)]}' { "log": [ { "commit": "3d7cb6b04c3f3115719235cc6866b10326de34cd", "tree": "9d86b319bf46cbf0fd0a9b3e1d46276474230c8e", "parents": [ "334c0ef6429f261c7f53dc035632435ffbc0c60d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 14:03:01 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 14:03:01 2022 -0700" }, "message": "Linux 5.19\n" }, { "commit": "334c0ef6429f261c7f53dc035632435ffbc0c60d", "tree": "aa3dc0d26424cde763a8507fcd2080e7cf456103", "parents": [ "89caf575400a9296e45b7de97b07d2fbf249c1f5", "8dc592c41f38735306d1f1dc0b183601379c6d94" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:52:20 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:52:20 2022 -0700" }, "message": "Merge tag \u0027clk-fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux\n\nPull clk fix from Stephen Boyd:\n \"One-liner fix of a NULL pointer deref in the Allwinner clk driver\"\n\n* tag \u0027clk-fixes-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:\n clk: sunxi-ng: Fix H6 RTC clock definition\n" }, { "commit": "89caf575400a9296e45b7de97b07d2fbf249c1f5", "tree": "d61fbf4ef873a69bf453c7fa6d67b6ce63766fef", "parents": [ "5e4823e6dab8fcc897757c1be68d157a369e4bb5", "ea304a8b89fd0d6cf94ee30cb139dc23d9f1a62f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:26:53 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:26:53 2022 -0700" }, "message": "Merge tag \u0027x86_urgent_for_v5.19\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Borislav Petkov:\n\n - Update the \u0027mitigations\u003d\u0027 kernel param documentation\n\n - Check the IBPB feature flag before enabling IBPB in firmware calls\n because cloud vendors\u0027 fantasy when it comes to creating guest\n configurations is unlimited\n\n - Unexport sev_es_ghcb_hv_call() before 5.19 releases now that HyperV\n doesn\u0027t need it anymore\n\n - Remove dead CONFIG_* items\n\n* tag \u0027x86_urgent_for_v5.19\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n docs/kernel-parameters: Update descriptions for \"mitigations\u003d\" param with retbleed\n x86/bugs: Do not enable IBPB at firmware entry when IBPB is not available\n Revert \"x86/sev: Expose sev_es_ghcb_hv_call() for use by HyperV\"\n x86/configs: Update configs in x86_debug.config\n" }, { "commit": "5e4823e6dab8fcc897757c1be68d157a369e4bb5", "tree": "6958c902b6b0754d14d54ccb0cca1f38cfa83072", "parents": [ "cd2715b7929bcaf6e38d13ac512e0a9bba8df10d", "6eebd5fb20838f5971ba17df9f55cc4f84a31053" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:21:13 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:21:13 2022 -0700" }, "message": "Merge tag \u0027locking_urgent_for_v5.19\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull locking fix from Borislav Petkov:\n\n - Avoid rwsem lockups in certain situations when handling the handoff\n bit\n\n* tag \u0027locking_urgent_for_v5.19\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n locking/rwsem: Allow slowpath writer to ignore handoff bit if not set by first waiter\n" }, { "commit": "cd2715b7929bcaf6e38d13ac512e0a9bba8df10d", "tree": "c38eec98fe24a049557d4804ca5a541bdc9e3815", "parents": [ "6a010258447d386186ca10cd374d888ac66ffe1a", "5e2805d5379619c4a2e3ae4994e73b36439f4bad" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:12:58 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Jul 31 09:12:58 2022 -0700" }, "message": "Merge tag \u0027edac_urgent_for_v5.19\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras\n\nPull EDAC fixes from Borislav Petkov:\n\n - Relax the condition under which the DIMM label in ghes_edac is set in\n order to accomodate an HPE BIOS which sets only the device but not\n the bank\n\n - Two forgotten fixes to synopsys_edac when handling error interrupts\n\n* tag \u0027edac_urgent_for_v5.19\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:\n EDAC/ghes: Set the DIMM label unconditionally\n EDAC/synopsys: Re-enable the error interrupts on v3 hw\n EDAC/synopsys: Use the correct register to disable the error interrupt on v3 hw\n" }, { "commit": "6a010258447d386186ca10cd374d888ac66ffe1a", "tree": "57ec6d03fb4e72fe86a86f7a9b2037e485403463", "parents": [ "620725263f4222b3c94d4ee19846835feec0ad69", "ec85bd369fd2bfaed6f45dd678706429d4f75b48" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 30 17:24:16 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 30 17:24:16 2022 -0700" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.armlinux.org.uk/~rmk/linux-arm\n\nPull ARM fixes from Russell King:\n \"Last set of ARM fixes for 5.19:\n\n - fix for MAX_DMA_ADDRESS overflow\n\n - fix for find_*_bit performing an out of bounds memory access\"\n\n* tag \u0027for-linus\u0027 of git://git.armlinux.org.uk/~rmk/linux-arm:\n ARM: findbit: fix overflowing offset\n ARM: 9216/1: Fix MAX_DMA_ADDRESS overflow\n" }, { "commit": "6eebd5fb20838f5971ba17df9f55cc4f84a31053", "tree": "ecb5fd0a4bb1de91b66c83b2a74ecdb5247175f0", "parents": [ "e0dccc3b76fb35bb257b4118367a883073d7390e" ], "author": { "name": "Waiman Long", "email": "longman@redhat.com", "time": "Wed Jun 22 16:04:19 2022 -0400" }, "committer": { "name": "Peter Zijlstra", "email": "peterz@infradead.org", "time": "Sat Jul 30 10:58:28 2022 +0200" }, "message": "locking/rwsem: Allow slowpath writer to ignore handoff bit if not set by first waiter\n\nWith commit d257cc8cb8d5 (\"locking/rwsem: Make handoff bit handling more\nconsistent\"), the writer that sets the handoff bit can be interrupted\nout without clearing the bit if the wait queue isn\u0027t empty. This disables\nreader and writer optimistic lock spinning and stealing.\n\nNow if a non-first writer in the queue is somehow woken up or a new\nwaiter enters the slowpath, it can\u0027t acquire the lock. This is not the\ncase before commit d257cc8cb8d5 as the writer that set the handoff bit\nwill clear it when exiting out via the out_nolock path. This is less\nefficient as the busy rwsem stays in an unlock state for a longer time.\n\nIn some cases, this new behavior may cause lockups as shown in [1] and\n[2].\n\nThis patch allows a non-first writer to ignore the handoff bit if it\nis not originally set or initiated by the first waiter. This patch is\nshown to be effective in fixing the lockup problem reported in [1].\n\n[1] https://lore.kernel.org/lkml/20220617134325.GC30825@techsingularity.net/\n[2] https://lore.kernel.org/lkml/3f02975c-1a9d-be20-32cf-f1d8e3dfafcc@oracle.com/\n\nFixes: d257cc8cb8d5 (\"locking/rwsem: Make handoff bit handling more consistent\")\nSigned-off-by: Waiman Long \u003clongman@redhat.com\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nAcked-by: John Donnelly \u003cjohn.p.donnelly@oracle.com\u003e\nTested-by: Mel Gorman \u003cmgorman@techsingularity.net\u003e\nLink: https://lore.kernel.org/r/20220622200419.778799-1-longman@redhat.com\n" }, { "commit": "620725263f4222b3c94d4ee19846835feec0ad69", "tree": "7d6de25a6d4a94207a1b20dad70639fd06b5a2fe", "parents": [ "8a91f86f3e9e9608210166c70aaad4919018c0e7", "8a295dbbaf7292c582a40ce469c326f472d51f66" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 21:02:35 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 21:02:35 2022 -0700" }, "message": "Merge tag \u0027mm-hotfixes-stable-2022-07-29\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm\n\nPull misc fixes from Andrew Morton:\n \"Two hotfixes, both cc:stable\"\n\n* tag \u0027mm-hotfixes-stable-2022-07-29\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:\n mm/hmm: fault non-owner device private entries\n page_alloc: fix invalid watermark check on a negative value\n" }, { "commit": "8a91f86f3e9e9608210166c70aaad4919018c0e7", "tree": "1d0e0e3a909fc8634410275c2b575130ecdc8138", "parents": [ "e65c6a46df94c8d76ea1129eb2d4564670c6f214", "eda3953b6a805d6df87a4c51058493ec88bfc622" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 16:07:35 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 16:07:35 2022 -0700" }, "message": "Merge tag \u0027block-5.19-2022-07-29\u0027 of git://git.kernel.dk/linux-block\n\nPull block fix from Jens Axboe:\n \"Just a single fix for NVMe, yet another quirk addition\"\n\n* tag \u0027block-5.19-2022-07-29\u0027 of git://git.kernel.dk/linux-block:\n nvme-pci: Crucial P2 has bogus namespace ids\n" }, { "commit": "e65c6a46df94c8d76ea1129eb2d4564670c6f214", "tree": "f04307d21fc3a7d8fcb3cb3845420eeacaf4816d", "parents": [ "1c8ac1c4af178ebcdf6fd4a0ba11716e2c814593", "ce156c8a1811c96a243590abd0e9b5a3b72c1f3a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 13:25:31 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 13:25:31 2022 -0700" }, "message": "Merge tag \u0027drm-fixes-2022-07-30\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull more drm fixes from Dave Airlie:\n \"Maxime had the dog^Wmailing list server eat his homework^Wmisc pull\n request.\n\n Two more small fixes, one in nouveau svm code and the other in\n simpledrm.\n\n nouveau:\n - page migration fix\n\n simpledrm:\n - fix mode_valid return value\"\n\n* tag \u0027drm-fixes-2022-07-30\u0027 of git://anongit.freedesktop.org/drm/drm:\n nouveau/svm: Fix to migrate all requested pages\n drm/simpledrm: Fix return type of simpledrm_simple_display_pipe_mode_valid()\n" }, { "commit": "ce156c8a1811c96a243590abd0e9b5a3b72c1f3a", "tree": "313f6470a3ddd22e129f74248b44b35bc76bc65c", "parents": [ "f16a2f593d0095e82e6b7f9d776f869c8ab45952", "66cee9097e2b74ff3c8cc040ce5717c521a0c3fa" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sat Jul 30 06:09:48 2022 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Sat Jul 30 06:09:57 2022 +1000" }, "message": "Merge tag \u0027drm-misc-fixes-2022-07-29\u0027 of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes\n\nOne fix to fix simpledrm mode_valid return value, and one for page\nmigration in nouveau\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/20220729094514.sfzhc3gqjgwgal62@penduick\n" }, { "commit": "1c8ac1c4af178ebcdf6fd4a0ba11716e2c814593", "tree": "1764cff0c585c4a83c194c87dce87e0f574497aa", "parents": [ "bb83c99d3d0aa2f51ba0b2abd7115002620118e6", "f5c2976e0cb0f6236013bfb479868531b04f61d4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 13:07:03 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 13:07:03 2022 -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 fixes, three in drivers.\n\n The two biggest fixes are ufs and the remaining driver and core fix\n are small and obvious (and the core fix is low risk)\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: ufs: core: Fix a race condition related to device management\n scsi: core: Fix warning in scsi_alloc_sgtables()\n scsi: ufs: host: Hold reference returned by of_parse_phandle()\n scsi: mpt3sas: Stop fw fault watchdog work item during system shutdown\n" }, { "commit": "ea304a8b89fd0d6cf94ee30cb139dc23d9f1a62f", "tree": "7939302cc3e49aead42ded5d04b2673745e4d198", "parents": [ "571c30b1a88465a1c85a6f7762609939b9085a15" ], "author": { "name": "Eiichi Tsukata", "email": "eiichi.tsukata@nutanix.com", "time": "Thu Jul 28 04:39:07 2022 +0000" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Fri Jul 29 20:47:07 2022 +0200" }, "message": "docs/kernel-parameters: Update descriptions for \"mitigations\u003d\" param with retbleed\n\nUpdates descriptions for \"mitigations\u003doff\" and \"mitigations\u003dauto,nosmt\"\nwith the respective retbleed\u003d settings.\n\nSigned-off-by: Eiichi Tsukata \u003ceiichi.tsukata@nutanix.com\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nCc: corbet@lwn.net\nLink: https://lore.kernel.org/r/20220728043907.165688-1-eiichi.tsukata@nutanix.com\n" }, { "commit": "8a295dbbaf7292c582a40ce469c326f472d51f66", "tree": "07cb545d852fb7c56a2bb9796090b127b374126b", "parents": [ "9282012fc0aa248b77a69f5eb802b67c5a16bb13" ], "author": { "name": "Ralph Campbell", "email": "rcampbell@nvidia.com", "time": "Mon Jul 25 11:36:14 2022 -0700" }, "committer": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Fri Jul 29 11:33:37 2022 -0700" }, "message": "mm/hmm: fault non-owner device private entries\n\nIf hmm_range_fault() is called with the HMM_PFN_REQ_FAULT flag and a\ndevice private PTE is found, the hmm_range::dev_private_owner page is used\nto determine if the device private page should not be faulted in. \nHowever, if the device private page is not owned by the caller,\nhmm_range_fault() returns an error instead of calling migrate_to_ram() to\nfault in the page.\n\nFor example, if a page is migrated to GPU private memory and a RDMA fault\ncapable NIC tries to read the migrated page, without this patch it will\nget an error. With this patch, the page will be migrated back to system\nmemory and the NIC will be able to read the data.\n\nLink: https://lkml.kernel.org/r/20220727000837.4128709-2-rcampbell@nvidia.com\nLink: https://lkml.kernel.org/r/20220725183615.4118795-2-rcampbell@nvidia.com\nFixes: 08ddddda667b (\"mm/hmm: check the device private page owner in hmm_range_fault()\")\nSigned-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nReported-by: Felix Kuehling \u003cfelix.kuehling@amd.com\u003e\nReviewed-by: Alistair Popple \u003capopple@nvidia.com\u003e\nCc: Philip Yang \u003cPhilip.Yang@amd.com\u003e\nCc: Jason Gunthorpe \u003cjgg@nvidia.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "9282012fc0aa248b77a69f5eb802b67c5a16bb13", "tree": "999ab1fd2d2f53e0fa19b3cc0c17f373106a2c58", "parents": [ "1f7ea54727caaa6701a15af0cbeddfdb015b2869" ], "author": { "name": "Jaewon Kim", "email": "jaewon31.kim@samsung.com", "time": "Mon Jul 25 18:52:12 2022 +0900" }, "committer": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Fri Jul 29 11:33:37 2022 -0700" }, "message": "page_alloc: fix invalid watermark check on a negative value\n\nThere was a report that a task is waiting at the\nthrottle_direct_reclaim. The pgscan_direct_throttle in vmstat was\nincreasing.\n\nThis is a bug where zone_watermark_fast returns true even when the free\nis very low. The commit f27ce0e14088 (\"page_alloc: consider highatomic\nreserve in watermark fast\") changed the watermark fast to consider\nhighatomic reserve. But it did not handle a negative value case which\ncan be happened when reserved_highatomic pageblock is bigger than the\nactual free.\n\nIf watermark is considered as ok for the negative value, allocating\ncontexts for order-0 will consume all free pages without direct reclaim,\nand finally free page may become depleted except highatomic free.\n\nThen allocating contexts may fall into throttle_direct_reclaim. This\nsymptom may easily happen in a system where wmark min is low and other\nreclaimers like kswapd does not make free pages quickly.\n\nHandle the negative case by using MIN.\n\nLink: https://lkml.kernel.org/r/20220725095212.25388-1-jaewon31.kim@samsung.com\nFixes: f27ce0e14088 (\"page_alloc: consider highatomic reserve in watermark fast\")\nSigned-off-by: Jaewon Kim \u003cjaewon31.kim@samsung.com\u003e\nReported-by: GyeongHwan Hong \u003cgh21.hong@samsung.com\u003e\nAcked-by: Mel Gorman \u003cmgorman@techsingularity.net\u003e\nCc: Minchan Kim \u003cminchan@kernel.org\u003e\nCc: Baoquan He \u003cbhe@redhat.com\u003e\nCc: Vlastimil Babka \u003cvbabka@suse.cz\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Michal Hocko \u003cmhocko@kernel.org\u003e\nCc: Yong-Taek Lee \u003cytk.lee@samsung.com\u003e\nCc: \u003cstable@vger.kerenl.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "bb83c99d3d0aa2f51ba0b2abd7115002620118e6", "tree": "97577d7e6b5ed4ae2f6cdf20d55190f01f7645ae", "parents": [ "4b20426d044163df0ff9d868e9767ca9e65dd36c", "9a241805673ec0a826b7ddf84b00f4e03adb0a5e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 11:26:28 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 11:26:28 2022 -0700" }, "message": "Merge tag \u0027perf-tools-fixes-for-v5.19-2022-07-29\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux\n\nPull perf tools fixes from Arnaldo Carvalho de Melo:\n\n - Fix addresses for bss symbols, describing variables used in resolving\n data access in tools such as \u0027perf c2c\u0027 and \u0027perf mem\u0027.\n\n - Skip symbols if SHF_ALLOC flag is not set, a technique used for\n listing deprecated symbols, its addresses are zeros, so not useful.\n\n - Remove undefined behavior from bpf_perf_object__next() when dealing\n with an empty bpf_objects_list list.\n\n - Make a ARM CoreSight disasm script work with both python2 and\n python3.\n\n - Sync x86\u0027s cpufeatures header with with the kernel sources.\n\n* tag \u0027perf-tools-fixes-for-v5.19-2022-07-29\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:\n perf bpf: Remove undefined behavior from bpf_perf_object__next()\n perf symbol: Skip symbols if SHF_ALLOC flag is not set\n perf symbol: Correct address for bss symbols\n perf scripts python: Let script to be python2 compliant\n tools headers cpufeatures: Sync with the kernel sources\n" }, { "commit": "4b20426d044163df0ff9d868e9767ca9e65dd36c", "tree": "ba90cbc9f820c5635da7fd18acbbf4804e7ac87a", "parents": [ "506e6dfb0f1131f104df66ccae1197bdf8f1b6b0", "46a4d679ef88285ea17c3e1e4fed330be2044f21" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 11:20:40 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 11:20:40 2022 -0700" }, "message": "Merge tag \u0027wq-for-5.19-rc8-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq\n\nPull workqueue fix from Tejun Heo:\n \"Just one commit to suppress a spurious warning added during the 5.19\n cycle\"\n\n* tag \u0027wq-for-5.19-rc8-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:\n workqueue: Avoid a false warning in unbind_workers()\n" }, { "commit": "506e6dfb0f1131f104df66ccae1197bdf8f1b6b0", "tree": "78e5dabc912d76bc90dc6545265942ec1dcb0522", "parents": [ "e4d8b09d671f0ee92d36c48b65d2f34affcdd531", "d295ad34f236c3518634fb6403d4c0160456e470" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 10:57:26 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 10:57:26 2022 -0700" }, "message": "Merge tag \u0027pm-5.19-rc9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fix from Rafael Wysocki:\n \"Make some false positive RCU splats resulting from a recent intel_idle\n driver change go away (Waiman Long)\"\n\n* tag \u0027pm-5.19-rc9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n intel_idle: Fix false positive RCU splats due to incorrect hardirqs state\n" }, { "commit": "46a4d679ef88285ea17c3e1e4fed330be2044f21", "tree": "041b7b34242dc7f44b8bd6bf894839e778926b12", "parents": [ "8bee9dd953b69c634d1c9a3241a8b357469ad4aa" ], "author": { "name": "Lai Jiangshan", "email": "jiangshan.ljs@antgroup.com", "time": "Fri Jul 29 17:44:38 2022 +0800" }, "committer": { "name": "Tejun Heo", "email": "tj@kernel.org", "time": "Fri Jul 29 07:49:02 2022 -1000" }, "message": "workqueue: Avoid a false warning in unbind_workers()\n\nDoing set_cpus_allowed_ptr() with wq_unbound_cpumask can be possible\nfails and trigger the false warning.\n\nUse cpu_possible_mask instead when wq_unbound_cpumask has no active CPUs.\n\nIt is very easy to trigger the warning:\n Set wq_unbound_cpumask to a small set of CPUs.\n Offline all the CPUs of wq_unbound_cpumask.\n Offline an extra CPU and trigger the warning.\n\nFixes: 10a5a651e3af (\"workqueue: Restrict kworker in the offline CPU pool running on housekeeping CPUs\")\nSigned-off-by: Lai Jiangshan \u003cjiangshan.ljs@antgroup.com\u003e\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n" }, { "commit": "e4d8b09d671f0ee92d36c48b65d2f34affcdd531", "tree": "7dfa16ad4e5a5408335aa17e9993b4f29d4f7e3a", "parents": [ "a95eb1d086dcc579d52ca4c34742516f6434d1f2", "88bd24d73d5bfa1b7b97a9221ff320fc44ef401a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 10:46:03 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 10:46:03 2022 -0700" }, "message": "Merge tag \u0027riscv-for-linus-5.19-rc9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux\n\nPull RISC-V fix from Palmer Dabbelt:\n \"A build fix for \u0027make vdso_install\u0027 that avoids an issue trying to\n install the compat VDSO\"\n\n* tag \u0027riscv-for-linus-5.19-rc9\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:\n riscv: compat: vdso: Fix vdso_install target\n" }, { "commit": "a95eb1d086dcc579d52ca4c34742516f6434d1f2", "tree": "10033fc1e634d94c7d98baf45d27952ff1fb9ee2", "parents": [ "9d928d9b78beec5d4b8afde9c144919b979685f2", "45b53c9051770c0d9145083a328548745ee2e75b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 10:10:30 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 10:10:30 2022 -0700" }, "message": "Merge tag \u0027loongarch-fixes-5.19-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson\n\nPull LoongArch fixes from Huacai Chen:\n\n - Fix cache size calculation, stack protection attributes, ptrace\u0027s\n fpr_set and \"ROM Size\" in boardinfo\n\n - Some cleanups and improvements of assembly\n\n - Some cleanups of unused code and useless code\n\n* tag \u0027loongarch-fixes-5.19-5\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson:\n LoongArch: Fix wrong \"ROM Size\" of boardinfo\n LoongArch: Fix missing fcsr in ptrace\u0027s fpr_set\n LoongArch: Fix shared cache size calculation\n LoongArch: Disable executable stack by default\n LoongArch: Remove unused variables\n LoongArch: Remove clock setting during cpu hotplug stage\n LoongArch: Remove useless header compiler.h\n LoongArch: Remove several syntactic sugar macros for branches\n LoongArch: Re-tab the assembly files\n LoongArch: Simplify \"BGT foo, zero\" with BGTZ\n LoongArch: Simplify \"BLT foo, zero\" with BLTZ\n LoongArch: Simplify \"BEQ/BNE foo, zero\" with BEQZ/BNEZ\n LoongArch: Use the \"move\" pseudo-instruction where applicable\n LoongArch: Use the \"jr\" pseudo-instruction where applicable\n LoongArch: Use ABI names of registers where appropriate\n" }, { "commit": "9d928d9b78beec5d4b8afde9c144919b979685f2", "tree": "fcb32435cea66dd52b48c6ba3169c57897df781a", "parents": [ "6e2c0490769ef8a95b61304389116ccc85c53e12", "c653c591789b3acfa4bf6ae45d5af4f330e50a91" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 09:57:07 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Jul 29 09:57:07 2022 -0700" }, "message": "Merge tag \u0027powerpc-5.19-6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull powerpc fixes from Michael Ellerman:\n\n - Re-enable the new amdgpu display engine for powerpc, as long as the\n compiler is correctly configured.\n\n - Disable stack variable initialisation in prom_init to fix GCC 12\n allmodconfig.\n\nThanks to Dan Horák and Sudip Mukherjee.\n\n* tag \u0027powerpc-5.19-6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n drm/amdgpu: Re-enable DCN for 64-bit powerpc\n powerpc/64s: Disable stack variable initialisation for prom_init\n" }, { "commit": "45b53c9051770c0d9145083a328548745ee2e75b", "tree": "218c360338a1bb64b85f9fd05500b35a101100a3", "parents": [ "b0f3bdc00240fc9d7bf0f2a076943122d168c95e" ], "author": { "name": "Tiezhu Yang", "email": "yangtiezhu@loongson.cn", "time": "Thu Jul 21 17:53:01 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:33 2022 +0800" }, "message": "LoongArch: Fix wrong \"ROM Size\" of boardinfo\n\nWe can see the \"ROM Size\" is different in the following outputs:\n\n[root@linux loongson]# cat /sys/firmware/loongson/boardinfo\nBIOS Information\nVendor : Loongson\nVersion : vUDK2018-LoongArch-V2.0.pre-beta8\nROM Size : 63 KB\nRelease Date : 06/15/2022\n\nBoard Information\nManufacturer : Loongson\nBoard Name : Loongson-LS3A5000-7A1000-1w-A2101\nFamily : LOONGSON64\n\n[root@linux loongson]# dmidecode | head -11\n...\nHandle 0x0000, DMI type 0, 26 bytes\nBIOS Information\n\tVendor: Loongson\n\tVersion: vUDK2018-LoongArch-V2.0.pre-beta8\n\tRelease Date: 06/15/2022\n\tROM Size: 4 MB\n\nAccording to \"BIOS Information (Type 0) structure\" in the SMBIOS\nReference Specification [1], it shows 64K * (n+1) is the size of\nthe physical device containing the BIOS if the size is less than\n16M.\n\nAdditionally, we can see the related code in dmidecode [2]:\n\n u64 s \u003d { .l \u003d (code1 + 1) \u003c\u003c 6 };\n\nSo the output of dmidecode is correct, the output of boardinfo\nis wrong, fix it.\n\nBy the way, at present no need to consider the size is 16M or\ngreater on LoongArch, because it is usually 4M or 8M which is\nenough to use.\n\n[1] https://www.dmtf.org/sites/default/files/standards/documents/DSP0134_3.6.0.pdf\n[2] https://git.savannah.nongnu.org/cgit/dmidecode.git/tree/dmidecode.c#n347\n\nFixes: 628c3bb40e9a (\"LoongArch: Add boot and setup routines\")\nReviewed-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Tiezhu Yang \u003cyangtiezhu@loongson.cn\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "b0f3bdc00240fc9d7bf0f2a076943122d168c95e", "tree": "44e91945af7f64a47da280d48449fae1873ee093", "parents": [ "1aea29d7c3569e5b6c40e73c51e9f4b2142c96ef" ], "author": { "name": "Qi Hu", "email": "huqi@loongson.cn", "time": "Thu Jul 14 14:25:50 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:33 2022 +0800" }, "message": "LoongArch: Fix missing fcsr in ptrace\u0027s fpr_set\n\nIn file ptrace.c, function fpr_set does not copy fcsr data from ubuf\nto kbuf. That\u0027s the reason why fcsr cannot be modified by ptrace.\n\nThis patch fixs this problem and allows users using ptrace to modify\nthe fcsr.\n\nCo-developed-by: Xu Li \u003clixu@loongson.cn\u003e\nSigned-off-by: Qi Hu \u003chuqi@loongson.cn\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "1aea29d7c3569e5b6c40e73c51e9f4b2142c96ef", "tree": "2273750572cd4be482eb7268c8230a40f7e798e8", "parents": [ "317980e6b4d03884429f2cdaf51efd28f01b71b0" ], "author": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Wed Jul 13 18:00:41 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:33 2022 +0800" }, "message": "LoongArch: Fix shared cache size calculation\n\nCurrent calculation of shared cache size is from the node (die) scope,\nbut we hope \u0027lscpu\u0027 to show the shared cache size of the whole package\nfor multi-die chips (e.g., Loongson-3C5000L, which contains 4 dies in\none package). So fix it by multiplying nodes_per_package.\n\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "317980e6b4d03884429f2cdaf51efd28f01b71b0", "tree": "d0eea6b90a303630ec4cb36983dd0e7b261f1248", "parents": [ "3a3a4f7a65e3ff7ad395afc8c41ac317c8667546" ], "author": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Tue Jul 26 20:43:11 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Disable executable stack by default\n\nDisable executable stack for LoongArch by default, as all modern\narchitectures do.\n\nReported-by: Andreas Schwab \u003cschwab@suse.de\u003e\nSuggested-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nLink: https://sourceware.org/pipermail/binutils/2022-July/121992.html\nTested-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nTested-by: Xi Ruoyao \u003cxry111@xry111.site\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "3a3a4f7a65e3ff7ad395afc8c41ac317c8667546", "tree": "d5e23f0efe876d35a07875d4ed9704da3d3b0871", "parents": [ "71610ab1d017e131a9888ef8acd035284fb0e1dd" ], "author": { "name": "Bibo Mao", "email": "maobibo@loongson.cn", "time": "Wed Jul 20 15:21:52 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Remove unused variables\n\nThere are some variables never used or referenced, this patch\nremoves these varaibles and make the code cleaner.\n\nReviewed-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Bibo Mao \u003cmaobibo@loongson.cn\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "71610ab1d017e131a9888ef8acd035284fb0e1dd", "tree": "38f705fd1cf80f188a5aad59ce4b8e25cecdb04d", "parents": [ "f62b7626cb79dfbfe292145b7ebeee4dc63c9499" ], "author": { "name": "Bibo Mao", "email": "maobibo@loongson.cn", "time": "Wed Jul 20 15:21:51 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Remove clock setting during cpu hotplug stage\n\nOn physical machine we can save power by disabling clock of hot removed\ncpu. However as different platforms require different methods to\nconfigure clocks, the code is platform-specific, and probably belongs to\nfirmware/pmu or cpu regulator, rather than generic arch/loongarch code.\n\nAlso, there is no such register on QEMU virt machine since the\nclock/frequency regulation is not emulated.\n\nThis patch removes the hard-coded clock register accesses in generic\nLoongArch cpu hotplug flow.\n\nReviewed-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Bibo Mao \u003cmaobibo@loongson.cn\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "f62b7626cb79dfbfe292145b7ebeee4dc63c9499", "tree": "22fbfd489c110c2af439723e9a4384812531ee25", "parents": [ "ab6e57a69df515cc9231b578de5b820f9ba3d0be" ], "author": { "name": "Jun Yi", "email": "yijun@loongson.cn", "time": "Thu Jul 21 19:10:49 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Remove useless header compiler.h\n\nThe content of LoongArch\u0027s compiler.h is trivial, with some unused\nanywhere, so inline the definitions and remove the header.\n\nSigned-off-by: Jun Yi \u003cyijun@loongson.cn\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "ab6e57a69df515cc9231b578de5b820f9ba3d0be", "tree": "5a9af39cc0ede5a40028783a1345c872147d0f40", "parents": [ "f5c3c22f21b6a002e371afdcc9180a2fa47dc267" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:15 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Remove several syntactic sugar macros for branches\n\nThese syntactic sugars have been supported by upstream binutils from the\nbeginning, so no need to patch them locally.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "f5c3c22f21b6a002e371afdcc9180a2fa47dc267", "tree": "410717336fcfa47fed29fe3ecc50ea78d8f0e067", "parents": [ "1fdb9a92495a6b6996530d27781892796e22f08b" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:22 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Re-tab the assembly files\n\nReflow the *.S files for better stylistic consistency, namely hard tabs\nafter mnemonic position, and vertical alignment of the first operand\nwith hard tabs. Tab width is obviously 8. Some pre-existing intra-block\nvertical alignments are preserved.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "1fdb9a92495a6b6996530d27781892796e22f08b", "tree": "05a8a33f18dd9b21e06402162d5542c3406803d6", "parents": [ "d1bc75d7595b237f78b594509ea7cc159f98cae9" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:21 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Simplify \"BGT foo, zero\" with BGTZ\n\nSupport for the syntactic sugar is present in upstream binutils port\nfrom the beginning. Use it for shorter lines and better consistency.\nGenerated code should be identical.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "d1bc75d7595b237f78b594509ea7cc159f98cae9", "tree": "f88a8d2cfb3295b219c7a3fb21a53274357a6a45", "parents": [ "d47b2dc87c58154052daf8ac0f9229db5c7890cc" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:20 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Simplify \"BLT foo, zero\" with BLTZ\n\nSupport for the syntactic sugar is present in upstream binutils port\nfrom the beginning. Use it for shorter lines and better consistency.\nGenerated code should be identical.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "d47b2dc87c58154052daf8ac0f9229db5c7890cc", "tree": "78ace17ff4fec7785d9a0622d51357afc6810087", "parents": [ "57ce5d3eefacfaadfe2ed0a3a85713d1ae6287b9" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:19 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Simplify \"BEQ/BNE foo, zero\" with BEQZ/BNEZ\n\nWhile B{EQ,NE}Z and B{EQ,NE} are different instructions, and the vastly\nexpanded range for branch destination does not really matter in the few\ncases touched, use the B{EQ,NE}Z where possible for shorter lines and\nbetter consistency (e.g. some places used \"BEQ foo, zero\", while some\nused \"BEQ zero, foo\").\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "57ce5d3eefacfaadfe2ed0a3a85713d1ae6287b9", "tree": "627915b42bd823231e06585855e74f4ccd6688cc", "parents": [ "07b480695d24d1c9f27bb60fd4b980ae87e8bc1e" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:18 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Use the \"move\" pseudo-instruction where applicable\n\nSome of the assembly code in the LoongArch port likely originated\nfrom a time when the assembler did not support pseudo-instructions like\n\"move\" or \"jr\", so the desugared form was used and readability suffers\n(to a minor degree) as a result.\n\nAs the upstream toolchain supports these pseudo-instructions from the\nbeginning, migrate the existing few usages to them for better\nreadability.