On 1/14/19 12:21 PM, Liu Bo wrote: > Our test reported the following stack, and vmcore showed that > ->inflight counter is -1. > > [ffffc9003fcc38d0] __schedule at ffffffff8173d95d > [ffffc9003fcc3958] schedule at ffffffff8173de26 > [ffffc9003fcc3970] io_schedule at ffffffff810bb6b6 > [ffffc9003fcc3988] blkcg_iolatency_throttle at ffffffff813911cb > [ffffc9003fcc3a20] rq_qos_throttle at ffffffff813847f3 > [ffffc9003fcc3a48] blk_mq_make_request at ffffffff8137468a > [ffffc9003fcc3b08] generic_make_request at ffffffff81368b49 > [ffffc9003fcc3b68] submit_bio at ffffffff81368d7d > [ffffc9003fcc3bb8] ext4_io_submit at ffffffffa031be00 [ext4] > [ffffc9003fcc3c00] ext4_writepages at ffffffffa03163de [ext4] > [ffffc9003fcc3d68] do_writepages at ffffffff811c49ae > [ffffc9003fcc3d78] __filemap_fdatawrite_range at ffffffff811b6188 > [ffffc9003fcc3e30] filemap_write_and_wait_range at ffffffff811b6301 > [ffffc9003fcc3e60] ext4_sync_file at ffffffffa030cee8 [ext4] > [ffffc9003fcc3ea8] vfs_fsync_range at ffffffff8128594b > [ffffc9003fcc3ee8] do_fsync at ffffffff81285abd > [ffffc9003fcc3f18] sys_fsync at ffffffff81285d50 > [ffffc9003fcc3f28] do_syscall_64 at ffffffff81003c04 > [ffffc9003fcc3f50] entry_SYSCALL_64_after_swapgs at ffffffff81742b8e > > The ->inflight counter may be negative (-1) if > > 0) blk-throttle had been enabled when the IO was issued, so its bio > has a associated blkg, > > 1) blk-iolatency was disabled when the IO was issued, so iolatency_grp > in this blkg was not available by then, > > 2) blk-iolatency was enabled before this IO reached its endio, so that > iolatency_grp became available when the IO did the endio. > > 3) the ->inflight counter is decreased from 0 to -1. > > This uses atomic_dec_is_positive() instead to avoid the negative > inflight counter. The problem with that is that it'll hide a lot of other issues, too. Any way we can either track if this rqw is in flight, and only dec if it is, or quiesce when enabling? -- Jens Axboe
