Frank van der Linden
2009-Jan-20 23:36 UTC
[Xen-devel] [PATCH] Move smp_num_siblings to the per-cpu data
Make the num_siblings CPU parameter per-CPU. While it is unlikely that a system has a different number of siblings for different physical CPUs, make this parameter per-CPU for consistency, and deal with the (so far theoretical) case that this is different per CPU package. Signed-off-by: Frank van der Linden <Frank.Vanderlinden@Sun.COM> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Keir Fraser
2009-Jan-21 11:50 UTC
Re: [Xen-devel] [PATCH] Move smp_num_siblings to the per-cpu data
On 20/01/2009 23:36, "Frank van der Linden" <Frank.Vanderlinden@Sun.COM> wrote:> Make the num_siblings CPU parameter per-CPU. > > While it is unlikely that a system has a different number of siblings > for different physical CPUs, make this parameter per-CPU for > consistency, and deal with the (so far theoretical) case that this is > different per CPU package. > > Signed-off-by: Frank van der Linden <Frank.Vanderlinden@Sun.COM>It''s a pretty odd thing to decide to knock together one day. Do you have an underlying near-term motive for adding this flexibility? -- Keir _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Frank Van Der Linden
2009-Jan-21 16:18 UTC
Re: [Xen-devel] [PATCH] Move smp_num_siblings to the per-cpu data
Keir Fraser wrote:> On 20/01/2009 23:36, "Frank van der Linden" <Frank.Vanderlinden@Sun.COM> > wrote: > > >> Make the num_siblings CPU parameter per-CPU. >> >> While it is unlikely that a system has a different number of siblings >> for different physical CPUs, make this parameter per-CPU for >> consistency, and deal with the (so far theoretical) case that this is >> different per CPU package. >> >> Signed-off-by: Frank van der Linden <Frank.Vanderlinden@Sun.COM> >> > > It''s a pretty odd thing to decide to knock together one day. Do you have an > underlying near-term motive for adding this flexibility? > > -- Keir > > >I originally did this as part of an extension to the MCE support hypercall. We needed some detailed physical CPU information, plus some MSR info. When implementing this, and extracting infro from cpu_data[], I noticed the smp_num_siblings inconsistency, so I changed it. The other changes will be submitted as a separate patch, after they''ve been discussed with AMD & Intel. - Frank _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel