Re: NUMA aware slab allocator V2

[Andrew Morton]

Christoph Lameter <clameter@xxxxxxxxxxxx> wrote:
>
> On Thu, 12 May 2005, Andrew Morton wrote:
> 
> > Christoph Lameter <clameter@xxxxxxxxxxxx> wrote:
> > >
> > > This patch allows kmalloc_node to be as fast as kmalloc by introducing
> > >  node specific page lists for partial, free and full slabs.
> >
> > This patch causes the ppc64 G5 to lock up fairly early in boot.  It's
> > pretty much a default config:
> > http://www.zip.com.au/~akpm/linux/patches/stuff/config-pmac
> >
> > No serial port, no debug environment, but no useful-looking error messages
> > either.  See http://www.zip.com.au/~akpm/linux/patches/stuff/dsc02516.jpg
> 
> I got rc4-mm1 and booted it on an x86_64 machines with similar
> configuration (no NUMA but SMP, numa slab uncommented) but multiple
> configurations worked fine (apart from another error attempting to
> initialize a nonexistand second cpu by the NMI handler that I described
> in another email to you). I have no ppc64 available.
> 
> Could we boot the box without quiet so that we can get better debug
> messages?

OK, I'll try that, but I doubt if it'll give much more info.

> Did the box boot okay without the patch?

Yup, I tested base 2.6.12-rc4 and 2.6.12-rc4+the-patch-you-sent.
-
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/