Messages in this thread | | | Date | Tue, 10 Mar 2015 19:37:13 -0300 | From | Marcelo Tosatti <> | Subject | Re: [PATCH] x86: svm: make wbinvd faster |
| |
On Tue, Mar 10, 2015 at 12:01:31PM +0100, Radim Krčmář wrote: > 2015-03-09 20:28-0300, Marcelo Tosatti: > > On Fri, Feb 27, 2015 at 06:19:18PM -0600, Joel Schopp wrote: > > > From: David Kaplan <David.Kaplan@amd.com> > > > No need to re-decode WBINVD since we know what it is from the intercept. > > > > > > Signed-off-by: David Kaplan <David.Kaplan@amd.com> > > > [extracted from larger unlrelated patch, forward ported, tested] > > > Signed-off-by: Joel Schopp <joel.schopp@amd.com> > > > > Can't you disable the intercept if need_emulate_wbinvd(vcpu) == false? > > I don't think we want to: it should be faster to intercept and ignore > than to invalidate all caches. The exit doesn't affect other physical > cores and costs just about 10(?) L3 cache misses.
Yes, right.
| |