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "07b480695d24d1c9f27bb60fd4b980ae87e8bc1e", "tree": "4f6d6209f66bcc90748dbf5b371e11595cdcf86b", "parents": [ "d8e7f201a4cf148c3801cdc9603963061d28d64f" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:17 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Use the \"jr\" pseudo-instruction where applicable\n\nSome of the assembly code in the LoongArch port likely originated\nfrom a time when the assembler did not support pseudo-instructions like\n\"move\" or \"jr\", so the desugared form was used and readability suffers\n(to a minor degree) as a result.\n\nAs the upstream toolchain supports these pseudo-instructions from the\nbeginning, migrate the existing few usages to them for better\nreadability.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "d8e7f201a4cf148c3801cdc9603963061d28d64f", "tree": "5bbe5881044a3dcb4ad37fc59a49f9c9e815c3db", "parents": [ "e0dccc3b76fb35bb257b4118367a883073d7390e" ], "author": { "name": "WANG Xuerui", "email": "git@xen0n.name", "time": "Tue Jul 26 23:57:16 2022 +0800" }, "committer": { "name": "Huacai Chen", "email": "chenhuacai@loongson.cn", "time": "Fri Jul 29 18:22:32 2022 +0800" }, "message": "LoongArch: Use ABI names of registers where appropriate\n\nSome of the assembly in the LoongArch port seem to come from a\nprehistoric time, when the assembler didn\u0027t even have support for the\nABI names we all come to know and love, thus used raw register numbers\nwhich hampered readability.\n\nThe usages are found with a regex match inside arch/loongarch, then\nmanually adjusted for those non-definitions.\n\nSigned-off-by: WANG Xuerui \u003cgit@xen0n.name\u003e\nSigned-off-by: Huacai Chen \u003cchenhuacai@loongson.cn\u003e\n" }, { "commit": "ec85bd369fd2bfaed6f45dd678706429d4f75b48", "tree": "f57a23467a04861d936011986cb005740199659f", "parents": [ "fb0fd3469ead5b937293c213daa1f589b4b7ce46" ], "author": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Tue Jul 26 23:51:48 2022 +0100" }, "committer": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Fri Jul 29 09:54:26 2022 +0100" }, "message": "ARM: findbit: fix overflowing offset\n\nWhen offset is larger than the size of the bit array, we should not\nattempt to access the array as we can perform an access beyond the\nend of the array. Fix this by changing the pre-condition.\n\nUsing \"cmp r2, r1; bhs ...\" covers us for the size \u003d\u003d 0 case, since\nthis will always take the branch when r1 is zero, irrespective of\nthe value of r2. This means we can fix this bug without adding any\nadditional code!\n\nTested-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\n" }, { "commit": "571c30b1a88465a1c85a6f7762609939b9085a15", "tree": "4a33b6d2120a2cfb9f132816b43199914d439113", "parents": [ "5bb6c1d1126ebcbcd6314f80d82f50b021a9e351" ], "author": { "name": "Thadeu Lima de Souza Cascardo", "email": "cascardo@canonical.com", "time": "Thu Jul 28 09:26:02 2022 -0300" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Fri Jul 29 10:02:35 2022 +0200" }, "message": "x86/bugs: Do not enable IBPB at firmware entry when IBPB is not available\n\nSome cloud hypervisors do not provide IBPB on very recent CPU processors,\nincluding AMD processors affected by Retbleed.\n\nUsing IBPB before firmware calls on such systems would cause a GPF at boot\nlike the one below. Do not enable such calls when IBPB support is not\npresent.\n\n EFI Variables Facility v0.08 2004-May-17\n general protection fault, maybe for address 0x1: 0000 [#1] PREEMPT SMP NOPTI\n CPU: 0 PID: 24 Comm: kworker/u2:1 Not tainted 5.19.0-rc8+ #7\n Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.0.0 02/06/2015\n Workqueue: efi_rts_wq efi_call_rts\n RIP: 0010:efi_call_rts\n Code: e8 37 33 58 ff 41 bf 48 00 00 00 49 89 c0 44 89 f9 48 83 c8 01 4c 89 c2 48 c1 ea 20 66 90 b9 49 00 00 00 b8 01 00 00 00 31 d2 \u003c0f\u003e 30 e8 7b 9f 5d ff e8 f6 f8 ff ff 4c 89 f1 4c 89 ea 4c 89 e6 48\n RSP: 0018:ffffb373800d7e38 EFLAGS: 00010246\n RAX: 0000000000000001 RBX: 0000000000000006 RCX: 0000000000000049\n RDX: 0000000000000000 RSI: ffff94fbc19d8fe0 RDI: ffff94fbc1b2b300\n RBP: ffffb373800d7e70 R08: 0000000000000000 R09: 0000000000000000\n R10: 000000000000000b R11: 000000000000000b R12: ffffb3738001fd78\n R13: ffff94fbc2fcfc00 R14: ffffb3738001fd80 R15: 0000000000000048\n FS: 0000000000000000(0000) GS:ffff94fc3da00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: ffff94fc30201000 CR3: 000000006f610000 CR4: 00000000000406f0\n Call Trace:\n \u003cTASK\u003e\n ? __wake_up\n process_one_work\n worker_thread\n ? rescuer_thread\n kthread\n ? kthread_complete_and_exit\n ret_from_fork\n \u003c/TASK\u003e\n Modules linked in:\n\nFixes: 28a99e95f55c (\"x86/amd: Use IBPB for firmware calls\")\nReported-by: Dimitri John Ledkov \u003cdimitri.ledkov@canonical.com\u003e\nSigned-off-by: Thadeu Lima de Souza Cascardo \u003ccascardo@canonical.com\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: https://lore.kernel.org/r/20220728122602.2500509-1-cascardo@canonical.com\n" }, { "commit": "6e2c0490769ef8a95b61304389116ccc85c53e12", "tree": "598d79be81d223c2eacea35a2f381f68db85a6fe", "parents": [ "33ea1340bafe1f394e5bf96fceef73e9771d066b", "f16a2f593d0095e82e6b7f9d776f869c8ab45952" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 20:34:59 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 20:34:59 2022 -0700" }, "message": "Merge tag \u0027drm-fixes-2022-07-29\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fix from Dave Airlie:\n \"Quiet extra week, just a single fix for i915 workaround with execlist\n backend.\n\n i915:\n\n - Further reset robustness improvements for execlists [Wa_22011802037]\"\n\n* tag \u0027drm-fixes-2022-07-29\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/i915/reset: Add additional steps for Wa_22011802037 for execlist backend\n" }, { "commit": "f16a2f593d0095e82e6b7f9d776f869c8ab45952", "tree": "d3c6821767840645887e1cfd3929dd6fb901d889", "parents": [ "e0dccc3b76fb35bb257b4118367a883073d7390e", "a7a47a5dfa9a9692a41764ee9ab4054f12924a42" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 29 11:39:13 2022 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Jul 29 11:39:13 2022 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2022-07-28-1\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\n- Further reset robustness improvements for execlists [Wa_22011802037] (Umesh Nerlige Ramappa)\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/YuJIWaEbKcs/q0NY@tursulin-desk\n" }, { "commit": "66cee9097e2b74ff3c8cc040ce5717c521a0c3fa", "tree": "5da17daacb6baf645430a1741634464a2738f2ad", "parents": [ "0c09bc33aa8e9dc867300acaadc318c2f0d85a1e" ], "author": { "name": "Alistair Popple", "email": "apopple@nvidia.com", "time": "Wed Jul 20 16:27:45 2022 +1000" }, "committer": { "name": "Lyude Paul", "email": "lyude@redhat.com", "time": "Thu Jul 28 16:43:31 2022 -0400" }, "message": "nouveau/svm: Fix to migrate all requested pages\n\nUsers may request that pages from an OpenCL SVM allocation be migrated\nto the GPU with clEnqueueSVMMigrateMem(). In Nouveau this will call into\nnouveau_dmem_migrate_vma() to do the migration. If the total range to be\nmigrated exceeds SG_MAX_SINGLE_ALLOC the pages will be migrated in\nchunks of size SG_MAX_SINGLE_ALLOC. However a typo in updating the\nstarting address means that only the first chunk will get migrated.\n\nFix the calculation so that the entire range will get migrated if\npossible.\n\nSigned-off-by: Alistair Popple \u003capopple@nvidia.com\u003e\nFixes: e3d8b0890469 (\"drm/nouveau/svm: map pages after migration\")\nReviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nReviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\nSigned-off-by: Lyude Paul \u003clyude@redhat.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20220720062745.960701-1-apopple@nvidia.com\nCc: \u003cstable@vger.kernel.org\u003e # v5.8+\n" }, { "commit": "33ea1340bafe1f394e5bf96fceef73e9771d066b", "tree": "3b1d816103a2d02692e4845e5c4570ed6f4afecd", "parents": [ "e64ab2dbd882933b65cd82ff6235d705ad65dbb6", "4d3d3a1b244fd54629a6b7047f39a7bbc8d11910" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 11:54:59 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 11:54:59 2022 -0700" }, "message": "Merge tag \u0027net-5.19-final\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from Jakub Kicinski:\n \"Including fixes from bluetooth and netfilter, no known blockers for\n the release.\n\n Current release - regressions:\n\n - wifi: mac80211: do not abuse fq.lock in ieee80211_do_stop(), fix\n taking the lock before its initialized\n\n - Bluetooth: mgmt: fix double free on error path\n\n Current release - new code bugs:\n\n - eth: ice: fix tunnel checksum offload with fragmented traffic\n\n Previous releases - regressions:\n\n - tcp: md5: fix IPv4-mapped support after refactoring, don\u0027t take the\n pure v6 path\n\n - Revert \"tcp: change pingpong threshold to 3\", improving detection\n of interactive sessions\n\n - mld: fix netdev refcount leak in mld_{query | report}_work() due to\n a race\n\n - Bluetooth:\n - always set event mask on suspend, avoid early wake ups\n - L2CAP: fix use-after-free caused by l2cap_chan_put\n\n - bridge: do not send empty IFLA_AF_SPEC attribute\n\n Previous releases - always broken:\n\n - ping6: fix memleak in ipv6_renew_options()\n\n - sctp: prevent null-deref caused by over-eager error paths\n\n - virtio-net: fix the race between refill work and close, resulting\n in NAPI scheduled after close and a BUG()\n\n - macsec:\n - fix three netlink parsing bugs\n - avoid breaking the device state on invalid change requests\n - fix a memleak in another error path\n\n Misc:\n\n - dt-bindings: net: ethernet-controller: rework \u0027fixed-link\u0027 schema\n\n - two more batches of sysctl data race adornment\"\n\n* tag \u0027net-5.19-final\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (67 commits)\n stmmac: dwmac-mediatek: fix resource leak in probe\n ipv6/addrconf: fix a null-ptr-deref bug for ip6_ptr\n net: ping6: Fix memleak in ipv6_renew_options().\n net/funeth: Fix fun_xdp_tx() and XDP packet reclaim\n sctp: leave the err path free in sctp_stream_init to sctp_stream_free\n sfc: disable softirqs for ptp TX\n ptp: ocp: Select CRC16 in the Kconfig.\n tcp: md5: fix IPv4-mapped support\n virtio-net: fix the race between refill work and close\n mptcp: Do not return EINPROGRESS when subflow creation succeeds\n Bluetooth: L2CAP: Fix use-after-free caused by l2cap_chan_put\n Bluetooth: Always set event mask on suspend\n Bluetooth: mgmt: Fix double free on error path\n wifi: mac80211: do not abuse fq.lock in ieee80211_do_stop()\n ice: do not setup vlan for loopback VSI\n ice: check (DD | EOF) bits on Rx descriptor rather than (EOP | RS)\n ice: Fix VSIs unable to share unicast MAC\n ice: Fix tunnel checksum offload with fragmented traffic\n ice: Fix max VLANs available for VF\n netfilter: nft_queue: only allow supported familes and hooks\n ...\n" }, { "commit": "4d3d3a1b244fd54629a6b7047f39a7bbc8d11910", "tree": "86507bdbfcd1e69f12dcf756af11f3f9fb702dac", "parents": [ "85f0173df35e5462d89947135a6a5599c6c3ef6f" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Thu Jul 28 14:52:09 2022 +0300" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Thu Jul 28 10:43:04 2022 -0700" }, "message": "stmmac: dwmac-mediatek: fix resource leak in probe\n\nIf mediatek_dwmac_clks_config() fails, then call stmmac_remove_config_dt()\nbefore returning. Otherwise it is a resource leak.\n\nFixes: fa4b3ca60e80 (\"stmmac: dwmac-mediatek: fix clock issue\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nLink: https://lore.kernel.org/r/YuJ4aZyMUlG6yGGa@kili\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "85f0173df35e5462d89947135a6a5599c6c3ef6f", "tree": "f4b4fb6dee5fec51395e64d1964914f4731f0f1f", "parents": [ "e27326009a3d247b831eda38878c777f6f4eb3d1" ], "author": { "name": "Ziyang Xuan", "email": "william.xuanziyang@huawei.com", "time": "Thu Jul 28 09:33:07 2022 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Thu Jul 28 10:42:44 2022 -0700" }, "message": "ipv6/addrconf: fix a null-ptr-deref bug for ip6_ptr\n\nChange net device\u0027s MTU to smaller than IPV6_MIN_MTU or unregister\ndevice while matching route. That may trigger null-ptr-deref bug\nfor ip6_ptr probability as following.\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\nBUG: KASAN: null-ptr-deref in find_match.part.0+0x70/0x134\nRead of size 4 at addr 0000000000000308 by task ping6/263\n\nCPU: 2 PID: 263 Comm: ping6 Not tainted 5.19.0-rc7+ #14\nCall trace:\n dump_backtrace+0x1a8/0x230\n show_stack+0x20/0x70\n dump_stack_lvl+0x68/0x84\n print_report+0xc4/0x120\n kasan_report+0x84/0x120\n __asan_load4+0x94/0xd0\n find_match.part.0+0x70/0x134\n __find_rr_leaf+0x408/0x470\n fib6_table_lookup+0x264/0x540\n ip6_pol_route+0xf4/0x260\n ip6_pol_route_output+0x58/0x70\n fib6_rule_lookup+0x1a8/0x330\n ip6_route_output_flags_noref+0xd8/0x1a0\n ip6_route_output_flags+0x58/0x160\n ip6_dst_lookup_tail+0x5b4/0x85c\n ip6_dst_lookup_flow+0x98/0x120\n rawv6_sendmsg+0x49c/0xc70\n inet_sendmsg+0x68/0x94\n\nReproducer as following:\nFirstly, prepare conditions:\n$ip netns add ns1\n$ip netns add ns2\n$ip link add veth1 type veth peer name veth2\n$ip link set veth1 netns ns1\n$ip link set veth2 netns ns2\n$ip netns exec ns1 ip -6 addr add 2001:0db8:0:f101::1/64 dev veth1\n$ip netns exec ns2 ip -6 addr add 2001:0db8:0:f101::2/64 dev veth2\n$ip netns exec ns1 ifconfig veth1 up\n$ip netns exec ns2 ifconfig veth2 up\n$ip netns exec ns1 ip -6 route add 2000::/64 dev veth1 metric 1\n$ip netns exec ns2 ip -6 route add 2001::/64 dev veth2 metric 1\n\nSecondly, execute the following two commands in two ssh windows\nrespectively:\n$ip netns exec ns1 sh\n$while true; do ip -6 addr add 2001:0db8:0:f101::1/64 dev veth1; ip -6 route add 2000::/64 dev veth1 metric 1; ping6 2000::2; done\n\n$ip netns exec ns1 sh\n$while true; do ip link set veth1 mtu 1000; ip link set veth1 mtu 1500; sleep 5; done\n\nIt is because ip6_ptr has been assigned to NULL in addrconf_ifdown() firstly,\nthen ip6_ignore_linkdown() accesses ip6_ptr directly without NULL check.\n\n\tcpu0\t\t\tcpu1\nfib6_table_lookup\n__find_rr_leaf\n\t\t\taddrconf_notify [ NETDEV_CHANGEMTU ]\n\t\t\taddrconf_ifdown\n\t\t\tRCU_INIT_POINTER(dev-\u003eip6_ptr, NULL)\nfind_match\nip6_ignore_linkdown\n\nSo we can add NULL check for ip6_ptr before using in ip6_ignore_linkdown() to\nfix the null-ptr-deref bug.\n\nFixes: dcd1f572954f (\"net/ipv6: Remove fib6_idev\")\nSigned-off-by: Ziyang Xuan \u003cwilliam.xuanziyang@huawei.com\u003e\nReviewed-by: David Ahern \u003cdsahern@kernel.org\u003e\nLink: https://lore.kernel.org/r/20220728013307.656257-1-william.xuanziyang@huawei.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e27326009a3d247b831eda38878c777f6f4eb3d1", "tree": "06e744bc3fa3e2dfa08f917fe30351e029f8a77a", "parents": [ "51a83391d77bb0f7ff0aef06ca4c7f5aa9e80b4c" ], "author": { "name": "Kuniyuki Iwashima", "email": "kuniyu@amazon.com", "time": "Wed Jul 27 18:22:20 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Thu Jul 28 10:42:08 2022 -0700" }, "message": "net: ping6: Fix memleak in ipv6_renew_options().\n\nWhen we close ping6 sockets, some resources are left unfreed because\npingv6_prot is missing sk-\u003esk_prot-\u003edestroy(). As reported by\nsyzbot [0], just three syscalls leak 96 bytes and easily cause OOM.\n\n struct ipv6_sr_hdr *hdr;\n char data[24] \u003d {0};\n int fd;\n\n hdr \u003d (struct ipv6_sr_hdr *)data;\n hdr-\u003ehdrlen \u003d 2;\n hdr-\u003etype \u003d IPV6_SRCRT_TYPE_4;\n\n fd \u003d socket(AF_INET6, SOCK_DGRAM, NEXTHDR_ICMP);\n setsockopt(fd, IPPROTO_IPV6, IPV6_RTHDR, data, 24);\n close(fd);\n\nTo fix memory leaks, let\u0027s add a destroy function.\n\nNote the socket() syscall checks if the GID is within the range of\nnet.ipv4.ping_group_range. The default value is [1, 0] so that no\nGID meets the condition (1 \u003c\u003d GID \u003c\u003d 0). Thus, the local DoS does\nnot succeed until we change the default value. However, at least\nUbuntu/Fedora/RHEL loosen it.\n\n $ cat /usr/lib/sysctl.d/50-default.conf\n ...\n -net.ipv4.ping_group_range \u003d 0 2147483647\n\nAlso, there could be another path reported with these options, and\nsome of them require CAP_NET_RAW.\n\n setsockopt\n IPV6_ADDRFORM (inet6_sk(sk)-\u003epktoptions)\n IPV6_RECVPATHMTU (inet6_sk(sk)-\u003erxpmtu)\n IPV6_HOPOPTS (inet6_sk(sk)-\u003eopt)\n IPV6_RTHDRDSTOPTS (inet6_sk(sk)-\u003eopt)\n IPV6_RTHDR (inet6_sk(sk)-\u003eopt)\n IPV6_DSTOPTS (inet6_sk(sk)-\u003eopt)\n IPV6_2292PKTOPTIONS (inet6_sk(sk)-\u003eopt)\n\n getsockopt\n IPV6_FLOWLABEL_MGR (inet6_sk(sk)-\u003eipv6_fl_list)\n\nFor the record, I left a different splat with syzbot\u0027s one.