)]}' { "log": [ { "commit": "e99332e7b4cda6e60f5b5916cf9943a79dbef902", "tree": "c6deee5dadd157c64c0775766192a0fe8ea818e1", "parents": [ "2e28f3b13a41b8a7d36a73ddf4bb41972a9c1dd9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 17:50:03 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 17:50:03 2020 -0700" }, "message": "gcc-10: mark more functions __init to avoid section mismatch warnings\n\nIt seems that for whatever reason, gcc-10 ends up not inlining a couple\nof functions that used to be inlined before. Even if they only have one\nsingle callsite - it looks like gcc may have decided that the code was\nunlikely, and not worth inlining.\n\nThe code generation difference is harmless, but caused a few new section\nmismatch errors, since the (now no longer inlined) function wasn\u0027t in\nthe __init section, but called other init functions:\n\n Section mismatch in reference from the function kexec_free_initrd() to the function .init.text:free_initrd_mem()\n Section mismatch in reference from the function tpm2_calc_event_log_size() to the function .init.text:early_memremap()\n Section mismatch in reference from the function tpm2_calc_event_log_size() to the function .init.text:early_memunmap()\n\nSo add the appropriate __init annotation to make modpost not complain.\nIn both cases there were trivially just a single callsite from another\n__init function.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "2e28f3b13a41b8a7d36a73ddf4bb41972a9c1dd9", "tree": "2d53b1fc9d0c4707ff2128ae846f262c4f65f1ff", "parents": [ "1a263ae60b04de959d9ce9caea4889385eefcc7b", "73cb8e2a5863ccc5215660f5123db621bd57dff7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 16:24:16 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 16:24:16 2020 -0700" }, "message": "Merge tag \u0027riscv-for-linus-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V fixes from Palmer Dabbelt:\n \"A smattering of fixes and cleanups:\n\n - Dead code removal.\n\n - Exporting riscv_cpuid_to_hartid_mask for modules.\n\n - Per-CPU tracking of ISA features.\n\n - Setting max_pfn correctly when probing memory.\n\n - Adding a note to the VDSO so glibc can check the kernel\u0027s version\n without a uname().\n\n - A fix to force the bootloader to initialize the boot spin tables,\n which still get used as a fallback when SBI-0.1 is enabled\"\n\n* tag \u0027riscv-for-linus-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n RISC-V: Remove unused code from STRICT_KERNEL_RWX\n riscv: force __cpu_up_ variables to put in data section\n riscv: add Linux note to vdso\n riscv: set max_pfn to the PFN of the last page\n RISC-V: Remove N-extension related defines\n RISC-V: Add bitmap reprensenting ISA features common across CPUs\n RISC-V: Export riscv_cpuid_to_hartid_mask() API\n" }, { "commit": "1a263ae60b04de959d9ce9caea4889385eefcc7b", "tree": "0c1f705db3fa09f7584873a3bcbf85dddac9249a", "parents": [ "adc71920969870dfa54e8f40dac8616284832d02" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 15:58:04 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 15:58:04 2020 -0700" }, "message": "gcc-10: avoid shadowing standard library \u0027free()\u0027 in crypto\n\ngcc-10 has started warning about conflicting types for a few new\nbuilt-in functions, particularly \u0027free()\u0027.\n\nThis results in warnings like:\n\n crypto/xts.c:325:13: warning: conflicting types for built-in function ‘free’; expected ‘void(void *)’ [-Wbuiltin-declaration-mismatch]\n\nbecause the crypto layer had its local freeing functions called\n\u0027free()\u0027.\n\nGcc-10 is in the wrong here, since that function is marked \u0027static\u0027, and\nthus there is no chance of confusion with any standard library function\nnamespace.\n\nBut the simplest thing to do is to just use a different name here, and\navoid this gcc mis-feature.\n\n[ Side note: gcc knowing about \u0027free()\u0027 is in itself not the\n mis-feature: the semantics of \u0027free()\u0027 are special enough that a\n compiler can validly do special things when seeing it.\n\n So the mis-feature here is that gcc thinks that \u0027free()\u0027 is some\n restricted name, and you can\u0027t shadow it as a local static function.\n\n Making the special \u0027free()\u0027 semantics be a function attribute rather\n than tied to the name would be the much better model ]\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "adc71920969870dfa54e8f40dac8616284832d02", "tree": "bb96fe43b00c4129f7161bb8a115197d376477a6", "parents": [ "5a76021c2eff7fcf2f0918a08fd8a37ce7922921" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 15:45:21 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 15:45:21 2020 -0700" }, "message": "gcc-10: disable \u0027restrict\u0027 warning for now\n\ngcc-10 now warns about passing aliasing pointers to functions that take\nrestricted pointers.\n\nThat\u0027s actually a great warning, and if we ever start using \u0027restrict\u0027\nin the kernel, it might be quite useful. But right now we don\u0027t, and it\nturns out that the only thing this warns about is an idiom where we have\ndeclared a few functions to be \"printf-like\" (which seems to make gcc\npick up the restricted pointer thing), and then we print to the same\nbuffer that we also use as an input.\n\nAnd people do that as an odd concatenation pattern, with code like this:\n\n #define sysfs_show_gen_prop(buffer, fmt, ...) \\\n snprintf(buffer, PAGE_SIZE, \"%s\"fmt, buffer, __VA_ARGS__)\n\nwhere we have \u0027buffer\u0027 as both the destination of the final result, and\nas the initial argument.\n\nYes, it\u0027s a bit questionable. And outside of the kernel, people do have\nstandard declarations like\n\n int snprintf( char *restrict buffer, size_t bufsz,\n const char *restrict format, ... );\n\nwhere that output buffer is marked as a restrict pointer that cannot\nalias with any other arguments.\n\nBut in the context of the kernel, that \u0027use snprintf() to concatenate to\nthe end result\u0027 does work, and the pattern shows up in multiple places.\nAnd we have not marked our own version of snprintf() as taking restrict\npointers, so the warning is incorrect for now, and gcc picks it up on\nits own.\n\nIf we do start using \u0027restrict\u0027 in the kernel (and it might be a good\nidea if people find places where it matters), we\u0027ll need to figure out\nhow to avoid this issue for snprintf and friends. But in the meantime,\nthis warning is not useful.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5a76021c2eff7fcf2f0918a08fd8a37ce7922921", "tree": "96fb6bc668e9f84a946676df3afb10f4093000c4", "parents": [ "44720996e2d79e47d508b0abe99b931a726a3197" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 15:40:52 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 15:40:52 2020 -0700" }, "message": "gcc-10: disable \u0027stringop-overflow\u0027 warning for now\n\nThis is the final array bounds warning removal for gcc-10 for now.\n\nAgain, the warning is good, and we should re-enable all these warnings\nwhen we have converted all the legacy array declaration cases to\nflexible arrays. But in the meantime, it\u0027s just noise.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "44720996e2d79e47d508b0abe99b931a726a3197", "tree": "731517d0e160daa835abc1735d0e9fc31cd2b1fe", "parents": [ "5c45de21a2223fe46cf9488c99a7fbcf01527670" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 14:52:44 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 14:52:44 2020 -0700" }, "message": "gcc-10: disable \u0027array-bounds\u0027 warning for now\n\nThis is another fine warning, related to the \u0027zero-length-bounds\u0027 one,\nbut hitting the same historical code in the kernel.\n\nBecause C didn\u0027t historically support flexible array members, we have\ncode that instead uses a one-sized array, the same way we have cases of\nzero-sized arrays.\n\nThe one-sized arrays come from either not wanting to use the gcc\nzero-sized array extension, or from a slight convenience-feature, where\nparticularly for strings, the size of the structure now includes the\nallocation for the final NUL character.\n\nSo with a \"char name[1];\" at the end of a structure, you can do things\nlike\n\n v \u003d my_malloc(sizeof(struct vendor) + strlen(name));\n\nand avoid the \"+1\" for the terminator.\n\nYes, the modern way to do that is with a flexible array, and using\n\u0027offsetof()\u0027 instead of \u0027sizeof()\u0027, and adding the \"+1\" by hand. That\nalso technically gets the size \"more correct\" in that it avoids any\nalignment (and thus padding) issues, but this is another long-term\ncleanup thing that will not happen for 5.7.\n\nSo disable the warning for now, even though it\u0027s potentially quite\nuseful. Having a slew of warnings that then hide more urgent new issues\nis not an improvement.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5c45de21a2223fe46cf9488c99a7fbcf01527670", "tree": "bd55b54971a70c8974d87d35399fc922997ccd90", "parents": [ "78a5255ffb6a1af189a83e493d916ba1c54d8c75" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 14:30:29 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 14:30:29 2020 -0700" }, "message": "gcc-10: disable \u0027zero-length-bounds\u0027 warning for now\n\nThis is a fine warning, but we still have a number of zero-length arrays\nin the kernel that come from the traditional gcc extension. Yes, they\nare getting converted to flexible arrays, but in the meantime the gcc-10\nwarning about zero-length bounds is very verbose, and is hiding other\nissues.\n\nI missed one actual build failure because it was hidden among hundreds\nof lines of warning. Thankfully I caught it on the second go before\npushing things out, but it convinced me that I really need to disable\nthe new warnings for now.\n\nWe\u0027ll hopefully be all done with our conversion to flexible arrays in\nthe not too distant future, and we can then re-enable this warning.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "78a5255ffb6a1af189a83e493d916ba1c54d8c75", "tree": "ff0c8ef475d1a5921fa761a96f01582a444fef0d", "parents": [ "1d3962ae3b3d3a945f7fd5c651cf170a27521a35" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 13:57:10 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 13:57:10 2020 -0700" }, "message": "Stop the ad-hoc games with -Wno-maybe-initialized\n\nWe have some rather random rules about when we accept the\n\"maybe-initialized\" warnings, and when we don\u0027t.\n\nFor example, we consider it unreliable for gcc versions \u003c 4.9, but also\nif -O3 is enabled, or if optimizing for size. And then various kernel\nconfig options disabled it, because they know that they trigger that\nwarning by confusing gcc sufficiently (ie PROFILE_ALL_BRANCHES).\n\nAnd now gcc-10 seems to be introducing a lot of those warnings too, so\nit falls under the same heading as 4.9 did.\n\nAt the same time, we have a very straightforward way to _enable_ that\nwarning when wanted: use \"W\u003d2\" to enable more warnings.\n\nSo stop playing these ad-hoc games, and just disable that warning by\ndefault, with the known and straight-forward \"if you want to work on the\nextra compiler warnings, use W\u003d123\".\n\nWould it be great to have code that is always so obvious that it never\nconfuses the compiler whether a variable is used initialized or not?\nYes, it would. In a perfect world, the compilers would be smarter, and\nour source code would be simpler.\n\nThat\u0027s currently not the world we live in, though.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "1d3962ae3b3d3a945f7fd5c651cf170a27521a35", "tree": "c673e2dfc42377aa9ee7c4cb4e28538f060e28e7", "parents": [ "d5eeab8d7e269e8cfc53b915bccd7bd30485bcbf", "63ff822358b276137059520cf16e587e8073e80f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 12:02:09 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 09 12:02:09 2020 -0700" }, "message": "Merge tag \u0027io_uring-5.7-2020-05-08\u0027 of git://git.kernel.dk/linux-block\n\nPull io_uring fixes from Jens Axboe:\n\n - Fix finish_wait() balancing in file cancelation (Xiaoguang)\n\n - Ensure early cleanup of resources in ring map failure (Xiaoguang)\n\n - Ensure IORING_OP_SLICE does the right file mode checks (Pavel)\n\n - Remove file opening from openat/openat2/statx, it\u0027s not needed and\n messes with O_PATH\n\n* tag \u0027io_uring-5.7-2020-05-08\u0027 of git://git.kernel.dk/linux-block:\n io_uring: don\u0027t use \u0027fd\u0027 for openat/openat2/statx\n splice: move f_mode checks to do_{splice,tee}()\n io_uring: handle -EFAULT properly in io_uring_setup()\n io_uring: fix mismatched finish_wait() calls in io_uring_cancel_files()\n" }, { "commit": "d5eeab8d7e269e8cfc53b915bccd7bd30485bcbf", "tree": "082c96f2037a69ef6dea3a4d2d5046d41a2cf4b6", "parents": [ "eb24fdd8e6f5c6bb95129748a1801c6476492aba", "b36522150e5b85045f868768d46fbaaa034174b2" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 10:36:56 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 10:36:56 2020 -0700" }, "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 minor fixes, all in drivers (qla2xxx, ibmvfc, ibmvscsi)\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: ibmvscsi: Fix WARN_ON during event pool release\n scsi: ibmvfc: Don\u0027t send implicit logouts prior to NPIV login\n scsi: qla2xxx: Delete all sessions before unregister local nvme port\n scsi: qla2xxx: Fix hang when issuing nvme disconnect-all in NPIV\n" }, { "commit": "eb24fdd8e6f5c6bb95129748a1801c6476492aba", "tree": "da951f5852d918181639567d92bc7d987739b1c3", "parents": [ "4334f30ebf395b204c6cbeabf371a5a998d6ba7c", "12ae44a40a1be891bdc6463f8c7072b4ede746ef" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 10:27:00 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 10:27:00 2020 -0700" }, "message": "Merge tag \u0027ceph-for-5.7-rc5\u0027 of git://github.com/ceph/ceph-client\n\nPull ceph fixes from Ilya Dryomov:\n \"Fixes for an endianness handling bug that prevented mounts on\n big-endian arches, a spammy log message and a couple error paths.\n\n Also included a MAINTAINERS update\"\n\n* tag \u0027ceph-for-5.7-rc5\u0027 of git://github.com/ceph/ceph-client:\n ceph: demote quotarealm lookup warning to a debug message\n MAINTAINERS: remove myself as ceph co-maintainer\n ceph: fix double unlock in handle_cap_export()\n ceph: fix special error code in ceph_try_get_caps()\n ceph: fix endianness bug when handling MDS session feature bits\n" }, { "commit": "12ae44a40a1be891bdc6463f8c7072b4ede746ef", "tree": "39a2939dccdc27d9ed099ce77a89e0fff7c29ebd", "parents": [ "3a5ccecd9af71220e8b303f439ebccbc48385305" ], "author": { "name": "Luis Henriques", "email": "lhenriques@suse.com", "time": "Tue May 05 13:59:02 2020 +0100" }, "committer": { "name": "Ilya Dryomov", "email": "idryomov@gmail.com", "time": "Fri May 08 18:44:40 2020 +0200" }, "message": "ceph: demote quotarealm lookup warning to a debug message\n\nA misconfigured cephx can easily result in having the kernel client\nflooding the logs with:\n\n ceph: Can\u0027t lookup inode 1 (err: -13)\n\nChange this message to debug level.\n\nCc: stable@vger.kernel.org\nURL: https://tracker.ceph.com/issues/44546\nSigned-off-by: Luis Henriques \u003clhenriques@suse.com\u003e\nReviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\nSigned-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n" }, { "commit": "4334f30ebf395b204c6cbeabf371a5a998d6ba7c", "tree": "82cd890c8ca8b2a75197edd9c42b9cd6cfb61cf5", "parents": [ "c61529f6f59c9e136432cde96f9a053a51ad2a1e", "f0e1d3ac2d7c16a5d2c9d67f5a61133db7681af8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 09:11:53 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 09:11:53 2020 -0700" }, "message": "Merge tag \u0027char-misc-5.7-rc5\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 driver fixes for 5.7-rc5 that resolve a number of\n minor reported issues:\n\n - mhi bus driver fixes found as people actually use the code\n\n - phy driver fixes and compat string additions\n\n - most driver fix due to link order changing when the core moved out\n of staging\n\n - mei driver fix\n\n - interconnect build warning fix\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027char-misc-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n bus: mhi: core: Fix channel device name conflict\n bus: mhi: core: Fix typo in comment\n bus: mhi: core: Offload register accesses to the controller\n bus: mhi: core: Remove link_status() callback\n bus: mhi: core: Make sure to powerdown if mhi_sync_power_up fails\n bus: mhi: Fix parsing of mhi_flags\n mei: me: disable mei interface on LBG servers.