)]}' { "log": [ { "commit": "98d54f81e36ba3bf92172791eba5ca5bd813989b", "tree": "997265da22447fdbb0e890b532c6aca9938d42f5", "parents": [ "e70869821a469029aeb056b4da96e5815cc5830e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 01 16:38:46 2020 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 01 16:38:46 2020 -0600" }, "message": "Linux 5.6-rc4\n" }, { "commit": "e70869821a469029aeb056b4da96e5815cc5830e", "tree": "874df815e0f61bf91d2d8a931201f8642d9df39d", "parents": [ "f853ed90e2e49f4f6e585f995ffb25e093fe4df6", "37b0b6b8b99c0e1c1f11abbe7cf49b6d03795b3f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 01 16:35:08 2020 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 01 16:35:08 2020 -0600" }, "message": "Merge tag \u0027ext4_for_linus_stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4\n\nPull ext4 fixes from Ted Ts\u0027o:\n \"Two more bug fixes (including a regression) for 5.6\"\n\n* tag \u0027ext4_for_linus_stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:\n ext4: potential crash on allocation error in ext4_alloc_flex_bg_array()\n jbd2: fix data races at struct journal_head\n" }, { "commit": "f853ed90e2e49f4f6e585f995ffb25e093fe4df6", "tree": "a30f39aacf3d517f3787762bfb7d3917c828f634", "parents": [ "fb279f4e238617417b132a550f24c1e86d922558", "86f7e90ce840aa1db407d3ea6e9b3a52b2ce923c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 01 15:16:35 2020 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 01 15:16:35 2020 -0600" }, "message": "Merge tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm\n\nPull KVM fixes from Paolo Bonzini:\n \"More bugfixes, including a few remaining \"make W\u003d1\" issues such as too\n large frame sizes on some configurations.\n\n On the ARM side, the compiler was messing up shadow stacks between EL1\n and EL2 code, which is easily fixed with __always_inline\"\n\n* tag \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/virt/kvm/kvm:\n KVM: VMX: check descriptor table exits on instruction emulation\n kvm: x86: Limit the number of \"kvm: disabled by bios\" messages\n KVM: x86: avoid useless copy of cpufreq policy\n KVM: allow disabling -Werror\n KVM: x86: allow compiling as non-module with W\u003d1\n KVM: Pre-allocate 1 cpumask variable per cpu for both pv tlb and pv ipis\n KVM: Introduce pv check helpers\n KVM: let declaration of kvm_get_running_vcpus match implementation\n KVM: SVM: allocate AVIC data structures based on kvm_amd module parameter\n arm64: Ask the compiler to __always_inline functions used by KVM at HYP\n KVM: arm64: Define our own swab32() to avoid a uapi static inline\n KVM: arm64: Ask the compiler to __always_inline functions used at HYP\n kvm: arm/arm64: Fold VHE entry/exit work into kvm_vcpu_run_vhe()\n KVM: arm/arm64: Fix up includes for trace.h\n" }, { "commit": "86f7e90ce840aa1db407d3ea6e9b3a52b2ce923c", "tree": "7679ea76c6ff795ed0d53bc755aa43d77b4ce7a5", "parents": [ "e951445f4d3b5d0df69c0c5d18ab1e9058c20e52" ], "author": { "name": "Oliver Upton", "email": "oupton@google.com", "time": "Sat Feb 29 11:30:14 2020 -0800" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Sun Mar 01 19:26:31 2020 +0100" }, "message": "KVM: VMX: check descriptor table exits on instruction emulation\n\nKVM emulates UMIP on hardware that doesn\u0027t support it by setting the\n\u0027descriptor table exiting\u0027 VM-execution control and performing\ninstruction emulation. When running nested, this emulation is broken as\nKVM refuses to emulate L2 instructions by default.\n\nCorrect this regression by allowing the emulation of descriptor table\ninstructions if L1 hasn\u0027t requested \u0027descriptor table exiting\u0027.\n\nFixes: 07721feee46b (\"KVM: nVMX: Don\u0027t emulate instructions in guest mode\")\nReported-by: Jan Kiszka \u003cjan.kiszka@web.de\u003e\nCc: stable@vger.kernel.org\nCc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nCc: Jim Mattson \u003cjmattson@google.com\u003e\nSigned-off-by: Oliver Upton \u003coupton@google.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "fb279f4e238617417b132a550f24c1e86d922558", "tree": "b5fee8cdf5794e924a98f90cfcf8635476c27bc5", "parents": [ "7557c1b3f715624f3bd4c809f03771ce9384eb7b", "38b17afb0ebb9ecd41418d3c08bcf9198af4349d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 29 19:16:46 2020 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 29 19:16:46 2020 -0600" }, "message": "Merge branch \u0027i2c/for-current-fixed\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux\n\nPull i2c fixes from Wolfram Sang:\n \"I2C has three driver bugfixes for you. We agreed on the Mac regression\n to go in via I2C\"\n\n* \u0027i2c/for-current-fixed\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:\n macintosh: therm_windtunnel: fix regression when instantiating devices\n i2c: altera: Fix potential integer overflow\n i2c: jz4780: silence log flood on txabrt\n" }, { "commit": "37b0b6b8b99c0e1c1f11abbe7cf49b6d03795b3f", "tree": "e5b072f6e69c2b726c70c865015763d395680d6a", "parents": [ "6c5d911249290f41f7b50b43344a7520605b1acb" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@oracle.com", "time": "Fri Feb 28 12:22:56 2020 +0300" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sat Feb 29 17:48:08 2020 -0500" }, "message": "ext4: potential crash on allocation error in ext4_alloc_flex_bg_array()\n\nIf sbi-\u003es_flex_groups_allocated is zero and the first allocation fails\nthen this code will crash. The problem is that \"i--\" will set \"i\" to\n-1 but when we compare \"i \u003e\u003d sbi-\u003es_flex_groups_allocated\" then the -1\nis type promoted to unsigned and becomes UINT_MAX. Since UINT_MAX\nis more than zero, the condition is true so we call kvfree(new_groups[-1]).\nThe loop will carry on freeing invalid memory until it crashes.\n\nFixes: 7c990728b99e (\"ext4: fix potential race between s_flex_groups online resizing and access\")\nReviewed-by: Suraj Jitindar Singh \u003csurajjs@amazon.com\u003e\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nCc: stable@kernel.org\nLink: https://lore.kernel.org/r/20200228092142.7irbc44yaz3by7nb@kili.mountain\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "38b17afb0ebb9ecd41418d3c08bcf9198af4349d", "tree": "4007a8cc82228efe42c4c06386a22be45e0338a0", "parents": [ "54498e8070e19e74498a72c7331348143e7e1f8c" ], "author": { "name": "Wolfram Sang", "email": "wsa@the-dreams.de", "time": "Tue Feb 25 15:12:29 2020 +0100" }, "committer": { "name": "Wolfram Sang", "email": "wsa@the-dreams.de", "time": "Sat Feb 29 21:13:22 2020 +0100" }, "message": "macintosh: therm_windtunnel: fix regression when instantiating devices\n\nRemoving attach_adapter from this driver caused a regression for at\nleast some machines. Those machines had the sensors described in their\nDT, too, so they didn\u0027t need manual creation of the sensor devices. The\nold code worked, though, because manual creation came first. Creation of\nDT devices then failed later and caused error logs, but the sensors\nworked nonetheless because of the manually created devices.\n\nWhen removing attach_adaper, manual creation now comes later and loses\nthe race. The sensor devices were already registered via DT, yet with\nanother binding, so the driver could not be bound to it.\n\nThis fix refactors the code to remove the race and only manually creates\ndevices if there are no DT nodes present. Also, the DT binding is updated\nto match both, the DT and manually created devices. Because we don\u0027t\nknow which device creation will be used at runtime, the code to start\nthe kthread is moved to do_probe() which will be called by both methods.\n\nFixes: 3e7bed52719d (\"macintosh: therm_windtunnel: drop using attach_adapter\")\nLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d201723\nReported-by: Erhard Furtner \u003cerhard_f@mailbox.org\u003e\nTested-by: Erhard Furtner \u003cerhard_f@mailbox.org\u003e\nAcked-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e (powerpc)\nSigned-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\nCc: stable@kernel.org # v4.19+\n" }, { "commit": "6c5d911249290f41f7b50b43344a7520605b1acb", "tree": "13428fb7182fb2df1c5e1d5df135fed3b661017c", "parents": [ "f8788d86ab28f61f7b46eb6be375f8a726783636" ], "author": { "name": "Qian Cai", "email": "cai@lca.pw", "time": "Fri Feb 21 23:31:11 2020 -0500" }, "committer": { "name": "Theodore Ts\u0027o", "email": "tytso@mit.edu", "time": "Sat Feb 29 13:40:02 2020 -0500" }, "message": "jbd2: fix data races at struct journal_head\n\njournal_head::b_transaction and journal_head::b_next_transaction could\nbe accessed concurrently as noticed by KCSAN,\n\n LTP: starting fsync04\n /dev/zero: Can\u0027t open blockdev\n EXT4-fs (loop0): mounting ext3 file system using the ext4 subsystem\n EXT4-fs (loop0): mounted filesystem with ordered data mode. Opts: (null)\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\n BUG: KCSAN: data-race in __jbd2_journal_refile_buffer [jbd2] / jbd2_write_access_granted [jbd2]\n\n write to 0xffff99f9b1bd0e30 of 8 bytes by task 25721 on cpu 70:\n __jbd2_journal_refile_buffer+0xdd/0x210 [jbd2]\n __jbd2_journal_refile_buffer at fs/jbd2/transaction.c:2569\n jbd2_journal_commit_transaction+0x2d15/0x3f20 [jbd2]\n (inlined by) jbd2_journal_commit_transaction at fs/jbd2/commit.c:1034\n kjournald2+0x13b/0x450 [jbd2]\n kthread+0x1cd/0x1f0\n ret_from_fork+0x27/0x50\n\n read to 0xffff99f9b1bd0e30 of 8 bytes by task 25724 on cpu 68:\n jbd2_write_access_granted+0x1b2/0x250 [jbd2]\n jbd2_write_access_granted at fs/jbd2/transaction.c:1155\n jbd2_journal_get_write_access+0x2c/0x60 [jbd2]\n __ext4_journal_get_write_access+0x50/0x90 [ext4]\n ext4_mb_mark_diskspace_used+0x158/0x620 [ext4]\n ext4_mb_new_blocks+0x54f/0xca0 [ext4]\n ext4_ind_map_blocks+0xc79/0x1b40 [ext4]\n ext4_map_blocks+0x3b4/0x950 [ext4]\n _ext4_get_block+0xfc/0x270 [ext4]\n ext4_get_block+0x3b/0x50 [ext4]\n __block_write_begin_int+0x22e/0xae0\n __block_write_begin+0x39/0x50\n ext4_write_begin+0x388/0xb50 [ext4]\n generic_perform_write+0x15d/0x290\n ext4_buffered_write_iter+0x11f/0x210 [ext4]\n ext4_file_write_iter+0xce/0x9e0 [ext4]\n new_sync_write+0x29c/0x3b0\n __vfs_write+0x92/0xa0\n vfs_write+0x103/0x260\n ksys_write+0x9d/0x130\n __x64_sys_write+0x4c/0x60\n do_syscall_64+0x91/0xb05\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\n 5 locks held by fsync04/25724:\n #0: ffff99f9911093f8 (sb_writers#13){.+.+}, at: vfs_write+0x21c/0x260\n #1: ffff99f9db4c0348 (\u0026sb-\u003es_type-\u003ei_mutex_key#15){+.+.}, at: ext4_buffered_write_iter+0x65/0x210 [ext4]\n #2: ffff99f5e7dfcf58 (jbd2_handle){++++}, at: start_this_handle+0x1c1/0x9d0 [jbd2]\n #3: ffff99f9db4c0168 (\u0026ei-\u003ei_data_sem){++++}, at: ext4_map_blocks+0x176/0x950 [ext4]\n #4: ffffffff99086b40 (rcu_read_lock){....}, at: jbd2_write_access_granted+0x4e/0x250 [jbd2]\n irq event stamp: 1407125\n hardirqs last enabled at (1407125): [\u003cffffffff980da9b7\u003e] __find_get_block+0x107/0x790\n hardirqs last disabled at (1407124): [\u003cffffffff980da8f9\u003e] __find_get_block+0x49/0x790\n softirqs last enabled at (1405528): [\u003cffffffff98a0034c\u003e] __do_softirq+0x34c/0x57c\n softirqs last disabled at (1405521): [\u003cffffffff97cc67a2\u003e] irq_exit+0xa2/0xc0\n\n Reported by Kernel Concurrency Sanitizer on:\n CPU: 68 PID: 25724 Comm: fsync04 Tainted: G L 5.6.0-rc2-next-20200221+ #7\n Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019\n\nThe plain reads are outside of jh-\u003eb_state_lock critical section which result\nin data races. Fix them by adding pairs of READ|WRITE_ONCE().\n\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Qian Cai \u003ccai@lca.pw\u003e\nLink: https://lore.kernel.org/r/20200222043111.2227-1-cai@lca.pw\nSigned-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n" }, { "commit": "7557c1b3f715624f3bd4c809f03771ce9384eb7b", "tree": "4ea11f34ea4c4c35099fa420205dd18982c60666", "parents": [ "29795de0d242a5ba45904b36a5fb67e38a304cb7", "03264ddde2453f6877a7d637d84068079632a3c5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 29 09:58:47 2020 -0600" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Feb 29 09:58:47 2020 -0600" }, "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 small fixes.\n\n Three are in drivers for fairly obvious bugs. The fourth is a set of\n regressions introduced by the compat_ioctl changes because some of the\n compat updates wrongly replaced .ioctl instead of .compat_ioctl\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n scsi: compat_ioctl: cdrom: Replace .ioctl with .compat_ioctl in four appropriate places\n scsi: zfcp: fix wrong data and display format of SFP+ temperature\n scsi: sd_sbc: Fix sd_zbc_report_zones()\n scsi: libfc: free response frame from GPN_ID\n" }, { "commit": "29795de0d242a5ba45904b36a5fb67e38a304cb7", "tree": "c24f966799e251c307d73e7da4a0876f4a8c5b36", "parents": [ "2edc78b9a4b868d7bfee4f87ea29f2df19b6e955", "5901b51f3e5d9129da3e59b10cc76e4cc983e940" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 11:51:53 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 11:51:53 2020 -0800" }, "message": "Merge tag \u0027pci-v5.6-fixes-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci\n\nPull PCI fixes from Bjorn Helgaas:\n\n - Fix build issue on 32-bit ARM with old compilers (Marek Szyprowski)\n\n - Update MAINTAINERS for recent Cadence driver file move (Lukas\n Bulwahn)\n\n* tag \u0027pci-v5.6-fixes-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:\n MAINTAINERS: Correct Cadence PCI driver path\n PCI: brcmstb: Fix build on 32bit ARM platforms with older compilers\n" }, { "commit": "2edc78b9a4b868d7bfee4f87ea29f2df19b6e955", "tree": "e719289213d2691ad2a89a9576f72e9b670cbf74", "parents": [ "74dea5d99d19236608914d2c556134e4cdc21c60", "5b8ea58b6a338cb981670c4408225331aedb4b89" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 11:43:30 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 11:43:30 2020 -0800" }, "message": "Merge tag \u0027block-5.6-2020-02-28\u0027 of git://git.kernel.dk/linux-block\n\nPull block fixes from Jens Axboe:\n\n - Passthrough insertion fix (Ming)\n\n - Kill off some unused arguments (John)\n\n - blktrace RCU fix (Jan)\n\n - Dead fields removal for null_blk (Dongli)\n\n - NVMe polled IO fix (Bijan)\n\n* tag \u0027block-5.6-2020-02-28\u0027 of git://git.kernel.dk/linux-block:\n nvme-pci: Hold cq_poll_lock while completing CQEs\n blk-mq: Remove some unused function arguments\n null_blk: remove unused fields in \u0027nullb_cmd\u0027\n blktrace: Protect q-\u003eblk_trace with RCU\n blk-mq: insert passthrough request into hctx-\u003edispatch directly\n" }, { "commit": "74dea5d99d19236608914d2c556134e4cdc21c60", "tree": "4406cd9b8fcb78103587aa4d06a67480b6812e4c", "parents": [ "c60c04021353c55b133519804734415f647f08bd", "d876836204897b6d7d911f942084f69a1e9d5c4d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 11:39:14 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 11:39:14 2020 -0800" }, "message": "Merge tag \u0027io_uring-5.6-2020-02-28\u0027 of git://git.kernel.dk/linux-block\n\nPull io_uring fixes from Jens Axboe:\n\n - Fix for a race with IOPOLL used with SQPOLL (Xiaoguang)\n\n - Only show -\u003efdinfo if procfs is enabled (Tobias)\n\n - Fix for a chain with multiple personalities in the SQEs\n\n - Fix for a missing free of personality idr on exit\n\n - Removal of the spin-for-work optimization\n\n - Fix for next work lookup on request completion\n\n - Fix for non-vec read/write result progation in case of links\n\n - Fix for a fileset references on switch\n\n - Fix for a recvmsg/sendmsg 32-bit compatability mode\n\n* tag \u0027io_uring-5.6-2020-02-28\u0027 of git://git.kernel.dk/linux-block:\n io_uring: fix 32-bit compatability with sendmsg/recvmsg\n io_uring: define and set show_fdinfo only if procfs is enabled\n io_uring: drop file set ref put/get on switch\n io_uring: import_single_range() returns 0/-ERROR\n io_uring: pick up link work on submit reference drop\n io-wq: ensure work-\u003etask_pid is cleared on init\n io-wq: remove spin-for-work optimization\n io_uring: fix poll_list race for SETUP_IOPOLL|SETUP_SQPOLL\n io_uring: fix personality idr leak\n io_uring: handle multiple personalities in link chains\n" }, { "commit": "5b8ea58b6a338cb981670c4408225331aedb4b89", "tree": "abe811e8684f25b41397639a2c41b54466bbc8a1", "parents": [ "cae740a04b4d6d5166f19ee5faf04ea2a1f34b3d", "9515743bfb39c61aaf3d4f3219a645c8d1fe9a0e" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Feb 28 10:02:36 2020 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Fri Feb 28 10:02:36 2020 -0700" }, "message": "Merge branch \u0027nvme-5.6-rc4\u0027 of git://git.infradead.org/nvme into block-5.6\n\nPull NVMe fix from Keith.