004981dce19f0127dabc1b4ec927d202a", "tree": "3e51cc659f77932742a2c749a2d9da441806cfb2", "parents": [ "b2bdba1cbc84cadb14393d0101a5bfd38d342e0a" ], "author": { "name": "Horatiu Vultur", "email": "horatiu.vultur@microchip.com", "time": "Sat Feb 06 22:47:34 2021 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 08 16:20:57 2021 -0800" }, "message": "switchdev: mrp: Remove SWITCHDEV_ATTR_ID_MRP_PORT_STAT\n\nNow that MRP started to use also SWITCHDEV_ATTR_ID_PORT_STP_STATE to\nnotify HW, then SWITCHDEV_ATTR_ID_MRP_PORT_STAT is not used anywhere\nelse, therefore we can remove it.\n\nFixes: c284b545900830 (\"switchdev: mrp: Extend switchdev API to offload MRP\")\nSigned-off-by: Horatiu Vultur \u003choratiu.vultur@microchip.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b2bdba1cbc84cadb14393d0101a5bfd38d342e0a", "tree": "008b5d113b1f3eee52aa5e78f809956dfd2f871e", "parents": [ "3aa6bce9af0e25b735c9c1263739a5639a336ae8" ], "author": { "name": "Horatiu Vultur", "email": "horatiu.vultur@microchip.com", "time": "Sat Feb 06 22:47:33 2021 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 08 16:20:57 2021 -0800" }, "message": "bridge: mrp: Fix the usage of br_mrp_port_switchdev_set_state\n\nThe function br_mrp_port_switchdev_set_state was called both with MRP\nport state and STP port state, which is an issue because they don\u0027t\nmatch exactly.\n\nTherefore, update the function to be used only with STP port state and\nuse the id SWITCHDEV_ATTR_ID_PORT_STP_STATE.\n\nThe choice of using STP over MRP is that the drivers already implement\nSWITCHDEV_ATTR_ID_PORT_STP_STATE and already in SW we update the port\nSTP state.\n\nFixes: 9a9f26e8f7ea30 (\"bridge: mrp: Connect MRP API with the switchdev API\")\nFixes: fadd409136f0f2 (\"bridge: switchdev: mrp: Implement MRP API for switchdev\")\nFixes: 2f1a11ae11d222 (\"bridge: mrp: Add MRP interface.\")\nReported-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nSigned-off-by: Horatiu Vultur \u003choratiu.vultur@microchip.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3aa6bce9af0e25b735c9c1263739a5639a336ae8", "tree": "044f908132ba3671debf843068c8f9c23728d464", "parents": [ "ce7536bc7398e2ae552d2fabb7e0e371a9f1fe46" ], "author": { "name": "Edwin Peer", "email": "edwin.peer@broadcom.com", "time": "Fri Feb 05 17:37:32 2021 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Feb 08 16:18:58 2021 -0800" }, "message": "net: watchdog: hold device global xmit lock during tx disable\n\nPrevent netif_tx_disable() running concurrently with dev_watchdog() by\ntaking the device global xmit lock. Otherwise, the recommended:\n\n\tnetif_carrier_off(dev);\n\tnetif_tx_disable(dev);\n\ndriver shutdown sequence can happen after the watchdog has already\nchecked carrier, resulting in possible false alarms. This is because\nnetif_tx_lock() only sets the frozen bit without maintaining the locks\non the individual queues.\n\nFixes: c3f26a269c24 (\"netdev: Fix lockdep warnings in multiqueue configurations.\")\nSigned-off-by: Edwin Peer \u003cedwin.peer@broadcom.com\u003e\nReviewed-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "664899e85c1312e51d2761e7f8b2f25d053e8489", "tree": "2f9749a0bb017e4ef4358bc18e6f2708bd616280", "parents": [ "07998281c268592963e1cd623fe6ab0270b65ae4" ], "author": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Mon Feb 08 13:20:47 2021 +0100" }, "committer": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Tue Feb 09 00:50:14 2021 +0100" }, "message": "netfilter: nftables: relax check for stateful expressions in set definition\n\nRestore the original behaviour where users are allowed to add an element\nwith any stateful expression if the set definition specifies no stateful\nexpressions. Make sure upper maximum number of stateful expressions of\nNFT_SET_EXPR_MAX is not reached.