\n phy: qualcomm: usb-hs-28nm: Prepare clocks in init\n MAINTAINERS: Add Vinod Koul as Generic PHY co-maintainer\n interconnect: qcom: Move the static keyword to the front of declaration\n most: core: use function subsys_initcall()\n bus: mhi: core: Fix a NULL vs IS_ERR check in mhi_create_devices()\n phy: qcom-qusb2: Re add \"qcom,sdm845-qusb2-phy\" compat string\n phy: tegra: Select USB_COMMON for usb_get_maximum_speed()\n" }, { "commit": "c61529f6f59c9e136432cde96f9a053a51ad2a1e", "tree": "e72cbda75f28dfe4a7a67533ef11a120c0edee02", "parents": [ "e7a1c733fecd6bf71f87f74b36ab51bb3eafd8b3", "2a15483b401c0b07e44b43b95414e36f32c02f32" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 09:06:34 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 09:06:34 2020 -0700" }, "message": "Merge tag \u0027driver-core-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core\n\nPull driver core fixes from Greg KH:\n \"Here are a number of small driver core fixes for 5.7-rc5 to resolve a\n bunch of reported issues with the current tree.\n\n Biggest here are the reverts and patches from John Stultz to resolve a\n bunch of deferred probe regressions we have been seeing in 5.7-rc\n right now.\n\n Along with those are some other smaller fixes:\n\n - coredump crash fix\n\n - devlink fix for when permissive mode was enabled\n\n - amba and platform device dma_parms fixes\n\n - component error silenced for when deferred probe happens\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027driver-core-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:\n regulator: Revert \"Use driver_deferred_probe_timeout for regulator_init_complete_work\"\n driver core: Ensure wait_for_device_probe() waits until the deferred_probe_timeout fires\n driver core: Use dev_warn() instead of dev_WARN() for deferred_probe_timeout warnings\n driver core: Revert default driver_deferred_probe_timeout value to 0\n component: Silence bind error on -EPROBE_DEFER\n driver core: Fix handling of fw_devlink\u003dpermissive\n coredump: fix crash when umh is disabled\n amba: Initialize dma_parms for amba devices\n driver core: platform: Initialize dma_parms for platform devices\n" }, { "commit": "e7a1c733fecd6bf71f87f74b36ab51bb3eafd8b3", "tree": "61dfcb05783742ee5ddc13d4bdb6f73d321cd001", "parents": [ "cbd0e4821373ab10833c38edc3b858e3fa1a2976", "769acc3656d93aaacada814939743361d284fd87" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 09:03:49 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 09:03:49 2020 -0700" }, "message": "Merge tag \u0027staging-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging driver fixes from Greg KH:\n \"Here are three small driver fixes for 5.7-rc5.\n\n Two of these are documentation fixes:\n\n - MAINTAINERS update due to removed driver\n\n - removing Wolfram from the ks7010 driver TODO file\n\n The other patch is a real fix:\n\n - fix gasket driver to proper check the return value of a call\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027staging-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:\n staging: gasket: Check the return value of gasket_get_bar_index()\n staging: ks7010: remove me from CC list\n MAINTAINERS: remove entry after hp100 driver removal\n" }, { "commit": "cbd0e4821373ab10833c38edc3b858e3fa1a2976", "tree": "62f3cc0069c0fe572a3c4eec44f00816c3e6e59a", "parents": [ "0a0b96b2e27d75c74e254abf8efe17628aa427ad", "2ae11c46d5fdc46cb396e35911c713d271056d35" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:56:16 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:56:16 2020 -0700" }, "message": "Merge tag \u0027tty-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull tty/serial fixes from Greg KH:\n \"Here are three small TTY/Serial/VT fixes for 5.7-rc5:\n\n - revert for the bcm63xx driver \"fix\" that was incorrect\n\n - vt unicode console bugfix\n\n - xilinx_uartps console driver fix\n\n All of these have been in linux next with no reported issues\"\n\n* tag \u0027tty-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n tty: xilinx_uartps: Fix missing id assignment to the console\n vt: fix unicode console freeing with a common interface\n Revert \"tty: serial: bcm63xx: fix missing clk_put() in bcm63xx_uart\"\n" }, { "commit": "0a0b96b2e27d75c74e254abf8efe17628aa427ad", "tree": "f6811af99b67123c936f80416fcf0809596e92a0", "parents": [ "775a8e0316bd5443a0f3faf771b1bdfb8f8e8342", "91edf63d5022bd0464788ffb4acc3d5febbaf81d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:54:00 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:54:00 2020 -0700" }, "message": "Merge tag \u0027usb-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB fixes from Greg KH:\n \"Here are some small USB fixes for 5.7-rc5 to resolve some reported\n issues:\n\n - syzbot found problems fixed\n\n - usbfs dma mapping fix\n\n - typec bugfixs\n\n - chipidea bugfix\n\n - usb4/thunderbolt fix\n\n - new device ids/quirks\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027usb-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:\n usb: chipidea: msm: Ensure proper controller reset using role switch API\n usb: typec: mux: intel: Handle alt mode HPD_HIGH\n usb: usbfs: correct kernel-\u003euser page attribute mismatch\n usb: typec: intel_pmc_mux: Fix the property names\n USB: core: Fix misleading driver bug report\n USB: serial: qcserial: Add DW5816e support\n USB: uas: add quirk for LaCie 2Big Quadra\n thunderbolt: Check return value of tb_sw_read() in usb4_switch_op()\n USB: serial: garmin_gps: add sanity checking for data length\n" }, { "commit": "775a8e0316bd5443a0f3faf771b1bdfb8f8e8342", "tree": "5559a06b9802a57a6c912fe459f9466049307415", "parents": [ "af38553c661207f96464e15f3506bf788daee474", "a9fe6f18cde03c20facbf75dc910a372c1c1025b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:49:34 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:49:34 2020 -0700" }, "message": "Merge tag \u0027drm-fixes-2020-05-08\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Another pretty normal week. I didn\u0027t get any i915 fixes yet, so next\n week I\u0027d expect double the usual i915, but otherwise a bunch of amdgpu\n and some scattered other fixes.\n\n hdcp:\n - fix HDCP regression\n\n amdgpu:\n - Runtime PM fixes\n - DC fix for PPC\n - Misc DC fixes\n\n virtio:\n - fix context ordering issue\n\n sun4i:\n - old gcc warning fix\n\n ingenic-drm:\n - missing module support\"\n\n* tag \u0027drm-fixes-2020-05-08\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/amd/display: Prevent dpcd reads with passive dongles\n drm/amd/display: fix counter in wait_for_no_pipes_pending\n drm/amd/display: Update DCN2.1 DV Code Revision\n drm: Fix HDCP failures when SRM fw is missing\n sun6i: dsi: fix gcc-4.8\n drm: ingenic-drm: add MODULE_DEVICE_TABLE\n drm/virtio: create context before RESOURCE_CREATE_2D in 3D mode\n drm/amd/display: work around fp code being emitted outside of DC_FP_START/END\n drm/amdgpu/dc: Use WARN_ON_ONCE for ASSERT\n drm/amdgpu: drop redundant cg/pg ungate on runpm enter\n drm/amdgpu: move kfd suspend after ip_suspend_phase1\n" }, { "commit": "af38553c661207f96464e15f3506bf788daee474", "tree": "5b58816c3cd8126736065c831dc3bc5a61cbea12", "parents": [ "79dede78c0573618e3137d3d8cbf78c84e25fabd", "14f69140ff9c92a0928547ceefb153a842e8492c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:41:09 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri May 08 08:41:09 2020 -0700" }, "message": "Merge branch \u0027akpm\u0027 (patches from Andrew)\n\nMerge misc fixes from Andrew Morton:\n \"14 fixes and one selftest to verify the ipc fixes herein\"\n\n* emailed patches from Andrew Morton \u003cakpm@linux-foundation.org\u003e:\n mm: limit boost_watermark on small zones\n ubsan: disable UBSAN_ALIGNMENT under COMPILE_TEST\n mm/vmscan: remove unnecessary argument description of isolate_lru_pages()\n epoll: atomically remove wait entry on wake up\n kselftests: introduce new epoll60 testcase for catching lost wakeups\n percpu: make pcpu_alloc() aware of current gfp context\n mm/slub: fix incorrect interpretation of s-\u003eoffset\n scripts/gdb: repair rb_first() and rb_last()\n eventpoll: fix missing wakeup for ovflist in ep_poll_callback\n arch/x86/kvm/svm/sev.c: change flag passed to GUP fast in sev_pin_memory()\n scripts/decodecode: fix trapping instruction formatting\n kernel/kcov.c: fix typos in kcov_remote_start documentation\n mm/page_alloc: fix watchdog soft lockups during set_zone_contiguous()\n mm, memcg: fix error return value of mem_cgroup_css_alloc()\n ipc/mqueue.c: change __do_notify() to bypass check_kill_permission()\n" }, { "commit": "a9fe6f18cde03c20facbf75dc910a372c1c1025b", "tree": "6c1477a1448c33d4c6043b11e32d5db92086166b", "parents": [ "c61b0b97ef10ab7f4022707c516b5c32df2bd738", "5fe89a6acd668cbd1817fcdef5caa9fee568c2e8" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 08 15:02:49 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 08 15:04:25 2020 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2020-05-07\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\nA few minor fixes for an ordering issue in virtio, an (old) gcc warning\nin sun4i, a probe issue in ingenic-drm and a regression in the HDCP\nsupport.\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Maxime Ripard \u003cmaxime@cerno.tech\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20200507160130.id64niqgf5wsha4u@gilmour.lan\n" }, { "commit": "c61b0b97ef10ab7f4022707c516b5c32df2bd738", "tree": "1fade149964235e6438de0283aa050c80a74ac35", "parents": [ "0e698dfa282211e414076f9dc7e83c1c288314fd", "e6142dd511425cb827b5db869f489eb81f5f994d" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 08 13:31:38 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri May 08 13:31:39 2020 +1000" }, "message": "Merge tag \u0027amd-drm-fixes-5.7-2020-05-06\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\namd-drm-fixes-5.7-2020-05-06:\n\namdgpu:\n- Runtime PM fixes\n- DC fix for PPC\n- Misc DC fixes\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexdeucher@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20200506212257.3893-1-alexander.deucher@amd.com\n" }, { "commit": "79dede78c0573618e3137d3d8cbf78c84e25fabd", "tree": "35396e8f3a4beb6db4f7887a8ad302148a768dcb", "parents": [ "192ffb7515839b1cc8457e0a8c1e09783de019d3", "54261af473be4c5481f6196064445d2945f2bdab" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:43:13 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:43:13 2020 -0700" }, "message": "Merge branch \u0027for-v5.7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security\n\nPull security subsystem fix from James Morris:\n \"Fix the default value of fs_context_parse_param hook\"\n\n* \u0027for-v5.7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:\n security: Fix the default value of fs_context_parse_param hook\n" }, { "commit": "14f69140ff9c92a0928547ceefb153a842e8492c", "tree": "9d72ebf46f71a8d8a9d6b02b4ad37e243cda55a8", "parents": [ "8d58f222e85f01da0c0e1fc1e77986c86de889e2" ], "author": { "name": "Henry Willard", "email": "henry.willard@oracle.com", "time": "Thu May 07 18:36:27 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:21 2020 -0700" }, "message": "mm: limit boost_watermark on small zones\n\nCommit 1c30844d2dfe (\"mm: reclaim small amounts of memory when an\nexternal fragmentation event occurs\") adds a boost_watermark() function\nwhich increases the min watermark in a zone by at least\npageblock_nr_pages or the number of pages in a page block.\n\nOn Arm64, with 64K pages and 512M huge pages, this is 8192 pages or\n512M. It does this regardless of the number of managed pages managed in\nthe zone or the likelihood of success.\n\nThis can put the zone immediately under water in terms of allocating\npages from the zone, and can cause a small machine to fail immediately\ndue to OoM. Unlike set_recommended_min_free_kbytes(), which\nsubstantially increases min_free_kbytes and is tied to THP,\nboost_watermark() can be called even if THP is not active.\n\nThe problem is most likely to appear on architectures such as Arm64\nwhere pageblock_nr_pages is very large.\n\nIt is desirable to run the kdump capture kernel in as small a space as\npossible to avoid wasting memory. In some architectures, such as Arm64,\nthere are restrictions on where the capture kernel can run, and\ntherefore, the space available. A capture kernel running in 768M can\nfail due to OoM immediately after boost_watermark() sets the min in zone\nDMA32, where most of the memory is, to 512M. It fails even though there\nis over 500M of free memory. With boost_watermark() suppressed, the\ncapture kernel can run successfully in 448M.\n\nThis patch limits boost_watermark() to boosting a zone\u0027s min watermark\nonly when there are enough pages that the boost will produce positive\nresults. In this case that is estimated to be four times as many pages\nas pageblock_nr_pages.\n\nMel said:\n\n: There is no harm in marking it stable. Clearly it does not happen very\n: often but it\u0027s not impossible. 32-bit x86 is a lot less common now\n: which would previously have been vulnerable to triggering this easily.\n: ppc64 has a larger base page size but typically only has one zone.\n: arm64 is likely the most vulnerable, particularly when CMA is\n: configured with a small movable zone.\n\nFixes: 1c30844d2dfe (\"mm: reclaim small amounts of memory when an external fragmentation event occurs\")\nSigned-off-by: Henry Willard \u003chenry.willard@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nAcked-by: Mel Gorman \u003cmgorman@techsingularity.net\u003e\nCc: Vlastimil Babka \u003cvbabka@suse.cz\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/1588294148-6586-1-git-send-email-henry.willard@oracle.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8d58f222e85f01da0c0e1fc1e77986c86de889e2", "tree": "cc48a1ad2ce4e222e539256c16c9b54287124cca", "parents": [ "17e34526f0a8f81a214d1ee6f7d8ad2a9c9bae33" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Thu May 07 18:36:23 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:21 2020 -0700" }, "message": "ubsan: disable UBSAN_ALIGNMENT under COMPILE_TEST\n\nThe documentation for UBSAN_ALIGNMENT already mentions that it should\nnot be used on all*config builds (and for efficient-unaligned-access\narchitectures), so just refactor the Kconfig to correctly implement this\nso randconfigs will stop creating insane images that freak out objtool\nunder CONFIG_UBSAN_TRAP (due to the false positives producing functions\nthat never return, etc).\n\nLink: http://lkml.kernel.org/r/202005011433.C42EA3E2D@keescook\nFixes: 0887a7ebc977 (\"ubsan: add trap instrumentation option\")\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Link: https://lore.kernel.org/linux-next/202004231224.D6B3B650@keescook/\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Andrey Ryabinin \u003caryabinin@virtuozzo.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "17e34526f0a8f81a214d1ee6f7d8ad2a9c9bae33", "tree": "dcec89e5f0b55adde372030583e45ba9bc7ead7f", "parents": [ "412895f03cbf9633298111cb4dfde13b7720e2c5" ], "author": { "name": "Qiwu Chen", "email": "qiwuchen55@gmail.com", "time": "Thu May 07 18:36:20 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:21 2020 -0700" }, "message": "mm/vmscan: remove unnecessary argument description of isolate_lru_pages()\n\nSince commit a9e7c39fa9fd9 (\"mm/vmscan.c: remove 7th argument of\nisolate_lru_pages()\"), the explanation of \u0027mode\u0027 argument has been\nunnecessary. Let\u0027s remove it.\n\nSigned-off-by: Qiwu Chen \u003cchenqiwu@xiaomi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nLink: http://lkml.kernel.org/r/20200501090346.2894-1-chenqiwu@xiaomi.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "412895f03cbf9633298111cb4dfde13b7720e2c5", "tree": "7e604e64f5f70225195b69898567de8fb2bb7fc3", "parents": [ "474328c06e3ee75bb6b92826fec90fdc8ef3c573" ], "author": { "name": "Roman Penyaev", "email": "rpenyaev@suse.de", "time": "Thu May 07 18:36:16 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:21 2020 -0700" }, "message": "epoll: atomically remove wait entry on wake up\n\nThis patch does two things:\n\n - fixes a lost wakeup introduced by commit 339ddb53d373 (\"fs/epoll:\n remove unnecessary wakeups of nested epoll\")\n\n - improves performance for events delivery.\n\nThe description of the problem is the following: if N (\u003e1) threads are\nwaiting on ep-\u003ewq for new events and M (\u003e1) events come, it is quite\nlikely that \u003e1 wakeups hit the same wait queue entry, because there is\nquite a big window between __add_wait_queue_exclusive() and the\nfollowing __remove_wait_queue() calls in ep_poll() function.\n\nThis can lead to lost wakeups, because thread, which was woken up, can\nhandle not all the events in -\u003erdllist. (in better words the problem is\ndescribed here: https://lkml.org/lkml/2019/10/7/905)\n\nThe idea of the current patch is to use init_wait() instead of\ninit_waitqueue_entry().\n\nInternally init_wait() sets autoremove_wake_function as a callback,\nwhich removes the wait entry atomically (under the wq locks) from the\nlist, thus the next coming wakeup hits the next wait entry in the wait\nqueue, thus preventing lost wakeups.\n\nProblem is very well reproduced by the epoll60 test case [1].\n\nWait entry removal on wakeup has also performance benefits, because\nthere is no need to take a ep-\u003elock and remove wait entry from the queue\nafter the successful wakeup. Here is the timing output of the epoll60\ntest case:\n\n With explicit wakeup from ep_scan_ready_list() (the state of the\n code prior 339ddb53d373):\n\n real 0m6.970s\n user 0m49.786s\n sys 0m0.113s\n\n After this patch:\n\n real 0m5.220s\n user 0m36.879s\n sys 0m0.019s\n\nThe other testcase is the stress-epoll [2], where one thread consumes\nall the events and other threads produce many events:\n\n With explicit wakeup from ep_scan_ready_list() (the state of the\n code prior 339ddb53d373):\n\n threads events/ms run-time ms\n 8 5427 1474\n 16 6163 2596\n 32 6824 4689\n 64 7060 9064\n 128 6991 18309\n\n After this patch:\n\n threads events/ms run-time ms\n 8 5598 1429\n 16 7073 2262\n 32 7502 4265\n 64 7640 8376\n 128 7634 16767\n\n (number of \"events/ms\" represents event bandwidth, thus higher is\n better; number of \"run-time ms\" represents overall time spent\n doing the benchmark, thus lower is better)\n\n[1] tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c\n[2] https://github.com/rouming/test-tools/blob/master/stress-epoll.c\n\nSigned-off-by: Roman Penyaev \u003crpenyaev@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Jason Baron \u003cjbaron@akamai.com\u003e\nCc: Khazhismel Kumykov \u003ckhazhy@google.com\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Heiher \u003cr@hev.cc\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200430130326.1368509-2-rpenyaev@suse.de\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "474328c06e3ee75bb6b92826fec90fdc8ef3c573", "tree": "76282d78ae14c5384232bc28e9af8299d4b2d636", "parents": [ "28307d938fb2e4056ed4c982c06d1503d7719813" ], "author": { "name": "Roman Penyaev", "email": "rpenyaev@suse.de", "time": "Thu May 07 18:36:13 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:21 2020 -0700" }, "message": "kselftests: introduce new epoll60 testcase for catching lost wakeups\n\nThis test case catches lost wake up introduced by commit 339ddb53d373\n(\"fs/epoll: remove unnecessary wakeups of nested epoll\")\n\nThe test is simple: we have 10 threads and 10 event fds. Each thread\ncan harvest only 1 event. 1 producer fires all 10 events at once and\nwaits that all 10 events will be observed by 10 threads.\n\nIn case of lost wakeup epoll_wait() will timeout and 0 will be returned.\n\nTest case catches two sort of problems: forgotten wakeup on event, which\nhits the -\u003eovflist list, this problem was fixed by:\n\n 5a2513239750 (\"eventpoll: fix missing wakeup for ovflist in ep_poll_callback\")\n\nthe other problem is when several sequential events hit the same waiting\nthread, thus other waiters get no wakeups. Problem is fixed in the\nfollowing patch.\n\nSigned-off-by: Roman Penyaev \u003crpenyaev@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Khazhismel Kumykov \u003ckhazhy@google.com\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Heiher \u003cr@hev.cc\u003e\nCc: Jason Baron \u003cjbaron@akamai.com\u003e\nLink: http://lkml.kernel.org/r/20200430130326.1368509-1-rpenyaev@suse.de\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "28307d938fb2e4056ed4c982c06d1503d7719813", "tree": "c4eb77ac70ae42492eb89c9c13f28410415090f8", "parents": [ "cbfc35a48609ceac978791e3ab9dde0c01f8cb20" ], "author": { "name": "Filipe Manana", "email": "fdmanana@suse.com", "time": "Thu May 07 18:36:10 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:21 2020 -0700" }, "message": "percpu: make pcpu_alloc() aware of current gfp context\n\nSince 5.7-rc1, on btrfs we have a percpu counter initialization for\nwhich we always pass a GFP_KERNEL gfp_t argument (this happens since\ncommit 2992df73268f78 (\"btrfs: Implement DREW lock\")).\n\nThat is safe in some contextes but not on others where allowing fs\nreclaim could lead to a deadlock because we are either holding some\nbtrfs lock needed for a transaction commit or holding a btrfs\ntransaction handle open. Because of that we surround the call to the\nfunction that initializes the percpu counter with a NOFS context using\nmemalloc_nofs_save() (this is done at btrfs_init_fs_root()).\n\nHowever it turns out that this is not enough to prevent a possible\ndeadlock because percpu_alloc() determines if it is in an atomic context\nby looking exclusively at the gfp flags passed to it (GFP_KERNEL in this\ncase) and it is not aware that a NOFS context is set.\n\nBecause percpu_alloc() thinks it is in a non atomic context it locks the\npcpu_alloc_mutex. This can result in a btrfs deadlock when\npcpu_balance_workfn() is running, has acquired that mutex and is waiting\nfor reclaim, while the btrfs task that called percpu_counter_init() (and\ntherefore percpu_alloc()) is holding either the btrfs commit_root\nsemaphore or a transaction handle (done fs/btrfs/backref.c:\niterate_extent_inodes()), which prevents reclaim from finishing as an\nattempt to commit the current btrfs transaction will deadlock.\n\nLockdep reports this issue with the following trace:\n\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n WARNING: possible circular locking dependency detected\n 5.6.0-rc7-btrfs-next-77 #1 Not tainted\n ------------------------------------------------------\n kswapd0/91 is trying to acquire lock:\n ffff8938a3b3fdc8 (\u0026delayed_node-\u003emutex){+.+.}, at: __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs]\n\n but task is already holding lock:\n ffffffffb4f0dbc0 (fs_reclaim){+.+.}, at: __fs_reclaim_acquire+0x5/0x30\n\n which lock already depends on the new lock.\n\n the existing dependency chain (in reverse order) is:\n\n -\u003e #4 (fs_reclaim){+.+.}:\n fs_reclaim_acquire.part.0+0x25/0x30\n __kmalloc+0x5f/0x3a0\n pcpu_create_chunk+0x19/0x230\n pcpu_balance_workfn+0x56a/0x680\n process_one_work+0x235/0x5f0\n worker_thread+0x50/0x3b0\n kthread+0x120/0x140\n ret_from_fork+0x3a/0x50\n\n -\u003e #3 (pcpu_alloc_mutex){+.+.}:\n __mutex_lock+0xa9/0xaf0\n pcpu_alloc+0x480/0x7c0\n __percpu_counter_init+0x50/0xd0\n btrfs_drew_lock_init+0x22/0x70 [btrfs]\n btrfs_get_fs_root+0x29c/0x5c0 [btrfs]\n resolve_indirect_refs+0x120/0xa30 [btrfs]\n find_parent_nodes+0x50b/0xf30 [btrfs]\n btrfs_find_all_leafs+0x60/0xb0 [btrfs]\n iterate_extent_inodes+0x139/0x2f0 [btrfs]\n iterate_inodes_from_logical+0xa1/0xe0 [btrfs]\n btrfs_ioctl_logical_to_ino+0xb4/0x190 [btrfs]\n btrfs_ioctl+0x165a/0x3130 [btrfs]\n ksys_ioctl+0x87/0xc0\n __x64_sys_ioctl+0x16/0x20\n do_syscall_64+0x5c/0x260\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\n -\u003e #2 (\u0026fs_info-\u003ecommit_root_sem){++++}:\n down_write+0x38/0x70\n btrfs_cache_block_group+0x2ec/0x500 [btrfs]\n find_free_extent+0xc6a/0x1600 [btrfs]\n btrfs_reserve_extent+0x9b/0x180 [btrfs]\n btrfs_alloc_tree_block+0xc1/0x350 [btrfs]\n alloc_tree_block_no_bg_flush+0x4a/0x60 [btrfs]\n __btrfs_cow_block+0x122/0x5a0 [btrfs]\n btrfs_cow_block+0x106/0x240 [btrfs]\n commit_cowonly_roots+0x55/0x310 [btrfs]\n btrfs_commit_transaction+0x509/0xb20 [btrfs]\n sync_filesystem+0x74/0x90\n generic_shutdown_super+0x22/0x100\n kill_anon_super+0x14/0x30\n btrfs_kill_super+0x12/0x20 [btrfs]\n deactivate_locked_super+0x31/0x70\n cleanup_mnt+0x100/0x160\n task_work_run+0x93/0xc0\n exit_to_usermode_loop+0xf9/0x100\n do_syscall_64+0x20d/0x260\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\n -\u003e #1 (\u0026space_info-\u003egroups_sem){++++}:\n down_read+0x3c/0x140\n find_free_extent+0xef6/0x1600 [btrfs]\n btrfs_reserve_extent+0x9b/0x180 [btrfs]\n btrfs_alloc_tree_block+0xc1/0x350 [btrfs]\n alloc_tree_block_no_bg_flush+0x4a/0x60 [btrfs]\n __btrfs_cow_block+0x122/0x5a0 [btrfs]\n btrfs_cow_block+0x106/0x240 [btrfs]\n btrfs_search_slot+0x50c/0xd60 [btrfs]\n btrfs_lookup_inode+0x3a/0xc0 [btrfs]\n __btrfs_update_delayed_inode+0x90/0x280 [btrfs]\n __btrfs_commit_inode_delayed_items+0x81f/0x870 [btrfs]\n __btrfs_run_delayed_items+0x8e/0x180 [btrfs]\n btrfs_commit_transaction+0x31b/0xb20 [btrfs]\n iterate_supers+0x87/0xf0\n ksys_sync+0x60/0xb0\n __ia32_sys_sync+0xa/0x10\n do_syscall_64+0x5c/0x260\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\n -\u003e #0 (\u0026delayed_node-\u003emutex){+.+.}:\n __lock_acquire+0xef0/0x1c80\n lock_acquire+0xa2/0x1d0\n __mutex_lock+0xa9/0xaf0\n __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs]\n btrfs_evict_inode+0x40d/0x560 [btrfs]\n evict+0xd9/0x1c0\n dispose_list+0x48/0x70\n prune_icache_sb+0x54/0x80\n super_cache_scan+0x124/0x1a0\n do_shrink_slab+0x176/0x440\n shrink_slab+0x23a/0x2c0\n shrink_node+0x188/0x6e0\n balance_pgdat+0x31d/0x7f0\n kswapd+0x238/0x550\n kthread+0x120/0x140\n ret_from_fork+0x3a/0x50\n\n other info that might help us debug this:\n\n Chain exists of:\n \u0026delayed_node-\u003emutex --\u003e pcpu_alloc_mutex --\u003e fs_reclaim\n\n Possible unsafe locking scenario:\n\n CPU0 CPU1\n ---- ----\n lock(fs_reclaim);\n lock(pcpu_alloc_mutex);\n lock(fs_reclaim);\n lock(\u0026delayed_node-\u003emutex);\n\n *** DEADLOCK ***\n\n 3 locks held by kswapd0/91:\n #0: (fs_reclaim){+.+.}, at: __fs_reclaim_acquire+0x5/0x30\n #1: (shrinker_rwsem){++++}, at: shrink_slab+0x12f/0x2c0\n #2: (\u0026type-\u003es_umount_key#43){++++}, at: trylock_super+0x16/0x50\n\n stack backtrace:\n CPU: 1 PID: 91 Comm: kswapd0 Not tainted 5.6.0-rc7-btrfs-next-77 #1\n Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-0-ga698c8995f-prebuilt.qemu.org 04/01/2014\n Call Trace:\n dump_stack+0x8f/0xd0\n check_noncircular+0x170/0x190\n __lock_acquire+0xef0/0x1c80\n lock_acquire+0xa2/0x1d0\n __mutex_lock+0xa9/0xaf0\n __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs]\n btrfs_evict_inode+0x40d/0x560 [btrfs]\n evict+0xd9/0x1c0\n dispose_list+0x48/0x70\n prune_icache_sb+0x54/0x80\n super_cache_scan+0x124/0x1a0\n do_shrink_slab+0x176/0x440\n shrink_slab+0x23a/0x2c0\n shrink_node+0x188/0x6e0\n balance_pgdat+0x31d/0x7f0\n kswapd+0x238/0x550\n kthread+0x120/0x140\n ret_from_fork+0x3a/0x50\n\nThis could be fixed by making btrfs pass GFP_NOFS instead of GFP_KERNEL\nto percpu_counter_init() in contextes where it is not reclaim safe,\nhowever that type of approach is discouraged since\nmemalloc_[nofs|noio]_save() were introduced. Therefore this change\nmakes pcpu_alloc() look up into an existing nofs/noio context before\ndeciding whether it is in an atomic context or not.\n\nSigned-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nAcked-by: Dennis Zhou \u003cdennis@kernel.org\u003e\nCc: Tejun Heo \u003ctj@kernel.org\u003e\nCc: Christoph Lameter \u003ccl@linux.com\u003e\nLink: http://lkml.kernel.org/r/20200430164356.15543-1-fdmanana@kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "cbfc35a48609ceac978791e3ab9dde0c01f8cb20", "tree": "9797f682c1e5800fbd8a73a9ca953636b3806f58", "parents": [ "50e36be1fb9572b2e4f2753340bdce3116bf2ce7" ], "author": { "name": "Waiman Long", "email": "longman@redhat.com", "time": "Thu May 07 18:36:06 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "mm/slub: fix incorrect interpretation of s-\u003eoffset\n\nIn a couple of places in the slub memory allocator, the code uses\n\"s-\u003eoffset\" as a check to see if the free pointer is put right after the\nobject. That check is no longer true with commit 3202fa62fb43 (\"slub:\nrelocate freelist pointer to middle of object\").\n\nAs a result, echoing \"1\" into the validate sysfs file, e.g. of dentry,\nmay cause a bunch of \"Freepointer corrupt\" error reports like the\nfollowing to appear with the system in panic afterwards.\n\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG dentry(666:pmcd.service) (Tainted: G B): Freepointer corrupt\n -----------------------------------------------------------------------------\n\nTo fix it, use the check \"s-\u003eoffset \u003d\u003d s-\u003einuse\" in the new helper\nfunction freeptr_outside_object() instead. Also add another helper\nfunction get_info_end() to return the end of info block (inuse + free\npointer if not overlapping with object).\n\nFixes: 3202fa62fb43 (\"slub: relocate freelist pointer to middle of object\")\nSigned-off-by: Waiman Long \u003clongman@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Rafael Aquini \u003caquini@redhat.com\u003e\nCc: Christoph Lameter \u003ccl@linux.com\u003e\nCc: Vitaly Nikolenko \u003cvnik@duasynt.com\u003e\nCc: Silvio Cesare \u003csilvio.cesare@gmail.com\u003e\nCc: Pekka Enberg \u003cpenberg@kernel.org\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Joonsoo Kim \u003ciamjoonsoo.kim@lge.com\u003e\nCc: Markus Elfring \u003cMarkus.Elfring@web.de\u003e\nCc: Changbin Du \u003cchangbin.du@gmail.com\u003e\nLink: http://lkml.kernel.org/r/20200429135328.26976-1-longman@redhat.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "50e36be1fb9572b2e4f2753340bdce3116bf2ce7", "tree": "235c826e4a8eba5d50110ade00c96e60bb9bbd33", "parents": [ "0c54a6a44bf3d41e76ce3f583a6ece267618df2e" ], "author": { "name": "Aymeric Agon-Rambosson", "email": "aymeric.agon@yandex.com", "time": "Thu May 07 18:36:03 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "scripts/gdb: repair rb_first() and rb_last()\n\nThe current implementations of the rb_first() and rb_last() gdb\nfunctions have a variable that references itself in its instanciation,\nwhich causes the function to throw an error if a specific condition on\nthe argument is met. The original author rather intended to reference\nthe argument and made a typo. Referring the argument instead makes the\nfunction work as intended.\n\nSigned-off-by: Aymeric Agon-Rambosson \u003caymeric.agon@yandex.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\nCc: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\nCc: Kieran Bingham \u003ckbingham@kernel.org\u003e\nCc: Douglas Anderson \u003cdianders@chromium.org\u003e\nCc: Nikolay Borisov \u003cn.borisov.lkml@gmail.com\u003e\nCc: Jackie Liu \u003cliuyun01@kylinos.cn\u003e\nCc: Jason Wessel \u003cjason.wessel@windriver.com\u003e\nLink: http://lkml.kernel.org/r/20200427051029.354840-1-aymeric.agon@yandex.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0c54a6a44bf3d41e76ce3f583a6ece267618df2e", "tree": "b16b25712931adde867ee34e1afbbd48b450cc79", "parents": [ "996ed22c7a5251d76dcdfe5026ef8230e90066d9" ], "author": { "name": "Khazhismel Kumykov", "email": "khazhy@google.com", "time": "Thu May 07 18:35:59 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "eventpoll: fix missing wakeup for ovflist in ep_poll_callback\n\nIn the event that we add to ovflist, before commit 339ddb53d373\n(\"fs/epoll: remove unnecessary wakeups of nested epoll\") we would be\nwoken up by ep_scan_ready_list, and did no wakeup in ep_poll_callback.\n\nWith that wakeup removed, if we add to ovflist here, we may never wake\nup. Rather than adding back the ep_scan_ready_list wakeup - which was\nresulting in unnecessary wakeups, trigger a wake-up in ep_poll_callback.\n\nWe noticed that one of our workloads was missing wakeups starting with\n339ddb53d373 and upon manual inspection, this wakeup seemed missing to me.\nWith this patch added, we no longer see missing wakeups. I haven\u0027t yet\ntried to make a small reproducer, but the existing kselftests in\nfilesystem/epoll passed for me with this patch.\n\n[khazhy@google.com: use if/elif instead of goto + cleanup suggested by Roman]\n Link: http://lkml.kernel.org/r/20200424190039.192373-1-khazhy@google.com\nFixes: 339ddb53d373 (\"fs/epoll: remove unnecessary wakeups of nested epoll\")\nSigned-off-by: Khazhismel Kumykov \u003ckhazhy@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Roman Penyaev \u003crpenyaev@suse.de\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Roman Penyaev \u003crpenyaev@suse.de\u003e\nCc: Heiher \u003cr@hev.cc\u003e\nCc: Jason Baron \u003cjbaron@akamai.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200424025057.118641-1-khazhy@google.