\n\n* \u0027nvme-5.6-rc4\u0027 of git://git.infradead.org/nvme:\n nvme-pci: Hold cq_poll_lock while completing CQEs\n" }, { "commit": "c60c04021353c55b133519804734415f647f08bd", "tree": "193679c93495cef83ebbaeda3261cf246e27bda3", "parents": [ "36428598126ec20329537825ecd855ac77334301", "cabe17d0173ab04bd3f87b8199ae75f43f1ea473" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 09:02:18 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 09:02:18 2020 -0800" }, "message": "Merge tag \u0027acpi-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI fixes from Rafael Wysocki:\n \"Fix a couple of configuration issues in the ACPI watchdog (WDAT)\n driver (Mika Westerberg) and make it possible to disable that driver\n at boot time in case it still does not work as expected (Jean\n Delvare)\"\n\n* tag \u0027acpi-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n ACPI: watchdog: Set default timeout in probe\n ACPI: watchdog: Fix gas-\u003eaccess_width usage\n ACPICA: Introduce ACPI_ACCESS_BYTE_WIDTH() macro\n ACPI: watchdog: Allow disabling WDAT at boot\n" }, { "commit": "36428598126ec20329537825ecd855ac77334301", "tree": "2c3f089aaa6122450e04e5d8c7cfaf886740d243", "parents": [ "bfeb4f9977348daaaf7283ff365d81f7ee95940a", "189c6967fe61729fa7e9cdc32690b08fc7eebe41" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 08:49:52 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 08:49:52 2020 -0800" }, "message": "Merge tag \u0027pm-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fixes from Rafael Wysocki:\n \"Fix a recent cpufreq initialization regression (Rafael Wysocki),\n revert a devfreq commit that made incompatible changes and broke user\n land on some systems (Orson Zhai), drop a stale reference to a\n document that has gone away recently (Jonathan Neuschäfer), and fix a\n typo in a hibernation code comment (Alexandre Belloni)\"\n\n* tag \u0027pm-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n cpufreq: Fix policy initialization for internal governor drivers\n Revert \"PM / devfreq: Modify the device name as devfreq(X) for sysfs\"\n PM / hibernate: fix typo \"reserverd_size\" -\u003e \"reserved_size\"\n Documentation: power: Drop reference to interface.rst\n" }, { "commit": "bfeb4f9977348daaaf7283ff365d81f7ee95940a", "tree": "41fadae29769848070ce7483979d0e432009ef35", "parents": [ "45d0b75b98bf1de4b3a5b09188c75f3bfa3152b0", "0dda2ddb7ded1395189e95d43106469687c07795" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 08:34:47 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Feb 28 08:34:47 2020 -0800" }, "message": "Merge tag \u0027zonefs-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs\n\nPull zonefs fixes from Damien Le Moal:\n \"Two fixes in here:\n\n - Revert the initial decision to silently ignore IOCB_NOWAIT for\n asynchronous direct IOs to sequential zone files. Instead, return\n an error to the user to signal that the feature is not supported\n (from Christoph)\n\n - A fix to zonefs Kconfig to select FS_IOMAP to avoid build failures\n if no other file system already selected this option (from\n Johannes)\"\n\n* tag \u0027zonefs-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs:\n zonefs: select FS_IOMAP\n zonefs: fix IOCB_NOWAIT handling\n" }, { "commit": "e951445f4d3b5d0df69c0c5d18ab1e9058c20e52", "tree": "173541185ed04ca711273da2cb2ebd89dcf75753", "parents": [ "ef935c25fd648a17c27af5d1738b1884f78c5b75", "e43f1331e2ef913b8c566920c9af75e0ccdd1d3f" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 11:46:59 2020 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 11:50:06 2020 +0100" }, "message": "Merge tag \u0027kvmarm-fixes-5.6-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD\n\nKVM/arm fixes for 5.6, take #1\n\n- Fix compilation on 32bit\n- Move VHE guest entry/exit into the VHE-specific entry code\n- Make sure all functions called by the non-VHE HYP code is tagged as __always_inline\n" }, { "commit": "ef935c25fd648a17c27af5d1738b1884f78c5b75", "tree": "119ef5b87f3ea7eff4af959af74959366a7ece5b", "parents": [ "aaec7c03de92c35a96966631989950e6e27662db" ], "author": { "name": "Erwan Velu", "email": "erwanaliasr1@gmail.com", "time": "Thu Feb 27 19:00:46 2020 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 11:37:20 2020 +0100" }, "message": "kvm: x86: Limit the number of \"kvm: disabled by bios\" messages\n\nIn older version of systemd(219), at boot time, udevadm is called with :\n\t/usr/bin/udevadm trigger --type\u003ddevices --action\u003dadd\"\n\nThis program generates an echo \"add\" in /sys/devices/system/cpu/cpu\u003cx\u003e/uevent,\nleading to the \"kvm: disabled by bios\" message in case of your Bios disabled\nthe virtualization extensions.\n\nOn a modern system running up to 256 CPU threads, this pollutes the Kernel logs.\n\nThis patch offers to ratelimit this message to avoid any userspace program triggering\nthis uevent printing this message too often.\n\nThis patch is only a workaround but greatly reduce the pollution without\nbreaking the current behavior of printing a message if some try to instantiate\nKVM on a system that doesn\u0027t support it.\n\nNote that recent versions of systemd (\u003e239) do not have trigger this behavior.\n\nThis patch will be useful at least for some using older systemd with recent Kernels.\n\nSigned-off-by: Erwan Velu \u003ce.velu@criteo.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "189c6967fe61729fa7e9cdc32690b08fc7eebe41", "tree": "b193917f9e4f958da908a4cd4cd46d5b5d2bfb85", "parents": [ "f5739cb0b56590d68d8df8a44659893b6d0084c3", "b0c609ab2057d0953fa05e7566f0c0e8a28fa9e1", "1df97a02a937abf8473ccf0ef12edefcb11671de" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Feb 28 11:00:50 2020 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Feb 28 11:00:50 2020 +0100" }, "message": "Merge branches \u0027pm-sleep\u0027 and \u0027pm-devfreq\u0027\n\n* pm-sleep:\n PM / hibernate: fix typo \"reserverd_size\" -\u003e \"reserved_size\"\n Documentation: power: Drop reference to interface.rst\n\n* pm-devfreq:\n Revert \"PM / devfreq: Modify the device name as devfreq(X) for sysfs\"\n" }, { "commit": "aaec7c03de92c35a96966631989950e6e27662db", "tree": "ea7bca180557b531b94b6a1a3d81aaca81bd8c89", "parents": [ "4f337faf1c55e55bdc49df13fcb3a3c45655899e" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:49:10 2020 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:54:50 2020 +0100" }, "message": "KVM: x86: avoid useless copy of cpufreq policy\n\nstruct cpufreq_policy is quite big and it is not a good idea\nto allocate one on the stack. Just use cpufreq_cpu_get and\ncpufreq_cpu_put which is even simpler.\n\nReported-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "4f337faf1c55e55bdc49df13fcb3a3c45655899e", "tree": "362de3a83d19657b9bf767c0250b8bcda33803e2", "parents": [ "575b255c1663c8fccc41fe965dcac281e3113c65" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:42:31 2020 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:45:28 2020 +0100" }, "message": "KVM: allow disabling -Werror\n\nRestrict -Werror to well-tested configurations and allow disabling it\nvia Kconfig.\n\nReported-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "575b255c1663c8fccc41fe965dcac281e3113c65", "tree": "fc4937fbd94f1376803b0847ec16816ad05333f3", "parents": [ "8a9442f49c72bde43f982e53b74526ac37d3565b" ], "author": { "name": "Valdis Klētnieks", "email": "valdis.kletnieks@vt.edu", "time": "Thu Feb 27 21:49:52 2020 -0500" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:35:37 2020 +0100" }, "message": "KVM: x86: allow compiling as non-module with W\u003d1\n\nCompile error with CONFIG_KVM_INTEL\u003dy and W\u003d1:\n\n CC arch/x86/kvm/vmx/vmx.o\narch/x86/kvm/vmx/vmx.c:68:32: error: \u0027vmx_cpu_id\u0027 defined but not used [-Werror\u003dunused-const-variable\u003d]\n 68 | static const struct x86_cpu_id vmx_cpu_id[] \u003d {\n | ^~~~~~~~~~\ncc1: all warnings being treated as errors\n\nWhen building with \u003dy, the MODULE_DEVICE_TABLE macro doesn\u0027t generate a\nreference to the structure (or any code at all). This makes W\u003d1 compiles\nunhappy.\n\nWrap both in a #ifdef to avoid the issue.\n\nSigned-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n[Do the same for CONFIG_KVM_AMD. - Paolo]\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "8a9442f49c72bde43f982e53b74526ac37d3565b", "tree": "26afe2e6c08863d26927695bdd4208cab4e28ad2", "parents": [ "a262bca3aba03f0696995beb223c610e47533db3" ], "author": { "name": "Wanpeng Li", "email": "wanpengli@tencent.com", "time": "Tue Feb 18 09:08:24 2020 +0800" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:34:25 2020 +0100" }, "message": "KVM: Pre-allocate 1 cpumask variable per cpu for both pv tlb and pv ipis\n\nNick Desaulniers Reported:\n\n When building with:\n $ make CC\u003dclang arch/x86/ CFLAGS\u003d-Wframe-larger-than\u003d1000\n The following warning is observed:\n arch/x86/kernel/kvm.c:494:13: warning: stack frame size of 1064 bytes in\n function \u0027kvm_send_ipi_mask_allbutself\u0027 [-Wframe-larger-than\u003d]\n static void kvm_send_ipi_mask_allbutself(const struct cpumask *mask, int\n vector)\n ^\n Debugging with:\n https://github.com/ClangBuiltLinux/frame-larger-than\n via:\n $ python3 frame_larger_than.py arch/x86/kernel/kvm.o \\\n kvm_send_ipi_mask_allbutself\n points to the stack allocated `struct cpumask newmask` in\n `kvm_send_ipi_mask_allbutself`. The size of a `struct cpumask` is\n potentially large, as it\u0027s CONFIG_NR_CPUS divided by BITS_PER_LONG for\n the target architecture. CONFIG_NR_CPUS for X86_64 can be as high as\n 8192, making a single instance of a `struct cpumask` 1024 B.\n\nThis patch fixes it by pre-allocate 1 cpumask variable per cpu and use it for\nboth pv tlb and pv ipis..\n\nReported-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nAcked-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nReviewed-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nSigned-off-by: Wanpeng Li \u003cwanpengli@tencent.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "a262bca3aba03f0696995beb223c610e47533db3", "tree": "43f947857a1b297da542748a45bb33cd7f928d7e", "parents": [ "fcd07f9adc7dacc2532695cf9dd2284d49e716ff" ], "author": { "name": "Wanpeng Li", "email": "wanpengli@tencent.com", "time": "Tue Feb 18 09:08:23 2020 +0800" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:34:19 2020 +0100" }, "message": "KVM: Introduce pv check helpers\n\nIntroduce some pv check helpers for consistency.\n\nSuggested-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nReviewed-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\nSigned-off-by: Wanpeng Li \u003cwanpengli@tencent.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "fcd07f9adc7dacc2532695cf9dd2284d49e716ff", "tree": "f3dfa29c35eff077b102af6390eeb6d9a889f6b7", "parents": [ "7943f4acea3caf0b6d5b6cdfce7d5a2b4a9aa608" ], "author": { "name": "Christian Borntraeger", "email": "borntraeger@de.ibm.com", "time": "Fri Feb 28 09:49:41 2020 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:33:57 2020 +0100" }, "message": "KVM: let declaration of kvm_get_running_vcpus match implementation\n\nSparse notices that declaration and implementation do not match:\narch/s390/kvm/../../../virt/kvm/kvm_main.c:4435:17: warning: incorrect type in return expression (different address spaces)\narch/s390/kvm/../../../virt/kvm/kvm_main.c:4435:17: expected struct kvm_vcpu [noderef] \u003casn:3\u003e **\narch/s390/kvm/../../../virt/kvm/kvm_main.c:4435:17: got struct kvm_vcpu *[noderef] \u003casn:3\u003e *\n\nSigned-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "7943f4acea3caf0b6d5b6cdfce7d5a2b4a9aa608", "tree": "bafefb073f8bbe836e51354a19bbd5ecfa662cd2", "parents": [ "a93236fcbe1d0248461b29c0f87cb0b510c94e6f" ], "author": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Tue Feb 25 08:54:26 2020 +0100" }, "committer": { "name": "Paolo Bonzini", "email": "pbonzini@redhat.com", "time": "Fri Feb 28 10:33:17 2020 +0100" }, "message": "KVM: SVM: allocate AVIC data structures based on kvm_amd module parameter\n\nEven if APICv is disabled at startup, the backing page and ir_list need\nto be initialized in case they are needed later. The only case in\nwhich this can be skipped is for userspace irqchip, and that must be\ndone because avic_init_backing_page dereferences vcpu-\u003earch.apic\n(which is NULL for userspace irqchip).\n\nTested-by: rmuncrief@humanavance.com\nFixes: https://bugzilla.kernel.org/show_bug.cgi?id\u003d206579\nReviewed-by: Miaohe Lin \u003clinmiaohe@huawei.com\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n" }, { "commit": "45d0b75b98bf1de4b3a5b09188c75f3bfa3152b0", "tree": "6b0830a6228d89f5b3f3421a13dbf9df5fc83f91", "parents": [ "7058b837899fc978c9f8a033fa29ab07360a85c8", "f091bf39700dd086ab244c823f389556fed0c513" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 21:52:18 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 21:52:18 2020 -0800" }, "message": "Merge tag \u0027drm-fixes-2020-02-28\u0027 of git://anongit.freedesktop.org/drm/drm\n\nPull drm fixes from Dave Airlie:\n \"Just some fixes for this week: amdgpu, radeon and i915.\n\n The main i915 one is a regression Gen7 (Ivybridge/Haswell), this moves\n them back from trying to use the full-ppgtt support to the aliasing\n version it used to use due to gpu hangs. Otherwise it\u0027s pretty quiet.\n\n amdgpu:\n - Drop DRIVER_USE_AGP\n - Fix memory leak in GPU reset\n - Resume fix for raven\n\n radeon:\n - Drop DRIVER_USE_AGP\n\n i915:\n - downgrade gen7 back to aliasing-ppgtt to avoid GPU hangs\n - shrinker fix\n - pmu leak and double free fixes\n - gvt user after free and virtual display reset fixes\n - randconfig build fix\"\n\n* tag \u0027drm-fixes-2020-02-28\u0027 of git://anongit.freedesktop.org/drm/drm:\n drm/radeon: Inline drm_get_pci_dev\n drm/amdgpu: Drop DRIVER_USE_AGP\n drm/i915: Avoid recursing onto active vma from the shrinker\n drm/i915/pmu: Avoid using globals for PMU events\n drm/i915/pmu: Avoid using globals for CPU hotplug state\n drm/i915/gtt: Downgrade gen7 (ivb, byt, hsw) back to aliasing-ppgtt\n drm/i915: fix header test with GCOV\n amdgpu/gmc_v9: save/restore sdpif regs during S3\n drm/amdgpu: fix memory leak during TDR test(v2)\n drm/i915/gvt: Fix orphan vgpu dmabuf_objs\u0027 lifetime\n drm/i915/gvt: Separate display reset from ALL_ENGINES reset\n" }, { "commit": "f091bf39700dd086ab244c823f389556fed0c513", "tree": "5a48e931ead489fec17e0e0c599f6963c162fa68", "parents": [ "e180af1970d535f9d19309e91b0cd4379248b19f", "8e9a400c706ee5bcbc052d3ec9f87cfdbbd3f5cb" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 28 12:40:41 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 28 12:40:49 2020 +1000" }, "message": "Merge tag \u0027drm-intel-fixes-2020-02-27\u0027 of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes\n\ndrm/i915 fixes for v5.6-rc4:\n- downgrade gen7 back to aliasing-ppgtt to avoid GPU hangs\n- shrinker fix\n- pmu leak and double free fixes\n- gvt user after free and virtual display reset fixes\n- randconfig build fix\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\n\nFrom: Jani Nikula \u003cjani.nikula@intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/874kvcsh00.fsf@intel.com\n" }, { "commit": "e180af1970d535f9d19309e91b0cd4379248b19f", "tree": "6e116a67784e43a97b2f5e439679fde33dc5a104", "parents": [ "f8788d86ab28f61f7b46eb6be375f8a726783636", "eb12c957735b582607e5842a06d1f4c62e185c1d" ], "author": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 28 12:30:18 2020 +1000" }, "committer": { "name": "Dave Airlie", "email": "airlied@redhat.com", "time": "Fri Feb 28 12:30:20 2020 +1000" }, "message": "Merge tag \u0027amd-drm-fixes-5.6-2020-02-26\u0027 of git://people.freedesktop.org/~agd5f/linux into drm-fixes\n\namd-drm-fixes-5.6-2020-02-26:\n\namdgpu:\n- Drop DRIVER_USE_AGP\n- Fix memory leak in GPU reset\n- Resume fix for raven\n\nradeon:\n- Drop DRIVER_USE_AGP\n\nSigned-off-by: Dave Airlie \u003cairlied@redhat.com\u003e\nFrom: Alex Deucher \u003calexdeucher@gmail.