Re: [PATCH 2/2] perf/x86/intel/uncore: With > 8 nodes, get pci bus die id from NUMA info

2021-01-12 Thread Steve Wahl
On Tue, Jan 12, 2021 at 04:07:15PM +0100, Peter Zijlstra wrote: > On Mon, Jan 11, 2021 at 09:45:16AM -0600, Steve Wahl wrote: > > On Mon, Jan 11, 2021 at 02:00:33PM +0100, Peter Zijlstra wrote: > > > On Fri, Jan 08, 2021 at 09:35:49AM -0600, Steve Wahl wrote: > > > > > > > > > > + /

Re: [PATCH 2/2] perf/x86/intel/uncore: With > 8 nodes, get pci bus die id from NUMA info

2021-01-12 Thread Peter Zijlstra
On Mon, Jan 11, 2021 at 09:45:16AM -0600, Steve Wahl wrote: > On Mon, Jan 11, 2021 at 02:00:33PM +0100, Peter Zijlstra wrote: > > On Fri, Jan 08, 2021 at 09:35:49AM -0600, Steve Wahl wrote: > > > > > > > + /* > > > + * The nodeid and idmap registers only contain enough > > > +

Re: [PATCH 2/2] perf/x86/intel/uncore: With > 8 nodes, get pci bus die id from NUMA info

2021-01-11 Thread Steve Wahl
On Mon, Jan 11, 2021 at 02:00:33PM +0100, Peter Zijlstra wrote: > On Fri, Jan 08, 2021 at 09:35:49AM -0600, Steve Wahl wrote: > > > > + /* > > +* The nodeid and idmap registers only contain enough > > +* information to handle 8 nodes. On systems with more > > +

Re: [PATCH 2/2] perf/x86/intel/uncore: With > 8 nodes, get pci bus die id from NUMA info

2021-01-11 Thread Peter Zijlstra
On Fri, Jan 08, 2021 at 09:35:49AM -0600, Steve Wahl wrote: > + /* > + * The nodeid and idmap registers only contain enough > + * information to handle 8 nodes. On systems with more > + * than 8 nodes, we need to rely on NUMA information, > +

[PATCH 2/2] perf/x86/intel/uncore: With > 8 nodes, get pci bus die id from NUMA info

2021-01-08 Thread Steve Wahl
The registers used to determine which die a pci bus belongs to don't contain enough information to uniquely specify more than 8 dies, so when more than 8 dies are present, use NUMA information instead. Continue to use the previous method for 8 or fewer because it works there, and covers cases of N