2019-11-05 14:27:36

by Trond Myklebust

Subject: [PATCH] SUNRPC: Avoid RPC delays when exiting suspend

Jon Hunter: "I have been tracking down another suspend/NFS related
issue where again I am seeing random delays exiting suspend. The delays
can be up to a couple minutes in the worst case and this is causing a
suspend test we have to fail."

Change the use of a deferrable work to a standard delayed one.

Reported-by: Jon Hunter <[email protected]>
Fixes: 7e0a0e38fcfea ("SUNRPC: Replace the queue timer with a delayed work function")
Signed-off-by: Trond Myklebust <[email protected]>
---
net/sunrpc/sched.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 360afe153193..987c4b1f0b17 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -260,7 +260,7 @@ static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const c
rpc_reset_waitqueue_priority(queue);
queue->qlen = 0;
queue->timer_list.expires = 0;
- INIT_DEFERRABLE_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
+ INIT_DELAYED_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
INIT_LIST_HEAD(&queue->timer_list.list);
rpc_assign_waitqueue_name(queue, qname);
}
--
2.23.0

2019-11-06 11:16:05

by Jon Hunter

Subject: Re: [PATCH] SUNRPC: Avoid RPC delays when exiting suspend

On 05/11/2019 14:21, Trond Myklebust wrote:
> Jon Hunter: "I have been tracking down another suspend/NFS related
> issue where again I am seeing random delays exiting suspend. The delays
> can be up to a couple minutes in the worst case and this is causing a
> suspend test we have to fail."
>
> Change the use of a deferrable work to a standard delayed one.
>
> Reported-by: Jon Hunter <[email protected]>
> Fixes: 7e0a0e38fcfea ("SUNRPC: Replace the queue timer with a delayed work function")
> Signed-off-by: Trond Myklebust <[email protected]>
> ---
> net/sunrpc/sched.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
> index 360afe153193..987c4b1f0b17 100644
> --- a/net/sunrpc/sched.c
> +++ b/net/sunrpc/sched.c
> @@ -260,7 +260,7 @@ static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const c
> rpc_reset_waitqueue_priority(queue);
> queue->qlen = 0;
> queue->timer_list.expires = 0;
> - INIT_DEFERRABLE_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
> + INIT_DELAYED_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
> INIT_LIST_HEAD(&queue->timer_list.list);
> rpc_assign_waitqueue_name(queue, qname);
> }

Thanks!

Tested-by: Jon Hunter <[email protected]>

Cheers
Jon

--
nvpublic

2019-11-18 10:22:42

by Jon Hunter

Subject: Re: [PATCH] SUNRPC: Avoid RPC delays when exiting suspend

Hi Trond,

On 06/11/2019 11:15, Jon Hunter wrote:
>
> On 05/11/2019 14:21, Trond Myklebust wrote:
>> Jon Hunter: "I have been tracking down another suspend/NFS related
>> issue where again I am seeing random delays exiting suspend. The delays
>> can be up to a couple minutes in the worst case and this is causing a
>> suspend test we have to fail."
>>
>> Change the use of a deferrable work to a standard delayed one.
>>
>> Reported-by: Jon Hunter <[email protected]>
>> Fixes: 7e0a0e38fcfea ("SUNRPC: Replace the queue timer with a delayed work function")
>> Signed-off-by: Trond Myklebust <[email protected]>
>> ---
>> net/sunrpc/sched.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
>> index 360afe153193..987c4b1f0b17 100644
>> --- a/net/sunrpc/sched.c
>> +++ b/net/sunrpc/sched.c
>> @@ -260,7 +260,7 @@ static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const c
>> rpc_reset_waitqueue_priority(queue);
>> queue->qlen = 0;
>> queue->timer_list.expires = 0;
>> - INIT_DEFERRABLE_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
>> + INIT_DELAYED_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
>> INIT_LIST_HEAD(&queue->timer_list.list);
>> rpc_assign_waitqueue_name(queue, qname);
>> }
>
> Thanks!
>
> Tested-by: Jon Hunter <[email protected]>

I see this is now applied in -next, but I am seeing the failures on
mainline. Any chance we could still get this into v5.4?

Cheers
Jon

--
nvpublic