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20200227034106.3912-1-alexander.deucher@amd.com\n" }, { "commit": "7058b837899fc978c9f8a033fa29ab07360a85c8", "tree": "94a8180a13f40f58ba07b01f9b129c72a278835c", "parents": [ "a2f0b878c3ca531a1706cb2a8b079cea3b17bafc", "3ee339eb28959629db33aaa2b8cde4c63c6289eb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 16:34:41 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 16:34:41 2020 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from David Miller:\n\n 1) Fix leak in nl80211 AP start where we leak the ACL memory, from\n Johannes Berg.\n\n 2) Fix double mutex unlock in mac80211, from Andrei Otcheretianski.\n\n 3) Fix RCU stall in ipset, from Jozsef Kadlecsik.\n\n 4) Fix devlink locking in devlink_dpipe_table_register, from Madhuparna\n Bhowmik.\n\n 5) Fix race causing TX hang in ll_temac, from Esben Haabendal.\n\n 6) Stale eth hdr pointer in br_dev_xmit(), from Nikolay Aleksandrov.\n\n 7) Fix TX hash calculation bounds checking wrt. tc rules, from Amritha\n Nambiar.\n\n 8) Size netlink responses properly in schedule action code to take into\n consideration TCA_ACT_FLAGS. From Jiri Pirko.\n\n 9) Fix firmware paths for mscc PHY driver, from Antoine Tenart.\n\n10) Don\u0027t register stmmac notifier multiple times, from Aaro Koskinen.\n\n11) Various rmnet bug fixes, from Taehee Yoo.\n\n12) Fix vsock deadlock in vsock transport release, from Stefano\n Garzarella.\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (61 commits)\n net: dsa: mv88e6xxx: Fix masking of egress port\n mlxsw: pci: Wait longer before accessing the device after reset\n sfc: fix timestamp reconstruction at 16-bit rollover points\n vsock: fix potential deadlock in transport-\u003erelease()\n unix: It\u0027s CONFIG_PROC_FS not CONFIG_PROCFS\n net: rmnet: fix packet forwarding in rmnet bridge mode\n net: rmnet: fix bridge mode bugs\n net: rmnet: use upper/lower device infrastructure\n net: rmnet: do not allow to change mux id if mux id is duplicated\n net: rmnet: remove rcu_read_lock in rmnet_force_unassociate_device()\n net: rmnet: fix suspicious RCU usage\n net: rmnet: fix NULL pointer dereference in rmnet_changelink()\n net: rmnet: fix NULL pointer dereference in rmnet_newlink()\n net: phy: marvell: don\u0027t interpret PHY status unless resolved\n mlx5: register lag notifier for init network namespace only\n unix: define and set show_fdinfo only if procfs is enabled\n hinic: fix a bug of rss configuration\n hinic: fix a bug of setting hw_ioctxt\n hinic: fix a irq affinity bug\n net/smc: check for valid ib_client_data\n ...\n" }, { "commit": "5901b51f3e5d9129da3e59b10cc76e4cc983e940", "tree": "bcdcb484f3e2a6609ccce32bc5d830a2b1458b71", "parents": [ "73a7a271b3eee7b83f29b13866163776f1cbef89" ], "author": { "name": "Lukas Bulwahn", "email": "lukas.bulwahn@gmail.com", "time": "Fri Feb 21 19:54:02 2020 +0100" }, "committer": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Thu Feb 27 15:49:28 2020 -0600" }, "message": "MAINTAINERS: Correct Cadence PCI driver path\n\nde80f95ccb9c (\"PCI: cadence: Move all files to per-device cadence\ndirectory\") moved files of the PCI cadence drivers, but did not update the\nMAINTAINERS entry.\n\nSince then, ./scripts/get_maintainer.pl --self-test complains:\n\n warning: no file matches F: drivers/pci/controller/pcie-cadence*\n\nRepair the MAINTAINERS entry.\n\nLink: https://lore.kernel.org/r/20200221185402.4703-1-lukas.bulwahn@gmail.com\nSigned-off-by: Lukas Bulwahn \u003clukas.bulwahn@gmail.com\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n" }, { "commit": "d876836204897b6d7d911f942084f69a1e9d5c4d", "tree": "86e904e35ae5a6bdfea02665e869e0be78d7677b", "parents": [ "bebdb65e077267957f48e43d205d4a16cc7b8161" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Feb 27 14:17:49 2020 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Feb 27 14:17:49 2020 -0700" }, "message": "io_uring: fix 32-bit compatability with sendmsg/recvmsg\n\nWe must set MSG_CMSG_COMPAT if we\u0027re in compatability mode, otherwise\nthe iovec import for these commands will not do the right thing and fail\nthe command with -EINVAL.\n\nFound by running the test suite compiled as 32-bit.\n\nCc: stable@vger.kernel.org\nFixes: aa1fa28fc73e (\"io_uring: add support for recvmsg()\")\nFixes: 0fa03c624d8f (\"io_uring: add support for sendmsg()\")\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "3ee339eb28959629db33aaa2b8cde4c63c6289eb", "tree": "1fdf38711d96fd421de69d346cb668eb8d34f79a", "parents": [ "ac004e84164e27d69017731a97b11402a69d854b" ], "author": { "name": "Andrew Lunn", "email": "andrew@lunn.ch", "time": "Thu Feb 27 21:20:49 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 12:29:09 2020 -0800" }, "message": "net: dsa: mv88e6xxx: Fix masking of egress port\n\nAdd missing ~ to the usage of the mask.\n\nReported-by: Kevin Benson \u003cKevin.Benson@zii.aero\u003e\nReported-by: Chris Healy \u003cChris.Healy@zii.aero\u003e\nFixes: 5c74c54ce6ff (\"net: dsa: mv88e6xxx: Split monitor port configuration\")\nSigned-off-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ac004e84164e27d69017731a97b11402a69d854b", "tree": "ad88760ed5936fe96c05a4dcd43d4983c4f659e6", "parents": [ "23797b98909f34b75fd130369bde86f760db69d0" ], "author": { "name": "Amit Cohen", "email": "amitc@mellanox.com", "time": "Thu Feb 27 21:07:53 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 12:09:22 2020 -0800" }, "message": "mlxsw: pci: Wait longer before accessing the device after reset\n\nDuring initialization the driver issues a reset to the device and waits\nfor 100ms before checking if the firmware is ready. The waiting is\nnecessary because before that the device is irresponsive and the first\nread can result in a completion timeout.\n\nWhile 100ms is sufficient for Spectrum-1 and Spectrum-2, it is\ninsufficient for Spectrum-3.\n\nFix this by increasing the timeout to 200ms.\n\nFixes: da382875c616 (\"mlxsw: spectrum: Extend to support Spectrum-3 ASIC\")\nSigned-off-by: Amit Cohen \u003camitc@mellanox.com\u003e\nSigned-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\nSigned-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "23797b98909f34b75fd130369bde86f760db69d0", "tree": "8ab3f20d8390108d66db979b51aa460bc14aa2b2", "parents": [ "3f74957fcbeab703297ed0f135430414ed7e0dd0" ], "author": { "name": "Alex Maftei (amaftei)", "email": "amaftei@solarflare.com", "time": "Wed Feb 26 17:33:19 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 12:04:49 2020 -0800" }, "message": "sfc: fix timestamp reconstruction at 16-bit rollover points\n\nWe can\u0027t just use the top bits of the last sync event as they could be\noff-by-one every 65,536 seconds, giving an error in reconstruction of\n65,536 seconds.\n\nThis patch uses the difference in the bottom 16 bits (mod 2^16) to\ncalculate an offset that needs to be applied to the last sync event to\nget to the current time.\n\nSigned-off-by: Alexandru-Mihai Maftei \u003camaftei@solarflare.com\u003e\nAcked-by: Martin Habets \u003cmhabets@solarflare.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3f74957fcbeab703297ed0f135430414ed7e0dd0", "tree": "1b8cadd1963b397e027e022e1ce2f04f7b378d6d", "parents": [ "5c05a164d441a1792791175e4959ea9df12f7e2b" ], "author": { "name": "Stefano Garzarella", "email": "sgarzare@redhat.com", "time": "Wed Feb 26 11:58:18 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 12:03:56 2020 -0800" }, "message": "vsock: fix potential deadlock in transport-\u003erelease()\n\nSome transports (hyperv, virtio) acquire the sock lock during the\n.release() callback.\n\nIn the vsock_stream_connect() we call vsock_assign_transport(); if\nthe socket was previously assigned to another transport, the\nvsk-\u003etransport-\u003erelease() is called, but the sock lock is already\nheld in the vsock_stream_connect(), causing a deadlock reported by\nsyzbot:\n\n INFO: task syz-executor280:9768 blocked for more than 143 seconds.\n Not tainted 5.6.0-rc1-syzkaller #0\n \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n syz-executor280 D27912 9768 9766 0x00000000\n Call Trace:\n context_switch kernel/sched/core.c:3386 [inline]\n __schedule+0x934/0x1f90 kernel/sched/core.c:4082\n schedule+0xdc/0x2b0 kernel/sched/core.c:4156\n __lock_sock+0x165/0x290 net/core/sock.c:2413\n lock_sock_nested+0xfe/0x120 net/core/sock.c:2938\n virtio_transport_release+0xc4/0xd60 net/vmw_vsock/virtio_transport_common.c:832\n vsock_assign_transport+0xf3/0x3b0 net/vmw_vsock/af_vsock.c:454\n vsock_stream_connect+0x2b3/0xc70 net/vmw_vsock/af_vsock.c:1288\n __sys_connect_file+0x161/0x1c0 net/socket.c:1857\n __sys_connect+0x174/0x1b0 net/socket.c:1874\n __do_sys_connect net/socket.c:1885 [inline]\n __se_sys_connect net/socket.c:1882 [inline]\n __x64_sys_connect+0x73/0xb0 net/socket.c:1882\n do_syscall_64+0xfa/0x790 arch/x86/entry/common.c:294\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n\nTo avoid this issue, this patch remove the lock acquiring in the\n.release() callback of hyperv and virtio transports, and it holds\nthe lock when we call vsk-\u003etransport-\u003erelease() in the vsock core.\n\nReported-by: syzbot+731710996d79d0d58fbc@syzkaller.appspotmail.com\nFixes: 408624af4c89 (\"vsock: use local transport when it is loaded\")\nSigned-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nReviewed-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "5c05a164d441a1792791175e4959ea9df12f7e2b", "tree": "272bf41dd873c2ac27d47eae716e616221b297c6", "parents": [ "795c03a5d09c37f153affa0bdb416c16f678bd7e" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:52:35 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:52:35 2020 -0800" }, "message": "unix: It\u0027s CONFIG_PROC_FS not CONFIG_PROCFS\n\nFixes: 3a12500ed5dd (\"unix: define and set show_fdinfo only if procfs is enabled\")\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "795c03a5d09c37f153affa0bdb416c16f678bd7e", "tree": "bf4be7bd700ab44340fb4378717c52ca9e5f37e4", "parents": [ "b82cf17ff1957ec35eaee7dc519c365ecd06ba38", "ad3cc31b599ea80f06b29ebdc18b3a39878a48d6" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "Merge branch \u0027net-rmnet-fix-several-bugs\u0027\n\nTaehee Yoo says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: rmnet: fix several bugs\n\nThis patchset is to fix several bugs in RMNET module.\n\n1. The first patch fixes NULL-ptr-deref in rmnet_newlink().\nWhen rmnet interface is being created, it uses IFLA_LINK\nwithout checking NULL.\nSo, if userspace doesn\u0027t set IFLA_LINK, panic will occur.\nIn this patch, checking NULL pointer code is added.\n\n2. The second patch fixes NULL-ptr-deref in rmnet_changelink().\nTo get real device in rmnet_changelink(), it uses IFLA_LINK.\nBut, IFLA_LINK should not be used in rmnet_changelink().\n\n3. The third patch fixes suspicious RCU usage in rmnet_get_port().\nrmnet_get_port() uses rcu_dereference_rtnl().\nBut, rmnet_get_port() is used by datapath.\nSo, rcu_dereference_bh() should be used instead of rcu_dereference_rtnl().\n\n4. The fourth patch fixes suspicious RCU usage in\nrmnet_force_unassociate_device().\nRCU critical section should not be scheduled.\nBut, unregister_netdevice_queue() in the rmnet_force_unassociate_device()\nwould be scheduled.\nSo, the RCU warning occurs.\nIn this patch, the rcu_read_lock() in the rmnet_force_unassociate_device()\nis removed because it\u0027s unnecessary.\n\n5. The fifth patch fixes duplicate MUX ID case.\nRMNET MUX ID is unique.\nSo, rmnet interface isn\u0027t allowed to be created, which have\na duplicate MUX ID.\nBut, only rmnet_newlink() checks this condition, rmnet_changelink()\ndoesn\u0027t check this.\nSo, duplicate MUX ID case would happen.\n\n6. The sixth patch fixes upper/lower interface relationship problems.\nWhen IFLA_LINK is used, the upper/lower infrastructure should be used.\nBecause it checks the maximum depth of upper/lower interfaces and it also\nchecks circular interface relationship, etc.\nIn this patch, netdev_upper_dev_link() is used.\n\n7. The seventh patch fixes bridge related problems.\na) -\u003endo_del_slave() doesn\u0027t work.\nb) It couldn\u0027t detect circular upper/lower interface relationship.\nc) It couldn\u0027t prevent stack overflow because of too deep depth\nof upper/lower interface\nd) It doesn\u0027t check the number of lower interfaces.\ne) Panics because of several reasons.\nThese problems are actually the same problem.\nSo, this patch fixes these problems.\n\n8. The eighth patch fixes packet forwarding issue in bridge mode\nPacket forwarding is not working in rmnet bridge mode.\nBecause when a packet is forwarded, skb_push() for an ethernet header\nis needed. But it doesn\u0027t call skb_push().\nSo, the ethernet header will be lost.\n\nChange log:\n - update commit logs.\n - drop two patches in this patchset because of wrong target branch.\n - (\"net: rmnet: add missing module alias\")\n - (\"net: rmnet: print error message when command fails\")\n - remove unneessary rcu_read_lock() in the third patch.\n - use rcu_dereference_bh() instead of rcu_dereference in third patch.\n - do not allow to add a bridge device if rmnet interface is already\n bridge mode in the seventh patch.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "ad3cc31b599ea80f06b29ebdc18b3a39878a48d6", "tree": "bf4be7bd700ab44340fb4378717c52ca9e5f37e4", "parents": [ "d939b6d30bea1a2322bc536b12be0a7c4c2bccd7" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:26:15 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: fix packet forwarding in rmnet bridge mode\n\nPacket forwarding is not working in rmnet bridge mode.\nBecause when a packet is forwarded, skb_push() for an ethernet header\nis needed. But it doesn\u0027t call skb_push().\nSo, the ethernet header will be lost.\n\nTest commands:\n modprobe rmnet\n ip netns add nst\n ip netns add nst2\n ip link add veth0 type veth peer name veth1\n ip link add veth2 type veth peer name veth3\n ip link set veth1 netns nst\n ip link set veth3 netns nst2\n\n ip link add rmnet0 link veth0 type rmnet mux_id 1\n ip link set veth2 master rmnet0\n ip link set veth0 up\n ip link set veth2 up\n ip link set rmnet0 up\n ip a a 192.168.100.1/24 dev rmnet0\n\n ip netns exec nst ip link set veth1 up\n ip netns exec nst ip a a 192.168.100.2/24 dev veth1\n ip netns exec nst2 ip link set veth3 up\n ip netns exec nst2 ip a a 192.168.100.3/24 dev veth3\n ip netns exec nst2 ping 192.168.100.2\n\nFixes: 60d58f971c10 (\"net: qualcomm: rmnet: Implement bridge mode\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d939b6d30bea1a2322bc536b12be0a7c4c2bccd7", "tree": "521efd669297fc30751877dbf15aba5e51fa9de3", "parents": [ "037f9cdf72fb8a7ff9ec2b5dd05336ec1492bdf1" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:26:02 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: fix bridge mode bugs\n\nIn order to attach a bridge interface to the rmnet interface,\n\"master\" operation is used.\n(e.g. ip link set dummy1 master rmnet0)\nBut, in the rmnet_add_bridge(), which is a callback of -\u003endo_add_slave()\ndoesn\u0027t register lower interface.\nSo, -\u003endo_del_slave() doesn\u0027t work.\nThere are other problems too.\n1. It couldn\u0027t detect circular upper/lower interface relationship.\n2. It couldn\u0027t prevent stack overflow because of too deep depth\nof upper/lower interface\n3. It doesn\u0027t check the number of lower interfaces.\n4. Panics because of several reasons.\n\nThe root problem of these issues is actually the same.\nSo, in this patch, these all problems will be fixed.\n\nTest commands:\n modprobe rmnet\n ip link add dummy0 type dummy\n ip link add rmnet0 link dummy0 type rmnet mux_id 1\n ip link add dummy1 master rmnet0 type dummy\n ip link add dummy2 master rmnet0 type dummy\n ip link del rmnet0\n ip link del dummy2\n ip link del dummy1\n\nSplat looks like:\n[ 41.867595][ T1164] general protection fault, probably for non-canonical address 0xdffffc0000000101I\n[ 41.869993][ T1164] KASAN: null-ptr-deref in range [0x0000000000000808-0x000000000000080f]\n[ 41.872950][ T1164] CPU: 0 PID: 1164 Comm: ip Not tainted 5.6.0-rc1+ #447\n[ 41.873915][ T1164] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 41.875161][ T1164] RIP: 0010:rmnet_unregister_bridge.isra.6+0x71/0xf0 [rmnet]\n[ 41.876178][ T1164] Code: 48 89 ef 48 89 c6 5b 5d e9 fc fe ff ff e8 f7 f3 ff ff 48 8d b8 08 08 00 00 48 ba 00 7\n[ 41.878925][ T1164] RSP: 0018:ffff8880c4d0f188 EFLAGS: 00010202\n[ 41.879774][ T1164] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000101\n[ 41.887689][ T1164] RDX: dffffc0000000000 RSI: ffffffffb8cf64f0 RDI: 0000000000000808\n[ 41.888727][ T1164] RBP: ffff8880c40e4000 R08: ffffed101b3c0e3c R09: 0000000000000001\n[ 41.889749][ T1164] R10: 0000000000000001 R11: ffffed101b3c0e3b R12: 1ffff110189a1e3c\n[ 41.890783][ T1164] R13: ffff8880c4d0f200 R14: ffffffffb8d56160 R15: ffff8880ccc2c000\n[ 41.891794][ T1164] FS: 00007f4300edc0c0(0000) GS:ffff8880d9c00000(0000) knlGS:0000000000000000\n[ 41.892953][ T1164] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[ 41.893800][ T1164] CR2: 00007f43003bc8c0 CR3: 00000000ca53e001 CR4: 00000000000606f0\n[ 41.894824][ T1164] Call Trace:\n[ 41.895274][ T1164] ? rcu_is_watching+0x2c/0x80\n[ 41.895895][ T1164] rmnet_config_notify_cb+0x1f7/0x590 [rmnet]\n[ 41.896687][ T1164] ? rmnet_unregister_bridge.isra.6+0xf0/0xf0 [rmnet]\n[ 41.897611][ T1164] ? rmnet_unregister_bridge.isra.6+0xf0/0xf0 [rmnet]\n[ 41.898508][ T1164] ? __module_text_address+0x13/0x140\n[ 41.899162][ T1164] notifier_call_chain+0x90/0x160\n[ 41.899814][ T1164] rollback_registered_many+0x660/0xcf0\n[ 41.900544][ T1164] ? netif_set_real_num_tx_queues+0x780/0x780\n[ 41.901316][ T1164] ? __lock_acquire+0xdfe/0x3de0\n[ 41.901958][ T1164] ? memset+0x1f/0x40\n[ 41.902468][ T1164] ? __nla_validate_parse+0x98/0x1ab0\n[ 41.903166][ T1164] unregister_netdevice_many.part.133+0x13/0x1b0\n[ 41.903988][ T1164] rtnl_delete_link+0xbc/0x100\n[ ... ]\n\nFixes: 60d58f971c10 (\"net: qualcomm: rmnet: Implement bridge mode\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "037f9cdf72fb8a7ff9ec2b5dd05336ec1492bdf1", "tree": "524c914a51ea33926d93043ff66d0ddd185dc801", "parents": [ "1dc49e9d164cd7e11c81279c83db84a147e14740" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:25:43 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: use upper/lower device infrastructure\n\nnetdev_upper_dev_link() is useful to manage lower/upper interfaces.\nAnd this function internally validates looping, maximum depth.\nAll or most virtual interfaces that could have a real interface\n(e.g. macsec, macvlan, ipvlan etc.) use lower/upper infrastructure.\n\nTest commands:\n modprobe rmnet\n ip link add dummy0 type dummy\n ip link add rmnet1 link dummy0 type rmnet mux_id 1\n for i in {2..100}\n do\n let A\u003d$i-1\n ip link add rmnet$i link rmnet$A type rmnet mux_id $i\n done\n ip link del dummy0\n\nThe purpose of the test commands is to make stack overflow.\n\nSplat looks like:\n[ 52.411438][ T1395] BUG: KASAN: slab-out-of-bounds in find_busiest_group+0x27e/0x2c00\n[ 52.413218][ T1395] Write of size 64 at addr ffff8880c774bde0 by task ip/1395\n[ 52.414841][ T1395]\n[ 52.430720][ T1395] CPU: 1 PID: 1395 Comm: ip Not tainted 5.6.0-rc1+ #447\n[ 52.496511][ T1395] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 52.513597][ T1395] Call Trace:\n[ 52.546516][ T1395]\n[ 52.558773][ T1395] Allocated by task 3171537984:\n[ 52.588290][ T1395] BUG: unable to handle page fault for address: ffffffffb999e260\n[ 52.589311][ T1395] #PF: supervisor read access in kernel mode\n[ 52.590529][ T1395] #PF: error_code(0x0000) - not-present page\n[ 52.591374][ T1395] PGD d6818067 P4D d6818067 PUD d6819063 PMD 0\n[ 52.592288][ T1395] Thread overran stack, or stack corrupted\n[ 52.604980][ T1395] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN PTI\n[ 52.605856][ T1395] CPU: 1 PID: 1395 Comm: ip Not tainted 5.6.0-rc1+ #447\n[ 52.611764][ T1395] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 52.621520][ T1395] RIP: 0010:stack_depot_fetch+0x10/0x30\n[ 52.622296][ T1395] Code: ff e9 f9 fe ff ff 48 89 df e8 9c 1d 91 ff e9 ca fe ff ff cc cc cc cc cc cc cc 89 f8 0\n[ 52.627887][ T1395] RSP: 0018:ffff8880c774bb60 EFLAGS: 00010006\n[ 52.628735][ T1395] RAX: 00000000001f8880 RBX: ffff8880c774d140 RCX: 0000000000000000\n[ 52.631773][ T1395] RDX: 000000000000001d RSI: ffff8880c774bb68 RDI: 0000000000003ff0\n[ 52.649584][ T1395] RBP: ffffea00031dd200 R08: ffffed101b43e403 R09: ffffed101b43e403\n[ 52.674857][ T1395] R10: 0000000000000001 R11: ffffed101b43e402 R12: ffff8880d900e5c0\n[ 52.678257][ T1395] R13: ffff8880c774c000 R14: 0000000000000000 R15: dffffc0000000000\n[ 52.694541][ T1395] FS: 00007fe867f6e0c0(0000) GS:ffff8880da000000(0000) knlGS:0000000000000000\n[ 52.764039][ T1395] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[ 52.815008][ T1395] CR2: ffffffffb999e260 CR3: 00000000c26aa005 CR4: 00000000000606e0\n[ 52.862312][ T1395] Call Trace:\n[ 52.887133][ T1395] Modules linked in: dummy rmnet veth openvswitch nsh nf_conncount nf_nat nf_conntrack nf_dex\n[ 52.936749][ T1395] CR2: ffffffffb999e260\n[ 52.965695][ T1395] ---[ end trace 7e32ca99482dbb31 ]---\n[ 52.966556][ T1395] RIP: 0010:stack_depot_fetch+0x10/0x30\n[ 52.971083][ T1395] Code: ff e9 f9 fe ff ff 48 89 df e8 9c 1d 91 ff e9 ca fe ff ff cc cc cc cc cc cc cc 89 f8 0\n[ 53.003650][ T1395] RSP: 0018:ffff8880c774bb60 EFLAGS: 00010006\n[ 53.043183][ T1395] RAX: 00000000001f8880 RBX: ffff8880c774d140 RCX: 0000000000000000\n[ 53.076480][ T1395] RDX: 000000000000001d RSI: ffff8880c774bb68 RDI: 0000000000003ff0\n[ 53.093858][ T1395] RBP: ffffea00031dd200 R08: ffffed101b43e403 R09: ffffed101b43e403\n[ 53.112795][ T1395] R10: 0000000000000001 R11: ffffed101b43e402 R12: ffff8880d900e5c0\n[ 53.139837][ T1395] R13: ffff8880c774c000 R14: 0000000000000000 R15: dffffc0000000000\n[ 53.141500][ T1395] FS: 00007fe867f6e0c0(0000) GS:ffff8880da000000(0000) knlGS:0000000000000000\n[ 53.143343][ T1395] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[ 53.152007][ T1395] CR2: ffffffffb999e260 CR3: 00000000c26aa005 CR4: 00000000000606e0\n[ 53.156459][ T1395] Kernel panic - not syncing: Fatal exception\n[ 54.213570][ T1395] Shutting down cpus with NMI\n[ 54.354112][ T1395] Kernel Offset: 0x33000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0x)\n[ 54.355687][ T1395] Rebooting in 5 seconds..\n\nFixes: b37f78f234bf (\"net: qualcomm: rmnet: Fix crash on real dev unregistration\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1dc49e9d164cd7e11c81279c83db84a147e14740", "tree": "076aa32842a8b54275d9cb226f8146c970b178ee", "parents": [ "c026d970102e9af9958edefb4a015702c6aab636" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:25:19 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: do not allow to change mux id if mux id is duplicated\n\nBasically, duplicate mux id isn\u0027t be allowed.\nSo, the creation of rmnet will be failed if there is duplicate mux id\nis existing.\nBut, changelink routine doesn\u0027t check duplicate mux id.\n\nTest commands:\n modprobe rmnet\n ip link add dummy0 type dummy\n ip link add rmnet0 link dummy0 type rmnet mux_id 1\n ip link add rmnet1 link dummy0 type rmnet mux_id 2\n ip link set rmnet1 type rmnet mux_id 1\n\nFixes: 23790ef12082 (\"net: qualcomm: rmnet: Allow to configure flags for existing devices\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c026d970102e9af9958edefb4a015702c6aab636", "tree": "90a7cd42b179cd7446da9324295e59ba3964e84f", "parents": [ "102210f7664442d8c0ce332c006ea90626df745b" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:25:05 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: remove rcu_read_lock in rmnet_force_unassociate_device()\n\nThe notifier_call() of the slave interface removes rmnet interface with\nunregister_netdevice_queue().\nBut, before calling unregister_netdevice_queue(), it acquires\nrcu readlock.\nIn the RCU critical section, sleeping isn\u0027t be allowed.\nBut, unregister_netdevice_queue() internally calls synchronize_net(),\nwhich would sleep.\nSo, suspicious RCU usage warning occurs.\n\nTest commands:\n modprobe rmnet\n ip link add dummy0 type dummy\n ip link add dummy1 type dummy\n ip link add rmnet0 link dummy0 type rmnet mux_id 1\n ip link set dummy1 master rmnet0\n ip link del dummy0\n\nSplat looks like:\n[ 79.639245][ T1195] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ 79.640134][ T1195] WARNING: suspicious RCU usage\n[ 79.640852][ T1195] 5.6.0-rc1+ #447 Not tainted\n[ 79.641657][ T1195] -----------------------------\n[ 79.642472][ T1195] ./include/linux/rcupdate.h:273 Illegal context switch in RCU read-side critical section!\n[ 79.644043][ T1195]\n[ 79.644043][ T1195] other info that might help us debug this:\n[ 79.644043][ T1195]\n[ 79.645682][ T1195]\n[ 79.645682][ T1195] rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n[ 79.646980][ T1195] 2 locks held by ip/1195:\n[ 79.647629][ T1195] #0: ffffffffa3cf64f0 (rtnl_mutex){+.+.}, at: rtnetlink_rcv_msg+0x457/0x890\n[ 79.649312][ T1195] #1: ffffffffa39256c0 (rcu_read_lock){....}, at: rmnet_config_notify_cb+0xf0/0x590 [rmnet]\n[ 79.651717][ T1195]\n[ 79.651717][ T1195] stack backtrace:\n[ 79.652650][ T1195] CPU: 3 PID: 1195 Comm: ip Not tainted 5.6.0-rc1+ #447\n[ 79.653702][ T1195] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 79.655037][ T1195] Call Trace:\n[ 79.655560][ T1195] dump_stack+0x96/0xdb\n[ 79.656252][ T1195] ___might_sleep+0x345/0x440\n[ 79.656994][ T1195] synchronize_net+0x18/0x30\n[ 79.661132][ T1195] netdev_rx_handler_unregister+0x40/0xb0\n[ 79.666266][ T1195] rmnet_unregister_real_device+0x42/0xb0 [rmnet]\n[ 79.667211][ T1195] rmnet_config_notify_cb+0x1f7/0x590 [rmnet]\n[ 79.668121][ T1195] ? rmnet_unregister_bridge.isra.6+0xf0/0xf0 [rmnet]\n[ 79.669166][ T1195] ? rmnet_unregister_bridge.isra.6+0xf0/0xf0 [rmnet]\n[ 79.670286][ T1195] ? __module_text_address+0x13/0x140\n[ 79.671139][ T1195] notifier_call_chain+0x90/0x160\n[ 79.671973][ T1195] rollback_registered_many+0x660/0xcf0\n[ 79.672893][ T1195] ? netif_set_real_num_tx_queues+0x780/0x780\n[ 79.675091][ T1195] ? __lock_acquire+0xdfe/0x3de0\n[ 79.675825][ T1195] ? memset+0x1f/0x40\n[ 79.676367][ T1195] ? __nla_validate_parse+0x98/0x1ab0\n[ 79.677290][ T1195] unregister_netdevice_many.part.133+0x13/0x1b0\n[ 79.678163][ T1195] rtnl_delete_link+0xbc/0x100\n[ ... ]\n\nFixes: ceed73a2cf4a (\"drivers: net: ethernet: qualcomm: rmnet: Initial implementation\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "102210f7664442d8c0ce332c006ea90626df745b", "tree": "f8d48c4c67fed981257fc2dcdf976e38425ad3d0", "parents": [ "1eb1f43a6e37282348a41e3d68f5e9a6a4359212" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:24:45 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: fix suspicious RCU usage\n\nrmnet_get_port() internally calls rcu_dereference_rtnl(),\nwhich checks RTNL.\nBut rmnet_get_port() could be called by packet path.\nThe packet path is not protected by RTNL.\nSo, the suspicious RCU usage problem occurs.\n\nTest commands:\n modprobe rmnet\n ip netns add nst\n ip link add veth0 type veth peer name veth1\n ip link set veth1 netns nst\n ip link add rmnet0 link veth0 type rmnet mux_id 1\n ip netns exec nst ip link add rmnet1 link veth1 type rmnet mux_id 1\n ip netns exec nst ip link set veth1 up\n ip netns exec nst ip link set rmnet1 up\n ip netns exec nst ip a a 192.168.100.2/24 dev rmnet1\n ip link set veth0 up\n ip link set rmnet0 up\n ip a a 192.168.100.1/24 dev rmnet0\n ping 192.168.100.2\n\nSplat looks like:\n[ 146.630958][ T1174] WARNING: suspicious RCU usage\n[ 146.631735][ T1174] 5.6.0-rc1+ #447 Not tainted\n[ 146.632387][ T1174] -----------------------------\n[ 146.633151][ T1174] drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c:386 suspicious rcu_dereference_check() !\n[ 146.634742][ T1174]\n[ 146.634742][ T1174] other info that might help us debug this:\n[ 146.634742][ T1174]\n[ 146.645992][ T1174]\n[ 146.645992][ T1174] rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n[ 146.646937][ T1174] 5 locks held by ping/1174:\n[ 146.647609][ T1174] #0: ffff8880c31dea70 (sk_lock-AF_INET){+.+.}, at: raw_sendmsg+0xab8/0x2980\n[ 146.662463][ T1174] #1: ffffffff93925660 (rcu_read_lock_bh){....}, at: ip_finish_output2+0x243/0x2150\n[ 146.671696][ T1174] #2: ffffffff93925660 (rcu_read_lock_bh){....}, at: __dev_queue_xmit+0x213/0x2940\n[ 146.673064][ T1174] #3: ffff8880c19ecd58 (\u0026dev-\u003eqdisc_running_key#7){+...}, at: ip_finish_output2+0x714/0x2150\n[ 146.690358][ T1174] #4: ffff8880c5796898 (\u0026dev-\u003eqdisc_xmit_lock_key#3){+.-.}, at: sch_direct_xmit+0x1e2/0x1020\n[ 146.699875][ T1174]\n[ 146.699875][ T1174] stack backtrace:\n[ 146.701091][ T1174] CPU: 0 PID: 1174 Comm: ping Not tainted 5.6.0-rc1+ #447\n[ 146.705215][ T1174] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 146.706565][ T1174] Call Trace:\n[ 146.707102][ T1174] dump_stack+0x96/0xdb\n[ 146.708007][ T1174] rmnet_get_port.part.9+0x76/0x80 [rmnet]\n[ 146.709233][ T1174] rmnet_egress_handler+0x107/0x420 [rmnet]\n[ 146.710492][ T1174] ? sch_direct_xmit+0x1e2/0x1020\n[ 146.716193][ T1174] rmnet_vnd_start_xmit+0x3d/0xa0 [rmnet]\n[ 146.717012][ T1174] dev_hard_start_xmit+0x160/0x740\n[ 146.717854][ T1174] sch_direct_xmit+0x265/0x1020\n[ 146.718577][ T1174] ? register_lock_class+0x14d0/0x14d0\n[ 146.719429][ T1174] ? dev_watchdog+0xac0/0xac0\n[ 146.723738][ T1174] ? __dev_queue_xmit+0x15fd/0x2940\n[ 146.724469][ T1174] ? lock_acquire+0x164/0x3b0\n[ 146.725172][ T1174] __dev_queue_xmit+0x20c7/0x2940\n[ ... ]\n\nFixes: ceed73a2cf4a (\"drivers: net: ethernet: qualcomm: rmnet: Initial implementation\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1eb1f43a6e37282348a41e3d68f5e9a6a4359212", "tree": "a3f5b137d02ef02cf11c0c39b222605d962fc448", "parents": [ "93b5cbfa9636d385126f211dca9efa7e3f683202" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:24:26 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: fix NULL pointer dereference in rmnet_changelink()\n\nIn the rmnet_changelink(), it uses IFLA_LINK without checking\nNULL pointer.\ntb[IFLA_LINK] could be NULL pointer.\nSo, NULL-ptr-deref could occur.\n\nrmnet already has a lower interface (real_dev).\nSo, after this patch, rmnet_changelink() does not use IFLA_LINK anymore.\n\nTest commands:\n modprobe rmnet\n ip link add dummy0 type dummy\n ip link add rmnet0 link dummy0 type rmnet mux_id 1\n ip link set rmnet0 type rmnet mux_id 2\n\nSplat looks like:\n[ 90.578726][ T1131] general protection fault, probably for non-canonical address 0xdffffc0000000000I\n[ 90.581121][ T1131] KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]\n[ 90.582380][ T1131] CPU: 2 PID: 1131 Comm: ip Not tainted 5.6.0-rc1+ #447\n[ 90.584285][ T1131] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 90.587506][ T1131] RIP: 0010:rmnet_changelink+0x5a/0x8a0 [rmnet]\n[ 90.588546][ T1131] Code: 83 ec 20 48 c1 ea 03 80 3c 02 00 0f 85 6f 07 00 00 48 8b 5e 28 48 b8 00 00 00 00 00 0\n[ 90.591447][ T1131] RSP: 0018:ffff8880ce78f1b8 EFLAGS: 00010247\n[ 90.592329][ T1131] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffff8880ce78f8b0\n[ 90.593253][ T1131] RDX: 0000000000000000 RSI: ffff8880ce78f4a0 RDI: 0000000000000004\n[ 90.594058][ T1131] RBP: ffff8880cf543e00 R08: 0000000000000002 R09: 0000000000000002\n[ 90.594859][ T1131] R10: ffffffffc0586a40 R11: 0000000000000000 R12: ffff8880ca47c000\n[ 90.595690][ T1131] R13: ffff8880ca47c000 R14: ffff8880cf545000 R15: 0000000000000000\n[ 90.596553][ T1131] FS: 00007f21f6c7e0c0(0000) GS:ffff8880da400000(0000) knlGS:0000000000000000\n[ 90.597504][ T1131] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[ 90.599418][ T1131] CR2: 0000556e413db458 CR3: 00000000c917a002 CR4: 00000000000606e0\n[ 90.600289][ T1131] Call Trace:\n[ 90.600631][ T1131] __rtnl_newlink+0x922/0x1270\n[ 90.601194][ T1131] ? lock_downgrade+0x6e0/0x6e0\n[ 90.601724][ T1131] ? rtnl_link_unregister+0x220/0x220\n[ 90.602309][ T1131] ? lock_acquire+0x164/0x3b0\n[ 90.602784][ T1131] ? is_bpf_image_address+0xff/0x1d0\n[ 90.603331][ T1131] ? rtnl_newlink+0x4c/0x90\n[ 90.603810][ T1131] ? kernel_text_address+0x111/0x140\n[ 90.604419][ T1131] ? __kernel_text_address+0xe/0x30\n[ 90.604981][ T1131] ? unwind_get_return_address+0x5f/0xa0\n[ 90.605616][ T1131] ? create_prof_cpu_mask+0x20/0x20\n[ 90.606304][ T1131] ? arch_stack_walk+0x83/0xb0\n[ 90.606985][ T1131] ? stack_trace_save+0x82/0xb0\n[ 90.607656][ T1131] ? stack_trace_consume_entry+0x160/0x160\n[ 90.608503][ T1131] ? deactivate_slab.isra.78+0x2c5/0x800\n[ 90.609336][ T1131] ? kasan_unpoison_shadow+0x30/0x40\n[ 90.610096][ T1131] ? kmem_cache_alloc_trace+0x135/0x350\n[ 90.610889][ T1131] ? rtnl_newlink+0x4c/0x90\n[ 90.611512][ T1131] rtnl_newlink+0x65/0x90\n[ ... ]\n\nFixes: 23790ef12082 (\"net: qualcomm: rmnet: Allow to configure flags for existing devices\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "93b5cbfa9636d385126f211dca9efa7e3f683202", "tree": "abf49eff22a8438a361f24b0a4af911cf462f678", "parents": [ "b82cf17ff1957ec35eaee7dc519c365ecd06ba38" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Thu Feb 27 12:23:52 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:45:07 2020 -0800" }, "message": "net: rmnet: fix NULL pointer dereference in rmnet_newlink()\n\nrmnet registers IFLA_LINK interface as a lower interface.