summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2025-12-12i3c: master: Fix confusing cleanup.h syntaxKrzysztof Kozlowski
Initializing automatic __free variables to NULL without need (e.g. branches with different allocations), followed by actual allocation is in contrary to explicit coding rules guiding cleanup.h: "Given that the "__free(...) = NULL" pattern for variables defined at the top of the function poses this potential interdependency problem the recommendation is to always define and assign variables in one statement and not group variable definitions at the top of the function when __free() is used." Code does not have a bug, but is less readable and uses discouraged coding practice, so fix that by moving declaration to the place of assignment. Not that other existing usage of __free() in this context is a corret exception initialized to NULL, because the actual allocation is branched in if(). Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com> Link: https://patch.msgid.link/20251208020750.4727-3-krzysztof.kozlowski@oss.qualcomm.com Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
2025-12-12i3c: master: cleanup callback .priv_xfers()Frank Li
Remove the .priv_xfers() callback from the framework after all master controller drivers have switched to use the new .i3c_xfers() callback. Signed-off-by: Frank Li <Frank.Li@nxp.com> Tested-by: Tommaso Merciai <tommaso.merciai.xr@bp.renesas.com> Link: https://patch.msgid.link/20251203-i3c_xfer_cleanup_master-v2-2-7dd94d04ee2d@nxp.com Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
2025-12-12drm/xe: Restore engine registers before restarting schedulers after GT resetJan Maslak
During GT reset recovery in do_gt_restart(), xe_uc_start() was called before xe_reg_sr_apply_mmio() restored engine-specific registers. This created a race window where the scheduler could run jobs before hardware state was fully restored. This caused failures in eudebug tests (xe_exec_sip_eudebug@breakpoint- waitsip-*) where TD_CTL register (containing TD_CTL_GLOBAL_DEBUG_ENABLE) wasn't restored before jobs started executing. Breakpoints would fail to trigger SIP entry because the debug enable bit wasn't set yet. Fix by moving xe_uc_start() after all MMIO register restoration, including engine registers and CCS mode configuration, ensuring all hardware state is fully restored before any jobs can be scheduled. Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Signed-off-by: Jan Maslak <jan.maslak@intel.com> Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Signed-off-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251210145618.169625-2-jan.maslak@intel.com
2025-12-12block: fix race between wbt_enable_default and IO submissionMing Lei
When wbt_enable_default() is moved out of queue freezing in elevator_change(), it can cause the wbt inflight counter to become negative (-1), leading to hung tasks in the writeback path. Tasks get stuck in wbt_wait() because the counter is in an inconsistent state. The issue occurs because wbt_enable_default() could race with IO submission, allowing the counter to be decremented before proper initialization. This manifests as: rq_wait[0]: inflight: -1 has_waiters: True rwb_enabled() checks the state, which can be updated exactly between wbt_wait() (rq_qos_throttle()) and wbt_track()(rq_qos_track()), then the inflight counter will become negative. And results in hung task warnings like: task:kworker/u24:39 state:D stack:0 pid:14767 Call Trace: rq_qos_wait+0xb4/0x150 wbt_wait+0xa9/0x100 __rq_qos_throttle+0x24/0x40 blk_mq_submit_bio+0x672/0x7b0 ... Fix this by: 1. Splitting wbt_enable_default() into: - __wbt_enable_default(): Returns true if wbt_init() should be called - wbt_enable_default(): Wrapper for existing callers (no init) - wbt_init_enable_default(): New function that checks and inits WBT 2. Using wbt_init_enable_default() in blk_register_queue() to ensure proper initialization during queue registration 3. Move wbt_init() out of wbt_enable_default() which is only for enabling disabled wbt from bfq and iocost, and wbt_init() isn't needed. Then the original lock warning can be avoided. 4. Removing the ELEVATOR_FLAG_ENABLE_WBT_ON_EXIT flag and its handling code since it's no longer needed This ensures WBT is properly initialized before any IO can be submitted, preventing the counter from going negative. Cc: Nilay Shroff <nilay@linux.ibm.com> Cc: Yu Kuai <yukuai@fnnas.com> Cc: Guangwu Zhang <guazhang@redhat.com> Fixes: 78c271344b6f ("block: move wbt_enable_default() out of queue freezing from sched ->exit()") Signed-off-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: add user copy test casesCaleb Sander Mateos