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "996ed22c7a5251d76dcdfe5026ef8230e90066d9", "tree": "7d36617f8a17687412ede1f5da6f780a68926a89", "parents": [ "e08df079b23e2e982df15aa340bfbaf50f297504" ], "author": { "name": "Janakarajan Natarajan", "email": "Janakarajan.Natarajan@amd.com", "time": "Thu May 07 18:35:56 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "arch/x86/kvm/svm/sev.c: change flag passed to GUP fast in sev_pin_memory()\n\nWhen trying to lock read-only pages, sev_pin_memory() fails because\nFOLL_WRITE is used as the flag for get_user_pages_fast().\n\nCommit 73b0140bf0fe (\"mm/gup: change GUP fast to use flags rather than a\nwrite \u0027bool\u0027\") updated the get_user_pages_fast() call sites to use\nflags, but incorrectly updated the call in sev_pin_memory(). As the\noriginal coding of this call was correct, revert the change made by that\ncommit.\n\nFixes: 73b0140bf0fe (\"mm/gup: change GUP fast to use flags rather than a write \u0027bool\u0027\")\nSigned-off-by: Janakarajan Natarajan \u003cJanakarajan.Natarajan@amd.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\nCc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nCc: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nCc: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nCc: Wanpeng Li \u003cwanpengli@tencent.com\u003e\nCc: Jim Mattson \u003cjmattson@google.com\u003e\nCc: Joerg Roedel \u003cjoro@8bytes.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Borislav Petkov \u003cbp@alien8.de\u003e\nCc: \"H . Peter Anvin\" \u003chpa@zytor.com\u003e\nCc: Mike Marshall \u003chubcap@omnibond.com\u003e\nCc: Brijesh Singh \u003cbrijesh.singh@amd.com\u003e\nLink: http://lkml.kernel.org/r/20200423152419.87202-1-Janakarajan.Natarajan@amd.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e08df079b23e2e982df15aa340bfbaf50f297504", "tree": "2b5af094e8837e7d08875b8c88c4fb70d9e7e7e9", "parents": [ "324cfb19567c80ed71d7a02f1d5ff4621902f4c3" ], "author": { "name": "Ivan Delalande", "email": "colona@arista.com", "time": "Thu May 07 18:35:53 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "scripts/decodecode: fix trapping instruction formatting\n\nIf the trapping instruction contains a \u0027:\u0027, for a memory access through\nsegment registers for example, the sed substitution will insert the \u0027*\u0027\nmarker in the middle of the instruction instead of the line address:\n\n\t2b: 65 48 0f c7 0f cmpxchg16b %gs:*(%rdi) \u003c-- trapping instruction\n\nI started to think I had forgotten some quirk of the assembly syntax\nbefore noticing that it was actually coming from the script. Fix it to\nadd the address marker at the right place for these instructions:\n\n\t28: 49 8b 06 mov (%r14),%rax\n\t2b:* 65 48 0f c7 0f cmpxchg16b %gs:(%rdi) \u003c-- trapping instruction\n\t30: 0f 94 c0 sete %al\n\nFixes: 18ff44b189e2 (\"scripts/decodecode: make faulting insn ptr more robust\")\nSigned-off-by: Ivan Delalande \u003ccolona@arista.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Borislav Petkov \u003cbp@suse.de\u003e\nLink: http://lkml.kernel.org/r/20200419223653.GA31248@visor\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "324cfb19567c80ed71d7a02f1d5ff4621902f4c3", "tree": "7114dfbf26eba78d3b65f2aea48858369a1f45fd", "parents": [ "e84fe99b68ce353c37ceeecc95dce9696c976556" ], "author": { "name": "Maciej Grochowski", "email": "maciej.grochowski@pm.me", "time": "Thu May 07 18:35:49 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "kernel/kcov.c: fix typos in kcov_remote_start documentation\n\nSigned-off-by: Maciej Grochowski \u003cmaciej.grochowski@pm.me\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nLink: http://lkml.kernel.org/r/20200420030259.31674-1-maciek.grochowski@gmail.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e84fe99b68ce353c37ceeecc95dce9696c976556", "tree": "d7163339a36226679c493628f01ffe1d8986ff72", "parents": [ "11d6761218d19ca06ae5387f4e3692c4fa9e7493" ], "author": { "name": "David Hildenbrand", "email": "david@redhat.com", "time": "Thu May 07 18:35:46 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "mm/page_alloc: fix watchdog soft lockups during set_zone_contiguous()\n\nWithout CONFIG_PREEMPT, it can happen that we get soft lockups detected,\ne.g., while booting up.\n\n watchdog: BUG: soft lockup - CPU#0 stuck for 22s! [swapper/0:1]\n CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.6.0-next-20200331+ #4\n Hardware name: Red Hat KVM, BIOS 1.11.1-4.module+el8.1.0+4066+0f1aadab 04/01/2014\n RIP: __pageblock_pfn_to_page+0x134/0x1c0\n Call Trace:\n set_zone_contiguous+0x56/0x70\n page_alloc_init_late+0x166/0x176\n kernel_init_freeable+0xfa/0x255\n kernel_init+0xa/0x106\n ret_from_fork+0x35/0x40\n\nThe issue becomes visible when having a lot of memory (e.g., 4TB)\nassigned to a single NUMA node - a system that can easily be created\nusing QEMU. Inside VMs on a hypervisor with quite some memory\novercommit, this is fairly easy to trigger.\n\nSigned-off-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nReviewed-by: Pavel Tatashin \u003cpasha.tatashin@soleen.com\u003e\nReviewed-by: Pankaj Gupta \u003cpankaj.gupta.linux@gmail.com\u003e\nReviewed-by: Baoquan He \u003cbhe@redhat.com\u003e\nReviewed-by: Shile Zhang \u003cshile.zhang@linux.alibaba.com\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: Kirill Tkhai \u003cktkhai@virtuozzo.com\u003e\nCc: Shile Zhang \u003cshile.zhang@linux.alibaba.com\u003e\nCc: Pavel Tatashin \u003cpasha.tatashin@soleen.com\u003e\nCc: Daniel Jordan \u003cdaniel.m.jordan@oracle.com\u003e\nCc: Michal Hocko \u003cmhocko@kernel.org\u003e\nCc: Alexander Duyck \u003calexander.duyck@gmail.com\u003e\nCc: Baoquan He \u003cbhe@redhat.com\u003e\nCc: Oscar Salvador \u003cosalvador@suse.de\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/20200416073417.5003-1-david@redhat.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "11d6761218d19ca06ae5387f4e3692c4fa9e7493", "tree": "3290a247cd155247a76c43735b9cec84f908d3ec", "parents": [ "b5f2006144c6ae941726037120fa1001ddede784" ], "author": { "name": "Yafang Shao", "email": "laoar.shao@gmail.com", "time": "Thu May 07 18:35:43 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "mm, memcg: fix error return value of mem_cgroup_css_alloc()\n\nWhen I run my memcg testcase which creates lots of memcgs, I found\nthere\u0027re unexpected out of memory logs while there\u0027re still enough\navailable free memory. The error log is\n\n mkdir: cannot create directory \u0027foo.65533\u0027: Cannot allocate memory\n\nThe reason is when we try to create more than MEM_CGROUP_ID_MAX memcgs,\nan -ENOMEM errno will be set by mem_cgroup_css_alloc(), but the right\nerrno should be -ENOSPC \"No space left on device\", which is an\nappropriate errno for userspace\u0027s failed mkdir.\n\nAs the errno really misled me, we should make it right. After this\npatch, the error log will be\n\n mkdir: cannot create directory \u0027foo.65533\u0027: No space left on device\n\n[akpm@linux-foundation.org: s/EBUSY/ENOSPC/, per Michal]\n[akpm@linux-foundation.org: s/EBUSY/ENOSPC/, per Michal]\nFixes: 73f576c04b94 (\"mm: memcontrol: fix cgroup creation failure after many small jobs\")\nSuggested-by: Matthew Wilcox \u003cwilly@infradead.org\u003e\nSigned-off-by: Yafang Shao \u003claoar.shao@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nAcked-by: Michal Hocko \u003cmhocko@kernel.org\u003e\nAcked-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Vladimir Davydov \u003cvdavydov.dev@gmail.com\u003e\nLink: http://lkml.kernel.org/r/20200407063621.GA18914@dhcp22.suse.cz\nLink: http://lkml.kernel.org/r/1586192163-20099-1-git-send-email-laoar.shao@gmail.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b5f2006144c6ae941726037120fa1001ddede784", "tree": "b5f3df12dee3889685ca560c5aa5ea63781c0376", "parents": [ "a811c1fa0a02c062555b54651065899437bacdbe" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Thu May 07 18:35:39 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 19:27:20 2020 -0700" }, "message": "ipc/mqueue.c: change __do_notify() to bypass check_kill_permission()\n\nCommit cc731525f26a (\"signal: Remove kernel interal si_code magic\")\nchanged the value of SI_FROMUSER(SI_MESGQ), this means that mq_notify() no\nlonger works if the sender doesn\u0027t have rights to send a signal.\n\nChange __do_notify() to use do_send_sig_info() instead of kill_pid_info()\nto avoid check_kill_permission().\n\nThis needs the additional notify.sigev_signo !\u003d 0 check, shouldn\u0027t we\nchange do_mq_notify() to deny sigev_signo \u003d\u003d 0 ?\n\nTest-case:\n\n\t#include \u003csignal.h\u003e\n\t#include \u003cmqueue.h\u003e\n\t#include \u003cunistd.h\u003e\n\t#include \u003csys/wait.h\u003e\n\t#include \u003cassert.h\u003e\n\n\tstatic int notified;\n\n\tstatic void sigh(int sig)\n\t{\n\t\tnotified \u003d 1;\n\t}\n\n\tint main(void)\n\t{\n\t\tsignal(SIGIO, sigh);\n\n\t\tint fd \u003d mq_open(\"/mq\", O_RDWR|O_CREAT, 0666, NULL);\n\t\tassert(fd \u003e\u003d 0);\n\n\t\tstruct sigevent se \u003d {\n\t\t\t.sigev_notify\t\u003d SIGEV_SIGNAL,\n\t\t\t.sigev_signo\t\u003d SIGIO,\n\t\t};\n\t\tassert(mq_notify(fd, \u0026se) \u003d\u003d 0);\n\n\t\tif (!fork()) {\n\t\t\tassert(setuid(1) \u003d\u003d 0);\n\t\t\tmq_send(fd, \"\",1,0);\n\t\t\treturn 0;\n\t\t}\n\n\t\twait(NULL);\n\t\tmq_unlink(\"/mq\");\n\t\tassert(notified);\n\t\treturn 0;\n\t}\n\n[manfred@colorfullife.com: 1) Add self_exec_id evaluation so that the implementation matches do_notify_parent 2) use PIDTYPE_TGID everywhere]\nFixes: cc731525f26a (\"signal: Remove kernel interal si_code magic\")\nReported-by: Yoji \u003cyoji.fujihar.min@gmail.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nSigned-off-by: Manfred Spraul \u003cmanfred@colorfullife.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nAcked-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nCc: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nCc: \u003c1vier1@web.de\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/e2a782e4-eab9-4f5c-c749-c07a8f7a4e66@colorfullife.com\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "192ffb7515839b1cc8457e0a8c1e09783de019d3", "tree": "a0c710bd51591401ad466564bea0f5bb819cf6cc", "parents": [ "9ecc4d775f61fb3fd26a1b8cc70a56a98b3b40d3", "8842604446d1f005abcbf8c63c12eabdb5695094" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 15:27:11 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 15:27:11 2020 -0700" }, "message": "Merge tag \u0027trace-v5.7-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing fixes from Steven Rostedt:\n\n - Fix bootconfig causing kernels to fail with CONFIG_BLK_DEV_RAM\n enabled\n\n - Fix allocation leaks in bootconfig tool\n\n - Fix a double initialization of a variable\n\n - Fix API bootconfig usage from kprobe boot time events\n\n - Reject NULL location for kprobes\n\n - Fix crash caused by preempt delay module not cleaning up kthread\n correctly\n\n - Add vmalloc_sync_mappings() to prevent x86_64 page faults from\n recursively faulting from tracing page faults\n\n - Fix comment in gpu/trace kerneldoc header\n\n - Fix documentation of how to create a trace event class\n\n - Make the local tracing_snapshot_instance_cond() function static\n\n* tag \u0027trace-v5.7-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n tools/bootconfig: Fix resource leak in apply_xbc()\n tracing: Make tracing_snapshot_instance_cond() static\n tracing: Fix doc mistakes in trace sample\n gpu/trace: Minor comment updates for gpu_mem_total tracepoint\n tracing: Add a vmalloc_sync_mappings() for safe measure\n tracing: Wait for preempt irq delay thread to finish\n tracing/kprobes: Reject new event if loc is NULL\n tracing/boottime: Fix kprobe event API usage\n tracing/kprobes: Fix a double initialization typo\n bootconfig: Fix to remove bootconfig data from initrd while boot\n" }, { "commit": "9ecc4d775f61fb3fd26a1b8cc70a56a98b3b40d3", "tree": "4a98f96f584550499be8e87e7bf8d9de9f4282a2", "parents": [ "6e7f2eacf09811d092c1b41263108ac7fe0d089d", "66d69e081b526b6a6031f0d3ca8ddff71e5707a5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 15:22:08 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 15:22:08 2020 -0700" }, "message": "Merge tag \u0027linux-kselftest-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest\n\nPull kselftest fixes from Shuah Khan:\n \"ftrace test fixes and a fix to kvm Makefile for relocatable\n native/cross builds and installs\"\n\n* tag \u0027linux-kselftest-5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:\n selftests: fix kvm relocatable native/cross builds and installs\n selftests/ftrace: Make XFAIL green color\n ftrace/selftest: make unresolved cases cause failure if --fail-unresolved set\n ftrace/selftests: workaround cgroup RT scheduling issues\n" }, { "commit": "63ff822358b276137059520cf16e587e8073e80f", "tree": "fdffa1a1cf9a869305c5759ef1aa8e36d2ece1db", "parents": [ "90da2e3f25c8b4d742b2687b8fed8fc4eb8851da" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu May 07 14:56:15 2020 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu May 07 14:56:15 2020 -0600" }, "message": "io_uring: don\u0027t use \u0027fd\u0027 for openat/openat2/statx\n\nWe currently make some guesses as when to open this fd, but in reality\nwe have no business (or need) to do so at all. In fact, it makes certain\nthings fail, like O_PATH.\n\nRemove the fd lookup from these opcodes, we\u0027re just passing the \u0027fd\u0027 to\ngeneric helpers anyway. With that, we can also remove the special casing\nof fd values in io_req_needs_file(), and the \u0027fd_non_neg\u0027 check that\nwe have. And we can ensure that we only read sqe-\u003efd once.\n\nThis fixes O_PATH usage with openat/openat2, and ditto statx path side\noddities.\n\nCc: stable@vger.kernel.org: # v5.6\nReported-by: Max Kellermann \u003cmk@cm4all.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "8842604446d1f005abcbf8c63c12eabdb5695094", "tree": "90c8adbc1fa5d018bf2a99aebb4ad941856c22d8", "parents": [ "192b7993b3ff92b62b687e940e5e88fa0218d764" ], "author": { "name": "Yunfeng Ye", "email": "yeyunfeng@huawei.com", "time": "Thu May 07 17:23:36 2020 +0800" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu May 07 14:18:27 2020 -0400" }, "message": "tools/bootconfig: Fix resource leak in apply_xbc()\n\nFix the @data and @fd allocations that are leaked in the error path of\napply_xbc().\n\nLink: http://lkml.kernel.org/r/583a49c9-c27a-931d-e6c2-6f63a4b18bea@huawei.com\n\nAcked-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nSigned-off-by: Yunfeng Ye \u003cyeyunfeng@huawei.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "192b7993b3ff92b62b687e940e5e88fa0218d764", "tree": "8678e4369f35b8cf4b839299380c8accf630bdff", "parents": [ "f094a233e1d5b1c61cc797d204aa28b611058827" ], "author": { "name": "Zou Wei", "email": "zou_wei@huawei.com", "time": "Thu Apr 23 12:08:25 2020 +0800" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu May 07 13:32:58 2020 -0400" }, "message": "tracing: Make tracing_snapshot_instance_cond() static\n\nFix the following sparse warning:\n\nkernel/trace/trace.c:950:6: warning: symbol \u0027tracing_snapshot_instance_cond\u0027\nwas not declared. Should it be static?\n\nLink: http://lkml.kernel.org/r/1587614905-48692-1-git-send-email-zou_wei@huawei.com\n\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nSigned-off-by: Zou Wei \u003czou_wei@huawei.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "f094a233e1d5b1c61cc797d204aa28b611058827", "tree": "134f567314525e3b6224c5631a01c4602c4bbd71", "parents": [ "386c82a70319d42dba4f1b30e5e7076f2b4d8c2f" ], "author": { "name": "Wei Yang", "email": "richard.weiyang@gmail.com", "time": "Tue Apr 28 21:49:59 2020 +0000" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu May 07 13:32:57 2020 -0400" }, "message": "tracing: Fix doc mistakes in trace sample\n\nAs the example below shows, DECLARE_EVENT_CLASS() is used instead of\nDEFINE_EVENT_CLASS().\n\nLink: http://lkml.kernel.org/r/20200428214959.11259-1-richard.weiyang@gmail.com\n\nSigned-off-by: Wei Yang \u003crichard.weiyang@gmail.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "386c82a70319d42dba4f1b30e5e7076f2b4d8c2f", "tree": "f584d53b7514b90c8d98f197b9b824c9cf674366", "parents": [ "11f5efc3ab66284f7aaacc926e9351d658e2577b" ], "author": { "name": "Yiwei Zhang", "email": "zzyiwei@google.com", "time": "Tue Apr 28 15:08:25 2020 -0700" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu May 07 13:32:57 2020 -0400" }, "message": "gpu/trace: Minor comment updates for gpu_mem_total tracepoint\n\nThis change updates the improper comment for the \u0027size\u0027 attribute in the\ntracepoint definition. Most gfx drivers pre-fault in physical pages\ninstead of making virtual allocations. So we drop the \u0027Virtual\u0027 keyword\nhere and leave this to the implementations.