* Babu Moger (babu.mo...@amd.com) wrote:
>
> > -----Original Message-----
> > From: Igor Mammedov <imamm...@redhat.com>
> > Sent: Wednesday, August 26, 2020 8:31 AM
> > To: Daniel P. Berrangé <berra...@redhat.com>
> > Cc: Moger, Babu <babu.mo...@amd.com>; pbonz...@redhat.com;
> > r...@twiddle.net; ehabk...@redhat.com; qemu-devel@nongnu.org;
> > m...@redhat.com; Michal Privoznik <mpriv...@redhat.com>
> > Subject: Re: [PATCH v5 0/8] Remove EPYC mode apicid decode and use generic
> > decode
> >
> > On Wed, 26 Aug 2020 13:50:59 +0100
> > Daniel P. Berrangé <berra...@redhat.com> wrote:
> >
> > > On Wed, Aug 26, 2020 at 02:38:49PM +0200, Igor Mammedov wrote:
> > > > On Fri, 21 Aug 2020 17:12:19 -0500
> > > > Babu Moger <babu.mo...@amd.com> wrote:
> > > >
> > > > > To support some of the complex topology, we introduced EPYC mode
> > apicid decode.
> > > > > But, EPYC mode decode is running into problems. Also it can become
> > > > > quite a maintenance problem in the future. So, it was decided to
> > > > > remove that code and use the generic decode which works for
> > > > > majority of the topology. Most of the SPECed configuration would
> > > > > work just fine. With some non-SPECed user inputs, it will create some
> > > > > sub-
> > optimal configuration.
> > > > > Here is the discussion thread.
> > > > > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > > > lore.kernel.org%2Fqemu-devel%2Fc0bcc1a6-1d84-a6e7-e468-
> > d5b437c1b25
> > > > >
> > 4%40amd.com%2F&data=02%7C01%7Cbabu.moger%40amd.com%7C8a5c
> > 52f92
> > > > >
> > 3f04082a40808d849c43d49%7C3dd8961fe4884e608e11a82d994e183d%7C0%7
> > C0
> > > > >
> > %7C637340454473508873&sdata=VnW28H1v4XwK3GaNGFxu%2BhwiMeA
> > YO%2B
> > > > > 3WAzo3DeY5Ha8%3D&reserved=0
> > > > >
> > > > > This series removes all the EPYC mode specific apicid changes and
> > > > > use the generic apicid decode.
> > > >
> > > > the main difference between EPYC and all other CPUs is that it
> > > > requires numa configuration (it's not optional) so we need an extra
> No, That is not true. Because of that assumption we made all these apicid
> changes. And here we are now.
>
> AMD supports varies mixed configurations. In case of EPYC-Rome, we have
> NPS1, NPS2 and NPS4(Numa Nodes per socket). In case of NPS1, basically we
> have all the cores in a socket under one numa node. This is non-numa
> configuration.
> Looking at the various configurations and also discussing internally, it
> is not advisable to have (epyc && !numa) check.
Indeed on real hardware, I don't think we always see NUMA; my single
socket, 16 core/32 thread 7302P Dell box, shows the kernel printing
'No NUMA configuration found...Faking a node.'
So if real hardware hasn't got a NUMA node, what's the real problem?
Dave
> > > > patch on top of this series to enfoce that, i.e:
> > > >
> > > > if (epyc && !numa)
> > > > error("EPYC cpu requires numa to be configured")
> > >
> > > Please no. This will break 90% of current usage of the EPYC CPU in
> > > real world QEMU deployments. That is way too user hostile to introduce
> > > as a requirement.
> > >
> > > Why do we need to force this ? People have been successfuly using
> > > EPYC CPUs without NUMA in QEMU for years now.
> > >
> > > It might not match behaviour of bare metal silicon, but that hasn't
> > > obviously caused the world to come crashing down.
> > So far it produces warning in linux kernel (RHBZ1728166), (resulting
> > performance
> > might be suboptimal), but I haven't seen anyone reporting crashes yet.
> >
> >
> > What other options do we have?
> > Perhaps we can turn on strict check for new machine types only, so old
> > configs
> > can keep broken topology (CPUID), while new ones would require -numa and
> > produce correct topology.
> >
> >
> > >
> > > Regards,
> > > Daniel
>
>
--
Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
