1. 02 Oct, 2020 1 commit
  2. 29 Sep, 2020 1 commit
    • yangerkun's avatar
      blk-mq: call commit_rqs while list empty but error happen · 632bfb63
      yangerkun authored
      Blk-mq should call commit_rqs once 'bd.last != true' and no more
      request will come(so virtscsi can kick the virtqueue, e.g.). We already
      do that in 'blk_mq_dispatch_rq_list/blk_mq_try_issue_list_directly' while
      list not empty and 'queued > 0'. However, we can seen the same scene
      once the last request in list call queue_rq and return error like
      BLK_STS_IOERR which will not requeue the request, and lead that list
      empty but need call commit_rqs too(Or the request for virtscsi will stay
      timeout until other request kick virtqueue).
      
      We found this problem by do fsstress test with offline/online virtscsi
      device repeat quickly.
      
      Fixes: d666ba98 ("blk-mq: add mq_ops->commit_rqs()")
      Reported-by: default avatarzhangyi (F) <yi.zhang@huawei.com>
      Signed-off-by: default avataryangerkun <yangerkun@huawei.com>
      Reviewed-by: default avatarMing Lei <ming.lei@redhat.com>
      Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
      632bfb63
  3. 25 Sep, 2020 1 commit
  4. 24 Sep, 2020 1 commit
    • Jens Axboe's avatar
      Merge tag 'nvme-5.9-2020-09-24' of git://git.infradead.org/nvme into block-5.9 · 9754d6cb
      Jens Axboe authored
      Pull NVMe fixes from Christoph:
      
      "nvme fixes for 5.9
      
        - fix error during controller probe that cause double free irqs
          (Keith Busch)
        - FC connection establishment fix (James Smart)
        - properly handle completions for invalid tags (Xianting Tian)
        - pass the correct nsid to the command effects and supported log
          (Chaitanya Kulkarni)"
      
      * tag 'nvme-5.9-2020-09-24' of git://git.infradead.org/nvme:
        nvme-core: don't use NVME_NSID_ALL for command effects and supported log
        nvme-fc: fail new connections to a deleted host or remote port
        nvme-pci: fix NULL req in completion handler
        nvme: return errors for hwmon init
      9754d6cb
  5. 23 Sep, 2020 1 commit
  6. 22 Sep, 2020 3 commits
  7. 17 Sep, 2020 3 commits
    • Jens Axboe's avatar
      Merge tag 'nvme-5.9-2020-09-17' of git://git.infradead.org/nvme into block-5.9 · 4a2dd2c7
      Jens Axboe authored
      Pull NVMe fixes from Christoph:
      
      "nvme fixes for 5.9
      
        - another quirk for the controller from hell (David Milburn)
        - fix a Kconfig dependency (Necip Fazil Yildiran)
        - char devices / passthrough refcount fixes (Chaitanya Kulkarni)"
      
      * tag 'nvme-5.9-2020-09-17' of git://git.infradead.org/nvme:
        nvmet: get transport reference for passthru ctrl
        nvme-core: get/put ctrl and transport module in nvme_dev_open/release()
        nvme-tcp: fix kconfig dependency warning when !CRYPTO
        nvme-pci: disable the write zeros command for Intel 600P/P3100
      4a2dd2c7
    • Christoph Hellwig's avatar
      nvmet: get transport reference for passthru ctrl · 3a6b0761
      Christoph Hellwig authored
      Grab a reference to the transport driver to ensure it can't be unloaded
      while a passthrough controller is active.
      
      Fixes: c1fef73f ("nvmet: add passthru code to process commands")
      Reported-by: default avatarChaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
      Signed-off-by: default avatarChristoph Hellwig <hch@lst.de>
      Reviewed-by: default avatarLogan Gunthorpe <logang@deltatee.com>
      3a6b0761
    • Chaitanya Kulkarni's avatar
      nvme-core: get/put ctrl and transport module in nvme_dev_open/release() · 52a3974f
      Chaitanya Kulkarni authored
      Get and put the reference to the ctrl in the nvme_dev_open() and
      nvme_dev_release() before and after module get/put for ctrl in char
      device file operations.
      