\n\nFixes: 8cfd9b0f8515 (\"netfilter: nftables: generalize set expressions support\")\nFixes: 48b0ae046ee9 (\"netfilter: nftables: netlink support for several set element expressions\")\nSigned-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "07998281c268592963e1cd623fe6ab0270b65ae4", "tree": "1f35dd93ba3f633448b4bff33ca7e6005247f063", "parents": [ "ce7536bc7398e2ae552d2fabb7e0e371a9f1fe46" ], "author": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Fri Feb 05 12:56:43 2021 +0100" }, "committer": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Tue Feb 09 00:04:14 2021 +0100" }, "message": "netfilter: conntrack: skip identical origin tuple in same zone only\n\nThe origin skip check needs to re-test the zone. Else, we might skip\na colliding tuple in the reply direction.\n\nThis only occurs when using \u0027directional zones\u0027 where origin tuples\nreside in different zones but the reply tuples share the same zone.\n\nThis causes the new conntrack entry to be dropped at confirmation time\nbecause NAT clash resolution was elided.\n\nFixes: 4e35c1cb9460240 (\"netfilter: nf_nat: skip nat clash resolution for same-origin entries\")\nSigned-off-by: Florian Westphal \u003cfw@strlen.de\u003e\nSigned-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "ce7536bc7398e2ae552d2fabb7e0e371a9f1fe46", "tree": "c0b4ee4ca792efce5a8df27e236b3c9f4a4274dc", "parents": [ "af8085f3a4712c57d0dd415ad543bac85780375c" ], "author": { "name": "Stefano Garzarella", "email": "sgarzare@redhat.com", "time": "Mon Feb 08 15:44:54 2021 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 08 13:27:46 2021 -0800" }, "message": "vsock/virtio: update credit only if socket is not closed\n\nIf the socket is closed or is being released, some resources used by\nvirtio_transport_space_update() such as \u0027vsk-\u003etrans\u0027 may be released.\n\nTo avoid a use after free bug we should only update the available credit\nwhen we are sure the socket is still open and we have the lock held.\n\nFixes: 06a8fc78367d (\"VSOCK: Introduce virtio_vsock_common.ko\")\nSigned-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nAcked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nLink: https://lore.kernel.org/r/20210208144454.84438-1-sgarzare@redhat.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e0756cfc7d7cd08c98a53b6009c091a3f6a50be6", "tree": "723acaa839265332e5b027c687b8523217deb260", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3", "256cfdd6fdf70c6fcf0f7c8ddb0ebd73ce8f3bc9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 08 11:32:39 2021 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 08 11:32:39 2021 -0800" }, "message": "Merge tag \u0027trace-v5.11-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fix from Steven Rostedt:\n \"Fix output of top level event tracing \u0027enable\u0027 file.\n\n When writing a tool for enabling events in the tracing system, an\n anomaly was discovered. The top level event \u0027enable\u0027 file would never\n show \u00271\u0027 when all events were enabled.\n\n The system and event \u0027enable\u0027 files worked as expected.\n\n The reason was because the top level event \u0027enable\u0027 file included the\n \u0027ftrace\u0027 tracer events, which are not controlled by the \u0027enable\u0027 file\n and would cause the output to be wrong. This appears to have been a\n bug since it was created\"\n\n* tag \u0027trace-v5.11-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n tracing: Do not count ftrace events in top level enable output\n" }, { "commit": "af8085f3a4712c57d0dd415ad543bac85780375c", "tree": "2dbdc0e116d3ae862ff448bdcb434dc3ef8f0757", "parents": [ "225353c070fda18a23785e34e1eec2be508a3a3c" ], "author": { "name": "NeilBrown", "email": "neilb@suse.de", "time": "Fri Feb 05 11:36:30 2021 +1100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Feb 08 10:15:49 2021 -0800" }, "message": "net: fix iteration for sctp transport seq_files\n\nThe sctp transport seq_file iterators take a reference to the transport\nin the -\u003estart and -\u003enext functions and releases the reference in the\n-\u003eshow function. The preferred handling for such resources is to\nrelease them in the subsequent -\u003enext or -\u003estop function call.