From: Yu Kuai <yukuai3@xxxxxxxxxx>
While bdev_count_inflight is interating all cpus, if some IOs are issued
from traversed cpu and then completed from the cpu that is not traversed
yet:
cpu0
cpu1
bdev_count_inflight
//for_each_possible_cpu
// cpu0 is 0
infliht += 0
// issue a io
blk_account_io_start
// cpu0 inflight ++
cpu2
// the io is done
blk_account_io_done
// cpu2 inflight --
// cpu 1 is 0
inflight += 0
// cpu2 is -1
inflight += -1
...
In this case, the total inflight will be -1, causing lots of false
warning. Fix the problem by removing the warning.
Noted there is still a valid warning for nvme-mpath(From Yi) that is not
fixed yet.
Fixes: f5482ee5edb9 ("block: WARN if bdev inflight counter is negative")
Reported-by: Yi Zhang <yi.zhang@xxxxxxxxxx>
Closes: https://lore.kernel.org/linux-block/aFtUXy-lct0WxY2w@xxxxxxxxxxxxx/T/#mae89155a5006463d0a21a4a2c35ae0034b26a339
Reported-and-tested-by: Calvin Owens <calvin@xxxxxxxxxx>
Closes: https://lore.kernel.org/linux-block/aFtUXy-lct0WxY2w@xxxxxxxxxxxxx/T/#m1d935a00070bf95055d0ac84e6075158b08acaef
Reported-by: Dave Chinner <david@xxxxxxxxxxxxx>
Closes: https://lore.kernel.org/linux-block/aFuypjqCXo9-5_En@xxxxxxxxxxxxxxxxxxx/
Signed-off-by: Yu Kuai <yukuai3@xxxxxxxxxx>
---
block/genhd.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/block/genhd.c b/block/genhd.c
index 8171a6bc3210..680fa717082f 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -141,9 +141,14 @@ static void bdev_count_inflight_rw(struct block_device *part,
}
}
- if (WARN_ON_ONCE((int)inflight[READ] < 0))
+ /*
+ * While iterating all cpus, some IOs might issued from traversed cpu
+ * and then completed from the cpu that is not traversed yet, causing
+ * the inflight number to be negative.
+ */
+ if ((int)inflight[READ] < 0)
inflight[READ] = 0;
- if (WARN_ON_ONCE((int)inflight[WRITE] < 0))
+ if ((int)inflight[WRITE] < 0)
inflight[WRITE] = 0;
}
--
2.39.2