The ublk selftests cover every data copy mode except user copy. Add tests for user copy based on the existing test suite: - generic_14 ("basic recover function verification (user copy)") based on generic_04 and generic_05 - null_03 ("basic IO test with user copy") based on null_01 and null_02 - loop_06 ("write and verify over user copy") based on loop_01 and loop_03 - loop_07 ("mkfs & mount & umount with user copy") based on loop_02 and loop_04 - stripe_05 ("write and verify test on user copy") based on stripe_03 - stripe_06 ("mkfs & mount & umount on user copy") based on stripe_02 and stripe_04 - stress_06 ("run IO and remove device (user copy)") based on stress_01 and stress_03 - stress_07 ("run IO and kill ublk server (user copy)") based on stress_02 and stress_04 Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: add support for user copy to kublkCaleb Sander Mateos
The ublk selftests mock ublk server kublk supports every data copy mode except user copy. Add support for user copy to kublk, enabled via the --user_copy (-u) command line argument. On writes, issue pread() calls to copy the write data into the ublk_io's buffer before dispatching the write to the target implementation. On reads, issue pwrite() calls to copy read data from the ublk_io's buffer before committing the request. Copy in 2 KB chunks to provide some coverage of the offseting logic. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: forbid multiple data copy modesCaleb Sander Mateos
The kublk mock ublk server allows multiple data copy mode arguments to be passed on the command line (--zero_copy, --get_data, and --auto_zc). The ublk device will be created with all the requested feature flags, however kublk will only use one of the modes to interact with request data (arbitrarily preferring auto_zc over zero_copy over get_data). To clarify the intent of the test, don't allow multiple data copy modes to be specified. --zero_copy and --auto_zc are allowed together for --auto_zc_fallback, which uses both copy modes. Don't set UBLK_F_USER_COPY for zero_copy, as it's a separate feature. Fix the test cases in test_stress_05 passing --get_data along with --zero_copy or --auto_zc. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: don't share backing files between ublk serversCaleb Sander Mateos
stress_04 is missing a wait between blocks of tests, meaning multiple ublk servers will be running in parallel using the same backing files. Add a wait after each section to ensure each backing file is in use by a single ublk server at a time. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: use auto_zc for PER_IO_DAEMON tests in stress_04Caleb Sander Mateos
stress_04 is described as "run IO and kill ublk server(zero copy)" but the --per_io_tasks tests cases don't use zero copy. Plus, one of the test cases is duplicated. Add --auto_zc to these test cases and --auto_zc_fallback to one of the duplicated ones. This matches the test cases in stress_03. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: fix fio arguments in run_io_and_recover()Caleb Sander Mateos
run_io_and_recover() invokes fio with --size="${size}", but the variable size doesn't exist. Thus, the argument expands to --size=, which causes fio to exit immediately with an error without issuing any I/O. Pass the value for size as the first argument to the function. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: remove unused ios map in seq_io.btCaleb Sander Mateos
The ios map populated by seq_io.bt is never read, so remove it. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: correct last_rw map type in seq_io.btCaleb Sander Mateos
The last_rw map is initialized with a value of 0 but later assigned the value args.sector + args.nr_sector, which has type sector_t = u64. bpftrace complains about the type mismatch between int64 and uint64: trace/seq_io.bt:18:3-59: ERROR: Type mismatch for @last_rw: trying to assign value of type 'uint64' when map already contains a value of type 'int64' @last_rw[$dev, str($2)] = (args.sector + args.nr_sector); Cast the initial value to uint64 so bpftrace will load the program. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests: ublk: fix overflow in ublk_queue_auto_zc_fallback()Ming Lei
The functions ublk_queue_use_zc(), ublk_queue_use_auto_zc(), and ublk_queue_auto_zc_fallback() were returning int, but performing bitwise AND on q->flags which is __u64. When a flag bit is set in the upper 32 bits (beyond INT_MAX), the result of the bitwise AND operation could overflow when cast to int, leading to incorrect boolean evaluation. For example, if UBLKS_Q_AUTO_BUF_REG_FALLBACK is 0x8000000000000000: - (u64)flags & 0x8000000000000000 = 0x8000000000000000 - Cast to int: undefined behavior / incorrect value - Used in if(): may evaluate incorrectly Fix by: 1. Changing return type from int to bool for semantic correctness 2. Using !! to explicitly convert to boolean (0 or 1) This ensures the functions return proper boolean values regardless of which bit position the flags occupy in the 64-bit field. Fixes: c3a6d48f86da ("selftests: ublk: remove ublk queue self-defined flags") Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12block: move around bio flagging helpersPavel Begunkov