      Introduce char_dev relase function, get/put the controller and module
      which allows us to fix the potential Oops which can be easily reproduced
      with a passthru ctrl (although the problem also exists with pure user
      access):
      
      Entering kdb (current=0xffff8887f8290000, pid 3128) on processor 30 Oops: (null)
      due to oops @ 0xffffffffa01019ad
      CPU: 30 PID: 3128 Comm: bash Tainted: G        W  OE     5.8.0-rc4nvme-5.9+ #35
      Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.4
      RIP: 0010:nvme_free_ctrl+0x234/0x285 [nvme_core]
      Code: 57 10 a0 e8 73 bf 02 e1 ba 3d 11 00 00 48 c7 c6 98 33 10 a0 48 c7 c7 1d 57 10 a0 e8 5b bf 02 e1 8
      RSP: 0018:ffffc90001d63de0 EFLAGS: 00010246
      RAX: ffffffffa05c0440 RBX: ffff8888119e45a0 RCX: 0000000000000000
      RDX: 0000000000000000 RSI: ffff8888177e9550 RDI: ffff8888119e43b0
      RBP: ffff8887d4768000 R08: 0000000000000000 R09: 0000000000000000
      R10: 0000000000000000 R11: ffffc90001d63c90 R12: ffff8888119e43b0
      R13: ffff8888119e5108 R14: dead000000000100 R15: ffff8888119e5108
      FS:  00007f1ef27b0740(0000) GS:ffff888817600000(0000) knlGS:0000000000000000
      CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
      CR2: ffffffffa05c0470 CR3: 00000007f6bee000 CR4: 00000000003406e0
      Call Trace:
       device_release+0x27/0x80
       kobject_put+0x98/0x170
       nvmet_passthru_ctrl_disable+0x4a/0x70 [nvmet]
       nvmet_passthru_enable_store+0x4c/0x90 [nvmet]
       configfs_write_file+0xe6/0x150
       vfs_write+0xba/0x1e0
       ksys_write+0x5f/0xe0
       do_syscall_64+0x52/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xa9
      RIP: 0033:0x7f1ef1eb2840
      Code: Bad RIP value.
      RSP: 002b:00007fffdbff0eb8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
      RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007f1ef1eb2840
      RDX: 0000000000000002 RSI: 00007f1ef27d2000 RDI: 0000000000000001
      RBP: 00007f1ef27d2000 R08: 000000000000000a R09: 00007f1ef27b0740
      R10: 0000000000000001 R11: 0000000000000246 R12: 00007f1ef2186400
      R13: 0000000000000002 R14: 0000000000000001 R15: 0000000000000000
      
      With this patch fix we take the module ref count in nvme_dev_open() and
      release that ref count in newly introduced nvme_dev_release().
      Signed-off-by: default avatarChaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
      Signed-off-by: default avatarChristoph Hellwig <hch@lst.de>
      52a3974f
  8. 15 Sep, 2020 3 commits
  9. 10 Sep, 2020 1 commit
    • Jens Axboe's avatar
      Merge tag 'nvme-5.9-2020-09-10' of git://git.infradead.org/nvme into block-5.9 · fd04358e
      Jens Axboe authored
      Pull NVMe fixes from Christoph.
      
