[Xen-changelog] [xen master] x86/levelling: Restrict non-architectural OSXSAVE handling to emulated CPUID

[patchbot@xxxxxxx]

commit 3b7cac5232012e167b284aba738fef1eceda33f8
Author:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
AuthorDate: Tue Aug 23 11:10:12 2016 +0100
Commit:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
CommitDate: Thu Sep 1 11:41:03 2016 +0100

    x86/levelling: Restrict non-architectural OSXSAVE handling to emulated CPUID
    
    There is no need to extend the workaround to the faulted CPUID view, as
    Linux's dependence on the workaround is stricly via the emulated view.
    
    This causes a guest kernel faulted CPUID to observe architectural behaviour
    with respect to its CR4.OSXSAVE setting.
    
    Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
    Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
---
 xen/arch/x86/traps.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/xen/arch/x86/traps.c b/xen/arch/x86/traps.c
index c4a0f7a..ce924d8 100644
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -1003,6 +1003,8 @@ void pv_cpuid(struct cpu_user_regs *regs)
              *
              * Therefore, the leaking of Xen's OSXSAVE setting has become a
              * defacto part of the PV ABI and can't reasonably be corrected.
+             * It can however be restricted to only the enlightened CPUID
+             * view, as seen by the guest kernel.
              *
              * The following situations and logic now applies:
              *
@@ -1016,14 +1018,18 @@ void pv_cpuid(struct cpu_user_regs *regs)
              *
              * - Enlightened CPUID or CPUID faulting available:
              *    Xen can fully control what is seen here.  Guest kernels need
-             *    to see the leaked OSXSAVE, but guest userspace is given
-             *    architectural behaviour, to reflect the guest kernels
-             *    intentions.
+             *    to see the leaked OSXSAVE via the enlightened path, but
+             *    guest userspace and the native is given architectural
+             *    behaviour.
+             *
+             *    Emulated vs Faulted CPUID is distinguised based on whether a
+             *    #UD or #GP is currently being serviced.
              */
             /* OSXSAVE cleared by pv_featureset.  Fast-forward CR4 back in. */
-            if ( (guest_kernel_mode(curr, regs) &&
-                  (read_cr4() & X86_CR4_OSXSAVE)) ||
-                 (curr->arch.pv_vcpu.ctrlreg[4] & X86_CR4_OSXSAVE) )
+            if ( (curr->arch.pv_vcpu.ctrlreg[4] & X86_CR4_OSXSAVE) ||
+                 (regs->entry_vector == TRAP_invalid_op &&
+                  guest_kernel_mode(curr, regs) &&
+                  (read_cr4() & X86_CR4_OSXSAVE)) )
                 c |= cpufeat_mask(X86_FEATURE_OSXSAVE);
 
             /*
--
generated by git-patchbot for /home/xen/git/xen.git#master

_______________________________________________
Xen-changelog mailing list
Xen-changelog@xxxxxxxxxxxxx
https://lists.xenproject.org/xen-changelog