[Qemu-devel] [PATCH 12/12] kvm: fix 80000001.EDX supported bit filtering

[Marcelo Tosatti]

From: Gleb Natapov <address@hidden>

On AMD some bits from 1.EDX are reported in 80000001.EDX. The mask used
to copy bits from 1.EDX to 80000001.EDX is incorrect resulting in
unsupported features passed into a guest.

Signed-off-by: Gleb Natapov <address@hidden>
Signed-off-by: Marcelo Tosatti <address@hidden>
---
 target-i386/kvm.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 676aa60..d6b12ed 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -111,7 +111,7 @@ uint32_t kvm_arch_get_supported_cpuid(CPUState *env, 
uint32_t function, int reg)
                      * so add missing bits according to the AMD spec:
                      */
                     cpuid_1_edx = kvm_arch_get_supported_cpuid(env, 1, R_EDX);
-                    ret |= cpuid_1_edx & 0xdfeff7ff;
+                    ret |= cpuid_1_edx & 0x183f7ff;
                     break;
                 }
                 break;
-- 
1.6.6.1