\n\nSince Commit 1f4aace60b0e (\"fs/seq_file.c: simplify seq_file iteration\ncode and interface\") there is no guarantee that -\u003eshow will be called\nafter -\u003enext, so this function can now leak references.\n\nSo move the sctp_transport_put() call to -\u003enext and -\u003estop.\n\nFixes: 1f4aace60b0e (\"fs/seq_file.c: simplify seq_file iteration code and interface\")\nReported-by: Xin Long \u003clucien.xin@gmail.com\u003e\nSigned-off-by: NeilBrown \u003cneilb@suse.de\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "2ade0d60939bcd54197c133b03b460fe62a4ec47", "tree": "abc76294ac6e62e42c2de92a77c14a3c085908b6", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Jarkko Sakkinen", "email": "jarkko@kernel.org", "time": "Mon Feb 08 00:14:01 2021 +0200" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Mon Feb 08 19:11:30 2021 +0100" }, "message": "x86/sgx: Maintain encl-\u003erefcount for each encl-\u003emm_list entry\n\nThis has been shown in tests:\n\n[ +0.000008] WARNING: CPU: 3 PID: 7620 at kernel/rcu/srcutree.c:374 cleanup_srcu_struct+0xed/0x100\n\nThis is essentially a use-after free, although SRCU notices it as\nan SRCU cleanup in an invalid context.\n\n\u003d\u003d Background \u003d\u003d\n\nSGX has a data structure (struct sgx_encl_mm) which keeps per-mm SGX\nmetadata. This is separate from struct sgx_encl because, in theory,\nan enclave can be mapped from more than one mm. sgx_encl_mm includes\na pointer back to the sgx_encl.\n\nThis means that sgx_encl must have a longer lifetime than all of the\nsgx_encl_mm\u0027s that point to it. That\u0027s usually the case: sgx_encl_mm\nis freed only after the mmu_notifier is unregistered in sgx_release().\n\nHowever, there\u0027s a race. If the process is exiting,\nsgx_mmu_notifier_release() can be called in parallel with sgx_release()\ninstead of being called *by* it. The mmu_notifier path keeps encl_mm\nalive past when sgx_encl can be freed. This inverts the lifetime rules\nand means that sgx_mmu_notifier_release() can access a freed sgx_encl.\n\n\u003d\u003d Fix \u003d\u003d\n\nIncrease encl-\u003erefcount when encl_mm-\u003eencl is established. Release\nthis reference when encl_mm is freed. This ensures that encl outlives\nencl_mm.\n\n [ bp: Massage commit message. ]\n\nFixes: 1728ab54b4be (\"x86/sgx: Add a page reclaimer\")\nReported-by: Haitao Huang \u003chaitao.huang@linux.intel.com\u003e\nSigned-off-by: Jarkko Sakkinen \u003cjarkko@kernel.org\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nAcked-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\nLink: https://lkml.kernel.org/r/20210207221401.29933-1-jarkko@kernel.org\n" }, { "commit": "fe0af09074bfeb46a35357e67635eefe33cdfc49", "tree": "17e269f0d5b882d9df1abcbfaed36ba79808d967", "parents": [ "92bf22614b21a2706f4993b278017e437f7785b3" ], "author": { "name": "Ard Biesheuvel", "email": "ardb@kernel.org", "time": "Sat Feb 06 09:49:37 2021 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Mon Feb 08 13:46:53 2021 +0100" }, "message": "Revert \"ACPICA: Interpreter: fix memory leak by using existing buffer\"\n\nThis reverts commit 32cf1a12cad43358e47dac8014379c2f33dfbed4.\n\nThe \u0027exisitng buffer\u0027 in this case is the firmware provided table, and\nwe should not modify that in place. This fixes a crash on arm64 with\ninitrd table overrides, in which case the DSDT is not mapped with\nread/write permissions.\n\nReported-by: Shawn Guo \u003cshawn.guo@linaro.org\u003e\nSigned-off-by: Ard Biesheuvel \u003cardb@kernel.org\u003e\nTested-by: Shawn Guo \u003cshawn.guo@linaro.org\u003e\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" } ], "next": "d11a1d08a082a7dc0ada423d2b2e26e9b6f2525c" }