\n\n unreferenced object 0xffff888006270c60 (size 96):\n comm \"repro2\", pid 231, jiffies 4294696626 (age 13.118s)\n hex dump (first 32 bytes):\n 01 00 00 00 44 00 00 00 00 00 00 00 00 00 00 00 ....D...........\n 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n backtrace:\n [\u003c00000000f6bc7ea9\u003e] sock_kmalloc (net/core/sock.c:2564 net/core/sock.c:2554)\n [\u003c000000006d699550\u003e] do_ipv6_setsockopt.constprop.0 (net/ipv6/ipv6_sockglue.c:715)\n [\u003c00000000c3c3b1f5\u003e] ipv6_setsockopt (net/ipv6/ipv6_sockglue.c:1024)\n [\u003c000000007096a025\u003e] __sys_setsockopt (net/socket.c:2254)\n [\u003c000000003a8ff47b\u003e] __x64_sys_setsockopt (net/socket.c:2265 net/socket.c:2262 net/socket.c:2262)\n [\u003c000000007c409dcb\u003e] do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80)\n [\u003c00000000e939c4a9\u003e] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120)\n\n[0]: https://syzkaller.appspot.com/bug?extid\u003da8430774139ec3ab7176\n\nFixes: 6d0bfe226116 (\"net: ipv6: Add IPv6 support to the ping socket.\")\nReported-by: syzbot+a8430774139ec3ab7176@syzkaller.appspotmail.com\nReported-by: Ayushman Dutta \u003cayudutta@amazon.com\u003e\nSigned-off-by: Kuniyuki Iwashima \u003ckuniyu@amazon.com\u003e\nReviewed-by: David Ahern \u003cdsahern@kernel.org\u003e\nReviewed-by: Eric Dumazet \u003cedumazet@google.com\u003e\nLink: https://lore.kernel.org/r/20220728012220.46918-1-kuniyu@amazon.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e64ab2dbd882933b65cd82ff6235d705ad65dbb6", "tree": "6207812d3aa830b193c472afc934424f8a45dd08", "parents": [ "e0339f036ef4beb9b20f0b6532a1e0ece7f594c6" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 10:31:12 2022 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 10:06:49 2022 -0700" }, "message": "watch_queue: Fix missing locking in add_watch_to_object()\n\nIf a watch is being added to a queue, it needs to guard against\ninterference from addition of a new watch, manual removal of a watch and\nremoval of a watch due to some other queue being destroyed.\n\nKEYCTL_WATCH_KEY guards against this for the same {key,queue} pair by\nholding the key-\u003esem writelocked and by holding refs on both the key and\nthe queue - but that doesn\u0027t prevent interaction from other {key,queue}\npairs.\n\nWhile add_watch_to_object() does take the spinlock on the event queue,\nit doesn\u0027t take the lock on the source\u0027s watch list. The assumption was\nthat the caller would prevent that (say by taking key-\u003esem) - but that\ndoesn\u0027t prevent interference from the destruction of another queue.\n\nFix this by locking the watcher list in add_watch_to_object().\n\nFixes: c73be61cede5 (\"pipe: Add general notification queue support\")\nReported-by: syzbot+03d7b43290037d1f87ca@syzkaller.appspotmail.com\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\ncc: keyrings@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "e0339f036ef4beb9b20f0b6532a1e0ece7f594c6", "tree": "2d768f0d5ac77df7c883295ba1842ab09cb38c70", "parents": [ "6e7765cb477a9753670d4351d14de93f1e9dbbd4" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Thu Jul 28 10:31:06 2022 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 28 10:06:49 2022 -0700" }, "message": "watch_queue: Fix missing rcu annotation\n\nSince __post_watch_notification() walks wlist-\u003ewatchers with only the\nRCU read lock held, we need to use RCU methods to add to the list (we\nalready use RCU methods to remove from the list).\n\nFix add_watch_to_object() to use hlist_add_head_rcu() instead of\nhlist_add_head() for that list.\n\nFixes: c73be61cede5 (\"pipe: Add general notification queue support\")\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "51a83391d77bb0f7ff0aef06ca4c7f5aa9e80b4c", "tree": "36686d10aaf8f1c24dac791cde91cb24e61b49dd", "parents": [ "bf84719df765f1a832a770f3cd0cdc8bee20c91f" ], "author": { "name": "Dimitris Michailidis", "email": "d.michailidis@fungible.com", "time": "Tue Jul 26 14:59:23 2022 -0700" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Thu Jul 28 12:54:10 2022 +0200" }, "message": "net/funeth: Fix fun_xdp_tx() and XDP packet reclaim\n\nThe current implementation of fun_xdp_tx(), used for XPD_TX, is\nincorrect in that it takes an address/length pair and later releases it\nwith page_frag_free(). It is OK for XDP_TX but the same code is used by\nndo_xdp_xmit. In that case it loses the XDP memory type and releases the\npacket incorrectly for some of the types. Assorted breakage follows.\n\nChange fun_xdp_tx() to take xdp_frame and rely on xdp_return_frame() in\nreclaim.\n\nFixes: db37bc177dae (\"net/funeth: add the data path\")\nSigned-off-by: Dimitris Michailidis \u003cdmichail@fungible.com\u003e\nLink: https://lore.kernel.org/r/20220726215923.7887-1-dmichail@fungible.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "bf84719df765f1a832a770f3cd0cdc8bee20c91f", "tree": "09d45d974928b980e89791e97c9df99f3c682b74", "parents": [ "181d8d2066c000ba0a0e6940a7ad80f1a0e68e9d", "cc019545a238518fa9da1e2a889f6e1bb1005a63" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Jul 27 19:56:28 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Jul 27 19:56:28 2022 -0700" }, "message": "Merge branch \u0027100GbE\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue\n\nTony Nguyen says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nIntel Wired LAN Driver Updates 2022-07-26\n\nThis series contains updates to ice driver only.\n\nPrzemyslaw corrects accounting for VF VLANs to allow for correct number\nof VLANs for untrusted VF. He also correct issue with checksum offload\non VXLAN tunnels.\n\nAni allows for two VSIs to share the same MAC address.\n\nMaciej corrects checked bits for descriptor completion of loopback\n\n* \u0027100GbE\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:\n ice: do not setup vlan for loopback VSI\n ice: check (DD | EOF) bits on Rx descriptor rather than (EOP | RS)\n ice: Fix VSIs unable to share unicast MAC\n ice: Fix tunnel checksum offload with fragmented traffic\n ice: Fix max VLANs available for VF\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20220726204646.2171589-1-anthony.l.nguyen@intel.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "181d8d2066c000ba0a0e6940a7ad80f1a0e68e9d", "tree": "0a6e824f8046ec6563de603554f86ddd1fecd093", "parents": [ "67c3b611d92fc238c43734878bc3e232ab570c79" ], "author": { "name": "Xin Long", "email": "lucien.xin@gmail.com", "time": "Mon Jul 25 18:11:06 2022 -0400" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Jul 27 18:23:22 2022 -0700" }, "message": "sctp: leave the err path free in sctp_stream_init to sctp_stream_free\n\nA NULL pointer dereference was reported by Wei Chen:\n\n BUG: kernel NULL pointer dereference, address: 0000000000000000\n RIP: 0010:__list_del_entry_valid+0x26/0x80\n Call Trace:\n \u003cTASK\u003e\n sctp_sched_dequeue_common+0x1c/0x90\n sctp_sched_prio_dequeue+0x67/0x80\n __sctp_outq_teardown+0x299/0x380\n sctp_outq_free+0x15/0x20\n sctp_association_free+0xc3/0x440\n sctp_do_sm+0x1ca7/0x2210\n sctp_assoc_bh_rcv+0x1f6/0x340\n\nThis happens when calling sctp_sendmsg without connecting to server first.\nIn this case, a data chunk already queues up in send queue of client side\nwhen processing the INIT_ACK from server in sctp_process_init() where it\ncalls sctp_stream_init() to alloc stream_in. If it fails to alloc stream_in\nall stream_out will be freed in sctp_stream_init\u0027s err path. Then in the\nasoc freeing it will crash when dequeuing this data chunk as stream_out\nis missing.\n\nAs we can\u0027t free stream out before dequeuing all data from send queue, and\nthis patch is to fix it by moving the err path stream_out/in freeing in\nsctp_stream_init() to sctp_stream_free() which is eventually called when\nfreeing the asoc in sctp_association_free(). This fix also makes the code\nin sctp_process_init() more clear.\n\nNote that in sctp_association_init() when it fails in sctp_stream_init(),\nsctp_association_free() will not be called, and in that case it should\ngo to \u0027stream_free\u0027 err path to free stream instead of \u0027fail_init\u0027.\n\nFixes: 5bbbbe32a431 (\"sctp: introduce stream scheduler foundations\")\nReported-by: Wei Chen \u003charperchen1110@gmail.com\u003e\nSigned-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\nLink: https://lore.kernel.org/r/831a3dc100c4908ff76e5bcc363be97f2778bc0b.1658787066.git.lucien.xin@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "67c3b611d92fc238c43734878bc3e232ab570c79", "tree": "6bd0db5ae924e9a78c7b184fd830f431c60437f0", "parents": [ "0c104556267242d922a3def60be8092b280e4fee" ], "author": { "name": "Alejandro Lucero", "email": "alejandro.lucero-palau@amd.com", "time": "Tue Jul 26 08:45:04 2022 +0200" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Jul 27 18:20:43 2022 -0700" }, "message": "sfc: disable softirqs for ptp TX\n\nSending a PTP packet can imply to use the normal TX driver datapath but\ninvoked from the driver\u0027s ptp worker. The kernel generic TX code\ndisables softirqs and preemption before calling specific driver TX code,\nbut the ptp worker does not. Although current ptp driver functionality\ndoes not require it, there are several reasons for doing so:\n\n 1) The invoked code is always executed with softirqs disabled for non\n PTP packets.\n 2) Better if a ptp packet transmission is not interrupted by softirq\n handling which could lead to high latencies.\n 3) netdev_xmit_more used by the TX code requires preemption to be\n disabled.\n\nIndeed a solution for dealing with kernel preemption state based on static\nkernel configuration is not possible since the introduction of dynamic\npreemption level configuration at boot time using the static calls\nfunctionality.\n\nFixes: f79c957a0b537 (\"drivers: net: sfc: use netdev_xmit_more helper\")\nSigned-off-by: Alejandro Lucero \u003calejandro.lucero-palau@amd.com\u003e\nLink: https://lore.kernel.org/r/20220726064504.49613-1-alejandro.lucero-palau@amd.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "0c104556267242d922a3def60be8092b280e4fee", "tree": "9101d33095b04612ddd4a4d3daadd5c2b87c0714", "parents": [ "e62d2e110356093c034998e093675df83057e511" ], "author": { "name": "Jonathan Lemon", "email": "jonathan.lemon@gmail.com", "time": "Tue Jul 26 15:06:04 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Jul 27 18:11:34 2022 -0700" }, "message": "ptp: ocp: Select CRC16 in the Kconfig.\n\nThe crc16() function is used to check the firmware validity, but\nthe library was not explicitly selected.\n\nFixes: 3c3673bde50c (\"ptp: ocp: Add firmware header checks\")\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nSigned-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nAcked-by: Vadim Fedorenko \u003cvadfed@fb.com\u003e\nLink: https://lore.kernel.org/r/20220726220604.1339972-1-jonathan.lemon@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "8dc592c41f38735306d1f1dc0b183601379c6d94", "tree": "c61c08fca7c1ac67fcdf5a726e669bba71946760", "parents": [ "25c2a075eb6a3031813b6051bd10dfc22c36a2a4" ], "author": { "name": "Jernej Skrabec", "email": "jernej.skrabec@gmail.com", "time": "Tue Jul 19 20:37:25 2022 +0200" }, "committer": { "name": "Stephen Boyd", "email": "sboyd@kernel.org", "time": "Wed Jul 27 16:45:58 2022 -0700" }, "message": "clk: sunxi-ng: Fix H6 RTC clock definition\n\nWhile RTC clock was added in H616 ccu_common list, it was not in H6\nlist. That caused invalid pointer dereference like this:\n\nUnable to handle kernel NULL pointer dereference at virtual address 000000000000020c\nMem abort info:\n ESR \u003d 0x96000004\n EC \u003d 0x25: DABT (current EL), IL \u003d 32 bits\n SET \u003d 0, FnV \u003d 0\n EA \u003d 0, S1PTW \u003d 0\n FSC \u003d 0x04: level 0 translation fault\nData abort info:\n ISV \u003d 0, ISS \u003d 0x00000004\n CM \u003d 0, WnR \u003d 0\nuser pgtable: 4k pages, 48-bit VAs, pgdp\u003d000000004d574000\n[000000000000020c] pgd\u003d0000000000000000, p4d\u003d0000000000000000\nInternal error: Oops: 96000004 [#1] PREEMPT SMP\nCPU: 3 PID: 339 Comm: cat Tainted: G B 5.18.0-rc1+ #1352\nHardware name: Tanix TX6 (DT)\npstate: 00000005 (nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE\u003d--)\npc : ccu_gate_is_enabled+0x48/0x74\nlr : ccu_gate_is_enabled+0x40/0x74\nsp : ffff80000c0b76d0\nx29: ffff80000c0b76d0 x28: 00000000016e3600 x27: 0000000000000000\nx26: 0000000000000000 x25: 0000000000000002 x24: ffff00000952fe08\nx23: ffff800009611400 x22: ffff00000952fe79 x21: 0000000000000000\nx20: 0000000000000001 x19: ffff80000aad6f08 x18: 0000000000000000\nx17: 2d2d2d2d2d2d2d2d x16: 2d2d2d2d2d2d2d2d x15: 2d2d2d2d2d2d2d2d\nx14: 0000000000000000 x13: 00000000f2f2f2f2 x12: ffff700001816e89\nx11: 1ffff00001816e88 x10: ffff700001816e88 x9 : dfff800000000000\nx8 : ffff80000c0b7447 x7 : 0000000000000001 x6 : ffff700001816e88\nx5 : ffff80000c0b7440 x4 : 0000000000000001 x3 : ffff800008935c50\nx2 : dfff800000000000 x1 : 0000000000000000 x0 : 000000000000020c\nCall trace:\n ccu_gate_is_enabled+0x48/0x74\n clk_core_is_enabled+0x7c/0x1c0\n clk_summary_show_subtree+0x1dc/0x334\n clk_summary_show_subtree+0x250/0x334\n clk_summary_show_subtree+0x250/0x334\n clk_summary_show_subtree+0x250/0x334\n clk_summary_show_subtree+0x250/0x334\n clk_summary_show+0x90/0xdc\n seq_read_iter+0x248/0x6d4\n seq_read+0x17c/0x1fc\n full_proxy_read+0x90/0xf0\n vfs_read+0xdc/0x28c\n ksys_read+0xc8/0x174\n __arm64_sys_read+0x44/0x5c\n invoke_syscall+0x60/0x190\n el0_svc_common.constprop.0+0x7c/0x160\n do_el0_svc+0x38/0xa0\n el0_svc+0x68/0x160\n el0t_64_sync_handler+0x10c/0x140\n el0t_64_sync+0x18c/0x190\nCode: d1006260 97e5c981 785e8260 8b0002a0 (b9400000)\n---[ end trace 0000000000000000 ]---\n\nFix that by adding rtc clock to H6 ccu_common list too.\n\nFixes: 38d321b61bda (\"clk: sunxi-ng: h6-r: Add RTC gate clock\")\nSigned-off-by: Jernej Skrabec \u003cjernej.skrabec@gmail.com\u003e\nLink: https://lore.kernel.org/r/20220719183725.2605141-1-jernej.skrabec@gmail.com\nReviewed-by: Samuel Holland \u003csamuel@sholland.org\u003e\nSigned-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n" }, { "commit": "e62d2e110356093c034998e093675df83057e511", "tree": "1f814c250ba21d175b77c33f04156427a2ef22d1", "parents": [ "5a159128faff151b7fe5f4eb0f310b1e0a2d56bf" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue Jul 26 11:57:43 2022 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Wed Jul 27 10:18:21 2022 -0700" }, "message": "tcp: md5: fix IPv4-mapped support\n\nAfter the blamed commit, IPv4 SYN packets handled\nby a dual stack IPv6 socket are dropped, even if\nperfectly valid.\n\n$ nstat | grep MD5\nTcpExtTCPMD5Failure 5 0.0\n\nFor a dual stack listener, an incoming IPv4 SYN packet\nwould call tcp_inbound_md5_hash() with @family \u003d\u003d AF_INET,\nwhile tp-\u003eaf_specific is pointing to tcp_sock_ipv6_specific.\n\nOnly later when an IPv4-mapped child is created, tp-\u003eaf_specific\nis changed to tcp_sock_ipv6_mapped_specific.\n\nFixes: 7bbb765b7349 (\"net/tcp: Merge TCP-MD5 inbound callbacks\")\nReported-by: Brian Vazquez \u003cbrianvv@google.com\u003e\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReviewed-by: David Ahern \u003cdsahern@kernel.org\u003e\nReviewed-by: Dmitry Safonov \u003cdima@arista.com\u003e\nTested-by: Leonard Crestez \u003ccdleonard@gmail.com\u003e\nLink: https://lore.kernel.org/r/20220726115743.2759832-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "fb0fd3469ead5b937293c213daa1f589b4b7ce46", "tree": "92382cedc981a203680589e7d5d69c18a5315ce3", "parents": [ "29589ca09a74cfc0c50ad002e298bf4b8e69e0bd" ], "author": { "name": "Florian Fainelli", "email": "f.fainelli@gmail.com", "time": "Tue Jul 19 17:33:21 2022 +0100" }, "committer": { "name": "Russell King (Oracle)", "email": "rmk+kernel@armlinux.org.uk", "time": "Wed Jul 27 17:55:01 2022 +0100" }, "message": "ARM: 9216/1: Fix MAX_DMA_ADDRESS overflow\n\nCommit 26f09e9b3a06 (\"mm/memblock: add memblock memory allocation apis\")\nadded a check to determine whether arm_dma_zone_size is exceeding the\namount of kernel virtual address space available between the upper 4GB\nvirtual address limit and PAGE_OFFSET in order to provide a suitable\ndefinition of MAX_DMA_ADDRESS that should fit within the 32-bit virtual\naddress space. The quantity used for comparison was off by a missing\ntrailing 0, leading to MAX_DMA_ADDRESS to be overflowing a 32-bit\nquantity.