| Message ID | 20180727211407.24318-1-josef@toxicpanda.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | block: don't do cgroup accounting for split bio's | expand |
On 7/27/18 3:14 PM, Josef Bacik wrote: > We need to check in blkcg_bio_issue_check if the bio is flagged as > QUEUE_ENTERED, because if it is then we've already accounted for the IO > in the cgroup stats. That's a bit of a flag overload, I think that at least warrants a comment.
On Fri, Jul 27, 2018 at 03:17:11PM -0600, Jens Axboe wrote: > On 7/27/18 3:14 PM, Josef Bacik wrote: > > We need to check in blkcg_bio_issue_check if the bio is flagged as > > QUEUE_ENTERED, because if it is then we've already accounted for the IO > > in the cgroup stats. > > That's a bit of a flag overload, I think that at least warrants a > comment. Yeah I'm not a fan of it, but there's nothing else to indicate that the bio is a split bio other than checking if it's endio is the chain endio. I'll respin with the comment. Thanks, Josef
diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index 3bed5e02a873..a977c31688ae 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -767,7 +767,8 @@ static inline bool blkcg_bio_issue_check(struct request_queue *q, throtl = blk_throtl_bio(q, blkg, bio); - if (!throtl) { + if (!throtl && + !bio_flagged(bio, BIO_QUEUE_ENTERED)) { blkg = blkg ?: q->root_blkg; blkg_rwstat_add(&blkg->stat_bytes, bio->bi_opf, bio->bi_iter.bi_size);
We need to check in blkcg_bio_issue_check if the bio is flagged as QUEUE_ENTERED, because if it is then we've already accounted for the IO in the cgroup stats. Reported-by: Tejun Heo <tj@kernel.org> Signed-off-by: Josef Bacik <josef@toxicpanda.com> --- include/linux/blk-cgroup.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)