We'll need bio_flagged() earlier in bio.h for later patches, move it together with all related helpers, and mark the bio_flagged()'s bio argument as const. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-12-12selftests/sched_ext: flush stdout before test to avoid log spamEmil Tsalapatis
The sched_ext selftests runner runs each test in the same process, with each test possibly forking multiple times. When the main runner has not flushed its stdout, the children inherit the buffered output for previous tests and emit it during exit. This causes log spam. Make sure stdout/stderr is fully flushed before each test. Cc: Ihor Solodrai <ihor.solodrai@linux.dev> Signed-off-by: Emil Tsalapatis <emil@etsalapatis.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2025-12-12accel/amdxdna: Fix race where send ring appears full due to delayed head updateLizhi Hou
The firmware sends a response and interrupts the driver before advancing the mailbox send ring head pointer. As a result, the driver may observe the response and attempt to send a new request before the firmware has updated the head pointer. In this window, the send ring still appears full, causing the driver to incorrectly fail the send operation. This race can be triggered more easily in a multithreaded environment, leading to unexpected and spurious "send ring full" failures. To address this, poll the send ring head pointer for up to 100us before returning a full-ring condition. This allows the firmware time to update the head pointer. Fixes: b87f920b9344 ("accel/amdxdna: Support hardware mailbox") Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Lizhi Hou <lizhi.hou@amd.com> Link: https://patch.msgid.link/20251211045125.1724604-1-lizhi.hou@amd.com
2025-12-13Merge tag 'loongarch-6.19' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson Pull LoongArch updates from Huacai Chen: - Add basic LoongArch32 support Note: Build infrastructures of LoongArch32 are not enabled yet, because we need to adjust irqchip drivers and wait for GNU toolchain be upstream first. - Select HAVE_ARCH_BITREVERSE in Kconfig - Fix build and boot for CONFIG_RANDSTRUCT - Correct the calculation logic of thread_count - Some bug fixes and other small changes * tag 'loongarch-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson: (22 commits) LoongArch: Adjust default config files for 32BIT/64BIT LoongArch: Adjust VDSO/VSYSCALL for 32BIT/64BIT LoongArch: Adjust misc routines for 32BIT/64BIT LoongArch: Adjust user accessors for 32BIT/64BIT LoongArch: Adjust system call for 32BIT/64BIT LoongArch: Adjust module loader for 32BIT/64BIT LoongArch: Adjust time routines for 32BIT/64BIT LoongArch: Adjust process management for 32BIT/64BIT LoongArch: Adjust memory management for 32BIT/64BIT LoongArch: Adjust boot & setup for 32BIT/64BIT LoongArch: Adjust common macro definitions for 32BIT/64BIT LoongArch: Add adaptive CSR accessors for 32BIT/64BIT LoongArch: Add atomic operations for 32BIT/64BIT LoongArch: Add new PCI ID for pci_fixup_vgadev() LoongArch: Add and use some macros for AVEC LoongArch: Correct the calculation logic of thread_count LoongArch: Use unsigned long for _end and _text LoongArch: Use __pmd()/__pte() for swap entry conversions LoongArch: Fix arch_dup_task_struct() for CONFIG_RANDSTRUCT LoongArch: Fix build errors for CONFIG_RANDSTRUCT ...
2025-12-12drm/xe: Increase TDF timeoutJagmeet Randhawa
There are some corner cases where flushing transient data may take slightly longer than the 150us timeout we currently allow. Update the driver to use a 300us timeout instead based on the latest guidance from the hardware team. An update to the bspec to formally document this is expected to arrive soon. Fixes: c01c6066e6fa ("drm/xe/device: implement transient flush") Signed-off-by: Jagmeet Randhawa <jagmeet.randhawa@intel.com> Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patch.msgid.link/0201b1d6ec64d3651fcbff1ea21026efa915126a.1765487866.git.jagmeet.randhawa@intel.com Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
2025-12-12drm/xe/cri: Enable I2C controllerRaag Jadav
Enable I2C controller for Crescent Island and while at it, rely on has_i2c flag instead of manual platform checks. Signed-off-by: Raag Jadav <raag.jadav@intel.com> Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patch.msgid.link/20251128084414.306265-1-raag.jadav@intel.com Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