\nBut, IFLA_LINK could be NULL.\nIn the current code, rmnet doesn\u0027t check IFLA_LINK.\nSo, panic would occur.\n\nTest commands:\n modprobe rmnet\n ip link add rmnet0 type rmnet mux_id 1\n\nSplat looks like:\n[ 36.826109][ T1115] general protection fault, probably for non-canonical address 0xdffffc0000000000I\n[ 36.838817][ T1115] KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]\n[ 36.839908][ T1115] CPU: 1 PID: 1115 Comm: ip Not tainted 5.6.0-rc1+ #447\n[ 36.840569][ T1115] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006\n[ 36.841408][ T1115] RIP: 0010:rmnet_newlink+0x54/0x510 [rmnet]\n[ 36.841986][ T1115] Code: 83 ec 18 48 c1 e9 03 80 3c 01 00 0f 85 d4 03 00 00 48 8b 6a 28 48 b8 00 00 00 00 00 c\n[ 36.843923][ T1115] RSP: 0018:ffff8880b7e0f1c0 EFLAGS: 00010247\n[ 36.844756][ T1115] RAX: dffffc0000000000 RBX: ffff8880d14cca00 RCX: 1ffff11016fc1e99\n[ 36.845859][ T1115] RDX: 0000000000000000 RSI: ffff8880c3d04000 RDI: 0000000000000004\n[ 36.846961][ T1115] RBP: 0000000000000000 R08: ffff8880b7e0f8b0 R09: ffff8880b6ac2d90\n[ 36.848020][ T1115] R10: ffffffffc0589a40 R11: ffffed1016d585b7 R12: ffffffff88ceaf80\n[ 36.848788][ T1115] R13: ffff8880c3d04000 R14: ffff8880b7e0f8b0 R15: ffff8880c3d04000\n[ 36.849546][ T1115] FS: 00007f50ab3360c0(0000) GS:ffff8880da000000(0000) knlGS:0000000000000000\n[ 36.851784][ T1115] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n[ 36.852422][ T1115] CR2: 000055871afe5ab0 CR3: 00000000ae246001 CR4: 00000000000606e0\n[ 36.853181][ T1115] Call Trace:\n[ 36.853514][ T1115] __rtnl_newlink+0xbdb/0x1270\n[ 36.853967][ T1115] ? lock_downgrade+0x6e0/0x6e0\n[ 36.854420][ T1115] ? rtnl_link_unregister+0x220/0x220\n[ 36.854936][ T1115] ? lock_acquire+0x164/0x3b0\n[ 36.855376][ T1115] ? is_bpf_image_address+0xff/0x1d0\n[ 36.855884][ T1115] ? rtnl_newlink+0x4c/0x90\n[ 36.856304][ T1115] ? kernel_text_address+0x111/0x140\n[ 36.856857][ T1115] ? __kernel_text_address+0xe/0x30\n[ 36.857440][ T1115] ? unwind_get_return_address+0x5f/0xa0\n[ 36.858063][ T1115] ? create_prof_cpu_mask+0x20/0x20\n[ 36.858644][ T1115] ? arch_stack_walk+0x83/0xb0\n[ 36.859171][ T1115] ? stack_trace_save+0x82/0xb0\n[ 36.859710][ T1115] ? stack_trace_consume_entry+0x160/0x160\n[ 36.860357][ T1115] ? deactivate_slab.isra.78+0x2c5/0x800\n[ 36.860928][ T1115] ? kasan_unpoison_shadow+0x30/0x40\n[ 36.861520][ T1115] ? kmem_cache_alloc_trace+0x135/0x350\n[ 36.862125][ T1115] ? rtnl_newlink+0x4c/0x90\n[ 36.864073][ T1115] rtnl_newlink+0x65/0x90\n[ ... ]\n\nFixes: ceed73a2cf4a (\"drivers: net: ethernet: qualcomm: rmnet: Initial implementation\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a2f0b878c3ca531a1706cb2a8b079cea3b17bafc", "tree": "30b41a1a34946ac4a87da7dd6ebaaa4475c989d1", "parents": [ "278de45e145bedff90072fb47bafdde8393f6a80", "eabc8bcb292fb9a5757b0c8ab7751f41b0a104f8" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:26:33 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:26:33 2020 -0800" }, "message": "Merge tag \u0027kbuild-fixes-v5.6-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild\n\nPull Kbuild fixes from Masahiro Yamada:\n\n - fix missed rebuild of DT schema check\n\n - add some phony targets to PHONY\n\n - fix comments and documents\n\n* tag \u0027kbuild-fixes-v5.6-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:\n kbuild: get rid of trailing slash from subdir- example\n kbuild: add dt_binding_check to PHONY in a correct place\n kbuild: add dtbs_check to PHONY\n kbuild: remove unneeded semicolon at the end of cmd_dtb_check\n kbuild: fix DT binding schema rule to detect command line changes\n kbuild: remove wrong documentation about mandatory-y\n kbuild: add comment for V\u003d2 mode\n" }, { "commit": "b82cf17ff1957ec35eaee7dc519c365ecd06ba38", "tree": "2760bc9d209720e387df4bb60f45f48c8b26d704", "parents": [ "e387f7d5fccf95299135c88b799184c3bef6a705" ], "author": { "name": "Russell King", "email": "rmk+kernel@armlinux.org.uk", "time": "Thu Feb 27 09:44:49 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:17:21 2020 -0800" }, "message": "net: phy: marvell: don\u0027t interpret PHY status unless resolved\n\nDon\u0027t attempt to interpret the PHY specific status register unless\nthe PHY is indicating that the resolution is valid.\n\nReviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\nReviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e387f7d5fccf95299135c88b799184c3bef6a705", "tree": "3075f2442b89022be55f97c2e489beba3a05cf59", "parents": [ "3a12500ed5dd21a63da779ac73503f11085bbc1c" ], "author": { "name": "Jiri Pirko", "email": "jiri@mellanox.com", "time": "Thu Feb 27 08:22:10 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:16:14 2020 -0800" }, "message": "mlx5: register lag notifier for init network namespace only\n\nThe current code causes problems when the unregistering netdevice could\nbe different then the registering one.\n\nSince the check in mlx5_lag_netdev_event() does not allow any other\nnetwork namespace anyway, fix this by registerting the lag notifier\nper init network namespace only.\n\nFixes: d48834f9d4b4 (\"mlx5: Use dev_net netdevice notifier registrations\")\nSigned-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nTested-by: Aya Levin \u003cayal@mellanox.com\u003e\nAcked-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "278de45e145bedff90072fb47bafdde8393f6a80", "tree": "44b17e79b1aab1e53ef83b78cd0656e57c1ac4ef", "parents": [ "e46bfaba593c36de591a1153746af6bcb40ef67c", "4eb1b01de5b9d8596d6c103efcf1a15cfc1bedf7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:13:27 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:13:27 2020 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid\n\nPull HID subsystem fixes from Jiri Kosina:\n\n - syzkaller-reported error handling fixes in various drivers, from\n various people\n\n - increase of HID report buffer size to 8K, which is apparently needed\n by certain modern devices\n\n - a few new device-ID-specific fixes / quirks\n\n - battery charging status reporting fix in logitech-hidpp, from Filipe\n Laíns\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:\n HID: hid-bigbenff: fix race condition for scheduled work during removal\n HID: hid-bigbenff: call hid_hw_stop() in case of error\n HID: hid-bigbenff: fix general protection fault caused by double kfree\n HID: i2c-hid: add Trekstor Surfbook E11B to descriptor override\n HID: alps: Fix an error handling path in \u0027alps_input_configured()\u0027\n HID: hiddev: Fix race in in hiddev_disconnect()\n HID: core: increase HID report buffer size to 8KiB\n HID: core: fix off-by-one memset in hid_report_raw_event()\n HID: apple: Add support for recent firmware on Magic Keyboards\n HID: ite: Only bind to keyboard USB interface on Acer SW5-012 keyboard dock\n HID: logitech-hidpp: BatteryVoltage: only read chargeStatus if extPower is active\n" }, { "commit": "3a12500ed5dd21a63da779ac73503f11085bbc1c", "tree": "f762bf883ec807bc278655c263e3a3e372224779", "parents": [ "f4979b41f374c9a5c562df6e428e82629f882df0" ], "author": { "name": "Tobias Klauser", "email": "tklauser@distanz.ch", "time": "Wed Feb 26 18:29:53 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:11:57 2020 -0800" }, "message": "unix: define and set show_fdinfo only if procfs is enabled\n\nFollow the pattern used with other *_show_fdinfo functions and only\ndefine unix_show_fdinfo and set it in proto_ops if CONFIG_PROCFS\nis set.\n\nFixes: 3c32da19a858 (\"unix: Show number of pending scm files of receive queue in fdinfo\")\nSigned-off-by: Tobias Klauser \u003ctklauser@distanz.ch\u003e\nReviewed-by: Kirill Tkhai \u003cktkhai@virtuozzo.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f4979b41f374c9a5c562df6e428e82629f882df0", "tree": "d9f82ce7ceb189dca1d10c9ae32bae5ddc539e58", "parents": [ "a2f2ef4a54c0d97aa6a8386f4ff23f36ebb488cf", "386d4716fd91869e07c731657f2cde5a33086516" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:08:01 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:08:01 2020 -0800" }, "message": "Merge branch \u0027hinic-BugFixes\u0027\n\nLuo bin says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nhinic: BugFixes\n\nthe bug fixed in patch #2 has been present since the first commit.\nthe bugs fixed in patch #1 and patch #3 have been present since the\nfollowing commits:\npatch #1: 352f58b0d9f2 (\"net-next/hinic: Set Rxq irq to specific cpu for NUMA\")\npatch #3: 421e9526288b (\"hinic: add rss support\")\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "386d4716fd91869e07c731657f2cde5a33086516", "tree": "d9f82ce7ceb189dca1d10c9ae32bae5ddc539e58", "parents": [ "d2ed69ce9ed3477e2a9527e6b89fe4689d99510e" ], "author": { "name": "Luo bin", "email": "luobin9@huawei.com", "time": "Thu Feb 27 06:34:44 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:08:01 2020 -0800" }, "message": "hinic: fix a bug of rss configuration\n\nshould use real receive queue number to configure hw rss\nindirect table rather than maximal queue number\n\nSigned-off-by: Luo bin \u003cluobin9@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d2ed69ce9ed3477e2a9527e6b89fe4689d99510e", "tree": "28cd11fbc579908299298e05b3e9fae13dd09b54", "parents": [ "0bff777bd0cba73ad4cd0145696ad284d7e6a99f" ], "author": { "name": "Luo bin", "email": "luobin9@huawei.com", "time": "Thu Feb 27 06:34:43 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:08:01 2020 -0800" }, "message": "hinic: fix a bug of setting hw_ioctxt\n\na reserved field is used to signify prime physical function index\nin the latest firmware version, so we must assign a value to it\ncorrectly\n\nSigned-off-by: Luo bin \u003cluobin9@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0bff777bd0cba73ad4cd0145696ad284d7e6a99f", "tree": "98de438b4e3d5e792e98f3d5b05e79a3b042e080", "parents": [ "a2f2ef4a54c0d97aa6a8386f4ff23f36ebb488cf" ], "author": { "name": "Luo bin", "email": "luobin9@huawei.com", "time": "Thu Feb 27 06:34:42 2020 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu Feb 27 11:08:01 2020 -0800" }, "message": "hinic: fix a irq affinity bug\n\ncan not use a local variable as an input parameter of\nirq_set_affinity_hint\n\nSigned-off-by: Luo bin \u003cluobin9@huawei.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e46bfaba593c36de591a1153746af6bcb40ef67c", "tree": "11b252d12ab86c0b2559449504da2b4c8b6ce3eb", "parents": [ "ed5fa5591866f95be1fe75cd267cf9df2c0390f5", "adc10f5b0a03606e30c704cff1f0283a696d0260" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:07:13 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:07:13 2020 -0800" }, "message": "Merge tag \u0027docs-5.6-fixes\u0027 of git://git.lwn.net/linux\n\nPull documentation fixes from Jonathan Corbet:\n \"A pair of docs-build fixes\"\n\n* tag \u0027docs-5.6-fixes\u0027 of git://git.lwn.net/linux:\n docs: Fix empty parallelism argument\n docs: remove MPX from the x86 toc\n" }, { "commit": "ed5fa5591866f95be1fe75cd267cf9df2c0390f5", "tree": "b483dc1440127f7617925a062dbed0cef143cbcf", "parents": [ "bfdc6d91a25f4545bcd1b12e3219af4838142ef1", "756125289285f6e55a03861bf4b6257aa3d19a93" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:01:22 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Feb 27 11:01:22 2020 -0800" }, "message": "Merge tag \u0027audit-pr-20200226\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit\n\nPull audit fixes from Paul Moore:\n \"Two fixes for problems found by syzbot:\n\n - Moving audit filter structure fields into a union caused some\n problems in the code which populates that filter structure.\n\n We keep the union (that idea is a good one), but we are fixing the\n code so that it doesn\u0027t needlessly set fields in the union and mess\n up the error handling.\n\n - The audit_receive_msg() function wasn\u0027t validating user input as\n well as it should in all cases, we add the necessary checks\"\n\n* tag \u0027audit-pr-20200226\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit:\n audit: always check the netlink payload length in audit_receive_msg()\n audit: fix error handling in audit_data_to_entry()\n" }, { "commit": "9515743bfb39c61aaf3d4f3219a645c8d1fe9a0e", "tree": "abe811e8684f25b41397639a2c41b54466bbc8a1", "parents": [ "cae740a04b4d6d5166f19ee5faf04ea2a1f34b3d" ], "author": { "name": "Bijan Mottahedeh", "email": "bijan.mottahedeh@oracle.com", "time": "Wed Feb 26 18:53:43 2020 -0800" }, "committer": { "name": "Keith Busch", "email": "kbusch@kernel.org", "time": "Fri Feb 28 01:32:14 2020 +0900" }, "message": "nvme-pci: Hold cq_poll_lock while completing CQEs\n\nCompletions need to consumed in the same order the controller submitted\nthem, otherwise future completion entries may overwrite ones we haven\u0027t\nhandled yet. Hold the nvme queue\u0027s poll lock while completing new CQEs to\nprevent another thread from freeing command tags for reuse out-of-order.\n\nFixes: dabcefab45d3 (\"nvme: provide optimized poll function for separate poll queues\")\nSigned-off-by: Bijan Mottahedeh \u003cbijan.mottahedeh@oracle.com\u003e\nReviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\nReviewed-by: Jens Axboe \u003caxboe@kernel.dk\u003e\nSigned-off-by: Keith Busch \u003ckbusch@kernel.org\u003e\n" }, { "commit": "73a7a271b3eee7b83f29b13866163776f1cbef89", "tree": "1ce72a0ec1afce3dbb5d805f28ab657a08889398", "parents": [ "bb6d3fb354c5ee8d6bde2d576eb7220ea09862b9" ], "author": { "name": "Marek Szyprowski", "email": "m.szyprowski@samsung.com", "time": "Thu Feb 27 12:51:46 2020 +0100" }, "committer": { "name": "Bjorn Helgaas", "email": "bhelgaas@google.com", "time": "Thu Feb 27 08:06:20 2020 -0600" }, "message": "PCI: brcmstb: Fix build on 32bit ARM platforms with older compilers\n\nSome older compilers have no implementation for the helper for 64-bit\nunsigned division/modulo, so linking pcie-brcmstb driver causes the\n\"undefined reference to `__aeabi_uldivmod\u0027\" error.\n\n*rc_bar2_size is always a power of two, because it is calculated as:\n\"1ULL \u003c\u003c fls64(entry-\u003eres-\u003eend - entry-\u003eres-\u003estart)\", so the modulo\noperation in the subsequent check can be replaced by a simple logical\nAND with a proper mask.\n\nLink: https://lore.kernel.org/r/20200227115146.24515-1-m.szyprowski@samsung.com\nFixes: c0452137034b (\"PCI: brcmstb: Add Broadcom STB PCIe host controller driver\")\nSigned-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nAcked-by: Nicolas Saenz Julienne \u003cnsaenzjulienne@suse.de\u003e\nAcked-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n" }, { "commit": "bebdb65e077267957f48e43d205d4a16cc7b8161", "tree": "4b6b3fdfa7411b8ba3e2ca57a920e64c7ea2e9f7", "parents": [ "dd3db2a34cff14e152da7c8e320297719a35abf9" ], "author": { "name": "Tobias Klauser", "email": "tklauser@distanz.ch", "time": "Wed Feb 26 18:38:32 2020 +0100" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Thu Feb 27 06:56:21 2020 -0700" }, "message": "io_uring: define and set show_fdinfo only if procfs is enabled\n\nFollow the pattern used with other *_show_fdinfo functions and only\ndefine and use io_uring_show_fdinfo and its helper functions if\nCONFIG_PROC_FS is set.\n\nFixes: 87ce955b24c9 (\"io_uring: add -\u003eshow_fdinfo() for the io_uring file descriptor\")\nSigned-off-by: Tobias Klauser \u003ctklauser@distanz.ch\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "1df97a02a937abf8473ccf0ef12edefcb11671de", "tree": "d8e76c0c64f92487d7fcb54da2fb08fcb1bc8a11", "parents": [ "f8788d86ab28f61f7b46eb6be375f8a726783636", "66d0e797bf095d407479c89952d42b1d96ef0a7f" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Feb 27 11:21:23 2020 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Feb 27 11:21:23 2020 +0100" }, "message": "Merge tag \u0027devfreq-fixes-for-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux\n\nPull a devfreq fix for 5.6-rc4 from Chanwoo Choi:\n\n\"Revert \"PM / devfreq: Modify the device name as devfreq(X) for sysfs\"\n - This changes as devfreq(X) cause break some user space applications\n such as Android HAL from Unisoc and Hikey. In result, decide to revert it\n for preventing the HAL layer problem.