)]}' { "log": [ { "commit": "0a50438c84363bd37fe18fe432888ae9a074dcab", "tree": "1e3f95e5709d1ba184f80fb91e71f3d39fd2b0a9", "parents": [ "e49d033bddf5b565044e2abe4241353959bc9120", "89e28ce60cb65971c73359c66d076aa20a395cd5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 05 09:35:58 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 05 09:35:58 2021 -0700" }, "message": "Merge branch \u0027for-5.12-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq\n\nPull workqueue fixes from Tejun Heo:\n \"Two workqueue fixes.\n\n One is around debugobj and poses no risk. The other is to prevent the\n stall watchdog from firing spuriously in certain conditions. Not as\n trivial as debugobj change but is still fairly low risk\"\n\n* \u0027for-5.12-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:\n workqueue/watchdog: Make unbound workqueues aware of touch_softlockup_watchdog() 84;0;0c84;0;0c There are two workqueue-specific watchdog timestamps:\n workqueue: Move the position of debug_work_activate() in __queue_work()\n" }, { "commit": "e49d033bddf5b565044e2abe4241353959bc9120", "tree": "3d211178ef8164474a7d787acf5616cbf8a08949", "parents": [ "829933ef05a951c8ff140e814656d73e74915faf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 04 14:15:36 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 04 14:15:36 2021 -0700" }, "message": "Linux 5.12-rc6\n" }, { "commit": "829933ef05a951c8ff140e814656d73e74915faf", "tree": "843c9b245f9af0127bb1dc53d469cfa08af29003", "parents": [ "2023a53bdf41b7646b1d384b6816af06309f73a5" ], "author": { "name": "Zheyu Ma", "email": "zheyuma97@gmail.com", "time": "Sat Apr 03 06:58:36 2021 +0000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Apr 04 14:05:45 2021 -0700" }, "message": "firewire: nosy: Fix a use-after-free bug in nosy_ioctl()\n\nFor each device, the nosy driver allocates a pcilynx structure.\nA use-after-free might happen in the following scenario:\n\n 1. Open nosy device for the first time and call ioctl with command\n NOSY_IOC_START, then a new client A will be malloced and added to\n doubly linked list.\n 2. Open nosy device for the second time and call ioctl with command\n NOSY_IOC_START, then a new client B will be malloced and added to\n doubly linked list.\n 3. Call ioctl with command NOSY_IOC_START for client A, then client A\n will be readded to the doubly linked list. Now the doubly linked\n list is messed up.\n 4. Close the first nosy device and nosy_release will be called. In\n nosy_release, client A will be unlinked and freed.\n 5. Close the second nosy device, and client A will be referenced,\n resulting in UAF.\n\nThe root cause of this bug is that the element in the doubly linked list\nis reentered into the list.\n\nFix this bug by adding a check before inserting a client. If a client\nis already in the linked list, don\u0027t insert it.\n\nThe following KASAN report reveals it:\n\n BUG: KASAN: use-after-free in nosy_release+0x1ea/0x210\n Write of size 8 at addr ffff888102ad7360 by task poc\n CPU: 3 PID: 337 Comm: poc Not tainted 5.12.0-rc5+ #6\n Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014\n Call Trace:\n nosy_release+0x1ea/0x210\n __fput+0x1e2/0x840\n task_work_run+0xe8/0x180\n exit_to_user_mode_prepare+0x114/0x120\n syscall_exit_to_user_mode+0x1d/0x40\n entry_SYSCALL_64_after_hwframe+0x44/0xae\n\n Allocated by task 337:\n nosy_open+0x154/0x4d0\n misc_open+0x2ec/0x410\n chrdev_open+0x20d/0x5a0\n do_dentry_open+0x40f/0xe80\n path_openat+0x1cf9/0x37b0\n do_filp_open+0x16d/0x390\n do_sys_openat2+0x11d/0x360\n __x64_sys_open+0xfd/0x1a0\n do_syscall_64+0x33/0x40\n entry_SYSCALL_64_after_hwframe+0x44/0xae\n\n Freed by task 337:\n kfree+0x8f/0x210\n nosy_release+0x158/0x210\n __fput+0x1e2/0x840\n task_work_run+0xe8/0x180\n exit_to_user_mode_prepare+0x114/0x120\n syscall_exit_to_user_mode+0x1d/0x40\n entry_SYSCALL_64_after_hwframe+0x44/0xae\n\n The buggy address belongs to the object at ffff888102ad7300 which belongs to the cache kmalloc-128 of size 128\n The buggy address is located 96 bytes inside of 128-byte region [ffff888102ad7300, ffff888102ad7380)\n\n[ Modified to use \u0027list_empty()\u0027 inside proper lock - Linus ]\n\nLink: https://lore.kernel.org/lkml/1617433116-5930-1-git-send-email-zheyuma97@gmail.com/\nReported-and-tested-by: 马哲宇 (Zheyu Ma) \u003czheyuma97@gmail.com\u003e\nSigned-off-by: Zheyu Ma \u003czheyuma97@gmail.com\u003e\nCc: Greg Kroah-Hartman \u003cgreg@kroah.com\u003e\nCc: Stefan Richter \u003cstefanr@s5r6.in-berlin.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "89e28ce60cb65971c73359c66d076aa20a395cd5", "tree": "f6185526512cce4a8eb5903926739b51715524a8", "parents": [ "0687c66b5f666b5ad433f4e94251590d9bc9d10e" ], "author": { "name": "Wang Qing", "email": "wangqing@vivo.com", "time": "Wed Mar 24 19:40:29 2021 +0800" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Sun Apr 04 13:26:49 2021 -0400" }, "message": "workqueue/watchdog: Make unbound workqueues aware of touch_softlockup_watchdog()\n84;0;0c84;0;0c\nThere are two workqueue-specific watchdog timestamps:\n\n + @wq_watchdog_touched_cpu (per-CPU) updated by\n touch_softlockup_watchdog()\n\n + @wq_watchdog_touched (global) updated by\n touch_all_softlockup_watchdogs()\n\nwatchdog_timer_fn() checks only the global @wq_watchdog_touched for\nunbound workqueues. As a result, unbound workqueues are not aware\nof touch_softlockup_watchdog(). The watchdog might report a stall\neven when the unbound workqueues are blocked by a known slow code.\n\nSolution:\ntouch_softlockup_watchdog() must touch also the global @wq_watchdog_touched\ntimestamp.\n\nThe global timestamp can no longer be used for bound workqueues because\nit is now updated from all CPUs. Instead, bound workqueues have to check\nonly @wq_watchdog_touched_cpu and these timestamps have to be updated for\nall CPUs in touch_all_softlockup_watchdogs().\n\nBeware:\nThe change might cause the opposite problem. An unbound workqueue\nmight get blocked on CPU A because of a real softlockup. The workqueue\nwatchdog would miss it when the timestamp got touched on CPU B.\n\nIt is acceptable because softlockups are detected by softlockup\nwatchdog. The workqueue watchdog is there to detect stalls where\na work never finishes, for example, because of dependencies of works\nqueued into the same workqueue.\n\nV3:\n- Modify the commit message clearly according to Petr\u0027s suggestion.\n\nSigned-off-by: Wang Qing \u003cwangqing@vivo.com\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n" }, { "commit": "0687c66b5f666b5ad433f4e94251590d9bc9d10e", "tree": "2518008cd800c0073c604a540ebaf7dd4485a5df", "parents": [ "2023a53bdf41b7646b1d384b6816af06309f73a5" ], "author": { "name": "Zqiang", "email": "qiang.zhang@windriver.com", "time": "Thu Feb 18 11:16:49 2021 +0800" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Sun Apr 04 13:26:46 2021 -0400" }, "message": "workqueue: Move the position of debug_work_activate() in __queue_work()\n\nThe debug_work_activate() is called on the premise that\nthe work can be inserted, because if wq be in WQ_DRAINING\nstatus, insert work may be failed.\n\nFixes: e41e704bc4f4 (\"workqueue: improve destroy_workqueue() debuggability\")\nSigned-off-by: Zqiang \u003cqiang.zhang@windriver.com\u003e\nReviewed-by: Lai Jiangshan \u003cjiangshanlai@gmail.com\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n" }, { "commit": "2023a53bdf41b7646b1d384b6816af06309f73a5", "tree": "aae0480ea74530eeb570f5d01441a5cb8dbad03b", "parents": [ "d83e98f9d8c88cbae1b05fa5751bddfcf0e222b2", "1683f7de65dbf0a2c6a7d639173fe92430a28930" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 15:42:45 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 15:42:45 2021 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://github.com/openrisc/linux\n\nPull OpenRISC fix from Stafford Horne:\n \"Fix duplicate header include in Litex SOC driver\"\n\n* tag \u0027for-linus\u0027 of git://github.com/openrisc/linux:\n soc: litex: Remove duplicated header file inclusion\n" }, { "commit": "d83e98f9d8c88cbae1b05fa5751bddfcf0e222b2", "tree": "002c25e5248a56e6f3a780c5641c3a86146858f8", "parents": [ "8e29be3468d4565dd95fbb098df0d7a79ee60d71", "e82ad4853948382d37ac512b27a3e70b6f01c103" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 14:26:47 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 14:26:47 2021 -0700" }, "message": "Merge tag \u0027io_uring-5.12-2021-04-03\u0027 of git://git.kernel.dk/linux-block\n\nPOull io_uring fix from Jens Axboe:\n \"Just fixing a silly braino in a previous patch, where we\u0027d end up\n failing to compile if CONFIG_BLOCK isn\u0027t enabled.\n\n Not that a lot of people do that, but kernel bot spotted it and it\u0027s\n probably prudent to just flush this out now before -rc6.\n\n Sorry about that, none of my test compile configs have !CONFIG_BLOCK\"\n\n* tag \u0027io_uring-5.12-2021-04-03\u0027 of git://git.kernel.dk/linux-block:\n io_uring: fix !CONFIG_BLOCK compilation failure\n" }, { "commit": "1683f7de65dbf0a2c6a7d639173fe92430a28930", "tree": "d88e968d0ed8638262a8eb1be3ffb5fea03c9356", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3" ], "author": { "name": "Zhen Lei", "email": "thunder.leizhen@huawei.com", "time": "Wed Mar 31 15:06:43 2021 +0200" }, "committer": { "name": "Stafford Horne", "email": "shorne@gmail.com", "time": "Sun Apr 04 05:46:46 2021 +0900" }, "message": "soc: litex: Remove duplicated header file inclusion\n\nThe header file \u003clinux/errno.h\u003e is already included above and can be\nremoved here.\n\nSigned-off-by: Zhen Lei \u003cthunder.leizhen@huawei.com\u003e\nSigned-off-by: Mateusz Holenko \u003cmholenko@antmicro.com\u003e\nSigned-off-by: Stafford Horne \u003cshorne@gmail.com\u003e\n" }, { "commit": "8e29be3468d4565dd95fbb098df0d7a79ee60d71", "tree": "95b50a51dfe33b81aa50bf95cb525b62c58438d5", "parents": [ "7fd7d5c20129d2227b95cbe567b24559f144b77c", "ff132c5f93c06bd4432bbab5c369e468653bdec4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 12:15:01 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 12:15:01 2021 -0700" }, "message": "Merge tag \u0027gfs2-v5.12-rc2-fixes2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2\n\nPull gfs2 fixes from Andreas Gruenbacher:\n \"Two more gfs2 fixes\"\n\n* tag \u0027gfs2-v5.12-rc2-fixes2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:\n gfs2: report \"already frozen/thawed\" errors\n gfs2: Flag a withdraw if init_threads() fails\n" }, { "commit": "7fd7d5c20129d2227b95cbe567b24559f144b77c", "tree": "7283d5fa9fd855f16b11999fe9efdbe7b2dee385", "parents": [ "9c2ef23e4dae122d2b18e834d90f8bd4dda48fe6", "1adbc2941eee8acbe3c7dc6b51cdbc5a9bf19565" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 11:52:18 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 11:52:18 2021 -0700" }, "message": "Merge tag \u0027riscv-for-linus-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V fixes from Palmer Dabbelt:\n \"A handful of fixes for 5.12:\n\n - fix a stack tracing regression related to \"const register asm\"\n variables, which have unexpected behavior.\n\n - ensure the value to be written by put_user() is evaluated before\n enabling access to userspace memory..\n\n - align the exception vector table correctly, so we don\u0027t rely on the\n firmware\u0027s handling of unaligned accesses.\n\n - build fix to make NUMA depend on MMU, which triggered on some\n randconfigs\"\n\n* tag \u0027riscv-for-linus-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n riscv: Make NUMA depend on MMU\n riscv: remove unneeded semicolon\n riscv,entry: fix misaligned base for excp_vect_table\n riscv: evaluate put_user() arg before enabling user access\n riscv: Drop const annotation for sp\n" }, { "commit": "9c2ef23e4dae122d2b18e834d90f8bd4dda48fe6", "tree": "7810e410e8b43e98a86d5ab485d0574f66b2aca3", "parents": [ "fa16199500c8863da145870f01d61617d967b6c3", "53f1d31708f6240e4615b0927df31f182e389e2f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:49:38 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:49:38 2021 -0700" }, "message": "Merge tag \u0027powerpc-5.12-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull powerpc fixes from Michael Ellerman:\n \"Fix a bug on pseries where spurious wakeups from H_PROD would prevent\n partition migration from succeeding.\n\n Fix oopses seen in pcpu_alloc(), caused by parallel faults of the\n percpu mapping causing us to corrupt the protection key used for the\n mapping, and cause a fatal key fault.\n\n Thanks to Aneesh Kumar K.V, Murilo Opsfelder Araujo, and Nathan Lynch\"\n\n* tag \u0027powerpc-5.12-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n powerpc/mm/book3s64: Use the correct storage key value when calling H_PROTECT\n powerpc/pseries/mobility: handle premature return from H_JOIN\n powerpc/pseries/mobility: use struct for shared state\n" }, { "commit": "fa16199500c8863da145870f01d61617d967b6c3", "tree": "1039b30fb3edd34f2912909a7188963f7831da4e", "parents": [ "f5664825fc2055ed9a0e4988cfc3aeb199dce520", "37df9f3fedb6aeaff5564145e8162aab912c9284" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:42:20 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:42:20 2021 -0700" }, "message": "Merge tag \u0027hyperv-fixes-signed-20210402\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux\n\nPull Hyper-V fixes from Wei Liu:\n \"One fix from Lu Yunlong for a double free in hvfb_probe\"\n\n* tag \u0027hyperv-fixes-signed-20210402\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:\n video: hyperv_fb: Fix a double free in hvfb_probe\n" }, { "commit": "f5664825fc2055ed9a0e4988cfc3aeb199dce520", "tree": "a1a50df754efe6c7ff7a45e9daa3dc8fcbb5b1d0", "parents": [ "a443930a3e8eaf70c8768cc2c923d1d85b71262f", "f0acf637d60ffcef3ccb6e279f743e587b3c7359" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:14:47 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:14:47 2021 -0700" }, "message": "Merge tag \u0027driver-core-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core\n\nPull driver core fix from Greg KH:\n \"Here is a single driver core fix for a reported problem with differed\n probing. It has been in linux-next for a while with no reported\n problems\"\n\n* tag \u0027driver-core-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:\n driver core: clear deferred probe reason on probe retry\n" }, { "commit": "a443930a3e8eaf70c8768cc2c923d1d85b71262f", "tree": "abbe7ebe213902d322cfa0862ba21461217566ab", "parents": [ "3e707eb6b8aa715847c17279f9c17c8dca2d639b", "3756b6578709c55819742f6ba0c18f93e8901397" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:05:16 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:05:16 2021 -0700" }, "message": "Merge tag \u0027char-misc-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char/misc driver fixes from Greg KH:\n \"Here are a few small driver char/misc changes for 5.12-rc6.\n\n Nothing major here, a few fixes for reported issues:\n\n - interconnect fixes for problems found\n\n - fbcon syzbot-found fix\n\n - extcon fixes\n\n - firmware stratix10 bugfix\n\n - MAINTAINERS file update.\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027char-misc-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n drivers: video: fbcon: fix NULL dereference in fbcon_cursor()\n mei: allow map and unmap of client dma buffer only for disconnected client\n MAINTAINERS: Add linux-phy list and patchwork\n interconnect: Fix kerneldoc warning\n firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0\n extcon: Fix error handling in extcon_dev_register\n extcon: Add stubs for extcon_register_notifier_all() functions\n interconnect: core: fix error return code of icc_link_destroy()\n interconnect: qcom: msm8939: remove rpm-ids from non-RPM nodes\n" }, { "commit": "3e707eb6b8aa715847c17279f9c17c8dca2d639b", "tree": "416f745496b68fec8e6270a3e7d7d52f844df882", "parents": [ "0d2c5a9e15179806fd9ff24409d2c0d689fc422f", "e78836ae76d20f38eed8c8c67f21db97529949da" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:03:51 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:03:51 2021 -0700" }, "message": "Merge tag \u0027staging-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging driver fixes from Greg KH:\n \"Here are two rtl8192e staging driver fixes for reported problems.\n\n Both of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027staging-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:\n staging: rtl8192e: Change state information from u16 to u8\n staging: rtl8192e: Fix incorrect source in memcpy()\n" }, { "commit": "0d2c5a9e15179806fd9ff24409d2c0d689fc422f", "tree": "ad9fd3f7a70da76cb5a45af128ba48ed15b7ca53", "parents": [ "de879a8d37ff86b8111f94a64fc82c81c1dddd89", "29d96eb261345c8d888e248ae79484e681be2faa" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:00:53 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 10:00:53 2021 -0700" }, "message": "Merge tag \u0027tty-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull serial driver fix from Greg KH:\n \"Here is a single serial driver fix for 5.12-rc6. Is is a revert of a\n change that showed up in 5.9 that has been reported to cause problems.\n\n It has been in linux-next for a while with no reported issues\"\n\n* tag \u0027tty-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n soc: qcom-geni-se: Cleanup the code to remove proxy votes\n" }, { "commit": "de879a8d37ff86b8111f94a64fc82c81c1dddd89", "tree": "f59bf761cfe04d55557df97f8e943291bd37d706", "parents": [ "57fbdb15ec427ca3a6f35d4b71fc90ca9af301ea", "93f672804bf2d7a49ef3fd96827ea6290ca1841e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 09:56:22 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 09:56:22 2021 -0700" }, "message": "Merge tag \u0027usb-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB fixes from Greg KH:\n \"Here are a few small USB driver fixes for 5.12-rc6 to resolve reported\n problems.\n\n They include:\n\n - a number of cdc-acm fixes for reported problems. It seems more\n people are using this driver lately...\n\n - dwc3 driver fixes for reported problems, and fixes for the fixes :)\n\n - dwc2 driver fixes for reported issues.\n\n - musb driver fix.\n\n - new USB quirk additions.\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027usb-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (23 commits)\n usb: dwc2: Prevent core suspend when port connection flag is 0\n usb: dwc2: Fix HPRT0.PrtSusp bit setting for HiKey 960 board.\n usb: musb: Fix suspend with devices connected for a64\n usb: xhci-mtk: fix broken streams issue on 0.96 xHCI\n usb: dwc3: gadget: Clear DEP flags after stop transfers in ep disable\n usbip: vhci_hcd fix shift out-of-bounds in vhci_hub_control()\n USB: quirks: ignore remote wake-up on Fibocom L850-GL LTE modem\n USB: cdc-acm: do not log successful probe on later errors\n USB: cdc-acm: always claim data interface\n USB: cdc-acm: use negation for NULL checks\n USB: cdc-acm: clean up probe error labels\n USB: cdc-acm: drop redundant driver-data reset\n USB: cdc-acm: drop redundant driver-data assignment\n USB: cdc-acm: fix use-after-free after probe failure\n USB: cdc-acm: fix double free on probe failure\n USB: cdc-acm: downgrade message to debug\n USB: cdc-acm: untangle a circular dependency between callback and softint\n cdc-acm: fix BREAK rx code path adding necessary calls\n usb: gadget: udc: amd5536udc_pci fix null-ptr-dereference\n usb: dwc3: pci: Enable dis_uX_susphy_quirk for Intel Merrifield\n ...\n" }, { "commit": "57fbdb15ec427ca3a6f35d4b71fc90ca9af301ea", "tree": "2b32b6d00d3069339db03aafb5365f253c2941a4", "parents": [ "d93a0d43e3d0ba9e19387be4dae4a8d5b175a8d7", "9e67600ed6b8565da4b85698ec659b5879a6c1c6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 09:07:35 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Apr 03 09:07:35 2021 -0700" }, "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 \"A single fix to iscsi for a rare race condition which can cause a\n kernel panic\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: iscsi: Fix race condition between login and sync thread\n" }, { "commit": "e82ad4853948382d37ac512b27a3e70b6f01c103", "tree": "3ec5fd7f77d36b8f1429726f9ff5956ae9684daa", "parents": [ "230d50d448acb6639991440913299e50cacf1daf" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Apr 02 19:45:34 2021 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Apr 02 19:45:34 2021 -0600" }, "message": "io_uring: fix !CONFIG_BLOCK compilation failure\n\nkernel test robot correctly pinpoints a compilation failure if\nCONFIG_BLOCK isn\u0027t set:\n\nfs/io_uring.c: In function \u0027__io_complete_rw\u0027:\n\u003e\u003e fs/io_uring.c:2509:48: error: implicit declaration of function \u0027io_rw_should_reissue\u0027; did you mean \u0027io_rw_reissue\u0027? [-Werror\u003dimplicit-function-declaration]\n 2509 | if ((res \u003d\u003d -EAGAIN || res \u003d\u003d -EOPNOTSUPP) \u0026\u0026 io_rw_should_reissue(req)) {\n | ^~~~~~~~~~~~~~~~~~~~\n | io_rw_reissue\n cc1: some warnings being treated as errors\n\nEnsure that we have a stub declaration of io_rw_should_reissue() for\n!CONFIG_BLOCK.\n\nFixes: 230d50d448ac (\"io_uring: move reissue into regular IO path\")\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "d93a0d43e3d0ba9e19387be4dae4a8d5b175a8d7", "tree": "e49badc7daf845034102081c934aa767652aee17", "parents": [ "1faccb63949988e4cfdfba2e9d2c3476c6d690e5", "f06c609645ecd043c79380fac94145926603fb33" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 16:13:13 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 16:13:13 2021 -0700" }, "message": "Merge tag \u0027block-5.12-2021-04-02\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n\n - Remove comment that never came to fruition in 22 years of development\n (Christoph)\n\n - Remove unused request flag (Christoph)\n\n - Fix for null_blk fake timeout handling (Damien)\n\n - Fix for IOCB_NOWAIT being ignored for O_DIRECT on raw bdevs (Pavel)\n\n - Error propagation fix for multiple split bios (Yufen)\n\n* tag \u0027block-5.12-2021-04-02\u0027 of git://git.kernel.dk/linux-block:\n block: remove the unused RQF_ALLOCED flag\n block: update a few comments in uapi/linux/blkpg.h\n block: don\u0027t ignore REQ_NOWAIT for direct IO\n null_blk: fix command timeout completion handling\n block: only update parent bi_status when bio fail\n" }, { "commit": "1faccb63949988e4cfdfba2e9d2c3476c6d690e5", "tree": "3d780eb6d309f84adb9af24402039445d084a549", "parents": [ "0a84c2e440f74cbb2064084cc7ff4f74f5565ba7", "230d50d448acb6639991440913299e50cacf1daf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 16:08:19 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 16:08:19 2021 -0700" }, "message": "Merge tag \u0027io_uring-5.12-2021-04-02\u0027 of git://git.kernel.dk/linux-block\n\nPull io_uring fixes from Jens Axboe:\n \"Nothing really major in here, and finally nothing really related to\n signals. A few minor fixups related to the threading changes, and some\n general fixes, that\u0027s it.\n\n There\u0027s the pending gdb-get-confused-about-arch, but that\u0027s more of a\n cosmetic issue, nothing that hinder use of it. And given that other\n archs will likely be affected by that oddity too, better to postpone\n any changes there until 5.13 imho\"\n\n* tag \u0027io_uring-5.12-2021-04-02\u0027 of git://git.kernel.dk/linux-block:\n io_uring: move reissue into regular IO path\n io_uring: fix EIOCBQUEUED iter revert\n io_uring/io-wq: protect against sprintf overflow\n io_uring: don\u0027t mark S_ISBLK async work as unbounded\n io_uring: drop sqd lock before handling signals for SQPOLL\n io_uring: handle setup-failed ctx in kill_timeouts\n io_uring: always go for cancellation spin on exec\n" }, { "commit": "0a84c2e440f74cbb2064084cc7ff4f74f5565ba7", "tree": "936d07e859fedf80c4f3f20c073ff4734227ce3e", "parents": [ "9314a0e9c711b0c092158ee9e0ed24d5ea25c90a", "91463ebff32d3e6fc71134784ae6aa91402bfe3d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 15:34:17 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 15:34:17 2021 -0700" }, "message": "Merge tag \u0027acpi-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI fixes from Rafael Wysocki:\n \"These fix an ACPI tables management issue, an issue related to the\n ACPI enumeration of devices and CPU wakeup in the ACPI processor\n driver.\n\n Specifics:\n\n - Ensure that the memory occupied by ACPI tables on x86 will always\n be reserved to prevent it from being allocated for other purposes\n which was possible in some cases (Rafael Wysocki).\n\n - Fix the ACPI device enumeration code to prevent it from attempting\n to evaluate the _STA control method for devices with unmet\n dependencies which is likely to fail (Hans de Goede).\n\n - Fix the handling of CPU0 wakeup in the ACPI processor driver to\n prevent CPU0 online failures from occurring (Vitaly Kuznetsov)\"\n\n* tag \u0027acpi-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n ACPI: processor: Fix CPU0 wakeup in acpi_idle_play_dead()\n ACPI: scan: Fix _STA getting called on devices with unmet dependencies\n ACPI: tables: x86: Reserve memory occupied by ACPI tables\n" }, { "commit": "9314a0e9c711b0c092158ee9e0ed24d5ea25c90a", "tree": "8853c610c8d893f2e83ee2c1275a8e47b8abdd2d", "parents": [ "05de45383bd134fcb2b7d70d35ebb0bb50b5e4aa", "ac1790ad78f8f0cf9a588ffb530c700ad758e8b6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 15:17:08 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 15:17:08 2021 -0700" }, "message": "Merge tag \u0027pm-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fixes from Rafael Wysocki:\n \"These fix a race condition and an ordering issue related to using\n device links in the runtime PM framework and two kerneldoc comments in\n cpufreq.\n\n Specifics:\n\n - Fix race condition related to the handling of supplier devices\n during consumer device probe and fix the order of decrementation of\n two related reference counters in the runtime PM core code handling\n supplier devices (Adrian Hunter).\n\n - Fix kerneldoc comments in cpufreq that have not been updated along\n with the functions documented by them (Geert Uytterhoeven)\"\n\n* tag \u0027pm-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n PM: runtime: Fix race getting/putting suppliers at probe\n PM: runtime: Fix ordering in pm_runtime_get_suppliers()\n cpufreq: Fix scaling_{available,boost}_frequencies_show() comments\n" }, { "commit": "f06c609645ecd043c79380fac94145926603fb33", "tree": "19edd2da80633c6db547bafaa089397cdb963d58", "parents": [ "b9c6cdc37ee1fe5866d3b1c10efb9d03191a76af" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Fri Apr 02 19:17:46 2021 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Apr 02 11:18:31 2021 -0600" }, "message": "block: remove the unused RQF_ALLOCED flag\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "b9c6cdc37ee1fe5866d3b1c10efb9d03191a76af", "tree": "054be6209a4cc48b7204e541eb420ee5373fc315", "parents": [ "f8b78caf21d5bc3fcfc40c18898f9d52ed1451a5" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Fri Apr 02 19:17:31 2021 +0200" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Apr 02 11:18:30 2021 -0600" }, "message": "block: update a few comments in uapi/linux/blkpg.h\n\nThe big top of the file comment talk about grand plans that never\nhappened, so remove them to not confuse the readers. Also mark the\ndevname and volname fields as ignored as they were never used by the\nkernel.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "05de45383bd134fcb2b7d70d35ebb0bb50b5e4aa", "tree": "e22acb05830c78538a5b98589a6de4c118df4536", "parents": [ "1678e493d530e7977cce34e59a86bb86f3c5631e", "9deb193af69d3fd6dd8e47f292b67c805a787010" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 08:39:00 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Apr 02 08:39:00 2021 -0700" }, "message": "Merge tag \u0027trace-v5.12-rc5-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fix from Steven Rostedt:\n \"Fix stack trace entry size to stop showing garbage\n\n The macro that creates both the structure and the format displayed to\n user space for the stack trace event was changed a while ago to fix\n the parsing by user space tooling. But this change also modified the\n structure used to store the stack trace event. It changed the caller\n array field from [0] to [8].