)]}' { "log": [ { "commit": "ab9a7e27044b87ff2be47b8f8e095400e7fccc44", "tree": "3a349c6d12b30f89c466fdb4e2fb70d3712d32dd", "parents": [ "e00d996a4317aff5351c4338dd97d390225412c2" ], "author": { "name": "Mark Rutland", "email": "mark.rutland@arm.com", "time": "Tue Mar 10 12:09:12 2020 +0000" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Wed Mar 18 23:28:17 2020 -0400" }, "message": "random: avoid warnings for !CONFIG_NUMA builds\n\nAs crng_initialize_secondary() is only called by do_numa_crng_init(),\nand the latter is under ifdeffery for CONFIG_NUMA, when CONFIG_NUMA is\nnot selected the compiler will warn that the former is unused:\n\n| drivers/char/random.c:820:13: warning: \u0027crng_initialize_secondary\u0027 defined but not used [-Wunused-function]\n| 820 | static void crng_initialize_secondary(struct crng_state *crng)\n| | ^~~~~~~~~~~~~~~~~~~~~~~~~\n\nStephen reports that this happens for x86_64 noallconfig builds.\n\nWe could move crng_initialize_secondary() and crng_init_try_arch() under\nthe CONFIG_NUMA ifdeffery, but this has the unfortunate property of\nseparating them from crng_initialize_primary() and\ncrng_init_try_arch_early() respectively. Instead, let\u0027s mark\ncrng_initialize_secondary() as __maybe_unused.\n\nLink: https://lore.kernel.org/r/20200310121747.GA49602@lakrids.cambridge.arm.com\nFixes: 5cbe0f13b51a (\"random: split primary/secondary crng init paths\")\nReported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "e00d996a4317aff5351c4338dd97d390225412c2", "tree": "11489ed4f80a83bcc579fcaae90e0542af494597", "parents": [ "69efea712f5b0489e67d07565aad5c94e09a3e52" ], "author": { "name": "Qian Cai", "email": "cai@lca.pw", "time": "Tue Feb 25 11:27:04 2020 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Feb 27 23:21:53 2020 -0500" }, "message": "random: fix data races at timer_rand_state\n\nFields in \"struct timer_rand_state\" could be accessed concurrently.\nLockless plain reads and writes result in data races. Fix them by adding\npairs of READ|WRITE_ONCE(). The data races were reported by KCSAN,\n\n BUG: KCSAN: data-race in add_timer_randomness / add_timer_randomness\n\n write to 0xffff9f320a0a01d0 of 8 bytes by interrupt on cpu 22:\n add_timer_randomness+0x100/0x190\n add_timer_randomness at drivers/char/random.c:1152\n add_disk_randomness+0x85/0x280\n scsi_end_request+0x43a/0x4a0\n scsi_io_completion+0xb7/0x7e0\n scsi_finish_command+0x1ed/0x2a0\n scsi_softirq_done+0x1c9/0x1d0\n blk_done_softirq+0x181/0x1d0\n __do_softirq+0xd9/0x57c\n irq_exit+0xa2/0xc0\n do_IRQ+0x8b/0x190\n ret_from_intr+0x0/0x42\n cpuidle_enter_state+0x15e/0x980\n cpuidle_enter+0x69/0xc0\n call_cpuidle+0x23/0x40\n do_idle+0x248/0x280\n cpu_startup_entry+0x1d/0x1f\n start_secondary+0x1b2/0x230\n secondary_startup_64+0xb6/0xc0\n\n no locks held by swapper/22/0.\n irq event stamp: 32871382\n _raw_spin_unlock_irqrestore+0x53/0x60\n _raw_spin_lock_irqsave+0x21/0x60\n _local_bh_enable+0x21/0x30\n irq_exit+0xa2/0xc0\n\n read to 0xffff9f320a0a01d0 of 8 bytes by interrupt on cpu 2:\n add_timer_randomness+0xe8/0x190\n add_disk_randomness+0x85/0x280\n scsi_end_request+0x43a/0x4a0\n scsi_io_completion+0xb7/0x7e0\n scsi_finish_command+0x1ed/0x2a0\n scsi_softirq_done+0x1c9/0x1d0\n blk_done_softirq+0x181/0x1d0\n __do_softirq+0xd9/0x57c\n irq_exit+0xa2/0xc0\n do_IRQ+0x8b/0x190\n ret_from_intr+0x0/0x42\n cpuidle_enter_state+0x15e/0x980\n cpuidle_enter+0x69/0xc0\n call_cpuidle+0x23/0x40\n do_idle+0x248/0x280\n cpu_startup_entry+0x1d/0x1f\n start_secondary+0x1b2/0x230\n secondary_startup_64+0xb6/0xc0\n\n no locks held by swapper/2/0.\n irq event stamp: 37846304\n _raw_spin_unlock_irqrestore+0x53/0x60\n _raw_spin_lock_irqsave+0x21/0x60\n _local_bh_enable+0x21/0x30\n irq_exit+0xa2/0xc0\n\n Reported by Kernel Concurrency Sanitizer on:\n Hardware name: HP ProLiant BL660c Gen9, BIOS I38 10/17/2018\n\nLink: https://lore.kernel.org/r/1582648024-13111-1-git-send-email-cai@lca.pw\nSigned-off-by: Qian Cai \u003ccai@lca.pw\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "69efea712f5b0489e67d07565aad5c94e09a3e52", "tree": "c433e3e0653a16ef5ade701b9a41cd5efaaeba31", "parents": [ "23ae0c17b89cfeb511aa0770735a319d35597072" ], "author": { "name": "Jason A. Donenfeld", "email": "Jason@zx2c4.com", "time": "Fri Feb 21 21:10:37 2020 +0100" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Feb 27 23:21:52 2020 -0500" }, "message": "random: always use batched entropy for get_random_u{32,64}\n\nIt turns out that RDRAND is pretty slow. Comparing these two\nconstructions:\n\n for (i \u003d 0; i \u003c CHACHA_BLOCK_SIZE; i +\u003d sizeof(ret))\n arch_get_random_long(\u0026ret);\n\nand\n\n long buf[CHACHA_BLOCK_SIZE / sizeof(long)];\n extract_crng((u8 *)buf);\n\nit amortizes out to 352 cycles per long for the top one and 107 cycles\nper long for the bottom one, on Coffee Lake Refresh, Intel Core i9-9880H.\n\nAnd importantly, the top one has the drawback of not benefiting from the\nreal rng, whereas the bottom one has all the nice benefits of using our\nown chacha rng. As get_random_u{32,64} gets used in more places (perhaps\nbeyond what it was originally intended for when it was introduced as\nget_random_{int,long} back in the md5 monstrosity era), it seems like it\nmight be a good thing to strengthen its posture a tiny bit. Doing this\nshould only be stronger and not any weaker because that pool is already\ninitialized with a bunch of rdrand data (when available). This way, we\nget the benefits of the hardware rng as well as our own rng.\n\nAnother benefit of this is that we no longer hit pitfalls of the recent\nstream of AMD bugs in RDRAND. One often used code pattern for various\nthings is:\n\n do {\n \tval \u003d get_random_u32();\n } while (hash_table_contains_key(val));\n\nThat recent AMD bug rendered that pattern useless, whereas we\u0027re really\nvery certain that chacha20 output will give pretty distributed numbers,\nno matter what.\n\nSo, this simplification seems better both from a security perspective\nand from a performance perspective.\n\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\nReviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nLink: https://lore.kernel.org/r/20200221201037.30231-1-Jason@zx2c4.com\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "23ae0c17b89cfeb511aa0770735a319d35597072", "tree": "da94eefee6cbdb6316b64e530860e29bacf41fd3", "parents": [ "ead5084cdf5af51445d219800c2ac8b01eb85f2f" ], "author": { "name": "Richard Henderson", "email": "richard.henderson@linaro.org", "time": "Mon Feb 10 13:00:15 2020 +0000" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Feb 27 23:21:52 2020 -0500" }, "message": "random: Make RANDOM_TRUST_CPU depend on ARCH_RANDOM\n\nListing the set of host architectures does not scale.\nDepend instead on the existence of the architecture rng.\n\nThis will allow RANDOM_TRUST_CPU to be selected on arm64. Today\nARCH_RANDOM is only selected by x86, s390, and powerpc, so this does not\nadversely affect other architectures.\n\nSigned-off-by: Richard Henderson \u003crichard.henderson@linaro.org\u003e\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nLink: https://lore.kernel.org/r/20200210130015.17664-5-mark.rutland@arm.com\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "ead5084cdf5af51445d219800c2ac8b01eb85f2f", "tree": "37129018903b602c137a0c7801a2a2eb241582c0", "parents": [ "253d3194c2b58152fe830fd27c2fd83ebc6fe5ee" ], "author": { "name": "Mark Rutland", "email": "mark.rutland@arm.com", "time": "Mon Feb 10 13:00:14 2020 +0000" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Feb 27 23:21:52 2020 -0500" }, "message": "arm64: add credited/trusted RNG support\n\nCurrently arm64 doesn\u0027t initialize the primary CRNG in a (potentially)\ntrusted manner as we only detect the presence of the RNG once secondary\nCPUs are up.\n\nNow that the core RNG code distinguishes the early initialization of the\nprimary CRNG, we can implement arch_get_random_seed_long_early() to\nsupport this.\n\nThis patch does so.\n\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Mark Brown \u003cbroonie@kernel.org\u003e\nCc: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nLink: https://lore.kernel.org/r/20200210130015.17664-4-mark.rutland@arm.com\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "253d3194c2b58152fe830fd27c2fd83ebc6fe5ee", "tree": "e9e6c341e7642b3e5cade3e18b57e2e049b75060", "parents": [ "5cbe0f13b51ac2fb2fd55902cff8d0077fc084c0" ], "author": { "name": "Mark Rutland", "email": "mark.rutland@arm.com", "time": "Mon Feb 10 13:00:13 2020 +0000" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Feb 27 23:21:52 2020 -0500" }, "message": "random: add arch_get_random_*long_early()\n\nSome architectures (e.g. arm64) can have heterogeneous CPUs, and the\nboot CPU may be able to provide entropy while secondary CPUs cannot. On\nsuch systems, arch_get_random_long() and arch_get_random_seed_long()\nwill fail unless support for RNG instructions has been detected on all\nCPUs. This prevents the boot CPU from being able to provide\n(potentially) trusted entropy when seeding the primary CRNG.\n\nTo make it possible to seed the primary CRNG from the boot CPU without\nadversely affecting the runtime versions of arch_get_random_long() and\narch_get_random_seed_long(), this patch adds new early versions of the\nfunctions used when initializing the primary CRNG.\n\nDefault implementations are provided atop of the existing\narch_get_random_long() and arch_get_random_seed_long() so that only\narchitectures with such constraints need to provide the new helpers.\n\nThere should be no functional change as a result of this patch.\n\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Mark Brown \u003cbroonie@kernel.org\u003e\nCc: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nLink: https://lore.kernel.org/r/20200210130015.17664-3-mark.rutland@arm.com\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "5cbe0f13b51ac2fb2fd55902cff8d0077fc084c0", "tree": "7ac004926a2a07309afa3fca58702ea376a742c4", "parents": [ "f8788d86ab28f61f7b46eb6be375f8a726783636" ], "author": { "name": "Mark Rutland", "email": "mark.rutland@arm.com", "time": "Mon Feb 10 13:00:12 2020 +0000" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Thu Feb 27 23:21:52 2020 -0500" }, "message": "random: split primary/secondary crng init paths\n\nCurrently crng_initialize() is used for both the primary CRNG and\nsecondary CRNGs. While we wish to share common logic, we need to do a\nnumber of additional things for the primary CRNG, and this would be\neasier to deal with were these handled in separate functions.\n\nThis patch splits crng_initialize() into crng_initialize_primary() and\ncrng_initialize_secondary(), with common logic factored out into a\ncrng_init_try_arch() helper.\n\nThere should be no functional change as a result of this patch.\n\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Mark Brown \u003cbroonie@kernel.org\u003e\nCc: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nLink: https://lore.kernel.org/r/20200210130015.17664-2-mark.rutland@arm.com\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "f8788d86ab28f61f7b46eb6be375f8a726783636", "tree": "05d39df8fb0dd6fabe8343fac9297d09316627b6", "parents": [ "d2eee25858f246051b49c42c411629c78513e2a8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 16:17:42 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 16:17:42 2020 -0800" }, "message": "Linux 5.6-rc3\n" }, { "commit": "d2eee25858f246051b49c42c411629c78513e2a8", "tree": "c80c21fcce7f67a108a1c9050d1860769ce745f9", "parents": [ "a3163ca03f9913ba2c2fb6a06305f3dca98adfd1", "a5ae50dea9111db63d30d700766dd5509602f7ad" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 09:43:50 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 09:43:50 2020 -0800" }, "message": "Merge tag \u0027for-5.6-rc2-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux\n\nPull btrfs fixes from David Sterba:\n \"These are fixes that were found during testing with help of error\n injection, plus some other stable material.\n\n There\u0027s a fixup to patch added to rc1 causing locking in wrong context\n warnings, tests found one more deadlock scenario. The patches are\n tagged for stable, two of them now in the queue but we\u0027d like all\n three released at the same time.\n\n I\u0027m not happy about fixes to fixes in such a fast succession during\n rcs, but I hope we found all the fallouts of commit 28553fa992cb\n (\u0027Btrfs: fix race between shrinking truncate and fiemap\u0027)\"\n\n* tag \u0027for-5.6-rc2-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:\n Btrfs: fix deadlock during fast fsync when logging prealloc extents beyond eof\n Btrfs: fix btrfs_wait_ordered_range() so that it waits for all ordered extents\n btrfs: fix bytes_may_use underflow in prealloc error condtition\n btrfs: handle logged extent failure properly\n btrfs: do not check delayed items are empty for single transaction cleanup\n btrfs: reset fs_root to NULL on error in open_ctree\n btrfs: destroy qgroup extent records on transaction abort\n" }, { "commit": "a3163ca03f9913ba2c2fb6a06305f3dca98adfd1", "tree": "64a91a3b12056688c929a8d11a09fda96aabab95", "parents": [ "c6188dff33fba320826e87e387ae6efffab0525d", "9db176bceb5c5df4990486709da386edadc6bd1d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 09:42:19 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 09:42:19 2020 -0800" }, "message": "Merge tag \u0027ext4_for_linus_stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4\n\nPull ext4 fixes from Ted Ts\u0027o:\n \"More miscellaneous ext4 bug fixes (all stable fodder)\"\n\n* tag \u0027ext4_for_linus_stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:\n ext4: fix mount failure with quota configured as module\n jbd2: fix ocfs2 corrupt when clearing block group bits\n ext4: fix race between writepages and enabling EXT4_EXTENTS_FL\n ext4: rename s_journal_flag_rwsem to s_writepages_rwsem\n ext4: fix potential race between s_flex_groups online resizing and access\n ext4: fix potential race between s_group_info online resizing and access\n ext4: fix potential race between online resizing and write operations\n ext4: add cond_resched() to __ext4_find_entry()\n ext4: fix a data race in EXT4_I(inode)-\u003ei_disksize\n" }, { "commit": "c6188dff33fba320826e87e387ae6efffab0525d", "tree": "650a315c620239054b5ee0f0af7d5aeb76cb257f", "parents": [ "dca132a60f226f4cbaa98807518a5ca6cff112ce", "99db590b083fa2bc60adfcb5c839a62db4ef1d79" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 09:37:41 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Feb 23 09:37:41 2020 -0800" }, "message": "Merge tag \u0027csky-for-linus-5.6-rc3\u0027 of git://github.com/c-sky/csky-linux\n\nPull csky updates from Guo Ren:\n \"Sorry, I missed 5.6-rc1 merge window, but in this pull request the\n most are the fixes and the rests are between fixes and features. The\n only outside modification is the MAINTAINERS file update with our\n mailing list.\n\n - cache flush implementation fixes\n\n - ftrace modify panic fix\n\n - CONFIG_SMP boot problem fix\n\n - fix pt_regs saving for atomic.S\n\n - fix fixaddr_init without highmem.\n\n - fix stack protector support\n\n - fix fake Tightly-Coupled Memory code compile and use\n\n - fix some typos and coding convention\"\n\n* tag \u0027csky-for-linus-5.6-rc3\u0027 of git://github.com/c-sky/csky-linux: (23 commits)\n csky: Replace \u003clinux/clk-provider.h\u003e by \u003clinux/of_clk.h\u003e\n csky: Implement copy_thread_tls\n csky: Add PCI support\n csky: Minimize defconfig to support buildroot config.fragment\n csky: Add setup_initrd check code\n csky: Cleanup old Kconfig options\n arch/csky: fix some Kconfig typos\n csky: Fixup compile warning for three unimplemented syscalls\n csky: Remove unused cache implementation\n csky: Fixup ftrace modify panic\n csky: Add flush_icache_mm to defer flush icache all\n csky: Optimize abiv2 copy_to_user_page with VM_EXEC\n csky: Enable defer flush_dcache_page for abiv2 cpus (807/810/860)\n csky: Remove unnecessary flush_icache_* implementation\n csky: Support icache flush without specific instructions\n csky/Kconfig: Add Kconfig.platforms to support some drivers\n csky/smp: Fixup boot failed when CONFIG_SMP\n csky: Set regs-\u003eusp to kernel sp, when the exception is from kernel\n csky/mm: Fixup export invalid_pte_table symbol\n csky: Separate fixaddr_init from highmem\n ...\n" }, { "commit": "99db590b083fa2bc60adfcb5c839a62db4ef1d79", "tree": "809ad2261dddd6f5dd9f096afda83656945ee6a1", "parents": [ "0b9f386c4be6493d282aab0af6f9b70c62142777" ], "author": { "name": "Geert Uytterhoeven", "email": "geert+renesas@glider.be", "time": "Wed Feb 12 11:10:58 2020 +0100" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sun Feb 23 12:48:55 2020 +0800" }, "message": "csky: Replace \u003clinux/clk-provider.h\u003e by \u003clinux/of_clk.h\u003e\n\nThe C-Sky platform code is not a clock provider, and just needs to call\nof_clk_init().