Re: 2.6.22-rc3 hibernate(?) fails totally - regression (xfs on raid6)

[David Chinner]

On Wed, Jun 13, 2007 at 12:16:36PM +0100, David Greaves wrote:
> Linus Torvalds wrote:
> >
> >On Fri, 8 Jun 2007, David Greaves wrote:
> >>positive: I can now get sysrq-t :)
> >
> >Ok, so color me confused,
> So what do you think that makes me <grin>
> 
> >and maybe I have missed some of the emails or 
> >skimmed over them too fast (there's been too many of them ;),
> 
> You may have missed these 'tests' with rc4+Tejun's fix:
> * clean boot, unmounting the xfs fs : normal hibernate/resume
> * clean boot, remount ro xfs fs : normal hibernate/resume
> * clean boot, touch; sync; echo 1 > /proc/sys/vm/drop_caches: normal 
> hibernate/resume
> * clean boot, touch; sync; echo 2 > /proc/sys/vm/drop_caches: hang 
> hibernating
> * clean boot, touch; sync; echo 3 > /proc/sys/vm/drop_caches: hang 
> hibernating
> 
> Dave asked me to do them but hasn't responded yet.

Sorry 'bout that. Bit busy ATM.

What I was effectively looking for was whether it was data or metadata
that was causing the problems. From the above, it would appear that
dropping the page cache (echo 1 > drop caches) allows a successful
hibernate/resume. Next step would have been to isolate which cache
being dropped made the difference (e.g. a file or a bdev cache?).

However, it is clear from the back traces that there is something
unwell with md/sata code, so I don't think this needs to be tracked
any further from the filesystem perspective.

Cheers,

Dave.
-- 
Dave Chinner
Principal Engineer
SGI Australian Software Group
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/