\"\n\n* tag \u0027devfreq-fixes-for-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux:\n Revert \"PM / devfreq: Modify the device name as devfreq(X) for sysfs\"\n" }, { "commit": "f5739cb0b56590d68d8df8a44659893b6d0084c3", "tree": "f7aff4b0f4f5171fbe1a69b625f7051b551907ee", "parents": [ "f8788d86ab28f61f7b46eb6be375f8a726783636" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Feb 26 22:39:27 2020 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Thu Feb 27 08:57:48 2020 +0100" }, "message": "cpufreq: Fix policy initialization for internal governor drivers\n\nBefore commit 1e4f63aecb53 (\"cpufreq: Avoid creating excessively\nlarge stack frames\") the initial value of the policy field in struct\ncpufreq_policy set by the driver\u0027s -\u003einit() callback was implicitly\npassed from cpufreq_init_policy() to cpufreq_set_policy() if the\ndefault governor was neither \"performance\" nor \"powersave\". After\nthat commit, however, cpufreq_init_policy() must take that case into\nconsideration explicitly and handle it as appropriate, so make that\nhappen.\n\nFixes: 1e4f63aecb53 (\"cpufreq: Avoid creating excessively large stack frames\")\nLink: https://lore.kernel.org/linux-pm/39fb762880c27da110086741315ca8b111d781cd.camel@gmail.com/\nReported-by: Artem Bityutskiy \u003cdedekind1@gmail.com\u003e\nCc: 5.4+ \u003cstable@vger.kernel.org\u003e # 5.4+\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nAcked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n" }, { "commit": "a2f2ef4a54c0d97aa6a8386f4ff23f36ebb488cf", "tree": "ec405d8d9f4372a0fbda16a353967123a93bd9e1", "parents": [ "474a31e13a4e9749fb3ee55794d69d0f17ee0998" ], "author": { "name": "Karsten Graul", "email": "kgraul@linux.ibm.com", "time": "Wed Feb 26 17:52:46 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:56:25 2020 -0800" }, "message": "net/smc: check for valid ib_client_data\n\nIn smc_ib_remove_dev() check if the provided ib device was actually\ninitialized for SMC before.\n\nReported-by: syzbot+84484ccebdd4e5451d91@syzkaller.appspotmail.com\nFixes: a4cf0443c414 (\"smc: introduce SMC as an IB-client\")\nSigned-off-by: Karsten Graul \u003ckgraul@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "474a31e13a4e9749fb3ee55794d69d0f17ee0998", "tree": "6b114e11a9ed3ed7935ecb249122743d20d65120", "parents": [ "c87a9d6fc6d555e4981f2ded77d9a8cce950743e" ], "author": { "name": "Aaro Koskinen", "email": "aaro.koskinen@nokia.com", "time": "Wed Feb 26 18:49:01 2020 +0200" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:55:14 2020 -0800" }, "message": "net: stmmac: fix notifier registration\n\nWe cannot register the same netdev notifier multiple times when probing\nstmmac devices. Register the notifier only once in module init, and also\nmake debugfs creation/deletion safe against simultaneous notifier call.\n\nFixes: 481a7d154cbb (\"stmmac: debugfs entry name is not be changed when udev rename device name.\")\nSigned-off-by: Aaro Koskinen \u003caaro.koskinen@nokia.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c87a9d6fc6d555e4981f2ded77d9a8cce950743e", "tree": "df88b071c480b67cb5f8c94203aaa6866717b870", "parents": [ "dc24f8b4ecd3d6c4153a1ec1bc2006ab32a41b8d" ], "author": { "name": "Antoine Tenart", "email": "antoine.tenart@bootlin.com", "time": "Wed Feb 26 16:26:50 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:53:09 2020 -0800" }, "message": "net: phy: mscc: fix firmware paths\n\nThe firmware paths for the VSC8584 PHYs not not contain the leading\n\u0027microchip/\u0027 directory, as used in linux-firmware, resulting in an\nerror when probing the driver. This patch fixes it.\n\nFixes: a5afc1678044 (\"net: phy: mscc: add support for VSC8584 PHY\")\nSigned-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nReviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "dc24f8b4ecd3d6c4153a1ec1bc2006ab32a41b8d", "tree": "a09ac86eeaa642f87dff19225f42ff5207f68bb9", "parents": [ "4f31c532ad400a34dbdd836c204ed964d1ec2da5" ], "author": { "name": "Paolo Abeni", "email": "pabeni@redhat.com", "time": "Wed Feb 26 12:19:03 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:49:50 2020 -0800" }, "message": "mptcp: add dummy icsk_sync_mss()\n\nsyzbot noted that the master MPTCP socket lacks the icsk_sync_mss\ncallback, and was able to trigger a null pointer dereference:\n\nBUG: kernel NULL pointer dereference, address: 0000000000000000\nPGD 8e171067 P4D 8e171067 PUD 93fa2067 PMD 0\nOops: 0010 [#1] PREEMPT SMP KASAN\nCPU: 0 PID: 8984 Comm: syz-executor066 Not tainted 5.6.0-rc2-syzkaller #0\nHardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011\nRIP: 0010:0x0\nCode: Bad RIP value.\nRSP: 0018:ffffc900020b7b80 EFLAGS: 00010246\nRAX: 1ffff110124ba600 RBX: 0000000000000000 RCX: ffff88809fefa600\nRDX: ffff8880994cdb18 RSI: 0000000000000000 RDI: ffff8880925d3140\nRBP: ffffc900020b7bd8 R08: ffffffff870225be R09: fffffbfff140652a\nR10: fffffbfff140652a R11: 0000000000000000 R12: ffff8880925d35d0\nR13: ffff8880925d3140 R14: dffffc0000000000 R15: 1ffff110124ba6ba\nFS: 0000000001a0b880(0000) GS:ffff8880aea00000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: ffffffffffffffd6 CR3: 00000000a6d6f000 CR4: 00000000001406f0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\nCall Trace:\n cipso_v4_sock_setattr+0x34b/0x470 net/ipv4/cipso_ipv4.c:1888\n netlbl_sock_setattr+0x2a7/0x310 net/netlabel/netlabel_kapi.c:989\n smack_netlabel security/smack/smack_lsm.c:2425 [inline]\n smack_inode_setsecurity+0x3da/0x4a0 security/smack/smack_lsm.c:2716\n security_inode_setsecurity+0xb2/0x140 security/security.c:1364\n __vfs_setxattr_noperm+0x16f/0x3e0 fs/xattr.c:197\n vfs_setxattr fs/xattr.c:224 [inline]\n setxattr+0x335/0x430 fs/xattr.c:451\n __do_sys_fsetxattr fs/xattr.c:506 [inline]\n __se_sys_fsetxattr+0x130/0x1b0 fs/xattr.c:495\n __x64_sys_fsetxattr+0xbf/0xd0 fs/xattr.c:495\n do_syscall_64+0xf7/0x1c0 arch/x86/entry/common.c:294\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\nRIP: 0033:0x440199\nCode: 18 89 d0 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01 f0 ff ff 0f 83 fb 13 fc ff c3 66 2e 0f 1f 84 00 00 00 00\nRSP: 002b:00007ffcadc19e48 EFLAGS: 00000246 ORIG_RAX: 00000000000000be\nRAX: ffffffffffffffda RBX: 00000000004002c8 RCX: 0000000000440199\nRDX: 0000000020000200 RSI: 00000000200001c0 RDI: 0000000000000003\nRBP: 00000000006ca018 R08: 0000000000000003 R09: 00000000004002c8\nR10: 0000000000000009 R11: 0000000000000246 R12: 0000000000401a20\nR13: 0000000000401ab0 R14: 0000000000000000 R15: 0000000000000000\nModules linked in:\nCR2: 0000000000000000\n\nAddress the issue adding a dummy icsk_sync_mss callback.\nTo properly sync the subflows mss and options list we need some\nadditional infrastructure, which will land to net-next.\n\nReported-by: syzbot+f4dfece964792d80b139@syzkaller.appspotmail.com\nFixes: 2303f994b3e1 (\"mptcp: Associate MPTCP context with TCP socket\")\nSigned-off-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nReviewed-by: Mat Martineau \u003cmathew.j.martineau@linux.intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4f31c532ad400a34dbdd836c204ed964d1ec2da5", "tree": "27c10b7532610b657ff1ea9ba58a3c37972e9471", "parents": [ "f596c87005f7b1baeb7d62d9a9e25d68c3dfae10" ], "author": { "name": "Sudheesh Mavila", "email": "sudheesh.mavila@amd.com", "time": "Wed Feb 26 12:40:45 2020 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:41:42 2020 -0800" }, "message": "net: phy: corrected the return value for genphy_check_and_restart_aneg and genphy_c45_check_and_restart_aneg\n\nWhen auto-negotiation is not required, return value should be zero.\n\nChanges v1-\u003ev2:\n- improved comments and code as Andrew Lunn and Heiner Kallweit suggestion\n- fixed issue in genphy_c45_check_and_restart_aneg as Russell King\n suggestion.\n\nFixes: 2a10ab043ac5 (\"net: phy: add genphy_check_and_restart_aneg()\")\nFixes: 1af9f16840e9 (\"net: phy: add genphy_c45_check_and_restart_aneg()\")\nSigned-off-by: Sudheesh Mavila \u003csudheesh.mavila@amd.com\u003e\nReviewed-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nReviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f596c87005f7b1baeb7d62d9a9e25d68c3dfae10", "tree": "7fbf1945f6e80678ad03f8f2c59b06b92224e699", "parents": [ "b6f6118901d1e867ac9177bbff3b00b185bd4fdc" ], "author": { "name": "yangerkun", "email": "yangerkun@huawei.com", "time": "Wed Feb 26 11:54:35 2020 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:37:06 2020 -0800" }, "message": "slip: not call free_netdev before rtnl_unlock in slip_open\n\nAs the description before netdev_run_todo, we cannot call free_netdev\nbefore rtnl_unlock, fix it by reorder the code.\n\nSigned-off-by: yangerkun \u003cyangerkun@huawei.com\u003e\nReviewed-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "b6f6118901d1e867ac9177bbff3b00b185bd4fdc", "tree": "86ffc1065b3adc5934a1c0b142c0597f2d29912a", "parents": [ "51e3dfa8906ace90c809235b3d3afebc166b6433" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Tue Feb 25 11:52:29 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:20:58 2020 -0800" }, "message": "ipv6: restrict IPV6_ADDRFORM operation\n\nIPV6_ADDRFORM is able to transform IPv6 socket to IPv4 one.\nWhile this operation sounds illogical, we have to support it.\n\nOne of the things it does for TCP socket is to switch sk-\u003esk_prot\nto tcp_prot.\n\nWe now have other layers playing with sk-\u003esk_prot, so we should make\nsure to not interfere with them.\n\nThis patch makes sure sk_prot is the default pointer for TCP IPv6 socket.\n\nsyzbot reported :\nBUG: kernel NULL pointer dereference, address: 0000000000000000\nPGD a0113067 P4D a0113067 PUD a8771067 PMD 0\nOops: 0010 [#1] PREEMPT SMP KASAN\nCPU: 0 PID: 10686 Comm: syz-executor.0 Not tainted 5.6.0-rc2-syzkaller #0\nHardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011\nRIP: 0010:0x0\nCode: Bad RIP value.\nRSP: 0018:ffffc9000281fce0 EFLAGS: 00010246\nRAX: 1ffffffff15f48ac RBX: ffffffff8afa4560 RCX: dffffc0000000000\nRDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8880a69a8f40\nRBP: ffffc9000281fd10 R08: ffffffff86ed9b0c R09: ffffed1014d351f5\nR10: ffffed1014d351f5 R11: 0000000000000000 R12: ffff8880920d3098\nR13: 1ffff1101241a613 R14: ffff8880a69a8f40 R15: 0000000000000000\nFS: 00007f2ae75db700(0000) GS:ffff8880aea00000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: ffffffffffffffd6 CR3: 00000000a3b85000 CR4: 00000000001406f0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\nCall Trace:\n inet_release+0x165/0x1c0 net/ipv4/af_inet.c:427\n __sock_release net/socket.c:605 [inline]\n sock_close+0xe1/0x260 net/socket.c:1283\n __fput+0x2e4/0x740 fs/file_table.c:280\n ____fput+0x15/0x20 fs/file_table.c:313\n task_work_run+0x176/0x1b0 kernel/task_work.c:113\n tracehook_notify_resume include/linux/tracehook.h:188 [inline]\n exit_to_usermode_loop arch/x86/entry/common.c:164 [inline]\n prepare_exit_to_usermode+0x480/0x5b0 arch/x86/entry/common.c:195\n syscall_return_slowpath+0x113/0x4a0 arch/x86/entry/common.c:278\n do_syscall_64+0x11f/0x1c0 arch/x86/entry/common.c:304\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\nRIP: 0033:0x45c429\nCode: ad b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01 f0 ff ff 0f 83 7b b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00\nRSP: 002b:00007f2ae75dac78 EFLAGS: 00000246 ORIG_RAX: 0000000000000036\nRAX: 0000000000000000 RBX: 00007f2ae75db6d4 RCX: 000000000045c429\nRDX: 0000000000000001 RSI: 000000000000011a RDI: 0000000000000004\nRBP: 000000000076bf20 R08: 0000000000000038 R09: 0000000000000000\nR10: 0000000020000180 R11: 0000000000000246 R12: 00000000ffffffff\nR13: 0000000000000a9d R14: 00000000004ccfb4 R15: 000000000076bf2c\nModules linked in:\nCR2: 0000000000000000\n---[ end trace 82567b5207e87bae ]---\nRIP: 0010:0x0\nCode: Bad RIP value.\nRSP: 0018:ffffc9000281fce0 EFLAGS: 00010246\nRAX: 1ffffffff15f48ac RBX: ffffffff8afa4560 RCX: dffffc0000000000\nRDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8880a69a8f40\nRBP: ffffc9000281fd10 R08: ffffffff86ed9b0c R09: ffffed1014d351f5\nR10: ffffed1014d351f5 R11: 0000000000000000 R12: ffff8880920d3098\nR13: 1ffff1101241a613 R14: ffff8880a69a8f40 R15: 0000000000000000\nFS: 00007f2ae75db700(0000) GS:ffff8880aea00000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: ffffffffffffffd6 CR3: 00000000a3b85000 CR4: 00000000001406f0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n\nFixes: 604326b41a6f (\"bpf, sockmap: convert to generic sk_msg interface\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nReported-by: syzbot+1938db17e275e85dc328@syzkaller.appspotmail.com\nCc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "51e3dfa8906ace90c809235b3d3afebc166b6433", "tree": "d5a11fdd383811480c980d78eec9dc47521c1797", "parents": [ "402482a6a78e5c61d8a2ec6311fc5b4aca392cd6" ], "author": { "name": "Ursula Braun", "email": "ubraun@linux.ibm.com", "time": "Tue Feb 25 16:34:36 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 20:18:07 2020 -0800" }, "message": "net/smc: fix cleanup for linkgroup setup failures\n\nIf an SMC connection to a certain peer is setup the first time,\na new linkgroup is created. In case of setup failures, such a\nlinkgroup is unusable and should disappear. As a first step the\nlinkgroup is removed from the linkgroup list in smc_lgr_forget().\n\nThere are 2 problems:\nsmc_listen_decline() might be called before linkgroup creation\nresulting in a crash due to calling smc_lgr_forget() with\nparameter NULL.\nIf a setup failure occurs after linkgroup creation, the connection\nis never unregistered from the linkgroup, preventing linkgroup\nfreeing.\n\nThis patch introduces an enhanced smc_lgr_cleanup_early() function\nwhich\n* contains a linkgroup check for early smc_listen_decline()\n invocations\n* invokes smc_conn_free() to guarantee unregistering of the\n connection.\n* schedules fast linkgroup removal of the unusable linkgroup\n\nAnd the unused function smcd_conn_free() is removed from smc_core.h.\n\nFixes: 3b2dec2603d5b (\"net/smc: restructure client and server code in af_smc\")\nFixes: 2a0674fffb6bc (\"net/smc: improve abnormal termination of link groups\")\nSigned-off-by: Ursula Braun \u003cubraun@linux.ibm.com\u003e\nSigned-off-by: Karsten Graul \u003ckgraul@linux.ibm.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "402482a6a78e5c61d8a2ec6311fc5b4aca392cd6", "tree": "28bc47fbe8fec85bde81e8661fce2e42a17f424c", "parents": [ "1521a67e6016664941f0917d50cb20053a8826a2" ], "author": { "name": "Nicolas Saenz Julienne", "email": "nsaenzjulienne@suse.de", "time": "Tue Feb 25 14:11:59 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 17:12:30 2020 -0800" }, "message": "net: bcmgenet: Clear ID_MODE_DIS in EXT_RGMII_OOB_CTRL when not needed\n\nOutdated Raspberry Pi 4 firmware might configure the external PHY as\nrgmii although the kernel currently sets it as rgmii-rxid. This makes\nconnections unreliable as ID_MODE_DIS is left enabled. To avoid this,\nexplicitly clear that bit whenever we don\u0027t need it.\n\nFixes: da38802211cc (\"net: bcmgenet: Add RGMII_RXID support\")\nSigned-off-by: Nicolas Saenz Julienne \u003cnsaenzjulienne@suse.de\u003e\nAcked-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1521a67e6016664941f0917d50cb20053a8826a2", "tree": "692edd8f0b0997214f5d945870e357438088cdf4", "parents": [ "2eb51c75dcb354f8aef03d7648318b24630632e1" ], "author": { "name": "Jiri Pirko", "email": "jiri@mellanox.com", "time": "Tue Feb 25 13:54:12 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 17:10:44 2020 -0800" }, "message": "sched: act: count in the size of action flags bitfield\n\nThe put of the flags was added by the commit referenced in fixes tag,\nhowever the size of the message was not extended accordingly.\n\nFix this by adding size of the flags bitfield to the message size.\n\nFixes: e38226786022 (\"net: sched: update action implementations to support flags\")\nSigned-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eabc8bcb292fb9a5757b0c8ab7751f41b0a104f8", "tree": "20223c54b76b12fbef5149e5adc125dc8279ff3f", "parents": [ "c473a8d03ea8e03ca9d649b0b6d72fbcf6212c05" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:44:58 2020 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 10:03:27 2020 +0900" }, "message": "kbuild: get rid of trailing slash from subdir- example\n\nobj-* needs a trailing slash for a directory, but subdir-* does not.\n\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\n" }, { "commit": "2eb51c75dcb354f8aef03d7648318b24630632e1", "tree": "a8215309b5994be1239a270d11da6459b2d4b302", "parents": [ "98c5f7d44fef309e692c24c6d71131ee0f0871fb" ], "author": { "name": "Madhuparna Bhowmik", "email": "madhuparnabhowmik10@gmail.com", "time": "Tue Feb 25 17:57:45 2020 +0530" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 16:59:18 2020 -0800" }, "message": "net: core: devlink.c: Use built-in RCU list checking\n\nlist_for_each_entry_rcu() has built-in RCU and lock checking.\n\nPass cond argument to list_for_each_entry_rcu() to silence\nfalse lockdep warning when CONFIG_PROVE_RCU_LIST is enabled.\n\nThe devlink-\u003elock is held when devlink_dpipe_table_find()\nis called in non RCU read side section. Therefore, pass struct devlink\nto devlink_dpipe_table_find() for lockdep checking.\n\nSigned-off-by: Madhuparna Bhowmik \u003cmadhuparnabhowmik10@gmail.com\u003e\nReviewed-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "98c5f7d44fef309e692c24c6d71131ee0f0871fb", "tree": "46565463a1391ba4eb79f266fa4e5cc8c6b3568b", "parents": [ "574b238f64594cc0d87aad3f716ebab49fb663fa" ], "author": { "name": "Florian Fainelli", "email": "f.fainelli@gmail.com", "time": "Mon Feb 24 15:56:32 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 16:38:23 2020 -0800" }, "message": "net: dsa: bcm_sf2: Forcibly configure IMP port for 1Gb/sec\n\nWe are still experiencing some packet loss with the existing advanced\ncongestion buffering (ACB) settings with the IMP port configured for\n2Gb/sec, so revert to conservative link speeds that do not produce\npacket loss until this is resolved.\n\nFixes: 8f1880cbe8d0 (\"net: dsa: bcm_sf2: Configure IMP port for 2Gb/sec\")\nFixes: de34d7084edd (\"net: dsa: bcm_sf2: Only 7278 supports 2Gb/sec IMP port\")\nSigned-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nReviewed-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "574b238f64594cc0d87aad3f716ebab49fb663fa", "tree": "1b3c40577b00ee55e2435315e7230b487f399c09", "parents": [ "9a005c3898aa07cd5cdca77b7096814e6c478c92", "99b79c3900d4627672c85d9f344b5b0f06bc2a4d" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 16:30:17 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 16:30:17 2020 -0800" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf\n\nPablo Neira Ayuso says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nNetfilter fixes for net\n\nThe following patchset contains Netfilter fixes:\n\n1) Perform garbage collection from workqueue to fix rcu detected\n stall in ipset hash set types, from Jozsef Kadlecsik.\n\n2) Fix the forceadd evaluation path, also from Jozsef.\n\n3) Fix nft_set_pipapo selftest, from Stefano Brivio.\n\n4) Crash when add-flush-add element in pipapo set, also from Stefano.\n Add test to cover this crash.\n\n5) Remove sysctl entry under mutex in hashlimit, from Cong Wang.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "bfdc6d91a25f4545bcd1b12e3219af4838142ef1", "tree": "ce77383323b6e86d4dbe7578e687aeda178b3a8b", "parents": [ "91ad64a84e9e63e2906ae714dfa3933dd3f64c64", "0cbb4f9c69827decf56519c2f63918f16904ede5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 15:54:52 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 15:54:52 2020 -0800" }, "message": "Merge tag \u0027tag-chrome-platform-fixes-for-v5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chrome-platform/linux\n\nPull chrome platform fix from Benson Leung:\n \"Fix a build warning\"\n\n* tag \u0027tag-chrome-platform-fixes-for-v5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chrome-platform/linux:\n platform/chrome: wilco_ec: Include asm/unaligned instead of linux/ path\n" }, { "commit": "9a005c3898aa07cd5cdca77b7096814e6c478c92", "tree": "520c16b499bd822b8d3dcd546cd4fd8baea48361", "parents": [ "e34f1753eebc428c312527662eb1b529cf260240" ], "author": { "name": "Jonathan Lemon", "email": "jonathan.lemon@gmail.com", "time": "Mon Feb 24 15:29:09 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 15:52:33 2020 -0800" }, "message": "bnxt_en: add newline to netdev_*() format strings\n\nAdd missing newlines to netdev_* format strings so the lines\naren\u0027t buffered by the printk subsystem.\n\nNitpicked-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nAcked-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "99b79c3900d4627672c85d9f344b5b0f06bc2a4d", "tree": "96bd29993af77183af6ccedc4df190ffc284c391", "parents": [ "0954df70fba743d8cdaa09ccf6ba8e4ad09628de" ], "author": { "name": "Cong Wang", "email": "xiyou.wangcong@gmail.com", "time": "Wed Feb 12 22:53:52 2020 -0800" }, "committer": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Wed Feb 26 23:25:07 2020 +0100" }, "message": "netfilter: xt_hashlimit: unregister proc file before releasing mutex\n\nBefore releasing the global mutex, we only unlink the hashtable\nfrom the hash list, its proc file is still not unregistered at\nthis point. So syzbot could trigger a race condition where a\nparallel htable_create() could register the same file immediately\nafter the mutex is released.\n\nMove htable_remove_proc_entry() back to mutex protection to\nfix this. And, fold htable_destroy() into htable_put() to make\nthe code slightly easier to understand.\n\nReported-and-tested-by: syzbot+d195fd3b9a364ddd6731@syzkaller.appspotmail.com\nFixes: c4a3922d2d20 (\"netfilter: xt_hashlimit: reduce hashlimit_mutex scope for htable_put()\")\nSigned-off-by: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\nSigned-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" }, { "commit": "8e9a400c706ee5bcbc052d3ec9f87cfdbbd3f5cb", "tree": "42af2b553702b69f7d78d3d5de10def05756e929", "parents": [ "238734262142075056653b4de091458e0ca858f2", "b549c252b1292aea959cd9b83537fcb9384a6112" ], "author": { "name": "Jani Nikula", "email": "jani.nikula@intel.com", "time": "Wed Feb 26 22:58:24 2020 +0200" }, "committer": { "name": "Jani Nikula", "email": "jani.nikula@intel.com", "time": "Wed Feb 26 22:58:25 2020 +0200" }, "message": "Merge tag \u0027gvt-fixes-2020-02-26\u0027 of https://github.com/intel/gvt-linux into drm-intel-fixes\n\ngvt-fixes-2020-02-26\n\n- Fix virtual display reset (Tina)\n- Fix one use-after-free for dmabuf (Tina)\n\nSigned-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\nFrom: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\nLink: https://patchwork.freedesktop.org/patch/msgid/20200226103016.GC10413@zhen-hp.sh.intel.com\n" }, { "commit": "e34f1753eebc428c312527662eb1b529cf260240", "tree": "667163607addcc7d8c0ab4af5ab726fdfa71bcf0", "parents": [ "6e11d1578fba8d09d03a286740ffcf336d53928c" ], "author": { "name": "Michal Kubecek", "email": "mkubecek@suse.cz", "time": "Mon Feb 24 20:42:12 2020 +0100" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 11:27:31 2020 -0800" }, "message": "ethtool: limit bitset size\n\nSyzbot reported that ethnl_compact_sanity_checks() can be tricked into\nreading past the end of ETHTOOL_A_BITSET_VALUE and ETHTOOL_A_BITSET_MASK\nattributes and even the message by passing a value between (u32)(-31)\nand (u32)(-1) as ETHTOOL_A_BITSET_SIZE.\n\nThe problem is that DIV_ROUND_UP(attr_nbits, 32) is 0 for such values so\nthat zero length ETHTOOL_A_BITSET_VALUE will pass the length check but\nethnl_bitmap32_not_zero() check would try to access up to 512 MB of\nattribute \"payload\".\n\nPrevent this overflow byt limiting the bitset size. Technically, compact\nbitset format would allow bitset sizes up to almost 2^18 (so that the\nnest size does not exceed U16_MAX) but bitsets used by ethtool are much\nshorter. S16_MAX, the largest value which can be directly used as an\nupper limit in policy, should be a reasonable compromise.\n\nFixes: 10b518d4e6dd (\"ethtool: netlink bitset handling\")\nReported-by: syzbot+7fd4ed5b4234ab1fdccd@syzkaller.appspotmail.com\nReported-by: syzbot+709b7a64d57978247e44@syzkaller.appspotmail.com\nReported-by: syzbot+983cb8fb2d17a7af549d@syzkaller.appspotmail.com\nSigned-off-by: Michal Kubecek \u003cmkubecek@suse.cz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6e11d1578fba8d09d03a286740ffcf336d53928c", "tree": "28dd117c6d920b22221c937220d1b10ce734d2b4", "parents": [ "a8e41f6033a0c5633d55d6e35993c9e2005d872f" ], "author": { "name": "Amritha Nambiar", "email": "amritha.nambiar@intel.com", "time": "Mon Feb 24 10:56:00 2020 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Feb 26 11:14:10 2020 -0800" }, "message": "net: Fix Tx hash bound checking\n\nFixes the lower and upper bounds when there are multiple TCs and\ntraffic is on the the same TC on the same device.\n\nThe lower bound is represented by \u0027qoffset\u0027 and the upper limit for\nhash value is \u0027qcount + qoffset\u0027. This gives a clean Rx to Tx queue\nmapping when there are multiple TCs, as the queue indices for upper TCs\nwill be offset by \u0027qoffset\u0027.\n\nv2: Fixed commit description based on comments.\n\nFixes: 1b837d489e06 (\"net: Revoke export for __skb_tx_hash, update it to just be static skb_tx_hash\")\nFixes: eadec877ce9c (\"net: Add support for subordinate traffic classes to netdev_pick_tx\")\nSigned-off-by: Amritha Nambiar \u003camritha.nambiar@intel.com\u003e\nReviewed-by: Alexander Duyck \u003calexander.h.duyck@linux.intel.com\u003e\nReviewed-by: Sridhar Samudrala \u003csridhar.samudrala@intel.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "eb12c957735b582607e5842a06d1f4c62e185c1d", "tree": "16b2c37a50065c9d8635c81982563522026b5443", "parents": [ "8a3bddf67ce88b96531fb22c5a75d7f4dc41d155" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sat Feb 22 18:54:32 2020 +0100" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Feb 26 14:02:41 2020 -0500" }, "message": "drm/radeon: Inline drm_get_pci_dev\n\nIt\u0027s the last user, and more importantly, it\u0027s the last non-legacy\nuser of anything in drm_pci.c.\n\nThe only tricky bit is the agp initialization. But a close look shows\nthat radeon does not use the drm_agp midlayer (the main use of that is\ndrm_bufs for legacy drivers), and instead could use the agp subsystem\ndirectly (like nouveau does already). Hence we can just pull this in\ntoo.\n\nA further step would be to entirely drop the use of drm_device-\u003eagp,\nbut feels like too much churn just for this patch.\n\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\nCc: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\nCc: \"David (ChunMing) Zhou\" \u003cDavid1.Zhou@amd.com\u003e\nCc: amd-gfx@lists.freedesktop.org\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nReviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "8a3bddf67ce88b96531fb22c5a75d7f4dc41d155", "tree": "4addec5ae855e8057c2f6279d373d23be7e9e2bc", "parents": [ "a3ed353cf8015ba84a0407a5dc3ffee038166ab0" ], "author": { "name": "Daniel Vetter", "email": "daniel.vetter@ffwll.ch", "time": "Sat Feb 22 18:54:31 2020 +0100" }, "committer": { "name": "Alex Deucher", "email": "alexander.deucher@amd.com", "time": "Wed Feb 26 14:02:41 2020 -0500" }, "message": "drm/amdgpu: Drop DRIVER_USE_AGP\n\nThis doesn\u0027t do anything except auto-init drm_agp support when you\ncall drm_get_pci_dev(). Which amdgpu stopped doing with\n\ncommit b58c11314a1706bf094c489ef5cb28f76478c704\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Jun 2 17:16:31 2017 -0400\n\n drm/amdgpu: drop deprecated drm_get_pci_dev and drm_put_dev\n\nNo idea whether this was intentional or accidental breakage, but I\nguess anyone who manages to boot a this modern gpu behind an agp\nbridge deserves a price. A price I never expect anyone to ever collect\n:-)\n\nCc: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\nCc: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nCc: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nCc: Evan Quan \u003cevan.quan@amd.com\u003e\nCc: \"Tianci.Yin\" \u003ctianci.yin@amd.com\u003e\nCc: \"Marek Olšák\" \u003cmarek.olsak@amd.com\u003e\nCc: Hans de Goede \u003chdegoede@redhat.com\u003e\nReviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nReviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nSigned-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\nSigned-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\nCc: stable@vger.kernel.org\n" }, { "commit": "91ad64a84e9e63e2906ae714dfa3933dd3f64c64", "tree": "a6751cdf19320e5e5a4cd0ab876de5f1146a1d0e", "parents": [ "b98cce1ef5c5ecb8028531ec2305c4591c9b6ca1", "2910b5aa6f545c044173a5cab3dbb7f43e23916d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 10:34:42 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 10:34:42 2020 -0800" }, "message": "Merge tag \u0027trace-v5.6-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace\n\nPull tracing and bootconfig updates:\n \"Fixes and changes to bootconfig before it goes live in a release.\n\n Change in API of bootconfig (before it comes live in a release):\n - Have a magic value \"BOOTCONFIG\" in initrd to know a bootconfig\n exists\n - Set CONFIG_BOOT_CONFIG to \u0027n\u0027 by default\n - Show error if \"bootconfig\" on cmdline but not compiled in\n - Prevent redefining the same value\n - Have a way to append values\n - Added a SELECT BLK_DEV_INITRD to fix a build failure\n\n Synthetic event fixes:\n - Switch to raw_smp_processor_id() for recording CPU value in preempt\n section. (No care for what the value actually is)\n - Fix samples always recording u64 values\n - Fix endianess\n - Check number of values matches number of fields\n - Fix a printing bug\n\n Fix of trace_printk() breaking postponed start up tests\n\n Make a function static that is only used in a single file\"\n\n* tag \u0027trace-v5.6-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:\n bootconfig: Fix CONFIG_BOOTTIME_TRACING dependency issue\n bootconfig: Add append value operator support\n bootconfig: Prohibit re-defining value on same key\n bootconfig: Print array as multiple commands for legacy command line\n bootconfig: Reject subkey and value on same parent key\n tools/bootconfig: Remove unneeded error message silencer\n bootconfig: Add bootconfig magic word for indicating bootconfig explicitly\n bootconfig: Set CONFIG_BOOT_CONFIG\u003dn by default\n tracing: Clear trace_state when starting trace\n bootconfig: Mark boot_config_checksum() static\n tracing: Disable trace_printk() on post poned tests\n tracing: Have synthetic event test use raw_smp_processor_id()\n tracing: Fix number printing bug in print_synth_event()\n tracing: Check that number of vals matches number of synth event fields\n tracing: Make synth_event trace functions endian-correct\n tracing: Make sure synth_event_trace() example always uses u64\n" }, { "commit": "b98cce1ef5c5ecb8028531ec2305c4591c9b6ca1", "tree": "bc383fadf29133ea406051aac1539ef0b82365d8", "parents": [ "2fcc74178f6dc06ecffcd1f601ca3a3e87b3012c", "be886ba90cce2fb2f5a4dbcda8f3be3fd1b2f484" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 10:28:59 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 10:28:59 2020 -0800" }, "message": "Merge tag \u0027linux-kselftest-kunit-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest\n\nPull Kunit fixes from Shuah Khan:\n \"This Kselftest kunit update consists of fixes to documentation and\n the run-time tool from Brendan Higgins and Heidi Fahim\"\n\n* tag \u0027linux-kselftest-kunit-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:\n kunit: run kunit_tool from any directory\n kunit: test: Improve error messages for kunit_tool when kunitconfig is invalid\n Documentation: kunit: fixed sphinx error in code block\n" }, { "commit": "2fcc74178f6dc06ecffcd1f601ca3a3e87b3012c", "tree": "708049263a646f12b540fd8a3cf627e11fc0d3d8", "parents": [ "cfe2ce49b9da3959015e94b08f7494ade3ee0c49", "ef89d0545132d685f73da6f58b7e7fe002536f91" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 10:06:56 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 10:06:56 2020 -0800" }, "message": "Merge tag \u0027linux-kselftest-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest\n\nPull Kselftest fixes from Shuah Khan:\n\n - fixes to TIMEOUT failures and out-of-tree compilation compilation\n errors from Michael Ellerman.