\n\n Even though the size in the ring buffer is dynamic and can be\n something other than 8 (user space knows how to handle this), the 8\n extra words was not accounted for when reserving the event on the ring\n buffer, and added 8 more entries, due to the calculation of\n \"sizeof(*entry) + nr_entries * sizeof(long)\", as the sizeof(*entry)\n now contains 8 entries.\n\n The size of the caller field needs to be subtracted from the size of\n the entry to create the correct allocation size\"\n\n* tag \u0027trace-v5.12-rc5-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n tracing: Fix stack trace event size\n" }, { "commit": "230d50d448acb6639991440913299e50cacf1daf", "tree": "80c9957edc04cc1203dd3741c26193417dcd697e", "parents": [ "07204f21577a1d882f0259590c3553fe6a476381" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Apr 01 20:41:15 2021 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Apr 02 09:24:20 2021 -0600" }, "message": "io_uring: move reissue into regular IO path\n\nIt\u0027s non-obvious how retry is done for block backed files, when it happens\noff the kiocb done path. It also makes it tricky to deal with the iov_iter\nhandling.\n\nJust mark the req as needing a reissue, and handling it from the\nsubmission path instead. This makes it directly obvious that we\u0027re not\nre-importing the iovec from userspace past the submit point, and it means\nthat we can just reuse our usual -EAGAIN retry path from the read/write\nhandling.\n\nAt some point in the future, we\u0027ll gain the ability to always reliably\nreturn -EAGAIN through the stack. A previous attempt on the block side\ndidn\u0027t pan out and got reverted, hence the need to check for this\ninformation out-of-band right now.\n\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "91463ebff32d3e6fc71134784ae6aa91402bfe3d", "tree": "6d0d63cf617d3cb8bd15cf90e9f08dc69244ba77", "parents": [ "8cdddd182bd7befae6af49c5fd612893f55d6ccb", "1a1c130ab7575498eed5bcf7220037ae09cd1f8a", "3e759425cc3cf9a43392309819d34c65a3644c59" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Apr 02 16:57:56 2021 +0200" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Apr 02 16:57:56 2021 +0200" }, "message": "Merge branches \u0027acpi-tables\u0027 and \u0027acpi-scan\u0027\n\n* acpi-tables:\n ACPI: tables: x86: Reserve memory occupied by ACPI tables\n\n* acpi-scan:\n ACPI: scan: Fix _STA getting called on devices with unmet dependencies\n" }, { "commit": "ac1790ad78f8f0cf9a588ffb530c700ad758e8b6", "tree": "68a14844bef8284f51442b3f829cc5d30e87d2bb", "parents": [ "9dfacc54a8661bc8be6e08cffee59596ec59f263", "a95d8f581901c560ff1b63bc02107eb34a6efb1b" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Apr 02 16:45:58 2021 +0200" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Apr 02 16:45:58 2021 +0200" }, "message": "Merge branch \u0027pm-cpufreq\u0027\n\n* pm-cpufreq:\n cpufreq: Fix scaling_{available,boost}_frequencies_show() comments\n" }, { "commit": "f8b78caf21d5bc3fcfc40c18898f9d52ed1451a5", "tree": "a0647e87fb73c8e7599cff05766f93f38d603de9", "parents": [ "de3510e52b0a398261271455562458003b8eea62" ], "author": { "name": "Pavel Begunkov", "email": "asml.silence@gmail.com", "time": "Fri Nov 20 17:10:28 2020 +0000" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Apr 02 08:34:30 2021 -0600" }, "message": "block: don\u0027t ignore REQ_NOWAIT for direct IO\n\nIf IOCB_NOWAIT is set on submission, then that needs to get propagated to\nREQ_NOWAIT on the block side. Otherwise we completely lose this\ninformation, and any issuer of IOCB_NOWAIT IO will potentially end up\nblocking on eg request allocation on the storage side.\n\nSigned-off-by: Pavel Begunkov \u003casml.silence@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "1adbc2941eee8acbe3c7dc6b51cdbc5a9bf19565", "tree": "4ebf29b0783b84963ae1f8f3f9239fa88db8c6da", "parents": [ "9d8c7d92015ece9a2139a259cef781a41845d2c0" ], "author": { "name": "Kefeng Wang", "email": "wangkefeng.wang@huawei.com", "time": "Tue Mar 30 21:25:31 2021 +0800" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Thu Apr 01 21:37:08 2021 -0700" }, "message": "riscv: Make NUMA depend on MMU\n\nNUMA is useless when NOMMU, and it leads some build error,\nmake it depend on MMU.\n\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nSigned-off-by: Kefeng Wang \u003cwangkefeng.wang@huawei.com\u003e\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "9d8c7d92015ece9a2139a259cef781a41845d2c0", "tree": "af3f2d15fab2fd2527b3452d37d308bf0cfa28e2", "parents": [ "ac8d0b901f0033b783156ab2dc1a0e73ec42409b" ], "author": { "name": "Yang Li", "email": "yang.lee@linux.alibaba.com", "time": "Mon Mar 22 16:38:36 2021 +0800" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Thu Apr 01 21:37:06 2021 -0700" }, "message": "riscv: remove unneeded semicolon\n\nEliminate the following coccicheck warning:\n./arch/riscv/mm/kasan_init.c:219:2-3: Unneeded semicolon\n\nReported-by: Abaci Robot \u003cabaci@linux.alibaba.com\u003e\nSigned-off-by: Yang Li \u003cyang.lee@linux.alibaba.com\u003e\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "ac8d0b901f0033b783156ab2dc1a0e73ec42409b", "tree": "c30bba6025078108ef26dadf653ac8083b0941af", "parents": [ "285a76bb2cf51b0c74c634f2aaccdb93e1f2a359" ], "author": { "name": "Zihao Yu", "email": "yuzihao@ict.ac.cn", "time": "Wed Mar 17 16:17:25 2021 +0800" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Thu Apr 01 21:37:05 2021 -0700" }, "message": "riscv,entry: fix misaligned base for excp_vect_table\n\nIn RV64, the size of each entry in excp_vect_table is 8 bytes. If the\nbase of the table is not 8-byte aligned, loading an entry in the table\nwill raise a misaligned exception. Although such exception will be\nhandled by opensbi/bbl, this still causes performance degradation.\n\nSigned-off-by: Zihao Yu \u003cyuzihao@ict.ac.cn\u003e\nReviewed-by: Anup Patel \u003canup@brainfault.org\u003e\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "285a76bb2cf51b0c74c634f2aaccdb93e1f2a359", "tree": "a771004fd36a9ed36d9c819b89eec528f67fa08f", "parents": [ "23c1075ae83adaf14ea3f727c40368799f80bccc" ], "author": { "name": "Ben Dooks", "email": "ben.dooks@codethink.co.uk", "time": "Mon Mar 29 10:57:49 2021 +0100" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Thu Apr 01 21:37:04 2021 -0700" }, "message": "riscv: evaluate put_user() arg before enabling user access\n\nThe \u003casm/uaccess.h\u003e header has a problem with put_user(a, ptr) if\nthe \u0027a\u0027 is not a simple variable, such as a function. This can lead\nto the compiler producing code as so:\n\n1:\tenable_user_access()\n2:\tevaluate \u0027a\u0027 into register \u0027r\u0027\n3:\tput \u0027r\u0027 to \u0027ptr\u0027\n4:\tdisable_user_acess()\n\nThe issue is that \u0027a\u0027 is now being evaluated with the user memory\nprotections disabled. So we try and force the evaulation by assigning\n\u0027x\u0027 to __val at the start, and hoping the compiler barriers in\n enable_user_access() do the job of ordering step 2 before step 1.\n\nThis has shown up in a bug where \u0027a\u0027 sleeps and thus schedules out\nand loses the SR_SUM flag. This isn\u0027t sufficient to fully fix, but\nshould reduce the window of opportunity. The first instance of this\nwe found is in scheudle_tail() where the code does:\n\n$ less -N kernel/sched/core.c\n\n4263 if (current-\u003eset_child_tid)\n4264 put_user(task_pid_vnr(current), current-\u003eset_child_tid);\n\nHere, the task_pid_vnr(current) is called within the block that has\nenabled the user memory access. This can be made worse with KASAN\nwhich makes task_pid_vnr() a rather large call with plenty of\nopportunity to sleep.\n\nSigned-off-by: Ben Dooks \u003cben.dooks@codethink.co.uk\u003e\nReported-by: syzbot+e74b94fe601ab9552d69@syzkaller.appspotmail.com\nSuggested-by: Arnd Bergman \u003carnd@arndb.de\u003e\n\n--\nChanges since v1:\n- fixed formatting and updated the patch description with more info\n\nChanges since v2:\n- fixed commenting on __put_user() (schwab@linux-m68k.org)\n\nChange since v3:\n- fixed RFC in patch title. Should be ready to merge.\n\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "23c1075ae83adaf14ea3f727c40368799f80bccc", "tree": "9205e422692ca9b03ddf72092fd030063e30baf9", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3" ], "author": { "name": "Kefeng Wang", "email": "wangkefeng.wang@huawei.com", "time": "Wed Mar 17 23:08:38 2021 +0800" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Thu Apr 01 21:37:03 2021 -0700" }, "message": "riscv: Drop const annotation for sp\n\nThe const annotation should not be used for \u0027sp\u0027, or it will\nbecome read only and lead to bad stack output.\n\nFixes: dec822771b01 (\"riscv: stacktrace: Move register keyword to beginning of declaration\")\nSigned-off-by: Kefeng Wang \u003cwangkefeng.wang@huawei.com\u003e\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "1678e493d530e7977cce34e59a86bb86f3c5631e", "tree": "738494d912fa09953de941a6c8752e3a7cf3f2f5", "parents": [ "6905b1dc3c32a094f0da61bd656a740f0a97d592", "6a3193cdd5e5b96ac65f04ee42555c216da332af" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 17:57:43 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 17:57:43 2021 -0700" }, "message": "Merge tag \u0027lto-v5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux\n\nPull LTO fix from Kees Cook:\n \"It seems that there is a bug in ld.bfd when doing module section\n merging.\n\n As explicit merging is only needed for LTO, the work-around is to only\n do it under LTO, leaving the original section layout choices alone\n under normal builds:\n\n - Only perform explicit module section merges under LTO (Sean\n Christopherson)\"\n\n* tag \u0027lto-v5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:\n kbuild: lto: Merge module sections if and only if CONFIG_LTO_CLANG is enabled\n" }, { "commit": "6a3193cdd5e5b96ac65f04ee42555c216da332af", "tree": "b61f86cf72c116c8df62e778709fa3c8037db509", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3" ], "author": { "name": "Sean Christopherson", "email": "seanjc@google.com", "time": "Mon Mar 22 16:44:38 2021 -0700" }, "committer": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Thu Apr 01 14:15:59 2021 -0700" }, "message": "kbuild: lto: Merge module sections if and only if CONFIG_LTO_CLANG is enabled\n\nMerge module sections only when using Clang LTO. With ld.bfd, merging\nsections does not appear to update the symbol tables for the module,\ne.g. \u0027readelf -s\u0027 shows the value that a symbol would have had, if\nsections were not merged. ld.lld does not show this problem.\n\nThe stale symbol table breaks gdb\u0027s function disassembler, and presumably\nother things, e.g.\n\n gdb -batch -ex \"file arch/x86/kvm/kvm.ko\" -ex \"disassemble kvm_init\"\n\nreads the wrong bytes and dumps garbage.\n\nFixes: dd2776222abb (\"kbuild: lto: merge module sections\")\nCc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nSigned-off-by: Sean Christopherson \u003cseanjc@google.com\u003e\nReviewed-by: Sami Tolvanen \u003csamitolvanen@google.com\u003e\nTested-by: Sami Tolvanen \u003csamitolvanen@google.com\u003e\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nLink: https://lore.kernel.org/r/20210322234438.502582-1-seanjc@google.com\n" }, { "commit": "6905b1dc3c32a094f0da61bd656a740f0a97d592", "tree": "980c9c459b46dedf175a173f30582dee3cd29a14", "parents": [ "a80314c327a937ff1213288adf0d11414c40a898", "55626ca9c6909d077eca71bccbe15fef6e5ad917" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 12:42:55 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 12:42:55 2021 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nPull kvm fixes from Paolo Bonzini:\n \"It\u0027s a bit larger than I (and probably you) would like by the time we\n get to -rc6, but perhaps not entirely unexpected since the changes in\n the last merge window were larger than usual.\n\n x86:\n - Fixes for missing TLB flushes with TDP MMU\n\n - Fixes for race conditions in nested SVM\n\n - Fixes for lockdep splat with Xen emulation\n\n - Fix for kvmclock underflow\n\n - Fix srcdir !\u003d builddir builds\n\n - Other small cleanups\n\n ARM:\n - Fix GICv3 MMIO compatibility probing\n\n - Prevent guests from using the ARMv8.4 self-hosted tracing\n extension\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm:\n selftests: kvm: Check that TSC page value is small after KVM_SET_CLOCK(0)\n KVM: x86: Prevent \u0027hv_clock-\u003esystem_time\u0027 from going negative in kvm_guest_time_update()\n KVM: x86: disable interrupts while pvclock_gtod_sync_lock is taken\n KVM: x86: reduce pvclock_gtod_sync_lock critical sections\n KVM: SVM: ensure that EFER.SVME is set when running nested guest or on nested vmexit\n KVM: SVM: load control fields from VMCB12 before checking them\n KVM: x86/mmu: Don\u0027t allow TDP MMU to yield when recovering NX pages\n KVM: x86/mmu: Ensure TLBs are flushed for TDP MMU during NX zapping\n KVM: x86/mmu: Ensure TLBs are flushed when yielding during GFN range zap\n KVM: make: Fix out-of-source module builds\n selftests: kvm: make hardware_disable_test less verbose\n KVM: x86/vPMU: Forbid writing to MSR_F15H_PERF MSRs when guest doesn\u0027t have X86_FEATURE_PERFCTR_CORE\n KVM: x86: remove unused declaration of kvm_write_tsc()\n KVM: clean up the unused argument\n tools/kvm_stat: Add restart delay\n KVM: arm64: Fix CPU interface MMIO compatibility detection\n KVM: arm64: Disable guest access to trace filter controls\n KVM: arm64: Hide system instruction access to Trace registers\n" }, { "commit": "a80314c327a937ff1213288adf0d11414c40a898", "tree": "28a6c8e1d85a394fcb73104128f1c3d6189a2849", "parents": [ "ffd9fb546d498b2105dca6efc58a5cd0c17f0807", "6fdb8e5aba6a33fe5f1a0bd1bcf0cf2884437ead" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 12:19:03 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 12:19:03 2021 -0700" }, "message": "Merge tag \u0027drm-fixes-2021-04-02\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Things have settled down in time for Easter, a random smattering of\n small fixes across a few drivers.