Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-17630

osc_lru_shrink() should not block scheduling for long

Details

    • Improvement
    • Resolution: Fixed
    • Minor
    • Lustre 2.16.0
    • None
    • None
    • 3
    • 9223372036854775807

    Description

      osc_lru_shrink() can take quite long as the number of cached pages can be high. in such a case another threads may starve for CPU cycles and the kernel will be printing warnings:

      Nov  9 03:11:25 foxtrot3 kernel: NMI watchdog: BUG: soft lockup - CPU#23 stuck for 22s! [ptlrpcd_01_10:3531]
      ...
      Mar  1 14:03:47 woodbass46 kernel: Call Trace:
      Mar  1 14:03:47 woodbass46 kernel: _raw_spin_lock+0x1c/0x20
      Mar  1 14:03:47 woodbass46 kernel: osc_page_delete+0x2f2/0x470 [osc]
      Mar  1 14:03:47 woodbass46 kernel: cl_page_delete0+0x71/0x1f0 [obdclass]
      Mar  1 14:03:47 woodbass46 kernel: cl_page_delete+0x2a/0xf0 [obdclass]
      Mar  1 14:03:47 woodbass46 kernel: discard_pagevec+0x6f/0x150 [osc]
      Mar  1 14:03:47 woodbass46 kernel: osc_lru_shrink+0x80c/0x840 [osc]
      Mar  1 14:03:47 woodbass46 kernel: lru_queue_work+0xee/0x220 [osc]
      Mar  1 14:03:47 woodbass46 kernel: work_interpreter+0x32/0x110 [ptlrpc]
      Mar  1 14:03:47 woodbass46 kernel: ptlrpc_check_set.part.29+0x41d/0x1e80 [ptlrpc]
      Mar  1 14:03:47 woodbass46 kernel: ? __switch_to+0x183/0x480
      Mar  1 14:03:47 woodbass46 kernel: ptlrpcd_check+0x3d5/0x5b0 [ptlrpc]
      Mar  1 14:03:47 woodbass46 kernel: ptlrpcd+0x374/0x4b0 [ptlrpc]
      

      Attachments

        Issue Links

          Activity

            [LU-17630] osc_lru_shrink() should not block scheduling for long
            pjones Peter Jones added a comment -

            Merged for 2.16

            pjones Peter Jones added a comment - Merged for 2.16

            "Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/54346/
            Subject: LU-17630 osc: add cond_resched() to osc_lru_shrink()
            Project: fs/lustre-release
            Branch: master
            Current Patch Set:
            Commit: 69eb7b89c7f36ec6a8970e87fc8859207f4b9c0c

            gerrit Gerrit Updater added a comment - "Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/54346/ Subject: LU-17630 osc: add cond_resched() to osc_lru_shrink() Project: fs/lustre-release Branch: master Current Patch Set: Commit: 69eb7b89c7f36ec6a8970e87fc8859207f4b9c0c

            Oleg observed that a similar problem was hit in LU-16343.

            adilger Andreas Dilger added a comment - Oleg observed that a similar problem was hit in LU-16343 .

            "Alex Zhuravlev <bzzz@whamcloud.com>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/54346
            Subject: LU-17630 osc: add cond_resched() to osc_lru_shrink()
            Project: fs/lustre-release
            Branch: master
            Current Patch Set: 1
            Commit: 0646bb533293e89f6f42c4dbcd82739e135a17a7

            gerrit Gerrit Updater added a comment - "Alex Zhuravlev <bzzz@whamcloud.com>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/54346 Subject: LU-17630 osc: add cond_resched() to osc_lru_shrink() Project: fs/lustre-release Branch: master Current Patch Set: 1 Commit: 0646bb533293e89f6f42c4dbcd82739e135a17a7

            People

              bzzz Alex Zhuravlev
              bzzz Alex Zhuravlev
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: