1. 06 3月, 2021 26 次提交
  2. 05 3月, 2021 14 次提交
    • J
      io_uring: make SQPOLL thread parking saner · 86e0d676
      Jens Axboe 提交于
      We have this weird true/false return from parking, and then some of the
      callers decide to look at that. It can lead to unbalanced parks and
      sqd locking. Have the callers check the thread status once it's parked.
      We know we have the lock at that point, so it's either valid or it's NULL.
      
      Fix race with parking on thread exit. We need to be careful here with
      ordering of the sdq->lock and the IO_SQ_THREAD_SHOULD_PARK bit.
      
      Rename sqd->completion to sqd->parked to reflect that this is the only
      thing this completion event doesn.
      Signed-off-by: NJens Axboe <axboe@kernel.dk>
      86e0d676
    • J
      io-wq: kill hashed waitqueue before manager exits · 09ca6c40
      Jens Axboe 提交于
      If we race with shutting down the io-wq context and someone queueing
      a hashed entry, then we can exit the manager with it armed. If it then
      triggers after the manager has exited, we can have a use-after-free where
      io_wqe_hash_wake() attempts to wake a now gone manager process.
      
      Move the killing of the hashed write queue into the manager itself, so
      that we know we've killed it before the task exits.
      
      Fixes: e941894e ("io-wq: make buffered file write hashed work map per-ctx")
      Signed-off-by: NJens Axboe <axboe@kernel.dk>
      09ca6c40
    • J
      io_uring: clear IOCB_WAITQ for non -EIOCBQUEUED return · b5b0ecb7
      Jens Axboe 提交于
      The callback can only be armed, if we get -EIOCBQUEUED returned. It's
      important that we clear the WAITQ bit for other cases, otherwise we can
      queue for async retry and filemap will assume that we're armed and
      return -EAGAIN instead of just blocking for the IO.
      
      Cc: stable@vger.kernel.org # 5.9+
      Signed-off-by: NJens Axboe <axboe@kernel.dk>
      b5b0ecb7
    • J
      io_uring: don't keep looping for more events if we can't flush overflow · ca0a2651
      Jens Axboe 提交于
      It doesn't make sense to wait for more events to come in, if we can't
      even flush the overflow we already have to the ring. Return -EBUSY for
      that condition, just like we do for attempts to submit with overflow
      pending.
      
      Cc: stable@vger.kernel.org # 5.11
      Signed-off-by: NJens Axboe <axboe@kernel.dk>
      ca0a2651
    • J
      io_uring: move to using create_io_thread() · 46fe18b1
      Jens Axboe 提交于
      This allows us to do task creation and setup without needing to use
      completions to try and synchronize with the starting thread. Get rid of
      the old io_wq_fork_thread() wrapper, and the 'wq' and 'worker' startup
      completion events - we can now do setup before the task is running.
      Signed-off-by: NJens Axboe <axboe@kernel.dk>
      46fe18b1
    • R
      Merge branch 'powercap' · 7bff4c26
      Rafael J. Wysocki 提交于
      * powercap:
        powercap/drivers/dtpm: Add the experimental label to the option description
        powercap/drivers/dtpm: Fix root node initialization
      7bff4c26
    • M
      nvmet: model_number must be immutable once set · d9f273b7
      Max Gurtovoy 提交于
      In case we have already established connection to nvmf target, it
      shouldn't be allowed to change the model_number. E.g. if someone will
      identify ctrl and get model_number of "my_model" later on will change
      the model_numbel via configfs to "my_new_model" this will break the NVMe
      specification for "Get Log Page – Persistent Event Log" that refers to
      Model Number as: "This field contains the same value as reported in the
      Model Number field of the Identify Controller data structure, bytes
      63:24."
      
      Although it doesn't mentioned explicitly that this field can't be
      changed, we can assume it.
      
      So allow setting this field only once: using configfs or in the first
      identify ctrl operation.
      Signed-off-by: NMax Gurtovoy <mgurtovoy@nvidia.com>
      Signed-off-by: NChristoph Hellwig <hch@lst.de>
      d9f273b7
    • M
      nvme-fabrics: fix kato initialization · 32feb6de
      Martin George 提交于
      Currently kato is initialized to NVME_DEFAULT_KATO for both
      discovery & i/o controllers. This is a problem specifically
      for non-persistent discovery controllers since it always ends
      up with a non-zero kato value. Fix this by initializing kato
      to zero instead, and ensuring various controllers are assigned
      appropriate kato values as follows:
      
      non-persistent controllers  - kato set to zero
      persistent controllers      - kato set to NVMF_DEV_DISC_TMO
                                    (or any positive int via nvme-cli)
      i/o controllers             - kato set to NVME_DEFAULT_KATO
                                    (or any positive int via nvme-cli)
      Signed-off-by: NMartin George <marting@netapp.com>
      Signed-off-by: NChristoph Hellwig <hch@lst.de>
      32feb6de
    • D
      nvme-hwmon: Return error code when registration fails · 78570f88
      Daniel Wagner 提交于
      The hwmon pointer wont be NULL if the registration fails. Though the
      exit code path will assign it to ctrl->hwmon_device. Later
      nvme_hwmon_exit() will try to free the invalid pointer. Avoid this by
      returning the error code from hwmon_device_register_with_info().
      