\n\n - declutter git status fix from Christophe Leroy\n\n* tag \u0027linux-kselftest-5.6-rc4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:\n selftests/rseq: Fix out-of-tree compilation\n selftests: Install settings files to fix TIMEOUT failures\n selftest/lkdtm: Don\u0027t pollute \u0027git status\u0027\n" }, { "commit": "cfe2ce49b9da3959015e94b08f7494ade3ee0c49", "tree": "c6f2fa8f024fbe30c486addcfb600c634ad53309", "parents": [ "fda31c50292a5062332fa0343c084bd9f46604d9" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Wed Feb 26 07:39:29 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 09:59:58 2020 -0800" }, "message": "Revert \"KVM: x86: enable -Werror\"\n\nThis reverts commit ead68df94d248c80fdbae220ae5425eb5af2e753.\n\nUsing the -Werror flag breaks the build for me due to mostly harmless\nKASAN or similar warnings:\n\n arch/x86/kvm/x86.c: In function ‘kvm_timer_init’:\n arch/x86/kvm/x86.c:7209:1: error: the frame size of 1112 bytes is larger than 1024 bytes [-Werror\u003dframe-larger-than\u003d]\n\nFeel free to add a CONFIG_WERROR if you care strong enough, but don\u0027t\nbreak peoples builds for absolutely no good reason.\n\nSigned-off-by: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "fda31c50292a5062332fa0343c084bd9f46604d9", "tree": "90686674317d132008a54e6c8c7d726a4713f7c7", "parents": [ "c5f86891185c408b2241ba9a82ae8622d8386aff" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Feb 24 12:47:14 2020 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Feb 26 09:54:03 2020 -0800" }, "message": "signal: avoid double atomic counter increments for user accounting\n\nWhen queueing a signal, we increment both the users count of pending\nsignals (for RLIMIT_SIGPENDING tracking) and we increment the refcount\nof the user struct itself (because we keep a reference to the user in\nthe signal structure in order to correctly account for it when freeing).\n\nThat turns out to be fairly expensive, because both of them are atomic\nupdates, and particularly under extreme signal handling pressure on big\nmachines, you can get a lot of cache contention on the user struct.\nThat can then cause horrid cacheline ping-pong when you do these\nmultiple accesses.\n\nSo change the reference counting to only pin the user for the _first_\npending signal, and to unpin it when the last pending signal is\ndequeued. That means that when a user sees a lot of concurrent signal\nqueuing - which is the only situation when this matters - the only\natomic access needed is generally the \u0027sigpending\u0027 count update.\n\nThis was noticed because of a particularly odd timing artifact on a\ndual-socket 96C/192T Cascade Lake platform: when you get into bad\ncontention, on that machine for some reason seems to be much worse when\nthe contention happens in the upper 32-byte half of the cacheline.\n\nAs a result, the kernel test robot will-it-scale \u0027signal1\u0027 benchmark had\nan odd performance regression simply due to random alignment of the\n\u0027struct user_struct\u0027 (and pointed to a completely unrelated and\napparently nonsensical commit for the regression).\n\nAvoiding the double increments (and decrements on the dequeueing side,\nof course) makes for much less contention and hugely improved\nperformance on that will-it-scale microbenchmark.\n\nQuoting Feng Tang:\n\n \"It makes a big difference, that the performance score is tripled! bump\n from original 17000 to 54000. Also the gap between 5.0-rc6 and\n 5.0-rc6+Jiri\u0027s patch is reduced to around 2%\"\n\n[ The \"2% gap\" is the odd cacheline placement difference on that\n platform: under the extreme contention case, the effect of which half\n of the cacheline was hot was 5%, so with the reduced contention the\n odd timing artifact is reduced too ]\n\nIt does help in the non-contended case too, but is not nearly as\nnoticeable.\n\nReported-and-tested-by: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nCc: Huang, Ying \u003cying.huang@intel.com\u003e\nCc: Philip Li \u003cphilip.li@intel.com\u003e\nCc: Andi Kleen \u003candi.kleen@intel.com\u003e\nCc: Jiri Olsa \u003cjolsa@redhat.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "dd3db2a34cff14e152da7c8e320297719a35abf9", "tree": "5657a87b68b44aeb12b9cb426d242a4bbf7f6a80", "parents": [ "3a9015988b3d41027cda61f4fdeaaeee73be8b24" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 26 10:23:43 2020 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 26 10:53:33 2020 -0700" }, "message": "io_uring: drop file set ref put/get on switch\n\nDan reports that he triggered a warning on ring exit doing some testing:\n\npercpu ref (io_file_data_ref_zero) \u003c\u003d 0 (0) after switching to atomic\nWARNING: CPU: 3 PID: 0 at lib/percpu-refcount.c:160 percpu_ref_switch_to_atomic_rcu+0xe8/0xf0\nModules linked in:\nCPU: 3 PID: 0 Comm: swapper/3 Not tainted 5.6.0-rc3+ #5648\nHardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014\nRIP: 0010:percpu_ref_switch_to_atomic_rcu+0xe8/0xf0\nCode: e7 ff 55 e8 eb d2 80 3d bd 02 d2 00 00 75 8b 48 8b 55 d8 48 c7 c7 e8 70 e6 81 c6 05 a9 02 d2 00 01 48 8b 75 e8 e8 3a d0 c5 ff \u003c0f\u003e 0b e9 69 ff ff ff 90 55 48 89 fd 53 48 89 f3 48 83 ec 28 48 83\nRSP: 0018:ffffc90000110ef8 EFLAGS: 00010292\nRAX: 0000000000000045 RBX: 7fffffffffffffff RCX: 0000000000000000\nRDX: 0000000000000045 RSI: ffffffff825be7a5 RDI: ffffffff825bc32c\nRBP: ffff8881b75eac38 R08: 000000042364b941 R09: 0000000000000045\nR10: ffffffff825beb40 R11: ffffffff825be78a R12: 0000607e46005aa0\nR13: ffff888107dcdd00 R14: 0000000000000000 R15: 0000000000000009\nFS: 0000000000000000(0000) GS:ffff8881b9d80000(0000) knlGS:0000000000000000\nCS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\nCR2: 00007f49e6a5ea20 CR3: 00000001b747c004 CR4: 00000000001606e0\nDR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\nDR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\nCall Trace:\n \u003cIRQ\u003e\n rcu_core+0x1e4/0x4d0\n __do_softirq+0xdb/0x2f1\n irq_exit+0xa0/0xb0\n smp_apic_timer_interrupt+0x60/0x140\n apic_timer_interrupt+0xf/0x20\n \u003c/IRQ\u003e\nRIP: 0010:default_idle+0x23/0x170\nCode: ff eb ab cc cc cc cc 0f 1f 44 00 00 41 54 55 53 65 8b 2d 10 96 92 7e 0f 1f 44 00 00 e9 07 00 00 00 0f 00 2d 21 d0 51 00 fb f4 \u003c65\u003e 8b 2d f6 95 92 7e 0f 1f 44 00 00 5b 5d 41 5c c3 65 8b 05 e5 95\n\nTurns out that this is due to percpu_ref_switch_to_atomic() only\ngrabbing a reference to the percpu refcount if it\u0027s not already in\natomic mode. io_uring drops a ref and re-gets it when switching back to\npercpu mode. We attempt to protect against this with the FFD_F_ATOMIC\nbit, but that isn\u0027t reliable.\n\nWe don\u0027t actually need to juggle these refcounts between atomic and\npercpu switch, we can just do them when we\u0027ve switched to atomic mode.\nThis removes the need for FFD_F_ATOMIC, which wasn\u0027t reliable.\n\nFixes: 05f3fb3c5397 (\"io_uring: avoid ring quiesce for fixed file set unregister and update\")\nReported-by: Dan Melnic \u003cdmm@fb.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "cae740a04b4d6d5166f19ee5faf04ea2a1f34b3d", "tree": "ef8f77378eab19e730f18edfe88cd5bd954d850a", "parents": [ "93d7c3185893b185e7f4347f8986b9b521254a6e" ], "author": { "name": "John Garry", "email": "john.garry@huawei.com", "time": "Wed Feb 26 20:10:15 2020 +0800" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 26 10:34:41 2020 -0700" }, "message": "blk-mq: Remove some unused function arguments\n\nThe struct blk_mq_hw_ctx pointer argument in blk_mq_put_tag(),\nblk_mq_poll_nsecs(), and blk_mq_poll_hybrid_sleep() is unused, so remove\nit.\n\nOverall obj code size shows a minor reduction, before:\n text\t data\t bss\t dec\t hex\tfilename\n 27306\t 1312\t 0\t 28618\t 6fca\tblock/blk-mq.o\n 4303\t 272\t 0\t 4575\t 11df\tblock/blk-mq-tag.o\n\nafter:\n 27282\t 1312\t 0\t 28594\t 6fb2\tblock/blk-mq.o\n 4311\t 272\t 0\t 4583\t 11e7\tblock/blk-mq-tag.o\n\nReviewed-by: Johannes Thumshirn \u003cjohannes.thumshirn@wdc.com\u003e\nReviewed-by: Hannes Reinecke \u003chare@suse.de\u003e\nSigned-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n--\nThis minor patch had been carried as part of the blk-mq shared tags RFC,\nI\u0027d rather not carry it anymore as it required rebasing, so now or never..\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "c473a8d03ea8e03ca9d649b0b6d72fbcf6212c05", "tree": "1bfa77049c52dfd9f29f3b5d8783323821521f2b", "parents": [ "964a596db8db8c77c9903dd05655696696e6b3ad" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Sun Feb 23 04:04:34 2020 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:28:15 2020 +0900" }, "message": "kbuild: add dt_binding_check to PHONY in a correct place\n\nThe dt_binding_check is added to PHONY, but it is invisible when\n$(dtstree) is empty. So, it is not specified as phony for\nARCH\u003dx86 etc.\n\nAdd it to PHONY outside the ifneq ... endif block.\n\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\nAcked-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "964a596db8db8c77c9903dd05655696696e6b3ad", "tree": "99427310579daa507e69d28774ca38591281207b", "parents": [ "fd63fab48f143f73b534821408a303241ed174f9" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Sun Feb 23 04:04:33 2020 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:26:01 2020 +0900" }, "message": "kbuild: add dtbs_check to PHONY\n\nThe dtbs_check should be a phony target, but currently it is not\nspecified so.\n\n\u0027make dtbs_check\u0027 works even if a file named \u0027dtbs_check\u0027 exists\nbecause it depends on another phony target, scripts_dtc, but we\nshould not rely on it.\n\nAdd dtbs_check to PHONY.\n\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\nAcked-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "fd63fab48f143f73b534821408a303241ed174f9", "tree": "045a9e89055798e87e57e6c358d8192aa4ff9462", "parents": [ "7a04960560640ac5b0b89461f7757322b57d0c7a" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Sun Feb 23 04:04:32 2020 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:24:34 2020 +0900" }, "message": "kbuild: remove unneeded semicolon at the end of cmd_dtb_check\n\nThis trailing semicolon is unneeded.\n\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\nAcked-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "7a04960560640ac5b0b89461f7757322b57d0c7a", "tree": "05d5b17d913fc1bd55afeb61f843df257b43b022", "parents": [ "eccbde4f6c2b6ebc52b3e9103e6f2f73f5a9f79a" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Sun Feb 23 04:04:31 2020 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:23:12 2020 +0900" }, "message": "kbuild: fix DT binding schema rule to detect command line changes\n\nThis if_change_rule is not working properly; it cannot detect any\ncommand line change.\n\nThe reason is because cmd-check in scripts/Kbuild.include compares\n$(cmd_$@) and $(cmd_$1), but cmd_dtc_dt_yaml does not exist here.\n\nFor if_change_rule to work properly, the stem part of cmd_* and rule_*\nmust match. Because this cmd_and_fixdep invokes cmd_dtc, this rule must\nbe named rule_dtc.\n\nFixes: 4f0e3a57d6eb (\"kbuild: Add support for DT binding schema checks\")\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\nAcked-by: Rob Herring \u003crobh@kernel.org\u003e\n" }, { "commit": "eccbde4f6c2b6ebc52b3e9103e6f2f73f5a9f79a", "tree": "c200df9ceaf850c5ce480f3276be663a324aafea", "parents": [ "505b12b3861bc79d1b81c815faaf4910469a7006" ], "author": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Wed Feb 19 10:15:19 2020 +0900" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:19:30 2020 +0900" }, "message": "kbuild: remove wrong documentation about mandatory-y\n\nThis sentence does not make sense in the section about mandatory-y.\n\nThis seems to be a copy-paste mistake of commit fcc8487d477a (\"uapi:\nexport all headers under uapi directories\").\n\nThe correct description would be \"The convention is to list one\nmandatory-y per line ...\".\n\nI just removed it instead of fixing it. If such information is needed,\nit could be commented in include/asm-generic/Kbuild and\ninclude/uapi/asm-generic/Kbuild.\n\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\n" }, { "commit": "505b12b3861bc79d1b81c815faaf4910469a7006", "tree": "47b2462d422aac1c820c4c070217eaddd4466bdf", "parents": [ "f8788d86ab28f61f7b46eb6be375f8a726783636" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Wed Feb 12 20:40:57 2020 -0800" }, "committer": { "name": "Masahiro Yamada", "email": "masahiroy@kernel.org", "time": "Thu Feb 27 02:19:07 2020 +0900" }, "message": "kbuild: add comment for V\u003d2 mode\n\nComplete the comments for valid values of KBUILD_VERBOSE,\nspecifically for KBUILD_VERBOSE\u003d2.\n\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nSigned-off-by: Masahiro Yamada \u003cmasahiroy@kernel.org\u003e\n" }, { "commit": "3a9015988b3d41027cda61f4fdeaaeee73be8b24", "tree": "4722ab78b6f19ec409294d30fa281f5184f5a45c", "parents": [ "2a44f46781617c5040372b59da33553a02b1f46d" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Tue Feb 25 17:48:55 2020 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 26 07:06:57 2020 -0700" }, "message": "io_uring: import_single_range() returns 0/-ERROR\n\nUnlike the other core import helpers, import_single_range() returns 0 on\nsuccess, not the length imported. This means that links that depend on\nthe result of non-vec based IORING_OP_{READ,WRITE} that were added for\n5.5 get errored when they should not be.\n\nFixes: 3a6820f2bb8a (\"io_uring: add non-vectored read/write commands\")\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "2a44f46781617c5040372b59da33553a02b1f46d", "tree": "2f7e9657213e6df2d545e8567e007901b741530e", "parents": [ "2d141dd2caa78fbaf87b57c27769bdc14975ab3d" ], "author": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Tue Feb 25 13:25:41 2020 -0700" }, "committer": { "name": "Jens Axboe", "email": "axboe@kernel.dk", "time": "Wed Feb 26 07:05:30 2020 -0700" }, "message": "io_uring: pick up link work on submit reference drop\n\nIf work completes inline, then we should pick up a dependent link item\nin __io_queue_sqe() as well. If we don\u0027t do so, we\u0027re forced to go async\nwith that item, which is suboptimal.\n\nThis also fixes an issue with io_put_req_find_next(), which always looks\nup the next work item. That should only be done if we\u0027re dropping the\nlast reference to the request, to prevent multiple lookups of the same\nwork item.\n\nOutside of being a fix, this also enables a good cleanup series for 5.7,\nwhere we never have to pass \u0027nxt\u0027 around or into the work handlers.\n\nReviewed-by: Pavel Begunkov \u003casml.silence@gmail.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n" }, { "commit": "0954df70fba743d8cdaa09ccf6ba8e4ad09628de", "tree": "90a12c252dd66279acf91c41e23d49dc5d2ae31c", "parents": [ "212d58c106fd0f2704664be2bb173e14cb4e86d3" ], "author": { "name": "Stefano Brivio", "email": "sbrivio@redhat.com", "time": "Fri Feb 21 03:04:22 2020 +0100" }, "committer": { "name": "Pablo Neira Ayuso", "email": "pablo@netfilter.org", "time": "Wed Feb 26 14:33:09 2020 +0100" }, "message": "selftests: nft_concat_range: Add test for reported add/flush/add issue\n\nAdd a specific test for the crash reported by Phil Sutter and addressed\nin the previous patch. The test cases that, in my intention, should\nhave covered these cases, that is, the ones from the \u0027concurrency\u0027\nsection, don\u0027t run these sequences tightly enough and spectacularly\nfailed to catch this.\n\nWhile at it, define a convenient way to add these kind of tests, by\nadding a \"reported issues\" test section.\n\nIt\u0027s more convenient, for this particular test, to execute the set\nsetup in its own function. However, future test cases like this one\nmight need to call setup functions, and will typically need no tools\nother than nft, so allow for this in check_tools().\n\nThe original form of the reproducer used here was provided by Phil.\n\nReported-by: Phil Sutter \u003cphil@nwl.cc\u003e\nSigned-off-by: Stefano Brivio \u003csbrivio@redhat.com\u003e\nSigned-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n" } ], "next": "212d58c106fd0f2704664be2bb173e14cb4e86d3" }