\n\nHence it can include \u003clinux/of_clk.h\u003e instead of \u003clinux/clk-provider.h\u003e.\n\nSigned-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "dca132a60f226f4cbaa98807518a5ca6cff112ce", "tree": "d6e196846eee6d731a787e4570c503ef46d1869a", "parents": [ "f3cc24942e955604232e3b4b0dac6a54cbc7c679", "51dede9c05df2b78acd6dcf6a17d21f0877d2d7b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 18:02:10 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 18:02:10 2020 -0800" }, "message": "Merge tag \u0027ras-urgent-2020-02-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull RAS fixes from Thomas Gleixner:\n \"Two fixes for the AMD MCE driver:\n\n - Populate the per CPU MCA bank descriptor pointer only after it has\n been completely set up to prevent a use-after-free in case that one\n of the subsequent initialization step fails\n\n - Implement a proper release function for the sysfs entries of MCA\n threshold controls instead of freeing the memory right in the CPU\n teardown code, which leads to another use-after-free when the\n associated sysfs file is opened and accessed\"\n\n* tag \u0027ras-urgent-2020-02-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/mce/amd: Fix kobject lifetime\n x86/mce/amd: Publish the bank pointer only after setup has succeeded\n" }, { "commit": "f3cc24942e955604232e3b4b0dac6a54cbc7c679", "tree": "7419b6244e8278f687b4156a05a7e6855d3bcfce", "parents": [ "fca1037864a9fb75c9c48cb65649b24de8101ec4", "2546287c5fb363a0165933ae2181c92f03e701d0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 17:25:46 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 17:25:46 2020 -0800" }, "message": "Merge tag \u0027irq-urgent-2020-02-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq fixes from Thomas Gleixner:\n \"Two fixes for the irq core code which are follow ups to the recent MSI\n fixes:\n\n - The WARN_ON which was put into the MSI setaffinity callback for\n paranoia reasons actually triggered via a callchain which escaped\n when all the possible ways to reach that code were analyzed.\n\n The proc/irq/$N/*affinity interfaces have a quirk which came in\n when ALPHA moved to the generic interface: In case that the written\n affinity mask does not contain any online CPU it calls into ALPHAs\n magic auto affinity setting code.\n\n A few years later this mechanism was also made available to x86 for\n no good reasons and in a way which circumvents all sanity checks\n for interrupts which cannot have their affinity set from process\n context on X86 due to the way the X86 interrupt delivery works.\n\n It would be possible to make this work properly, but there is no\n point in doing so. If the interrupt is not yet started then the\n affinity setting has no effect and if it is started already then it\n is already assigned to an online CPU so there is no point to\n randomly move it to some other CPU. Just return EINVAL as the code\n has done before that change forever.\n\n - The new MSI quirk bit in the irq domain flags turned out to be\n already occupied, which escaped the author and the reviewers\n because the already in use bits were 0,6,2,3,4,5 listed in that\n order.\n\n That bit 6 was simply overlooked because the ordering was straight\n forward linear otherwise. So the new bit ended up being a\n duplicate.\n\n Fix it up by switching the oddball 6 to the obvious 1\"\n\n* tag \u0027irq-urgent-2020-02-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n genirq/irqdomain: Make sure all irq domain flags are distinct\n genirq/proc: Reject invalid affinity masks (again)\n" }, { "commit": "fca1037864a9fb75c9c48cb65649b24de8101ec4", "tree": "1f7b74fa822afc3998c34a88c11d4526b9b09885", "parents": [ "0a115e5f23b948be369faf14d3bccab283830f56", "21b5ee59ef18e27d85810584caf1f7ddc705ea83" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 17:08:16 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 17:08:16 2020 -0800" }, "message": "Merge tag \u0027x86-urgent-2020-02-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Thomas Gleixner:\n \"Two fixes for x86:\n\n - Remove the __force_oder definiton from the kaslr boot code as it is\n already defined in the page table code which makes GCC 10 builds\n fail because it changed the default to -fno-common.\n\n - Address the AMD erratum 1054 concerning the IRPERF capability and\n enable the Instructions Retired fixed counter on machines which are\n not affected by the erratum\"\n\n* tag \u0027x86-urgent-2020-02-22\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/cpu/amd: Enable the fixed Instructions Retired counter IRPERF\n x86/boot/compressed: Don\u0027t declare __force_order in kaslr_64.c\n" }, { "commit": "0a115e5f23b948be369faf14d3bccab283830f56", "tree": "f3fda89eedf9d18bd2406b87be07b7468460e740", "parents": [ "b88025ea47ec8aea47f0c283d182ab26bae2970d", "4c5fd3b791a06021084b42d5610400f846d206b5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:38:20 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:38:20 2020 -0800" }, "message": "Merge tag \u0027zonefs-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs\n\nPull zonefs fix from Damien Le Moal:\n \"A single patch fixing typos in the documentation file\"\n\n* tag \u0027zonefs-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs:\n zonefs: fix documentation typos etc.\n" }, { "commit": "b88025ea47ec8aea47f0c283d182ab26bae2970d", "tree": "653b575f66d4fe2bfb64c7f6890c783e19aeae47", "parents": [ "f6c69b7f51456a914da11c4ad1b14eba933d36aa", "c7849be9cc2dd2754c48ddbaca27c2de6d80a95d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:12:55 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:12:55 2020 -0800" }, "message": "Merge tag \u0027io_uring-5.6-2020-02-22\u0027 of git://git.kernel.dk/linux-block\n\nPull io_uring fixes from Jens Axboe:\n \"Here\u0027s a small collection of fixes that were queued up:\n\n - Remove unnecessary NULL check (Dan)\n\n - Missing io_req_cancelled() call in fallocate (Pavel)\n\n - Put the cleanup check for aux data in the right spot (Pavel)\n\n - Two fixes for SQPOLL (Stefano, Xiaoguang)\"\n\n* tag \u0027io_uring-5.6-2020-02-22\u0027 of git://git.kernel.dk/linux-block:\n io_uring: fix __io_iopoll_check deadlock in io_sq_thread\n io_uring: prevent sq_thread from spinning when it should stop\n io_uring: fix use-after-free by io_cleanup_req()\n io_uring: remove unnecessary NULL checks\n io_uring: add missing io_req_cancelled()\n" }, { "commit": "f6c69b7f51456a914da11c4ad1b14eba933d36aa", "tree": "9b87e8ac2fce1a84ee7640eda4513be80de97e1a", "parents": [ "b98b809c0a1311239605e8d2ce1583191d3efd5b", "ae7bbc091351a4c6ebddfbe36eb5eb7a149cd7a2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:09:06 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:09:06 2020 -0800" }, "message": "Merge tag \u0027block-5.6-2020-02-22\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n \"Just a set of NVMe fixes via Keith\"\n\n* tag \u0027block-5.6-2020-02-22\u0027 of git://git.kernel.dk/linux-block:\n nvme-multipath: Fix memory leak with ana_log_buf\n nvme: Fix uninitialized-variable warning\n nvme-pci: Use single IRQ vector for old Apple models\n nvme/pci: Add sleep quirk for Samsung and Toshiba drives\n" }, { "commit": "b98b809c0a1311239605e8d2ce1583191d3efd5b", "tree": "583fd17a9c458cb52922ed089667150d08d7386e", "parents": [ "5b442b1a4e9a19def99309b9831e276c21352ac1", "807b9515b7d044cf77df31f1af9d842a76ecd5cb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:00:52 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 11:00:52 2020 -0800" }, "message": "Merge tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi\n\nPull SCSI fixes from James Bottomley:\n \"Four non-core fixes.\n\n Two are reverts of target fixes which turned out to have unwanted side\n effects, one is a revert of an RDMA fix with the same problem and the\n final one fixes an incorrect warning about memory allocation failures\n in megaraid_sas (the driver actually reduces the allocation size until\n it succeeds)\"\n\nSigned-off-by: James E.J. Bottomley \u003cjejb@linux.ibm.com\u003e\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: Revert \"target: iscsi: Wait for all commands to finish before freeing a session\"\n scsi: Revert \"RDMA/isert: Fix a recently introduced regression related to logout\"\n scsi: megaraid_sas: silence a warning\n scsi: Revert \"target/core: Inline transport_lun_remove_cmd()\"\n" }, { "commit": "5b442b1a4e9a19def99309b9831e276c21352ac1", "tree": "b79879d9c4dc23837c44d506b8df9aeb28d79e36", "parents": [ "fea630215a9e767fd3917b2cb09ec3ded58f88a2", "e61d2392253220477813b43412090dccdcac601f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 10:52:54 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 10:52:54 2020 -0800" }, "message": "Merge tag \u0027hwmon-for-v5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging\n\nPull hwmon fixes from Guenter Roeck:\n\n - Fix crash in w83627ehf driver seen with W83627DHG-P\n\n - Fix lockdep splat in acpi_power_meter driver\n\n - Fix xdpe12284 documentation Sphinx warnings\n\n* tag \u0027hwmon-for-v5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:\n hwmon: (w83627ehf) Fix crash seen with W83627DHG-P\n hwmon: (acpi_power_meter) Fix lockdep splat\n Documentation/hwmon: fix xdpe12284 Sphinx warnings\n" }, { "commit": "fea630215a9e767fd3917b2cb09ec3ded58f88a2", "tree": "64bf050aa225d192ea9fedbafd5e760e88e27513", "parents": [ "591dd4c10146e541b9f7a461cb6b638ab313718d", "854bdbae9058bcf09b0add70b6047bc9ca776de2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 10:49:59 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 10:49:59 2020 -0800" }, "message": "Merge tag \u0027devicetree-fixes-for-5.6-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux\n\nPull devicetree fixes deom Rob Herring:\n \"A handful of fixes in DT bindings for MDIO bus, Allwinner CSI, OMAP\n HSMMC, and Tegra124 EMC\"\n\n* tag \u0027devicetree-fixes-for-5.6-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:\n dt-bindings: media: csi: Fix clocks description\n dt-bindings: media: csi: Add interconnects properties\n dt-bindings: net: mdio: remove compatible string from example\n dt-bindings: memory-controller: Update example for Tegra124 EMC\n dt-bindings: mmc: omap-hsmmc: Fix SDIO interrupt\n" }, { "commit": "591dd4c10146e541b9f7a461cb6b638ab313718d", "tree": "5f9d8ff37008fe47bb369985ff6418f3fc0428e8", "parents": [ "54dedb5b571d2fb0d65c3957ecfa9b32ce28d7f0", "2db01da8d25f0420c411e788a9e1ba39269ae37b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 10:43:41 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 22 10:43:41 2020 -0800" }, "message": "Merge tag \u0027s390-5.6-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 fixes from Vasily Gorbik:\n\n - Remove ieee_emulation_warnings sysctl which is a dead code.\n\n - Avoid triggering rebuild of the kernel during make install.\n\n - Enable protected virtualization guest support in default configs.\n\n - Fix cio_ignore seq_file .next function to increase position index.\n And use kobj_to_dev instead of container_of in cio code.\n\n - Fix storage block address lists to contain absolute addresses in qdio\n code.\n\n - Few clang warnings and spelling fixes.\n\n* tag \u0027s390-5.6-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n s390/qdio: fill SBALEs with absolute addresses\n s390/qdio: fill SL with absolute addresses\n s390: remove obsolete ieee_emulation_warnings\n s390: make \u0027install\u0027 not depend on vmlinux\n s390/kaslr: Fix casts in get_random\n s390/mm: Explicitly compare PAGE_DEFAULT_KEY against zero in storage_key_init_range\n s390/pkey/zcrypt: spelling s/crytp/crypt/\n s390/cio: use kobj_to_dev() API\n s390/defconfig: enable CONFIG_PROTECTED_VIRTUALIZATION_GUEST\n s390/cio: cio_ignore_proc_seq_next should increase position index\n" }, { "commit": "c7849be9cc2dd2754c48ddbaca27c2de6d80a95d", "tree": "f097b2326d01be9834bcad66d20b1818464db2f1", "parents": [ "7143b5ac5750f404ff3a594b34fdf3fc2f99f828" ], "author": { "name": "Xiaoguang Wang", "email": "xiaoguang.wang@linux.alibaba.com", "time": "Sat Feb 22 14:46:05 2020 +0800" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Sat Feb 22 07:45:03 2020 -0700" }, "message": "io_uring: fix __io_iopoll_check deadlock in io_sq_thread\n\nSince commit a3a0e43fd770 (\"io_uring: don\u0027t enter poll loop if we have\nCQEs pending\"), if we already events pending, we won\u0027t enter poll loop.\nIn case SETUP_IOPOLL and SETUP_SQPOLL are both enabled, if app has\nbeen terminated and don\u0027t reap pending events which are already in cq\nring, and there are some reqs in poll_list, io_sq_thread will enter\n__io_iopoll_check(), and find pending events, then return, this loop\nwill never have a chance to exit.\n\nI have seen this issue in fio stress tests, to fix this issue, let\nio_sq_thread call io_iopoll_getevents() with argument \u0027min\u0027 being zero,\nand remove __io_iopoll_check().\n\nFixes: a3a0e43fd770 (\"io_uring: don\u0027t enter poll loop if we have CQEs pending\")\nSigned-off-by: Xiaoguang Wang \u003cxiaoguang.wang@linux.alibaba.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "9db176bceb5c5df4990486709da386edadc6bd1d", "tree": "de32c5c0baeff9ae16aab39018cf7e9ce97dfbba", "parents": [ "8eedabfd66b68a4623beec0789eac54b8c9d0fb6" ], "author": { "name": "Jan Kara", "email": "jack@suse.cz", "time": "Fri Feb 21 11:08:35 2020 +0100" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 19:32:07 2020 -0500" }, "message": "ext4: fix mount failure with quota configured as module\n\nWhen CONFIG_QFMT_V2 is configured as a module, the test in\next4_feature_set_ok() fails and so mount of filesystems with quota or\nproject features fails. Fix the test to use IS_ENABLED macro which\nworks properly even for modules.\n\nLink: https://lore.kernel.org/r/20200221100835.9332-1-jack@suse.cz\nFixes: d65d87a07476 (\"ext4: improve explanation of a mount failure caused by a misconfigured kernel\")\nSigned-off-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nCc: stable@kernel.org\n" }, { "commit": "8eedabfd66b68a4623beec0789eac54b8c9d0fb6", "tree": "a530536f93841d65956f65ed1a514179cef70347", "parents": [ "cb85f4d23f794e24127f3e562cb3b54b0803f456" ], "author": { "name": "wangyan", "email": "wangyan122@huawei.com", "time": "Thu Feb 20 21:46:14 2020 +0800" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 19:32:07 2020 -0500" }, "message": "jbd2: fix ocfs2 corrupt when clearing block group bits\n\nI found a NULL pointer dereference in ocfs2_block_group_clear_bits().\nThe running environment:\n\tkernel version: 4.19\n\tA cluster with two nodes, 5 luns mounted on two nodes, and do some\n\tfile operations like dd/fallocate/truncate/rm on every lun with storage\n\tnetwork disconnection.\n\nThe fallocate operation on dm-23-45 caused an null pointer dereference.\n\nThe information of NULL pointer dereference as follows:\n\t[577992.878282] JBD2: Error -5 detected when updating journal superblock for dm-23-45.\n\t[577992.878290] Aborting journal on device dm-23-45.\n\t...\n\t[577992.890778] JBD2: Error -5 detected when updating journal superblock for dm-24-46.