Skip to contentSkip to conversation

KVM: x86: Swap the dst and src operand for MOVNTDQA

PATCH Superseded Sean Christopherson 1 participant 3 messages May 6

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Sean Christopherson

author maintainer
#0 lore raw

PATCH 1/1 KVM: x86: Swap the dst and src operand for MOVNTDQA

Swap the MOVNTDQA operands, as MOVNTDQA does NOT in fact have "the same characteristics as 0F E7 (MOVNTDQ)"; MOVNTDQA loads from memory and stores to registers, while MOVNTDQ loads from registers and stores to memory.

Per the SDM:

MOVNTDQ - Move packed integer values in xmm1 to m128 using non-temporal
hint.

MOVNTDQA - Move double quadword from m128 to xmm1 using non-temporal hint
if WC memory type.

Fixes c57d9bafbd0b ("KVM: x86: Add support for emulating MOVNTDQA")

---

I have updates to the xsave KUT test to verify the bug and the fix.

 arch/x86/kvm/emulate.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/emulate.c +2 -2
index c8c6cc0406d6..0bb832a76868 100644
@@ -4450,7 +4450,7 @@ static const struct opcode twobyte_table[256] = {
 	/* 0xD0 - 0xDF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
 	/* 0xE0 - 0xEF */
-	N, N, N, N, N, N, N, GP(SrcReg | DstMem | ModRM | Mov, &pfx_0f_e7_0f_38_2a),
+	N, N, N, N, N, N, N, GP(SrcMem | DstReg | ModRM | Mov, &pfx_0f_e7_0f_38_2a),
 	N, N, N, N, N, N, N, N,
 	/* 0xF0 - 0xFF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N
@@ -4481,7 +4481,7 @@ static const struct opcode opcode_map_0f_38[256] = {
 	X16(N), X16(N),
 	/* 0x20 - 0x2f */
 	X8(N),
-	X2(N), GP(SrcReg | DstMem | ModRM | Mov | Aligned, &pfx_0f_e7_0f_38_2a), N, N, N, N, N,
+	X2(N), GP(SrcMem | DstReg | ModRM | Mov | Aligned, &pfx_0f_e7_0f_38_2a), N, N, N, N, N,
 	/* 0x30 - 0x7f */
 	X16(N), X16(N), X16(N), X16(N), X16(N),
 	/* 0x80 - 0xef */

base-commit: 6d35786de28116ecf78797a62b84e6bf3c45aa5a

Signature · 1 line
2.54.0.545.g6539524ca2-goog
author maintainer
#1 lore raw
Quoted 28 lines · Sean Christopherson

Swap the MOVNTDQA operands, as MOVNTDQA does NOT in fact have "the same characteristics as 0F E7 (MOVNTDQ)"; MOVNTDQA loads from memory and stores to registers, while MOVNTDQ loads from registers and stores to memory.

Per the SDM:

MOVNTDQ - Move packed integer values in xmm1 to m128 using non-temporal
hint.

MOVNTDQA - Move double quadword from m128 to xmm1 using non-temporal hint
if WC memory type.

Fixes c57d9bafbd0b ("KVM: x86: Add support for emulating MOVNTDQA")

---

I have updates to the xsave KUT test to verify the bug and the fix.

 arch/x86/kvm/emulate.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/emulate.c +0 -0
index c8c6cc0406d6..0bb832a76868 100644

arch/x86/kvm/emulate.c static const struct opcode twobyte_table[256] = { · line 4453

Quoted 5 lines
arch/x86/kvm/emulate.c +1 -1
@@ -4450,7 +4450,7 @@ static const struct opcode twobyte_table[256] = {
 	/* 0xD0 - 0xDF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
 	/* 0xE0 - 0xEF */
-	N, N, N, N, N, N, N, GP(SrcReg | DstMem | ModRM | Mov, &pfx_0f_e7_0f_38_2a),
+	N, N, N, N, N, N, N, GP(SrcMem | DstReg | ModRM | Mov, &pfx_0f_e7_0f_38_2a),

Aaaaand Josh pointed out that I wasn't paying attention and stomped all over MOVNTDQ. I'll figure out why my KUT changes didn't detect this, and send a v2.

Quoted 17 lines
arch/x86/kvm/emulate.c +1 -1
 	N, N, N, N, N, N, N, N,
 	/* 0xF0 - 0xFF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N
@@ -4481,7 +4481,7 @@ static const struct opcode opcode_map_0f_38[256] = {
 	X16(N), X16(N),
 	/* 0x20 - 0x2f */
 	X8(N),
-	X2(N), GP(SrcReg | DstMem | ModRM | Mov | Aligned, &pfx_0f_e7_0f_38_2a), N, N, N, N, N,
+	X2(N), GP(SrcMem | DstReg | ModRM | Mov | Aligned, &pfx_0f_e7_0f_38_2a), N, N, N, N, N,
 	/* 0x30 - 0x7f */
 	X16(N), X16(N), X16(N), X16(N), X16(N),
 	/* 0x80 - 0xef */

base-commit: 6d35786de28116ecf78797a62b84e6bf3c45aa5a

Signature · 1 line
2.54.0.545.g6539524ca2-goog
author maintainer
#2 lore raw
Quoted 24 lines · Sean Christopherson
Quoted 23 lines · Sean Christopherson

Swap the MOVNTDQA operands, as MOVNTDQA does NOT in fact have "the same characteristics as 0F E7 (MOVNTDQ)"; MOVNTDQA loads from memory and stores to registers, while MOVNTDQ loads from registers and stores to memory.

Per the SDM:

MOVNTDQ - Move packed integer values in xmm1 to m128 using non-temporal
hint.

MOVNTDQA - Move double quadword from m128 to xmm1 using non-temporal hint
if WC memory type.

Fixes c57d9bafbd0b ("KVM: x86: Add support for emulating MOVNTDQA")

---

I have updates to the xsave KUT test to verify the bug and the fix.

 arch/x86/kvm/emulate.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
Quoted 13 lines

arch/x86/kvm/emulate.c static const struct opcode twobyte_table[256] = { · line 4453

Quoted 10 lines
arch/x86/kvm/emulate.c +1 -1
index c8c6cc0406d6..0bb832a76868 100644
@@ -4450,7 +4450,7 @@ static const struct opcode twobyte_table[256] = {
 	/* 0xD0 - 0xDF */
 	N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N,
 	/* 0xE0 - 0xEF */
-	N, N, N, N, N, N, N, GP(SrcReg | DstMem | ModRM | Mov, &pfx_0f_e7_0f_38_2a),
+	N, N, N, N, N, N, N, GP(SrcMem | DstReg | ModRM | Mov, &pfx_0f_e7_0f_38_2a),

Aaaaand Josh pointed out that I wasn't paying attention and stomped all over MOVNTDQ. I'll figure out why my KUT changes didn't detect this

*sigh*

I made the same mistake when testing MOVNTDQA, and didn't learn from that mistake for the MOVNTDQ half. Comparing the src and dst buffers on a "src => reg => dst" sequence wont't detect mismatches if KVM wrongly does "reg => src, reg => dst".