\n\nThis was caught thanks to CONFIG_DEBUG_VIRTUAL on the bcm2711 platform\nwhere we define a dma_zone_size of 1GB and we have a PAGE_OFFSET value\nof 0xc000_0000 (CONFIG_VMSPLIT_3G) leading to MAX_DMA_ADDRESS being\n0x1_0000_0000 which overflows the unsigned long type used throughout\n__pa() and then __virt_addr_valid(). Because the virtual address passed\nto __virt_addr_valid() would now be 0, the function would loudly warn\nand flood the kernel log, thus making the platform unable to boot\nproperly.\n\nFixes: 26f09e9b3a06 (\"mm/memblock: add memblock memory allocation apis\")\nSigned-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nReviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Russell King (Oracle) \u003crmk+kernel@armlinux.org.uk\u003e\n" }, { "commit": "6e7765cb477a9753670d4351d14de93f1e9dbbd4", "tree": "69dd8df8a2a26df07fd386e950bd3153c4945973", "parents": [ "9d8a8616ee47f478a9f78ab97f55c3fbf71bb5f0", "e2a619ca0b38f2114347b7078b8a67d72d457a3d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jul 27 09:50:18 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jul 27 09:50:18 2022 -0700" }, "message": "Merge tag \u0027asm-generic-fixes-5.19-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic\n\nPull asm-generic fixes from Arnd Bergmann:\n \"Two more bug fixes for asm-generic, one addressing an incorrect\n Kconfig symbol reference and another one fixing a build failure for\n the perf tool on mips and possibly others\"\n\n* tag \u0027asm-generic-fixes-5.19-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:\n asm-generic: remove a broken and needless ifdef conditional\n tools: Fixed MIPS builds due to struct flock re-definition\n" }, { "commit": "9d8a8616ee47f478a9f78ab97f55c3fbf71bb5f0", "tree": "a6b8495a75dc82132efb1640ce4571af0043e75f", "parents": [ "39c3c396f8131f3db454c80e0fcfcdc54ed9ec01", "c5cdb9286913aa5a5ebb81bcca0c17df3b0e2c79" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jul 27 09:43:07 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Jul 27 09:43:07 2022 -0700" }, "message": "Merge tag \u0027soc-fixes-5.19-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc\n\nPull ARM SoC fixes from Arnd Bergmann:\n \"One last set of changes for the soc tree:\n\n - fix clock frequency on lan966x\n\n - fix incorrect GPIO numbers on some pxa machines\n\n - update Baolin\u0027s email address\"\n\n* tag \u0027soc-fixes-5.19-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:\n ARM: pxa2xx: Fix GPIO descriptor tables\n mailmap: update Baolin Wang\u0027s email\n ARM: dts: lan966x: fix sys_clk frequency\n" }, { "commit": "5bb6c1d1126ebcbcd6314f80d82f50b021a9e351", "tree": "bbcee8b20dea7ab814aa985d1164eea9f9331771", "parents": [ "871808fd6981bcc6bb48f71032f983ca77748e96" ], "author": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Wed Jul 27 13:24:21 2022 +0200" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Wed Jul 27 18:09:13 2022 +0200" }, "message": "Revert \"x86/sev: Expose sev_es_ghcb_hv_call() for use by HyperV\"\n\nThis reverts commit 007faec014cb5d26983c1f86fd08c6539b41392e.\n\nNow that hyperv does its own protocol negotiation:\n\n 49d6a3c062a1 (\"x86/Hyper-V: Add SEV negotiate protocol support in Isolation VM\")\n\nrevert this exposure of the sev_es_ghcb_hv_call() helper.\n\nCc: Wei Liu \u003cwei.liu@kernel.org\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nReviewed-by：Tianyu Lan \u003ctiala@microsoft.com\u003e\nLink: https://lore.kernel.org/r/20220614014553.1915929-1-ltykernel@gmail.com\n" }, { "commit": "871808fd6981bcc6bb48f71032f983ca77748e96", "tree": "13b52d457c35b62ad06dbdfcfdba9bc67a9d8a67", "parents": [ "e0dccc3b76fb35bb257b4118367a883073d7390e" ], "author": { "name": "Lukas Bulwahn", "email": "lukas.bulwahn@gmail.com", "time": "Fri Jul 22 14:18:15 2022 +0200" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Wed Jul 27 18:09:11 2022 +0200" }, "message": "x86/configs: Update configs in x86_debug.config\n\nCommit\n\n 4675ff05de2d (\"kmemcheck: rip it out\")\n\nremoved kmemcheck and its corresponding build config KMEMCHECK.\n\nCommit\n\n 0f620cefd775 (\"objtool: Rename \"VMLINUX_VALIDATION\" -\u003e \"NOINSTR_VALIDATION\"\")\n\nrenamed the debug config option.\n\nAdjust x86_debug.config to those changes in debug configs.\n\nSigned-off-by: Lukas Bulwahn \u003clukas.bulwahn@gmail.com\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nLink: https://lore.kernel.org/r/20220722121815.27535-1-lukas.bulwahn@gmail.com\n" }, { "commit": "eda3953b6a805d6df87a4c51058493ec88bfc622", "tree": "e368d405febd9cbc15720d15f68959d69735c7bb", "parents": [ "82e094f7bd988c02df27f8c8d81af8f750660b2a", "d6c52fa3e955b97f8eb3ac824d2a3e0af147b3ce" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Jul 27 10:03:40 2022 -0600" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Jul 27 10:03:40 2022 -0600" }, "message": "Merge tag \u0027nvme-5.19-2022-07-27\u0027 of git://git.infradead.org/nvme into block-5.19\n\nPull NVMe fix from Christoph:\n\n\"nvme fix for Linux 5.19\n\n - yet another duplicate ID quirk (Tobias Gruetzmacher)\"\n\n* tag \u0027nvme-5.19-2022-07-27\u0027 of git://git.infradead.org/nvme:\n nvme-pci: Crucial P2 has bogus namespace ids\n" }, { "commit": "9a241805673ec0a826b7ddf84b00f4e03adb0a5e", "tree": "b0cd7e5590084cea678bd5760e0624ff54ccd4be", "parents": [ "882528d2e77687c3ef26abb9c490f77a9c1f6e1a" ], "author": { "name": "Ian Rogers", "email": "irogers@google.com", "time": "Tue Jul 26 15:09:21 2022 -0700" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Wed Jul 27 11:19:39 2022 -0300" }, "message": "perf bpf: Remove undefined behavior from bpf_perf_object__next()\n\nbpf_perf_object__next() folded the last element in the list test with the\nempty list test. However, this meant that offsets were computed against\nnull and that a struct list_head was compared against a \u0027struct\nbpf_perf_object\u0027.\n\nWorking around this with clang\u0027s undefined behavior sanitizer required\n-fno-sanitize\u003dnull and -fno-sanitize\u003dobject-size.\n\nRemove the undefined behavior by using the regular Linux list APIs and\nhandling the starting case separately from the end testing case.\n\nLooking at uses like bpf_perf_object__for_each(), as the constant NULL\nor non-NULL argument can be constant propagated, the code is no less\nefficient.\n\nSigned-off-by: Ian Rogers \u003cirogers@google.com\u003e\nCc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nCc: Alexei Starovoitov \u003cast@kernel.org\u003e\nCc: Andrii Nakryiko \u003candrii@kernel.org\u003e\nCc: Christy Lee \u003cchristylee@fb.com\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Miaoqian Lin \u003clinmq006@gmail.com\u003e\nCc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Nathan Chancellor \u003cnathan@kernel.org\u003e\nCc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Stephane Eranian \u003ceranian@google.com\u003e\nCc: Tom Rix \u003ctrix@redhat.com\u003e\nCc: bpf@vger.kernel.org\nCc: llvm@lists.linux.dev\nLink: https://lore.kernel.org/r/20220726220921.2567761-1-irogers@google.com\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n" }, { "commit": "882528d2e77687c3ef26abb9c490f77a9c1f6e1a", "tree": "6b02b93266429559cb90d58b922268cf192f831b", "parents": [ "2d86612aacb7805f72873691a2644d7279ed0630" ], "author": { "name": "Leo Yan", "email": "leo.yan@linaro.org", "time": "Sun Jul 24 14:00:13 2022 +0800" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Wed Jul 27 11:17:50 2022 -0300" }, "message": "perf symbol: Skip symbols if SHF_ALLOC flag is not set\n\nSome symbols are observed with the \u0027st_value\u0027 field zeroed. E.g.\nlibc.so.6 in Ubuntu contains a symbol \u0027__evoke_link_warning_getwd\u0027 which\nresides in the \u0027.gnu.warning.getwd\u0027 section.\n\nUnlike normal sections, such kind of sections are used for linker\nwarning when a file calls deprecated functions, but they are not part of\nmemory images, the symbols in these sections should be dropped.\n\nThis patch checks the section attribute SHF_ALLOC bit, if the bit is not\nset, it skips symbols to avoid spurious ones.\n\nSuggested-by: Fangrui Song \u003cmaskray@google.com\u003e\nSigned-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\nAcked-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nCc: Chang Rui \u003cchangruinj@gmail.com\u003e\nCc: Ian Rogers \u003cirogers@google.com\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: https://lore.kernel.org/r/20220724060013.171050-3-leo.yan@linaro.org\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n" }, { "commit": "2d86612aacb7805f72873691a2644d7279ed0630", "tree": "e9315213dc8be52879bc89dfdba4687a8e260ae7", "parents": [ "b226521923aee7051f4b24df9be5bf07d53f0a2b" ], "author": { "name": "Leo Yan", "email": "leo.yan@linaro.org", "time": "Sun Jul 24 14:00:12 2022 +0800" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Wed Jul 27 11:17:50 2022 -0300" }, "message": "perf symbol: Correct address for bss symbols\n\nWhen using \u0027perf mem\u0027 and \u0027perf c2c\u0027, an issue is observed that tool\nreports the wrong offset for global data symbols. This is a common\nissue on both x86 and Arm64 platforms.\n\nLet\u0027s see an example, for a test program, below is the disassembly for\nits .bss section which is dumped with objdump:\n\n ...\n\n Disassembly of section .bss:\n\n 0000000000004040 \u003ccompleted.0\u003e:\n \t...\n\n 0000000000004080 \u003cbuf1\u003e:\n \t...\n\n 00000000000040c0 \u003cbuf2\u003e:\n \t...\n\n 0000000000004100 \u003cthread\u003e:\n \t...\n\nFirst we used \u0027perf mem record\u0027 to run the test program and then used\n\u0027perf --debug verbose\u003d4 mem report\u0027 to observe what\u0027s the symbol info\nfor \u0027buf1\u0027 and \u0027buf2\u0027 structures.\n\n # ./perf mem record -e ldlat-loads,ldlat-stores -- false_sharing.exe 8\n # ./perf --debug verbose\u003d4 mem report\n ...\n dso__load_sym_internal: adjusting symbol: st_value: 0x40c0 sh_addr: 0x4040 sh_offset: 0x3028\n symbol__new: buf2 0x30a8-0x30e8\n ...\n dso__load_sym_internal: adjusting symbol: st_value: 0x4080 sh_addr: 0x4040 sh_offset: 0x3028\n symbol__new: buf1 0x3068-0x30a8\n ...\n\nThe perf tool relies on libelf to parse symbols, in executable and\nshared object files, \u0027st_value\u0027 holds a virtual address; \u0027sh_addr\u0027 is\nthe address at which section\u0027s first byte should reside in memory, and\n\u0027sh_offset\u0027 is the byte offset from the beginning of the file to the\nfirst byte in the section. The perf tool uses below formula to convert\na symbol\u0027s memory address to a file address:\n\n file_address \u003d st_value - sh_addr + sh_offset\n ^\n ` Memory address\n\nWe can see the final adjusted address ranges for buf1 and buf2 are\n[0x30a8-0x30e8) and [0x3068-0x30a8) respectively, apparently this is\nincorrect, in the code, the structure for \u0027buf1\u0027 and \u0027buf2\u0027 specifies\ncompiler attribute with 64-byte alignment.\n\nThe problem happens for \u0027sh_offset\u0027, libelf returns it as 0x3028 which\nis not 64-byte aligned, combining with disassembly, it\u0027s likely libelf\ndoesn\u0027t respect the alignment for .bss section, therefore, it doesn\u0027t\nreturn the aligned value for \u0027sh_offset\u0027.\n\nSuggested by Fangrui Song, ELF file contains program header which\ncontains PT_LOAD segments, the fields p_vaddr and p_offset in PT_LOAD\nsegments contain the execution info. A better choice for converting\nmemory address to file address is using the formula:\n\n file_address \u003d st_value - p_vaddr + p_offset\n\nThis patch introduces elf_read_program_header() which returns the\nprogram header based on the passed \u0027st_value\u0027, then it uses the formula\nabove to calculate the symbol file address; and the debugging log is\nupdated respectively.\n\nAfter applying the change:\n\n # ./perf --debug verbose\u003d4 mem report\n ...\n dso__load_sym_internal: adjusting symbol: st_value: 0x40c0 p_vaddr: 0x3d28 p_offset: 0x2d28\n symbol__new: buf2 0x30c0-0x3100\n ...\n dso__load_sym_internal: adjusting symbol: st_value: 0x4080 p_vaddr: 0x3d28 p_offset: 0x2d28\n symbol__new: buf1 0x3080-0x30c0\n ...\n\nFixes: f17e04afaff84b5c (\"perf report: Fix ELF symbol parsing\")\nReported-by: Chang Rui \u003cchangruinj@gmail.com\u003e\nSuggested-by: Fangrui Song \u003cmaskray@google.com\u003e\nSigned-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\nAcked-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nCc: Ian Rogers \u003cirogers@google.com\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: https://lore.kernel.org/r/20220724060013.171050-2-leo.yan@linaro.org\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n" }, { "commit": "b226521923aee7051f4b24df9be5bf07d53f0a2b", "tree": "3acee484f451b7b1672510994772a11fdd2b8e19", "parents": [ "553de6e1157df63fc6cdfe4573e04c8edcbe68f2" ], "author": { "name": "Leo Yan", "email": "leo.yan@linaro.org", "time": "Mon Jul 25 18:42:20 2022 +0800" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Wed Jul 27 11:17:50 2022 -0300" }, "message": "perf scripts python: Let script to be python2 compliant\n\nThe mainline kernel can be used for relative old distros, e.g. RHEL 7.\nThe distro doesn\u0027t upgrade from python2 to python3, this causes the\nbuilding error that the python script is not python2 compliant.\n\nTo fix the building failure, this patch changes from the python f-string\nformat to traditional string format.\n\nFixes: 12fdd6c009da0d02 (\"perf scripts python: Support Arm CoreSight trace data disassembly\")\nReported-by: Akemi Yagi \u003ctoracat@elrepo.org\u003e\nSigned-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\nCc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nCc: ElRepo \u003ccontact@elrepo.org\u003e\nCc: Ian Rogers \u003cirogers@google.com\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Leo Yan \u003cleo.yan@linaro.org\u003e\nCc: Mark Rutland \u003cmark.rutland@arm.com\u003e\nCc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nLink: https://lore.kernel.org/r/20220725104220.1106663-1-leo.yan@linaro.org\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n" }, { "commit": "553de6e1157df63fc6cdfe4573e04c8edcbe68f2", "tree": "ae62a4e12bdf4ca6579c1ed9285fafbc0e034e30", "parents": [ "39c3c396f8131f3db454c80e0fcfcdc54ed9ec01" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Thu Jul 01 13:39:15 2021 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Wed Jul 27 11:17:50 2022 -0300" }, "message": "tools headers cpufeatures: Sync with the kernel sources\n\nTo pick the changes from:\n\n 28a99e95f55c6185 (\"x86/amd: Use IBPB for firmware calls\")\n\nThis only causes these perf files to be rebuilt:\n\n CC /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o\n CC /tmp/build/perf/bench/mem-memset-x86-64-asm.o\n\nAnd addresses this perf build warning:\n\n Warning: Kernel ABI header at \u0027tools/arch/x86/include/asm/cpufeatures.h\u0027 differs from latest version at \u0027arch/x86/include/asm/cpufeatures.h\u0027\n diff -u tools/arch/x86/include/asm/cpufeatures.h arch/x86/include/asm/cpufeatures.h\n\nCc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Ian Rogers \u003cirogers@google.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\nLink: https://lore.kernel.org/lkml/Yt6oWce9UDAmBAtX@kernel.org\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n" }, { "commit": "5a159128faff151b7fe5f4eb0f310b1e0a2d56bf", "tree": "a0f49d7be71892bab4a779587848421ae72f13ab", "parents": [ "b5177ed92bf6f9d90a2493ed51c1327e088be1df" ], "author": { "name": "Jason Wang", "email": "jasowang@redhat.com", "time": "Mon Jul 25 15:21:59 2022 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Jul 27 13:20:44 2022 +0100" }, "message": "virtio-net: fix the race between refill work and close\n\nWe try using cancel_delayed_work_sync() to prevent the work from\nenabling NAPI. This is insufficient since we don\u0027t disable the source\nof the refill work scheduling. This means an NAPI poll callback after\ncancel_delayed_work_sync() can schedule the refill work then can\nre-enable the NAPI that leads to use-after-free [1].