\n\n I\u0027m guessing though there might be some i915 and misc fixes out there\n I haven\u0027t gotten yet, but since today is a public holiday here, I\u0027m\n sending this early so I can have the day off, I\u0027ll see if more\n requests come in and decide what to do with them later.\n\n amdgpu:\n - Polaris idle power fix\n - VM fix\n - Vangogh S3 fix\n - Fixes for non-4K page sizes\n\n amdkfd:\n - dqm fence memory corruption fix\n\n tegra:\n - lockdep warning fix\n - runtine PM reference fix\n - display controller fix\n - PLL Fix\n\n imx:\n - memory leak in error path fix\n - LDB driver channel registration fix\n - oob array warning in LDB driver\n\n exynos\n - unused header file removal\"\n\n* tag \u0027drm-fixes-2021-04-02\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/amdgpu: check alignment on CPU page for bo map\n drm/amdgpu: Set a suitable dev_info.gart_page_size\n drm/amdgpu/vangogh: don\u0027t check for dpm in is_dpm_running when in suspend\n drm/amdkfd: dqm fence memory corruption\n drm/tegra: sor: Grab runtime PM reference across reset\n drm/tegra: dc: Restore coupling of display controllers\n gpu: host1x: Use different lock classes for each client\n drm/tegra: dc: Don\u0027t set PLL clock to 0Hz\n drm/amdgpu: fix offset calculation in amdgpu_vm_bo_clear_mappings()\n drm/amd/pm: no need to force MCLK to highest when no display connected\n drm/exynos/decon5433: Remove the unused include statements\n drm/imx: imx-ldb: fix out of bounds array access warning\n drm/imx: imx-ldb: Register LDB channel1 when it is the only channel to be used\n drm/imx: fix memory leak when fails to init\n" }, { "commit": "6fdb8e5aba6a33fe5f1a0bd1bcf0cf2884437ead", "tree": "08448614a633042bdcfcfcb7334377196f190346", "parents": [ "a0497251f2b055a137d62ed065286ba999647b3c", "33ce7f2f95cabb5834cf0906308a5cb6103976da" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 02 04:52:45 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 02 04:53:16 2021 +1000" }, "message": "Merge tag \u0027imx-drm-fixes-2021-04-01\u0027 of git://git.pengutronix.de/git/pza/linux into drm-fixes\n\ndrm/imx: imx-drm-core and imx-ldb fixes\n\nFix a memory leak in an error path during DRM device initialization,\nfix the LDB driver to register channel 1 even if channel 0 is unused,\nand fix an out of bounds array access warning in the LDB driver.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210401092235.GA13586@pengutronix.de\n" }, { "commit": "a0497251f2b055a137d62ed065286ba999647b3c", "tree": "743ee0d6bb8e8f017b2a8b1633e4b037a2ed3c4b", "parents": [ "dcdb7aa452ef9503a9a90dc807475e2d5b78ba21", "ac097aecfef0bb289ca53d2fe0b73fc7e1612a05" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 02 04:44:28 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Apr 02 04:44:39 2021 +1000" }, "message": "Merge tag \u0027drm/tegra/for-5.12-rc6\u0027 of ssh://git.freedesktop.org/git/tegra/linux into drm-fixes\n\ndrm/tegra: Fixes for v5.12-rc6\n\nThis contains a couple of fixes for various issues such as lockdep\nwarnings, runtime PM references, coupled display controllers and\nmisconfigured PLLs.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Thierry Reding \u003cthierry.reding@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210401163352.3348296-1-thierry.reding@gmail.com\n" }, { "commit": "9deb193af69d3fd6dd8e47f292b67c805a787010", "tree": "01d0f654ae5dc0806ec0e186a13eb22e58449054", "parents": [ "59300b36f85f254260c81d9dd09195fa49eb0f98" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Apr 01 13:54:40 2021 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Apr 01 14:06:33 2021 -0400" }, "message": "tracing: Fix stack trace event size\n\nCommit cbc3b92ce037 fixed an issue to modify the macros of the stack trace\nevent so that user space could parse it properly. Originally the stack\ntrace format to user space showed that the called stack was a dynamic\narray. But it is not actually a dynamic array, in the way that other\ndynamic event arrays worked, and this broke user space parsing for it. The\nupdate was to make the array look to have 8 entries in it. Helper\nfunctions were added to make it parse it correctly, as the stack was\ndynamic, but was determined by the size of the event stored.\n\nAlthough this fixed user space on how it read the event, it changed the\ninternal structure used for the stack trace event. It changed the array\nsize from [0] to [8] (added 8 entries). This increased the size of the\nstack trace event by 8 words. The size reserved on the ring buffer was the\nsize of the stack trace event plus the number of stack entries found in\nthe stack trace. That commit caused the amount to be 8 more than what was\nneeded because it did not expect the caller field to have any size. This\nproduced 8 entries of garbage (and reading random data) from the stack\ntrace event:\n\n \u003cidle\u003e-0 [002] d... 1976396.837549: \u003cstack trace\u003e\n \u003d\u003e trace_event_raw_event_sched_switch\n \u003d\u003e __traceiter_sched_switch\n \u003d\u003e __schedule\n \u003d\u003e schedule_idle\n \u003d\u003e do_idle\n \u003d\u003e cpu_startup_entry\n \u003d\u003e secondary_startup_64_no_verify\n \u003d\u003e 0xc8c5e150ffff93de\n \u003d\u003e 0xffff93de\n \u003d\u003e 0\n \u003d\u003e 0\n \u003d\u003e 0xc8c5e17800000000\n \u003d\u003e 0x1f30affff93de\n \u003d\u003e 0x00000004\n \u003d\u003e 0x200000000\n\nInstead, subtract the size of the caller field from the size of the event\nto make sure that only the amount needed to store the stack trace is\nreserved.\n\nLink: https://lore.kernel.org/lkml/your-ad-here.call-01617191565-ext-9692@work.hours/\n\nCc: stable@vger.kernel.org\nFixes: cbc3b92ce037 (\"tracing: Set kernel_stack\u0027s caller size properly\")\nReported-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nTested-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nAcked-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ffd9fb546d498b2105dca6efc58a5cd0c17f0807", "tree": "cad3b9dfd2bda76fe55c3c47689d54402513cbf4", "parents": [ "5d17c1ba3ee7271346f459d9015c13e7a5fb2b39", "417eadfdd9e25188465280edf3668ed163fda2d0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 10:09:31 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 10:09:31 2021 -0700" }, "message": "Merge tag \u0027sound-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound\n\nPull sound fixes from Takashi Iwai:\n \"Things seem calming down, only usual device-specific fixes for\n HD-audio and USB-audio at this time\"\n\n* tag \u0027sound-5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:\n ALSA: hda/realtek: fix mute/micmute LEDs for HP 640 G8\n ALSA: hda: Add missing sanity checks in PM prepare/complete callbacks\n ALSA: hda: Re-add dropped snd_poewr_change_state() calls\n ALSA: usb-audio: Apply sample rate quirk to Logitech Connect\n ALSA: hda/realtek: call alc_update_headset_mode() in hp_automute_hook\n ALSA: hda/realtek: fix a determine_headset_type issue for a Dell AIO\n" }, { "commit": "5d17c1ba3ee7271346f459d9015c13e7a5fb2b39", "tree": "75d46bc3cae0df2f18f707cc3d2c798d74b2d2a8", "parents": [ "e8d1895883da8048266a7da43b0a381588dca87d", "4e53d1701b574b1ee9d500b4913a1ece2fac8911" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 09:39:51 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 09:39:51 2021 -0700" }, "message": "Merge tag \u0027tomoyo-pr-20210401\u0027 of git://git.osdn.net/gitroot/tomoyo/tomoyo-test1\n\nPull tomory fix from Tetsuo Handa:\n \"An update on \u0027tomoyo: recognize kernel threads correctly\u0027 from Jens\n Axboe to not special case PF_IO_WORKER for PF_KTHREAD\"\n\n* tag \u0027tomoyo-pr-20210401\u0027 of git://git.osdn.net/gitroot/tomoyo/tomoyo-test1:\n tomoyo: don\u0027t special case PF_IO_WORKER for PF_KTHREAD\n" }, { "commit": "e8d1895883da8048266a7da43b0a381588dca87d", "tree": "41dfe90d7650bab60546b96df542b44db056cce1", "parents": [ "d19cc4bfbff1ae72c3505a00fb8ce0d3fa519e6c", "2c7e57a02708a69d0194f9ef2a7b7e54f5a0484a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 09:32:18 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Apr 01 09:32:18 2021 -0700" }, "message": "Merge tag \u0027xarray-5.12\u0027 of git://git.infradead.org/users/willy/xarray\n\nPull XArray fixes from Matthew Wilcox:\n \"My apologies for the lateness of this. I had a bug reported in the\n test suite, and when I started working on it, I realised I had two\n fixes sitting in the xarray tree since last November. Anyway,\n everything here is fixes, apart from adding xa_limit_16b. The test\n suite passes.\n\n Summary:\n\n - Fix a bug when splitting to a non-zero order\n\n - Documentation fix\n\n - Add a predefined 16-bit allocation limit\n\n - Various test suite fixes\"\n\n* tag \u0027xarray-5.12\u0027 of git://git.infradead.org/users/willy/xarray:\n idr test suite: Improve reporting from idr_find_test_1\n idr test suite: Create anchor before launching throbber\n idr test suite: Take RCU read lock in idr_find_test_1\n radix tree test suite: Register the main thread with the RCU library\n radix tree test suite: Fix compilation\n XArray: Add xa_limit_16b\n XArray: Fix splitting to non-zero orders\n XArray: Fix split documentation\n" }, { "commit": "07204f21577a1d882f0259590c3553fe6a476381", "tree": "a581bb61e2722e70687801c9c9ec7c907ffaa24c", "parents": [ "696ee88a7c50f96573f98aa76cc74286033140c1" ], "author": { "name": "Pavel Begunkov", "email": "asml.silence@gmail.com", "time": "Thu Apr 01 12:18:48 2021 +0100" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Apr 01 09:31:21 2021 -0600" }, "message": "io_uring: fix EIOCBQUEUED iter revert\n\niov_iter_revert() is done in completion handlers that happensf before\nread/write returns -EIOCBQUEUED, no need to repeat reverting afterwards.\nMoreover, even though it may appear being just a no-op, it\u0027s actually\nraces with 1) user forging a new iovec of a different size 2) reissue,\nthat is done via io-wq continues completely asynchronously.\n\nFixes: 3e6a0d3c7571c (\"io_uring: fix -EAGAIN retry with IOPOLL\")\nSigned-off-by: Pavel Begunkov \u003casml.silence@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "696ee88a7c50f96573f98aa76cc74286033140c1", "tree": "b215f2544b7f3c95374a7a1ebaa85262296d05e5", "parents": [ "4b982bd0f383db9132e892c0c5144117359a6289" ], "author": { "name": "Pavel Begunkov", "email": "asml.silence@gmail.com", "time": "Thu Apr 01 09:55:04 2021 +0100" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Apr 01 09:21:18 2021 -0600" }, "message": "io_uring/io-wq: protect against sprintf overflow\n\ntask_pid may be large enough to not fit into the left space of\nTASK_COMM_LEN-sized buffers and overflow in sprintf. We not so care\nabout uniqueness, so replace it with safer snprintf().\n\nReported-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nSigned-off-by: Pavel Begunkov \u003casml.silence@gmail.com\u003e\nLink: https://lore.kernel.org/r/1702c6145d7e1c46fbc382f28334c02e1a3d3994.1617267273.git.asml.silence@gmail.com\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "4b982bd0f383db9132e892c0c5144117359a6289", "tree": "9b069f16d20a21e408f6d3821372f0c7acdf9d3d", "parents": [ "82734c5b1b24c020d701cf90ccb075e43a5ccb07" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Apr 01 08:38:34 2021 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Apr 01 08:56:28 2021 -0600" }, "message": "io_uring: don\u0027t mark S_ISBLK async work as unbounded\n\nS_ISBLK is marked as unbounded work for async preparation, because it\ndoesn\u0027t match S_ISREG. That is incorrect, as any read/write to a block\ndevice is also a bounded operation. Fix it up and ensure that S_ISBLK\nisn\u0027t marked unbounded.\n\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "de3510e52b0a398261271455562458003b8eea62", "tree": "c79cf634290c61e154d18058bff8c4e4eadd5e90", "parents": [ "3edf5346e4f2ce2fa0c94651a90a8dda169565ee" ], "author": { "name": "Damien Le Moal", "email": "damien.lemoal@wdc.com", "time": "Thu Apr 01 07:52:44 2021 +0900" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Apr 01 07:03:46 2021 -0600" }, "message": "null_blk: fix command timeout completion handling\n\nMemory backed or zoned null block devices may generate actual request\ntimeout errors due to the submission path being blocked on memory\nallocation or zone locking. Unlike fake timeouts or injected timeouts,\nthe request submission path will call blk_mq_complete_request() or\nblk_mq_end_request() for these real timeout errors, causing a double\ncompletion and use after free situation as the block layer timeout\nhandler executes blk_mq_rq_timed_out() and __blk_mq_free_request() in\nblk_mq_check_expired(). This problem often triggers a NULL pointer\ndereference such as:\n\nBUG: kernel NULL pointer dereference, address: 0000000000000050\nRIP: 0010:blk_mq_sched_mark_restart_hctx+0x5/0x20\n...\nCall Trace:\n dd_finish_request+0x56/0x80\n blk_mq_free_request+0x37/0x130\n null_handle_cmd+0xbf/0x250 [null_blk]\n ? null_queue_rq+0x67/0xd0 [null_blk]\n blk_mq_dispatch_rq_list+0x122/0x850\n __blk_mq_do_dispatch_sched+0xbb/0x2c0\n __blk_mq_sched_dispatch_requests+0x13d/0x190\n blk_mq_sched_dispatch_requests+0x30/0x60\n __blk_mq_run_hw_queue+0x49/0x90\n process_one_work+0x26c/0x580\n worker_thread+0x55/0x3c0\n ? process_one_work+0x580/0x580\n kthread+0x134/0x150\n ? kthread_create_worker_on_cpu+0x70/0x70\n ret_from_fork+0x1f/0x30\n\nThis problem very often triggers when running the full btrfs xfstests\non a memory-backed zoned null block device in a VM with limited amount\nof memory.\n\nAvoid this by executing blk_mq_complete_request() in null_timeout_rq()\nonly for commands that are marked for a fake timeout completion using\nthe fake_timeout boolean in struct null_cmd. For timeout errors injected\nthrough debugfs, the timeout handler will execute\nblk_mq_complete_request()i as before. This is safe as the submission\npath does not execute complete requests in this case.