      "nvme fixes for 5.9
      
       - cancel async events before freeing them (David Milburn)
       - revert a broken race fix (James Smart)
       - fix command processing during resets (Sagi Grimberg)"
      
      * tag 'nvme-5.9-2020-09-10' of git://git.infradead.org/nvme:
        nvme-fabrics: allow to queue requests for live queues
        nvme-tcp: cancel async events before freeing event struct
        nvme-rdma: cancel async events before freeing event struct
        nvme-fc: cancel async events before freeing event struct
        nvme: Revert: Fix controller creation races with teardown flow
      fd04358e
  10. 09 Sep, 2020 2 commits
    • Ritesh Harjani's avatar
      block: Set same_page to false in __bio_try_merge_page if ret is false · 2cd896a5
      Ritesh Harjani authored
      If we hit the UINT_MAX limit of bio->bi_iter.bi_size and so we are anyway
      not merging this page in this bio, then it make sense to make same_page
      also as false before returning.
      
      Without this patch, we hit below WARNING in iomap.
      This mostly happens with very large memory system and / or after tweaking
      vm dirty threshold params to delay writeback of dirty data.
      
      WARNING: CPU: 18 PID: 5130 at fs/iomap/buffered-io.c:74 iomap_page_release+0x120/0x150
       CPU: 18 PID: 5130 Comm: fio Kdump: loaded Tainted: G        W         5.8.0-rc3 #6
       Call Trace:
        __remove_mapping+0x154/0x320 (unreliable)
        iomap_releasepage+0x80/0x180
        try_to_release_page+0x94/0xe0
        invalidate_inode_page+0xc8/0x110
        invalidate_mapping_pages+0x1dc/0x540
        generic_fadvise+0x3c8/0x450
        xfs_file_fadvise+0x2c/0xe0 [xfs]
        vfs_fadvise+0x3c/0x60
        ksys_fadvise64_64+0x68/0xe0
        sys_fadvise64+0x28/0x40
        system_call_exception+0xf8/0x1c0
        system_call_common+0xf0/0x278
      
      Fixes: cc90bc68 ("block: fix "check bi_size overflow before merge"")
      Reported-by: default avatarShivaprasad G Bhat <sbhat@linux.ibm.com>
      Suggested-by: default avatarChristoph Hellwig <hch@infradead.org>
      Signed-off-by: default avatarAnju T Sudhakar <anju@linux.vnet.ibm.com>
      Signed-off-by: default avatarRitesh Harjani <riteshh@linux.ibm.com>
      Reviewed-by: default avatarMing Lei <ming.lei@redhat.com>
      Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
      Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
      2cd896a5
    • Sagi Grimberg's avatar
      nvme-fabrics: allow to queue requests for live queues · 73a53799
      Sagi Grimberg authored
      Right now we are failing requests based on the controller state (which
      is checked inline in nvmf_check_ready) however we should definitely
      accept requests if the queue is live.
      
      When entering controller reset, we transition the controller into
      NVME_CTRL_RESETTING, and then return BLK_STS_RESOURCE for non-mpath
      requests (have blk_noretry_request set).
      
      This is also the case for NVME_REQ_USER for the wrong reason. There
      shouldn't be any reason for us to reject this I/O in a controller reset.
      We do want to prevent passthru commands on the admin queue because we
      need the controller to fully initialize first before we let user passthru
      admin commands to be issued.
      
      In a non-mpath setup, this means that the requests will simply be
      requeued over and over forever not allowing the q_usage_counter to drop
      its final reference, causing controller reset to hang if running
      concurrently with heavy I/O.
      
      Fixes: 35897b92 ("nvme-fabrics: fix and refine state checks in __nvmf_check_ready")
      Reviewed-by: default avatarJames Smart <james.smart@broadcom.com>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      Signed-off-by: default avatarChristoph Hellwig <hch@lst.de>
      73a53799
  11. 08 Sep, 2020 6 commits
  12. 03 Sep, 2020 1 commit
  13. 01 Sep, 2020 5 commits
  14. 29 Aug, 2020 1 commit
    • Jens Axboe's avatar
      Merge branch 'nvme-5.9-rc' of git://git.infradead.org/nvme into block-5.9 · 5d220bcd
      Jens Axboe authored
      Pull NVMe fixes from Sagi:
      
