Re: [OOPS] 2.6.21-rc6-git5 in cfq_dispatch_insert

[Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>]

Hi,

On Tuesday 17 April 2007, Neil Brown wrote:
> On Monday April 16, cebbert@redhat.com wrote:
> > 
> > cfq_dispatch_insert() was called with rq == 0. This one is getting really
> > annoying... and md is involved again (RAID0 this time.)
> 
> Yeah... weird.
> RAID0 is so light-weight and so different from RAID1 or RAID5 that I
> feel fairly safe concluding that the problem isn't in or near md.
> But that doesn't help you.
> 
> This really feels like a locking problem.
> 
> The problem occurs when ->next_rq is NULL, but ->sort_list.rb_node is
> not NULL.  That happens plenty of times in the code (particularly as
> the first request is inserted) but always under ->queue_lock so it
> should never be visible to cfq_dispatch_insert..
> 
> Except that drivers/scsi/ide-scsi.c:idescsi_eh_reset calls
> elv_next_request which could ultimately call __cfq_dispatch_requests
> without taking ->queue_lock (that I can see).  But you probably aren't
> using ide-scsi (does anyone?).

ide-scsi is holding ide_lock while calling elv_next_request()
(for ide ide_lock == ->queue_lock)

Also from the original report:

On Sunday 15 April 2007, Brad Campbell wrote:
> 
> The box is booted with PXE and runs an nfsroot. It's Debian 3.1. It has 2 SIL 3112 controllers in it 
> with 4 WD 200GB ATA drives all on PATA->SATA bridges.

and you can even see libata functions in the OOPS...

Bart