[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] [PATCH] x86emul: no need to set fault_suppression to false for VMOVNT*



When evex.opmsk is required to be zero there's no need for this, as it
won't have been set to true in the first place.

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
I've made similar changes to patches 16, 45, and 46 of v7 of the main
AVX512 series.

--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -5911,7 +5911,6 @@ x86_emulate(
     CASE_SIMD_PACKED_FP(_EVEX, 0x0f, 0x2b): /* vmovntp{s,d} [xyz]mm,mem */
         generate_exception_if(ea.type != OP_MEM || evex.opmsk, EXC_UD);
         sfence = true;
-        fault_suppression = false;
         /* fall through */
     CASE_SIMD_PACKED_FP(_EVEX, 0x0f, 0x10): /* vmovup{s,d} 
[xyz]mm/mem,[xyz]mm{k} */
     CASE_SIMD_SCALAR_FP(_EVEX, 0x0f, 0x10): /* vmovs{s,d} mem,xmm{k} */
@@ -6793,7 +6792,6 @@ x86_emulate(
         generate_exception_if(ea.type != OP_MEM || evex.opmsk || evex.w,
                               EXC_UD);
         sfence = true;
-        fault_suppression = false;
         /* fall through */
     case X86EMUL_OPC_EVEX_66(0x0f, 0x6f): /* vmovdqa{32,64} 
[xyz]mm/mem,[xyz]mm{k} */
     case X86EMUL_OPC_EVEX_F3(0x0f, 0x6f): /* vmovdqu{32,64} 
[xyz]mm/mem,[xyz]mm{k} */



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.