Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1339970 > unrolled thread

[PATCH V2 12/13] blk-throttle: restore shrinked cgroup share

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH V2 12/13] blk-throttle: restore shrinked cgroup share Shaohua Li <shli@fb.com> - 2016-02-22 23:10 +0100

#1339970 — [PATCH V2 12/13] blk-throttle: restore shrinked cgroup share

If a cgroup share gets shrinked but the cgroup hits its limit, the
shrink isn't optimal, we restore its original share.

Signed-off-by: Shaohua Li <shli@fb.com>
---
 block/blk-throttle.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 2271c7e..7db7d8e 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -2229,6 +2229,7 @@ bool blk_throtl_bio(struct request_queue *q, struct blkcg_gq *blkg,
 
 	sq = &tg->service_queue;
 
+again:
 	tg_check_new_weight(tg);
 	detect_inactive_cg(tg);
 	while (true) {
@@ -2238,8 +2239,18 @@ bool blk_throtl_bio(struct request_queue *q, struct blkcg_gq *blkg,
 		tg_update_perf(tg);
 
 		/* if above limits, break to queue */
-		if (!tg_may_dispatch(tg, bio, NULL))
+		if (!tg_may_dispatch(tg, bio, NULL)) {
+			/*
+			 * If the cg hits limit but its share is shrinked, the
+			 * shrink isn't optimal
+			 */
+			if (sq->acting_weight < sq->weight) {
+				/* pretend we are changing weight */
+				sq->new_weight = sq->weight;
+				goto again;
+			}
 			break;
+		}
 
 		/* within limits, let's charge and dispatch directly */
 		throtl_charge_bio(tg, bio);
-- 
2.6.5

[toc] | [standalone]

Back to top | Article view | linux.kernel

csiph-web