\n\nIn null_timeout_rq(), also make sure to set the command error field to\nBLK_STS_TIMEOUT and to propagate this error through to the request\ncompletion.\n\nReported-by: Johannes Thumshirn \u003cJohannes.Thumshirn@wdc.com\u003e\nSigned-off-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\nTested-by: Johannes Thumshirn \u003cJohannes.Thumshirn@wdc.com\u003e\nReviewed-by: Johannes Thumshirn \u003cJohannes.Thumshirn@wdc.com\u003e\nLink: https://lore.kernel.org/r/20210331225244.126426-1-damien.lemoal@wdc.com\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "2c7e57a02708a69d0194f9ef2a7b7e54f5a0484a", "tree": "4d747492461ee5616597ec53fe0d2ccd654d37cc", "parents": [ "094ffbd1d8eaa27ed426feb8530cb1456348b018" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:50:42 2021 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:50:42 2021 -0400" }, "message": "idr test suite: Improve reporting from idr_find_test_1\n\nInstead of just reporting an assertion failure, report enough information\nthat we can start diagnosing exactly went wrong.\n\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "094ffbd1d8eaa27ed426feb8530cb1456348b018", "tree": "e6b15a6235168d002fe515f9f5dd1754cdd2b149", "parents": [ "703586410da69eb40062e64d413ca33bd735917a" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:46:49 2021 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:50:19 2021 -0400" }, "message": "idr test suite: Create anchor before launching throbber\n\nThe throbber could race with creation of the anchor entry and cause the\nIDR to have zero entries in it, which would cause the test to fail.\n\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "703586410da69eb40062e64d413ca33bd735917a", "tree": "1519f5488d549bdd0333767b243c31a48382165d", "parents": [ "1bb4bd266cf39fd2fa711f2d265c558b92df1119" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:44:48 2021 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:44:48 2021 -0400" }, "message": "idr test suite: Take RCU read lock in idr_find_test_1\n\nWhen run on a single CPU, this test would frequently access already-freed\nmemory. Due to timing, this bug never showed up on multi-CPU tests.\n\nReported-by: Chris von Recklinghausen \u003ccrecklin@redhat.com\u003e\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "1bb4bd266cf39fd2fa711f2d265c558b92df1119", "tree": "9c38ae5cb23856d0efcef808ac667a41aedc14b5", "parents": [ "7487de534dcbe143e6f41da751dd3ffcf93b00ee" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Wed Mar 31 14:59:19 2021 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Apr 01 07:41:30 2021 -0400" }, "message": "radix tree test suite: Register the main thread with the RCU library\n\nSeveral test runners register individual worker threads with the\nRCU library, but neglect to register the main thread, which can lead\nto objects being freed while the main thread is in what appears to be\nan RCU critical section.\n\nReported-by: Chris von Recklinghausen \u003ccrecklin@redhat.com\u003e\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "8cdddd182bd7befae6af49c5fd612893f55d6ccb", "tree": "d082d5e17f703d2d18dcd6b729e734b592e47849", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3" ], "author": { "name": "Vitaly Kuznetsov", "email": "vkuznets@redhat.com", "time": "Wed Mar 24 16:22:19 2021 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Apr 01 13:37:55 2021 +0200" }, "message": "ACPI: processor: Fix CPU0 wakeup in acpi_idle_play_dead()\n\nCommit 496121c02127 (\"ACPI: processor: idle: Allow probing on platforms\nwith one ACPI C-state\") broke CPU0 hotplug on certain systems, e.g.\nI\u0027m observing the following on AWS Nitro (e.g r5b.xlarge but other\ninstance types are affected as well):\n\n # echo 0 \u003e /sys/devices/system/cpu/cpu0/online\n # echo 1 \u003e /sys/devices/system/cpu/cpu0/online\n \u003c10 seconds delay\u003e\n -bash: echo: write error: Input/output error\n\nIn fact, the above mentioned commit only revealed the problem and did\nnot introduce it. On x86, to wakeup CPU an NMI is being used and\nhlt_play_dead()/mwait_play_dead() loops are prepared to handle it:\n\n\t/*\n\t * If NMI wants to wake up CPU0, start CPU0.\n\t */\n\tif (wakeup_cpu0())\n\t\tstart_cpu0();\n\ncpuidle_play_dead() -\u003e acpi_idle_play_dead() (which is now being called on\nsystems where it wasn\u0027t called before the above mentioned commit) serves\nthe same purpose but it doesn\u0027t have a path for CPU0. What happens now on\nwakeup is:\n - NMI is sent to CPU0\n - wakeup_cpu0_nmi() works as expected\n - we get back to while (1) loop in acpi_idle_play_dead()\n - safe_halt() puts CPU0 to sleep again.\n\nThe straightforward/minimal fix is add the special handling for CPU0 on x86\nand that\u0027s what the patch is doing.\n\nFixes: 496121c02127 (\"ACPI: processor: idle: Allow probing on platforms with one ACPI C-state\")\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nCc: 5.10+ \u003cstable@vger.kernel.org\u003e # 5.10+\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "55626ca9c6909d077eca71bccbe15fef6e5ad917", "tree": "b52364355ce6812cb99333e47f47d4411710b6c2", "parents": [ "77fcbe823f002ad18426545351fa2fb94f8d5e61" ], "author": { "name": "Vitaly Kuznetsov", "email": "vkuznets@redhat.com", "time": "Fri Mar 26 16:55:51 2021 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:14:19 2021 -0400" }, "message": "selftests: kvm: Check that TSC page value is small after KVM_SET_CLOCK(0)\n\nAdd a test for the issue when KVM_SET_CLOCK(0) call could cause\nTSC page value to go very big because of a signedness issue around\nhv_clock-\u003esystem_time.\n\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nMessage-Id: \u003c20210326155551.17446-3-vkuznets@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "77fcbe823f002ad18426545351fa2fb94f8d5e61", "tree": "23ce711101f3aab33c6fea0e336d947bf34ea49a", "parents": [ "a83829f56c7ce17d5d05370820e185d9a23d3090" ], "author": { "name": "Vitaly Kuznetsov", "email": "vkuznets@redhat.com", "time": "Wed Mar 31 14:41:29 2021 +0200" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:14:19 2021 -0400" }, "message": "KVM: x86: Prevent \u0027hv_clock-\u003esystem_time\u0027 from going negative in kvm_guest_time_update()\n\nWhen guest time is reset with KVM_SET_CLOCK(0), it is possible for\n\u0027hv_clock-\u003esystem_time\u0027 to become a small negative number. This happens\nbecause in KVM_SET_CLOCK handling we set \u0027kvm-\u003earch.kvmclock_offset\u0027 based\non get_kvmclock_ns(kvm) but when KVM_REQ_CLOCK_UPDATE is handled,\nkvm_guest_time_update() does (masterclock in use case):\n\nhv_clock.system_time \u003d ka-\u003emaster_kernel_ns + v-\u003ekvm-\u003earch.kvmclock_offset;\n\nAnd \u0027master_kernel_ns\u0027 represents the last time when masterclock\ngot updated, it can precede KVM_SET_CLOCK() call. Normally, this is not a\nproblem, the difference is very small, e.g. I\u0027m observing\nhv_clock.system_time \u003d -70 ns. The issue comes from the fact that\n\u0027hv_clock.system_time\u0027 is stored as unsigned and \u0027system_time / 100\u0027 in\ncompute_tsc_page_parameters() becomes a very big number.\n\nUse \u0027master_kernel_ns\u0027 instead of get_kvmclock_ns() when masterclock is in\nuse and get_kvmclock_base_ns() when it\u0027s not to prevent \u0027system_time\u0027 from\ngoing negative.\n\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nMessage-Id: \u003c20210331124130.337992-2-vkuznets@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "a83829f56c7ce17d5d05370820e185d9a23d3090", "tree": "e275a6777f41ee7b54f97b298265da2eb2d24583", "parents": [ "c2c647f91aec192f45f0849c225f134183cf4e90" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Mar 25 14:11:14 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:14:19 2021 -0400" }, "message": "KVM: x86: disable interrupts while pvclock_gtod_sync_lock is taken\n\npvclock_gtod_sync_lock can be taken with interrupts disabled if the\npreempt notifier calls get_kvmclock_ns to update the Xen\nrunstate information:\n\n spin_lock include/linux/spinlock.h:354 [inline]\n get_kvmclock_ns+0x25/0x390 arch/x86/kvm/x86.c:2587\n kvm_xen_update_runstate+0x3d/0x2c0 arch/x86/kvm/xen.c:69\n kvm_xen_update_runstate_guest+0x74/0x320 arch/x86/kvm/xen.c:100\n kvm_xen_runstate_set_preempted arch/x86/kvm/xen.h:96 [inline]\n kvm_arch_vcpu_put+0x2d8/0x5a0 arch/x86/kvm/x86.c:4062\n\nSo change the users of the spinlock to spin_lock_irqsave and\nspin_unlock_irqrestore.\n\nReported-by: syzbot+b282b65c2c68492df769@syzkaller.appspotmail.com\nFixes: 30b5c851af79 (\"KVM: x86/xen: Add support for vCPU runstate information\")\nCc: David Woodhouse \u003cdwmw@amazon.co.uk\u003e\nCc: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "c2c647f91aec192f45f0849c225f134183cf4e90", "tree": "9e300bec50aef907eed3af159066e32870d6f4e2", "parents": [ "6ebae23c072bb865048d2b2500ee1fae71944263" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Mar 25 14:05:11 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:14:19 2021 -0400" }, "message": "KVM: x86: reduce pvclock_gtod_sync_lock critical sections\n\nThere is no need to include changes to vcpu-\u003erequests into\nthe pvclock_gtod_sync_lock critical section. The changes to\nthe shared data structures (in pvclock_update_vm_gtod_copy)\nalready occur under the lock.\n\nCc: David Woodhouse \u003cdwmw@amazon.co.uk\u003e\nCc: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "6ebae23c072bb865048d2b2500ee1fae71944263", "tree": "e59348839c40db9eef5cadaa6a2f96c381bf9c86", "parents": [ "825e34d3c9e21ad1243b4464d23739de3c63a48b", "3c346c0c60ab06a021d1c0884a0ef494bc4ee3a7" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed Mar 31 07:50:54 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:14:05 2021 -0400" }, "message": "Merge branch \u0027kvm-fix-svm-races\u0027 into kvm-master\n" }, { "commit": "3c346c0c60ab06a021d1c0884a0ef494bc4ee3a7", "tree": "860cb0fd3c84ff826ff267151bfe6c0bf17abd10", "parents": [ "a58d9166a756a0f4a6618e4f593232593d6df134" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed Mar 31 06:28:01 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:11:35 2021 -0400" }, "message": "KVM: SVM: ensure that EFER.SVME is set when running nested guest or on nested vmexit\n\nFixing nested_vmcb_check_save to avoid all TOC/TOU races\nis a bit harder in released kernels, so do the bare minimum\nby avoiding that EFER.SVME is cleared. This is problematic\nbecause svm_set_efer frees the data structures for nested\nvirtualization if EFER.SVME is cleared.\n\nAlso check that EFER.SVME remains set after a nested vmexit;\nclearing it could happen if the bit is zero in the save area\nthat is passed to KVM_SET_NESTED_STATE (the save area of the\nnested state corresponds to the nested hypervisor\u0027s state\nand is restored on the next nested vmexit).\n\nCc: stable@vger.kernel.org\nFixes: 2fcf4876ada (\"KVM: nSVM: implement on demand allocation of the nested state\")\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "a58d9166a756a0f4a6618e4f593232593d6df134", "tree": "c002769f24a16115567ec82cdd84994bcda72d52", "parents": [ "2c85ebc57b3e1817b6ce1a6b703928e113a90442" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed Mar 31 06:24:43 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Thu Apr 01 05:09:31 2021 -0400" }, "message": "KVM: SVM: load control fields from VMCB12 before checking them\n\nAvoid races between check and use of the nested VMCB controls. This\nfor example ensures that the VMRUN intercept is always reflected to the\nnested hypervisor, instead of being processed by the host. Without this\npatch, it is possible to end up with svm-\u003enested.hsave pointing to\nthe MSR permission bitmap for nested guests.\n\nThis bug is CVE-2021-29657.\n\nReported-by: Felix Wilhelm \u003cfwilhelm@google.com\u003e\nCc: stable@vger.kernel.org\nFixes: 2fcf4876ada (\"KVM: nSVM: implement on demand allocation of the nested state\")\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "dcdb7aa452ef9503a9a90dc807475e2d5b78ba21", "tree": "02f4cbac7d0486f44bb3f7ad58d65334f77ca2d0", "parents": [ "7344c82777a753c87973b1554899592390a95781", "e3512fb67093fabdf27af303066627b921ee9bd8" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 01 15:04:58 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 01 15:04:58 2021 +1000" }, "message": "Merge tag \u0027amd-drm-fixes-5.12-2021-03-31\u0027 of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes\n\namd-drm-fixes-5.12-2021-03-31:\n\namdgpu:\n- Polaris idle power fix\n- VM fix\n- Vangogh S3 fix\n- Fixes for non-4K page sizes\n\namdkfd:\n- dqm fence memory corruption fix\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexander.deucher@amd.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20210401020057.17831-1-alexander.deucher@amd.com\n" }, { "commit": "7344c82777a753c87973b1554899592390a95781", "tree": "25a1a370ce25dad775487582c3a31d913b466c34", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3", "6161a435c1910d07ee00cc25af010889010e1f08" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 01 13:30:49 2021 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Thu Apr 01 13:31:11 2021 +1000" }, "message": "Merge tag \u0027exynos-drm-fixes-for-v5.12-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into drm-fixes\n\nJust one cleanup which drops of_gpio.h inclusion.\n- This header file isn\u0027t used anymore so drop it.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Inki Dae \u003cinki.dae@samsung.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/1617016858-14081-1-git-send-email-inki.dae@samsung.com\n" }, { "commit": "e3512fb67093fabdf27af303066627b921ee9bd8", "tree": "2aca66059fe370eaab15213dbecfcdf237ef788e", "parents": [ "566c6e25f957ebdb0b6e8073ee291049118f47fb" ], "author": { "name": "Xℹ Ruoyao", "email": "xry111@mengyan1223.wang", "time": "Tue Mar 30 23:33:34 2021 +0800" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Mar 31 21:53:38 2021 -0400" }, "message": "drm/amdgpu: check alignment on CPU page for bo map\n\nThe page table of AMDGPU requires an alignment to CPU page so we should\ncheck ioctl parameters for it. Return -EINVAL if some parameter is\nunaligned to CPU page, instead of corrupt the page table sliently.