[PATCH] blk-iocost: account flush sequence latency only once

[PATCH] blk-iocost: account flush sequence latency only once

[Zhe Liu]

A flush sequence with a DATA phase can complete its original request
twice. Both completions invoke ioc_rqos_done(), causing the request to
contribute two latency samples and adding the same queue wait time
twice. This gives such requests disproportionate weight compared to
ordinary writes, distorting the device-wide latency and queue wait
feedback used to adjust iocost's virtual time rate.

Skip requests marked RQF_FLUSH_SEQ in ioc_rqos_done(). The flag is set
when the request enters the flush machinery and is still set when the
data phase completes; blk_flush_restore_request() clears it before the
final completion, so accounting happens only once, after the whole
sequence finishes. The remaining sample covers the request's lifetime
through any postflush, and its queue wait time is accumulated only
once.

Fixes: 7caa47151ab2e ("blkcg: implement blk-iocost")
Cc: stable@vger.kernel.org
Signed-off-by: Zhe Liu <liuzhe1@kylinos.cn>
---
 block/blk-iocost.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 2ae40138e041..601029221e6f 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -2868,6 +2868,9 @@ static void ioc_rqos_done(struct rq_qos *rqos, struct request *rq)
 	if (!ioc->enabled || !rq->alloc_time_ns || !rq->start_time_ns)
 		return;
 
+	if (rq->rq_flags & RQF_FLUSH_SEQ)
+		return;
+
 	switch (req_op(rq)) {
 	case REQ_OP_READ:
 		pidx = QOS_RLAT;
-- 
2.25.1

Re: [PATCH] blk-iocost: account flush sequence latency only once

[Tejun Heo]

On Thu, Oct 29, 2026 at 05:24:24PM +0800, Zhe Liu wrote:
> +	if (rq->rq_flags & RQF_FLUSH_SEQ)
> +		return;

Can you please wrap the test in unlikely()?

Acked-by: Tejun Heo <tj@kernel.org>

Thanks.

--
tejun

Re: [PATCH] blk-iocost: account flush sequence latency only once

[Tao Cui]

Hi, Zhe

在 2026/10/29 17:24, Zhe Liu 写道:
> A flush sequence with a DATA phase can complete its original request
> twice. Both completions invoke ioc_rqos_done(), causing the request to
> contribute two latency samples and adding the same queue wait time
> twice. This gives such requests disproportionate weight compared to
> ordinary writes, distorting the device-wide latency and queue wait
> feedback used to adjust iocost's virtual time rate.
> 
> Skip requests marked RQF_FLUSH_SEQ in ioc_rqos_done(). The flag is set
> when the request enters the flush machinery and is still set when the
> data phase completes; blk_flush_restore_request() clears it before the
> final completion, so accounting happens only once, after the whole
> sequence finishes. The remaining sample covers the request's lifetime
> through any postflush, and its queue wait time is accumulated only
> once.
> 
> Fixes: 7caa47151ab2e ("blkcg: implement blk-iocost")
> Cc: stable@vger.kernel.org
> Signed-off-by: Zhe Liu <liuzhe1@kylinos.cn>
> ---
>  block/blk-iocost.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/block/blk-iocost.c b/block/blk-iocost.c
> index 2ae40138e041..601029221e6f 100644
> --- a/block/blk-iocost.c
> +++ b/block/blk-iocost.c
> @@ -2868,6 +2868,9 @@ static void ioc_rqos_done(struct rq_qos *rqos, struct request *rq)
>  	if (!ioc->enabled || !rq->alloc_time_ns || !rq->start_time_ns)
>  		return;
>  
> +	if (rq->rq_flags & RQF_FLUSH_SEQ)
> +		return;
> +
>  	switch (req_op(rq)) {
>  	case REQ_OP_READ:
>  		pidx = QOS_RLAT;

Reviewed-by: Tao Cui <cuitao@kylinos.cn>