* [PATCH] KVM: x86 emulator: fix pusha instruction emulation
@ 2010-06-15 1:00 Wei Yongjun
2010-06-15 1:03 ` [PATCHv2] " Wei Yongjun
2010-06-16 14:09 ` [PATCH] " Marcelo Tosatti
0 siblings, 2 replies; 12+ messages in thread
From: Wei Yongjun @ 2010-06-15 1:00 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti; +Cc: kvm
emulate pusha instruction only writeback the last
EDI register, but the other registers which need
to be writeback is ignored. This patch fixed it.
---
arch/x86/kvm/emulate.c | 133 ++++++++++++++++++++++++++----------------------
1 files changed, 73 insertions(+), 60 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index a4c2dcd..c990db0 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -1553,6 +1553,64 @@ exception:
return X86EMUL_PROPAGATE_FAULT;
}
+static inline int writeback(struct x86_emulate_ctxt *ctxt,
+ struct x86_emulate_ops *ops)
+{
+ int rc;
+ struct decode_cache *c = &ctxt->decode;
+ u32 err;
+
+ switch (c->dst.type) {
+ case OP_REG:
+ /* The 4-byte case *is* correct:
+ * in 64-bit mode we zero-extend.
+ */
+ switch (c->dst.bytes) {
+ case 1:
+ *(u8 *)c->dst.ptr = (u8)c->dst.val;
+ break;
+ case 2:
+ *(u16 *)c->dst.ptr = (u16)c->dst.val;
+ break;
+ case 4:
+ *c->dst.ptr = (u32)c->dst.val;
+ break; /* 64b: zero-ext */
+ case 8:
+ *c->dst.ptr = c->dst.val;
+ break;
+ }
+ break;
+ case OP_MEM:
+ if (c->lock_prefix)
+ rc = ops->cmpxchg_emulated(
+ (unsigned long)c->dst.ptr,
+ &c->dst.orig_val,
+ &c->dst.val,
+ c->dst.bytes,
+ &err,
+ ctxt->vcpu);
+ else
+ rc = ops->write_emulated(
+ (unsigned long)c->dst.ptr,
+ &c->dst.val,
+ c->dst.bytes,
+ &err,
+ ctxt->vcpu);
+ if (rc == X86EMUL_PROPAGATE_FAULT)
+ emulate_pf(ctxt,
+ (unsigned long)c->dst.ptr, err);
+ if (rc != X86EMUL_CONTINUE)
+ return rc;
+ break;
+ case OP_NONE:
+ /* no writeback */
+ break;
+ default:
+ break;
+ }
+ return X86EMUL_CONTINUE;
+}
+
static inline void emulate_push(struct x86_emulate_ctxt *ctxt,
struct x86_emulate_ops *ops)
{
@@ -1651,11 +1709,12 @@ static int emulate_pop_sreg(struct x86_emulate_ctxt *ctxt,
return rc;
}
-static void emulate_pusha(struct x86_emulate_ctxt *ctxt,
+static int emulate_pusha(struct x86_emulate_ctxt *ctxt,
struct x86_emulate_ops *ops)
{
struct decode_cache *c = &ctxt->decode;
unsigned long old_esp = c->regs[VCPU_REGS_RSP];
+ int rc = X86EMUL_CONTINUE;
int reg = VCPU_REGS_RAX;
while (reg <= VCPU_REGS_RDI) {
@@ -1663,8 +1722,18 @@ static void emulate_pusha(struct x86_emulate_ctxt *ctxt,
(c->src.val = old_esp) : (c->src.val = c->regs[reg]);
emulate_push(ctxt, ops);
+
+ rc = writeback(ctxt, ops);
+ if (rc != X86EMUL_CONTINUE)
+ return rc;
+
++reg;
}
+
+ /* Disable writeback. */
+ c->dst.type = OP_NONE;
+
+ return rc;
}
static int emulate_popa(struct x86_emulate_ctxt *ctxt,
@@ -1817,64 +1886,6 @@ static int emulate_ret_far(struct x86_emulate_ctxt *ctxt,
return rc;
}
-static inline int writeback(struct x86_emulate_ctxt *ctxt,
- struct x86_emulate_ops *ops)
-{
- int rc;
- struct decode_cache *c = &ctxt->decode;
- u32 err;
-
- switch (c->dst.type) {
- case OP_REG:
- /* The 4-byte case *is* correct:
- * in 64-bit mode we zero-extend.