\n\nLink: http://lkml.kernel.org/r/20200428220825.169606-1-zzyiwei@google.com\n\nSigned-off-by: Yiwei Zhang \u003czzyiwei@google.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "11f5efc3ab66284f7aaacc926e9351d658e2577b", "tree": "929e6ecc8dec198f556309b70b54c799dcc752fe", "parents": [ "d16a8c31077e75ecb9427fbfea59b74eed00f698" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed May 06 10:36:18 2020 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu May 07 13:32:57 2020 -0400" }, "message": "tracing: Add a vmalloc_sync_mappings() for safe measure\n\nx86_64 lazily maps in the vmalloc pages, and the way this works with per_cpu\nareas can be complex, to say the least. Mappings may happen at boot up, and\nif nothing synchronizes the page tables, those page mappings may not be\nsynced till they are used. This causes issues for anything that might touch\none of those mappings in the path of the page fault handler. When one of\nthose unmapped mappings is touched in the page fault handler, it will cause\nanother page fault, which in turn will cause a page fault, and leave us in\na loop of page faults.\n\nCommit 763802b53a42 (\"x86/mm: split vmalloc_sync_all()\") split\nvmalloc_sync_all() into vmalloc_sync_unmappings() and\nvmalloc_sync_mappings(), as on system exit, it did not need to do a full\nsync on x86_64 (although it still needed to be done on x86_32). By chance,\nthe vmalloc_sync_all() would synchronize the page mappings done at boot up\nand prevent the per cpu area from being a problem for tracing in the page\nfault handler. But when that synchronization in the exit of a task became a\nnop, it caused the problem to appear.\n\nLink: https://lore.kernel.org/r/20200429054857.66e8e333@oasis.local.home\n\nCc: stable@vger.kernel.org\nFixes: 737223fbca3b1 (\"tracing: Consolidate buffer allocation code\")\nReported-by: \"Tzvetomir Stoyanov (VMware)\" \u003ctz.stoyanov@gmail.com\u003e\nSuggested-by: Joerg Roedel \u003cjroedel@suse.de\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "d16a8c31077e75ecb9427fbfea59b74eed00f698", "tree": "126e7e7715b2d4a92904fce097df68da3aca8a6d", "parents": [ "5b4dcd2d201a395ad4054067bfae4a07554fbd65" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed May 06 10:20:10 2020 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu May 07 13:32:40 2020 -0400" }, "message": "tracing: Wait for preempt irq delay thread to finish\n\nRunning on a slower machine, it is possible that the preempt delay kernel\nthread may still be executing if the module was immediately removed after\nadded, and this can cause the kernel to crash as the kernel thread might be\nexecuting after its code has been removed.\n\nThere\u0027s no reason that the caller of the code shouldn\u0027t just wait for the\ndelay thread to finish, as the thread can also be created by a trigger in\nthe sysfs code, which also has the same issues.\n\nLink: http://lore.kernel.org/r/5EA2B0C8.2080706@cn.fujitsu.com\n\nCc: stable@vger.kernel.org\nFixes: 793937236d1ee (\"lib: Add module for testing preemptoff/irqsoff latency tracers\")\nReported-by: Xiao Yang \u003cyangx.jy@cn.fujitsu.com\u003e\nReviewed-by: Xiao Yang \u003cyangx.jy@cn.fujitsu.com\u003e\nReviewed-by: Joel Fernandes \u003cjoel@joelfernandes.org\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "6e7f2eacf09811d092c1b41263108ac7fe0d089d", "tree": "a04c5c449fc4c44a994feca4e39aade19ab4989e", "parents": [ "8c16ec94dc767a4d8c48149d646e8c835512cf8f", "027d0c7101f50cf03aeea9eebf484afd4920c8d3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 09:55:58 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 09:55:58 2020 -0700" }, "message": "Merge tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 fix from Catalin Marinas:\n \"Avoid potential NULL dereference in huge_pte_alloc() on pmd_alloc()\n failure\"\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n arm64: hugetlb: avoid potential NULL dereference\n" }, { "commit": "8c16ec94dc767a4d8c48149d646e8c835512cf8f", "tree": "7efeecfe535f76f455cc92a0664d59e8fa56ce52", "parents": [ "de268ccb42d6ec5475ec5a5e60723b665d6e0af2", "2673cb6849722a4ffd74c27a9200a9ec43f64be3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 09:50:59 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 09:50:59 2020 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nPull kvm fixes from Paolo Bonzini:\n \"Bugfixes, mostly for ARM and AMD, and more documentation.\n\n Slightly bigger than usual because I couldn\u0027t send out what was\n pending for rc4, but there is nothing worrisome going on. I have more\n fixes pending for guest debugging support (gdbstub) but I will send\n them next week\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm: (22 commits)\n KVM: X86: Declare KVM_CAP_SET_GUEST_DEBUG properly\n KVM: selftests: Fix build for evmcs.h\n kvm: x86: Use KVM CPU capabilities to determine CR4 reserved bits\n KVM: VMX: Explicitly clear RFLAGS.CF and RFLAGS.ZF in VM-Exit RSB path\n docs/virt/kvm: Document configuring and running nested guests\n KVM: s390: Remove false WARN_ON_ONCE for the PQAP instruction\n kvm: ioapic: Restrict lazy EOI update to edge-triggered interrupts\n KVM: x86: Fixes posted interrupt check for IRQs delivery modes\n KVM: SVM: fill in kvm_run-\u003edebug.arch.dr[67]\n KVM: nVMX: Replace a BUG_ON(1) with BUG() to squash clang warning\n KVM: arm64: Fix 32bit PC wrap-around\n KVM: arm64: vgic-v4: Initialize GICv4.1 even in the absence of a virtual ITS\n KVM: arm64: Save/restore sp_el0 as part of __guest_enter\n KVM: arm64: Delete duplicated label in invalid_vector\n KVM: arm64: vgic-its: Fix memory leak on the error path of vgic_add_lpi()\n KVM: arm64: vgic-v3: Retire all pending LPIs on vcpu destroy\n KVM: arm: vgic-v2: Only use the virtual state when userspace accesses pending bits\n KVM: arm: vgic: Only use the virtual state when userspace accesses enable bits\n KVM: arm: vgic: Synchronize the whole guest on GIC{D,R}_I{S,C}ACTIVER read\n KVM: arm64: PSCI: Forbid 64bit functions for 32bit guests\n ...\n" }, { "commit": "de268ccb42d6ec5475ec5a5e60723b665d6e0af2", "tree": "14ac01ecb135033e68c2433c96ada79cde2fe449", "parents": [ "a811c1fa0a02c062555b54651065899437bacdbe", "8aebfffacfa379ba400da573a5bf9e49634e38cb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 09:48:37 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu May 07 09:48:37 2020 -0700" }, "message": "Merge tag \u0027configfs-for-5.7\u0027 of git://git.infradead.org/users/hch/configfs\n\nPull configfs fix from Christoph Hellwig:\n \"Fix a refcount leak in configfs_rmdir (Xiyu Yang)\"\n\n* tag \u0027configfs-for-5.7\u0027 of git://git.infradead.org/users/hch/configfs:\n configfs: fix config_item refcnt leak in configfs_rmdir()\n" }, { "commit": "90da2e3f25c8b4d742b2687b8fed8fc4eb8851da", "tree": "6937c8d00b23be9016972d3000838338f9a09a6d", "parents": [ "7f13657d141346125f4d0bb93eab4777f40c406e" ], "author": { "name": "Pavel Begunkov", "email": "asml.silence@gmail.com", "time": "Mon May 04 22:39:35 2020 +0300" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu May 07 09:45:07 2020 -0600" }, "message": "splice: move f_mode checks to do_{splice,tee}()\n\ndo_splice() is used by io_uring, as will be do_tee(). Move f_mode\nchecks from sys_{splice,tee}() to do_{splice,tee}(), so they\u0027re\nenforced for io_uring as well.\n\nFixes: 7d67af2c0134 (\"io_uring: add splice(2) support\")\nReported-by: Jann Horn \u003cjannh@google.com\u003e\nSigned-off-by: Pavel Begunkov \u003casml.silence@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "027d0c7101f50cf03aeea9eebf484afd4920c8d3", "tree": "fd9ceb1343c30dbaa368827a8a42274a3087056a", "parents": [ "1578e5d03112e3e9d37e1c4d95b6dfb734c73955" ], "author": { "name": "Mark Rutland", "email": "mark.rutland@arm.com", "time": "Tue May 05 13:59:30 2020 +0100" }, "committer": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Thu May 07 09:24:15 2020 +0100" }, "message": "arm64: hugetlb: avoid potential NULL dereference\n\nThe static analyzer in GCC 10 spotted that in huge_pte_alloc() we may\npass a NULL pmdp into pte_alloc_map() when pmd_alloc() returns NULL:\n\n| CC arch/arm64/mm/pageattr.o\n| CC arch/arm64/mm/hugetlbpage.o\n| from arch/arm64/mm/hugetlbpage.c:10:\n| arch/arm64/mm/hugetlbpage.c: In function ‘huge_pte_alloc’:\n| ./arch/arm64/include/asm/pgtable-types.h:28:24: warning: dereference of NULL ‘pmdp’ [CWE-690] [-Wanalyzer-null-dereference]\n| ./arch/arm64/include/asm/pgtable.h:436:26: note: in expansion of macro ‘pmd_val’\n| arch/arm64/mm/hugetlbpage.c:242:10: note: in expansion of macro ‘pte_alloc_map’\n| |arch/arm64/mm/hugetlbpage.c:232:10:\n| |./arch/arm64/include/asm/pgtable-types.h:28:24:\n| ./arch/arm64/include/asm/pgtable.h:436:26: note: in expansion of macro ‘pmd_val’\n| arch/arm64/mm/hugetlbpage.c:242:10: note: in expansion of macro ‘pte_alloc_map’\n\nThis can only occur when the kernel cannot allocate a page, and so is\nunlikely to happen in practice before other systems start failing.\n\nWe can avoid this by bailing out if pmd_alloc() fails, as we do earlier\nin the function if pud_alloc() fails.\n\nFixes: 66b3923a1a0f (\"arm64: hugetlb: add support for PTE contiguous bit\")\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nReported-by: Kyrill Tkachov \u003ckyrylo.tkachov@arm.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e # 4.5.x-\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n" }, { "commit": "91edf63d5022bd0464788ffb4acc3d5febbaf81d", "tree": "02e69f34df4b6de50e566fa09b8603420fc679f5", "parents": [ "084d7e78467534eedaf0d6c72086a6817023d839" ], "author": { "name": "Bryan O\u0027Donoghue", "email": "bryan.odonoghue@linaro.org", "time": "Thu May 07 08:49:18 2020 +0800" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Thu May 07 08:46:35 2020 +0200" }, "message": "usb: chipidea: msm: Ensure proper controller reset using role switch API\n\nCurrently we check to make sure there is no error state on the extcon\nhandle for VBUS when writing to the HS_PHY_GENCONFIG_2 register. When using\nthe USB role-switch API we still need to write to this register absent an\nextcon handle.\n\nThis patch makes the appropriate update to ensure the write happens if\nrole-switching is true.\n\nFixes: 05559f10ed79 (\"usb: chipidea: add role switch class support\")\nCc: stable \u003cstable@vger.kernel.org\u003e\nCc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nCc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nCc: linux-usb@vger.kernel.org\nCc: linux-kernel@vger.kernel.org\nCc: Stephen Boyd \u003cswboyd@chromium.org\u003e\nSigned-off-by: Bryan O\u0027Donoghue \u003cbryan.odonoghue@linaro.org\u003e\nSigned-off-by: Peter Chen \u003cpeter.chen@nxp.com\u003e\nLink: https://lore.kernel.org/r/20200507004918.25975-2-peter.chen@kernel.org\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" }, { "commit": "a811c1fa0a02c062555b54651065899437bacdbe", "tree": "5c166695aa39019519ac5c9fffd96eed83738f50", "parents": [ "b9388959ba507c7a48ac18e4aa3b63b8a910ed99", "16f8036086a929694c3c62f577bb5925fe4fd607" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 06 20:53:22 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 06 20:53:22 2020 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n\n 1) Fix reference count leaks in various parts of batman-adv, from Xiyu\n Yang.\n\n 2) Update NAT checksum even when it is zero, from Guillaume Nault.\n\n 3) sk_psock reference count leak in tls code, also from Xiyu Yang.\n\n 4) Sanity check TCA_FQ_CODEL_DROP_BATCH_SIZE netlink attribute in\n fq_codel, from Eric Dumazet.\n\n 5) Fix panic in choke_reset(), also from Eric Dumazet.\n\n 6) Fix VLAN accel handling in bnxt_fix_features(), from Michael Chan.\n\n 7) Disallow out of range quantum values in sch_sfq, from Eric Dumazet.\n\n 8) Fix crash in x25_disconnect(), from Yue Haibing.\n\n 9) Don\u0027t pass pointer to local variable back to the caller in\n nf_osf_hdr_ctx_init(), from Arnd Bergmann.\n\n10) Wireguard should use the ECN decap helper functions, from Toke\n Høiland-Jørgensen.\n\n11) Fix command entry leak in mlx5 driver, from Moshe Shemesh.\n\n12) Fix uninitialized variable access in mptcp\u0027s\n subflow_syn_recv_sock(), from Paolo Abeni.\n\n13) Fix unnecessary out-of-order ingress frame ordering in macsec, from\n Scott Dial.\n\n14) IPv6 needs to use a global serial number for dst validation just\n like ipv4, from David Ahern.\n\n15) Fix up PTP_1588_CLOCK deps, from Clay McClure.\n\n16) Missing NLM_F_MULTI flag in gtp driver netlink messages, from\n Yoshiyuki Kurauchi.\n\n17) Fix a regression in that dsa user port errors should not be fatal,\n from Florian Fainelli.\n\n18) Fix iomap leak in enetc driver, from Dejin Zheng.\n\n19) Fix use after free in lec_arp_clear_vccs(), from Cong Wang.\n\n20) Initialize protocol value earlier in neigh code paths when\n generating events, from Roman Mashak.\n\n21) netdev_update_features() must be called with RTNL mutex in macsec\n driver, from Antoine Tenart.\n\n22) Validate untrusted GSO packets even more strictly, from Willem de\n Bruijn.\n\n23) Wireguard decrypt worker needs a cond_resched(), from Jason\n Donenfeld.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (111 commits)\n net: flow_offload: skip hw stats check for FLOW_ACTION_HW_STATS_DONT_CARE\n MAINTAINERS: put DYNAMIC INTERRUPT MODERATION in proper order\n wireguard: send/receive: use explicit unlikely branch instead of implicit coalescing\n wireguard: selftests: initalize ipv6 members to NULL to squelch clang warning\n wireguard: send/receive: cond_resched() when processing worker ringbuffers\n wireguard: socket: remove errant restriction on looping to self\n wireguard: selftests: use normal kernel stack size on ppc64\n net: ethernet: ti: am65-cpsw-nuss: fix irqs type\n ionic: Use debugfs_create_bool() to export bool\n net: dsa: Do not leave DSA master with NULL netdev_ops\n net: dsa: remove duplicate assignment in dsa_slave_add_cls_matchall_mirred\n net: stricter validation of untrusted gso packets\n seg6: fix SRH processing to comply with RFC8754\n net: mscc: ocelot: ANA_AUTOAGE_AGE_PERIOD holds a value in seconds, not ms\n net: dsa: ocelot: the MAC table on Felix is twice as large\n net: dsa: sja1105: the PTP_CLK extts input reacts on both edges\n selftests: net: tcp_mmap: fix SO_RCVLOWAT setting\n net: hsr: fix incorrect type usage for protocol variable\n net: macsec: fix rtnl locking issue\n net: mvpp2: cls: Prevent buffer overflow in mvpp2_ethtool_cls_rule_del()\n ...\n" }, { "commit": "16f8036086a929694c3c62f577bb5925fe4fd607", "tree": "23ecfc61c1efb40e0f8fb93993cd3e1a6d870b19", "parents": [ "b0956956bbd7b8c93112269a1e9c8309b0ceb0e7" ], "author": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Wed May 06 20:34:50 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:13:10 2020 -0700" }, "message": "net: flow_offload: skip hw stats check for FLOW_ACTION_HW_STATS_DONT_CARE\n\nThis patch adds FLOW_ACTION_HW_STATS_DONT_CARE which tells the driver\nthat the frontend does not need counters, this hw stats type request\nnever fails. The FLOW_ACTION_HW_STATS_DISABLED type explicitly requests\nthe driver to disable the stats, however, if the driver cannot disable\ncounters, it bails out.\n\nTCA_ACT_HW_STATS_* maintains the 1:1 mapping with FLOW_ACTION_HW_STATS_*\nexcept by disabled which is mapped to FLOW_ACTION_HW_STATS_DISABLED\n(this is 0 in tc). Add tc_act_hw_stats() to perform the mapping between\nTCA_ACT_HW_STATS_* and FLOW_ACTION_HW_STATS_*.\n\nFixes: 319a1d19471e (\"flow_offload: check for basic action hw stats type\")\nSigned-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b0956956bbd7b8c93112269a1e9c8309b0ceb0e7", "tree": "2e4c8055a206110e3b3775f8589befaa620d1017", "parents": [ "d3f3e6acb26b171e4572aaaafc7d2e918b35be35" ], "author": { "name": "Lukas Bulwahn", "email": "lukas.bulwahn@gmail.com", "time": "Wed May 06 22:29:06 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:12:33 2020 -0700" }, "message": "MAINTAINERS: put DYNAMIC INTERRUPT MODERATION in proper order\n\nCommit 9b038086f06b (\"docs: networking: convert DIM to RST\") added a new\nfile entry to DYNAMIC INTERRUPT MODERATION to the end, and not following\nalphabetical order.