* [PATCH v2] blk: optimization for classic polling
[not found] ` <CGME20180209104227epcas5p26e19e332c00996a777a2c3b73a347405@epcas5p2.samsung.com>
@ 2018-02-09 21:12 ` Nitesh Shetty
0 siblings, 0 replies; only message in thread
From: Nitesh Shetty @ 2018-02-09 21:12 UTC (permalink / raw)
Cc: axboe, linux-block, linux-kernel, joshi.k, Nitesh Shetty
From: Nitesh Shetty <nj.shetty@samsung.com>
This removes the dependency on interrupts to wake up task. Set task
state as TASK_RUNNING, if need_resched() returns true,
while polling for IO completion.
Earlier, polling task used to sleep, relying on interrupt to wake it up.
This made some IO take very long when interrupt-coalescing is enabled in
NVMe.
Changes v1->v2:
-setting task state once in blk_poll, instead of multiple callers.
Signed-off-by: Nitesh Shetty <nj.shetty@samsung.com>
---
block/blk-mq.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index df93102..40285fe 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -3164,6 +3164,7 @@ static bool __blk_mq_poll(struct blk_mq_hw_ctx *hctx, struct request *rq)
cpu_relax();
}
+ set_current_state(TASK_RUNNING);
return false;
}
--
2.7.4
^ permalink raw reply [flat|nested] only message in thread