\n\t[577992.890908] __journal_remove_journal_head: freeing b_committed_data\n\t[577992.890916] (fallocate,88392,52):ocfs2_extend_trans:474 ERROR: status \u003d -30\n\t[577992.890918] __journal_remove_journal_head: freeing b_committed_data\n\t[577992.890920] (fallocate,88392,52):ocfs2_rotate_tree_right:2500 ERROR: status \u003d -30\n\t[577992.890922] __journal_remove_journal_head: freeing b_committed_data\n\t[577992.890924] (fallocate,88392,52):ocfs2_do_insert_extent:4382 ERROR: status \u003d -30\n\t[577992.890928] (fallocate,88392,52):ocfs2_insert_extent:4842 ERROR: status \u003d -30\n\t[577992.890928] __journal_remove_journal_head: freeing b_committed_data\n\t[577992.890930] (fallocate,88392,52):ocfs2_add_clusters_in_btree:4947 ERROR: status \u003d -30\n\t[577992.890933] __journal_remove_journal_head: freeing b_committed_data\n\t[577992.890939] __journal_remove_journal_head: freeing b_committed_data\n\t[577992.890949] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020\n\t[577992.890950] Mem abort info:\n\t[577992.890951] ESR \u003d 0x96000004\n\t[577992.890952] Exception class \u003d DABT (current EL), IL \u003d 32 bits\n\t[577992.890952] SET \u003d 0, FnV \u003d 0\n\t[577992.890953] EA \u003d 0, S1PTW \u003d 0\n\t[577992.890954] Data abort info:\n\t[577992.890955] ISV \u003d 0, ISS \u003d 0x00000004\n\t[577992.890956] CM \u003d 0, WnR \u003d 0\n\t[577992.890958] user pgtable: 4k pages, 48-bit VAs, pgdp \u003d 00000000f8da07a9\n\t[577992.890960] [0000000000000020] pgd\u003d0000000000000000\n\t[577992.890964] Internal error: Oops: 96000004 [#1] SMP\n\t[577992.890965] Process fallocate (pid: 88392, stack limit \u003d 0x00000000013db2fd)\n\t[577992.890968] CPU: 52 PID: 88392 Comm: fallocate Kdump: loaded Tainted: G W OE 4.19.36 #1\n\t[577992.890969] Hardware name: Huawei TaiShan 2280 V2/BC82AMDD, BIOS 0.98 08/25/2019\n\t[577992.890971] pstate: 60400009 (nZCv daif +PAN -UAO)\n\t[577992.891054] pc : _ocfs2_free_suballoc_bits+0x63c/0x968 [ocfs2]\n\t[577992.891082] lr : _ocfs2_free_suballoc_bits+0x618/0x968 [ocfs2]\n\t[577992.891084] sp : ffff0000c8e2b810\n\t[577992.891085] x29: ffff0000c8e2b820 x28: 0000000000000000\n\t[577992.891087] x27: 00000000000006f3 x26: ffffa07957b02e70\n\t[577992.891089] x25: ffff807c59d50000 x24: 00000000000006f2\n\t[577992.891091] x23: 0000000000000001 x22: ffff807bd39abc30\n\t[577992.891093] x21: ffff0000811d9000 x20: ffffa07535d6a000\n\t[577992.891097] x19: ffff000001681638 x18: ffffffffffffffff\n\t[577992.891098] x17: 0000000000000000 x16: ffff000080a03df0\n\t[577992.891100] x15: ffff0000811d9708 x14: 203d207375746174\n\t[577992.891101] x13: 73203a524f525245 x12: 20373439343a6565\n\t[577992.891103] x11: 0000000000000038 x10: 0101010101010101\n\t[577992.891106] x9 : ffffa07c68a85d70 x8 : 7f7f7f7f7f7f7f7f\n\t[577992.891109] x7 : 0000000000000000 x6 : 0000000000000080\n\t[577992.891110] x5 : 0000000000000000 x4 : 0000000000000002\n\t[577992.891112] x3 : ffff000001713390 x2 : 2ff90f88b1c22f00\n\t[577992.891114] x1 : ffff807bd39abc30 x0 : 0000000000000000\n\t[577992.891116] Call trace:\n\t[577992.891139] _ocfs2_free_suballoc_bits+0x63c/0x968 [ocfs2]\n\t[577992.891162] _ocfs2_free_clusters+0x100/0x290 [ocfs2]\n\t[577992.891185] ocfs2_free_clusters+0x50/0x68 [ocfs2]\n\t[577992.891206] ocfs2_add_clusters_in_btree+0x198/0x5e0 [ocfs2]\n\t[577992.891227] ocfs2_add_inode_data+0x94/0xc8 [ocfs2]\n\t[577992.891248] ocfs2_extend_allocation+0x1bc/0x7a8 [ocfs2]\n\t[577992.891269] ocfs2_allocate_extents+0x14c/0x338 [ocfs2]\n\t[577992.891290] __ocfs2_change_file_space+0x3f8/0x610 [ocfs2]\n\t[577992.891309] ocfs2_fallocate+0xe4/0x128 [ocfs2]\n\t[577992.891316] vfs_fallocate+0x11c/0x250\n\t[577992.891317] ksys_fallocate+0x54/0x88\n\t[577992.891319] __arm64_sys_fallocate+0x28/0x38\n\t[577992.891323] el0_svc_common+0x78/0x130\n\t[577992.891325] el0_svc_handler+0x38/0x78\n\t[577992.891327] el0_svc+0x8/0xc\n\nMy analysis process as follows:\nocfs2_fallocate\n __ocfs2_change_file_space\n ocfs2_allocate_extents\n ocfs2_extend_allocation\n ocfs2_add_inode_data\n ocfs2_add_clusters_in_btree\n ocfs2_insert_extent\n ocfs2_do_insert_extent\n ocfs2_rotate_tree_right\n ocfs2_extend_rotate_transaction\n ocfs2_extend_trans\n jbd2_journal_restart\n jbd2__journal_restart\n /* handle-\u003eh_transaction is NULL,\n * is_handle_aborted(handle) is true\n */\n handle-\u003eh_transaction \u003d NULL;\n start_this_handle\n return -EROFS;\n ocfs2_free_clusters\n _ocfs2_free_clusters\n _ocfs2_free_suballoc_bits\n ocfs2_block_group_clear_bits\n ocfs2_journal_access_gd\n __ocfs2_journal_access\n jbd2_journal_get_undo_access\n /* I think jbd2_write_access_granted() will\n * return true, because do_get_write_access()\n * will return -EROFS.\n */\n if (jbd2_write_access_granted(...)) return 0;\n do_get_write_access\n /* handle-\u003eh_transaction is NULL, it will\n * return -EROFS here, so do_get_write_access()\n * was not called.\n */\n if (is_handle_aborted(handle)) return -EROFS;\n /* bh2jh(group_bh) is NULL, caused NULL\n pointer dereference */\n undo_bg \u003d (struct ocfs2_group_desc *)\n bh2jh(group_bh)-\u003eb_committed_data;\n\nIf handle-\u003eh_transaction \u003d\u003d NULL, then jbd2_write_access_granted()\ndoes not really guarantee that journal_head will stay around,\nnot even speaking of its b_committed_data. The bh2jh(group_bh)\ncan be removed after ocfs2_journal_access_gd() and before call\n\"bh2jh(group_bh)-\u003eb_committed_data\". So, we should move\nis_handle_aborted() check from do_get_write_access() into\njbd2_journal_get_undo_access() and jbd2_journal_get_write_access()\nbefore the call to jbd2_write_access_granted().\n\nLink: https://lore.kernel.org/r/f72a623f-b3f1-381a-d91d-d22a1c83a336@huawei.com\nSigned-off-by: Yan Wang \u003cwangyan122@huawei.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nReviewed-by: Jun Piao \u003cpiaojun@huawei.com\u003e\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: stable@kernel.org\n" }, { "commit": "cb85f4d23f794e24127f3e562cb3b54b0803f456", "tree": "8a8bf16b434a1f9e78b91220ef8eca39cc31a47f", "parents": [ "bbd55937de8f2754adc5792b0f8e5ff7d9c0420e" ], "author": { "name": "Eric Biggers", "email": "ebiggers@google.com", "time": "Wed Feb 19 10:30:47 2020 -0800" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 19:32:07 2020 -0500" }, "message": "ext4: fix race between writepages and enabling EXT4_EXTENTS_FL\n\nIf EXT4_EXTENTS_FL is set on an inode while ext4_writepages() is running\non it, the following warning in ext4_add_complete_io() can be hit:\n\nWARNING: CPU: 1 PID: 0 at fs/ext4/page-io.c:234 ext4_put_io_end_defer+0xf0/0x120\n\nHere\u0027s a minimal reproducer (not 100% reliable) (root isn\u0027t required):\n\n while true; do\n sync\n done \u0026\n while true; do\n rm -f file\n touch file\n chattr -e file\n echo X \u003e\u003e file\n chattr +e file\n done\n\nThe problem is that in ext4_writepages(), ext4_should_dioread_nolock()\n(which only returns true on extent-based files) is checked once to set\nthe number of reserved journal credits, and also again later to select\nthe flags for ext4_map_blocks() and copy the reserved journal handle to\next4_io_end::handle. But if EXT4_EXTENTS_FL is being concurrently set,\nthe first check can see dioread_nolock disabled while the later one can\nsee it enabled, causing the reserved handle to unexpectedly be NULL.\n\nSince changing EXT4_EXTENTS_FL is uncommon, and there may be other races\nrelated to doing so as well, fix this by synchronizing changing\nEXT4_EXTENTS_FL with ext4_writepages() via the existing\ns_writepages_rwsem (previously called s_journal_flag_rwsem).\n\nThis was originally reported by syzbot without a reproducer at\nhttps://syzkaller.appspot.com/bug?extid\u003d2202a584a00fffd19fbf,\nbut now that dioread_nolock is the default I also started seeing this\nwhen running syzkaller locally.\n\nLink: https://lore.kernel.org/r/20200219183047.47417-3-ebiggers@kernel.org\nReported-by: syzbot+2202a584a00fffd19fbf@syzkaller.appspotmail.com\nFixes: 6b523df4fb5a (\"ext4: use transaction reservation for extent conversion in ext4_end_io\")\nSigned-off-by: Eric Biggers \u003cebiggers@google.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: stable@kernel.org\n" }, { "commit": "bbd55937de8f2754adc5792b0f8e5ff7d9c0420e", "tree": "a648c05a34cb6f5c4b5333f3fff5bbf7d5498119", "parents": [ "7c990728b99ed6fbe9c75fc202fce1172d9916da" ], "author": { "name": "Eric Biggers", "email": "ebiggers@google.com", "time": "Wed Feb 19 10:30:46 2020 -0800" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 19:32:07 2020 -0500" }, "message": "ext4: rename s_journal_flag_rwsem to s_writepages_rwsem\n\nIn preparation for making s_journal_flag_rwsem synchronize\next4_writepages() with changes to both the EXTENTS and JOURNAL_DATA\nflags (rather than just JOURNAL_DATA as it does currently), rename it to\ns_writepages_rwsem.\n\nLink: https://lore.kernel.org/r/20200219183047.47417-2-ebiggers@kernel.org\nSigned-off-by: Eric Biggers \u003cebiggers@google.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: stable@kernel.org\n" }, { "commit": "7c990728b99ed6fbe9c75fc202fce1172d9916da", "tree": "73fcf3b46fb35781ab76a4c8fb84be7f70f87c96", "parents": [ "df3da4ea5a0fc5d115c90d5aa6caa4dd433750a7" ], "author": { "name": "Suraj Jitindar Singh", "email": "surajjs@amazon.com", "time": "Tue Feb 18 19:08:51 2020 -0800" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 19:31:46 2020 -0500" }, "message": "ext4: fix potential race between s_flex_groups online resizing and access\n\nDuring an online resize an array of s_flex_groups structures gets replaced\nso it can get enlarged. If there is a concurrent access to the array and\nthis memory has been reused then this can lead to an invalid memory access.\n\nThe s_flex_group array has been converted into an array of pointers rather\nthan an array of structures. This is to ensure that the information\ncontained in the structures cannot get out of sync during a resize due to\nan accessor updating the value in the old structure after it has been\ncopied but before the array pointer is updated. Since the structures them-\nselves are no longer copied but only the pointers to them this case is\nmitigated.\n\nLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d206443\nLink: https://lore.kernel.org/r/20200221053458.730016-4-tytso@mit.edu\nSigned-off-by: Suraj Jitindar Singh \u003csurajjs@amazon.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nCc: stable@kernel.org\n" }, { "commit": "54dedb5b571d2fb0d65c3957ecfa9b32ce28d7f0", "tree": "b064258f87c6582a7301d99426bdb3ef56397984", "parents": [ "63f01d852c75366fb4d15ce217d12c48b69a4bcc", "8645e56a4ad6dcbf504872db7f14a2f67db88ef2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 16:10:10 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 16:10:10 2020 -0800" }, "message": "Merge tag \u0027for-linus-5.6-rc3-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip\n\nPull xen fixes from Juergen Gross:\n \"Two small fixes for Xen:\n\n - a fix to avoid warnings with new gcc\n\n - a fix for incorrectly disabled interrupts when calling\n _cond_resched()\"\n\n* tag \u0027for-linus-5.6-rc3-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:\n xen: Enable interrupts when calling _cond_resched()\n x86/xen: Distribute switch variables for initialization\n" }, { "commit": "63f01d852c75366fb4d15ce217d12c48b69a4bcc", "tree": "8d47f51afb6763b923fc3082f6a259456a2c77f7", "parents": [ "2865936259e27629fac422bc80c9b55ca1f108a5", "dcde237319e626d1ec3c9d8b7613032f0fd4663a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 16:03:36 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 16:03:36 2020 -0800" }, "message": "Merge tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 fixes from Will Deacon:\n \"It\u0027s all straightforward apart from the changes to mmap()/mremap() in\n relation to their handling of address arguments from userspace with\n non-zero tag bits in the upper byte.\n\n The change to brk() is necessary to fix a nasty user-visible\n regression in malloc(), but we tightened up mmap() and mremap() at the\n same time because they also allow the user to create virtual aliases\n by accident. It\u0027s much less likely than brk() to matter in practice,\n but enforcing the principle of \"don\u0027t permit the creation of mappings\n using tagged addresses\" leads to a straightforward ABI without having\n to worry about the \"but what if a crazy program did foo?\" aspect of\n things.\n\n Summary:\n\n - Fix regression in malloc() caused by ignored address tags in brk()\n\n - Add missing brackets around argument to untagged_addr() macro\n\n - Fix clang build when using binutils assembler\n\n - Fix silly typo in virtual memory map documentation\"\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n mm: Avoid creating virtual address aliases in brk()/mmap()/mremap()\n docs: arm64: fix trivial spelling enought to enough in memory.rst\n arm64: memory: Add missing brackets to untagged_addr() macro\n arm64: lse: Fix LSE atomics with LLVM\n" }, { "commit": "2865936259e27629fac422bc80c9b55ca1f108a5", "tree": "fc246a8d4008c381303b1775872afd3b8008ccd2", "parents": [ "0c0ddd6ae47c9238c18f475bcca675ca74c9dc31", "9eb425b2e04e0e3006adffea5bf5f227a896f128" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 15:57:56 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 15:57:56 2020 -0800" }, "message": "Merge tag \u0027powerpc-5.6-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull powerpc fixes from Michael Ellerman:\n \"Some more powerpc fixes for 5.6. This is two weeks worth as I was out\n sick last week:\n\n - Three fixes for the recently added VMAP_STACK on 32-bit.\n\n - Three fixes related to hugepages on 8xx (32-bit).\n\n - A fix for a bug in our transactional memory handling that could\n lead to a kernel crash if we saw a page fault during signal\n delivery.\n\n - A fix for a deadlock in our PCI EEH (Enhanced Error Handling) code.\n\n - A couple of other minor fixes.\n\n Thanks to: Christophe Leroy, Erhard F, Frederic Barrat, Gustavo Luiz\n Duarte, Larry Finger, Leonardo Bras, Oliver O\u0027Halloran, Sam Bobroff\"\n\n* tag \u0027powerpc-5.6-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n powerpc/entry: Fix an #if which should be an #ifdef in entry_32.S\n powerpc/xmon: Fix whitespace handling in getstring()\n powerpc/6xx: Fix power_save_ppc32_restore() with CONFIG_VMAP_STACK\n powerpc/chrp: Fix enter_rtas() with CONFIG_VMAP_STACK\n powerpc/32s: Fix DSI and ISI exceptions for CONFIG_VMAP_STACK\n powerpc/tm: Fix clearing MSR[TS] in current when reclaiming on signal delivery\n powerpc/8xx: Fix clearing of bits 20-23 in ITLB miss\n powerpc/hugetlb: Fix 8M hugepages on 8xx\n powerpc/hugetlb: Fix 512k hugepages on 8xx with 16k page size\n powerpc/eeh: Fix deadlock handling dead PHB\n" }, { "commit": "0c0ddd6ae47c9238c18f475bcca675ca74c9dc31", "tree": "cfca0dc266a6e4b8eaf6da5a54d2618cf88d010a", "parents": [ "bb65619e97323b8d62dd4d7549b9b7fe8a774706", "44144c809e39d64ff9931c7e8956c42b2baa89e6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 13:02:49 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 13:02:49 2020 -0800" }, "message": "Merge tag \u0027linux-watchdog-5.6-rc3\u0027 of git://www.linux-watchdog.org/linux-watchdog\n\nPull watchdog fixes from Wim Van Sebroeck:\n\n - mtk_wdt needs RESET_CONTROLLER to build\n\n - da9062 driver fixes:\n - fix power management ops\n - do not ping the hw during stop()\n - add dependency on I2C\n\n* tag \u0027linux-watchdog-5.6-rc3\u0027 of git://www.linux-watchdog.org/linux-watchdog:\n watchdog: da9062: Add dependency on I2C\n watchdog: da9062: fix power management ops\n watchdog: da9062: do not ping the hw during stop()\n watchdog: fix mtk_wdt.c RESET_CONTROLLER build error\n" }, { "commit": "bb65619e97323b8d62dd4d7549b9b7fe8a774706", "tree": "52704ee2ed5fb02f024026b2c28ab3783b21c65f", "parents": [ "e5553ac71e584c3aa443e211ca2afded6071b5b6", "74ba569a15a08b988bc059ad515980f51e85be79" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:57:05 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:57:05 2020 -0800" }, "message": "Merge tag \u0027char-misc-5.6-rc3\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 some small char/misc driver fixes for 5.6-rc3.\n\n Also included in here are some updates for some documentation files\n that I seem to be maintaining these days.\n\n The driver fixes are:\n - small fixes for the habanalabs driver\n - fsi driver bugfix\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027char-misc-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n Documentation/process: Swap out the ambassador for Canonical\n habanalabs: patched cb equals user cb in device memset\n habanalabs: do not halt CoreSight during hard reset\n habanalabs: halt the engines before hard-reset\n MAINTAINERS: remove unnecessary \u0027:\u0027 characters\n fsi: aspeed: add unspecified HAS_IOMEM dependency\n COPYING: state that all contributions really are covered by this file\n Documentation/process: Change Microsoft contact for embargoed hardware issues\n embargoed-hardware-issues: drop Amazon contact as the email address now bounces\n Documentation/process: Add Arm contact for embargoed HW issues\n" }, { "commit": "e5553ac71e584c3aa443e211ca2afded6071b5b6", "tree": "d3c505338119ba459a63e1891ed40483eeb60827", "parents": [ "ef11f1b76a59833cfadbe88a9513ca01b14a2b88", "9a4556bd8f23209c29f152e6a930b6a893b0fc81" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:53:53 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:53:53 2020 -0800" }, "message": "Merge tag \u0027staging-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging driver fixes from Greg KH:\n \"Here are some small staging driver fixes for 5.6-rc3, along with the\n removal of an unused/unneeded driver as well.\n\n The android vsoc driver is not needed anymore by anyone, so it was\n removed.