[Cbe-oss-dev] ps3 kernel lockdep issue?

Michael Ellerman michael at ellerman.id.au
Thu Oct 29 00:30:06 EST 2009


On Wed, 2009-10-28 at 11:22 +0200, Siarhei Siamashka wrote:
> On Tuesday 20 October 2009, Michael Ellerman wrote:
> > On Mon, 2009-10-19 at 11:06 -0700, Geoff Levand wrote:
> > > On 10/17/2009 06:29 AM, Siarhei Siamashka wrote:
> > > > Hi,
> > > >
> > > > I have the following stuff reported in dmesg log when booting ps3,
> > > > using 2.6.32-rc4 kernel. But checking older kernels (ex. 2.6.29) shows
> > > > that the issue has been there too. Any plans/ideas how to fix it?
> > >
> > > Please feel free to investigate.  I think you can get more
> > > help on the Linuxppc-dev mailing list for this:
> >
> > ?
> 
> ?

That wasn't directed at you :)

> > ps3_smp_setup_cpu() is causing a GFP_KERNEL allocation (via
> > request_irq()) but interrupts are still off that early in cpu bringup.
> 
> Please elaborate. Does it mean that this dmesg report message is a false
> positive? Or does it mean that this is a known problem (or a mostly harmless
> code ugliness) which is going to be fixed?

No it's not a false positive. Though it is mostly harmless. The only
case (I think) where it would cause real trouble is if the kernel needed
to do I/O in order to do the allocation, which this early in boot is
unlikely.

> I had some occasional hangs when trying to reboot or shutdown my PS3. I 
> suspect it could be caused by some rare race condition. That's why I'm a
> bit suspicious about different things.

I don't see any way this could be related to problems at shutdown.

cheers

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
URL: <http://lists.ozlabs.org/pipermail/cbe-oss-dev/attachments/20091029/b87df7d1/attachment.pgp>


More information about the cbe-oss-dev mailing list