diff options
| author | Jan Kara | 2016-06-28 09:04:01 +0200 |
|---|---|---|
| committer | Jens Axboe | 2016-06-28 16:21:48 +0200 |
| commit | 149321a611d5d41cebcf5f813a3bf45b3afe66ad (patch) | |
| tree | 55b2dcd151cd40b622c2c5e41d29d6cb1f38960e /net/appletalk | |
| parent | cfq-iosched: Convert slice_resid from u64 to s64 (diff) | |
| download | kernel-qcow2-linux-149321a611d5d41cebcf5f813a3bf45b3afe66ad.tar.gz kernel-qcow2-linux-149321a611d5d41cebcf5f813a3bf45b3afe66ad.tar.xz kernel-qcow2-linux-149321a611d5d41cebcf5f813a3bf45b3afe66ad.zip | |
cfq-iosched: Fix regression in bonnie++ rewrite performance
Commit 9a7f38c42c2 (cfq-iosched: Convert from jiffies to nanoseconds)
broke the condition for detecting starved sync IO in
cfq_completed_request() because rq->start_time remained in jiffies but
we compared it with nanosecond values. This manifested as a regression
in bonnie++ rewrite performance because we always ended up considering
sync IO starved and thus never increased async IO queue depth.
Since rq->start_time is used in a lot of places, converting it to ns
values would be non-trivial. So just revert the condition in CFQ to use
comparison with jiffies. This will lead to suboptimal results if
cfq_fifo_expire[1] will ever come close to 1 jiffie but so far we are
relatively far from that with the storage used with CFQ (the default
value is 128 ms).
Fixes: 9a7f38c42c2b92391d9dabaf9f51df7cfe5608e4
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Diffstat (limited to 'net/appletalk')
0 files changed, 0 insertions, 0 deletions