2025-12-12sched_ext: Fix missing post-enqueue handling in move_local_task_to_local_dsq()Tejun Heo
move_local_task_to_local_dsq() is used when moving a task from a non-local DSQ to a local DSQ on the same CPU. It directly manipulates the local DSQ without going through dispatch_enqueue() and was missing the post-enqueue handling that triggers preemption when SCX_ENQ_PREEMPT is set or the idle task is running. The function is used by move_task_between_dsqs() which backs scx_bpf_dsq_move() and may be called while the CPU is busy. Add local_dsq_post_enq() call to move_local_task_to_local_dsq(). As the dispatch path doesn't need post-enqueue handling, add SCX_RQ_IN_BALANCE early exit to keep consume_dispatch_q() behavior unchanged and avoid triggering unnecessary resched when scx_bpf_dsq_move() is used from the dispatch path. Fixes: 4c30f5ce4f7a ("sched_ext: Implement scx_bpf_dispatch[_vtime]_from_dsq()") Cc: stable@vger.kernel.org # v6.12+ Reviewed-by: Andrea Righi <arighi@nvidia.com> Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2025-12-12sched_ext: Factor out local_dsq_post_enq() from dispatch_enqueue()Tejun Heo
Factor out local_dsq_post_enq() which performs post-enqueue handling for local DSQs - triggering resched_curr() if SCX_ENQ_PREEMPT is specified or if the current CPU is idle. No functional change. This will be used by the next patch to fix move_local_task_to_local_dsq(). Cc: stable@vger.kernel.org # v6.12+ Reviewed-by: Andrea Righi <arighi@nvidia.com> Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2025-12-12btrfs: fix changeset leak on mmap write after failure to reserve metadataFilipe Manana
If the call to btrfs_delalloc_reserve_metadata() fails we jump to the 'out_noreserve' label and there we never free the extent_changeset allocated by the previous call to btrfs_check_data_free_space() (if qgroups are enabled). Fix this by calling extent_changeset_free() under the 'out_noreserve' label. Fixes: 6599716de2d6 ("btrfs: fix -ENOSPC mmap write failure on NOCOW files/extents") Reported-by: syzbot+2f8aa76e6acc9fce6638@syzkaller.appspotmail.com Link: https://lore.kernel.org/linux-btrfs/693a635a.a70a0220.33cd7b.0029.GAE@google.com/ Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-12-12btrfs: fix memory leak of fs_devices in degraded seed device pathDeepanshu Kartikey
In open_seed_devices(), when find_fsid() fails and we're in DEGRADED mode, a new fs_devices is allocated via alloc_fs_devices() but is never added to the seed_list before returning. This contrasts with the normal path where fs_devices is properly added via list_add(). If any error occurs later in read_one_dev() or btrfs_read_chunk_tree(), the cleanup code iterates seed_list to free seed devices, but this orphaned fs_devices is never found and never freed, causing a memory leak. Any devices allocated via add_missing_dev() and attached to this fs_devices are also leaked. Fix this by adding the newly allocated fs_devices to seed_list in the degraded path, consistent with the normal path. Fixes: 5f37583569442 ("Btrfs: move the missing device to its own fs device list") Reported-by: syzbot+eadd98df8bceb15d7fed@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=eadd98df8bceb15d7fed Tested-by: syzbot+eadd98df8bceb15d7fed@syzkaller.appspotmail.com Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-12-12drm: Fix object leak in DRM_IOCTL_GEM_CHANGE_HANDLEKarol Wachowski
Add missing drm_gem_object_put() call when drm_gem_object_lookup() successfully returns an object. This fixes a GEM object reference leak that can prevent driver modules from unloading when using prime buffers. Fixes: 53096728b891 ("drm: Add DRM prime interface to reassign GEM handle") Cc: <stable@vger.kernel.org> # v6.18+ Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com> Signed-off-by: Christian König <christian.koenig@amd.com> Link: https://lore.kernel.org/r/20251212134133.475218-1-karol.wachowski@linux.intel.com
2025-12-12drm/{i915, xe}/panic: move panic handling to parent interfaceJani Nikula
Move the panic handling to the display parent interface, making display more independent of i915 and xe driver implementations. Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patch.msgid.link/e27eca5424479e8936b786018d0af19a34f839f6.1765474612.git.jani.nikula@intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2025-12-12drm/i915/panic: move i915 specific panic implementation to i915Jani Nikula
