blkcg, writeback: Add wbc->no_cgroup_owner

When writeback IOs are bounced through async layers, the IOs should
only be accounted against the wbc from the original bdi writeback to
avoid confusing cgroup inode ownership arbitration.  Add
wbc->no_cgroup_owner to allow disabling wbc cgroup owner accounting.
This will be used make btrfs compression work well with cgroup IO
control.

v2: Renamed from no_wbc_acct to no_cgroup_owner and added comment as
    per Jan.

Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Tejun Heo 2019-06-27 13:39:50 -07:00 committed by Jens Axboe
parent 34e51a5e1a
commit 27b36d8fa8
2 changed files with 10 additions and 1 deletions

View File

@ -727,7 +727,7 @@ void wbc_account_cgroup_owner(struct writeback_control *wbc, struct page *page,
* behind a slow cgroup. Ultimately, we want pageout() to kick off
* regular writeback instead of writing things out itself.
*/
if (!wbc->wb)
if (!wbc->wb || wbc->no_cgroup_owner)
return;
css = mem_cgroup_css_from_page(page);

View File

@ -68,6 +68,15 @@ struct writeback_control {
unsigned for_reclaim:1; /* Invoked from the page allocator */
unsigned range_cyclic:1; /* range_start is cyclic */
unsigned for_sync:1; /* sync(2) WB_SYNC_ALL writeback */
/*
* When writeback IOs are bounced through async layers, only the
* initial synchronous phase should be accounted towards inode
* cgroup ownership arbitration to avoid confusion. Later stages
* can set the following flag to disable the accounting.
*/
unsigned no_cgroup_owner:1;
#ifdef CONFIG_CGROUP_WRITEBACK
struct bdi_writeback *wb; /* wb this writeback is issued under */
struct inode *inode; /* inode being written out */