\n\nReviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\nSigned-off-by: Xi Ruoyao \u003cxry111@mengyan1223.wang\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "566c6e25f957ebdb0b6e8073ee291049118f47fb", "tree": "8ffb89b0f1cf66e52bc11e6b44cb5c8da574bf02", "parents": [ "6951c3e4a260f65a16433833d2511e8796dc8625" ], "author": { "name": "Huacai Chen", "email": "chenhc@lemote.com", "time": "Tue Mar 30 23:33:33 2021 +0800" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Mar 31 21:53:38 2021 -0400" }, "message": "drm/amdgpu: Set a suitable dev_info.gart_page_size\n\nIn Mesa, dev_info.gart_page_size is used for alignment and it was\nset to AMDGPU_GPU_PAGE_SIZE(4KB). However, the page table of AMDGPU\ndriver requires an alignment on CPU pages. So, for non-4KB page system,\ngart_page_size should be max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE).\n\nSigned-off-by: Rui Wang \u003cwangr@lemote.com\u003e\nSigned-off-by: Huacai Chen \u003cchenhc@lemote.com\u003e\nLink: https://github.com/loongson-community/linux-stable/commit/caa9c0a1\n[Xi: rebased for drm-next, use max_t for checkpatch,\n and reworded commit message.]\nSigned-off-by: Xi Ruoyao \u003cxry111@mengyan1223.wang\u003e\nBugLink: https://gitlab.freedesktop.org/drm/amd/-/issues/1549\nTested-by: Dan Horák \u003cdan@danny.cz\u003e\nReviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "6951c3e4a260f65a16433833d2511e8796dc8625", "tree": "a492642b843e53fb6c832123d7cd5cc70d1e2d0b", "parents": [ "e92049ae4548ba09e53eaa9c8f6964b07ea274c9" ], "author": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Fri Mar 26 16:56:07 2021 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Mar 31 21:53:38 2021 -0400" }, "message": "drm/amdgpu/vangogh: don\u0027t check for dpm in is_dpm_running when in suspend\n\nDo the same thing we do for Renoir. We can check, but since\nthe sbios has started DPM, it will always return true which\ncauses the driver to skip some of the SMU init when it shouldn\u0027t.\n\nReviewed-by: Zhan Liu \u003czhan.liu@amd.com\u003e\nAcked-by: Evan Quan \u003cevan.quan@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "e92049ae4548ba09e53eaa9c8f6964b07ea274c9", "tree": "67fc315e4aff60df091f148c9fc57b5c72aaab83", "parents": [ "5e61b84f9d3ddfba73091f9fbc940caae1c9eb22" ], "author": { "name": "Qu Huang", "email": "jinsdb@126.com", "time": "Thu Jan 28 20:14:25 2021 +0800" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Mar 31 21:53:25 2021 -0400" }, "message": "drm/amdkfd: dqm fence memory corruption\n\nAmdgpu driver uses 4-byte data type as DQM fence memory,\nand transmits GPU address of fence memory to microcode\nthrough query status PM4 message. However, query status\nPM4 message definition and microcode processing are all\nprocessed according to 8 bytes. Fence memory only allocates\n4 bytes of memory, but microcode does write 8 bytes of memory,\nso there is a memory corruption.\n\nChanges since v1:\n * Change dqm-\u003efence_addr as a u64 pointer to fix this issue,\nalso fix up query_status and amdkfd_fence_wait_timeout function\nuses 64 bit fence value to make them consistent.\n\nSigned-off-by: Qu Huang \u003cjinsdb@126.com\u003e\nReviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nSigned-off-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "3edf5346e4f2ce2fa0c94651a90a8dda169565ee", "tree": "0832502839bb11e70327e3afd42ee3a0d61c3358", "parents": [ "e82fc7855749aa197740a60ef22c492c41ea5d5f" ], "author": { "name": "Yufen Yu", "email": "yuyufen@huawei.com", "time": "Wed Mar 31 07:53:59 2021 -0400" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Mar 31 19:18:04 2021 -0600" }, "message": "block: only update parent bi_status when bio fail\n\nFor multiple split bios, if one of the bio is fail, the whole\nshould return error to application. But we found there is a race\nbetween bio_integrity_verify_fn and bio complete, which return\nio success to application after one of the bio fail. The race as\nfollowing:\n\nsplit bio(READ) kworker\n\nnvme_complete_rq\nblk_update_request //split error\u003d0\n bio_endio\n bio_integrity_endio\n queue_work(kintegrityd_wq, \u0026bip-\u003ebip_work);\n\n bio_integrity_verify_fn\n bio_endio //split bio\n __bio_chain_endio\n if (!parent-\u003ebi_status)\n\n \u003cinterrupt entry\u003e\n nvme_irq\n blk_update_request //parent error\u003d7\n req_bio_endio\n bio-\u003ebi_status \u003d 7 //parent bio\n \u003cinterrupt exit\u003e\n\n parent-\u003ebi_status \u003d 0\n parent-\u003ebi_end_io() // return bi_status\u003d0\n\nThe bio has been split as two: split and parent. When split\nbio completed, it depends on kworker to do endio, while\nbio_integrity_verify_fn have been interrupted by parent bio\ncomplete irq handler. Then, parent bio-\u003ebi_status which have\nbeen set in irq handler will overwrite by kworker.\n\nIn fact, even without the above race, we also need to conside\nthe concurrency beteen mulitple split bio complete and update\nthe same parent bi_status. Normally, multiple split bios will\nbe issued to the same hctx and complete from the same irq\nvector. But if we have updated queue map between multiple split\nbios, these bios may complete on different hw queue and different\nirq vector. Then the concurrency update parent bi_status may\ncause the final status error.\n\nSuggested-by: Keith Busch \u003ckbusch@kernel.org\u003e\nSigned-off-by: Yufen Yu \u003cyuyufen@huawei.com\u003e\nReviewed-by: Ming Lei \u003cming.lei@redhat.com\u003e\nLink: https://lore.kernel.org/r/20210331115359.1125679-1-yuyufen@huawei.com\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "d19cc4bfbff1ae72c3505a00fb8ce0d3fa519e6c", "tree": "a667a3358c76a97e2be749bbcaaaebcada289a1c", "parents": [ "39192106d4efd482f96a0be8b7aaae7ec150d9ee", "59300b36f85f254260c81d9dd09195fa49eb0f98" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 31 10:14:55 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 31 10:14:55 2021 -0700" }, "message": "Merge tag \u0027trace-v5.12-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull ftrace fix from Steven Rostedt:\n \"Add check of order \u003c 0 before calling free_pages()\n\n The function addresses that are traced by ftrace are stored in pages,\n and the size is held in a variable. If there\u0027s some error in creating\n them, the allocate ones will be freed. In this case, it is possible\n that the order of pages to be freed may end up being negative due to a\n size of zero passed to get_count_order(), and then that negative\n number will cause free_pages() to free a very large section.\n\n Make sure that does not happen\"\n\n* tag \u0027trace-v5.12-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n ftrace: Check if pages were allocated before calling free_pages()\n" }, { "commit": "39192106d4efd482f96a0be8b7aaae7ec150d9ee", "tree": "96b8be9017ab306c1cef69f641f91fc91a39ba9c", "parents": [ "5e46d1b78a03d52306f21f77a4e4a144b6d31486", "ba845907b23a6584e5944f6fbffda3efb010c28b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 31 10:09:44 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Mar 31 10:09:44 2021 -0700" }, "message": "Merge tag \u0027pinctrl-v5.12-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl\n\nPull pin control fixes from Linus Walleij:\n \"Some overly ripe fixes for the v5.12 kernel. I should have sent\n earlier but had my head stuck in GDB.\n\n All are driver fixes:\n\n - Fix up some Intel GPIO base calculations.\n\n - Fix a register offset in the Microchip driver.\n\n - Fix suspend/resume bug in the Rockchip driver.\n\n - Default pull up strength in the Qualcomm LPASS driver.\n\n - Fix two pingroup offsets in the Qualcomm SC7280 driver.\n\n - Fix SDC1 register offset in the Qualcomm SC7280 driver.\n\n - Fix a nasty string concatenation in the Qualcomm SDX55 driver.\n\n - Check the REVID register to see if the device is real or\n virtualized during virtualization in the Intel driver\"\n\n* tag \u0027pinctrl-v5.12-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:\n pinctrl: intel: check REVID register value for device presence\n pinctrl: qcom: fix unintentional string concatenation\n pinctrl: qcom: sc7280: Fix SDC1_RCLK configurations\n pinctrl: qcom: sc7280: Fix SDC_QDSD_PINGROUP and UFS_RESET offsets\n pinctrl: qcom: lpass lpi: use default pullup/strength values\n pinctrl: rockchip: fix restore error in resume\n pinctrl: microchip-sgpio: Fix wrong register offset for IRQ trigger\n pinctrl: intel: Show the GPIO base calculation explicitly\n" }, { "commit": "825e34d3c9e21ad1243b4464d23739de3c63a48b", "tree": "d31ae5ffb1277c2cef0c2ae717a1fe841449697e", "parents": [ "6fb3084ab5d9331cfadf07c59cf4a0bd4059bf4a", "33a3164161fc86b9cc238f7f2aa2ccb1d5559b1c" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed Mar 31 07:45:19 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed Mar 31 07:45:41 2021 -0400" }, "message": "Merge commit \u0027kvm-tdp-fix-flushes\u0027 into kvm-master\n" }, { "commit": "5e46d1b78a03d52306f21f77a4e4a144b6d31486", "tree": "026600fa8503b621e67faac085d5fea26f8ea881", "parents": [ "6ac86aae89289121db784161fe318819778f7f2a" ], "author": { "name": "Tetsuo Handa", "email": "penguin-kernel@i-love.sakura.ne.jp", "time": "Sun Mar 21 23:37:49 2021 +0900" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 14:27:32 2021 -0700" }, "message": "reiserfs: update reiserfs_xattrs_initialized() condition\n\nsyzbot is reporting NULL pointer dereference at reiserfs_security_init()\n[1], for commit ab17c4f02156c4f7 (\"reiserfs: fixup xattr_root caching\")\nis assuming that REISERFS_SB(s)-\u003exattr_root !\u003d NULL in\nreiserfs_xattr_jcreate_nblocks() despite that commit made\nREISERFS_SB(sb)-\u003epriv_root !\u003d NULL \u0026\u0026 REISERFS_SB(s)-\u003exattr_root \u003d\u003d NULL\ncase possible.\n\nI guess that commit 6cb4aff0a77cc0e6 (\"reiserfs: fix oops while creating\nprivroot with selinux enabled\") wanted to check xattr_root !\u003d NULL\nbefore reiserfs_xattr_jcreate_nblocks(), for the changelog is talking\nabout the xattr root.\n\n The issue is that while creating the privroot during mount\n reiserfs_security_init calls reiserfs_xattr_jcreate_nblocks which\n dereferences the xattr root. The xattr root doesn\u0027t exist, so we get\n an oops.\n\nTherefore, update reiserfs_xattrs_initialized() to check both the\nprivroot and the xattr root.\n\nLink: https://syzkaller.appspot.com/bug?id\u003d8abaedbdeb32c861dc5340544284167dd0e46cde # [1]\nReported-and-tested-by: syzbot \u003csyzbot+690cb1e51970435f9775@syzkaller.appspotmail.com\u003e\nSigned-off-by: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\nFixes: 6cb4aff0a77c (\"reiserfs: fix oops while creating privroot with selinux enabled\")\nAcked-by: Jeff Mahoney \u003cjeffm@suse.com\u003e\nAcked-by: Jan Kara \u003cjack@suse.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "82734c5b1b24c020d701cf90ccb075e43a5ccb07", "tree": "582d2b194fefa60fd97b52e0b7cd2b7010610ebf", "parents": [ "51520426f4bc3e61cbbf7a39ccf4e411b665002d" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Mon Mar 29 06:52:44 2021 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Tue Mar 30 14:36:46 2021 -0600" }, "message": "io_uring: drop sqd lock before handling signals for SQPOLL\n\nDon\u0027t call into get_signal() with the sqd mutex held, it\u0027ll fail if we\u0027re\nfreezing the task and we\u0027ll get complaints on locks still being held:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nWARNING: iou-sqp-8386/8387 still has locks held!\n5.12.0-rc4-syzkaller #0 Not tainted\n------------------------------------\n1 lock held by iou-sqp-8386/8387:\n #0: ffff88801e1d2470 (\u0026sqd-\u003elock){+.+.}-{3:3}, at: io_sq_thread+0x24c/0x13a0 fs/io_uring.c:6731\n\n stack backtrace:\n CPU: 1 PID: 8387 Comm: iou-sqp-8386 Not tainted 5.12.0-rc4-syzkaller #0\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011\n Call Trace:\n __dump_stack lib/dump_stack.c:79 [inline]\n dump_stack+0x141/0x1d7 lib/dump_stack.c:120\n try_to_freeze include/linux/freezer.h:66 [inline]\n get_signal+0x171a/0x2150 kernel/signal.c:2576\n io_sq_thread+0x8d2/0x13a0 fs/io_uring.c:6748\n\nFold the get_signal() case in with the parking checks, as we need to drop\nthe lock in both cases, and since we need to be checking for parking when\njuggling the lock anyway.\n\nReported-by: syzbot+796d767eb376810256f5@syzkaller.appspotmail.com\nFixes: dbe1bdbb39db (\"io_uring: handle signals for IO threads like a normal thread\")\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "3e759425cc3cf9a43392309819d34c65a3644c59", "tree": "727bdb7d45b951154224978e5684804602ff3685", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3" ], "author": { "name": "Hans de Goede", "email": "hdegoede@redhat.com", "time": "Tue Mar 30 20:49:32 2021 +0200" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Tue Mar 30 21:36:20 2021 +0200" }, "message": "ACPI: scan: Fix _STA getting called on devices with unmet dependencies\n\nCommit 71da201f38df (\"ACPI: scan: Defer enumeration of devices with\n_DEP lists\") dropped the following 2 lines from acpi_init_device_object():\n\n\t/* Assume there are unmet deps until acpi_device_dep_initialize() runs */\n\tdevice-\u003edep_unmet \u003d 1;\n\nLeaving the initial value of dep_unmet at the 0 from the kzalloc(). This\ncauses the acpi_bus_get_status() call in acpi_add_single_object() to\nactually call _STA, even though there maybe unmet deps, leading to errors\nlike these:\n\n[ 0.123579] ACPI Error: No handler for Region [ECRM] (00000000ba9edc4c)\n [GenericSerialBus] (20170831/evregion-166)\n[ 0.123601] ACPI Error: Region GenericSerialBus (ID\u003d9) has no handler\n (20170831/exfldio-299)\n[ 0.123618] ACPI Error: Method parse/execution failed\n \\_SB.I2C1.BAT1._STA, AE_NOT_EXIST (20170831/psparse-550)\n\nFix this by re-adding the dep_unmet \u003d 1 initialization to\nacpi_init_device_object() and modifying acpi_bus_check_add() to make sure\nthat dep_unmet always gets setup there, overriding the initial 1 value.