The intel_panic.c implementation is i915 specific, and xe has its own. Move it to i915 core as i915_panic.c. Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patch.msgid.link/8dc7af0ae1f859d17b0be269a545146c5536d8fc.1765474612.git.jani.nikula@intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2025-12-12Merge tag 'libcrypto-fixes-for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux Pull crypto library fixes from Eric Biggers: "Fixes for some recent regressions as well as some longstanding issues: - Fix incorrect output from the arm64 NEON implementation of GHASH - Merge the ksimd scopes in the arm64 XTS code to reduce stack usage - Roll up the BLAKE2b round loop on 32-bit kernels to greatly reduce code size and stack usage - Add missing RISCV_EFFICIENT_VECTOR_UNALIGNED_ACCESS dependency - Fix chacha-riscv64-zvkb.S to not use frame pointer for data" * tag 'libcrypto-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux: crypto: arm64/ghash - Fix incorrect output from ghash-neon crypto/arm64: sm4/xts - Merge ksimd scopes to reduce stack bloat crypto/arm64: aes/xts - Use single ksimd scope to reduce stack bloat lib/crypto: blake2s: Replace manual unrolling with unrolled_full lib/crypto: blake2b: Roll up BLAKE2b round loop on 32-bit lib/crypto: riscv: Depend on RISCV_EFFICIENT_VECTOR_UNALIGNED_ACCESS lib/crypto: riscv/chacha: Avoid s0/fp register
2025-12-12Merge tag 'block-6.19-20251211' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux Pull block fixes from Jens Axboe: - Always initialize DMA state, fixing a potentially nasty issue on the block side - btrfs zoned write fix with cached zone reports - Fix corruption issues in bcache with chained bio's, and further make it clear that the chained IO handler is simply a marker, it's not code meant to be executed - Kill old code dealing with synchronous IO polling in the block layer, that has been dead for a long time. Only async polling is supported these days - Fix a lockdep issue in tag_set management, moving it to RCU - Fix an issue with ublks bio_vec iteration - Don't unconditionally enforce blocking issue of ublk control commands, allow some of them with non-blocking issue as they do not block * tag 'block-6.19-20251211' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux: blk-mq-dma: always initialize dma state blk-mq: delete task running check in blk_hctx_poll() block: fix cached zone reports on devices with native zone append block: Use RCU in blk_mq_[un]quiesce_tagset() instead of set->tag_list_lock ublk: don't mutate struct bio_vec in iteration block: prohibit calls to bio_chain_endio bcache: fix improper use of bi_end_io ublk: allow non-blocking ctrl cmds in IO_URING_F_NONBLOCK issue
2025-12-12Merge tag 'io_uring-6.19-20251211' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux Pull io_uring fix from Jens Axboe: "Single fix for io_uring headed to stable, fixing an issue introduced with the min_wait support earlier this year, where SQPOLL didn't get correctly woken if an event arrived once the event waiting has finished the min_wait portion. As we already have regression tests for this added and people reporting new failures there, let's get this one flushed out so it can bubble back down to stable as well" * tag 'io_uring-6.19-20251211' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux: io_uring: fix min_wait wakeups for SQPOLL
2025-12-12Merge tag 'v6.19-rc-smb3-server-fixes' of git://git.samba.org/ksmbdLinus Torvalds
Pull smb server fixes from Steve French: - minor cleanup - minor update to comment to avoid confusion about fs type * tag 'v6.19-rc-smb3-server-fixes' of git://git.samba.org/ksmbd: smb/server: add comment to FileSystemName of FileFsAttributeInformation smb/server: remove unused nterr.h smb/server: rename include guard in smb_common.h
2025-12-12Merge tag 'v6.19-rc-part2-smb3-client-fixes' of ↵Linus Torvalds
git://git.samba.org/sfrench/cifs-2.6 Pull smb client fixes from Steve French: - Fix incorrect error code defines - Add missing error code definitions - Add parenthesis around NT_STATUS code defines to fix checkpatch warnings - Remove some duplicated protocol definitions, moving to common code shared by client and server - Add missing protocol documentation reference (for change notify) - Correct struct definition (for duplicate_extents_to_file_ex) * tag 'v6.19-rc-part2-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6: smb/client: remove DeviceType Flags and Device Characteristics definitions smb: move File Attributes definitions into common/fscc.h smb: update struct duplicate_extents_to_file_ex smb: move file_notify_information to common/fscc.h smb: move SMB2 Notify Action Flags into common/smb2pdu.h smb: move notify completion filter flags into common/smb2pdu.h smb/client: add parentheses to NT error code definitions containing bitwise OR operator smb: add documentation references for smb2 change notify definitions smb/client: add 4 NT error code definitions smb/client: fix NT_STATUS_UNABLE_TO_FREE_VM value smb/client: fix NT_STATUS_DEVICE_DOOR_OPEN value smb/client: fix NT_STATUS_NO_DATA_DETECTED value