\n\n The other driver fixes are:\n - ashmem bugfixes\n - greybus audio driver bugfix\n - wireless driver bugfixes and tiny cleanups to error paths\n\n All of these have been in linux-next for a while now with no reported\n issues\"\n\n* tag \u0027staging-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:\n staging: rtl8723bs: Remove unneeded goto statements\n staging: rtl8188eu: Remove some unneeded goto statements\n staging: rtl8723bs: Fix potential overuse of kernel memory\n staging: rtl8188eu: Fix potential overuse of kernel memory\n staging: rtl8723bs: Fix potential security hole\n staging: rtl8188eu: Fix potential security hole\n staging: greybus: use after free in gb_audio_manager_remove_all()\n staging: android: Delete the \u0027vsoc\u0027 driver\n staging: rtl8723bs: fix copy of overlapping memory\n staging: android: ashmem: Disallow ashmem memory from being remapped\n staging: vt6656: fix sign of rx_dbm to bb_pre_ed_rssi.\n" }, { "commit": "ef11f1b76a59833cfadbe88a9513ca01b14a2b88", "tree": "77c99d0274bbc1dee76e29794aaad5b5704c6590", "parents": [ "cee853e825a9ab64b54bd53f991e98f832c47317", "07e6124a1a46b4b5a9b3cacc0c306b50da87abf5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:48:29 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:48:29 2020 -0800" }, "message": "Merge tag \u0027tty-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull tty/serial driver fixes from Greg KH:\n \"Here are a number of small tty and serial driver fixes for 5.6-rc3\n that resolve a bunch of reported issues.\n\n They are:\n - vt selection and ioctl fixes\n - serdev bugfix\n - atmel serial driver fixes\n - qcom serial driver fixes\n - other minor serial driver fixes\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027tty-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n vt: selection, close sel_buffer race\n vt: selection, handle pending signals in paste_selection\n serial: cpm_uart: call cpm_muram_init before registering console\n tty: serial: qcom_geni_serial: Fix RX cancel command failure\n serial: 8250: Check UPF_IRQ_SHARED in advance\n tty: serial: imx: setup the correct sg entry for tx dma\n vt: vt_ioctl: fix race in VT_RESIZEX\n vt: fix scrollback flushing on background consoles\n tty: serial: tegra: Handle RX transfer in PIO mode if DMA wasn\u0027t started\n tty/serial: atmel: manage shutdown in case of RS485 or ISO7816 mode\n serdev: ttyport: restore client ops on deregistration\n serial: ar933x_uart: set UART_CS_{RX,TX}_READY_ORIDE\n" }, { "commit": "cee853e825a9ab64b54bd53f991e98f832c47317", "tree": "32592a61b14c664eb361e5ab4e4f347b1ed6537f", "parents": [ "88f8bbfa94ce18eff7b322256ec4b5f885dea969", "8b978be8d067f58b29cbbf0579e88bb4a2d4350a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:44:53 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:44:53 2020 -0800" }, "message": "Merge tag \u0027usb-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB/Thunderbolt fixes from Greg KH:\n \"Here are a number of small USB driver fixes for 5.6-rc3.\n\n Included in here are:\n - MAINTAINER file updates\n - USB gadget driver fixes\n - usb core quirk additions and fixes for regressions\n - xhci driver fixes\n - usb serial driver id additions and fixes\n - thunderbolt bugfix\n\n Thunderbolt patches come in through here now that USB4 is really\n thunderbolt.\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027usb-5.6-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (34 commits)\n USB: misc: iowarrior: add support for the 100 device\n thunderbolt: Prevent crash if non-active NVMem file is read\n usb: gadget: udc-xilinx: Fix xudc_stop() kernel-doc format\n USB: misc: iowarrior: add support for the 28 and 28L devices\n USB: misc: iowarrior: add support for 2 OEMed devices\n USB: Fix novation SourceControl XL after suspend\n xhci: Fix memory leak when caching protocol extended capability PSI tables - take 2\n Revert \"xhci: Fix memory leak when caching protocol extended capability PSI tables\"\n MAINTAINERS: Sort entries in database for THUNDERBOLT\n usb: dwc3: debug: fix string position formatting mixup with ret and len\n usb: gadget: serial: fix Tx stall after buffer overflow\n usb: gadget: ffs: ffs_aio_cancel(): Save/restore IRQ flags\n usb: dwc2: Fix SET/CLEAR_FEATURE and GET_STATUS flows\n usb: dwc2: Fix in ISOC request length checking\n usb: gadget: composite: Support more than 500mA MaxPower\n usb: gadget: composite: Fix bMaxPower for SuperSpeedPlus\n usb: gadget: u_audio: Fix high-speed max packet size\n usb: dwc3: gadget: Check for IOC/LST bit in TRB-\u003ectrl fields\n USB: core: clean up endpoint-descriptor parsing\n USB: quirks: blacklist duplicate ep on Sound Devices USBPre2\n ...\n" }, { "commit": "88f8bbfa94ce18eff7b322256ec4b5f885dea969", "tree": "7839ca8098d0ae4d4f1fe40f7ffed3c8ff01eb03", "parents": [ "3dc55dba67231fc22352483f5ca737df96cdc1e6", "97d9a4e9619a822c5baf6a63e6f5b80fee4d4213" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:18:02 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 12:18:02 2020 -0800" }, "message": "Merge tag \u0027drm-fixes-2020-02-21\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Varied fixes for rc3.\n\n i915 is the largest, they are seeing some ACPI problems with their CI\n which hopefully get solved soon [1].\n\n msm has a bunch of fixes for new hw added in the merge, a bunch of\n amdgpu fixes, and nouveau adds support for some new firmwares for\n turing tu11x GPUs that were just released into linux-firmware by\n nvidia, they operate the same as the ones we already have for tu10x so\n should be fine to hook up.\n\n Otherwise it\u0027s just misc fixes for panfrost and sun4i.\n\n core:\n - Allow only one rotation argument, and allow zero rotation in video\n cmdline.\n\n i915:\n - Workaround missing Display Stream Compression (DSC) state readout\n by forcing modeset when its enabled at probe\n - Fix EHL port clock voltage level requirements\n - Fix queuing retire workers on the virtual engine\n - Fix use of partially initialized waiters\n - Stop using drm_pci_alloc/drm_pci/free\n - Fix rewind of RING_TAIL by forcing a context reload\n - Fix locking on resetting ring-\u003ehead\n - Propagate our bug filing URL change to stable kernels\n\n panfrost:\n - Small compiler warning fix for panfrost.\n - Fix when using performance counters in panfrost when using per fd\n address space.\n\n sun4xi:\n - Fix dt binding\n\n nouveau:\n - tu11x modesetting fix\n - ACR/GR firmware support for tu11x (fw is public now)\n\n msm:\n - fix UBWC on GPU and display side for sc7180\n - fix DSI suspend/resume issue encountered on sc7180\n - fix some breakage on so called \"linux-android\" devices\n (fallout from sc7180/a618 support, not seen earlier due to\n bootloader/firmware differences)\n - couple other misc fixes\n\n amdgpu:\n - HDCP fixes\n - xclk fix for raven\n - GFXOFF fixes\"\n\n[1] The Intel suspend testing should now be fixed by commit 63fb9623427f\n (\"ACPI: PM: s2idle: Check fixed wakeup events in acpi_s2idle_wake()\")\n\n* tag \u0027drm-fixes-2020-02-21\u0027 of git://anongit.freedesktop.org/drm/drm: (39 commits)\n drm/amdgpu/display: clean up hdcp workqueue handling\n drm/amdgpu: add is_raven_kicker judgement for raven1\n drm/i915/gt: Avoid resetting ring-\u003ehead outside of its timeline mutex\n drm/i915/execlists: Always force a context reload when rewinding RING_TAIL\n drm/i915: Wean off drm_pci_alloc/drm_pci_free\n drm/i915/gt: Protect defer_request() from new waiters\n drm/i915/gt: Prevent queuing retire workers on the virtual engine\n drm/i915/dsc: force full modeset whenever DSC is enabled at probe\n drm/i915/ehl: Update port clock voltage level requirements\n drm/i915: Update drm/i915 bug filing URL\n MAINTAINERS: Update drm/i915 bug filing URL\n drm/i915: Initialise basic fence before acquiring seqno\n drm/i915/gem: Require per-engine reset support for non-persistent contexts\n drm/nouveau/kms/gv100-: Re-set LUT after clearing for modesets\n drm/nouveau/gr/tu11x: initial support\n drm/nouveau/acr/tu11x: initial support\n drm/amdgpu/gfx10: disable gfxoff when reading rlc clock\n drm/amdgpu/gfx9: disable gfxoff when reading rlc clock\n drm/amdgpu/soc15: fix xclk for raven\n drm/amd/powerplay: always refetch the enabled features status on dpm enablement\n ...\n" }, { "commit": "3dc55dba67231fc22352483f5ca737df96cdc1e6", "tree": "5eec1f5ce42822cb744313ec394b761fab0f7ed6", "parents": [ "b0dd1eb220c06892e0a4098378c4296650f3f8db", "36a44bcdd8df092d76c11bc213e81c5817d4e302" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:59:51 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:59:51 2020 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n\n 1) Limit xt_hashlimit hash table size to avoid OOM or hung tasks, from\n Cong Wang.\n\n 2) Fix deadlock in xsk by publishing global consumer pointers when NAPI\n is finished, from Magnus Karlsson.\n\n 3) Set table field properly to RT_TABLE_COMPAT when necessary, from\n Jethro Beekman.\n\n 4) NLA_STRING attributes are not necessary NULL terminated, deal wiht\n that in IFLA_ALT_IFNAME. From Eric Dumazet.\n\n 5) Fix checksum handling in atlantic driver, from Dmitry Bezrukov.\n\n 6) Handle mtu\u003d\u003d0 devices properly in wireguard, from Jason A.\n Donenfeld.\n\n 7) Fix several lockdep warnings in bonding, from Taehee Yoo.\n\n 8) Fix cls_flower port blocking, from Jason Baron.\n\n 9) Sanitize internal map names in libbpf, from Toke Høiland-Jørgensen.\n\n10) Fix RDMA race in qede driver, from Michal Kalderon.\n\n11) Fix several false lockdep warnings by adding conditions to\n list_for_each_entry_rcu(), from Madhuparna Bhowmik.\n\n12) Fix sleep in atomic in mlx5 driver, from Huy Nguyen.\n\n13) Fix potential deadlock in bpf_map_do_batch(), from Yonghong Song.\n\n14) Hey, variables declared in switch statement before any case\n statements are not initialized. I learn something every day. Get\n rids of this stuff in several parts of the networking, from Kees\n Cook.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (99 commits)\n bnxt_en: Issue PCIe FLR in kdump kernel to cleanup pending DMAs.\n bnxt_en: Improve device shutdown method.\n net: netlink: cap max groups which will be considered in netlink_bind()\n net: thunderx: workaround BGX TX Underflow issue\n ionic: fix fw_status read\n net: disable BRIDGE_NETFILTER by default\n net: macb: Properly handle phylink on at91rm9200\n s390/qeth: fix off-by-one in RX copybreak check\n s390/qeth: don\u0027t warn for napi with 0 budget\n s390/qeth: vnicc Fix EOPNOTSUPP precedence\n openvswitch: Distribute switch variables for initialization\n net: ip6_gre: Distribute switch variables for initialization\n net: core: Distribute switch variables for initialization\n udp: rehash on disconnect\n net/tls: Fix to avoid gettig invalid tls record\n bpf: Fix a potential deadlock with bpf_map_do_batch\n bpf: Do not grab the bucket spinlock by default on htab batch ops\n ice: Wait for VF to be reset/ready before configuration\n ice: Don\u0027t tell the OS that link is going down\n ice: Don\u0027t reject odd values of usecs set by user\n ...\n" }, { "commit": "b0dd1eb220c06892e0a4098378c4296650f3f8db", "tree": "4a36d79ed2db95ed13b417963641f9fc0b1c31ff", "parents": [ "63fb9623427fbb44e3782233b6e4714057b76ff2", "bb8d00ff51a0c5e58cebd2e698a778f4e3d34d48" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:40:10 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:40:10 2020 -0800" }, "message": "Merge branch \u0027akpm\u0027 (patches from Andrew)\n\nMerge misc fixes from Andrew Morton:\n\n - A few y2038 fixes which missed the merge window while dependencies\n in NFS were being sorted out.\n\n - A bunch of fixes. Some minor, some not.\n\n* emailed patches from Andrew Morton \u003cakpm@linux-foundation.org\u003e:\n MAINTAINERS: use tabs for SAFESETID\n lib/stackdepot.c: fix global out-of-bounds in stack_slabs\n mm/sparsemem: pfn_to_page is not valid yet on SPARSEMEM\n mm/vmscan.c: don\u0027t round up scan size for online memory cgroup\n lib/string.c: update match_string() doc-strings with correct behavior\n mm/memcontrol.c: lost css_put in memcg_expand_shrinker_maps()\n mm/swapfile.c: fix a comment in sys_swapon()\n scripts/get_maintainer.pl: deprioritize old Fixes: addresses\n get_maintainer: remove uses of P: for maintainer name\n selftests/vm: add missed tests in run_vmtests\n include/uapi/linux/swab.h: fix userspace breakage, use __BITS_PER_LONG for swap\n Revert \"ipc,sem: remove uneeded sem_undo_list lock usage in exit_sem()\"\n y2038: hide timeval/timespec/itimerval/itimerspec types\n y2038: remove unused time32 interfaces\n y2038: remove ktime to/from timespec/timeval conversion\n" }, { "commit": "bb8d00ff51a0c5e58cebd2e698a778f4e3d34d48", "tree": "3598dfbb87bbd0d47baaa416a5d9d05dccea2ffc", "parents": [ "305e519ce48e935702c32241f07d393c3c8fed3e" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Thu Feb 20 20:04:33 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "MAINTAINERS: use tabs for SAFESETID\n\nUse tabs for indentation instead of spaces for SAFESETID. All (!) other\nentries in MAINTAINERS use tabs (according to my simple grepping).\n\nLink: http://lkml.kernel.org/r/2bb2e52a-2694-816d-57b4-6cabfadd6c1a@infradead.org\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nCc: Micah Morton \u003cmortonm@chromium.org\u003e\nCc: James Morris \u003cjmorris@namei.org\u003e\nCc: \"Serge E. Hallyn\" \u003cserge@hallyn.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "305e519ce48e935702c32241f07d393c3c8fed3e", "tree": "b4b47f134cac7739a57dd61debe61b61d1783a75", "parents": [ "18e19f195cd888f65643a77a0c6aee8f5be6439a" ], "author": { "name": "Alexander Potapenko", "email": "glider@google.com", "time": "Thu Feb 20 20:04:30 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "lib/stackdepot.c: fix global out-of-bounds in stack_slabs\n\nWalter Wu has reported a potential case in which init_stack_slab() is\ncalled after stack_slabs[STACK_ALLOC_MAX_SLABS - 1] has already been\ninitialized. In that case init_stack_slab() will overwrite\nstack_slabs[STACK_ALLOC_MAX_SLABS], which may result in a memory\ncorruption.\n\nLink: http://lkml.kernel.org/r/20200218102950.260263-1-glider@google.com\nFixes: cd11016e5f521 (\"mm, kasan: stackdepot implementation. Enable stackdepot for SLAB\")\nSigned-off-by: Alexander Potapenko \u003cglider@google.com\u003e\nReported-by: Walter Wu \u003cwalter-zh.wu@mediatek.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nCc: Kate Stewart \u003ckstewart@linuxfoundation.org\u003e\nCc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "18e19f195cd888f65643a77a0c6aee8f5be6439a", "tree": "8ec6c66cbb73b2712c93bc1c32487a6e2fe49fae", "parents": [ "76073c646f5f4999d763f471df9e38a5a912d70d" ], "author": { "name": "Wei Yang", "email": "richardw.yang@linux.intel.com", "time": "Thu Feb 20 20:04:27 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "mm/sparsemem: pfn_to_page is not valid yet on SPARSEMEM\n\nWhen we use SPARSEMEM instead of SPARSEMEM_VMEMMAP, pfn_to_page()\ndoesn\u0027t work before sparse_init_one_section() is called.\n\nThis leads to a crash when hotplug memory:\n\n BUG: unable to handle page fault for address: 0000000006400000\n #PF: supervisor write access in kernel mode\n #PF: error_code(0x0002) - not-present page\n PGD 0 P4D 0\n Oops: 0002 [#1] SMP PTI\n CPU: 3 PID: 221 Comm: kworker/u16:1 Tainted: G W 5.5.0-next-20200205+ #343\n Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.0.0 02/06/2015\n Workqueue: kacpi_hotplug acpi_hotplug_work_fn\n RIP: 0010:__memset+0x24/0x30\n Code: cc cc cc cc cc cc 0f 1f 44 00 00 49 89 f9 48 89 d1 83 e2 07 48 c1 e9 03 40 0f b6 f6 48 b8 01 01 01 01 01 01 01 01 48 0f af c6 \u003cf3\u003e 48 ab 89 d1 f3 aa 4c 89 c8 c3 90 49 89 f9 40 88 f0 48 89 d1 f3\n RSP: 0018:ffffb43ac0373c80 EFLAGS: 00010a87\n RAX: ffffffffffffffff RBX: ffff8a1518800000 RCX: 0000000000050000\n RDX: 0000000000000000 RSI: 00000000000000ff RDI: 0000000006400000\n RBP: 0000000000140000 R08: 0000000000100000 R09: 0000000006400000\n R10: 0000000000000000 R11: 0000000000000002 R12: 0000000000000000\n R13: 0000000000000028 R14: 0000000000000000 R15: ffff8a153ffd9280\n FS: 0000000000000000(0000) GS:ffff8a153ab00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 0000000006400000 CR3: 0000000136fca000 CR4: 00000000000006e0\n DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n Call Trace:\n sparse_add_section+0x1c9/0x26a\n __add_pages+0xbf/0x150\n add_pages+0x12/0x60\n add_memory_resource+0xc8/0x210\n __add_memory+0x62/0xb0\n acpi_memory_device_add+0x13f/0x300\n acpi_bus_attach+0xf6/0x200\n acpi_bus_scan+0x43/0x90\n acpi_device_hotplug+0x275/0x3d0\n acpi_hotplug_work_fn+0x1a/0x30\n process_one_work+0x1a7/0x370\n worker_thread+0x30/0x380\n kthread+0x112/0x130\n ret_from_fork+0x35/0x40\n\nWe should use memmap as it did.\n\nOn x86 the impact is limited to x86_32 builds, or x86_64 configurations\nthat override the default setting for SPARSEMEM_VMEMMAP.\n\nOther memory hotplug archs (arm64, ia64, and ppc) also default to\nSPARSEMEM_VMEMMAP\u003dy.