Re: [patch]blk-mq: blk_mq_tag_to_rq should handle flush request

[Ming Lei]

On Thu, Jun 5, 2014 at 12:33 AM, Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote:
> On Wed, Jun 04, 2014 at 10:28:40AM -0600, Jens Axboe wrote:
>> > Attachment patch should be enough.
>>
>> Yep, looks fine. Care to send as a proper patch, then I will include it?
>
> I think we an also drop the assignment to q->flush_rq->cmd_flags in
> flush_end_io now.
>

Yes, that is already done, :-)

Jens, attachment is the formal one, thanks.

Thanks,
-- 
Ming Lei
From 77eb2e96942deed528b2e5892a7d9aa3735c0c67 Mon Sep 17 00:00:00 2001
From: Ming Lei <tom.leiming@xxxxxxxxx>
Date: Thu, 5 Jun 2014 00:23:55 +0800
Subject: [PATCH] block: mq flush: clear flush_rq's tag in flush_end_io()

So that blk_mq_tag_to_rq() can decide if one request
is flush_rq by comparing the tag.

Signed-off-by: Ming Lei <tom.leiming@xxxxxxxxx>
---
 block/blk-flush.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block/blk-flush.c b/block/blk-flush.c
index ff87c66..8ffee4b 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -225,7 +225,7 @@ static void flush_end_io(struct request *flush_rq, int error)
 
 	if (q->mq_ops) {
 		spin_lock_irqsave(&q->mq_flush_lock, flags);
-		q->flush_rq->cmd_flags = 0;
+		q->flush_rq->tag = -1;
 	}
 
 	running = &q->flush_queue[q->flush_running_idx];
-- 
1.7.9.5