Messages in this thread | | | Date | Mon, 31 Oct 2016 15:22:36 -0400 | From | Yazen Ghannam <> | Subject | Re: [PATCH v2 2/2] x86/AMD: Group cpu_llc_id assignment by topology feature and family |
| |
> > > The NODEID_MSR feature only applies to Fam10h in which case the llc is at > > s/llc/LLC (Last Level Cache/ > > Let's try to have abbreviations written out in their first mention in the text. >
Okay.
> > the node level. > > > > The TOPOEXT feature is used on families 15h, 16h and 17h. So far we only > > see multiple last level caches if L3 caches are available. Otherwise, the > > cpu_llc_id will default to be the phys_proc_id. We have L3 caches only on > > families 15h and 17h. On Fam15h, the llc is at the node level. On Fam17h, > > s/llc/LLC/g >
Ack.
> > the llc is at the core complex level and can be found by right shifting the > ^^^ > > LLC >
Ack.
> > + > > + /* > > + * We may have multiple LLCs if L3 caches exist, so check if we > > + * have an L3 cache by looking at the L3 cache cpuid leaf. > > x86 instructions in caps please: CPUID >
Ack.
> > + */ > > + if (cpuid_edx(0x80000006)) { > > + if (c->x86 == 0x15) { > > + /* LLC is at the node level. */ > > + per_cpu(cpu_llc_id, cpu) = node_id; > > + > > + } else if (c->x86 == 0x17) { > > How about >= ? >
This APICID format is only valid for Fam17h. What I'm going for is that we fall back to a sensible default if we don't have a better assignment for a new family. At first I thought that phys_proc_id would be good but now I think node_id is better as a sensible default. I'll make this change in the V3 set.
> > Btw, please add for your next submission: > > Tested-by: Borislav Petkov <bp@suse.de> >
For both patches?
Thanks, Yazen
| |