      "- instance leak and io boundary fixes from Keith
       - fc locking fix from Christophe
       - various tcp/rdma reset during traffic fixes from Me
       - pci use-after-free fix from Tong
       - tcp target null deref fix from Ziye"
      
      * 'nvme-5.9-rc' of git://git.infradead.org/nvme:
        nvme-pci: cancel nvme device request before disabling
        nvme: only use power of two io boundaries
        nvme: fix controller instance leak
        nvmet-fc: Fix a missed _irqsave version of spin_lock in 'nvmet_fc_fod_op_done()'
        nvme: Fix NULL dereference for pci nvme controllers
        nvme-rdma: fix reset hang if controller died in the middle of a reset
        nvme-rdma: fix timeout handler
        nvme-rdma: serialize controller teardown sequences
        nvme-tcp: fix reset hang if controller died in the middle of a reset
        nvme-tcp: fix timeout handler
        nvme-tcp: serialize controller teardown sequences
        nvme: have nvme_wait_freeze_timeout return if it timed out
        nvme-fabrics: don't check state NVME_CTRL_NEW for request acceptance
        nvmet-tcp: Fix NULL dereference when a connect data comes in h2cdata pdu
      5d220bcd
  15. 28 Aug, 2020 10 commits
    • Tong Zhang's avatar
      nvme-pci: cancel nvme device request before disabling · 7ad92f65
      Tong Zhang authored
      This patch addresses an irq free warning and null pointer dereference
      error problem when nvme devices got timeout error during initialization.
      This problem happens when nvme_timeout() function is called while
      nvme_reset_work() is still in execution. This patch fixed the problem by
      setting flag of the problematic request to NVME_REQ_CANCELLED before
      calling nvme_dev_disable() to make sure __nvme_submit_sync_cmd() returns
      an error code and let nvme_submit_sync_cmd() fail gracefully.
      The following is console output.
      
      [   62.472097] nvme nvme0: I/O 13 QID 0 timeout, disable controller
      [   62.488796] nvme nvme0: could not set timestamp (881)
      [   62.494888] ------------[ cut here ]------------
      [   62.495142] Trying to free already-free IRQ 11
      [   62.495366] WARNING: CPU: 0 PID: 7 at kernel/irq/manage.c:1751 free_irq+0x1f7/0x370
      [   62.495742] Modules linked in:
      [   62.495902] CPU: 0 PID: 7 Comm: kworker/u4:0 Not tainted 5.8.0+ #8
      [   62.496206] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-48-gd9c812dda519-p4
      [   62.496772] Workqueue: nvme-reset-wq nvme_reset_work
      [   62.497019] RIP: 0010:free_irq+0x1f7/0x370
      [   62.497223] Code: e8 ce 49 11 00 48 83 c4 08 4c 89 e0 5b 5d 41 5c 41 5d 41 5e 41 5f c3 44 89 f6 48 c70
      [   62.498133] RSP: 0000:ffffa96800043d40 EFLAGS: 00010086
      [   62.498391] RAX: 0000000000000000 RBX: ffff9b87fc458400 RCX: 0000000000000000
      [   62.498741] RDX: 0000000000000001 RSI: 0000000000000096 RDI: ffffffff9693d72c
      [   62.499091] RBP: ffff9b87fd4c8f60 R08: ffffa96800043bfd R09: 0000000000000163
      [   62.499440] R10: ffffa96800043bf8 R11: ffffa96800043bfd R12: ffff9b87fd4c8e00
      [   62.499790] R13: ffff9b87fd4c8ea4 R14: 000000000000000b R15: ffff9b87fd76b000
      [   62.500140] FS:  0000000000000000(0000) GS:ffff9b87fdc00000(0000) knlGS:0000000000000000
      [   62.500534] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
      [   62.500816] CR2: 0000000000000000 CR3: 000000003aa0a000 CR4: 00000000000006f0
      [   62.501165] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
      [   62.501515] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
      [   62.501864] Call Trace:
      [   62.501993]  pci_free_irq+0x13/0x20
      [   62.502167]  nvme_reset_work+0x5d0/0x12a0
      [   62.502369]  ? update_load_avg+0x59/0x580
      [   62.502569]  ? ttwu_queue_wakelist+0xa8/0xc0
      [   62.502780]  ? try_to_wake_up+0x1a2/0x450
      [   62.502979]  process_one_work+0x1d2/0x390
      [   62.503179]  worker_thread+0x45/0x3b0
      [   62.503361]  ? process_one_work+0x390/0x390
      [   62.503568]  kthread+0xf9/0x130
      [   62.503726]  ? kthread_park+0x80/0x80
      [   62.503911]  ret_from_fork+0x22/0x30
      [   62.504090] ---[ end trace de9ed4a70f8d71e2 ]---
      [  123.912275] nvme nvme0: I/O 12 QID 0 timeout, disable controller
      [  123.914670] nvme nvme0: 1/0/0 default/read/poll queues
      [  123.916310] BUG: kernel NULL pointer dereference, address: 0000000000000000
      [  123.917469] #PF: supervisor write access in kernel mode
      [  123.917725] #PF: error_code(0x0002) - not-present page
      [  123.917976] PGD 0 P4D 0
      [  123.918109] Oops: 0002 [#1] SMP PTI
      [  123.918283] CPU: 0 PID: 7 Comm: kworker/u4:0 Tainted: G        W         5.8.0+ #8
      [  123.918650] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-48-gd9c812dda519-p4
      [  123.919219] Workqueue: nvme-reset-wq nvme_reset_work
      [  123.919469] RIP: 0010:__blk_mq_alloc_map_and_request+0x21/0x80
      [  123.919757] Code: 66 0f 1f 84 00 00 00 00 00 41 55 41 54 55 48 63 ee 53 48 8b 47 68 89 ee 48 89 fb 8b4
      [  123.920657] RSP: 0000:ffffa96800043d40 EFLAGS: 00010286
      [  123.920912] RAX: ffff9b87fc4fee40 RBX: ffff9b87fc8cb008 RCX: 0000000000000000
      [  123.921258] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff9b87fc618000
      [  123.921602] RBP: 0000000000000000 R08: ffff9b87fdc2c4a0 R09: ffff9b87fc616000
      [  123.921949] R10: 0000000000000000 R11: ffff9b87fffd1500 R12: 0000000000000000
      [  123.922295] R13: 0000000000000000 R14: ffff9b87fc8cb200 R15: ffff9b87fc8cb000
      [  123.922641] FS:  0000000000000000(0000) GS:ffff9b87fdc00000(0000) knlGS:0000000000000000
      [  123.923032] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
      [  123.923312] CR2: 0000000000000000 CR3: 000000003aa0a000 CR4: 00000000000006f0
      [  123.923660] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
      [  123.924007] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
      [  123.924353] Call Trace:
      [  123.924479]  blk_mq_alloc_tag_set+0x137/0x2a0
      [  123.924694]  nvme_reset_work+0xed6/0x12a0
      [  123.924898]  process_one_work+0x1d2/0x390
      [  123.925099]  worker_thread+0x45/0x3b0
      [  123.925280]  ? process_one_work+0x390/0x390
      [  123.925486]  kthread+0xf9/0x130
      [  123.925642]  ? kthread_park+0x80/0x80
      [  123.925825]  ret_from_fork+0x22/0x30
      [  123.926004] Modules linked in:
      [  123.926158] CR2: 0000000000000000
      [  123.926322] ---[ end trace de9ed4a70f8d71e3 ]---
      [  123.926549] RIP: 0010:__blk_mq_alloc_map_and_request+0x21/0x80
      [  123.926832] Code: 66 0f 1f 84 00 00 00 00 00 41 55 41 54 55 48 63 ee 53 48 8b 47 68 89 ee 48 89 fb 8b4
      [  123.927734] RSP: 0000:ffffa96800043d40 EFLAGS: 00010286
      [  123.927989] RAX: ffff9b87fc4fee40 RBX: ffff9b87fc8cb008 RCX: 0000000000000000
      [  123.928336] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff9b87fc618000
      [  123.928679] RBP: 0000000000000000 R08: ffff9b87fdc2c4a0 R09: ffff9b87fc616000
      [  123.929025] R10: 0000000000000000 R11: ffff9b87fffd1500 R12: 0000000000000000
      [  123.929370] R13: 0000000000000000 R14: ffff9b87fc8cb200 R15: ffff9b87fc8cb000
      [  123.929715] FS:  0000000000000000(0000) GS:ffff9b87fdc00000(0000) knlGS:0000000000000000
      [  123.930106] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
      [  123.930384] CR2: 0000000000000000 CR3: 000000003aa0a000 CR4: 00000000000006f0
      [  123.930731] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
      [  123.931077] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
      Co-developed-by: default avatarKeith Busch <kbusch@kernel.org>
      Signed-off-by: default avatarTong Zhang <ztong0001@gmail.com>
      Reviewed-by: default avatarKeith Busch <kbusch@kernel.org>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      7ad92f65
    • Keith Busch's avatar
      nvme: only use power of two io boundaries · e83d776f
      Keith Busch authored
      The kernel requires a power of two for boundaries because that's the
      only way it can efficiently split commands that cross them. A
      controller, however, may report a non-power of two boundary.
      
      The driver had been rounding the controller's value to one the kernel
      can use, but splitting on the wrong boundary provides no benefit on the
      device side, and incurs additional submission overhead from non-optimal
      splits.
      
      Don't provide any boundary hint if the controller's value can't be used
      and log a warning when first scanning a disk's unreported IO boundary.
      Since the chunk sector logic has grown, move it to a separate function.
      
      Cc: Martin K. Petersen <martin.petersen@oracle.com>
      Signed-off-by: default avatarKeith Busch <kbusch@kernel.org>
      Reviewed-by: default avatarMartin K. Petersen <martin.petersen@oracle.com>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      e83d776f
    • Keith Busch's avatar
      nvme: fix controller instance leak · 192f6c29
      Keith Busch authored
      If the driver has to unbind from the controller for an early failure
      before the subsystem has been set up, there won't be a subsystem holding
      the controller's instance, so the controller needs to free its own
      instance in this case.
      
      Fixes: 733e4b69 ("nvme: Assign subsys instance from first ctrl")
      Signed-off-by: default avatarKeith Busch <kbusch@kernel.org>
      Reviewed-by: default avatarChaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
      Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      192f6c29
    • Christophe JAILLET's avatar
      nvmet-fc: Fix a missed _irqsave version of spin_lock in 'nvmet_fc_fod_op_done()' · 70e37988
      Christophe JAILLET authored
      The way 'spin_lock()' and 'spin_lock_irqsave()' are used is not consistent
      in this function.
      
      Use 'spin_lock_irqsave()' also here, as there is no guarantee that
      interruptions are disabled at that point, according to surrounding code.
      
      Fixes: a97ec51b ("nvmet_fc: Rework target side abort handling")
      Signed-off-by: default avatarChristophe JAILLET <christophe.jaillet@wanadoo.fr>
      Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      70e37988
    • Sagi Grimberg's avatar
      nvme: Fix NULL dereference for pci nvme controllers · 7cd49f75
      Sagi Grimberg authored
      PCIe controllers do not have fabric opts, verify they exist before
      showing ctrl_loss_tmo or reconnect_delay attributes.
      