2025-12-12Merge tag 'nfs-for-6.19-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfsLinus Torvalds
Pull NFS client updates from Trond Myklebust: "Bugfixes: - Fix 'nlink' attribute update races when unlinking a file - Add missing initialisers for the directory verifier in various places - Don't regress the NFSv4 open state due to misordered racing replies - Ensure the NFSv4.x callback server uses the correct transport connection - Fix potential use-after-free races when shutting down the NFSv4.x callback server - Fix a pNFS layout commit crash - Assorted fixes to ensure correct propagation of mount options when the client crosses a filesystem boundary and triggers the VFS automount code - More localio fixes Features and cleanups: - Add initial support for basic directory delegations - SunRPC back channel code cleanups" * tag 'nfs-for-6.19-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (24 commits) NFSv4: Handle NFS4ERR_NOTSUPP errors for directory delegations nfs/localio: remove 61 byte hole from needless ____cacheline_aligned nfs/localio: remove alignment size checking in nfs_is_local_dio_possible NFS: Fix up the automount fs_context to use the correct cred NFS: Fix inheritance of the block sizes when automounting NFS: Automounted filesystems should inherit ro,noexec,nodev,sync flags Revert "nfs: ignore SB_RDONLY when mounting nfs" Revert "nfs: clear SB_RDONLY before getting superblock" Revert "nfs: ignore SB_RDONLY when remounting nfs" NFS: Add a module option to disable directory delegations NFS: Shortcut lookup revalidations if we have a directory delegation NFS: Request a directory delegation during RENAME NFS: Request a directory delegation on ACCESS, CREATE, and UNLINK NFS: Add support for sending GDD_GETATTR NFSv4/pNFS: Clear NFS_INO_LAYOUTCOMMIT in pnfs_mark_layout_stateid_invalid NFSv4.1: protect destroying and nullifying bc_serv structure SUNRPC: new helper function for stopping backchannel server SUNRPC: cleanup common code in backchannel request NFSv4.1: pass transport for callback shutdown NFSv4: ensure the open stateid seqid doesn't go backwards ...
2025-12-12rseq: Always inline rseq_debug_syscall_return()Eric Dumazet
To get the full benefit of: eaa9088d568c ("rseq: Use static branch for syscall exit debug when GENERIC_IRQ_ENTRY=y") clang needs an __always_inline instead of a plain inline qualifier: $ for i in {1..10}; do taskset -c 4 perf5 bench syscall basic -l 100000000 | grep "ops/sec"; done Before After ops/sec 15424491 15872221 +2.9% Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Link: https://patch.msgid.link/20251205100753.4073221-1-edumazet@google.com
2025-12-12bug: Hush suggest-attribute=format for __warn_printf()Brendan Jackman
Recent additions to this function cause GCC 14.3.0 to get excited (W=1) and suggest a missing attribute: lib/bug.c: In function '__warn_printf': lib/bug.c:187:25: error: function '__warn_printf' be a candidate for 'gnu_printf' format attribute [-Werror=suggest-attribute=format] 187 | vprintk(fmt, *args); | ^~~~~~~ Disable the diagnostic locally, following the pattern used for stuff like va_format(). Fixes: 5c47b7f3d1a9 ("bug: Add BUG_FORMAT_ARGS infrastructure") Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://patch.msgid.link/20251207-warn-printf-gcc-v1-1-b597d612b94b@google.com
2025-12-12bug: Let report_bug_entry() provide the correct bugaddrHeiko Carstens
report_bug_entry() always provides zero for bugaddr but could easily extract the correct address from the provided bug_entry. Just do that to have proper warning messages. E.g. adding an artificial: void foo(void) { WARN_ONCE(1, "bar"); } function generates this warning message: WARNING: arch/s390/kernel/setup.c:1017 at 0x0, CPU#0: swapper/0/0 ^^^ With the correct bug address this changes to: WARNING: arch/s390/kernel/setup.c:1017 at foo+0x1c/0x40, CPU#0: swapper/0/0 ^^^^^^^^^^^^^ Fixes: 7d2c27a0ec5e ("bug: Add report_bug_entry()") Signed-off-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://patch.msgid.link/20251208200658.3431511-1-hca@linux.ibm.com
2025-12-12drm/tests: Handle EDEADLK in set_up_atomic_state()José Expósito
Fedora/CentOS/RHEL CI is reporting intermittent failures while running the drm_validate_modeset test [1]: # drm_test_check_connector_changed_modeset: EXPECTATION FAILED at # drivers/gpu/drm/tests/drm_atomic_state_test.c:162 Expected ret == 0, but ret == -35 (0xffffffffffffffdd) Change the set_up_atomic_state() helper function to return on error and restart the atomic sequence when the returned error is EDEADLK. [1] https://s3.amazonaws.com/arr-cki-prod-trusted-artifacts/trusted-artifacts/2106744096/test_x86_64/11762450343/artifacts/jobwatch/logs/recipes/19797909/tasks/204139142/results/945095586/logs/dmesg.log Fixes: 73d934d7b6e3 ("drm/tests: Add test for drm_atomic_helper_commit_modeset_disables()") Closes: https://datawarehouse.cki-project.org/issue/4004 Reviewed-by: Maxime Ripard <mripard@kernel.org> Signed-off-by: José Expósito <jose.exposito89@gmail.com> Link: https://patch.msgid.link/20251104102535.12212-2-jose.exposito89@gmail.com
2025-12-12drm/tests: Handle EDEADLK in drm_test_check_valid_clones()José Expósito
Fedora/CentOS/RHEL CI is reporting intermittent failures while running the drm_test_check_valid_clones() KUnit test. The error log can be either [1]: # drm_test_check_valid_clones: ASSERTION FAILED at # drivers/gpu/drm/tests/drm_atomic_state_test.c:295 Expected ret == param->expected_result, but ret == -35 (0xffffffffffffffdd) param->expected_result == 0 (0x0) Or [2] depending on the test case: # drm_test_check_valid_clones: ASSERTION FAILED at # drivers/gpu/drm/tests/drm_atomic_state_test.c:295 Expected ret == param->expected_result, but ret == -35 (0xffffffffffffffdd) param->expected_result == -22 (0xffffffffffffffea) Restart the atomic sequence when EDEADLK is returned. [1] https://s3.amazonaws.com/arr-cki-prod-trusted-artifacts/trusted-artifacts/2113057246/test_x86_64/11802139999/artifacts/jobwatch/logs/recipes/19824965/tasks/204347800/results/946112713/logs/dmesg.log [2] https://s3.amazonaws.com/arr-cki-prod-trusted-artifacts/trusted-artifacts/2106744297/test_aarch64/11762450907/artifacts/jobwatch/logs/recipes/19797942/tasks/204139727/results/945094561/logs/dmesg.log Fixes: 88849f24e2ab ("drm/tests: Add test for drm_atomic_helper_check_modeset()") Closes: https://datawarehouse.cki-project.org/issue/4004 Reviewed-by: Maxime Ripard <mripard@kernel.org> Signed-off-by: José Expósito <jose.exposito89@gmail.com> Link: https://patch.msgid.link/20251104102535.12212-1-jose.exposito89@gmail.com
2025-12-12drm/tests: hdmi: Handle drm_kunit_helper_enable_crtc_connector() returning ↵José Expósito
EDEADLK Fedora/CentOS/RHEL CI is reporting intermittent failures while running the KUnit tests present in drm_hdmi_state_helper_test.c [1]. While the specific test causing the failure change between runs, all of them are caused by drm_kunit_helper_enable_crtc_connector() returning -EDEADLK. The error trace always follow this structure: # <test name>: ASSERTION FAILED at # drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c:<line> Expected ret == 0, but ret == -35 (0xffffffffffffffdd) As documented, if the drm_kunit_helper_enable_crtc_connector() function returns -EDEADLK (-35), the entire atomic sequence must be restarted. Handle this error code for all function calls. Closes: https://datawarehouse.cki-project.org/issue/4039 [1] Fixes: 6a5c0ad7e08e ("drm/tests: hdmi_state_helpers: Switch to new helper") Reviewed-by: Maxime Ripard <mripard@kernel.org> Signed-off-by: José Expósito <jose.exposito89@gmail.com> Link: https://patch.msgid.link/20251104102258.10026-1-jose.exposito89@gmail.com
2025-12-12Merge tag 'drm-intel-next-fixes-2025-12-12' of ↵Dave Airlie
https://gitlab.freedesktop.org/drm/i915/kernel into drm-next drm/i915 fixes for v6.19-rc1: - Fix format string truncation warning - FIx runtime PM reference during fbdev BO creation Signed-off-by: Dave Airlie <airlied@redhat.com> From: Jani Nikula <jani.nikula@intel.com> Link: https://patch.msgid.link/281309f78560bcceebac8d5c0511efe66baf641c@intel.com
2025-12-12perf/x86/intel: Fix NULL event dereference crash in handle_pmi_common()Evan Li
handle_pmi_common() may observe an active bit set in cpuc->active_mask while the corresponding cpuc->events[] entry has already been cleared, which leads to a NULL pointer dereference. This can happen when interrupt throttling stops all events in a group while PEBS processing is still in progress. perf_event_overflow() can trigger perf_event_throttle_group(), which stops the group and clears the cpuc->events[] entry, but the active bit may still be set when handle_pmi_common() iterates over the events. The following recent fix: 7e772a93eb61 ("perf/x86: Fix NULL event access and potential PEBS record loss") moved the cpuc->events[] clearing from x86_pmu_stop() to x86_pmu_del() and relied on cpuc->active_mask/pebs_enabled checks. However, handle_pmi_common() can still encounter a NULL cpuc->events[] entry despite the active bit being set. Add an explicit NULL check on the event pointer before using it, to cover this legitimate scenario and avoid the NULL dereference crash. Fixes: 7e772a93eb61 ("perf/x86: Fix NULL event access and potential PEBS record loss") Reported-by: kitta <kitta@linux.alibaba.com> Co-developed-by: kitta <kitta@linux.alibaba.com> Signed-off-by: Evan Li <evan.li@linux.alibaba.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://patch.msgid.link/20251212084943.2124787-1-evan.li@linux.alibaba.com Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220855
2025-12-12Input: lkkbd - disable pending work before freeing deviceMinseong Kim
lkkbd_interrupt() schedules lk->tq via schedule_work(), and the work handler lkkbd_reinit() dereferences the lkkbd structure and its serio/input_dev fields. lkkbd_disconnect() and error paths in lkkbd_connect() free the lkkbd structure without preventing the reinit work from being queued again until serio_close() returns. This can allow the work handler to run after the structure has been freed, leading to a potential use-after-free. Use disable_work_sync() instead of cancel_work_sync() to ensure the reinit work cannot be re-queued, and call it both in lkkbd_disconnect() and in lkkbd_connect() error paths after serio_open(). Signed-off-by: Minseong Kim <ii4gsp@gmail.com> Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20251212052314.16139-1-ii4gsp@gmail.com Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2025-12-12x86/boot/Documentation: Fix whitespace noise in boot.rstIngo Molnar
There's a lot of unnecessary whitespace damage in this file: space before tabs, etc., that has no formatting or readability effect or advantages. Fix them. Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://patch.msgid.link/176535283007.498.16442167388418039352.tip-bot2@tip-bot2
2025-12-11drm/xe/doc: Add documentation for Multi Queue Group GuC interfaceNiranjana Vishwanathapura
Add kernel documentation for Multi Queue group GuC interface. Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-36-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/doc: Add documentation for Multi Queue GroupNiranjana Vishwanathapura
Add kernel documentation for Multi Queue group and update the corresponding rst. Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-35-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/multi_queue: Support active group after primary is destroyedNiranjana Vishwanathapura
Add support to keep the group active after the primary queue is destroyed. Instead of killing the primary queue during exec_queue destroy ioctl, kill it when all the secondary queues of the group are killed. Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-34-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/multi_queue: Tracepoint supportNiranjana Vishwanathapura
Add xe_exec_queue_create_multi_queue event with multi-queue information. Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-33-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/multi_queue: Teardown group upon job timeoutNiranjana Vishwanathapura
Upon a job timeout, teardown the multi-queue group by triggering TDR on all queues of the multi-queue group and by skipping timeout checks in them. v5: Ban the group while triggering TDR for the guc reported errors Add FIXME in TDR to take multi-queue group off HW (Matt Brost) v6: Trigger cleanup of group only for multi-queue case Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-32-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/multi_queue: Reset GT upon CGP_SYNC failureNiranjana Vishwanathapura
If GuC doesn't response to CGP_SYNC message, trigger GT reset and cleanup of all the queues of the multi queue group. Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-31-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/multi_queue: Handle CGP context errorNiranjana Vishwanathapura
Trigger multi-queue context cleanup upon CGP context error notification from GuC. v4: Fix error message Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-30-niranjana.vishwanathapura@intel.com
2025-12-11drm/xe/multi_queue: Set QUEUE_DRAIN_MODE for Multi Queue batchesNiranjana Vishwanathapura
To properly support soft light restore between batches being arbitrated at the CFEG, PIPE_CONTROL instructions have a new bit in the first DW, QUEUE_DRAIN_MODE. When set, this indicates to the CFEG that it should only drain the current queue. Additionally we no longer want to set the CS_STALL bit for these multi queue queues as this causes the entire pipeline to stall waiting for completion of the prior batch, preventing this soft light restore from occurring between queues in a queue group. v4: Assert !multi_queue where applicable (Matt Roper) Bspec: 56551 Signed-off-by: Stuart Summers <stuart.summers@intel.com> Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patch.msgid.link/20251211010249.1647839-29-niranjana.vishwanathapura@intel.com
generated by cgit v1.2.3 (git 2.25.1) at 2026-05-22 15:06:25 +0000