\n\n[dan.j.williams@intel.com: changelog update]\n{rppt@linux.ibm.com: changelog update]\nLink: http://lkml.kernel.org/r/20200219030454.4844-1-bhe@redhat.com\nFixes: ba72b4c8cf60 (\"mm/sparsemem: support sub-section hotplug\")\nSigned-off-by: Wei Yang \u003crichardw.yang@linux.intel.com\u003e\nSigned-off-by: Baoquan He \u003cbhe@redhat.com\u003e\nAcked-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nReviewed-by: Baoquan He \u003cbhe@redhat.com\u003e\nReviewed-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Oscar Salvador \u003cosalvador@suse.de\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "76073c646f5f4999d763f471df9e38a5a912d70d", "tree": "2b23f10f9c07d063a77f1bcbf6b363ec90e62c9e", "parents": [ "c11d3fa0116a6bc832a9e387427caa16f8de5ef2" ], "author": { "name": "Gavin Shan", "email": "gshan@redhat.com", "time": "Thu Feb 20 20:04:24 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "mm/vmscan.c: don\u0027t round up scan size for online memory cgroup\n\nCommit 68600f623d69 (\"mm: don\u0027t miss the last page because of round-off\nerror\") makes the scan size round up to @denominator regardless of the\nmemory cgroup\u0027s state, online or offline. This affects the overall\nreclaiming behavior: the corresponding LRU list is eligible for\nreclaiming only when its size logically right shifted by @sc-\u003epriority\nis bigger than zero in the former formula.\n\nFor example, the inactive anonymous LRU list should have at least 0x4000\npages to be eligible for reclaiming when we have 60/12 for\nswappiness/priority and without taking scan/rotation ratio into account.\n\nAfter the roundup is applied, the inactive anonymous LRU list becomes\neligible for reclaiming when its size is bigger than or equal to 0x1000\nin the same condition.\n\n (0x4000 \u003e\u003e 12) * 60 / (60 + 140 + 1) \u003d 1\n ((0x1000 \u003e\u003e 12) * 60) + 200) / (60 + 140 + 1) \u003d 1\n\naarch64 has 512MB huge page size when the base page size is 64KB. The\nmemory cgroup that has a huge page is always eligible for reclaiming in\nthat case.\n\nThe reclaiming is likely to stop after the huge page is reclaimed,\nmeaing the further iteration on @sc-\u003epriority and the silbing and child\nmemory cgroups will be skipped. The overall behaviour has been changed.\nThis fixes the issue by applying the roundup to offlined memory cgroups\nonly, to give more preference to reclaim memory from offlined memory\ncgroup. It sounds reasonable as those memory is unlikedly to be used by\nanyone.\n\nThe issue was found by starting up 8 VMs on a Ampere Mustang machine,\nwhich has 8 CPUs and 16 GB memory. Each VM is given with 2 vCPUs and\n2GB memory. It took 264 seconds for all VMs to be completely up and\n784MB swap is consumed after that. With this patch applied, it took 236\nseconds and 60MB swap to do same thing. So there is 10% performance\nimprovement for my case. Note that KSM is disable while THP is enabled\nin the testing.\n\n total used free shared buff/cache available\n Mem: 16196 10065 2049 16 4081 3749\n Swap: 8175 784 7391\n total used free shared buff/cache available\n Mem: 16196 11324 3656 24 1215 2936\n Swap: 8175 60 8115\n\nLink: http://lkml.kernel.org/r/20200211024514.8730-1-gshan@redhat.com\nFixes: 68600f623d69 (\"mm: don\u0027t miss the last page because of round-off error\")\nSigned-off-by: Gavin Shan \u003cgshan@redhat.com\u003e\nAcked-by: Roman Gushchin \u003cguro@fb.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\t[4.20+]\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c11d3fa0116a6bc832a9e387427caa16f8de5ef2", "tree": "8efee1cd0afd84cb0dbd76db4a8b5047ea1a9689", "parents": [ "75866af62b439859d5146b7093ceb6b482852683" ], "author": { "name": "Alexandru Ardelean", "email": "alexandru.ardelean@analog.com", "time": "Thu Feb 20 20:04:21 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "lib/string.c: update match_string() doc-strings with correct behavior\n\nThere were a few attempts at changing behavior of the match_string()\nhelpers (i.e. \u0027match_string()\u0027 \u0026 \u0027sysfs_match_string()\u0027), to change \u0026\nextend the behavior according to the doc-string.\n\nBut the simplest approach is to just fix the doc-strings. The current\nbehavior is fine as-is, and some bugs were introduced trying to fix it.\n\nAs for extending the behavior, new helpers can always be introduced if\nneeded.\n\nThe match_string() helpers behave more like \u0027strncmp()\u0027 in the sense\nthat they go up to n elements or until the first NULL element in the\narray of strings.\n\nThis change updates the doc-strings with this info.\n\nLink: http://lkml.kernel.org/r/20200213072722.8249-1-alexandru.ardelean@analog.com\nSigned-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nAcked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: \"Tobin C . Harding\" \u003ctobin@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "75866af62b439859d5146b7093ceb6b482852683", "tree": "e7e0b1bffcc04b8a7cbae61084d8fb70255b8a78", "parents": [ "fed98ef4d8b665316479dd35cbd92d3e2ff470a3" ], "author": { "name": "Vasily Averin", "email": "vvs@virtuozzo.com", "time": "Thu Feb 20 20:04:18 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "mm/memcontrol.c: lost css_put in memcg_expand_shrinker_maps()\n\nfor_each_mem_cgroup() increases css reference counter for memory cgroup\nand requires to use mem_cgroup_iter_break() if the walk is cancelled.\n\nLink: http://lkml.kernel.org/r/c98414fb-7e1f-da0f-867a-9340ec4bd30b@virtuozzo.com\nFixes: 0a4465d34028 (\"mm, memcg: assign memcg-aware shrinkers bitmap to memcg\")\nSigned-off-by: Vasily Averin \u003cvvs@virtuozzo.com\u003e\nAcked-by: Kirill Tkhai \u003cktkhai@virtuozzo.com\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nReviewed-by: Roman Gushchin \u003cguro@fb.com\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Vladimir Davydov \u003cvdavydov.dev@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fed98ef4d8b665316479dd35cbd92d3e2ff470a3", "tree": "07e493a5bfc0ab14c8d02c796b79152f9cc14043", "parents": [ "0ef82fcefb99300ede6f4d38a8100845b2dc8e30" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Thu Feb 20 20:04:15 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "mm/swapfile.c: fix a comment in sys_swapon()\n\nclaim_swapfile now always takes i_rwsem.\n\nLink: http://lkml.kernel.org/r/20200114161225.309792-2-hch@lst.de\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0ef82fcefb99300ede6f4d38a8100845b2dc8e30", "tree": "898aea2effd80ac5816046680523049a29bd2faa", "parents": [ "ef0c08192ac09e29ddd676b3ca6c4a501f277f10" ], "author": { "name": "Douglas Anderson", "email": "dianders@chromium.org", "time": "Thu Feb 20 20:04:12 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "scripts/get_maintainer.pl: deprioritize old Fixes: addresses\n\nRecently, I found that get_maintainer was causing me to send emails to\nthe old addresses for maintainers. Since I usually just trust the\noutput of get_maintainer to know the right email address, I didn\u0027t even\nlook carefully and fired off two patch series that went to the wrong\nplace. Oops.\n\nThe problem was introduced recently when trying to add signatures from\nFixes. The problem was that these email addresses were added too early\nin the process of compiling our list of places to send. Things added to\nthe list earlier are considered more canonical and when we later added\nmaintainer entries we ended up deduplicating to the old address.\n\nHere are two examples using mainline commits (to make it easier to\nreplicate) for the two maintainers that I messed up recently:\n\n $ git format-patch d8549bcd0529~..d8549bcd0529\n $ ./scripts/get_maintainer.pl 0001-clk-Add-clk_hw*.patch | grep Boyd\n Stephen Boyd \u003csboyd@codeaurora.org\u003e...\n\n $ git format-patch 6d1238aa3395~..6d1238aa3395\n $ ./scripts/get_maintainer.pl 0001-arm64-dts-qcom-qcs404*.patch | grep Andy\n Andy Gross \u003candy.gross@linaro.org\u003e\n\nLet\u0027s move the adding of addresses from Fixes: to the end since the\nemail addresses from these are much more likely to be older.\n\nAfter this patch the above examples get the right addresses for the two\nexamples.\n\nLink: http://lkml.kernel.org/r/20200127095001.1.I41fba9f33590bfd92cd01960161d8384268c6569@changeid\nFixes: 2f5bd343694e (\"scripts/get_maintainer.pl: add signatures from Fixes: \u003cbadcommit\u003e lines in commit message\")\nSigned-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nAcked-by: Joe Perches \u003cjoe@perches.com\u003e\nCc: Stephen Boyd \u003csboyd@kernel.org\u003e\nCc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nCc: Andy Gross \u003cagross@kernel.org\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nCc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ef0c08192ac09e29ddd676b3ca6c4a501f277f10", "tree": "1b1c11603948ca6511ff7aac36cd0ee1d2b656e2", "parents": [ "9e69fa46275b63f670ced0f11095af1841c73fca" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Thu Feb 20 20:04:09 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "get_maintainer: remove uses of P: for maintainer name\n\nCommit 1ca84ed6425f (\"MAINTAINERS: Reclaim the P: tag for Maintainer\nEntry Profile\") changed the use of the \"P:\" tag from \"Person\" to\n\"Profile (ie: special subsystem coding styles and characteristics)\"\n\nChange how get_maintainer.pl parses the \"P:\" tag to match.\n\nLink: http://lkml.kernel.org/r/ca53823fc5d25c0be32ad937d0207a0589c08643.camel@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nAcked-by: Dan Williams \u003cdan.j.william@intel.com\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "9e69fa46275b63f670ced0f11095af1841c73fca", "tree": "6a7835508ea35901f7300e7d52db169309c7095d", "parents": [ "467d12f5c7842896d2de3ced74e4147ee29e97c8" ], "author": { "name": "SeongJae Park", "email": "sjpark@amazon.de", "time": "Thu Feb 20 20:04:06 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "selftests/vm: add missed tests in run_vmtests\n\nThe commits introducing \u0027mlock-random-test\u0027[1], \u0027map_fiex_noreplace\u0027[2],\nand \u0027thuge-gen\u0027[3] have not added those in the \u0027run_vmtests\u0027 script and\nthus the \u0027run_tests\u0027 command of kselftests doesn\u0027t run those. This\ncommit adds those in the script.\n\n\u0027gup_benchmark\u0027 and \u0027transhuge-stress\u0027 are also not included in the\n\u0027run_vmtests\u0027, but this commit does not add those because those are for\nperformance measurement rather than pass/fail tests.\n\n[1] commit 26b4224d9961 (\"selftests: expanding more mlock selftest\")\n[2] commit 91cbacc34512 (\"tools/testing/selftests/vm/map_fixed_noreplace.c: add test for MAP_FIXED_NOREPLACE\")\n[3] commit fcc1f2d5dd34 (\"selftests: add a test program for variable huge page sizes in mmap/shmget\")\n\nLink: http://lkml.kernel.org/r/20200206085144.29126-1-sj38.park@gmail.com\nSigned-off-by: SeongJae Park \u003csjpark@amazon.de\u003e\nCc: Uladzislau Rezki (Sony) \u003curezki@gmail.com\u003e\nCc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nCc: Shuah Khan \u003cshuah@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "467d12f5c7842896d2de3ced74e4147ee29e97c8", "tree": "c654c30e7df411761d9c964644b5abd477ac3c5e", "parents": [ "edf28f4061afe4c2d9eb1c3323d90e882c1d6800" ], "author": { "name": "Christian Borntraeger", "email": "borntraeger@de.ibm.com", "time": "Thu Feb 20 20:04:03 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "include/uapi/linux/swab.h: fix userspace breakage, use __BITS_PER_LONG for swap\n\nQEMU has a funny new build error message when I use the upstream kernel\nheaders:\n\n CC block/file-posix.o\n In file included from /home/cborntra/REPOS/qemu/include/qemu/timer.h:4,\n from /home/cborntra/REPOS/qemu/include/qemu/timed-average.h:29,\n from /home/cborntra/REPOS/qemu/include/block/accounting.h:28,\n from /home/cborntra/REPOS/qemu/include/block/block_int.h:27,\n from /home/cborntra/REPOS/qemu/block/file-posix.c:30:\n /usr/include/linux/swab.h: In function `__swab\u0027:\n /home/cborntra/REPOS/qemu/include/qemu/bitops.h:20:34: error: \"sizeof\" is not defined, evaluates to 0 [-Werror\u003dundef]\n 20 | #define BITS_PER_LONG (sizeof (unsigned long) * BITS_PER_BYTE)\n | ^~~~~~\n /home/cborntra/REPOS/qemu/include/qemu/bitops.h:20:41: error: missing binary operator before token \"(\"\n 20 | #define BITS_PER_LONG (sizeof (unsigned long) * BITS_PER_BYTE)\n | ^\n cc1: all warnings being treated as errors\n make: *** [/home/cborntra/REPOS/qemu/rules.mak:69: block/file-posix.o] Error 1\n rm tests/qemu-iotests/socket_scm_helper.o\n\nThis was triggered by commit d5767057c9a (\"uapi: rename ext2_swab() to\nswab() and share globally in swab.h\"). That patch is doing\n\n #include \u003casm/bitsperlong.h\u003e\n\nbut it uses BITS_PER_LONG.\n\nThe kernel file asm/bitsperlong.h provide only __BITS_PER_LONG.\n\nLet us use the __ variant in swap.h\n\nLink: http://lkml.kernel.org/r/20200213142147.17604-1-borntraeger@de.ibm.com\nFixes: d5767057c9a (\"uapi: rename ext2_swab() to swab() and share globally in swab.h\")\nSigned-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nCc: Yury Norov \u003cyury.norov@gmail.com\u003e\nCc: Allison Randal \u003callison@lohutok.net\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: William Breathitt Gray \u003cvilhelm.gray@gmail.com\u003e\nCc: Torsten Hilbrich \u003ctorsten.hilbrich@secunet.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "edf28f4061afe4c2d9eb1c3323d90e882c1d6800", "tree": "55fe4c350dde81a7ce4cfc342ff3356fcea30bb7", "parents": [ "c766d1472c70d25ad475cf56042af1652e792b23" ], "author": { "name": "Ioanna Alifieraki", "email": "ioanna-maria.alifieraki@canonical.com", "time": "Thu Feb 20 20:04:00 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "Revert \"ipc,sem: remove uneeded sem_undo_list lock usage in exit_sem()\"\n\nThis reverts commit a97955844807e327df11aa33869009d14d6b7de0.\n\nCommit a97955844807 (\"ipc,sem: remove uneeded sem_undo_list lock usage\nin exit_sem()\") removes a lock that is needed. This leads to a process\nlooping infinitely in exit_sem() and can also lead to a crash. There is\na reproducer available in [1] and with the commit reverted the issue\ndoes not reproduce anymore.\n\nUsing the reproducer found in [1] is fairly easy to reach a point where\none of the child processes is looping infinitely in exit_sem between\nfor(;;) and if (semid \u003d\u003d -1) block, while it\u0027s trying to free its last\nsem_undo structure which has already been freed by freeary().\n\nEach sem_undo struct is on two lists: one per semaphore set (list_id)\nand one per process (list_proc). The list_id list tracks undos by\nsemaphore set, and the list_proc by process.\n\nUndo structures are removed either by freeary() or by exit_sem(). The\nfreeary function is invoked when the user invokes a syscall to remove a\nsemaphore set. During this operation freeary() traverses the list_id\nassociated with the semaphore set and removes the undo structures from\nboth the list_id and list_proc lists.\n\nFor this case, exit_sem() is called at process exit. Each process\ncontains a struct sem_undo_list (referred to as \"ulp\") which contains\nthe head for the list_proc list. When the process exits, exit_sem()\ntraverses this list to remove each sem_undo struct. As in freeary(),\nwhenever a sem_undo struct is removed from list_proc, it is also removed\nfrom the list_id list.\n\nRemoving elements from list_id is safe for both exit_sem() and freeary()\ndue to sem_lock(). Removing elements from list_proc is not safe;\nfreeary() locks \u0026un-\u003eulp-\u003elock when it performs\nlist_del_rcu(\u0026un-\u003elist_proc) but exit_sem() does not (locking was\nremoved by commit a97955844807 (\"ipc,sem: remove uneeded sem_undo_list\nlock usage in exit_sem()\").\n\nThis can result in the following situation while executing the\nreproducer [1] : Consider a child process in exit_sem() and the parent\nin freeary() (because of semctl(sid[i], NSEM, IPC_RMID)).\n\n - The list_proc for the child contains the last two undo structs A and\n B (the rest have been removed either by exit_sem() or freeary()).\n\n - The semid for A is 1 and semid for B is 2.\n\n - exit_sem() removes A and at the same time freeary() removes B.\n\n - Since A and B have different semid sem_lock() will acquire different\n locks for each process and both can proceed.\n\nThe bug is that they remove A and B from the same list_proc at the same\ntime because only freeary() acquires the ulp lock. When exit_sem()\nremoves A it makes ulp-\u003elist_proc.next to point at B and at the same\ntime freeary() removes B setting B-\u003esemid\u003d-1.\n\nAt the next iteration of for(;;) loop exit_sem() will try to remove B.\n\nThe only way to break from for(;;) is for (\u0026un-\u003elist_proc \u003d\u003d\n\u0026ulp-\u003elist_proc) to be true which is not. Then exit_sem() will check if\nB-\u003esemid\u003d-1 which is and will continue looping in for(;;) until the\nmemory for B is reallocated and the value at B-\u003esemid is changed.