\n\nSince the work can enable NAPI, we can\u0027t simply disable NAPI before\ncalling cancel_delayed_work_sync(). So fix this by introducing a\ndedicated boolean to control whether or not the work could be\nscheduled from NAPI.\n\n[1]\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nBUG: KASAN: use-after-free in refill_work+0x43/0xd4\nRead of size 2 at addr ffff88810562c92e by task kworker/2:1/42\n\nCPU: 2 PID: 42 Comm: kworker/2:1 Not tainted 5.19.0-rc1+ #480\nHardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014\nWorkqueue: events refill_work\nCall Trace:\n \u003cTASK\u003e\n dump_stack_lvl+0x34/0x44\n print_report.cold+0xbb/0x6ac\n ? _printk+0xad/0xde\n ? refill_work+0x43/0xd4\n kasan_report+0xa8/0x130\n ? refill_work+0x43/0xd4\n refill_work+0x43/0xd4\n process_one_work+0x43d/0x780\n worker_thread+0x2a0/0x6f0\n ? process_one_work+0x780/0x780\n kthread+0x167/0x1a0\n ? kthread_exit+0x50/0x50\n ret_from_fork+0x22/0x30\n \u003c/TASK\u003e\n...\n\nFixes: b2baed69e605c (\"virtio_net: set/cancel work on ndo_open/ndo_stop\")\nSigned-off-by: Jason Wang \u003cjasowang@redhat.com\u003e\nAcked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\nReviewed-by: Xuan Zhuo \u003cxuanzhuo@linux.alibaba.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5e2805d5379619c4a2e3ae4994e73b36439f4bad", "tree": "c95a1f39b8ce01c55675480152241fb8d4d3bac5", "parents": [ "4bcffe941758ee17becb43af3b25487f848f6512" ], "author": { "name": "Toshi Kani", "email": "toshi.kani@hpe.com", "time": "Thu Jul 21 12:05:03 2022 -0600" }, "committer": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Wed Jul 27 10:42:52 2022 +0200" }, "message": "EDAC/ghes: Set the DIMM label unconditionally\n\nThe commit\n\n cb51a371d08e (\"EDAC/ghes: Setup DIMM label from DMI and use it in error reports\")\n\nenforced that both the bank and device strings passed to\ndimm_setup_label() are not NULL.\n\nHowever, there are BIOSes, for example on a\n\n HPE ProLiant DL360 Gen10/ProLiant DL360 Gen10, BIOS U32 03/15/2019\n\nwhich don\u0027t populate both strings:\n\n Handle 0x0020, DMI type 17, 84 bytes\n Memory Device\n Array Handle: 0x0013\n Error Information Handle: Not Provided\n Total Width: 72 bits\n Data Width: 64 bits\n Size: 32 GB\n Form Factor: DIMM\n Set: None\n Locator: PROC 1 DIMM 1 \u003c\u003d\u003d\u003d\u003d\u003d device\n Bank Locator: Not Specified \u003c\u003d\u003d\u003d\u003d\u003d bank\n\nThis results in a buffer overflow because ghes_edac_register() calls\nstrlen() on an uninitialized label, which had non-zero values left over\nfrom krealloc_array():\n\n detected buffer overflow in __fortify_strlen\n ------------[ cut here ]------------\n kernel BUG at lib/string_helpers.c:983!\n invalid opcode: 0000 [#1] PREEMPT SMP NOPTI\n CPU: 1 PID: 1 Comm: swapper/0 Tainted: G I 5.18.6-200.fc36.x86_64 #1\n Hardware name: HPE ProLiant DL360 Gen10/ProLiant DL360 Gen10, BIOS U32 03/15/2019\n RIP: 0010:fortify_panic\n ...\n Call Trace:\n \u003cTASK\u003e\n ghes_edac_register.cold\n ghes_probe\n platform_probe\n really_probe\n __driver_probe_device\n driver_probe_device\n __driver_attach\n ? __device_attach_driver\n bus_for_each_dev\n bus_add_driver\n driver_register\n acpi_ghes_init\n acpi_init\n ? acpi_sleep_proc_init\n do_one_initcall\n\nThe label contains garbage because the commit in Fixes reallocs the\nDIMMs array while scanning the system but doesn\u0027t clear the newly\nallocated memory.\n\nChange dimm_setup_label() to always initialize the label to fix the\nissue. Set it to the empty string in case BIOS does not provide both\nbank and device so that ghes_edac_register() can keep the default label\ngiven by edac_mc_alloc_dimms().\n\n [ bp: Rewrite commit message. ]\n\nFixes: b9cae27728d1f (\"EDAC/ghes: Scan the system once on driver init\")\nCo-developed-by: Robert Richter \u003crric@kernel.org\u003e\nSigned-off-by: Robert Richter \u003crric@kernel.org\u003e\nSigned-off-by: Toshi Kani \u003ctoshi.kani@hpe.com\u003e\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nTested-by: Robert Elliott \u003celliott@hpe.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nLink: https://lore.kernel.org/r/20220719220124.760359-1-toshi.kani@hpe.com\n" }, { "commit": "b5177ed92bf6f9d90a2493ed51c1327e088be1df", "tree": "b6e8b4eb8dd35c69e8564c9041c07a9a440df2e3", "parents": [ "e77ea97d2bd99b004e96c339ee22408c5475a52e" ], "author": { "name": "Mat Martineau", "email": "mathew.j.martineau@linux.intel.com", "time": "Mon Jul 25 13:52:31 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Jul 26 19:57:55 2022 -0700" }, "message": "mptcp: Do not return EINPROGRESS when subflow creation succeeds\n\nNew subflows are created within the kernel using O_NONBLOCK, so\nEINPROGRESS is the expected return value from kernel_connect().\n__mptcp_subflow_connect() has the correct logic to consider EINPROGRESS\nto be a successful case, but it has also used that error code as its\nreturn value.\n\nBefore v5.19 this was benign: all the callers ignored the return\nvalue. Starting in v5.19 there is a MPTCP_PM_CMD_SUBFLOW_CREATE generic\nnetlink command that does use the return value, so the EINPROGRESS gets\npropagated to userspace.\n\nMake __mptcp_subflow_connect() always return 0 on success instead.\n\nFixes: ec3edaa7ca6c (\"mptcp: Add handling of outgoing MP_JOIN requests\")\nFixes: 702c2f646d42 (\"mptcp: netlink: allow userspace-driven subflow establishment\")\nAcked-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nSigned-off-by: Mat Martineau \u003cmathew.j.martineau@linux.intel.com\u003e\nLink: https://lore.kernel.org/r/20220725205231.87529-1-mathew.j.martineau@linux.intel.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e77ea97d2bd99b004e96c339ee22408c5475a52e", "tree": "cc6975faf310ce8bdd1f38b17c8fe57b1625ad52", "parents": [ "e53f5293973181e8f557a7fef9a47f131fc3d4f0", "47f4f510ad586032b85c89a0773fbb011d412425" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Jul 26 19:53:09 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Jul 26 19:53:09 2022 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf\n\nFlorian Westphal says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnetfilter updates for net\n\nThree late fixes for netfilter:\n\n1) If nf_queue user requests packet truncation below size of l3 header,\n we corrupt the skb, then crash. Reject such requests.\n\n2) add cond_resched() calls when doing cycle detection in the\n nf_tables graph. This avoids softlockup warning with certain\n rulesets.\n\n3) Reject rulesets that use nftables \u0027queue\u0027 expression in family/chain\n combinations other than those that are supported. Currently the ruleset\n will load, but when userspace attempts to reinject you get WARN splat +\n packet drops.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:\n netfilter: nft_queue: only allow supported familes and hooks\n netfilter: nf_tables: add rescheduling points during loop detection walks\n netfilter: nf_queue: do not allow packet truncation below transport header offset\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20220726192056.13497-1-fw@strlen.de\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e53f5293973181e8f557a7fef9a47f131fc3d4f0", "tree": "eb18c458ebefa974fe5e2e7bae468036acd790a6", "parents": [ "aa40d5a43526cca9439a2b45fcfdcd016594dece", "d0be8347c623e0ac4202a1d4e0373882821f56b0" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Jul 26 19:48:24 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Jul 26 19:48:24 2022 -0700" }, "message": "Merge tag \u0027for-net-2022-07-26\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth\n\nLuiz Augusto von Dentz says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nbluetooth pull request for net:\n\n - Fix early wakeup after suspend\n - Fix double free on error\n - Fix use-after-free on l2cap_chan_put\n\n* tag \u0027for-net-2022-07-26\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth:\n Bluetooth: L2CAP: Fix use-after-free caused by l2cap_chan_put\n Bluetooth: Always set event mask on suspend\n Bluetooth: mgmt: Fix double free on error path\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/20220726221328.423714-1-luiz.dentz@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "39c3c396f8131f3db454c80e0fcfcdc54ed9ec01", "tree": "5e73a645381f47535903fbc6a16442a4a6ef2b85", "parents": [ "5de64d44968e4ae66ebdb0a2d08b443f189d3651", "1f7ea54727caaa6701a15af0cbeddfdb015b2869" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 19:38:46 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 19:38:46 2022 -0700" }, "message": "Merge tag \u0027mm-hotfixes-stable-2022-07-26\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm\n\nPull misc fixes from Andrew Morton:\n \"Thirteen hotfixes.\n\n Eight are cc:stable and the remainder are for post-5.18 issues or are\n too minor to warrant backporting\"\n\n* tag \u0027mm-hotfixes-stable-2022-07-26\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:\n mailmap: update Gao Xiang\u0027s email addresses\n userfaultfd: provide properly masked address for huge-pages\n Revert \"ocfs2: mount shared volume without ha stack\"\n hugetlb: fix memoryleak in hugetlb_mcopy_atomic_pte\n fs: sendfile handles O_NONBLOCK of out_fd\n ntfs: fix use-after-free in ntfs_ucsncmp()\n secretmem: fix unhandled fault in truncate\n mm/hugetlb: separate path for hwpoison entry in copy_hugetlb_page_range()\n mm: fix missing wake-up event for FSDAX pages\n mm: fix page leak with multiple threads mapping the same page\n mailmap: update Seth Forshee\u0027s email address\n tmpfs: fix the issue that the mount and remount results are inconsistent.\n mm: kfence: apply kmemleak_ignore_phys on early allocated pool\n" }, { "commit": "f5c2976e0cb0f6236013bfb479868531b04f61d4", "tree": "b3a7c35fabfa93adc7ead7522af7bcd6ab4c8bb2", "parents": [ "d9a434fa0c12ed5f7afe1e9dd30003ab5d059b85" ], "author": { "name": "Bart Van Assche", "email": "bvanassche@acm.org", "time": "Wed Jul 20 10:02:23 2022 -0700" }, "committer": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Tue Jul 26 21:59:29 2022 -0400" }, "message": "scsi: ufs: core: Fix a race condition related to device management\n\nIf a device management command completion happens after\nwait_for_completion_timeout() times out and before ufshcd_clear_cmds() is\ncalled, then the completion code may crash on the complete() call in\n__ufshcd_transfer_req_compl().\n\nFix the following crash:\n\n Unable to handle kernel NULL pointer dereference at virtual address 0000000000000008\n Call trace:\n complete+0x64/0x178\n __ufshcd_transfer_req_compl+0x30c/0x9c0\n ufshcd_poll+0xf0/0x208\n ufshcd_sl_intr+0xb8/0xf0\n ufshcd_intr+0x168/0x2f4\n __handle_irq_event_percpu+0xa0/0x30c\n handle_irq_event+0x84/0x178\n handle_fasteoi_irq+0x150/0x2e8\n __handle_domain_irq+0x114/0x1e4\n gic_handle_irq.31846+0x58/0x300\n el1_irq+0xe4/0x1c0\n efi_header_end+0x110/0x680\n __irq_exit_rcu+0x108/0x124\n __handle_domain_irq+0x118/0x1e4\n gic_handle_irq.31846+0x58/0x300\n el1_irq+0xe4/0x1c0\n cpuidle_enter_state+0x3ac/0x8c4\n do_idle+0x2fc/0x55c\n cpu_startup_entry+0x84/0x90\n kernel_init+0x0/0x310\n start_kernel+0x0/0x608\n start_kernel+0x4ec/0x608\n\nLink: https://lore.kernel.org/r/20220720170228.1598842-1-bvanassche@acm.org\nFixes: 5a0b0cb9bee7 (\"[SCSI] ufs: Add support for sending NOP OUT UPIU\")\nCc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nCc: Avri Altman \u003cavri.altman@wdc.com\u003e\nCc: Bean Huo \u003cbeanhuo@micron.com\u003e\nCc: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\nSigned-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n" }, { "commit": "d9a434fa0c12ed5f7afe1e9dd30003ab5d059b85", "tree": "5f4304ace4ca80ee85f8e15c1ce67d130c5ffaeb", "parents": [ "a3435afba87dc6cd83f5595e7607f3c40f93ef01" ], "author": { "name": "Jason Yan", "email": "yanaijie@huawei.com", "time": "Wed Jul 20 10:51:20 2022 +0800" }, "committer": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Tue Jul 26 21:54:30 2022 -0400" }, "message": "scsi: core: Fix warning in scsi_alloc_sgtables()\n\nAs explained in SG_IO howto[1]:\n\n\"If iovec_count is non-zero then \u0027dxfer_len\u0027 should be equal to the sum of\niov_len lengths. If not, the minimum of the two is the transfer length.\"\n\nWhen iovec_count is non-zero and dxfer_len is zero, the sg_io() just\ngenarated a null bio, and finally caused a warning below. To fix it, skip\ngenerating a bio for this request if dxfer_len is zero.\n\n[1] https://tldp.org/HOWTO/SCSI-Generic-HOWTO/x198.html\n\nWARNING: CPU: 2 PID: 3643 at drivers/scsi/scsi_lib.c:1032 scsi_alloc_sgtables+0xc7d/0xf70 drivers/scsi/scsi_lib.c:1032\nModules linked in:\n\nCPU: 2 PID: 3643 Comm: syz-executor397 Not tainted\n5.17.0-rc3-syzkaller-00316-gb81b1829e7e3 #0\nHardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-204/01/2014\nRIP: 0010:scsi_alloc_sgtables+0xc7d/0xf70 drivers/scsi/scsi_lib.c:1032\nCode: e7 fc 31 ff 44 89 f6 e8 c1 4e e7 fc 45 85 f6 0f 84 1a f5 ff ff e8\n93 4c e7 fc 83 c5 01 0f b7 ed e9 0f f5 ff ff e8 83 4c e7 fc \u003c0f\u003e 0b 41\n bc 0a 00 00 00 e9 2b fb ff ff 41 bc 09 00 00 00 e9 20 fb\nRSP: 0018:ffffc90000d07558 EFLAGS: 00010293\nRAX: 0000000000000000 RBX: ffff88801bfc96a0 RCX: 0000000000000000\nRDX: ffff88801c876000 RSI: ffffffff849060bd RDI: 0000000000000003\nRBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000\nR10: ffffffff849055b9 R11: 0000000000000000 R12: ffff888012b8c000\nR13: ffff88801bfc9580 R14: 0000000000000000 R15: ffff88801432c000\nFS: 00007effdec8e700(0000) GS:ffff88802cc00000(0000)\nknlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: 00007effdec6d718 CR3: 00000000206d6000 CR4: 0000000000150ee0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\nCall Trace:\n \u003cTASK\u003e\n scsi_setup_scsi_cmnd drivers/scsi/scsi_lib.c:1219 [inline]\n scsi_prepare_cmd drivers/scsi/scsi_lib.c:1614 [inline]\n scsi_queue_rq+0x283e/0x3630 drivers/scsi/scsi_lib.c:1730\n blk_mq_dispatch_rq_list+0x6ea/0x22e0 block/blk-mq.c:1851\n __blk_mq_sched_dispatch_requests+0x20b/0x410 block/blk-mq-sched.c:299\n blk_mq_sched_dispatch_requests+0xfb/0x180 block/blk-mq-sched.c:332\n __blk_mq_run_hw_queue+0xf9/0x350 block/blk-mq.c:1968\n __blk_mq_delay_run_hw_queue+0x5b6/0x6c0 block/blk-mq.c:2045\n blk_mq_run_hw_queue+0x30f/0x480 block/blk-mq.c:2096\n blk_mq_sched_insert_request+0x340/0x440 block/blk-mq-sched.c:451\n blk_execute_rq+0xcc/0x340 block/blk-mq.c:1231\n sg_io+0x67c/0x1210 drivers/scsi/scsi_ioctl.c:485\n scsi_ioctl_sg_io drivers/scsi/scsi_ioctl.c:866 [inline]\n scsi_ioctl+0xa66/0x1560 drivers/scsi/scsi_ioctl.c:921\n sd_ioctl+0x199/0x2a0 drivers/scsi/sd.c:1576\n blkdev_ioctl+0x37a/0x800 block/ioctl.c:588\n vfs_ioctl fs/ioctl.c:51 [inline]\n __do_sys_ioctl fs/ioctl.c:874 [inline]\n __se_sys_ioctl fs/ioctl.c:860 [inline]\n __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:860\n do_syscall_x64 arch/x86/entry/common.c:50 [inline]\n do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80\n entry_SYSCALL_64_after_hwframe+0x44/0xae\nRIP: 0033:0x7effdecdc5d9\nCode: 28 00 00 00 75 05 48 83 c4 28 c3 e8 81 14 00 00 90 48 89 f8 48 89\nf7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01\nf0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48\nRSP: 002b:00007effdec8e2f8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010\nRAX: ffffffffffffffda RBX: 00007effded664c0 RCX: 00007effdecdc5d9\nRDX: 0000000020002300 RSI: 0000000000002285 RDI: 0000000000000004\nRBP: 00007effded34034 R08: 0000000000000000 R09: 0000000000000000\nR10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000003\nR13: 00007effded34054 R14: 2f30656c69662f2e R15: 00007effded664c8\n\nLink: https://lore.kernel.org/r/20220720025120.3226770-1-yanaijie@huawei.com\nFixes: 25636e282fe9 (\"block: fix SG_IO vector request data length handling\")\nReported-by: syzbot+d44b35ecfb807e5af0b5@syzkaller.appspotmail.com\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nReviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\nSigned-off-by: Jason Yan \u003cyanaijie@huawei.com\u003e\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n" }, { "commit": "a3435afba87dc6cd83f5595e7607f3c40f93ef01", "tree": "56d70039cdf9d6acdf4c0fcdd1575824ec29809b", "parents": [ "0fde22c5420ed258ee538a760291c2f3935f6a01" ], "author": { "name": "Liang He", "email": "windhl@126.com", "time": "Tue Jul 19 15:15:29 2022 +0800" }, "committer": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Tue Jul 26 21:51:40 2022 -0400" }, "message": "scsi: ufs: host: Hold reference returned by of_parse_phandle()\n\nIn ufshcd_populate_vreg(), we should hold the reference returned by\nof_parse_phandle() and then use it to call of_node_put() for refcount\nbalance.\n\nLink: https://lore.kernel.org/r/20220719071529.1081166-1-windhl@126.com\nFixes: aa4976130934 (\"ufs: Add regulator enable support\")\nReviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\nSigned-off-by: Liang He \u003cwindhl@126.com\u003e\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n" }, { "commit": "0fde22c5420ed258ee538a760291c2f3935f6a01", "tree": "1935d290ed44b0740a68e333d608b102b8227029", "parents": [ "355bf2e036c954317ddc4a9618b4f7e38ea5a970" ], "author": { "name": "David Jeffery", "email": "djeffery@redhat.com", "time": "Fri Jul 22 10:24:48 2022 -0400" }, "committer": { "name": "Martin K. Petersen", "email": "martin.petersen@oracle.com", "time": "Tue Jul 26 21:40:43 2022 -0400" }, "message": "scsi: mpt3sas: Stop fw fault watchdog work item during system shutdown\n\nDuring system shutdown or reboot, mpt3sas will reset the firmware back to\nready state. However, the driver leaves running a watchdog work item\nintended to keep the firmware in operational state. This causes a second,\nunneeded reset on shutdown and moves the firmware back to operational\ninstead of in ready state as intended. And if the mpt3sas_fwfault_debug\nmodule parameter is set, this extra reset also panics the system.\n\nmpt3sas\u0027s scsih_shutdown needs to stop the watchdog before resetting the\nfirmware back to ready state.\n\nLink: https://lore.kernel.org/r/20220722142448.6289-1-djeffery@redhat.com\nFixes: fae21608c31c (\"scsi: mpt3sas: Transition IOC to Ready state during shutdown\")\nTested-by: Laurence Oberman \u003cloberman@redhat.com\u003e\nAcked-by: Sreekanth Reddy \u003csreekanth.reddy@broadcom.com\u003e\nSigned-off-by: David Jeffery \u003cdjeffery@redhat.com\u003e\nSigned-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n" }, { "commit": "1f7ea54727caaa6701a15af0cbeddfdb015b2869", "tree": "072f122819eeae44cc94c921b4c193c62ef28a4a", "parents": [ "d172b1a3bd065dd89234eac547fc62cf80681631" ], "author": { "name": "Gao Xiang", "email": "xiang@kernel.org", "time": "Tue Jul 19 23:42:46 2022 +0800" }, "committer": { "name": "akpm", "email": "akpm@linux-foundation.org", "time": "Tue Jul 26 18:25:01 2022 -0700" }, "message": "mailmap: update Gao Xiang\u0027s email addresses\n\nI\u0027ve been in Alibaba Cloud for more than one year, mainly to address\ncloud-native challenges (such as high-performance container images) for\nopen source communities.\n\nUpdate my email addresses on behalf of my current employer (Alibaba Cloud)\nto support all my (team) work in this area. Also add an outdated\n@redhat.com address of me.\n\nLink: https://lkml.kernel.org/r/20220719154246.62970-1-xiang@kernel.org\nSigned-off-by: Gao Xiang \u003cxiang@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "d172b1a3bd065dd89234eac547fc62cf80681631", "tree": "6e7a6d1a112a50d3607adcbd9787f2a08fe366ae", "parents": [ "c80af0c250c8f8a3c978aa5aafbe9c39b336b813" ], "author": { "name": "Nadav Amit", "email": "namit@vmware.com", "time": "Mon Jul 11 09:59:06 2022 -0700" }, "committer": { "name": "akpm", "email": "akpm@linux-foundation.org", "time": "Tue Jul 26 18:25:01 2022 -0700" }, "message": "userfaultfd: provide properly masked address for huge-pages\n\nCommit 824ddc601adc (\"userfaultfd: provide unmasked address on\npage-fault\") was introduced to fix an old bug, in which the offset in the\naddress of a page-fault was masked. Concerns were raised - although were\nnever backed by actual code - that some userspace code might break because\nthe bug has been around for quite a while. To address these concerns a\nnew flag was introduced, and only when this flag is set by the user,\nuserfaultfd provides the exact address of the page-fault.\n\nThe commit however had a bug, and if the flag is unset, the offset was\nalways masked based on a base-page granularity. Yet, for huge-pages, the\nbehavior prior to the commit was that the address is masked to the\nhuge-page granulrity.\n\nWhile there are no reports on real breakage, fix this issue. If the flag\nis unset, use the address with the masking that was done before.\n\nLink: https://lkml.kernel.org/r/20220711165906.2682-1-namit@vmware.com\nFixes: 824ddc601adc (\"userfaultfd: provide unmasked address on page-fault\")\nSigned-off-by: Nadav Amit \u003cnamit@vmware.com\u003e\nReported-by: James Houghton \u003cjthoughton@google.com\u003e\nReviewed-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nReviewed-by: Peter Xu \u003cpeterx@redhat.com\u003e\nReviewed-by: James Houghton \u003cjthoughton@google.com\u003e\nCc: David Hildenbrand \u003cdavid@redhat.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nCc: Andrea Arcangeli \u003caarcange@redhat.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "d0be8347c623e0ac4202a1d4e0373882821f56b0", "tree": "a9e3a6e8a6dc1cf2c345587e1413cd818b618f56", "parents": [ "ef61b6ea154464fefd8a6712d7a3b43b445c3d4a" ], "author": { "name": "Luiz Augusto von Dentz", "email": "luiz.von.dentz@intel.com", "time": "Thu Jul 21 09:10:50 2022 -0700" }, "committer": { "name": "Luiz Augusto von Dentz", "email": "luiz.von.dentz@intel.com", "time": "Tue Jul 26 13:35:24 2022 -0700" }, "message": "Bluetooth: L2CAP: Fix use-after-free caused by l2cap_chan_put\n\nThis fixes the following trace which is caused by hci_rx_work starting up\n*after* the final channel reference has been put() during sock_close() but\n*before* the references to the channel have been destroyed, so instead\nthe code now rely on kref_get_unless_zero/l2cap_chan_hold_unless_zero to\nprevent referencing a channel that is about to be destroyed.\n\n refcount_t: increment on 0; use-after-free.\n BUG: KASAN: use-after-free in refcount_dec_and_test+0x20/0xd0\n Read of size 4 at addr ffffffc114f5bf18 by task kworker/u17:14/705\n\n CPU: 4 PID: 705 Comm: kworker/u17:14 Tainted: G S W\n 4.14.234-00003-g1fb6d0bd49a4-dirty #28\n Hardware name: Qualcomm Technologies, Inc. SM8150 V2 PM8150\n Google Inc. MSM sm8150 Flame DVT (DT)\n Workqueue: hci0 hci_rx_work\n Call trace:\n dump_backtrace+0x0/0x378\n show_stack+0x20/0x2c\n dump_stack+0x124/0x148\n print_address_description+0x80/0x2e8\n __kasan_report+0x168/0x188\n kasan_report+0x10/0x18\n __asan_load4+0x84/0x8c\n refcount_dec_and_test+0x20/0xd0\n l2cap_chan_put+0x48/0x12c\n l2cap_recv_frame+0x4770/0x6550\n l2cap_recv_acldata+0x44c/0x7a4\n hci_acldata_packet+0x100/0x188\n hci_rx_work+0x178/0x23c\n process_one_work+0x35c/0x95c\n worker_thread+0x4cc/0x960\n kthread+0x1a8/0x1c4\n ret_from_fork+0x10/0x18\n\nCc: stable@kernel.org\nReported-by: Lee Jones \u003clee.jones@linaro.org\u003e\nSigned-off-by: Luiz Augusto von Dentz \u003cluiz.von.dentz@intel.com\u003e\nTested-by: Lee Jones \u003clee.jones@linaro.org\u003e\nSigned-off-by: Luiz Augusto von Dentz \u003cluiz.von.dentz@intel.com\u003e\n" }, { "commit": "ef61b6ea154464fefd8a6712d7a3b43b445c3d4a", "tree": "38ed3eb7fa840652dd1defceddd99e86e0ae4f69", "parents": [ "4b2f4e072fb2599b6a2e5e277f0d2b5705eaa630" ], "author": { "name": "Abhishek Pandit-Subedi", "email": "abhishekpandit@chromium.org", "time": "Mon Jul 25 15:34:21 2022 -0700" }, "committer": { "name": "Luiz Augusto von Dentz", "email": "luiz.von.dentz@intel.com", "time": "Tue Jul 26 13:35:13 2022 -0700" }, "message": "Bluetooth: Always set event mask on suspend\n\nWhen suspending, always set the event mask once disconnects are\nsuccessful. Otherwise, if wakeup is disallowed, the event mask is not\nset before suspend continues and can result in an early wakeup.\n\nFixes: 182ee45da083 (\"Bluetooth: hci_sync: Rework hci_suspend_notifier\")\nCc: stable@vger.kernel.org\nSigned-off-by: Abhishek Pandit-Subedi \u003cabhishekpandit@chromium.org\u003e\nSigned-off-by: Luiz Augusto von Dentz \u003cluiz.von.dentz@intel.com\u003e\n" }, { "commit": "4b2f4e072fb2599b6a2e5e277f0d2b5705eaa630", "tree": "68400f3da6163d14d7ca656cc832ab48a4757156", "parents": [ "9b134b1694ec8926926ba6b7b80884ea829245a0" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Tue Jul 26 18:13:44 2022 +0300" }, "committer": { "name": "Luiz Augusto von Dentz", "email": "luiz.von.dentz@intel.com", "time": "Tue Jul 26 13:32:40 2022 -0700" }, "message": "Bluetooth: mgmt: Fix double free on error path\n\nDon\u0027t call mgmt_pending_remove() twice (double free).\n\nFixes: 6b88eff43704 (\"Bluetooth: hci_sync: Refactor remove Adv Monitor\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Luiz Augusto von Dentz \u003cluiz.von.dentz@intel.com\u003e\n" }, { "commit": "aa40d5a43526cca9439a2b45fcfdcd016594dece", "tree": "57ec3310310ba0880c04e5b3fddade8da4a2a6a6", "parents": [ "9b134b1694ec8926926ba6b7b80884ea829245a0" ], "author": { "name": "Tetsuo Handa", "email": "penguin-kernel@I-love.SAKURA.ne.jp", "time": "Sun Jul 17 21:21:52 2022 +0900" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Tue Jul 26 13:23:05 2022 -0700" }, "message": "wifi: mac80211: do not abuse fq.lock in ieee80211_do_stop()\n\nlockdep complains use of uninitialized spinlock at ieee80211_do_stop() [1],\nfor commit f856373e2f31ffd3 (\"wifi: mac80211: do not wake queues on a vif\nthat is being stopped\") guards clear_bit() using fq.lock even before\nfq_init() from ieee80211_txq_setup_flows() initializes this spinlock.\n\nAccording to discussion [2], Toke was not happy with expanding usage of\nfq.lock. Since __ieee80211_wake_txqs() is called under RCU read lock, we\ncan instead use synchronize_rcu() for flushing ieee80211_wake_txqs().\n\nLink: https://syzkaller.appspot.com/bug?extid\u003deceab52db7c4b961e9d6 [1]\nLink: https://lkml.kernel.org/r/874k0zowh2.fsf@toke.dk [2]\nReported-by: syzbot \u003csyzbot+eceab52db7c4b961e9d6@syzkaller.appspotmail.com\u003e\nSigned-off-by: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\nFixes: f856373e2f31ffd3 (\"wifi: mac80211: do not wake queues on a vif that is being stopped\")\nTested-by: syzbot \u003csyzbot+eceab52db7c4b961e9d6@syzkaller.appspotmail.com\u003e\nAcked-by: Toke Høiland-Jørgensen \u003ctoke@kernel.org\u003e\nSigned-off-by: Kalle Valo \u003ckvalo@kernel.org\u003e\nLink: https://lore.kernel.org/r/9cc9b81d-75a3-3925-b612-9d0ad3cab82b@I-love.SAKURA.ne.jp\n[ pick up commit 3598cb6e1862 (\"wifi: mac80211: do not abuse fq.lock in ieee80211_do_stop()\") from -next]\nLink: https://lore.kernel.org/all/87o7xcq6qt.fsf@kernel.org/\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "cc019545a238518fa9da1e2a889f6e1bb1005a63", "tree": "258182232b3b673b8a3b60c0e18a3dc8ceecb708", "parents": [ "283d736ff7c7e96ac5b32c6c0de40372f8eb171e" ], "author": { "name": "Maciej Fijalkowski", "email": "maciej.fijalkowski@intel.com", "time": "Thu Jul 07 12:20:43 2022 +0200" }, "committer": { "name": "Tony Nguyen", "email": "anthony.l.nguyen@intel.com", "time": "Tue Jul 26 13:15:36 2022 -0700" }, "message": "ice: do not setup vlan for loopback VSI\n\nCurrently loopback test is failiing due to the error returned from\nice_vsi_vlan_setup(). Skip calling it when preparing loopback VSI.\n\nFixes: 0e674aeb0b77 (\"ice: Add handler for ethtool selftest\")\nSigned-off-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nTested-by: George Kuruvinakunnel \u003cgeorge.kuruvinakunnel@intel.com\u003e\nSigned-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n" }, { "commit": "283d736ff7c7e96ac5b32c6c0de40372f8eb171e", "tree": "08fbc83925390b851870a290f6b4e92d46873fa4", "parents": [ "5c8e3c7ff3e7bd7b938659be704f75cc746b697f" ], "author": { "name": "Maciej Fijalkowski", "email": "maciej.fijalkowski@intel.com", "time": "Thu Jul 07 12:20:42 2022 +0200" }, "committer": { "name": "Tony Nguyen", "email": "anthony.l.nguyen@intel.com", "time": "Tue Jul 26 13:15:36 2022 -0700" }, "message": "ice: check (DD | EOF) bits on Rx descriptor rather than (EOP | RS)\n\nTx side sets EOP and RS bits on descriptors to indicate that a\nparticular descriptor is the last one and needs to generate an irq when\nit was sent. These bits should not be checked on completion path\nregardless whether it\u0027s the Tx or the Rx. DD bit serves this purpose and\nit indicates that a particular descriptor is either for Rx or was\nsuccessfully Txed. EOF is also set as loopback test does not xmit\nfragmented frames.\n\nLook at (DD | EOF) bits setting in ice_lbtest_receive_frames() instead\nof EOP and RS pair.\n\nFixes: 0e674aeb0b77 (\"ice: Add handler for ethtool selftest\")\nSigned-off-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nTested-by: George Kuruvinakunnel \u003cgeorge.kuruvinakunnel@intel.com\u003e\nSigned-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n" }, { "commit": "5c8e3c7ff3e7bd7b938659be704f75cc746b697f", "tree": "7cf9c209715f6891fa3dac146795271d0199001e", "parents": [ "01658aeeada6f93c2924af94d895ff28d559690c" ], "author": { "name": "Anirudh Venkataramanan", "email": "anirudh.venkataramanan@intel.com", "time": "Thu Jul 21 10:03:09 2022 +0200" }, "committer": { "name": "Tony Nguyen", "email": "anthony.l.nguyen@intel.com", "time": "Tue Jul 26 13:15:36 2022 -0700" }, "message": "ice: Fix VSIs unable to share unicast MAC\n\nThe driver currently does not allow two VSIs in the same PF domain\nto have the same unicast MAC address. This is incorrect in the sense\nthat a policy decision is being made in the driver when it must be\nleft to the user. This approach was causing issues when rebooting\nthe system with VFs spawned not being able to change their MAC addresses.\nSuch errors were present in dmesg:\n\n[ 7921.068237] ice 0000:b6:00.2 ens2f2: Unicast MAC 6a:0d:e4:70:ca:d1 already\nexists on this PF. Preventing setting VF 7 unicast MAC address to 6a:0d:e4:70:ca:d1\n\nFix that by removing this restriction. Doing this also allows\nus to remove some additional code that\u0027s checking if a unicast MAC\nfilter already exists.\n\nFixes: 47ebc7b02485 (\"ice: Check if unicast MAC exists before setting VF MAC\")\nSigned-off-by: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\nSigned-off-by: Sylwester Dziedziuch \u003csylwesterx.dziedziuch@intel.com\u003e\nSigned-off-by: Mateusz Palczewski \u003cmateusz.palczewski@intel.com\u003e\nSigned-off-by: Jedrzej Jagielski \u003cjedrzej.jagielski@intel.com\u003e\nTested-by: Marek Szlosek \u003cmarek.szlosek@intel.com\u003e\nSigned-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n" }, { "commit": "01658aeeada6f93c2924af94d895ff28d559690c", "tree": "016f835f66e2617f989b0e7fdbbf0ec579e75bae", "parents": [ "1e308c6fb7127371f48a0fb9770ea0b30a6b5698" ], "author": { "name": "Przemyslaw Patynowski", "email": "przemyslawx.patynowski@intel.com", "time": "Mon Jul 18 13:34:27 2022 +0200" }, "committer": { "name": "Tony Nguyen", "email": "anthony.l.nguyen@intel.com", "time": "Tue Jul 26 13:15:36 2022 -0700" }, "message": "ice: Fix tunnel checksum offload with fragmented traffic\n\nFix checksum offload on VXLAN tunnels.\nIn case, when mpls protocol is not used, set l4 header to transport\nheader of skb. This fixes case, when user tries to offload checksums\nof VXLAN tunneled traffic.\n\nSteps for reproduction (requires link partner with tunnels):\nip l s enp130s0f0 up\nip a f enp130s0f0\nip a a 10.10.110.2/24 dev enp130s0f0\nip l s enp130s0f0 mtu 1600\nip link add vxlan12_sut type vxlan id 12 group 238.168.100.100 dev enp130s0f0 dstport 4789\nip l s vxlan12_sut up\nip a a 20.10.110.2/24 dev vxlan12_sut\niperf3 -c 20.10.110.1 #should connect\n\nOffload params: td_offset, cd_tunnel_params were\ncorrupted, due to l4 header pointing wrong address. NIC would then drop\nthose packets internally, due to incorrect TX descriptor data,\nwhich increased GLV_TEPC register.\n\nFixes: 69e66c04c672 (\"ice: Add mpls+tso support\")\nSigned-off-by: Przemyslaw Patynowski \u003cprzemyslawx.patynowski@intel.com\u003e\nSigned-off-by: Mateusz Palczewski \u003cmateusz.palczewski@intel.com\u003e\nSigned-off-by: Jedrzej Jagielski \u003cjedrzej.jagielski@intel.com\u003e\nTested-by: Gurucharan \u003cgurucharanx.g@intel.com\u003e (A Contingent worker at Intel)\nSigned-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n" }, { "commit": "1e308c6fb7127371f48a0fb9770ea0b30a6b5698", "tree": "b695b7a50dbe3bb69d801cea1ee8161bc712a166", "parents": [ "9b134b1694ec8926926ba6b7b80884ea829245a0" ], "author": { "name": "Przemyslaw Patynowski", "email": "przemyslawx.patynowski@intel.com", "time": "Mon Jul 04 15:46:12 2022 +0200" }, "committer": { "name": "Tony Nguyen", "email": "anthony.l.nguyen@intel.com", "time": "Tue Jul 26 13:15:18 2022 -0700" }, "message": "ice: Fix max VLANs available for VF\n\nLegacy VLAN implementation allows for untrusted VF to have 8 VLAN\nfilters, not counting VLAN 0 filters. Current VLAN_V2 implementation\nlowers available filters for VF, by counting in VLAN 0 filter for both\nTPIDs.