\n\nSo, ./scripts/checkpatch.pl -f MAINTAINERS complains:\n\n WARNING: Misordered MAINTAINERS entry - list file patterns in alphabetic\n order\n #5966: FILE: MAINTAINERS:5966:\n +F: lib/dim/\n +F: Documentation/networking/net_dim.rst\n\nReorder the file entries to keep MAINTAINERS nicely ordered.\n\nSigned-off-by: Lukas Bulwahn \u003clukas.bulwahn@gmail.com\u003e\nAcked-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d3f3e6acb26b171e4572aaaafc7d2e918b35be35", "tree": "6313edda1c8e183fe94a4a6832c4129568baa659", "parents": [ "6f5c27f9c6f89226f5b97476797e5abedd61912c", "243f2148937adc72bcaaa590d482d599c936efde" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:48 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:48 2020 -0700" }, "message": "Merge branch \u0027wireguard-fixes\u0027\n\nJason A. Donenfeld says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nwireguard fixes for 5.7-rc5\n\nWith Ubuntu and Debian having backported this into their kernels, we\u0027re\nfinally seeing testing from places we hadn\u0027t seen prior, which is nice.\nWith that comes more fixes:\n\n1) The CI for PPC64 was running with extremely small stacks for 64-bit,\n causing spurious crashes in surprising places.\n\n2) There\u0027s was an old leftover routing loop restriction, which no longer\n makes sense given the queueing architecture, and was causing problems\n for people who really did want nested routing.\n\n3) Not yielding our kthread on CONFIG_PREEMPT_VOLUNTARY systems caused\n RCU stalls and other issues, reported by Wang Jian, with the fix\n suggested by Sultan Alsawaf.\n\n4) Clang spewed warnings in a selftest for CONFIG_IPV6\u003dn, reported by\n Arnd Bergmann.\n\n5) A complicated if statement was simplified to an assignment while also\n making the likely/unlikely hinting more correct and simple, and\n increasing readability, suggested by Sultan.\n\nPatches (2) and (3) have Fixes: lines and are probably good candidates\nfor stable.\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": "243f2148937adc72bcaaa590d482d599c936efde", "tree": "6313edda1c8e183fe94a4a6832c4129568baa659", "parents": [ "4fed818ef54b08d4b29200e416cce65546ad5312" ], "author": { "name": "Jason A. Donenfeld", "email": "Jason@zx2c4.com", "time": "Wed May 06 15:33:06 2020 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:47 2020 -0700" }, "message": "wireguard: send/receive: use explicit unlikely branch instead of implicit coalescing\n\nIt\u0027s very unlikely that send will become true. It\u0027s nearly always false\nbetween 0 and 120 seconds of a session, and in most cases becomes true\nonly between 120 and 121 seconds before becoming false again. So,\nunlikely(send) is clearly the right option here.\n\nWhat happened before was that we had this complex boolean expression\nwith multiple likely and unlikely clauses nested. Since this is\nevaluated left-to-right anyway, the whole thing got converted to\nunlikely. So, we can clean this up to better represent what\u0027s going on.\n\nThe generated code is the same.\n\nSuggested-by: Sultan Alsawaf \u003csultan@kerneltoast.com\u003e\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4fed818ef54b08d4b29200e416cce65546ad5312", "tree": "dfa01477390797c40b3e09145ee1d4278b18e191", "parents": [ "4005f5c3c9d006157ba716594e0d70c88a235c5e" ], "author": { "name": "Jason A. Donenfeld", "email": "Jason@zx2c4.com", "time": "Wed May 06 15:33:05 2020 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:47 2020 -0700" }, "message": "wireguard: selftests: initalize ipv6 members to NULL to squelch clang warning\n\nWithout setting these to NULL, clang complains in certain\nconfigurations that have CONFIG_IPV6\u003dn:\n\nIn file included from drivers/net/wireguard/ratelimiter.c:223:\ndrivers/net/wireguard/selftest/ratelimiter.c:173:34: error: variable \u0027skb6\u0027 is uninitialized when used here [-Werror,-Wuninitialized]\n ret \u003d timings_test(skb4, hdr4, skb6, hdr6, \u0026test_count);\n ^~~~\ndrivers/net/wireguard/selftest/ratelimiter.c:123:29: note: initialize the variable \u0027skb6\u0027 to silence this warning\n struct sk_buff *skb4, *skb6;\n ^\n \u003d NULL\ndrivers/net/wireguard/selftest/ratelimiter.c:173:40: error: variable \u0027hdr6\u0027 is uninitialized when used here [-Werror,-Wuninitialized]\n ret \u003d timings_test(skb4, hdr4, skb6, hdr6, \u0026test_count);\n ^~~~\ndrivers/net/wireguard/selftest/ratelimiter.c:125:22: note: initialize the variable \u0027hdr6\u0027 to silence this warning\n struct ipv6hdr *hdr6;\n ^\n\nWe silence this warning by setting the variables to NULL as the warning\nsuggests.\n\nReported-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4005f5c3c9d006157ba716594e0d70c88a235c5e", "tree": "d275974f4516371281eb5d8dc1b508a02b2e242e", "parents": [ "b673e24aad36981f327a6570412ffa7754de8911" ], "author": { "name": "Jason A. Donenfeld", "email": "Jason@zx2c4.com", "time": "Wed May 06 15:33:04 2020 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:47 2020 -0700" }, "message": "wireguard: send/receive: cond_resched() when processing worker ringbuffers\n\nUsers with pathological hardware reported CPU stalls on CONFIG_\nPREEMPT_VOLUNTARY\u003dy, because the ringbuffers would stay full, meaning\nthese workers would never terminate. That turned out not to be okay on\nsystems without forced preemption, which Sultan observed. This commit\nadds a cond_resched() to the bottom of each loop iteration, so that\nthese workers don\u0027t hog the core. Note that we don\u0027t need this on the\nnapi poll worker, since that terminates after its budget is expended.\n\nSuggested-by: Sultan Alsawaf \u003csultan@kerneltoast.com\u003e\nReported-by: Wang Jian \u003clarkwang@gmail.com\u003e\nFixes: e7096c131e51 (\"net: WireGuard secure network tunnel\")\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b673e24aad36981f327a6570412ffa7754de8911", "tree": "7df3be82da0cc598286970cddbb764ae46fc0f4c", "parents": [ "a0fd7cc87a018df1a17f9d3f0bd994c1f22c6b34" ], "author": { "name": "Jason A. Donenfeld", "email": "Jason@zx2c4.com", "time": "Wed May 06 15:33:03 2020 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:47 2020 -0700" }, "message": "wireguard: socket: remove errant restriction on looping to self\n\nIt\u0027s already possible to create two different interfaces and loop\npackets between them. This has always been possible with tunnels in the\nkernel, and isn\u0027t specific to wireguard. Therefore, the networking stack\nalready needs to deal with that. At the very least, the packet winds up\nexceeding the MTU and is discarded at that point. So, since this is\nalready something that happens, there\u0027s no need to forbid the not very\nexceptional case of routing a packet back to the same interface; this\nloop is no different than others, and we shouldn\u0027t special case it, but\nrather rely on generic handling of loops in general. This also makes it\neasier to do interesting things with wireguard such as onion routing.\n\nAt the same time, we add a selftest for this, ensuring that both onion\nrouting works and infinite routing loops do not crash the kernel. We\nalso add a test case for wireguard interfaces nesting packets and\nsending traffic between each other, as well as the loop in this case\ntoo. We make sure to send some throughput-heavy traffic for this use\ncase, to stress out any possible recursion issues with the locks around\nworkqueues.\n\nFixes: e7096c131e51 (\"net: WireGuard secure network tunnel\")\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a0fd7cc87a018df1a17f9d3f0bd994c1f22c6b34", "tree": "37c518977817f9c4d9abbac924e3846ad59d9acf", "parents": [ "6f5c27f9c6f89226f5b97476797e5abedd61912c" ], "author": { "name": "Jason A. Donenfeld", "email": "Jason@zx2c4.com", "time": "Wed May 06 15:33:02 2020 -0600" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 20:03:47 2020 -0700" }, "message": "wireguard: selftests: use normal kernel stack size on ppc64\n\nWhile at some point it might have made sense to be running these tests\non ppc64 with 4k stacks, the kernel hasn\u0027t actually used 4k stacks on\n64-bit powerpc in a long time, and more interesting things that we test\ndon\u0027t really work when we deviate from the default (16k). So, we stop\npushing our luck in this commit, and return to the default instead of\nthe minimum.\n\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6f5c27f9c6f89226f5b97476797e5abedd61912c", "tree": "2c83bbab3ca9a60c06fe66d09544ccd03d3b0312", "parents": [ "0735ccc9d9a4ce09154395feb70b804de7cc6ebf" ], "author": { "name": "Grygorii Strashko", "email": "grygorii.strashko@ti.com", "time": "Tue May 05 19:31:26 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:53:00 2020 -0700" }, "message": "net: ethernet: ti: am65-cpsw-nuss: fix irqs type\n\nThe K3 INTA driver, which is source TX/RX IRQs for CPSW NUSS, defines IRQs\ntriggering type as EDGE by default, but triggering type for CPSW NUSS TX/RX\nIRQs has to be LEVEL as the EDGE triggering type may cause unnecessary IRQs\ntriggering and NAPI scheduling for empty queues. It was discovered with\nRT-kernel.\n\nFix it by explicitly specifying CPSW NUSS TX/RX IRQ type as\nIRQF_TRIGGER_HIGH.\n\nFixes: 93a76530316a (\"net: ethernet: ti: introduce am65x/j721e gigabit eth subsystem driver\")\nSigned-off-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0735ccc9d9a4ce09154395feb70b804de7cc6ebf", "tree": "f4e4ca1a451ac7fb416c120e2a8142305fe79de9", "parents": [ "050569fc8384c8056bacefcc246bcb2dfe574936" ], "author": { "name": "Geert Uytterhoeven", "email": "geert+renesas@glider.be", "time": "Tue May 05 15:28:09 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:46:33 2020 -0700" }, "message": "ionic: Use debugfs_create_bool() to export bool\n\nCurrently bool ionic_cq.done_color is exported using\ndebugfs_create_u8(), which requires a cast, preventing further compiler\nchecks.\n\nFix this by switching to debugfs_create_bool(), and dropping the cast.\n\nSigned-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nAcked-by: Shannon Nelson \u003csnelson@pensando.io\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "050569fc8384c8056bacefcc246bcb2dfe574936", "tree": "6b4fd984dc23bfc5a0d090b8194db878fc67f8c7", "parents": [ "657221598f820ff860972b67583fb91d9ab7caf4" ], "author": { "name": "Florian Fainelli", "email": "f.fainelli@gmail.com", "time": "Mon May 04 13:18:06 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:31:54 2020 -0700" }, "message": "net: dsa: Do not leave DSA master with NULL netdev_ops\n\nWhen ndo_get_phys_port_name() for the CPU port was added we introduced\nan early check for when the DSA master network device in\ndsa_master_ndo_setup() already implements ndo_get_phys_port_name(). When\nwe perform the teardown operation in dsa_master_ndo_teardown() we would\nnot be checking that cpu_dp-\u003eorig_ndo_ops was successfully allocated and\nnon-NULL initialized.\n\nWith network device drivers such as virtio_net, this leads to a NPD as\nsoon as the DSA switch hanging off of it gets torn down because we are\nnow assigning the virtio_net device\u0027s netdev_ops a NULL pointer.\n\nFixes: da7b9e9b00d4 (\"net: dsa: Add ndo_get_phys_port_name() for CPU port\")\nReported-by: Allen Pais \u003callen.pais@oracle.com\u003e\nSigned-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nTested-by: Allen Pais \u003callen.pais@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "657221598f820ff860972b67583fb91d9ab7caf4", "tree": "4dee26f1fb5be50f51a9d85d8acb280814e81740", "parents": [ "9274124f023b5c56dc4326637d4f787968b03607" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Mon May 04 22:58:56 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:30:35 2020 -0700" }, "message": "net: dsa: remove duplicate assignment in dsa_slave_add_cls_matchall_mirred\n\nThis was caused by a poor merge conflict resolution on my side. The\n\"act \u003d \u0026cls-\u003erule-\u003eaction.entries[0];\" assignment was already present in\nthe code prior to the patch mentioned below.\n\nFixes: e13c2075280e (\"net: dsa: refactor matchall mirred action to separate function\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9274124f023b5c56dc4326637d4f787968b03607", "tree": "0f46c00bd05149b9747ab50c1e290a56fb7b5ee6", "parents": [ "0cb7498f234e4e7d75187a8cae6c7c2053f2488a" ], "author": { "name": "Willem de Bruijn", "email": "willemb@google.com", "time": "Mon May 04 12:48:54 2020 -0400" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:23:06 2020 -0700" }, "message": "net: stricter validation of untrusted gso packets\n\nSyzkaller again found a path to a kernel crash through bad gso input:\na packet with transport header extending beyond skb_headlen(skb).\n\nTighten validation at kernel entry:\n\n- Verify that the transport header lies within the linear section.\n\n To avoid pulling linux/tcp.h, verify just sizeof tcphdr.\n tcp_gso_segment will call pskb_may_pull (th-\u003edoff * 4) before use.\n\n- Match the gso_type against the ip_proto found by the flow dissector.\n\nFixes: bfd5f4a3d605 (\"packet: Add GSO/csum offload support.\")\nReported-by: syzbot \u003csyzkaller@googlegroups.com\u003e\nSigned-off-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0cb7498f234e4e7d75187a8cae6c7c2053f2488a", "tree": "5afb7fe4c736967ca650195f9541ca9508ad9dc9", "parents": [ "6e0ddb6530b5010b33f91f0939029d2819b2dc3f" ], "author": { "name": "Ahmed Abdelsalam", "email": "ahabdels@gmail.com", "time": "Mon May 04 14:42:11 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:21:35 2020 -0700" }, "message": "seg6: fix SRH processing to comply with RFC8754\n\nThe Segment Routing Header (SRH) which defines the SRv6 dataplane is defined\nin RFC8754.\n\nRFC8754 (section 4.1) defines the SR source node behavior which encapsulates\npackets into an outer IPv6 header and SRH. The SR source node encodes the\nfull list of Segments that defines the packet path in the SRH. Then, the\nfirst segment from list of Segments is copied into the Destination address\nof the outer IPv6 header and the packet is sent to the first hop in its path\ntowards the destination.\n\nIf the Segment list has only one segment, the SR source node can omit the SRH\nas he only segment is added in the destination address.\n\nRFC8754 (section 4.1.1) defines the Reduced SRH, when a source does not\nrequire the entire SID list to be preserved in the SRH. A reduced SRH does\nnot contain the first segment of the related SR Policy (the first segment is\nthe one already in the DA of the IPv6 header), and the Last Entry field is\nset to n-2, where n is the number of elements in the SR Policy.\n\nRFC8754 (section 4.3.1.1) defines the SRH processing and the logic to\nvalidate the SRH (S09, S10, S11) which works for both reduced and\nnon-reduced behaviors.\n\nThis patch updates seg6_validate_srh() to validate the SRH as per RFC8754.\n\nSigned-off-by: Ahmed Abdelsalam \u003cahabdels@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6e0ddb6530b5010b33f91f0939029d2819b2dc3f", "tree": "501d3327e2d77a45a76ed54c0bb84caf9aaa8d4f", "parents": [ "0ba83aa037da456bb73868e5cc20558c5a644414", "c0d7eccbc76115b7eb337956c03d47d6a889cf8c" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:15:38 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:15:38 2020 -0700" }, "message": "Merge branch \u0027FDB-fixes-for-Felix-and-Ocelot-switches\u0027\n\nVladimir Oltean says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nFDB fixes for Felix and Ocelot switches\n\nThis series fixes the following problems:\n- Dynamically learnt addresses never expiring (neither for Ocelot nor\n for Felix)\n- Half of the FDB not visible in \u0027bridge fdb show\u0027 (for Felix only)\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": "c0d7eccbc76115b7eb337956c03d47d6a889cf8c", "tree": "501d3327e2d77a45a76ed54c0bb84caf9aaa8d4f", "parents": [ "21ce7f3e16fbf89faaf149cfe0f730edfc553914" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Mon May 04 01:20:27 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:15:25 2020 -0700" }, "message": "net: mscc: ocelot: ANA_AUTOAGE_AGE_PERIOD holds a value in seconds, not ms\n\nOne may notice that automatically-learnt entries \u0027never\u0027 expire, even\nthough the bridge configures the address age period at 300 seconds.\n\nActually the value written to hardware corresponds to a time interval\n1000 times higher than intended, i.e. 83 hours.\n\nFixes: a556c76adc05 (\"net: mscc: Add initial Ocelot switch support\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nReviewed-by: Florian Faineli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "21ce7f3e16fbf89faaf149cfe0f730edfc553914", "tree": "98fe33ca24e78f974bc88b13fa8c3f6d3a56278d", "parents": [ "0ba83aa037da456bb73868e5cc20558c5a644414" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Mon May 04 01:20:26 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 17:15:24 2020 -0700" }, "message": "net: dsa: ocelot: the MAC table on Felix is twice as large\n\nWhen running \u0027bridge fdb dump\u0027 on Felix, sometimes learnt and static MAC\naddresses would appear, sometimes they wouldn\u0027t.\n\nTurns out, the MAC table has 4096 entries on VSC7514 (Ocelot) and 8192\nentries on VSC9959 (Felix), so the existing code from the Ocelot common\nlibrary only dumped half of Felix\u0027s MAC table. They are both organized\nas a 4-way set-associative TCAM, so we just need a single variable\nindicating the correct number of rows.\n\nFixes: 56051948773e (\"net: dsa: ocelot: add driver for Felix switch family\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nReviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b9388959ba507c7a48ac18e4aa3b63b8a910ed99", "tree": "3cc8a186c187dfee2c15d941d4cb099aa4c1f39b", "parents": [ "3c40cdb0e93ec166f1fa4fee1eb62d45b5451515", "b31d1d2b1c3a8452f425b09ebd374ecd3ddd5179" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 06 16:40:14 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 06 16:40:14 2020 -0700" }, "message": "Merge tag \u0027tag-chrome-platform-fixes-for-v5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chrome-platform/linux\n\nPull chrome platform fix from Benson Leung:\n \"Fix a resource allocation issue in cros_ec_sensorhub.c\"\n\n* tag \u0027tag-chrome-platform-fixes-for-v5.7-rc5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chrome-platform/linux:\n platform/chrome: cros_ec_sensorhub: Allocate sensorhub resource before claiming sensors\n" }, { "commit": "0ba83aa037da456bb73868e5cc20558c5a644414", "tree": "cb40fdf38cb7eb8d87f2bba4686946a83a7ef4f3", "parents": [ "a84724178bd7081cf3bd5b558616dd6a9a4ca63b" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Wed May 06 20:48:13 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 15:03:14 2020 -0700" }, "message": "net: dsa: sja1105: the PTP_CLK extts input reacts on both edges\n\nIt looks like the sja1105 external timestamping input is not as generic\nas we thought. When fed a signal with 50% duty cycle, it will timestamp\nboth the rising and the falling edge. When fed a short pulse signal,\nonly the timestamp of the falling edge will be seen in the PTPSYNCTS\nregister, because that of the rising edge had been overwritten. So the\nmoral is: don\u0027t feed it short pulse inputs.\n\nLuckily this is not a complete deal breaker, as we can still work with\n1 Hz square waves. But the problem is that the extts polling period was\nnot dimensioned enough for this input signal. If we leave the period at\nhalf a second, we risk losing timestamps due to jitter in the measuring\nprocess. So we need to increase it to 4 times per second.\n\nAlso, the very least we can do to inform the user is to deny any other\nflags combination than with PTP_RISING_EDGE and PTP_FALLING_EDGE both\nset.\n\nFixes: 747e5eb31d59 (\"net: dsa: sja1105: configure the PTP_CLK pin as EXT_TS or PER_OUT\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nAcked-by: Richard Cochran \u003crichardcochran@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a84724178bd7081cf3bd5b558616dd6a9a4ca63b", "tree": "785e03e9831197f6be9ae736575127951ee83fb7", "parents": [ "f5dda315b63775fe00f5470687d40ce801ddfb38" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Wed May 06 09:21:15 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 15:01:30 2020 -0700" }, "message": "selftests: net: tcp_mmap: fix SO_RCVLOWAT setting\n\nSince chunk_size is no longer an integer, we can not\nuse it directly as an argument of setsockopt().\n\nThis patch should fix tcp_mmap for Big Endian kernels.\n\nFixes: 597b01edafac (\"selftests: net: avoid ptl lock contention in tcp_mmap\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\nCc: Arjun Roy \u003carjunroy@google.com\u003e\nAcked-by: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f5dda315b63775fe00f5470687d40ce801ddfb38", "tree": "8a5571238db196bf984016acf4726e720ee5369d", "parents": [ "29ca3cdfe13b2792b8624e6f769777e8cb387f9c" ], "author": { "name": "Murali Karicheri", "email": "m-karicheri2@ti.com", "time": "Wed May 06 11:41:07 2020 -0400" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 15:00:20 2020 -0700" }, "message": "net: hsr: fix incorrect type usage for protocol variable\n\nFix following sparse checker warning:-\n\nnet/hsr/hsr_slave.c:38:18: warning: incorrect type in assignment (different base types)\nnet/hsr/hsr_slave.c:38:18: expected unsigned short [unsigned] [usertype] protocol\nnet/hsr/hsr_slave.c:38:18: got restricted __be16 [usertype] h_proto\nnet/hsr/hsr_slave.c:39:25: warning: restricted __be16 degrades to integer\nnet/hsr/hsr_slave.c:39:57: warning: restricted __be16 degrades to integer\n\nSigned-off-by: Murali Karicheri \u003cm-karicheri2@ti.com\u003e\nAcked-by: Vinicius Costa Gomes \u003cvinicius.gomes@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "29ca3cdfe13b2792b8624e6f769777e8cb387f9c", "tree": "dad89c90cf9a6cab6b2243e9566dcc55964c3738", "parents": [ "722c0f00d4feea77475a5dc943b53d60824a1e4e" ], "author": { "name": "Antoine Tenart", "email": "antoine.tenart@bootlin.com", "time": "Wed May 06 15:58:30 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 14:34:38 2020 -0700" }, "message": "net: macsec: fix rtnl locking issue\n\nnetdev_update_features() must be called with the rtnl lock taken. Not\ndoing so triggers a warning, as ASSERT_RTNL() is used in\n__netdev_update_features(), the first function called by\nnetdev_update_features(). Fix this.\n\nFixes: c850240b6c41 (\"net: macsec: report real_dev features when HW offloading is enabled\")\nSigned-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "722c0f00d4feea77475a5dc943b53d60824a1e4e", "tree": "6b4b721795d1cf58cb0986efcd70af85e3c61451", "parents": [ "39bd16df7c31bb8cf5dfd0c88e42abd5ae10029d" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Wed May 06 13:16:56 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 14:15:46 2020 -0700" }, "message": "net: mvpp2: cls: Prevent buffer overflow in mvpp2_ethtool_cls_rule_del()\n\nThe \"info-\u003efs.location\" is a u32 that comes from the user via the\nethtool_set_rxnfc() function. We need to check for invalid values to\nprevent a buffer overflow.\n\nI copy and pasted this check from the mvpp2_ethtool_cls_rule_ins()\nfunction.\n\nFixes: 90b509b39ac9 (\"net: mvpp2: cls: Add Classification offload support\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "39bd16df7c31bb8cf5dfd0c88e42abd5ae10029d", "tree": "f0d1b06ad51170facfc101d6991082993fa55549", "parents": [ "bf5525f3a8e3248be5aa5defe5aaadd60e1c1ba1" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Wed May 06 13:16:22 2020 +0300" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 14:15:46 2020 -0700" }, "message": "net: mvpp2: prevent buffer overflow in mvpp22_rss_ctx()\n\nThe \"rss_context\" variable comes from the user via ethtool_get_rxfh().\nIt can be any u32 value except zero. Eventually it gets passed to\nmvpp22_rss_ctx() and if it is over MVPP22_N_RSS_TABLES (8) then it\nresults in an array overflow.\n\nFixes: 895586d5dc32 (\"net: mvpp2: cls: Use RSS contexts to handle RSS tables\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bf5525f3a8e3248be5aa5defe5aaadd60e1c1ba1", "tree": "af4918566cfaa90205362758a7ef9ba3b0c89626", "parents": [ "38212bb31fe923d0a2c6299bd2adfbb84cddef2a" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue May 05 20:51:06 2020 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 06 13:45:09 2020 -0700" }, "message": "selftests: net: tcp_mmap: clear whole tcp_zerocopy_receive struct\n\nWe added fields in tcp_zerocopy_receive structure,\nso make sure to clear all fields to not pass garbage to the kernel.\n\nWe were lucky because recent additions added \u0027out\u0027 parameters,\nstill we need to clean our reference implementation, before folks\ncopy/paste it.\n\nFixes: c8856c051454 (\"tcp-zerocopy: Return inq along with tcp receive zerocopy.\")\nFixes: 33946518d493 (\"tcp-zerocopy: Return sk_err (if set) along with tcp receive zerocopy.\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nCc: Arjun Roy \u003carjunroy@google.com\u003e\nCc: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\nAcked-by: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3c40cdb0e93ec166f1fa4fee1eb62d45b5451515", "tree": "17429ac34d2f544b38317d48793e4536adc51687", "parents": [ "dc56c5acd8505e1e7f776d62650f3850ad2ce8e7", "a9a8ba90fa5857c2c8a0e32eef2159cec717da11" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 06 10:20:00 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed May 06 10:20:00 2020 -0700" }, "message": "Merge branch \u0027linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6\n\nPull crypto fixes from Herbert Xu:\n \"This fixes a potential scheduling latency problem for the algorithms\n used by WireGuard\"\n\n* \u0027linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:\n crypto: arch/nhpoly1305 - process in explicit 4k chunks\n crypto: arch/lib - limit simd usage to 4k chunks\n" }, { "commit": "084d7e78467534eedaf0d6c72086a6817023d839", "tree": "54588155e38a74dc19d5913740a15d098dcdcda5", "parents": [ "7990be48ef4d87163940d6c04c349c93f0bd9ae7", "78d6de3cfbd342918d31cf68d0d2eda401338aef" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Wed May 06 17:26:35 2020 +0200" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Wed May 06 17:26:35 2020 +0200" }, "message": "Merge tag \u0027usb-serial-5.7-rc5\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial into usb-linus\n\nJohan writes:\n\nUSB-serial fixes for 5.7-rc5\n\nHere\u0027s a fix adding a missing input sanity check and a new modem device\nid.\n\nBoth have been in linux-next with no reported issues.\n\n* tag \u0027usb-serial-5.7-rc5\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial:\n USB: serial: qcserial: Add DW5816e support\n USB: serial: garmin_gps: add sanity checking for data length\n" }, { "commit": "5b4dcd2d201a395ad4054067bfae4a07554fbd65", "tree": "16296e48bb9730866ab354078efdcc478ef5eb61", "parents": [ "da0f1f4167e3af69e1d8b32d6d65195ddd2bfb64" ], "author": { "name": "Masami Hiramatsu", "email": "mhiramat@kernel.org", "time": "Sat Apr 25 14:49:26 2020 +0900" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed May 06 09:04:11 2020 -0400" }, "message": "tracing/kprobes: Reject new event if loc is NULL\n\nReject the new event which has NULL location for kprobes.\nFor kprobes, user must specify at least the location.\n\nLink: http://lkml.kernel.org/r/158779376597.6082.1411212055469099461.stgit@devnote2\n\nCc: Tom Zanussi \u003czanussi@kernel.org\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: stable@vger.kernel.org\nFixes: 2a588dd1d5d6 (\"tracing: Add kprobe event command generation functions\")\nSigned-off-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "da0f1f4167e3af69e1d8b32d6d65195ddd2bfb64", "tree": "3f3004ade391b9d890c55ab35942a949eb5471ed", "parents": [ "dcbd21c9fca5e954fd4e3d91884907eb6d47187e" ], "author": { "name": "Masami Hiramatsu", "email": "mhiramat@kernel.org", "time": "Sat Apr 25 14:49:17 2020 +0900" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed May 06 09:04:11 2020 -0400" }, "message": "tracing/boottime: Fix kprobe event API usage\n\nFix boottime kprobe events to use API correctly for\nmultiple events.\n\nFor example, when we set a multiprobe kprobe events in\nbootconfig like below,\n\n ftrace.event.kprobes.myevent {\n \tprobes \u003d \"vfs_read $arg1 $arg2\", \"vfs_write $arg1 $arg2\"\n }\n\nThis cause an error;\n\n trace_boot: Failed to add probe: p:kprobes/myevent (null) vfs_read $arg1 $arg2 vfs_write $arg1 $arg2\n\nThis shows the 1st argument becomes NULL and multiprobes\nare merged to 1 probe.\n\nLink: http://lkml.kernel.org/r/158779375766.6082.201939936008972838.stgit@devnote2\n\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: stable@vger.kernel.org\nFixes: 29a154810546 (\"tracing: Change trace_boot to use kprobe_event interface\")\nReviewed-by: Tom Zanussi \u003czanussi@kernel.org\u003e\nSigned-off-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "dcbd21c9fca5e954fd4e3d91884907eb6d47187e", "tree": "4a12e623b5097e57e90ded74fd1810d0d0b5bd27", "parents": [ "de462e5f10718517bacf2f84c8aa2804567ef7df" ], "author": { "name": "Masami Hiramatsu", "email": "mhiramat@kernel.org", "time": "Sat Apr 25 14:49:09 2020 +0900" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed May 06 09:04:11 2020 -0400" }, "message": "tracing/kprobes: Fix a double initialization typo\n\nFix a typo that resulted in an unnecessary double\ninitialization to addr.\n\nLink: http://lkml.kernel.org/r/158779374968.6082.2337484008464939919.stgit@devnote2\n\nCc: Tom Zanussi \u003czanussi@kernel.org\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: stable@vger.kernel.org\nFixes: c7411a1a126f (\"tracing/kprobe: Check whether the non-suffixed symbol is notrace\")\nSigned-off-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "de462e5f10718517bacf2f84c8aa2804567ef7df", "tree": "249437f9aee0e376efae8ee7831bfabbc95e81c1", "parents": [ "6a8b55ed4056ea5559ebe4f6a4b247f627870d4c" ], "author": { "name": "Masami Hiramatsu", "email": "mhiramat@kernel.org", "time": "Sun Apr 26 15:53:30 2020 +0900" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed May 06 09:04:11 2020 -0400" }, "message": "bootconfig: Fix to remove bootconfig data from initrd while boot\n\nIf there is a bootconfig data in the tail of initrd/initramfs,\ninitrd image sanity check caused an error while decompression\nstage as follows.\n\n[ 0.883882] Unpacking initramfs...\n[ 2.696429] Initramfs unpacking failed: invalid magic at start of compressed archive\n\nThis error will be ignored if CONFIG_BLK_DEV_RAM\u003dn,\nbut CONFIG_BLK_DEV_RAM\u003dy the kernel failed to mount rootfs\nand causes a panic.\n\nTo fix this issue, shrink down the initrd_end for removing\ntailing bootconfig data while boot the kernel.\n\nLink: http://lkml.kernel.org/r/158788401014.24243.17424755854115077915.stgit@devnote2\n\nCc: Borislav Petkov \u003cbp@alien8.de\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: stable@vger.kernel.org\nFixes: 7684b8582c24 (\"bootconfig: Load boot config from the tail of initrd\")\nSigned-off-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "2673cb6849722a4ffd74c27a9200a9ec43f64be3", "tree": "92318aa6661be398fb2a22ef91094f454fa72190", "parents": [ "495907ec36def1d28a44e2d1b5a51affe716aacf", "5615e74f48dcc982655543e979b6c3f3f877e6f6" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 08:09:17 2020 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 08:09:17 2020 -0400" }, "message": "Merge tag \u0027kvm-s390-master-5.7-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD\n\nKVM: s390: Fix for running nested uner z/VM\n\nThere are circumstances when running nested under z/VM that would trigger a\nWARN_ON_ONCE. Remove the WARN_ON_ONCE. Long term we certainly want to make this\ncode more robust and flexible, but just returning instead of WARNING makes\nguest bootable again.\n" }, { "commit": "495907ec36def1d28a44e2d1b5a51affe716aacf", "tree": "85e68a9c6ae3c373263ec9b9979f5bb28b8a74b9", "parents": [ "8ffdaf9155ebe517cdec5edbcca19ba6e7ee9c3c" ], "author": { "name": "Peter Xu", "email": "peterx@redhat.com", "time": "Tue May 05 11:47:50 2020 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 06:51:38 2020 -0400" }, "message": "KVM: X86: Declare KVM_CAP_SET_GUEST_DEBUG properly\n\nKVM_CAP_SET_GUEST_DEBUG should be supported for x86 however it\u0027s not declared\nas supported. My wild guess is that userspaces like QEMU are using \"#ifdef\nKVM_CAP_SET_GUEST_DEBUG\" to check for the capability instead, but that could be\nwrong because the compilation host may not be the runtime host.\n\nThe userspace might still want to keep the old \"#ifdef\" though to not break the\nguest debug on old kernels.