      Fixes: ed7770f6 ("nvme/hwmon: rework to avoid devm allocation")
      Signed-off-by: NDaniel Wagner <dwagner@suse.de>
      Signed-off-by: NChristoph Hellwig <hch@lst.de>
      78570f88
    • P
      nvme-pci: add quirks for Lexar 256GB SSD · 6e6a6828
      Pascal Terjan 提交于
      Add the NVME_QUIRK_NO_NS_DESC_LIST and NVME_QUIRK_IGNORE_DEV_SUBNQN
      quirks for this buggy device.
      
      Reported and tested in https://bugs.mageia.org/show_bug.cgi?id=28417Signed-off-by: NPascal Terjan <pterjan@google.com>
      Signed-off-by: NChristoph Hellwig <hch@lst.de>
      6e6a6828
    • Z
      nvme-pci: mark Kingston SKC2000 as not supporting the deepest power state · dc22c1c0
      Zoltán Böszörményi 提交于
      My 2TB SKC2000 showed the exact same symptoms that were provided
      in 538e4a8c ("nvme-pci: avoid the deepest sleep state on
      Kingston A2000 SSDs"), i.e. a complete NVME lockup that needed
      cold boot to get it back.
      
      According to some sources, the A2000 is simply a rebadged
      SKC2000 with a slightly optimized firmware.
      
      Adding the SKC2000 PCI ID to the quirk list with the same workaround
      as the A2000 made my laptop survive a 5 hours long Yocto bootstrap
      buildfest which reliably triggered the SSD lockup previously.
      Signed-off-by: NZoltán Böszörményi <zboszor@gmail.com>
      Signed-off-by: NChristoph Hellwig <hch@lst.de>
      dc22c1c0
    • J
      nvme-pci: mark Seagate Nytro XM1440 as QUIRK_NO_NS_DESC_LIST. · 5e112d3f
      Julian Einwag 提交于
      The kernel fails to fully detect these SSDs, only the character devices
      are present:
      
      [   10.785605] nvme nvme0: pci function 0000:04:00.0
      [   10.876787] nvme nvme1: pci function 0000:81:00.0
      [   13.198614] nvme nvme0: missing or invalid SUBNQN field.
      [   13.198658] nvme nvme1: missing or invalid SUBNQN field.
      [   13.206896] nvme nvme0: Shutdown timeout set to 20 seconds
      [   13.215035] nvme nvme1: Shutdown timeout set to 20 seconds
      [   13.225407] nvme nvme0: 16/0/0 default/read/poll queues
      [   13.233602] nvme nvme1: 16/0/0 default/read/poll queues
      [   13.239627] nvme nvme0: Identify Descriptors failed (8194)
      [   13.246315] nvme nvme1: Identify Descriptors failed (8194)
      
      Adding the NVME_QUIRK_NO_NS_DESC_LIST fixes this problem.
      
      BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=205679Signed-off-by: NJulian Einwag <jeinwag-nvme@marcapo.com>
      Signed-off-by: NChristoph Hellwig <hch@lst.de>
      Reviewed-by: NKeith Busch <kbusch@kernel.org>
      5e112d3f
    • L
      Merge tag 'drm-fixes-2021-03-05' of git://anongit.freedesktop.org/drm/drm · 280d542f
      Linus Torvalds 提交于
      Pull drm fixes from Dave Airlie:
       "More may show up but this is what I have at this stage: just a single
        nouveau regression fix, and a bunch of amdgpu fixes.
      
        amdgpu:
         - S0ix fix
         - Handle new NV12 SKU
         - Misc power fixes
         - Display uninitialized value fix
         - PCIE debugfs register access fix
      
        nouveau:
         - regression fix for gk104"
      
      * tag 'drm-fixes-2021-03-05' of git://anongit.freedesktop.org/drm/drm:
        drm/amdgpu: fix parameter error of RREG32_PCIE() in amdgpu_regs_pcie
        drm/amd/display: fix the return of the uninitialized value in ret
        drm/amdgpu: enable BACO runpm by default on sienna cichlid and navy flounder
        drm/amd/pm: correct Arcturus mmTHM_BACO_CNTL register address
        drm/amdgpu/swsmu/vangogh: Only use RLCPowerNotify msg for disable
        drm/amdgpu/pm: make unsupported power profile messages debug
        drm/amdgpu:disable VCN for Navi12 SKU
        drm/amdgpu: Only check for S0ix if AMD_PMC is configured
        drm/nouveau/fifo/gk104-gp1xx: fix creation of sw class
      280d542f
    • L
      Merge tag 'mkp-scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi · 44195bd7
      Linus Torvalds 提交于
      Pull iSCSI fixes from Martin Petersen:
       "Three fixes for missed iSCSI verification checks (and make the sysfs
        files use "sysfs_emit()" - that's what it is there for)"
      
      * tag 'mkp-scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi:
        scsi: iscsi: Verify lengths on passthrough PDUs
        scsi: iscsi: Ensure sysfs attributes are limited to PAGE_SIZE
        scsi: iscsi: Restrict sessions and handles to admin capabilities
      44195bd7
反馈
建议
客服 返回
顶部