lkml.org 
[lkml]   [2012]   [Jun]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: multi-second application stall in open()
Hey,

On Tue, Jun 26, 2012 at 03:37:31PM -0500, Josh Hunt wrote:
> The print "cfq schedule dispatch: busy_queues:1 rq_queued:1
> rq_in_driver:0 rq_in_flight[0]:0 [1]:0" is inside the if to call
> kblockd_schedule_work() so it is getting called, but you can see that
> we never see the workqueue_queue_work tracepoint being hit.

Hmm... looks like the work item is already queued. Can you please
move the "schedule dispatch" message after kblockd_schedule_work() and
print its return value too?

> Tejun - if you could let me know what other data I should be looking
> for/need to enable in the traces I would appreciate it.

The trace seems truncated and doesn't contain 9527.XXXX traces. Can
you please resend?

Thanks.

--
tejun


\
 
 \ /
  Last update: 2012-06-26 23:41    [W:0.710 / U:0.316 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site