\nFix this by counting only non zero VLAN filters.\nWithout this patch, untrusted VF would not be able to access 8 VLAN\nfilters.\n\nFixes: cc71de8fa133 (\"ice: Add support for VIRTCHNL_VF_OFFLOAD_VLAN_V2\")\nSigned-off-by: Przemyslaw Patynowski \u003cprzemyslawx.patynowski@intel.com\u003e\nSigned-off-by: Mateusz Palczewski \u003cmateusz.palczewski@intel.com\u003e\nTested-by: Marek Szlosek \u003cmarek.szlosek@intel.com\u003e\nSigned-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n" }, { "commit": "47f4f510ad586032b85c89a0773fbb011d412425", "tree": "7b5cf8909cb4ea535ab69853e72ce215947faed0", "parents": [ "81ea010667417ef3f218dfd99b69769fe66c2b67" ], "author": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Tue Jul 26 19:49:00 2022 +0200" }, "committer": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Tue Jul 26 21:12:42 2022 +0200" }, "message": "netfilter: nft_queue: only allow supported familes and hooks\n\nTrying to use \u0027queue\u0027 statement in ingress (for example)\ntriggers a splat on reinject:\n\nWARNING: CPU: 3 PID: 1345 at net/netfilter/nf_queue.c:291\n\n... because nf_reinject cannot find the ruleset head.\n\nThe netdev family doesn\u0027t support async resume at the moment anyway,\nso disallow loading such rulesets with a more appropriate\nerror message.\n\nv2: add \u0027validate\u0027 callback and also check hook points, v1 did\nallow ingress use in \u0027table inet\u0027, but that doesn\u0027t work either. (Pablo)\n\nSigned-off-by: Florian Westphal \u003cfw@strlen.de\u003e\nReviewed-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "81ea010667417ef3f218dfd99b69769fe66c2b67", "tree": "58b996afb7d57d0fe001a2e40518f6395ae8c0db", "parents": [ "99a63d36cb3ed5ca3aa6fcb64cffbeaf3b0fb164" ], "author": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Tue Jul 26 12:44:35 2022 +0200" }, "committer": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Tue Jul 26 21:12:42 2022 +0200" }, "message": "netfilter: nf_tables: add rescheduling points during loop detection walks\n\nAdd explicit rescheduling points during ruleset walk.\n\nSwitching to a faster algorithm is possible but this is a much\nsmaller change, suitable for nf tree.\n\nLink: https://bugzilla.netfilter.org/show_bug.cgi?id\u003d1460\nSigned-off-by: Florian Westphal \u003cfw@strlen.de\u003e\nAcked-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "99a63d36cb3ed5ca3aa6fcb64cffbeaf3b0fb164", "tree": "3c552bf8a966f236cf0bfb778f4b3a945150b2cc", "parents": [ "9b134b1694ec8926926ba6b7b80884ea829245a0" ], "author": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Tue Jul 26 12:42:06 2022 +0200" }, "committer": { "name": "Florian Westphal", "email": "fw@strlen.de", "time": "Tue Jul 26 21:12:42 2022 +0200" }, "message": "netfilter: nf_queue: do not allow packet truncation below transport header offset\n\nDomingo Dirutigliano and Nicola Guerrera report kernel panic when\nsending nf_queue verdict with 1-byte nfta_payload attribute.\n\nThe IP/IPv6 stack pulls the IP(v6) header from the packet after the\ninput hook.\n\nIf user truncates the packet below the header size, this skb_pull() will\nresult in a malformed skb (skb-\u003elen \u003c 0).\n\nFixes: 7af4cc3fa158 (\"[NETFILTER]: Add \"nfnetlink_queue\" netfilter queue handler over nfnetlink\")\nReported-by: Domingo Dirutigliano \u003cpwnzer0tt1@proton.me\u003e\nSigned-off-by: Florian Westphal \u003cfw@strlen.de\u003e\nReviewed-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "5de64d44968e4ae66ebdb0a2d08b443f189d3651", "tree": "cb77a416574ee71ec0d09c01e7d6239408d4ba34", "parents": [ "cdb281e63874086a650552d36c504ea717a0e0cb", "918e75f77af7d2e049bb70469ec0a2c12782d96a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 10:03:53 2022 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 10:03:53 2022 -0700" }, "message": "Merge tag \u0027s390-5.19-7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 fix from Alexander GordeevL\n\n - Prevent relatively slow PRNO TRNG random number operation from being\n called from interrupt context. That could for example cause some\n network loads to timeout.\n\n* tag \u0027s390-5.19-7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n s390/archrandom: prevent CPACF trng invocations in interrupt context\n" }, { "commit": "cdb281e63874086a650552d36c504ea717a0e0cb", "tree": "22d312d4d3875c486bdd61311d36a116fada06be", "parents": [ "e0dccc3b76fb35bb257b4118367a883073d7390e" ], "author": { "name": "Qi Zheng", "email": "zhengqi.arch@bytedance.com", "time": "Tue Jul 26 14:24:36 2022 +0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 26 09:21:43 2022 -0700" }, "message": "mm: fix NULL pointer dereference in wp_page_reuse()\n\nThe vmf-\u003epage can be NULL when the wp_page_reuse() is invoked by\nwp_pfn_shared(), it will cause the following panic:\n\n BUG: kernel NULL pointer dereference, address: 000000000000008\n #PF: supervisor read access in kernel mode\n #PF: error_code(0x0000) - not-present page\n PGD 0 P4D 0\n Oops: 0000 [#1] PREEMPT SMP PTI\n CPU: 18 PID: 923 Comm: Xorg Not tainted 5.19.0-rc8.bm.1-amd64 #263\n Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g14\n RIP: 0010:_compound_head+0x0/0x40\n [...]\n Call Trace:\n wp_page_reuse+0x1c/0xa0\n do_wp_page+0x1a5/0x3f0\n __handle_mm_fault+0x8cf/0xd20\n handle_mm_fault+0xd5/0x2a0\n do_user_addr_fault+0x1d0/0x680\n exc_page_fault+0x78/0x170\n asm_exc_page_fault+0x22/0x30\n\nTo fix it, this patch performs a NULL pointer check before dereferencing\nthe vmf-\u003epage.\n\nFixes: 6c287605fd56 (\"mm: remember exclusively mapped anonymous pages with PG_anon_exclusive\")\nSigned-off-by: Qi Zheng \u003czhengqi.arch@bytedance.com\u003e\nReviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "0c09bc33aa8e9dc867300acaadc318c2f0d85a1e", "tree": "7b125e986be4571dfd30419d14e5fd9de94b7f4e", "parents": [ "02c87df2480ac855d88ee308ce3fa857d9bd55a8" ], "author": { "name": "Nathan Chancellor", "email": "nathan@kernel.org", "time": "Mon Jul 25 16:36:29 2022 -0700" }, "committer": { "name": "Thomas Zimmermann", "email": "tzimmermann@suse.de", "time": "Tue Jul 26 17:18:45 2022 +0200" }, "message": "drm/simpledrm: Fix return type of simpledrm_simple_display_pipe_mode_valid()\n\nWhen booting a kernel compiled with clang\u0027s CFI protection\n(CONFIG_CFI_CLANG), there is a CFI failure in\ndrm_simple_kms_crtc_mode_valid() when trying to call\nsimpledrm_simple_display_pipe_mode_valid() through -\u003emode_valid():\n\n[ 0.322802] CFI failure (target: simpledrm_simple_display_pipe_mode_valid+0x0/0x8):\n...\n[ 0.324928] Call trace:\n[ 0.324969] __ubsan_handle_cfi_check_fail+0x58/0x60\n[ 0.325053] __cfi_check_fail+0x3c/0x44\n[ 0.325120] __cfi_slowpath_diag+0x178/0x200\n[ 0.325192] drm_simple_kms_crtc_mode_valid+0x58/0x80\n[ 0.325279] __drm_helper_update_and_validate+0x31c/0x464\n...\n\nThe -\u003emode_valid() member in \u0027struct drm_simple_display_pipe_funcs\u0027\nexpects a return type of \u0027enum drm_mode_status\u0027, not \u0027int\u0027. Correct it\nto fix the CFI failure.\n\nCc: stable@vger.kernel.org\nFixes: 11e8f5fd223b (\"drm: Add simpledrm driver\")\nLink: https://github.com/ClangBuiltLinux/linux/issues/1647\nReported-by: Tomasz Paweł Gajc \u003ctpgxyz@gmail.com\u003e\nSigned-off-by: Nathan Chancellor \u003cnathan@kernel.org\u003e\nSigned-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nReviewed-by: Sami Tolvanen \u003csamitolvanen@google.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20220725233629.223223-1-nathan@kernel.org\n" }, { "commit": "9b134b1694ec8926926ba6b7b80884ea829245a0", "tree": "c4fda61cabd1d8d778a8740342571c0959b4b416", "parents": [ "33881ab73d6f6b4f0a2a89aece86dd60e6e507c6" ], "author": { "name": "Benjamin Poirier", "email": "bpoirier@nvidia.com", "time": "Mon Jul 25 09:12:36 2022 +0900" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Jul 26 15:35:53 2022 +0200" }, "message": "bridge: Do not send empty IFLA_AF_SPEC attribute\n\nAfter commit b6c02ef54913 (\"bridge: Netlink interface fix.\"),\nbr_fill_ifinfo() started to send an empty IFLA_AF_SPEC attribute when a\nbridge vlan dump is requested but an interface does not have any vlans\nconfigured.\n\niproute2 ignores such an empty attribute since commit b262a9becbcb\n(\"bridge: Fix output with empty vlan lists\") but older iproute2 versions as\nwell as other utilities have their output changed by the cited kernel\ncommit, resulting in failed test cases. Regardless, emitting an empty\nattribute is pointless and inefficient.\n\nAvoid this change by canceling the attribute if no AF_SPEC data was added.\n\nFixes: b6c02ef54913 (\"bridge: Netlink interface fix.\")\nReviewed-by: Ido Schimmel \u003cidosch@nvidia.com\u003e\nSigned-off-by: Benjamin Poirier \u003cbpoirier@nvidia.com\u003e\nAcked-by: Nikolay Aleksandrov \u003crazor@blackwall.org\u003e\nLink: https://lore.kernel.org/r/20220725001236.95062-1-bpoirier@nvidia.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "33881ab73d6f6b4f0a2a89aece86dd60e6e507c6", "tree": "24991ed1a66df00db2a1e277aa3ac0f2ee8ad504", "parents": [ "b89fc26f741d9f9efb51cba3e9b241cf1380ec5a", "59e1be6f83b928a04189bbf3ab683a1fc6248db3" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Jul 26 13:05:46 2022 +0200" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Jul 26 13:05:46 2022 +0200" }, "message": "Merge branch \u0027octeontx2-minor-tc-fixes\u0027\n\nSubbaraya Sundeep says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nOcteontx2 minor tc fixes\n\nThis patch set fixes two problems found in tc code\nwrt to ratelimiting and when installing UDP/TCP filters.\n\nPatch 1: CN10K has different register format compared to\nCN9xx hence fixes that.\nPatch 2: Check flow mask also before installing a src/dst\nport filter, otherwise installing for one port installs for other one too.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://lore.kernel.org/r/1658650874-16459-1-git-send-email-sbhatta@marvell.com\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "59e1be6f83b928a04189bbf3ab683a1fc6248db3", "tree": "24991ed1a66df00db2a1e277aa3ac0f2ee8ad504", "parents": [ "b354eaeec8637d87003945439209251d76a2bb95" ], "author": { "name": "Subbaraya Sundeep", "email": "sbhatta@marvell.com", "time": "Sun Jul 24 13:51:14 2022 +0530" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Jul 26 13:05:44 2022 +0200" }, "message": "octeontx2-pf: Fix UDP/TCP src and dst port tc filters\n\nCheck the mask for non-zero value before installing tc filters\nfor L4 source and destination ports. Otherwise installing a\nfilter for source port installs destination port too and\nvice-versa.\n\nFixes: 1d4d9e42c240 (\"octeontx2-pf: Add tc flower hardware offload on ingress traffic\")\nSigned-off-by: Subbaraya Sundeep \u003csbhatta@marvell.com\u003e\nSigned-off-by: Sunil Goutham \u003csgoutham@marvell.com\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "b354eaeec8637d87003945439209251d76a2bb95", "tree": "559375d70ac79f744ad0b996dc68b4b82c4414f4", "parents": [ "b89fc26f741d9f9efb51cba3e9b241cf1380ec5a" ], "author": { "name": "Sunil Goutham", "email": "sgoutham@marvell.com", "time": "Sun Jul 24 13:51:13 2022 +0530" }, "committer": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Tue Jul 26 13:05:44 2022 +0200" }, "message": "octeontx2-pf: cn10k: Fix egress ratelimit configuration\n\nNIX_AF_TLXX_PIR/CIR register format has changed from OcteonTx2\nto CN10K. CN10K supports larger burst size. Fix burst exponent\nand burst mantissa configuration for CN10K.\n\nAlso fixed \u0027maxrate\u0027 from u32 to u64 since \u0027police.rate_bytes_ps\u0027\npassed by stack is also u64.\n\nFixes: e638a83f167e (\"octeontx2-pf: TC_MATCHALL egress ratelimiting offload\")\nSigned-off-by: Sunil Goutham \u003csgoutham@marvell.com\u003e\nSigned-off-by: Subbaraya Sundeep \u003csbhatta@marvell.com\u003e\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\n" }, { "commit": "b89fc26f741d9f9efb51cba3e9b241cf1380ec5a", "tree": "47550b33acac21610b5832143f9e4b6488080f09", "parents": [ "c7560d1203b7a1ea0b99a5c575547e95d564b2a8" ], "author": { "name": "Duoming Zhou", "email": "duoming@zju.edu.cn", "time": "Sat Jul 23 09:58:09 2022 +0800" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Jul 25 19:39:05 2022 -0700" }, "message": "sctp: fix sleep in atomic context bug in timer handlers\n\nThere are sleep in atomic context bugs in timer handlers of sctp\nsuch as sctp_generate_t3_rtx_event(), sctp_generate_probe_event(),\nsctp_generate_t1_init_event(), sctp_generate_timeout_event(),\nsctp_generate_t3_rtx_event() and so on.\n\nThe root cause is sctp_sched_prio_init_sid() with GFP_KERNEL parameter\nthat may sleep could be called by different timer handlers which is in\ninterrupt context.\n\nOne of the call paths that could trigger bug is shown below:\n\n (interrupt context)\nsctp_generate_probe_event\n sctp_do_sm\n sctp_side_effects\n sctp_cmd_interpreter\n sctp_outq_teardown\n sctp_outq_init\n sctp_sched_set_sched\n n-\u003einit_sid(..,GFP_KERNEL)\n sctp_sched_prio_init_sid //may sleep\n\nThis patch changes gfp_t parameter of init_sid in sctp_sched_set_sched()\nfrom GFP_KERNEL to GFP_ATOMIC in order to prevent sleep in atomic\ncontext bugs.\n\nFixes: 5bbbbe32a431 (\"sctp: introduce stream scheduler foundations\")\nSigned-off-by: Duoming Zhou \u003cduoming@zju.edu.cn\u003e\nAcked-by: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\nLink: https://lore.kernel.org/r/20220723015809.11553-1-duoming@zju.edu.cn\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "c7560d1203b7a1ea0b99a5c575547e95d564b2a8", "tree": "9ae41522750add3ed40077f03e70f59393c95d94", "parents": [ "5fcbb711024aac6d4db385623e6f2fdf019f7782" ], "author": { "name": "Vladimir Oltean", "email": "vladimir.oltean@nxp.com", "time": "Sat Jul 23 04:24:11 2022 +0300" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Jul 25 19:37:06 2022 -0700" }, "message": "net: dsa: fix reference counting for LAG FDBs\n\nDue to an invalid conflict resolution on my side while working on 2\ndifferent series (LAG FDBs and FDB isolation), dsa_switch_do_lag_fdb_add()\ndoes not store the database associated with a dsa_mac_addr structure.\n\nSo after adding an FDB entry associated with a LAG, dsa_mac_addr_find()\nfails to find it while deleting it, because \u0026a-\u003edb is zeroized memory\nfor all stored FDB entries of lag-\u003efdbs, and dsa_switch_do_lag_fdb_del()\nreturns -ENOENT rather than deleting the entry.\n\nFixes: c26933639b54 (\"net: dsa: request drivers to perform FDB isolation\")\nSigned-off-by: Vladimir Oltean \u003cvladimir.oltean@nxp.com\u003e\nLink: https://lore.kernel.org/r/20220723012411.1125066-1-vladimir.oltean@nxp.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "5fcbb711024aac6d4db385623e6f2fdf019f7782", "tree": "8fca83021cffdeb1369662c9d214e9187b3ad070", "parents": [ "9af0620de1e118666881376f6497d1785758b04c" ], "author": { "name": "Michal Maloszewski", "email": "michal.maloszewski@intel.com", "time": "Fri Jul 22 10:54:01 2022 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Jul 25 18:48:41 2022 -0700" }, "message": "i40e: Fix interface init with MSI interrupts (no MSI-X)\n\nFix the inability to bring an interface up on a setup with\nonly MSI interrupts enabled (no MSI-X).\nSolution is to add a default number of QPs \u003d 1. This is enough,\nsince without MSI-X support driver enables only a basic feature set.\n\nFixes: bc6d33c8d93f (\"i40e: Fix the number of queues available to be mapped for use\")\nSigned-off-by: Dawid Lukwinski \u003cdawid.lukwinski@intel.com\u003e\nSigned-off-by: Michal Maloszewski \u003cmichal.maloszewski@intel.com\u003e\nTested-by: Dave Switzer \u003cdavid.switzer@intel.com\u003e\nSigned-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\nLink: https://lore.kernel.org/r/20220722175401.112572-1-anthony.l.nguyen@intel.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" } ], "next": "c653c591789b3acfa4bf6ae45d5af4f330e50a91" }