\n\nAt that point, exit_sem() will crash attempting to unlink B from the\nlists (this can be easily triggered by running the reproducer [1] a\nsecond time).\n\nTo prove this scenario instrumentation was added to keep information\nabout each sem_undo (un) struct that is removed per process and per\nsemaphore set (sma).\n\n CPU0 CPU1\n [caller holds sem_lock(sma for A)] ...\n freeary() exit_sem()\n ... ...\n ... sem_lock(sma for B)\n spin_lock(A-\u003eulp-\u003elock) ...\n list_del_rcu(un_A-\u003elist_proc) list_del_rcu(un_B-\u003elist_proc)\n\nUndo structures A and B have different semid and sem_lock() operations\nproceed. However they belong to the same list_proc list and they are\nremoved at the same time. This results into ulp-\u003elist_proc.next\npointing to the address of B which is already removed.\n\nAfter reverting commit a97955844807 (\"ipc,sem: remove uneeded\nsem_undo_list lock usage in exit_sem()\") the issue was no longer\nreproducible.\n\n[1] https://bugzilla.redhat.com/show_bug.cgi?id\u003d1694779\n\nLink: http://lkml.kernel.org/r/20191211191318.11860-1-ioanna-maria.alifieraki@canonical.com\nFixes: a97955844807 (\"ipc,sem: remove uneeded sem_undo_list lock usage in exit_sem()\")\nSigned-off-by: Ioanna Alifieraki \u003cioanna-maria.alifieraki@canonical.com\u003e\nAcked-by: Manfred Spraul \u003cmanfred@colorfullife.com\u003e\nAcked-by: Herton R. Krzesinski \u003cherton@redhat.com\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: \u003cmalat@debian.org\u003e\nCc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nCc: Jay Vosburgh \u003cjay.vosburgh@canonical.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c766d1472c70d25ad475cf56042af1652e792b23", "tree": "de6847beef3a80352b3c1eb4fd50657199c80c1c", "parents": [ "412c53a680a97cb1ae2c0ab60230e193bee86387" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Feb 20 20:03:57 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "y2038: hide timeval/timespec/itimerval/itimerspec types\n\nThere are no in-kernel users remaining, but there may still be users that\ninclude linux/time.h instead of sys/time.h from user space, so leave the\ntypes available to user space while hiding them from kernel space.\n\nOnly the __kernel_old_* versions of these types remain now.\n\nLink: http://lkml.kernel.org/r/20200110154232.4104492-4-arnd@arndb.de\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nAcked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "412c53a680a97cb1ae2c0ab60230e193bee86387", "tree": "e45468f621d037bf6622c932573480149b70af0d", "parents": [ "595abbaff5db121428247a2e6ab368734472e101" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Feb 20 20:03:54 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "y2038: remove unused time32 interfaces\n\nNo users remain, so kill these off before we grow new ones.\n\nLink: http://lkml.kernel.org/r/20200110154232.4104492-3-arnd@arndb.de\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nAcked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "595abbaff5db121428247a2e6ab368734472e101", "tree": "1c0810a3a15fe44052620aaa4732b59fb07ac5dd", "parents": [ "ca7e1fd1026c5af6a533b4b5447e1d2f153e28f2" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Feb 20 20:03:50 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 11:22:15 2020 -0800" }, "message": "y2038: remove ktime to/from timespec/timeval conversion\n\nA couple of helpers are now obsolete and can be removed, so drivers can no\nlonger start using them and instead use y2038-safe interfaces.\n\nLink: http://lkml.kernel.org/r/20200110154232.4104492-2-arnd@arndb.de\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nAcked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "63fb9623427fbb44e3782233b6e4714057b76ff2", "tree": "6659a2f1e164368abeb538c677e1f63634fc7c54", "parents": [ "ebe7acadf5a9c4b67475e766e0e80ae6a2a79c61" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Feb 21 01:46:18 2020 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 21 10:01:25 2020 -0800" }, "message": "ACPI: PM: s2idle: Check fixed wakeup events in acpi_s2idle_wake()\n\nCommit fdde0ff8590b (\"ACPI: PM: s2idle: Prevent spurious SCIs from\nwaking up the system\") overlooked the fact that fixed events can wake\nup the system too and broke RTC wakeup from suspend-to-idle as a\nresult.\n\nFix this issue by checking the fixed events in acpi_s2idle_wake() in\naddition to checking wakeup GPEs and break out of the suspend-to-idle\nloop if the status bits of any enabled fixed events are set then.\n\nFixes: fdde0ff8590b (\"ACPI: PM: s2idle: Prevent spurious SCIs from waking up the system\")\nReported-and-tested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: 5.4+ \u003cstable@vger.kernel.org\u003e # 5.4+\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e61d2392253220477813b43412090dccdcac601f", "tree": "bcadd282c6f4c3564daeeafb48b17265bef778b0", "parents": [ "badcd4546d52ae4318f2bcfda0e47a1394b60e38" ], "author": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Thu Feb 20 06:29:48 2020 -0800" }, "committer": { "name": "Guenter Roeck", "email": "linux@roeck-us.net", "time": "Fri Feb 21 09:16:24 2020 -0800" }, "message": "hwmon: (w83627ehf) Fix crash seen with W83627DHG-P\n\nLoading the driver on a system with W83627DHG-P crashes as follows.\n\nw83627ehf: Found W83627DHG-P chip at 0x290\nBUG: kernel NULL pointer dereference, address: 0000000000000000\nPGD 0 P4D 0\nOops: 0000 [#1] SMP NOPTI\nCPU: 0 PID: 604 Comm: sensors Not tainted 5.6.0-rc2-00055-gca7e1fd1026c #29\nHardware name: /D425KT, BIOS MWPNT10N.86A.0132.2013.0726.1534 07/26/2013\nRIP: 0010:w83627ehf_read_string+0x27/0x70 [w83627ehf]\nCode: [... ]\nRSP: 0018:ffffb95980657df8 EFLAGS: 00010293\nRAX: 0000000000000000 RBX: ffff96caaa7f5218 RCX: 0000000000000000\nRDX: 0000000000000015 RSI: 0000000000000001 RDI: ffff96caa736ec08\nRBP: 0000000000000000 R08: ffffb95980657e20 R09: 0000000000000001\nR10: ffff96caaa635cc0 R11: 0000000000000000 R12: ffff96caa9f7cf00\nR13: ffff96caa9ec3d00 R14: ffff96caa9ec3d28 R15: ffff96caa9ec3d40\nFS: 00007fbc7c4e2740(0000) GS:ffff96caabc00000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: 0000000000000000 CR3: 0000000129d58000 CR4: 00000000000006f0\nCall Trace:\n ? cp_new_stat+0x12d/0x160\n hwmon_attr_show_string+0x37/0x70 [hwmon]\n dev_attr_show+0x14/0x50\n sysfs_kf_seq_show+0xb5/0x1b0\n seq_read+0xcf/0x460\n vfs_read+0x9b/0x150\n ksys_read+0x5f/0xe0\n do_syscall_64+0x48/0x190\n entry_SYSCALL_64_after_hwframe+0x44/0xa9\n...\n\nTemperature labels are not always present. Adjust sysfs attribute\nvisibility accordingly.\n\nReported-by: Meelis Roos \u003cmroos@linux.ee\u003e\nSuggested-by: Dr. David Alan Gilbert \u003clinux@treblig.org\u003e\nReviewed-by: Dr. David Alan Gilbert \u003clinux@treblig.org\u003e\nCc: Meelis Roos \u003cmroos@linux.ee\u003e\nCc: Dr. David Alan Gilbert \u003clinux@treblig.org\u003e\nFixes: 266cd5835947 (\"hwmon: (w83627ehf) convert to with_info interface\")\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n" }, { "commit": "ae7bbc091351a4c6ebddfbe36eb5eb7a149cd7a2", "tree": "cd8829335a71ad2dc9813b02f7353980986a80b0", "parents": [ "f25372ffc3f6c2684b57fb718219137e6ee2b64c", "3b7830904e17202524bad1974505a9bfc718d31f" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Feb 21 09:18:00 2020 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Feb 21 09:18:00 2020 -0700" }, "message": "Merge branch \u0027nvme-5.6-rc3\u0027 of git://git.infradead.org/nvme into block-5.6\n\nPull NVMe fixes from Keith.\n\n* \u0027nvme-5.6-rc3\u0027 of git://git.infradead.org/nvme:\n nvme-multipath: Fix memory leak with ana_log_buf\n nvme: Fix uninitialized-variable warning\n nvme-pci: Use single IRQ vector for old Apple models\n nvme/pci: Add sleep quirk for Samsung and Toshiba drives\n" }, { "commit": "7143b5ac5750f404ff3a594b34fdf3fc2f99f828", "tree": "d6bf0edaaf6f0f994d2043d91a91644b2e6ca461", "parents": [ "929a3af90f0f4bd7132d83552c1a98c83f60ef7e" ], "author": { "name": "Stefano Garzarella", "email": "sgarzare@redhat.com", "time": "Fri Feb 21 16:42:16 2020 +0100" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Feb 21 09:16:10 2020 -0700" }, "message": "io_uring: prevent sq_thread from spinning when it should stop\n\nThis patch drops \u0027cur_mm\u0027 before calling cond_resched(), to prevent\nthe sq_thread from spinning even when the user process is finished.\n\nBefore this patch, if the user process ended without closing the\nio_uring fd, the sq_thread continues to spin until the\n\u0027sq_thread_idle\u0027 timeout ends.\n\nIn the worst case where the \u0027sq_thread_idle\u0027 parameter is bigger than\nINT_MAX, the sq_thread will spin forever.\n\nFixes: 6c271ce2f1d5 (\"io_uring: add submission polling\")\nSigned-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "a5ae50dea9111db63d30d700766dd5509602f7ad", "tree": "8e9335ea7dc0484f65482c1d6918eaeeb0184858", "parents": [ "e75fd33b3f744f644061a4f9662bd63f5434f806" ], "author": { "name": "Filipe Manana", "email": "fdmanana@suse.com", "time": "Thu Feb 20 13:29:49 2020 +0000" }, "committer": { "name": "David Sterba", "email": "dsterba@suse.com", "time": "Fri Feb 21 16:21:19 2020 +0100" }, "message": "Btrfs: fix deadlock during fast fsync when logging prealloc extents beyond eof\n\nWhile logging the prealloc extents of an inode during a fast fsync we call\nbtrfs_truncate_inode_items(), through btrfs_log_prealloc_extents(), while\nholding a read lock on a leaf of the inode\u0027s root (not the log root, the\nfs/subvol root), and then that function locks the file range in the inode\u0027s\niotree. This can lead to a deadlock when:\n\n* the fsync is ranged\n\n* the file has prealloc extents beyond eof\n\n* writeback for a range different from the fsync range starts\n during the fsync\n\n* the size of the file is not sector size aligned\n\nBecause when finishing an ordered extent we lock first a file range and\nthen try to COW the fs/subvol tree to insert an extent item.\n\nThe following diagram shows how the deadlock can happen.\n\n CPU 1 CPU 2\n\n btrfs_sync_file()\n --\u003e for range [0, 1MiB)\n\n --\u003e inode has a size of\n 1MiB and has 1 prealloc\n extent beyond the\n i_size, starting at offset\n 4MiB\n\n flushes all delalloc for the\n range [0MiB, 1MiB) and waits\n for the respective ordered\n extents to complete\n\n --\u003e before task at CPU 1 locks the\n inode, a write into file range\n [1MiB, 2MiB + 1KiB) is made\n\n --\u003e i_size is updated to 2MiB + 1KiB\n\n --\u003e writeback is started for that\n range, [1MiB, 2MiB + 4KiB)\n --\u003e end offset rounded up to\n be sector size aligned\n\n btrfs_log_dentry_safe()\n btrfs_log_inode_parent()\n btrfs_log_inode()\n\n btrfs_log_changed_extents()\n btrfs_log_prealloc_extents()\n --\u003e does a search on the\n inode\u0027s root\n --\u003e holds a read lock on\n leaf X\n\n btrfs_finish_ordered_io()\n --\u003e locks range [1MiB, 2MiB + 4KiB)\n --\u003e end offset rounded up\n to be sector size aligned\n\n --\u003e tries to cow leaf X, through\n insert_reserved_file_extent()\n --\u003e already locked by the\n task at CPU 1\n\n btrfs_truncate_inode_items()\n\n --\u003e gets an i_size of\n 2MiB + 1KiB, which is\n not sector size\n aligned\n\n --\u003e tries to lock file\n range [2MiB, (u64)-1)\n --\u003e the start range\n is rounded down\n from 2MiB + 1K\n to 2MiB to be sector\n size aligned\n\n --\u003e but the subrange\n [2MiB, 2MiB + 4KiB) is\n already locked by\n task at CPU 2 which\n is waiting to get a\n write lock on leaf X\n for which we are\n holding a read lock\n\n *** deadlock ***\n\nThis results in a stack trace like the following, triggered by test case\ngeneric/561 from fstests:\n\n [ 2779.973608] INFO: task kworker/u8:6:247 blocked for more than 120 seconds.\n [ 2779.979536] Not tainted 5.6.0-rc2-btrfs-next-53 #1\n [ 2779.984503] \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n [ 2779.990136] kworker/u8:6 D 0 247 2 0x80004000\n [ 2779.990457] Workqueue: btrfs-endio-write btrfs_work_helper [btrfs]\n [ 2779.990466] Call Trace:\n [ 2779.990491] ? __schedule+0x384/0xa30\n [ 2779.990521] schedule+0x33/0xe0\n [ 2779.990616] btrfs_tree_read_lock+0x19e/0x2e0 [btrfs]\n [ 2779.990632] ? remove_wait_queue+0x60/0x60\n [ 2779.990730] btrfs_read_lock_root_node+0x2f/0x40 [btrfs]\n [ 2779.990782] btrfs_search_slot+0x510/0x1000 [btrfs]\n [ 2779.990869] btrfs_lookup_file_extent+0x4a/0x70 [btrfs]\n [ 2779.990944] __btrfs_drop_extents+0x161/0x1060 [btrfs]\n [ 2779.990987] ? mark_held_locks+0x6d/0xc0\n [ 2779.990994] ? __slab_alloc.isra.49+0x99/0x100\n [ 2779.991060] ? insert_reserved_file_extent.constprop.19+0x64/0x300 [btrfs]\n [ 2779.991145] insert_reserved_file_extent.constprop.19+0x97/0x300 [btrfs]\n [ 2779.991222] ? start_transaction+0xdd/0x5c0 [btrfs]\n [ 2779.991291] btrfs_finish_ordered_io+0x4f4/0x840 [btrfs]\n [ 2779.991405] btrfs_work_helper+0xaa/0x720 [btrfs]\n [ 2779.991432] process_one_work+0x26d/0x6a0\n [ 2779.991460] worker_thread+0x4f/0x3e0\n [ 2779.991481] ? process_one_work+0x6a0/0x6a0\n [ 2779.991489] kthread+0x103/0x140\n [ 2779.991499] ? kthread_create_worker_on_cpu+0x70/0x70\n [ 2779.991515] ret_from_fork+0x3a/0x50\n (...)\n [ 2780.026211] INFO: task fsstress:17375 blocked for more than 120 seconds.\n [ 2780.027480] Not tainted 5.6.0-rc2-btrfs-next-53 #1\n [ 2780.028482] \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n [ 2780.030035] fsstress D 0 17375 17373 0x00004000\n [ 2780.030038] Call Trace:\n [ 2780.030044] ? __schedule+0x384/0xa30\n [ 2780.030052] schedule+0x33/0xe0\n [ 2780.030075] lock_extent_bits+0x20c/0x320 [btrfs]\n [ 2780.030094] ? btrfs_truncate_inode_items+0xf4/0x1150 [btrfs]\n [ 2780.030098] ? rcu_read_lock_sched_held+0x59/0xa0\n [ 2780.030102] ? remove_wait_queue+0x60/0x60\n [ 2780.030122] btrfs_truncate_inode_items+0x133/0x1150 [btrfs]\n [ 2780.030151] ? btrfs_set_path_blocking+0xb2/0x160 [btrfs]\n [ 2780.030165] ? btrfs_search_slot+0x379/0x1000 [btrfs]\n [ 2780.030195] btrfs_log_changed_extents.isra.8+0x841/0x93e [btrfs]\n [ 2780.030202] ? do_raw_spin_unlock+0x49/0xc0\n [ 2780.030215] ? btrfs_get_num_csums+0x10/0x10 [btrfs]\n [ 2780.030239] btrfs_log_inode+0xf83/0x1124 [btrfs]\n [ 2780.030251] ? __mutex_unlock_slowpath+0x45/0x2a0\n [ 2780.030275] btrfs_log_inode_parent+0x2a0/0xe40 [btrfs]\n [ 2780.030282] ? dget_parent+0xa1/0x370\n [ 2780.030309] btrfs_log_dentry_safe+0x4a/0x70 [btrfs]\n [ 2780.030329] btrfs_sync_file+0x3f3/0x490 [btrfs]\n [ 2780.030339] do_fsync+0x38/0x60\n [ 2780.030343] __x64_sys_fdatasync+0x13/0x20\n [ 2780.030345] do_syscall_64+0x5c/0x280\n [ 2780.030348] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 2780.030356] RIP: 0033:0x7f2d80f6d5f0\n [ 2780.030361] Code: Bad RIP value.\n [ 2780.030362] RSP: 002b:00007ffdba3c8548 EFLAGS: 00000246 ORIG_RAX: 000000000000004b\n [ 2780.030364] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f2d80f6d5f0\n [ 2780.030365] RDX: 00007ffdba3c84b0 RSI: 00007ffdba3c84b0 RDI: 0000000000000003\n [ 2780.030367] RBP: 000000000000004a R08: 0000000000000001 R09: 00007ffdba3c855c\n [ 2780.030368] R10: 0000000000000078 R11: 0000000000000246 R12: 00000000000001f4\n [ 2780.030369] R13: 0000000051eb851f R14: 00007ffdba3c85f0 R15: 0000557a49220d90\n\nSo fix this by making btrfs_truncate_inode_items() not lock the range in\nthe inode\u0027s iotree when the target root is a log root, since it\u0027s not\nneeded to lock the range for log roots as the protection from the inode\u0027s\nlock and log_mutex are all that\u0027s needed.\n\nFixes: 28553fa992cb28 (\"Btrfs: fix race between shrinking truncate and fiemap\")\nCC: stable@vger.kernel.org # 4.4+\nReviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nSigned-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\nSigned-off-by: David Sterba \u003cdsterba@suse.com\u003e\n" }, { "commit": "3b7830904e17202524bad1974505a9bfc718d31f", "tree": "cd8829335a71ad2dc9813b02f7353980986a80b0", "parents": [ "15755854d53b4bbb0bb37a0fce66f0156cfc8a17" ], "author": { "name": "Logan Gunthorpe", "email": "logang@deltatee.com", "time": "Thu Feb 20 13:29:53 2020 -0700" }, "committer": { "name": "Keith Busch", "email": "kbusch@kernel.org", "time": "Fri Feb 21 23:52:25 2020 +0900" }, "message": "nvme-multipath: Fix memory leak with ana_log_buf\n\nkmemleak reports a memory leak with the ana_log_buf allocated by\nnvme_mpath_init():\n\nunreferenced object 0xffff888120e94000 (size 8208):\n comm \"nvme\", pid 6884, jiffies 4295020435 (age 78786.312s)\n hex dump (first 32 bytes):\n 00 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 ................\n 01 00 00 00 01 00 00 00 00 00 00 00 00 00 00 00 ................