On 8/25/22 9:25 PM, gumi@linux.alibaba.com wrote:

On 8/25/22 00:09, Gu Mi wrote:

quoted

The problem occurs in two async processes, One is when a new IO calls 
the blk_mq_start_request() interface to start sending,The other is 
that the block layer timer process calls the blk_mq_req_expired 
interface to check whether there is an IO timeout.

When an instruction out of sequence occurs between blk_add_timer and 
WRITE_ONCE(rq->state,MQ_RQ_IN_FLIGHT) in the interface 
blk_mq_start_request,at this time, the block timer is checking the new 
IO timeout, Since the req status has been set to MQ_RQ_IN_FLIGHT and 
req->deadline is 0 at this time, the new IO will be misjudged as a 
timeout.

Our repair plan is for the deadline to be 0, and we do not think that 
a timeout occurs. At the same time, because the jiffies of the 32-bit 
system will be reversed shortly after the system is turned on, we will 
add 1 jiffies to the deadline at this time.

Signed-off-by: Gu Mi <redacted>
---
v1->v2:

time_after_eq() can handle the overflow, so remove the change on 
32-bit in blk_add_timer().

  block/blk-mq.c | 2 ++
  1 file changed, 2 insertions(+)

diff --git a/block/blk-mq.c b/block/blk-mq.c index 4b90d2d..6defaa1 

100644

--- a/block/blk-mq.c
+++ b/block/blk-mq.c

@@ -1451,6 +1451,8 @@ static bool blk_mq_req_expired(struct request *rq, unsigned long *next)
  		return false;
  
  	deadline = READ_ONCE(rq->deadline);
+	if (unlikely(deadline == 0))
+		return false;
  	if (time_after_eq(jiffies, deadline))
  		return true;
  

rq->deadline == 0 can be a valid deadline value so the above patch
doesn't look right to me.

Gu, you need to fix your quoting of emails, these are impossible to
read.

That aside, I think there's a misunderstanding here. v1 has some
parts and v2 has others. Please post a v3 that has the hunk
that guarantees that deadline always has the lowest bit set if
assigned, and the !deadline check as well.

-- 
Jens Axboe