The attached patch adds the support of XSAVE/XRSTOR infrastructure for x86. XSAVE/XRSTOR manages the existing and future processor extended states on x86 architecture. XSAVE/XRSTOR related stuff is defined in Intel SDMs (The latest version is Nov. 2008): http://www.intel.com/products/processor/manuals/ The patch uses the classical CR0.TS based lazy algorithm. For now only FPU/SSE states (The bit 0 and 1 of XCR0) is supported. Signed-off-by: Dexuan Cui <dexuan.cui@intel.com> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On 26/11/08 09:43, "Cui, Dexuan" <dexuan.cui@intel.com> wrote:> The attached patch adds the support of XSAVE/XRSTOR infrastructure for x86. > XSAVE/XRSTOR manages the existing and future processor extended states on x86 > architecture. > > XSAVE/XRSTOR related stuff is defined in Intel SDMs (The latest version is > Nov. 2008): > http://www.intel.com/products/processor/manuals/ > > The patch uses the classical CR0.TS based lazy algorithm. > For now only FPU/SSE states (The bit 0 and 1 of XCR0) is supported. > > Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>Ah, XSAVE is back. Again, as of right now, it''s useful how? -- Keir _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
>>> "Cui, Dexuan" <dexuan.cui@intel.com> 26.11.08 10:43 >>> >The attached patch adds the support of XSAVE/XRSTOR infrastructure for x86. >XSAVE/XRSTOR manages the existing and future processor extended states on x86 architecture. > >XSAVE/XRSTOR related stuff is defined in Intel SDMs (The latest version is Nov. 2008): >http://www.intel.com/products/processor/manuals/ > >The patch uses the classical CR0.TS based lazy algorithm. >For now only FPU/SSE states (The bit 0 and 1 of XCR0) is supported. > >Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>I have to admit that I can''t see how save/restore is being taken care of. After all, the extra state logically would require vcpu_guest_context to be extended, but as that''s not possible, an alternative solution must be provided. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel