Messages in this thread |  | | | Date | Tue, 11 Dec 2007 09:52:32 +0100 | | From | Ingo Molnar <> | | Subject | Re: tipc_init(), WARNING: at arch/x86/mm/highmem_32.c:52, [2.6.24-rc4-git5: Reported regressions from 2.6.23] |
* Dave Jones <davej@redhat.com> wrote:
> On Sat, Dec 08, 2007 at 08:52:11PM +0100, Ingo Molnar wrote:
>
> > so even today's upstream kernel, which has 'ancient' SLUB code, SLAB and
> > SLUB have essentially the same linecount:
> >
> > $ wc -l mm/slab.c mm/slub.c
> > 4478 mm/slab.c
> > 4125 mm/slub.c
> >
> > (and while linecount != complexity, there is a strong relationship.)
> >
> > With SLAB having 10 years more test coverage and tuning.
>
> FWIW, the one thing slub does that slab doesn't that I find really
> nice is being enable to enable debugging at boot time rather than
> compile time.
yes, but that's largely due to "dont change SLAB because we've got SLUB"
resistence to SLAB patches. It's a 2 minute hack to implement this for
SLAB.
> We don't get many people running benchmarks against the Fedora kernel,
> so any scalability differences between slub/slab probably won't reach
> us until we start shipping betas of the next RHEL based on the same
> kernel.
>
> Which leaves my only other gripe. It broke slabtop.
that's actually a _bad_ ABI regression. Rafael, could you please add
this to the regressions list?
> There's an alternative implementation in Documentation/vm/slabinfo.c
> (why there not say, util-linux, home of current slabtop?)
the kernel should output /proc/slabinfo data with the same formatting,
period.
Ingo
|  |