\n\nSigned-off-by: Peter Xu \u003cpeterx@redhat.com\u003e\nMessage-Id: \u003c20200505154750.126300-1-peterx@redhat.com\u003e\n[Do the same for PPC and s390. - Paolo]\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "8ffdaf9155ebe517cdec5edbcca19ba6e7ee9c3c", "tree": "6361c701fbf5ddc3a7e13615bde59a5c07540b32", "parents": [ "139f7425fdf54f054463e7524b9f54c41af8407f" ], "author": { "name": "Peter Xu", "email": "peterx@redhat.com", "time": "Mon May 04 18:06:07 2020 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 06:51:36 2020 -0400" }, "message": "KVM: selftests: Fix build for evmcs.h\n\nI got this error when building kvm selftests:\n\n/usr/bin/ld: /home/xz/git/linux/tools/testing/selftests/kvm/libkvm.a(vmx.o):/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:222: multiple definition of `current_evmcs\u0027; /tmp/cco1G48P.o:/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:222: first defined here\n/usr/bin/ld: /home/xz/git/linux/tools/testing/selftests/kvm/libkvm.a(vmx.o):/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:223: multiple definition of `current_vp_assist\u0027; /tmp/cco1G48P.o:/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:223: first defined here\n\nI think it\u0027s because evmcs.h is included both in a test file and a lib file so\nthe structs have multiple declarations when linking. After all it\u0027s not a good\nhabit to declare structs in the header files.\n\nCc: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nSigned-off-by: Peter Xu \u003cpeterx@redhat.com\u003e\nMessage-Id: \u003c20200504220607.99627-1-peterx@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "139f7425fdf54f054463e7524b9f54c41af8407f", "tree": "c6a22176977632b38d79fe96aac1828070f53fb1", "parents": [ "c7cb2d650c9e78c03bd2d1c0db89891825f8c0f4" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue May 05 09:40:46 2020 -0400" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 06:51:36 2020 -0400" }, "message": "kvm: x86: Use KVM CPU capabilities to determine CR4 reserved bits\n\nUsing CPUID data can be useful for the processor compatibility\ncheck, but that\u0027s it. Using it to compute guest-reserved bits\ncan have both false positives (such as LA57 and UMIP which we\nare already handling) and false negatives: in particular, with\nthis patch we don\u0027t allow anymore a KVM guest to set CR4.PKE\nwhen CR4.PKE is clear on the host.\n\nFixes: b9dd21e104bc (\"KVM: x86: simplify handling of PKRU\")\nReported-by: Jim Mattson \u003cjmattson@google.com\u003e\nTested-by: Jim Mattson \u003cjmattson@google.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "c7cb2d650c9e78c03bd2d1c0db89891825f8c0f4", "tree": "b9997f467e84d76699e295b541a11cedf2563df0", "parents": [ "27abe57770ff1c4c23a60993816f02657ba94bd1" ], "author": { "name": "Sean Christopherson", "email": "sean.j.christopherson@intel.com", "time": "Tue May 05 20:53:55 2020 -0700" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 06:51:35 2020 -0400" }, "message": "KVM: VMX: Explicitly clear RFLAGS.CF and RFLAGS.ZF in VM-Exit RSB path\n\nClear CF and ZF in the VM-Exit path after doing __FILL_RETURN_BUFFER so\nthat KVM doesn\u0027t interpret clobbered RFLAGS as a VM-Fail. Filling the\nRSB has always clobbered RFLAGS, its current incarnation just happens\nclear CF and ZF in the processs. Relying on the macro to clear CF and\nZF is extremely fragile, e.g. commit 089dd8e53126e (\"x86/speculation:\nChange FILL_RETURN_BUFFER to work with objtool\") tweaks the loop such\nthat the ZF flag is always set.\n\nReported-by: Qian Cai \u003ccai@lca.pw\u003e\nCc: Rick Edgecombe \u003crick.p.edgecombe@intel.com\u003e\nCc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nCc: stable@vger.kernel.org\nFixes: f2fde6a5bcfcf (\"KVM: VMX: Move RSB stuffing to before the first RET after VM-Exit\")\nSigned-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nMessage-Id: \u003c20200506035355.2242-1-sean.j.christopherson@intel.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "27abe57770ff1c4c23a60993816f02657ba94bd1", "tree": "867daff50e73cc74a1b52130529071f891ef3e21", "parents": [ "8be8f932e3db5fe4ed178b8892eeffeab530273a" ], "author": { "name": "Kashyap Chamarthy", "email": "kchamart@redhat.com", "time": "Tue May 05 13:28:39 2020 +0200" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Wed May 06 05:45:47 2020 -0400" }, "message": "docs/virt/kvm: Document configuring and running nested guests\n\nThis is a rewrite of this[1] Wiki page with further enhancements. The\ndoc also includes a section on debugging problems in nested\nenvironments, among other improvements.\n\n[1] https://www.linux-kvm.org/page/Nested_Guests\n\nSigned-off-by: Kashyap Chamarthy \u003ckchamart@redhat.com\u003e\nMessage-Id: \u003c20200505112839.30534-1-kchamart@redhat.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "73cb8e2a5863ccc5215660f5123db621bd57dff7", "tree": "48e94f978d24d569cb2b918059105195c2132f58", "parents": [ "d6d5161280b3d57163f5310f0a0007cdeb729984" ], "author": { "name": "Atish Patra", "email": "atish.patra@wdc.com", "time": "Sun May 03 21:03:19 2020 -0700" }, "committer": { "name": "Palmer Dabbelt", "email": "palmerdabbelt@google.com", "time": "Tue May 05 17:02:14 2020 -0700" }, "message": "RISC-V: Remove unused code from STRICT_KERNEL_RWX\n\nThis patch removes the unused functions set_kernel_text_rw/ro.\nCurrently, it is not being invoked from anywhere and no other architecture\n(except arm) uses this code. Even in ARM, these functions are not invoked\nfrom anywhere currently.\n\nFixes: d27c3c90817e (\"riscv: add STRICT_KERNEL_RWX support\")\nSigned-off-by: Atish Patra \u003catish.patra@wdc.com\u003e\nReviewed-by: Zong Li \u003czong.li@sifive.com\u003e\nSigned-off-by: Palmer Dabbelt \u003cpalmerdabbelt@google.com\u003e\n" }, { "commit": "dc56c5acd8505e1e7f776d62650f3850ad2ce8e7", "tree": "12c1d4f63ffde24c264f8d5383440aaf987e8ddb", "parents": [ "47cf1b422e6093aee2a3e55d5e162112a2c69870", "f8a31eca47bec197f5ec5dc40ad675450c2058a5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 05 16:29:03 2020 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue May 05 16:29:03 2020 -0700" }, "message": "Merge tag \u0027platform-drivers-x86-v5.7-2\u0027 of git://git.infradead.org/linux-platform-drivers-x86\n\nPull x86 platform driver fixes from Andy Shevchenko:\n\n - Avoid loading asus-nb-wmi module on selected laptop models\n\n - Fix S0ix debug support for Jasper Lake PMC\n\n - Few fixes which have been reported by Hulk bot and others\n\n* tag \u0027platform-drivers-x86-v5.7-2\u0027 of git://git.infradead.org/linux-platform-drivers-x86:\n platform/x86: thinkpad_acpi: Remove always false \u0027value \u003c 0\u0027 statement\n platform/x86: intel_pmc_core: avoid unused-function warnings\n platform/x86: asus-nb-wmi: Do not load on Asus T100TA and T200TA\n platform/x86: intel_pmc_core: Change Jasper Lake S0ix debug reg map back to ICL\n platform/x86/intel-uncore-freq: make uncore_root_kobj static\n platform/x86: wmi: Make two functions static\n platform/x86: surface3_power: Fix a NULL vs IS_ERR() check in probe\n" }, { "commit": "38212bb31fe923d0a2c6299bd2adfbb84cddef2a", "tree": "5ee7a48a1599641bb542d1341f66c74c8bbe4e88", "parents": [ "755f5738ff981769211a0bfac709d514ef5b9f86" ], "author": { "name": "Roman Mashak", "email": "mrv@mojatatu.com", "time": "Fri May 01 21:34:18 2020 -0400" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 05 13:38:59 2020 -0700" }, "message": "neigh: send protocol value in neighbor create notification\n\nWhen a new neighbor entry has been added, event is generated but it does not\ninclude protocol, because its value is assigned after the event notification\nroutine has run, so move protocol assignment code earlier.\n\nFixes: df9b0e30d44c (\"neighbor: Add protocol attribute\")\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nSigned-off-by: Roman Mashak \u003cmrv@mojatatu.com\u003e\nReviewed-by: David Ahern \u003cdsahern@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e6142dd511425cb827b5db869f489eb81f5f994d", "tree": "26a90a317cb409e6d1a630d40713156caa89927c", "parents": [ "80797dd6f1a525d1160c463d6a9f9d29af182cbb" ], "author": { "name": "Aurabindo Pillai", "email": "aurabindo.pillai@amd.com", "time": "Wed Apr 22 14:37:33 2020 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue May 05 16:13:57 2020 -0400" }, "message": "drm/amd/display: Prevent dpcd reads with passive dongles\n\n[why]\nDuring hotplug, a DP port may be connected to the sink through\npassive adapter which does not support DPCD reads. Issuing reads\nwithout checking for this condition will result in errors\n\n[how]\nEnsure the link is in aux_mode before initiating operation that result\nin a DPCD read.\n\nSigned-off-by: Aurabindo Pillai \u003caurabindo.pillai@amd.com\u003e\nReviewed-by: Harry Wentland \u003cHarry.Wentland@amd.com\u003e\nAcked-by: Aurabindo Pillai \u003caurabindo.pillai@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "80797dd6f1a525d1160c463d6a9f9d29af182cbb", "tree": "7b0bd90315cd133c0ca533945eaf185b44e236ab", "parents": [ "b95e51eb9f2ee7b6d6c3203a2f75122349aa77be" ], "author": { "name": "Roman Li", "email": "roman.li@amd.com", "time": "Wed Feb 26 17:30:29 2020 -0500" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue May 05 16:13:15 2020 -0400" }, "message": "drm/amd/display: fix counter in wait_for_no_pipes_pending\n\n[Why]\nWait counter is not being reset for each pipe.\n\n[How]\nMove counter reset into pipe loop scope.\n\nSigned-off-by: Roman Li \u003croman.li@amd.com\u003e\nReviewed-by: Zhan Liu \u003cZhan.Liu@amd.com\u003e\nAcked-by: Aurabindo Pillai \u003caurabindo.pillai@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "b95e51eb9f2ee7b6d6c3203a2f75122349aa77be", "tree": "d16fede3e3ea59227ebaeb605af7b4cf178d4953", "parents": [ "59dfb0c64d3853d20dc84f4561f28d4f5a2ddc7d" ], "author": { "name": "Sung Lee", "email": "sung.lee@amd.com", "time": "Mon Apr 20 11:38:30 2020 -0400" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Tue May 05 16:07:02 2020 -0400" }, "message": "drm/amd/display: Update DCN2.1 DV Code Revision\n\n[WHY \u0026 HOW]\nThere is a problem in hscale_pixel_rate, the bug\ncauses DCN to be more optimistic (more likely to underflow)\nin upscale cases during prefetch.\nThis commit ports the fix from DV code to address these issues.\n\nSigned-off-by: Sung Lee \u003csung.lee@amd.com\u003e\nReviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nAcked-by: Aurabindo Pillai \u003caurabindo.pillai@amd.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n" }, { "commit": "7f13657d141346125f4d0bb93eab4777f40c406e", "tree": "ee786539b1ab574f47c0f9dd9c406a36176cc6c8", "parents": [ "d8f1b9716cfd1a1f74c0fedad40c5f65a25aa208" ], "author": { "name": "Xiaoguang Wang", "email": "xiaoguang.wang@linux.alibaba.com", "time": "Tue May 05 16:28:53 2020 +0800" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Tue May 05 13:18:11 2020 -0600" }, "message": "io_uring: handle -EFAULT properly in io_uring_setup()\n\nIf copy_to_user() in io_uring_setup() failed, we\u0027ll leak many kernel\nresources, which will be recycled until process terminates. This bug\ncan be reproduced by using mprotect to set params to PROT_READ. To fix\nthis issue, refactor io_uring_create() a bit to add a new \u0027struct\nio_uring_params __user *params\u0027 parameter and move the copy_to_user()\nin io_uring_setup() to io_uring_setup(), if copy_to_user() failed,\nwe can free kernel resource properly.\n\nSuggested-by: Jens Axboe \u003caxboe@kernel.dk\u003e\nSigned-off-by: Xiaoguang Wang \u003cxiaoguang.wang@linux.alibaba.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "755f5738ff981769211a0bfac709d514ef5b9f86", "tree": "a7b420b6af72f8fdb6a962da72933a3f115a9fff", "parents": [ "bd4af432cc71b5fbfe4833510359a6ad3ada250d" ], "author": { "name": "Dejin Zheng", "email": "zhengdejin5@gmail.com", "time": "Tue May 05 10:03:29 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 05 11:11:12 2020 -0700" }, "message": "net: broadcom: fix a mistake about ioremap resource\n\nCommit d7a5502b0bb8b (\"net: broadcom: convert to\ndevm_platform_ioremap_resource_byname()\") will broke this driver.\nidm_base and nicpm_base were optional, after this change, they are\nmandatory. it will probe fails with -22 when the dtb doesn\u0027t have them\ndefined. so revert part of this commit and make idm_base and nicpm_base\nas optional.\n\nFixes: d7a5502b0bb8bde (\"net: broadcom: convert to devm_platform_ioremap_resource_byname()\")\nReported-by: Jonathan Richardson \u003cjonathan.richardson@broadcom.com\u003e\nCc: Scott Branden \u003cscott.branden@broadcom.com\u003e\nCc: Ray Jui \u003cray.jui@broadcom.com\u003e\nCc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nCc: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Dejin Zheng \u003czhengdejin5@gmail.com\u003e\nAcked-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5fe89a6acd668cbd1817fcdef5caa9fee568c2e8", "tree": "e012109e6312bfd20826bdcdd775b8e6191f7702", "parents": [ "3a3a71f97c30983f1627c2c550d43566e9b634d2" ], "author": { "name": "Sean Paul", "email": "seanpaul@chromium.org", "time": "Tue Apr 14 15:02:55 2020 -0400" }, "committer": { "name": "Sean Paul", "email": "seanpaul@chromium.org", "time": "Tue May 05 14:01:53 2020 -0400" }, "message": "drm: Fix HDCP failures when SRM fw is missing\n\nThe SRM cleanup in 79643fddd6eb2 (\"drm/hdcp: optimizing the srm\nhandling\") inadvertently altered the behavior of HDCP auth when\nthe SRM firmware is missing. Before that patch, missing SRM was\ninterpreted as the device having no revoked keys. With that patch,\nif the SRM fw file is missing we reject _all_ keys.\n\nThis patch fixes that regression by returning success if the file\ncannot be found. It also checks the return value from request_srm such\nthat we won\u0027t end up trying to parse the ksv list if there is an error\nfetching it.\n\nFixes: 79643fddd6eb (\"drm/hdcp: optimizing the srm handling\")\nCc: stable@vger.kernel.org\nCc: Ramalingam C \u003cramalingam.c@intel.com\u003e\nCc: Sean Paul \u003csean@poorly.run\u003e\nCc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nCc: Maxime Ripard \u003cmripard@kernel.org\u003e\nCc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nCc: David Airlie \u003cairlied@linux.ie\u003e\nCc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\nCc: dri-devel@lists.freedesktop.org\nReviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\nSigned-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20200414190258.38873-1-sean@poorly.run\n\nChanges in v2:\n-Noticed a couple other things to clean up\nReviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n" }, { "commit": "f0e1d3ac2d7c16a5d2c9d67f5a61133db7681af8", "tree": "243032e62a5f34071e5b882bdd7b6cfa0e68268b", "parents": [ "af2e58818082ac0db29539444ca17eb1e77f6000" ], "author": { "name": "Jeffrey Hugo", "email": "jhugo@codeaurora.org", "time": "Fri May 01 00:35:55 2020 +0530" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@linuxfoundation.org", "time": "Tue May 05 16:59:54 2020 +0200" }, "message": "bus: mhi: core: Fix channel device name conflict\n\nWhen multiple instances of the same MHI product are present in a system,\nwe can see a splat from mhi_create_devices() - \"sysfs: cannot create\nduplicate filename\".\n\nThis is because the device names assigned to the MHI channel devices are\nnon-unique. They consist of the channel\u0027s name, and the channel\u0027s pipe\nid. For identical products, each instance is going to have the same\nset of channel (both in name and pipe id).\n\nTo fix this, we prepend the device name of the parent device that the\nMHI channels belong to. Since different instances of the same product\nshould have unique device names, this makes the MHI channel devices for\neach product also unique.\n\nAdditionally, remove the pipe id from the MHI channel device name. This\nis an internal detail to the MHI product that provides little value, and\nimposes too much device specific internal details to userspace. It is\nexpected that channel with a specific name (ie \"SAHARA\") has a specific\nclient, and it does not matter what pipe id that channel is enumerated on.\nThe pipe id is an internal detail between the MHI bus, and the hardware.\nThe client is not expected to make decisions based on the pipe id, and to\ndo so would require the client to have intimate knowledge of the hardware,\nwhich is inappropiate as it may violate the layering provided by the MHI\nbus. The limitation of doing this is that each product may only have one\ninstance of a channel by a unique name. This limitation is appropriate\ngiven the usecases of MHI channels.\n\nSigned-off-by: Jeffrey Hugo \u003cjhugo@codeaurora.org\u003e\nReviewed-by: Hemant Kumar \u003chemantk@codeaurora.org\u003e\nReviewed-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nSigned-off-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nLink: https://lore.kernel.org/r/20200430190555.32741-7-manivannan.sadhasivam@linaro.org\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n" } ], "next": "af2e58818082ac0db29539444ca17eb1e77f6000" }