\n\nThis re-fixes the issue initially fixed by\ncommit 63347db0affa (\"ACPI / scan: Use acpi_bus_get_status() to initialize\nACPI_TYPE_DEVICE devs\"), which introduced the removed\n\"device-\u003edep_unmet \u003d 1;\" statement.\n\nThis issue was noticed; and the fix tested on a Dell Venue 10 Pro 5055.\n\nFixes: 71da201f38df (\"ACPI: scan: Defer enumeration of devices with _DEP lists\")\nSuggested-by: Rafael J. Wysocki \u003crafael@kernel.org\u003e\nSigned-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nCc: 5.11+ \u003cstable@vger.kernel.org\u003e # 5.11+\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "6ac86aae89289121db784161fe318819778f7f2a", "tree": "7eb4d5d11fa77e502fd168853a658d92b1d0e787", "parents": [ "17860ccabff533748c85ea32904abd6bae990099", "84d572e634e28827d105746c922d8ada425e2d8b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 10:54:22 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 10:54:22 2021 -0700" }, "message": "Merge tag \u0027s390-5.12-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 updates from Heiko Carstens:\n\n - fix incorrect initialization and update of vdso data pages, which\n results in incorrect tod clock steering, and that\n clock_gettime(CLOCK_MONOTONIC_RAW, ...) returns incorrect values.\n\n - update MAINTAINERS for s390 vfio drivers\n\n* tag \u0027s390-5.12-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n MAINTAINERS: add backups for s390 vfio drivers\n s390/vdso: fix initializing and updating of vdso_data\n s390/vdso: fix tod_steering_delta type\n s390/vdso: copy tod_steering_delta value to vdso_data page\n" }, { "commit": "ac097aecfef0bb289ca53d2fe0b73fc7e1612a05", "tree": "578e0300e9c050e6405122f2fc925141ab9bbfef", "parents": [ "a31500fe7055451ed9043c8fff938dfa6f70ee37" ], "author": { "name": "Thierry Reding", "email": "treding@nvidia.com", "time": "Fri Mar 19 14:17:22 2021 +0100" }, "committer": { "name": "Thierry Reding", "email": "treding@nvidia.com", "time": "Tue Mar 30 19:51:39 2021 +0200" }, "message": "drm/tegra: sor: Grab runtime PM reference across reset\n\nThe SOR resets are exclusively shared with the SOR power domain. This\nmeans that exclusive access can only be granted temporarily and in order\nfor that to work, a rigorous sequence must be observed. To ensure that a\nsingle consumer gets exclusive access to a reset, each consumer must\nimplement a rigorous protocol using the reset_control_acquire() and\nreset_control_release() functions.\n\nHowever, these functions alone don\u0027t provide any guarantees at the\nsystem level. Drivers need to ensure that the only a single consumer has\naccess to the reset at the same time. In order for the SOR to be able to\nexclusively access its reset, it must therefore ensure that the SOR\npower domain is not powered off by holding on to a runtime PM reference\nto that power domain across the reset assert/deassert operation.\n\nThis used to work fine by accident, but was revealed when recently more\ndevices started to rely on the SOR power domain.\n\nFixes: 11c632e1cfd3 (\"drm/tegra: sor: Implement acquire/release for reset\")\nReported-by: Jonathan Hunter \u003cjonathanh@nvidia.com\u003e\nSigned-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n" }, { "commit": "7487de534dcbe143e6f41da751dd3ffcf93b00ee", "tree": "d0ea741d8933c2ed7862699fcb826fc1ab9c80d6", "parents": [ "df59d0a461bc5935232bf56a279e4d7a71c566a5" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Tue Mar 30 13:44:35 2021 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Tue Mar 30 13:44:35 2021 -0400" }, "message": "radix tree test suite: Fix compilation\n\nCommit 4bba4c4bb09a added tools/include/linux/compiler_types.h which\nincludes linux/compiler-gcc.h. Unfortunately, we had our own (empty)\ncompiler_types.h which overrode the one added by that commit, and\nso we lost the definition of __must_be_array(). Removing our empty\ncompiler_types.h fixes the problem and reduces our divergence from the\nrest of the tools.\n\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "df59d0a461bc5935232bf56a279e4d7a71c566a5", "tree": "f3997bffd62f1508714bfd896fdaee4e89b7d4b7", "parents": [ "3012110d71f41410932924e1d188f9eb57f1f824" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Tue Mar 30 13:40:27 2021 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Tue Mar 30 13:42:33 2021 -0400" }, "message": "XArray: Add xa_limit_16b\n\nA 16-bit limit is a more common limit than I had realised. Make it\ngenerally available.\n\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "3012110d71f41410932924e1d188f9eb57f1f824", "tree": "bdad57981759c2fe3fc329cef4577297ab9a76b8", "parents": [ "12efebab09e383bf366bec323bd296b7d90e1c43" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Thu Nov 19 08:32:31 2020 -0500" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Tue Mar 30 13:42:33 2021 -0400" }, "message": "XArray: Fix splitting to non-zero orders\n\nSplitting an order-4 entry into order-2 entries would leave the array\ncontaining pointers to 000040008000c000 instead of 000044448888cccc.\nThis is a one-character fix, but enhance the test suite to check this\ncase.\n\nReported-by: Zi Yan \u003cziy@nvidia.com\u003e\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "12efebab09e383bf366bec323bd296b7d90e1c43", "tree": "6213f469c529a0b7cb724f57fbebb396723fdc23", "parents": [ "17860ccabff533748c85ea32904abd6bae990099" ], "author": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Sat Oct 10 11:17:44 2020 -0400" }, "committer": { "name": "Matthew Wilcox (Oracle)", "email": "willy@infradead.org", "time": "Tue Mar 30 13:42:33 2021 -0400" }, "message": "XArray: Fix split documentation\n\nI wrote the documentation backwards; the new order of the entry is stored\nin the xas and the caller passes the old entry.\n\nReported-by: Zi Yan \u003cziy@nvidia.com\u003e\nSigned-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n" }, { "commit": "a31500fe7055451ed9043c8fff938dfa6f70ee37", "tree": "5d9292b52fc2ed77ef6f6dd91b8b40c5ab0bf96a", "parents": [ "a24f98176d1efae2c37d3438c57a624d530d9c33" ], "author": { "name": "Thierry Reding", "email": "treding@nvidia.com", "time": "Fri Mar 19 08:06:37 2021 +0100" }, "committer": { "name": "Thierry Reding", "email": "treding@nvidia.com", "time": "Tue Mar 30 19:40:43 2021 +0200" }, "message": "drm/tegra: dc: Restore coupling of display controllers\n\nCoupling of display controllers used to rely on runtime PM to take the\ncompanion controller out of reset. Commit fd67e9c6ed5a (\"drm/tegra: Do\nnot implement runtime PM\") accidentally broke this when runtime PM was\nremoved.\n\nRestore this functionality by reusing the hierarchical host1x client\nsuspend/resume infrastructure that\u0027s similar to runtime PM and which\nperfectly fits this use-case.\n\nFixes: fd67e9c6ed5a (\"drm/tegra: Do not implement runtime PM\")\nReported-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nReported-by: Paul Fertser \u003cfercerpav@gmail.com\u003e\nTested-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nSigned-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n" }, { "commit": "a24f98176d1efae2c37d3438c57a624d530d9c33", "tree": "47144591cb2d725eb77c59aaaa2103eb06b8892b", "parents": [ "f8fb97c915954fc6de6513cdf277103b5c6df7b3" ], "author": { "name": "Mikko Perttunen", "email": "mperttunen@nvidia.com", "time": "Mon Mar 29 16:38:27 2021 +0300" }, "committer": { "name": "Thierry Reding", "email": "treding@nvidia.com", "time": "Tue Mar 30 19:37:20 2021 +0200" }, "message": "gpu: host1x: Use different lock classes for each client\n\nTo avoid false lockdep warnings, give each client lock a different\nlock class, passed from the initialization site by macro.\n\nSigned-off-by: Mikko Perttunen \u003cmperttunen@nvidia.com\u003e\nSigned-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n" }, { "commit": "f8fb97c915954fc6de6513cdf277103b5c6df7b3", "tree": "b9409ee37de4ba274c863f43a7e20e5ea93c7bd0", "parents": [ "a38fd8748464831584a19438cbb3082b5a2dab15" ], "author": { "name": "Dmitry Osipenko", "email": "digetx@gmail.com", "time": "Tue Mar 02 16:15:06 2021 +0300" }, "committer": { "name": "Thierry Reding", "email": "treding@nvidia.com", "time": "Tue Mar 30 19:37:20 2021 +0200" }, "message": "drm/tegra: dc: Don\u0027t set PLL clock to 0Hz\n\nRGB output doesn\u0027t allow to change parent clock rate of the display and\nPCLK rate is set to 0Hz in this case. The tegra_dc_commit_state() shall\nnot set the display clock to 0Hz since this change propagates to the\nparent clock. The DISP clock is defined as a NODIV clock by the tegra-clk\ndriver and all NODIV clocks use the CLK_SET_RATE_PARENT flag.\n\nThis bug stayed unnoticed because by default PLLP is used as the parent\nclock for the display controller and PLLP silently skips the erroneous 0Hz\nrate changes because it always has active child clocks that don\u0027t permit\nrate changes. The PLLP isn\u0027t acceptable for some devices that we want to\nupstream (like Samsung Galaxy Tab and ASUS TF700T) due to a display panel\nclock rate requirements that can\u0027t be fulfilled by using PLLP and then the\nbug pops up in this case since parent clock is set to 0Hz, killing the\ndisplay output.\n\nDon\u0027t touch DC clock if pclk\u003d0 in order to fix the problem.\n\nSigned-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nSigned-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n" }, { "commit": "33a3164161fc86b9cc238f7f2aa2ccb1d5559b1c", "tree": "56dab11dd3b6f2a639998db7919db26e459fe21b", "parents": [ "048f49809c526348775425420fb5b8e84fd9a133" ], "author": { "name": "Sean Christopherson", "email": "seanjc@google.com", "time": "Thu Mar 25 13:01:19 2021 -0700" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:19:56 2021 -0400" }, "message": "KVM: x86/mmu: Don\u0027t allow TDP MMU to yield when recovering NX pages\n\nPrevent the TDP MMU from yielding when zapping a gfn range during NX\npage recovery. If a flush is pending from a previous invocation of the\nzapping helper, either in the TDP MMU or the legacy MMU, but the TDP MMU\nhas not accumulated a flush for the current invocation, then yielding\nwill release mmu_lock with stale TLB entries.\n\nThat being said, this isn\u0027t technically a bug fix in the current code, as\nthe TDP MMU will never yield in this case. tdp_mmu_iter_cond_resched()\nwill yield if and only if it has made forward progress, as defined by the\ncurrent gfn vs. the last yielded (or starting) gfn. Because zapping a\nsingle shadow page is guaranteed to (a) find that page and (b) step\nsideways at the level of the shadow page, the TDP iter will break its loop\nbefore getting a chance to yield.\n\nBut that is all very, very subtle, and will break at the slightest sneeze,\ne.g. zapping while holding mmu_lock for read would break as the TDP MMU\nwouldn\u0027t be guaranteed to see the present shadow page, and thus could step\nsideways at a lower level.\n\nCc: Ben Gardon \u003cbgardon@google.com\u003e\nSigned-off-by: Sean Christopherson \u003cseanjc@google.com\u003e\nMessage-Id: \u003c20210325200119.1359384-4-seanjc@google.com\u003e\n[Add lockdep assertion. - Paolo]\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "048f49809c526348775425420fb5b8e84fd9a133", "tree": "e0dfe4f3761e1b8bb77250314de82bf46833120b", "parents": [ "a835429cda91621fca915d80672a157b47738afb" ], "author": { "name": "Sean Christopherson", "email": "seanjc@google.com", "time": "Thu Mar 25 13:01:18 2021 -0700" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:19:55 2021 -0400" }, "message": "KVM: x86/mmu: Ensure TLBs are flushed for TDP MMU during NX zapping\n\nHonor the \"flush needed\" return from kvm_tdp_mmu_zap_gfn_range(), which\ndoes the flush itself if and only if it yields (which it will never do in\nthis particular scenario), and otherwise expects the caller to do the\nflush. If pages are zapped from the TDP MMU but not the legacy MMU, then\nno flush will occur.\n\nFixes: 29cf0f5007a2 (\"kvm: x86/mmu: NX largepage recovery for TDP MMU\")\nCc: stable@vger.kernel.org\nCc: Ben Gardon \u003cbgardon@google.com\u003e\nSigned-off-by: Sean Christopherson \u003cseanjc@google.com\u003e\nMessage-Id: \u003c20210325200119.1359384-3-seanjc@google.com\u003e\nReviewed-by: Ben Gardon \u003cbgardon@google.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "a835429cda91621fca915d80672a157b47738afb", "tree": "f7fc9e9c497b25df5d3ad9978ff47a7a7b290c94", "parents": [ "1e28eed17697bcf343c6743f0028cc3b5dd88bf0" ], "author": { "name": "Sean Christopherson", "email": "seanjc@google.com", "time": "Thu Mar 25 13:01:17 2021 -0700" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:19:55 2021 -0400" }, "message": "KVM: x86/mmu: Ensure TLBs are flushed when yielding during GFN range zap\n\nWhen flushing a range of GFNs across multiple roots, ensure any pending\nflush from a previous root is honored before yielding while walking the\ntables of the current root.\n\nNote, kvm_tdp_mmu_zap_gfn_range() now intentionally overwrites its local\n\"flush\" with the result to avoid redundant flushes. zap_gfn_range()\npreserves and return the incoming \"flush\", unless of course the flush was\nperformed prior to yielding and no new flush was triggered.\n\nFixes: 1af4a96025b3 (\"KVM: x86/mmu: Yield in TDU MMU iter even if no SPTES changed\")\nCc: stable@vger.kernel.org\nReviewed-by: Ben Gardon \u003cbgardon@google.com\u003e\nSigned-off-by: Sean Christopherson \u003cseanjc@google.com\u003e\nMessage-Id: \u003c20210325200119.1359384-2-seanjc@google.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "6fb3084ab5d9331cfadf07c59cf4a0bd4059bf4a", "tree": "69aefef61501e7a3bfe0d1f040705ac0336666b4", "parents": [ "f982fb62a304235397ec092936432199ca50ac4d" ], "author": { "name": "Siddharth Chandrasekaran", "email": "sidcha@amazon.de", "time": "Wed Mar 24 13:43:47 2021 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:07:10 2021 -0400" }, "message": "KVM: make: Fix out-of-source module builds\n\nBuilding kvm module out-of-source with,\n\n make -C $SRC O\u003d$BIN M\u003darch/x86/kvm\n\nfails to find \"irq.h\" as the include dir passed to cflags-y does not\nprefix the source dir. Fix this by prefixing $(srctree) to the include\ndir path.\n\nSigned-off-by: Siddharth Chandrasekaran \u003csidcha@amazon.de\u003e\nMessage-Id: \u003c20210324124347.18336-1-sidcha@amazon.de\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "f982fb62a304235397ec092936432199ca50ac4d", "tree": "9f8cf66ae155f42a7912fb0d8fb57f894cd3e7c2", "parents": [ "1973cadd4cca08eaeca944f60598f04ab0d80682" ], "author": { "name": "Vitaly Kuznetsov", "email": "vkuznets@redhat.com", "time": "Tue Mar 23 11:43:31 2021 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:07:10 2021 -0400" }, "message": "selftests: kvm: make hardware_disable_test less verbose\n\nhardware_disable_test produces 512 snippets like\n...\n main: [511] waiting semaphore\n run_test: [511] start vcpus\n run_test: [511] all threads launched\n main: [511] waiting 368us\n main: [511] killing child\n\nand this doesn\u0027t have much value, let\u0027s print this info with pr_debug().\n\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nMessage-Id: \u003c20210323104331.1354800-1-vkuznets@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "1973cadd4cca08eaeca944f60598f04ab0d80682", "tree": "bbac545a056e14aeabb38bef2c0145aa557f0c87", "parents": [ "ecaf088f53fcc893cd00c846f53042a536b9630d" ], "author": { "name": "Vitaly Kuznetsov", "email": "vkuznets@redhat.com", "time": "Tue Mar 23 09:45:15 2021 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:07:10 2021 -0400" }, "message": "KVM: x86/vPMU: Forbid writing to MSR_F15H_PERF MSRs when guest doesn\u0027t have X86_FEATURE_PERFCTR_CORE\n\nMSR_F15H_PERF_CTL0-5, MSR_F15H_PERF_CTR0-5 MSRs are only available when\nX86_FEATURE_PERFCTR_CORE CPUID bit was exposed to the guest. KVM, however,\nallows these MSRs unconditionally because kvm_pmu_is_valid_msr() -\u003e\namd_msr_idx_to_pmc() check always passes and because kvm_pmu_set_msr() -\u003e\namd_pmu_set_msr() doesn\u0027t fail.\n\nIn case of a counter (CTRn), no big harm is done as we only increase\ninternal PMC\u0027s value but in case of an eventsel (CTLn), we go deep into\nperf internals with a non-existing counter.\n\nNote, kvm_get_msr_common() just returns \u00270\u0027 when these MSRs don\u0027t exist\nand this also seems to contradict architectural behavior which is #GP\n(I did check one old Opteron host) but changing this status quo is a bit\nscarier.\n\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nMessage-Id: \u003c20210323084515.1346540-1-vkuznets@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "ecaf088f53fcc893cd00c846f53042a536b9630d", "tree": "95ccdf79c9f1744688bbe9e0476ee3a11e64f9f1", "parents": [ "d632826f26f2361e6ef18881611928036fac30e6" ], "author": { "name": "Dongli Zhang", "email": "dongli.zhang@oracle.com", "time": "Fri Mar 26 00:03:34 2021 -0700" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:07:09 2021 -0400" }, "message": "KVM: x86: remove unused declaration of kvm_write_tsc()\n\nkvm_write_tsc() was renamed and made static since commit 0c899c25d754\n(\"KVM: x86: do not attempt TSC synchronization on guest writes\"). Remove\nits unused declaration.\n\nSigned-off-by: Dongli Zhang \u003cdongli.zhang@oracle.com\u003e\nMessage-Id: \u003c20210326070334.12310-1-dongli.zhang@oracle.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "d632826f26f2361e6ef18881611928036fac30e6", "tree": "72128deea033e172ab8c7a9641ae42ff6dd48fae", "parents": [ "75f94ecbd0dfd2ac4e671f165f5ae864b7301422" ], "author": { "name": "Haiwei Li", "email": "lihaiwei@tencent.com", "time": "Sat Mar 13 13:10:32 2021 +0800" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:07:09 2021 -0400" }, "message": "KVM: clean up the unused argument\n\nkvm_msr_ignored_check function never uses vcpu argument. Clean up the\nfunction and invokers.\n\nSigned-off-by: Haiwei Li \u003clihaiwei@tencent.com\u003e\nMessage-Id: \u003c20210313051032.4171-1-lihaiwei.kernel@gmail.com\u003e\nReviewed-by: Keqian Zhu \u003czhukeqian1@huawei.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "75f94ecbd0dfd2ac4e671f165f5ae864b7301422", "tree": "27d31c6f1616fa92ae3dd8b753e95bcaee1e12d3", "parents": [ "41793e7f274e48c9c4a727e522eb22c875b77a0b" ], "author": { "name": "Stefan Raspl", "email": "raspl@linux.ibm.com", "time": "Thu Mar 25 13:29:49 2021 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:07:09 2021 -0400" }, "message": "tools/kvm_stat: Add restart delay\n\nIf this service is enabled and the system rebooted, Systemd\u0027s initial\nattempt to start this unit file may fail in case the kvm module is not\nloaded. Since we did not specify a delay for the retries, Systemd\nrestarts with a minimum delay a number of times before giving up and\ndisabling the service. Which means a subsequent kvm module load will\nhave kvm running without monitoring.\nAdding a delay to fix this.\n\nSigned-off-by: Stefan Raspl \u003craspl@linux.ibm.com\u003e\nMessage-Id: \u003c20210325122949.1433271-1-raspl@linux.ibm.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "41793e7f274e48c9c4a727e522eb22c875b77a0b", "tree": "3754de4a66f19db3f6b9c1486b409fc7b58ee9aa", "parents": [ "a5e13c6df0e41702d2b2c77c8ad41677ebb065b3", "af22df997d71c32304d6835a8b690281063b8010" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:06:42 2021 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Mar 30 13:06:42 2021 -0400" }, "message": "Merge tag \u0027kvmarm-fixes-5.12-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD\n\nKVM/arm64 fixes for 5.12, take #3\n\n- Fix GICv3 MMIO compatibility probing\n- Prevent guests from using the ARMv8.4 self-hosted tracing extension\n" }, { "commit": "17860ccabff533748c85ea32904abd6bae990099", "tree": "a693893cebd263144b6a35ebfd9c09dfbf152822", "parents": [ "e720e7d0e983bf05de80b231bccc39f1487f0f16", "e0146a108ce4d2c22b9510fd12268e3ee72a0161" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 09:49:36 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 09:49:36 2021 -0700" }, "message": "Merge tag \u0027vfio-v5.12-rc6\u0027 of git://github.com/awilliam/linux-vfio\n\nPull VFIO fixes from Alex Williamson:\n\n - Fix pfnmap batch carryover (Daniel Jordan)\n\n - Fix nvlink Kconfig dependency (Jason Gunthorpe)\n\n* tag \u0027vfio-v5.12-rc6\u0027 of git://github.com/awilliam/linux-vfio:\n vfio/nvlink: Add missing SPAPR_TCE_IOMMU depends\n vfio/type1: Empty batch for pfnmap pages\n" }, { "commit": "e720e7d0e983bf05de80b231bccc39f1487f0f16", "tree": "f58d4846d711e576e24fbd814580c38fce1edf75", "parents": [ "2bb25b3a748af6f11df42298e80b9863ed23f2b3" ], "author": { "name": "Ilya Lipnitskiy", "email": "ilya.lipnitskiy@gmail.com", "time": "Mon Mar 29 21:42:08 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 09:46:12 2021 -0700" }, "message": "mm: fix race by making init_zero_pfn() early_initcall\n\nThere are code paths that rely on zero_pfn to be fully initialized\nbefore core_initcall. For example, wq_sysfs_init() is a core_initcall\nfunction that eventually results in a call to kernel_execve, which\ncauses a page fault with a subsequent mmput. If zero_pfn is not\ninitialized by then it may not get cleaned up properly and result in an\nerror:\n\n BUG: Bad rss-counter state mm:(ptrval) type:MM_ANONPAGES val:1\n\nHere is an analysis of the race as seen on a MIPS device. On this\nparticular MT7621 device (Ubiquiti ER-X), zero_pfn is PFN 0 until\ninitialized, at which point it becomes PFN 5120:\n\n 1. wq_sysfs_init calls into kobject_uevent_env at core_initcall:\n kobject_uevent_env+0x7e4/0x7ec\n kset_register+0x68/0x88\n bus_register+0xdc/0x34c\n subsys_virtual_register+0x34/0x78\n wq_sysfs_init+0x1c/0x4c\n do_one_initcall+0x50/0x1a8\n kernel_init_freeable+0x230/0x2c8\n kernel_init+0x10/0x100\n ret_from_kernel_thread+0x14/0x1c\n\n 2. kobject_uevent_env() calls call_usermodehelper_exec() which executes\n kernel_execve asynchronously.\n\n 3. Memory allocations in kernel_execve cause a page fault, bumping the\n MM reference counter:\n add_mm_counter_fast+0xb4/0xc0\n handle_mm_fault+0x6e4/0xea0\n __get_user_pages.part.78+0x190/0x37c\n __get_user_pages_remote+0x128/0x360\n get_arg_page+0x34/0xa0\n copy_string_kernel+0x194/0x2a4\n kernel_execve+0x11c/0x298\n call_usermodehelper_exec_async+0x114/0x194\n\n 4. In case zero_pfn has not been initialized yet, zap_pte_range does\n not decrement the MM_ANONPAGES RSS counter and the BUG message is\n triggered shortly afterwards when __mmdrop checks the ref counters:\n __mmdrop+0x98/0x1d0\n free_bprm+0x44/0x118\n kernel_execve+0x160/0x1d8\n call_usermodehelper_exec_async+0x114/0x194\n ret_from_kernel_thread+0x14/0x1c\n\nTo avoid races such as described above, initialize init_zero_pfn at\nearly_initcall level. Depending on the architecture, ZERO_PAGE is\neither constant or gets initialized even earlier, at paging_init, so\nthere is no issue with initializing zero_pfn earlier.\n\nLink: https://lkml.kernel.org/r/CALCv0x2YqOXEAy2Q\u003dhafjhHCtTHVodChv1qpM\u003dniAXOpqEbt7w@mail.gmail.com\nSigned-off-by: Ilya Lipnitskiy \u003cilya.lipnitskiy@gmail.com\u003e\nCc: Hugh Dickins \u003chughd@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: stable@vger.kernel.org\nTested-by: 周琰杰 (Zhou Yanjie) \u003czhouyanjie@wanyeetech.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "2bb25b3a748af6f11df42298e80b9863ed23f2b3", "tree": "7f9dcdee96945742be43bd3ab7139a16dc67c629", "parents": [ "a080642d2f831cc34b68663c0db1c447d3807421", "9ae31e2ab293bf4d9c42e7079b156072f8a7f8ca" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 08:57:50 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 08:57:50 2021 -0700" }, "message": "Merge tag \u0027mips-fixes_5.12_3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux\n\nPull MIPS fix from Thomas Bogendoerfer:\n\n - Fix compile error with option MIPS_ELF_APPENDED_DTB\n\n* tag \u0027mips-fixes_5.12_3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:\n MIPS: kernel: setup.c: fix compilation error\n" }, { "commit": "a080642d2f831cc34b68663c0db1c447d3807421", "tree": "6fc9b8508e491f37e742c6bef4149dc65c50399f", "parents": [ "1e43c377a79f9189fea8f2711b399d4e8b4e609b", "a846738f8c3788d846ed1f587270d2f2e3d32432" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 08:55:47 2021 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Mar 30 08:55:47 2021 -0700" }, "message": "Merge tag \u0027for-linus-5.12b-rc6-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip\n\nPull xen fix from Juergen Gross:\n \"One Xen related security fix (XSA-371)\"\n\n* tag \u0027for-linus-5.12b-rc6-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:\n xen-blkback: don\u0027t leak persistent grants from xen_blkbk_map()\n" }, { "commit": "59300b36f85f254260c81d9dd09195fa49eb0f98", "tree": "a02466a8f75a624be15ce48ec686cab9d387782b", "parents": [ "83b62687a05205847d627f29126a8fee3c644335" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue Mar 30 09:58:38 2021 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue Mar 30 09:58:38 2021 -0400" }, "message": "ftrace: Check if pages were allocated before calling free_pages()\n\nIt is possible that on error pg-\u003esize can be zero when getting its order,\nwhich would return a -1 value. It is dangerous to pass in an order of -1\nto free_pages(). Check if order is greater than or equal to zero before\ncalling free_pages().\n\nLink: https://lore.kernel.org/lkml/20210330093916.432697c7@gandalf.local.home/\n\nReported-by: Abaci Robot \u003cabaci@linux.alibaba.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "9ae31e2ab293bf4d9c42e7079b156072f8a7f8ca", "tree": "cf2d58100f329a545d4d9411b659048e627b5a13", "parents": [ "3f6c515d723480bc8afd456b0a52438fe79128a8" ], "author": { "name": "Mauri Sandberg", "email": "sandberg@mailfence.com", "time": "Mon Mar 29 15:31:36 2021 +0300" }, "committer": { "name": "Thomas Bogendoerfer", "email": "tsbogend@alpha.franken.de", "time": "Tue Mar 30 14:51:23 2021 +0200" }, "message": "MIPS: kernel: setup.c: fix compilation error\n\nWith ath79_defconfig enabling CONFIG_MIPS_ELF_APPENDED_DTB gives a\ncompilation error. This patch fixes it.\n\nBuild log:\n...\n CC kernel/locking/percpu-rwsem.o\n../arch/mips/kernel/setup.c:46:39: error: conflicting types for\n\u0027__appended_dtb\u0027\n const char __section(\".appended_dtb\") __appended_dtb[0x100000];\n ^~~~~~~~~~~~~~\nIn file included from ../arch/mips/kernel/setup.c:34:\n../arch/mips/include/asm/bootinfo.h:118:13: note: previous declaration\nof \u0027__appended_dtb\u0027 was here\n extern char __appended_dtb[];\n ^~~~~~~~~~~~~~\n CC fs/attr.o\nmake[4]: *** [../scripts/Makefile.build:271: arch/mips/kernel/setup.o]\n Error 1\n...\n\nRoot cause seems to be:\nFixes: b83ba0b9df56 (\"MIPS: of: Introduce helper function to get DTB\")\n\nSigned-off-by: Mauri Sandberg \u003csandberg@mailfence.com\u003e\nReviewed-by: Thomas Bogendoerfer \u003ctsbogend@alpha.franken.de\u003e\nCc: trivial@kernel.org\nSigned-off-by: Thomas Bogendoerfer \u003ctsbogend@alpha.franken.de\u003e\n" }, { "commit": "417eadfdd9e25188465280edf3668ed163fda2d0", "tree": "fd2c92c8721489be73ef631c434095c5864a2027", "parents": [ "66affb7bb0dc0905155a1b2475261aa704d1ddb5" ], "author": { "name": "Jeremy Szu", "email": "jeremy.szu@canonical.com", "time": "Tue Mar 30 19:44:27 2021 +0800" }, "committer": { "name": "Takashi Iwai", "email": "tiwai@suse.de", "time": "Tue Mar 30 13:49:41 2021 +0200" }, "message": "ALSA: hda/realtek: fix mute/micmute LEDs for HP 640 G8\n\nThe HP EliteBook 640 G8 Notebook PC is using ALC236 codec which is\nusing 0x02 to control mute LED and 0x01 to control micmute LED.\nTherefore, add a quirk to make it works.\n\nSigned-off-by: Jeremy Szu \u003cjeremy.szu@canonical.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: https://lore.kernel.org/r/20210330114428.40490-1-jeremy.szu@canonical.com\nSigned-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n" } ], "next": "66affb7bb0dc0905155a1b2475261aa704d1ddb5" }