      Fixes: 764075fd ("nvme: expose reconnect_delay and ctrl_loss_tmo via sysfs")
      Reported-by: default avatarTobias Markus <tobias@markus-regensburg.de>
      Reviewed-by: default avatarKeith Busch <kbusch@kernel.org>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      7cd49f75
    • Sagi Grimberg's avatar
      nvme-rdma: fix reset hang if controller died in the middle of a reset · 2362acb6
      Sagi Grimberg authored
      If the controller becomes unresponsive in the middle of a reset, we
      will hang because we are waiting for the freeze to complete, but that
      cannot happen since we have commands that are inflight holding the
      q_usage_counter, and we can't blindly fail requests that times out.
      
      So give a timeout and if we cannot wait for queue freeze before
      unfreezing, fail and have the error handling take care how to
      proceed (either schedule a reconnect of remove the controller).
      Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      2362acb6
    • Sagi Grimberg's avatar
      nvme-rdma: fix timeout handler · 0475a8dc
      Sagi Grimberg authored
      When a request times out in a LIVE state, we simply trigger error
      recovery and let the error recovery handle the request cancellation,
      however when a request times out in a non LIVE state, we make sure to
      complete it immediately as it might block controller setup or teardown
      and prevent forward progress.
      
      However tearing down the entire set of I/O and admin queues causes
      freeze/unfreeze imbalance (q->mq_freeze_depth) because and is really
      an overkill to what we actually need, which is to just fence controller
      teardown that may be running, stop the queue, and cancel the request if
      it is not already completed.
      
      Now that we have the controller teardown_lock, we can safely serialize
      request cancellation. This addresses a hang caused by calling extra
      queue freeze on controller namespaces, causing unfreeze to not complete
      correctly.
      Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
      Reviewed-by: default avatarJames Smart <james.smart@broadcom.com>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      0475a8dc
    • Sagi Grimberg's avatar
      nvme-rdma: serialize controller teardown sequences · 5110f402
      Sagi Grimberg authored
      In the timeout handler we may need to complete a request because the
      request that timed out may be an I/O that is a part of a serial sequence
      of controller teardown or initialization. In order to complete the
      request, we need to fence any other context that may compete with us
      and complete the request that is timing out.
      
      In this case, we could have a potential double completion in case
      a hard-irq or a different competing context triggered error recovery
      and is running inflight request cancellation concurrently with the
      timeout handler.
      
      Protect using a ctrl teardown_lock to serialize contexts that may
      complete a cancelled request due to error recovery or a reset.
      Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
      Reviewed-by: default avatarJames Smart <james.smart@broadcom.com>
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      5110f402
    • Sagi Grimberg's avatar
      nvme-tcp: fix reset hang if controller died in the middle of a reset · e5c01f4f
      Sagi Grimberg authored
      If the controller becomes unresponsive in the middle of a reset, we will
      hang because we are waiting for the freeze to complete, but that cannot
      happen since we have commands that are inflight holding the
      q_usage_counter, and we can't blindly fail requests that times out.
      
      So give a timeout and if we cannot wait for queue freeze before
      unfreezing, fail and have the error handling take care how to proceed
      (either schedule a reconnect of remove the controller).
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      e5c01f4f
    • Sagi Grimberg's avatar
      nvme-tcp: fix timeout handler · 236187c4
      Sagi Grimberg authored
      When a request times out in a LIVE state, we simply trigger error
      recovery and let the error recovery handle the request cancellation,
      however when a request times out in a non LIVE state, we make sure to
      complete it immediately as it might block controller setup or teardown
      and prevent forward progress.
      
      However tearing down the entire set of I/O and admin queues causes
      freeze/unfreeze imbalance (q->mq_freeze_depth) because and is really
      an overkill to what we actually need, which is to just fence controller
      teardown that may be running, stop the queue, and cancel the request if
      it is not already completed.
      
      Now that we have the controller teardown_lock, we can safely serialize
      request cancellation. This addresses a hang caused by calling extra
      queue freeze on controller namespaces, causing unfreeze to not complete
      correctly.
      Signed-off-by: default avatarSagi Grimberg <sagi@grimberg.me>
      236187c4