\n backtrace:\n [\u003c00000000e2360188\u003e] kmalloc_order+0x97/0xc0\n [\u003c0000000079b18dd4\u003e] kmalloc_order_trace+0x24/0x100\n [\u003c00000000f50c0406\u003e] __kmalloc+0x24c/0x2d0\n [\u003c00000000f31a10b9\u003e] nvme_mpath_init+0x23c/0x2b0\n [\u003c000000005802589e\u003e] nvme_init_identify+0x75f/0x1600\n [\u003c0000000058ef911b\u003e] nvme_loop_configure_admin_queue+0x26d/0x280\n [\u003c00000000673774b9\u003e] nvme_loop_create_ctrl+0x2a7/0x710\n [\u003c00000000f1c7a233\u003e] nvmf_dev_write+0xc66/0x10b9\n [\u003c000000004199f8d0\u003e] __vfs_write+0x50/0xa0\n [\u003c0000000065466fef\u003e] vfs_write+0xf3/0x280\n [\u003c00000000b0db9a8b\u003e] ksys_write+0xc6/0x160\n [\u003c0000000082156b91\u003e] __x64_sys_write+0x43/0x50\n [\u003c00000000c34fbb6d\u003e] do_syscall_64+0x77/0x2f0\n [\u003c00000000bbc574c9\u003e] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\nnvme_mpath_init() is called by nvme_init_identify() which is called in\nmultiple places (nvme_reset_work(), nvme_passthru_end(), etc). This\nmeans nvme_mpath_init() may be called multiple times before\nnvme_mpath_uninit() (which is only called on nvme_free_ctrl()).\n\nWhen nvme_mpath_init() is called multiple times, it overwrites the\nana_log_buf pointer with a new allocation, thus leaking the previous\nallocation.\n\nTo fix this, free ana_log_buf before allocating a new one.\n\nFixes: 0d0b660f214dc490 (\"nvme: add ANA support\")\nCc: \u003cstable@vger.kernel.org\u003e\nReviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nSigned-off-by: Keith Busch \u003ckbusch@kernel.org\u003e\n" }, { "commit": "2546287c5fb363a0165933ae2181c92f03e701d0", "tree": "cfdc6c5fb308d2003199b8f623748a71de95056b", "parents": [ "cba6437a1854fde5934098ec3bd0ee83af3129f5" ], "author": { "name": "Zenghui Yu", "email": "yuzenghui@huawei.com", "time": "Fri Feb 21 10:07:25 2020 +0800" }, "committer": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Fri Feb 21 11:29:15 2020 +0100" }, "message": "genirq/irqdomain: Make sure all irq domain flags are distinct\n\nThis was noticed when printing debugfs for MSIs on my ARM64 server. The\nnew dstate IRQD_MSI_NOMASK_QUIRK came out surprisingly while it should only\nbe the x86 stuff for the time being...\n\nThe new MSI quirk flag uses the same bit as IRQ_DOMAIN_NAME_ALLOCATED which\nis oddly defined as bit 6 for no good reason.\n\nSwitch it to the non used bit 1.\n\nFixes: 6f1a4891a592 (\"x86/apic/msi: Plug non-maskable MSI affinity race\")\nSigned-off-by: Zenghui Yu \u003cyuzenghui@huawei.com\u003e\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: stable@vger.kernel.org\nLink: https://lkml.kernel.org/r/20200221020725.2038-1-yuzenghui@huawei.com\n" }, { "commit": "4c5fd3b791a06021084b42d5610400f846d206b5", "tree": "dcc991eb7882f018f2688b8461992d866bf03978", "parents": [ "11a48a5a18c63fd7621bb050228cebf13566e4d8" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Wed Feb 19 17:28:21 2020 -0800" }, "committer": { "name": "Damien Le Moal", "email": "damien.lemoal@wdc.com", "time": "Fri Feb 21 18:09:26 2020 +0900" }, "message": "zonefs: fix documentation typos etc.\n\nFix typos, spellos, etc. in zonefs.txt.\n\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nCc: Damien Le Moal \u003cDamien.LeMoal@wdc.com\u003e\nReviewed-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nSigned-off-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n" }, { "commit": "0b9f386c4be6493d282aab0af6f9b70c62142777", "tree": "41b2b3541a5d776249ddafa659408ba88ece742b", "parents": [ "5b49c82dadfe0f3741778f57385f964ec1514863" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Wed Feb 12 10:24:52 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Implement copy_thread_tls\n\nThis is required for clone3 which passes the TLS value through a\nstruct rather than a register.\n\nCc: Amanieu d\u0027Antras \u003camanieu@gmail.com\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "5b49c82dadfe0f3741778f57385f964ec1514863", "tree": "638078189f9c04cd55b7657be9f038d6e053165f", "parents": [ "dc2efc0028dd5c4bf0f7324806c9a196958aaad3" ], "author": { "name": "MaJun", "email": "majun258@linux.alibaba.com", "time": "Mon Jan 27 10:56:21 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Add PCI support\n\nAdd the pci related code for csky arch to support basic pci virtual\nfunction, such as qemu virt-pci-9pfs.\n\nSigned-off-by: MaJun \u003cmajun258@linux.alibaba.com\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "dc2efc0028dd5c4bf0f7324806c9a196958aaad3", "tree": "a7f1818d649648a3b2dc912efadfae0d338d3871", "parents": [ "d46869aaab7981f9153d3271228005a52ef18591" ], "author": { "name": "Ma Jun", "email": "majun258@linux.alibaba.com", "time": "Sun Feb 02 17:56:58 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Minimize defconfig to support buildroot config.fragment\n\nSome bsp (eg: buildroot) has defconfig.fragment design to add more\nconfigs into the defconfig in linux source code tree. For example,\nwe could put different cpu configs into different defconfig.fragments,\nbut they all use the same defconfig in Linux.\n\nSigned-off-by: Ma Jun \u003cmajun258@linux.alibaba.com\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "d46869aaab7981f9153d3271228005a52ef18591", "tree": "a3e7481fab17fd358f1a02f6bbfb34e97486674f", "parents": [ "4ec575b78521b4573e85aec451930cb040582455" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Tue Oct 08 14:25:13 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Add setup_initrd check code\n\nWe should give some necessary check for initrd just like other\narchitectures and it seems that setup_initrd() could be a common\ncode for all architectures.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "4ec575b78521b4573e85aec451930cb040582455", "tree": "3e259eab72074e03e197043dbc9ee83d57d6d037", "parents": [ "bebd26ab623616728d6e72b5c74a47bfff5287d8" ], "author": { "name": "Krzysztof Kozlowski", "email": "krzk@kernel.org", "time": "Thu Jan 30 20:22:40 2020 +0100" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Cleanup old Kconfig options\n\nCONFIG_CLKSRC_OF is gone since commit bb0eb050a577\n(\"clocksource/drivers: Rename CLKSRC_OF to TIMER_OF\"). The platform\nalready selects TIMER_OF.\n\nCONFIG_HAVE_DMA_API_DEBUG is gone since commit 6e88628d03dd (\"dma-debug:\nremove CONFIG_HAVE_DMA_API_DEBUG\").\n\nCONFIG_DEFAULT_DEADLINE is gone since commit f382fb0bcef4 (\"block:\nremove legacy IO schedulers\").\n\nSigned-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "bebd26ab623616728d6e72b5c74a47bfff5287d8", "tree": "3e9b3add17730aae39a510bb1c0049943f700f97", "parents": [ "2305f60b76110cb3e8658a4ae85d1f7eb0c66a5b" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Fri Jan 31 17:52:30 2020 -0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "arch/csky: fix some Kconfig typos\n\nFix wording in help text for the CPU_HAS_LDSTEX symbol.\n\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nSigned-off-by: Guo Ren \u003cguoren@kernel.org\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "2305f60b76110cb3e8658a4ae85d1f7eb0c66a5b", "tree": "1a04333e112a4ed2346c12ee532f718443d5a6dc", "parents": [ "9025fd48a8aeddade845afa353d4bbab7f19dbf2" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sun Feb 02 14:11:11 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Fixup compile warning for three unimplemented syscalls\n\nImplement fstat64, fstatat64, clone3 syscalls to fixup\nchecksyscalls.sh compile warnings.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "9025fd48a8aeddade845afa353d4bbab7f19dbf2", "tree": "5b71f8e8e62a80b8e4316b5f3e30d89dc322b470", "parents": [ "359ae00d12589c31cf103894d0f32588d523ca83" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sun Feb 02 10:58:38 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:25 2020 +0800" }, "message": "csky: Remove unused cache implementation\n\nOnly for coding convention, these codes are unnecessary for abiv2.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "359ae00d12589c31cf103894d0f32588d523ca83", "tree": "342c0370d09ec4e72d8f894212c79090cf53fad4", "parents": [ "997153b9a75c08d545ad45e6f8ceb432435d2425" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sun Feb 02 09:58:42 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Fixup ftrace modify panic\n\nDuring ftrace init, linux will replace all function prologues\n(call_mcout) with nops, but it need flush_dcache and\ninvalidate_icache to make it work. So flush_cache functions\ncouldn\u0027t be nested called by ftrace framework.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "997153b9a75c08d545ad45e6f8ceb432435d2425", "tree": "830915f73d05e7d094d2f293a712e05c0da1e6f5", "parents": [ "cc1f6563a92ced0889775d0587316d725b6e1a68" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Jan 31 20:33:10 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Add flush_icache_mm to defer flush icache all\n\nSome CPUs don\u0027t support icache.va instruction to maintain the whole\nsmp cores\u0027 icache. Using icache.all + IPI casue a lot on performace\nand using defer mechanism could reduce the number of calling icache\n_flush_all functions.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "cc1f6563a92ced0889775d0587316d725b6e1a68", "tree": "2338b9a232542393b39987b2a6bb5b16951c8ef0", "parents": [ "d936a7e708dcf22344c4420e8b0e36f5d5f8c073" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Mon Jan 27 19:57:29 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Optimize abiv2 copy_to_user_page with VM_EXEC\n\nOnly when vma is for VM_EXEC, we need sync dcache \u0026 icache. eg:\n - gdb ptrace modify user space instruction code area.\n\nAdd VM_EXEC condition to reduce unnecessary cache flush.\n\nThe abiv1 cpus\u0027 cache are all VIPT, so we still need to deal with\ndcache aliasing problem. But there is optimized way to use cache\ncolor, just like what\u0027s done in arch/csky/abiv1/inc/abi/page.h.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "d936a7e708dcf22344c4420e8b0e36f5d5f8c073", "tree": "b9b305756f2fb2696bb14f649b9ec5ccebb2858c", "parents": [ "a1176734132c630b50908c36563e05fb3599682c" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Mon Jan 27 01:20:36 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Enable defer flush_dcache_page for abiv2 cpus (807/810/860)\n\nInstead of flushing cache per update_mmu_cache() called, we use\nflush_dcache_page to reduce the frequency of flashing the cache.\n\nAs abiv2 cpus are all PIPT for icache \u0026 dcache, we needn\u0027t handle\ndcache aliasing problem. But their icache can\u0027t snoop dcache, so\nwe still need sync_icache_dcache in update_mmu_cache().\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "a1176734132c630b50908c36563e05fb3599682c", "tree": "cb054b34a4fb5e0a58c7c5318a5a0ec06432c88a", "parents": [ "761b4f694cb90b63ca2739ac8a8a176342636e5e" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sat Jan 25 00:37:09 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Remove unnecessary flush_icache_* implementation\n\nThe abiv2 CPUs are all PIPT cache, so there is no need to implement\nflush_icache_page function.\n\nThe function flush_icache_user_range hasn\u0027t been used, so just\nremove it.\n\nThe function flush_cache_range is not necessary for PIPT cache when\ntlb mapping changed.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "761b4f694cb90b63ca2739ac8a8a176342636e5e", "tree": "3af10af362b1feb39f151015ca4a96c537485745", "parents": [ "a736fa1ed772e3640e1bfaab36032c5a285d6a7b" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Wed Jan 22 11:15:14 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Support icache flush without specific instructions\n\nSome CPUs don\u0027t support icache specific instructions to flush icache\nlines in broadcast way. We use cpu control registers to flush local\nicache and use IPI to notify other cores.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "a736fa1ed772e3640e1bfaab36032c5a285d6a7b", "tree": "7d0e7df49e1e90b1ddaca58781a9fc92700f3d1f", "parents": [ "c9492737b25ca32679ba3163609d938c9abfd508" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sat Jan 11 13:44:32 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky/Kconfig: Add Kconfig.platforms to support some drivers\n\nSuch as snps,dw-apb-ictl\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "c9492737b25ca32679ba3163609d938c9abfd508", "tree": "ab61b49862355cefe149d1cda68090812740d53f", "parents": [ "f8e17c17b81070f38062dce79ca7f4541851dadd" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Tue Jan 07 12:21:25 2020 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky/smp: Fixup boot failed when CONFIG_SMP\n\nIf we use a non-ipi-support interrupt controller, it will cause panic here.\nWe should let cpu up and work with CONFIG_SMP, when we use a non-ipi intc.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "f8e17c17b81070f38062dce79ca7f4541851dadd", "tree": "0de930c6cf557570a44beb5c167384ca0692d791", "parents": [ "7f4a567332f035ab16b29010fbd04a0f10183c77" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Tue Dec 17 11:12:55 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Set regs-\u003eusp to kernel sp, when the exception is from kernel\n\nIn the past, we didn\u0027t care about kernel sp when saving pt_reg. But in some\ncases, we still need pt_reg-\u003eusp to represent the kernel stack before enter\nexception.\n\nFor cmpxhg in atomic.S, we need save and restore usp for above.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "7f4a567332f035ab16b29010fbd04a0f10183c77", "tree": "f1848fc1647826b7a0a3424c3e098c2c588af9c3", "parents": [ "f136008f31e91060d6b6e6e031cb2c827448e280" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Mon Dec 30 15:53:37 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky/mm: Fixup export invalid_pte_table symbol\n\nThere is no present bit in csky pmd hardware, so we need to prepare invalid_pte_table\nfor empty pmd entry and the functions (pmd_none \u0026 pmd_present) in pgtable.h need\ninvalid_pte_talbe to get result. If a module use these functions, we need export the\nsymbol for it.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\nCc: Mo Qihui \u003cqihui.mo@verisilicon.com\u003e\nCc: Zhange Jian \u003czhang_jian5@dahuatech.com\u003e\n" }, { "commit": "f136008f31e91060d6b6e6e031cb2c827448e280", "tree": "c3b978edb1107ca151b85c75c729f6f78c543bf4", "parents": [ "f525bb2c9e7cf1e3c43ab57704c9e1c836d30b34" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Sun Dec 01 22:34:19 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Separate fixaddr_init from highmem\n\nAfter fixaddr_init is separated from highmem, we could use tcm\nwithout highmem selected. (610 (abiv1) don\u0027t support highmem,\nbut it could use tcm now.)\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "f525bb2c9e7cf1e3c43ab57704c9e1c836d30b34", "tree": "6caa93dc823ca08b3f43bae3e22d70f1c92ebff9", "parents": [ "2f78c73f78c39dabc5c44ad8dd61fd6ec65636d6" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Wed Nov 27 08:44:33 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Tightly-Coupled Memory or Sram support\n\nThe implementation are not only used by TCM but also used by sram on\nSOC bus. It follow existed linux tcm software interface, so that old\ntcm application codes could be re-used directly.\n\nSoftware interface list in asm/tcm.h:\n - Variables/Const: \t__tcmdata, __tcmconst\n - Functions:\t\t__tcmfunc, __tcmlocalfunc\n - Malloc/Free:\t\ttcm_alloc, tcm_free\n\nIn linux menuconfig:\n - Choose a TCM contain instrctions + data or separated in ITCM/DTCM.\n - Determine TCM_BASE (DTCM_BASE) in phyiscal address.\n - Determine size of TCM or ITCM(DTCM) in page counts.\n\nHere is hello tcm example from Documentation/arm/tcm.rst which could\nbe directly used:\n\n/* Uninitialized data */\nstatic u32 __tcmdata tcmvar;\n/* Initialized data */\nstatic u32 __tcmdata tcmassigned \u003d 0x2BADBABEU;\n/* Constant */\nstatic const u32 __tcmconst tcmconst \u003d 0xCAFEBABEU;\n\nstatic void __tcmlocalfunc tcm_to_tcm(void)\n{\n\tint i;\n\tfor (i \u003d 0; i \u003c 100; i++)\n\t\ttcmvar ++;\n}\n\nstatic void __tcmfunc hello_tcm(void)\n{\n\t/* Some abstract code that runs in ITCM */\n\tint i;\n\tfor (i \u003d 0; i \u003c 100; i++) {\n\t\ttcmvar ++;\n\t}\n\ttcm_to_tcm();\n}\n\nstatic void __init test_tcm(void)\n{\n\tu32 *tcmem;\n\tint i;\n\n\thello_tcm();\n\tprintk(\"Hello TCM executed from ITCM RAM\\n\");\n\n\tprintk(\"TCM variable from testrun: %u @ %p\\n\", tcmvar, \u0026tcmvar);\n\ttcmvar \u003d 0xDEADBEEFU;\n\tprintk(\"TCM variable: 0x%x @ %p\\n\", tcmvar, \u0026tcmvar);\n\n\tprintk(\"TCM assigned variable: 0x%x @ %p\\n\", tcmassigned, \u0026tcmassigned);\n\n\tprintk(\"TCM constant: 0x%x @ %p\\n\", tcmconst, \u0026tcmconst);\n\n\t/* Allocate some TCM memory from the pool */\n\ttcmem \u003d tcm_alloc(20);\n\tif (tcmem) {\n\t\tprintk(\"TCM Allocated 20 bytes of TCM @ %p\\n\", tcmem);\n\t\ttcmem[0] \u003d 0xDEADBEEFU;\n\t\ttcmem[1] \u003d 0x2BADBABEU;\n\t\ttcmem[2] \u003d 0xCAFEBABEU;\n\t\ttcmem[3] \u003d 0xDEADBEEFU;\n\t\ttcmem[4] \u003d 0x2BADBABEU;\n\t\tfor (i \u003d 0; i \u003c 5; i++)\n\t\t\tprintk(\"TCM tcmem[%d] \u003d %08x\\n\", i, tcmem[i]);\n\t\ttcm_free(tcmem, 20);\n\t}\n}\n\nTODO:\n - Separate fixup mapping from highmem\n - Support abiv1\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "2f78c73f78c39dabc5c44ad8dd61fd6ec65636d6", "tree": "b9ba43f4bb9d709d612c77b3ed4c260a204a2d38", "parents": [ "fd1d98650ac0042d475155116e65fd17eb379542" ], "author": { "name": "Mao Han", "email": "han_mao@c-sky.com", "time": "Fri Oct 11 10:56:55 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "csky: Initial stack protector support\n\nThis is a basic -fstack-protector support without per-task canary\nswitching. The protector will report something like when stack\ncorruption is detected:\n\nIt\u0027s tested with strcpy local array overflow in sys_kill and get:\nstack-protector: Kernel stack is corrupted in: sys_kill+0x23c/0x23c\n\nTODO:\n - Support task switch for different cannary\n\nSigned-off-by: Mao Han \u003chan_mao@c-sky.com\u003e\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "fd1d98650ac0042d475155116e65fd17eb379542", "tree": "cc8499889cf31860ede0a47bc23804f1c3021084", "parents": [ "11a48a5a18c63fd7621bb050228cebf13566e4d8" ], "author": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Tue Oct 08 14:37:02 2019 +0800" }, "committer": { "name": "Guo Ren", "email": "guoren@linux.alibaba.com", "time": "Fri Feb 21 15:43:24 2020 +0800" }, "message": "MAINTAINERS: csky: Add mailing list for csky\n\nAdd mailing list and it\u0027s convenient for maintain C-SKY\nsubsystem.\n\nSigned-off-by: Guo Ren \u003cguoren@linux.alibaba.com\u003e\n" }, { "commit": "df3da4ea5a0fc5d115c90d5aa6caa4dd433750a7", "tree": "f9ff3e4480025acaef8bec07cbf806a7cd6dfb2b", "parents": [ "1d0c3924a92e69bfa91163bda83c12a994b4d106" ], "author": { "name": "Suraj Jitindar Singh", "email": "surajjs@amazon.com", "time": "Tue Feb 18 19:08:50 2020 -0800" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 00:38:12 2020 -0500" }, "message": "ext4: fix potential race between s_group_info online resizing and access\n\nDuring an online resize an array of pointers to s_group_info gets replaced\nso it can get enlarged. If there is a concurrent access to the array in\next4_get_group_info() and this memory has been reused then this can lead to\nan invalid memory access.\n\nLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d206443\nLink: https://lore.kernel.org/r/20200221053458.730016-3-tytso@mit.edu\nSigned-off-by: Suraj Jitindar Singh \u003csurajjs@amazon.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nReviewed-by: Balbir Singh \u003csblbir@amazon.com\u003e\nCc: stable@kernel.org\n" }, { "commit": "1d0c3924a92e69bfa91163bda83c12a994b4d106", "tree": "c8bdc41a140f8ef4f01d6b1b7f4bd7c86b3edecc", "parents": [ "9424ef56e13a1f14c57ea161eed3ecfdc7b2770e" ], "author": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sat Feb 15 16:40:37 2020 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Fri Feb 21 00:37:09 2020 -0500" }, "message": "ext4: fix potential race between online resizing and write operations\n\nDuring an online resize an array of pointers to buffer heads gets\nreplaced so it can get enlarged. If there is a racing block\nallocation or deallocation which uses the old array, and the old array\nhas gotten reused this can lead to a GPF or some other random kernel\nmemory getting modified.\n\nLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d206443\nLink: https://lore.kernel.org/r/20200221053458.730016-2-tytso@mit.edu\nReported-by: Suraj Jitindar Singh \u003csurajjs@amazon.com\u003e\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nCc: stable@kernel.org\n" }, { "commit": "97d9a4e9619a822c5baf6a63e6f5b80fee4d4213", "tree": "ff60199fe101f2e1ad6900de96b26b16471e98e7", "parents": [ "c1368b347fe81d9271412797fea8788dd5dc16b3", "15de9cb5c9c83a23be92b8f7a1178cead1486587" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 21 12:46:54 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 21 12:46:54 2020 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2020-02-20\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\ndrm/i915 fixes for v5.6-rc3:\n- Workaround missing Display Stream Compression (DSC) state readout by\n forcing modeset when its enabled at probe\n- Fix EHL port clock voltage level requirements\n- Fix queuing retire workers on the virtual engine\n- Fix use of partially initialized waiters\n- Stop using drm_pci_alloc/drm_pci/free\n- Fix rewind of RING_TAIL by forcing a context reload\n- Fix locking on resetting ring-\u003ehead\n- Propagate our bug filing URL change to stable kernels\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Jani Nikula \u003cjani.nikula@intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/87y2sxtsrd.fsf@intel.com\n" }, { "commit": "c1368b347fe81d9271412797fea8788dd5dc16b3", "tree": "22b7065a43062f3665b2e746a8a42cc2512972bb", "parents": [ "7c42545cad101fbb9811b55a3b9b697efba6555a", "dde2bb2da01e96c17f0a44b4a3cf72a30e66e3ef" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 21 12:30:23 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 21 12:31:16 2020 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2020-02-20\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\ndrm-misc-fixes for v5.6-rc3:\n- Fix dt binding for sunxi.\n- Allow only 1 rotation argument, and allow 0 rotation in video cmdline.\n- Small compiler warning fix for panfrost.\n- Fix when using performance counters in panfrost when using per fd address space.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/f5a6370d-9898-6c72-43e4-5bb56a99b6f2@linux.intel.com\n" }, { "commit": "36a44bcdd8df092d76c11bc213e81c5817d4e302", "tree": "f008ac7473b6837a34043a49d30566802c9b9a8a", "parents": [ "3a20773beeeeadec41477a5ba872175b778ff752", "8743db4a9acfd51f805ac0c87bcaae92c42d1061" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 16:05:42 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 16:05:42 2020 -0800" }, "message": "Merge branch \u0027bnxt_en-shutdown-and-kexec-kdump-related-fixes\u0027\n\nMichael Chan says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nbnxt_en: shutdown and kexec/kdump related fixes.\n\n2 small patches to fix kexec shutdown and kdump kernel driver init issues.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8743db4a9acfd51f805ac0c87bcaae92c42d1061", "tree": "f008ac7473b6837a34043a49d30566802c9b9a8a", "parents": [ "5567ae4a8d569d996d0d88d0eceb76205e4c7ce5" ], "author": { "name": "Vasundhara Volam", "email": "vasundhara-v.volam@broadcom.com", "time": "Thu Feb 20 17:26:35 2020 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 16:05:42 2020 -0800" }, "message": "bnxt_en: Issue PCIe FLR in kdump kernel to cleanup pending DMAs.\n\nIf crashed kernel does not shutdown the NIC properly, PCIe FLR\nis required in the kdump kernel in order to initialize all the\nfunctions properly.\n\nFixes: d629522e1d66 (\"bnxt_en: Reduce memory usage when running in kdump kernel.\")\nSigned-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nSigned-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5567ae4a8d569d996d0d88d0eceb76205e4c7ce5", "tree": "782cc6e64aac4e9216cf673489c97f39960c157c", "parents": [ "3a20773beeeeadec41477a5ba872175b778ff752" ], "author": { "name": "Vasundhara Volam", "email": "vasundhara-v.volam@broadcom.com", "time": "Thu Feb 20 17:26:34 2020 -0500" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 16:05:42 2020 -0800" }, "message": "bnxt_en: Improve device shutdown method.\n\nEspecially when bnxt_shutdown() is called during kexec, we need to\ndisable MSIX and disable Bus Master to completely quiesce the device.\nMake these 2 calls unconditionally in the shutdown method.\n\nFixes: c20dc142dd7b (\"bnxt_en: Disable bus master during PCI shutdown and driver unload.\")\nSigned-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nSigned-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3a20773beeeeadec41477a5ba872175b778ff752", "tree": "e6bf9989b00065892e06d785ceb4e198cdcbd914", "parents": [ "971617c3b761c876d686a2188220a33898c90e99" ], "author": { "name": "Nikolay Aleksandrov", "email": "nikolay@cumulusnetworks.com", "time": "Thu Feb 20 16:42:13 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 16:02:08 2020 -0800" }, "message": "net: netlink: cap max groups which will be considered in netlink_bind()\n\nSince nl_groups is a u32 we can\u0027t bind more groups via -\u003ebind\n(netlink_bind) call, but netlink has supported more groups via\nsetsockopt() for a long time and thus nlk-\u003engroups could be over 32.\nRecently I added support for per-vlan notifications and increased the\ngroups to 33 for NETLINK_ROUTE which exposed an old bug in the\nnetlink_bind() code causing out-of-bounds access on archs where unsigned\nlong is 32 bits via test_bit() on a local variable. Fix this by capping the\nmaximum groups in netlink_bind() to BITS_PER_TYPE(u32), effectively\ncapping them at 32 which is the minimum of allocated groups and the\nmaximum groups which can be bound via netlink_bind().\n\nCC: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCC: Richard Guy Briggs \u003crgb@redhat.com\u003e\nFixes: 4f520900522f (\"netlink: have netlink per-protocol bind function return an error code.\")\nReported-by: Erhard F. \u003cerhard_f@mailbox.org\u003e\nSigned-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "971617c3b761c876d686a2188220a33898c90e99", "tree": "29a9a168f861b8570e68ede14725eae4f331a143", "parents": [ "68b759a75d6257759d1e37ff13f2d0659baf1112" ], "author": { "name": "Tim Harvey", "email": "tharvey@gateworks.com", "time": "Wed Feb 19 15:19:36 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 15:49:20 2020 -0800" }, "message": "net: thunderx: workaround BGX TX Underflow issue\n\nWhile it is not yet understood why a TX underflow can easily occur\nfor SGMII interfaces resulting in a TX wedge. It has been found that\ndisabling/re-enabling the LMAC resolves the issue.\n\nSigned-off-by: Tim Harvey \u003ctharvey@gateworks.com\u003e\nReviewed-by: Robert Jones \u003crjones@gateworks.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "68b759a75d6257759d1e37ff13f2d0659baf1112", "tree": "39c0dcf49109280487aedc96532ac8a200c17b13", "parents": [ "98bda63e20daab95bdc084ce00459a4f622a0505" ], "author": { "name": "Shannon Nelson", "email": "snelson@pensando.io", "time": "Wed Feb 19 14:59:42 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 15:48:04 2020 -0800" }, "message": "ionic: fix fw_status read\n\nThe fw_status field is only 8 bits, so fix the read. Also,\nwe only want to look at the one status bit, to allow for future\nuse of the other bits, and watch for a bad PCI read.\n\nFixes: 97ca486592c0 (\"ionic: add heartbeat check\")\nSigned-off-by: Shannon Nelson \u003csnelson@pensando.io\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ebe7acadf5a9c4b67475e766e0e80ae6a2a79c61", "tree": "b200c91bc4e77bb62163dde440791298b4137428", "parents": [ "ca7e1fd1026c5af6a533b4b5447e1d2f153e28f2", "5780b9abd530982c2bb1018e2c52c05ab3c30b45" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 20 15:15:16 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 20 15:15:16 2020 -0800" }, "message": "Merge branch \u0027next-integrity\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/zohar/linux-integrity\n\nPull IMA fixes from Mimi Zohar:\n \"Two bug fixes and an associated change for each.\n\n The one that adds SM3 to the IMA list of supported hash algorithms is\n a simple change, but could be considered a new feature\"\n\n* \u0027next-integrity\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/zohar/linux-integrity:\n ima: add sm3 algorithm to hash algorithm configuration list\n crypto: rename sm3-256 to sm3 in hash_algo_name\n efi: Only print errors about failing to get certs if EFI vars are found\n x86/ima: use correct identifier for SetupMode variable\n" }, { "commit": "98bda63e20daab95bdc084ce00459a4f622a0505", "tree": "a06c9375c9287c3f8e9ebef8f7e88ea60e8671b4", "parents": [ "ac2fcfa9fd26db67d7000677c05629c34cc94564" ], "author": { "name": "Roman Kiryanov", "email": "rkir@google.com", "time": "Wed Feb 19 13:40:06 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 15:02:02 2020 -0800" }, "message": "net: disable BRIDGE_NETFILTER by default\n\nThe description says \u0027If unsure, say N.\u0027 but\nthe module is built as M by default (once\nthe dependencies are satisfied).\n\nWhen the module is selected (Y or M), it enables\nNETFILTER_FAMILY_BRIDGE and SKB_EXTENSIONS\nwhich alter kernel internal structures.\n\nWe (Android Studio Emulator) currently do not\nuse this module and think this it is more consistent\nto have it disabled by default as opposite to\ndisabling it explicitly to prevent enabling\nNETFILTER_FAMILY_BRIDGE and SKB_EXTENSIONS.\n\nSigned-off-by: Roman Kiryanov \u003crkir@google.com\u003e\nAcked-by: Florian Westphal \u003cfw@strlen.de\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ac2fcfa9fd26db67d7000677c05629c34cc94564", "tree": "8a5c0f13e8ac268fbcc653cee874e91d95da2b50", "parents": [ "0d5b8d7055652fd6b079671d00fa78fcd0e2cd7c" ], "author": { "name": "Alexandre Belloni", "email": "alexandre.belloni@bootlin.com", "time": "Wed Feb 19 15:15:51 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 15:00:31 2020 -0800" }, "message": "net: macb: Properly handle phylink on at91rm9200\n\nat91ether_init was handling the phy mode and speed but since the switch to\nphylink, the NCFGR register got overwritten by macb_mac_config(). The issue\nis that the RM9200_RMII bit and the MACB_CLK_DIV32 field are cleared\nbut never restored as they conflict with the PAE, GBE and PCSSEL bits.\n\nAdd new capability to differentiate between EMAC and the other versions of\nthe IP and use it to set and avoid clearing the relevant bits.\n\nAlso, this fixes a NULL pointer dereference in macb_mac_link_up as the EMAC\ndoesn\u0027t use any rings/bufffers/queues.\n\nFixes: 7897b071ac3b (\"net: macb: convert to phylink\")\nSigned-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8645e56a4ad6dcbf504872db7f14a2f67db88ef2", "tree": "b94527291919659a3aabfbe7082b040143b85579", "parents": [ "9038ec99ceb94fb8d93ade5e236b2928f0792c7c" ], "author": { "name": "Thomas Gleixner", "email": "tglx@linutronix.de", "time": "Wed Feb 19 18:30:26 2020 +0100" }, "committer": { "name": "Boris Ostrovsky", "email": "boris.ostrovsky@oracle.com", "time": "Thu Feb 20 16:40:38 2020 -0600" }, "message": "xen: Enable interrupts when calling _cond_resched()\n\nxen_maybe_preempt_hcall() is called from the exception entry point\nxen_do_hypervisor_callback with interrupts disabled.\n\n_cond_resched() evades the might_sleep() check in cond_resched() which\nwould have caught that and schedule_debug() unfortunately lacks a check\nfor irqs_disabled().\n\nEnable interrupts around the call and use cond_resched() to catch future\nissues.\n\nFixes: fdfd811ddde3 (\"x86/xen: allow privcmd hypercalls to be preempted\")\nSigned-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nLink: https://lore.kernel.org/r/878skypjrh.fsf@nanos.tec.linutronix.de\nReviewed-by: Juergen Gross \u003cjgross@suse.com\u003e\nSigned-off-by: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\n" }, { "commit": "0d5b8d7055652fd6b079671d00fa78fcd0e2cd7c", "tree": "8f3a001e656ec08d2279a6557ebfa08760250fcf", "parents": [ "16a556eeb7ed2dc3709fe2c5be76accdfa4901ab", "54a61fbc020fd2e305680871c453abcf7fc0339b" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 10:30:47 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 10:30:47 2020 -0800" }, "message": "Merge branch \u0027s390-fixes\u0027\n\nJulian Wiedmann says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\ns390/qeth: fixes 2020-02-20\n\nplease apply the following patch series for qeth to netdev\u0027s net tree.\n\nThis corrects three minor issues:\n1) return a more fitting errno when VNICC cmds are not supported,\n2) remove a bogus WARN in the NAPI code, and\n3) be _very_ pedantic about the RX copybreak.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "54a61fbc020fd2e305680871c453abcf7fc0339b", "tree": "8f3a001e656ec08d2279a6557ebfa08760250fcf", "parents": [ "420579dba126c6111b5a3dea062f21a7e4e647c6" ], "author": { "name": "Julian Wiedmann", "email": "jwi@linux.ibm.com", "time": "Thu Feb 20 15:54:56 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 20 10:30:47 2020 -0800" }, "message": "s390/qeth: fix off-by-one in RX copybreak check\n\nThe RX copybreak is intended as the _max_ value where the frame\u0027s data\nshould be copied. So for frame_len \u003d\u003d copybreak, don\u0027t build an SG skb.\n\nFixes: 4a71df50047f (\"qeth: new qeth device driver\")\nSigned-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" } ], "next": "420579dba126c6111b5a3dea062f21a7e4e647c6" }