- */
- switch (c->dst.bytes) {
- case 1:
- *(u8 *)c->dst.ptr = (u8)c->dst.val;
- break;
- case 2:
- *(u16 *)c->dst.ptr = (u16)c->dst.val;
- break;
- case 4:
- *c->dst.ptr = (u32)c->dst.val;
- break; /* 64b: zero-ext */
- case 8:
- *c->dst.ptr = c->dst.val;
- break;
- }
- break;
- case OP_MEM:
- if (c->lock_prefix)
- rc = ops->cmpxchg_emulated(
- (unsigned long)c->dst.ptr,
- &c->dst.orig_val,
- &c->dst.val,
- c->dst.bytes,
- &err,
- ctxt->vcpu);
- else
- rc = ops->write_emulated(
- (unsigned long)c->dst.ptr,
- &c->dst.val,
- c->dst.bytes,
- &err,
- ctxt->vcpu);
- if (rc == X86EMUL_PROPAGATE_FAULT)
- emulate_pf(ctxt,
- (unsigned long)c->dst.ptr, err);
- if (rc != X86EMUL_CONTINUE)
- return rc;
- break;
- case OP_NONE:
- /* no writeback */
- break;
- default:
- break;
- }
- return X86EMUL_CONTINUE;
-}
-
static inline void
setup_syscalls_segments(struct x86_emulate_ctxt *ctxt,
struct x86_emulate_ops *ops, struct desc_struct *cs,
@@ -2689,7 +2700,9 @@ special_insn:
goto done;
break;
case 0x60: /* pusha */
- emulate_pusha(ctxt, ops);
+ rc = emulate_pusha(ctxt, ops);
+ if (rc != X86EMUL_CONTINUE)
+ goto done;
break;
case 0x61: /* popa */
rc = emulate_popa(ctxt, ops);
--
1.7.0.4
^ permalink raw reply related [flat|nested] 12+ messages in thread* [PATCHv2] KVM: x86 emulator: fix pusha instruction emulation
2010-06-15 1:00 [PATCH] KVM: x86 emulator: fix pusha instruction emulation Wei Yongjun
@ 2010-06-15 1:03 ` Wei Yongjun
2010-06-15 2:10 ` [PATCH] test: add test for pusha and popa instructions Wei Yongjun
2010-06-15 11:00 ` [PATCHv2] KVM: x86 emulator: fix pusha instruction emulation Avi Kivity
2010-06-16 14:09 ` [PATCH] " Marcelo Tosatti
1 sibling, 2 replies; 12+ messages in thread
From: Wei Yongjun @ 2010-06-15 1:03 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti; +Cc: kvm
emulate pusha instruction only writeback the last
EDI register, but the other registers which need
to be writeback is ignored. This patch fixed it.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
---
v1 -> v2 add missing signed off by
---
arch/x86/kvm/emulate.c | 133 ++++++++++++++++++++++++++----------------------
1 files changed, 73 insertions(+), 60 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index a4c2dcd..c990db0 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -1553,6 +1553,64 @@ exception:
return X86EMUL_PROPAGATE_FAULT;
}
+static inline int writeback(struct x86_emulate_ctxt *ctxt,
+ struct x86_emulate_ops *ops)
+{
+ int rc;
+ struct decode_cache *c = &ctxt->decode;
+ u32 err;
+
+ switch (c->dst.type) {
+ case OP_REG:
+ /* The 4-byte case *is* correct:
+ * in 64-bit mode we zero-extend.
+ */
+ switch (c->dst.bytes) {
+ case 1:
+ *(u8 *)c->dst.ptr = (u8)c->dst.val;
+ break;
+ case 2:
+ *(u16 *)c->dst.ptr = (u16)c->dst.val;
+ break;
+ case 4:
+ *c->dst.ptr = (u32)c->dst.val;
+ break; /* 64b: zero-ext */
+ case 8:
+ *c->dst.ptr = c->dst.val;
+ break;
+ }
+ break;
+ case OP_MEM:
+ if (c->lock_prefix)
+ rc = ops->cmpxchg_emulated(
+ (unsigned long)c->dst.ptr,
+ &c->dst.orig_val,
+ &c->dst.val,
+ c->dst.bytes,
+ &err,
+ ctxt->vcpu);
+ else
+ rc = ops->write_emulated(
+ (unsigned long)c->dst.ptr,
+ &c->dst.val,
+ c->dst.bytes,
+ &err,
+ ctxt->vcpu);
+ if (rc == X86EMUL_PROPAGATE_FAULT)
+ emulate_pf(ctxt,
+ (unsigned long)c->dst.ptr, err);
+ if (rc != X86EMUL_CONTINUE)
+ return rc;
+ break;
+ case OP_NONE:
+ /* no writeback */
+ break;
+ default:
+ break;
+ }
+ return X86EMUL_CONTINUE;
+}
+
static inline void emulate_push(struct x86_emulate_ctxt *ctxt,
struct x86_emulate_ops *ops)
{
@@ -1651,11 +1709,12 @@ static int emulate_pop_sreg(struct x86_emulate_ctxt *ctxt,
return rc;
}
-static void emulate_pusha(struct x86_emulate_ctxt *ctxt,
+static int emulate_pusha(struct x86_emulate_ctxt *ctxt,
struct x86_emulate_ops *ops)
{
struct decode_cache *c = &ctxt->decode;
unsigned long old_esp = c->regs[VCPU_REGS_RSP];
+ int rc = X86EMUL_CONTINUE;
int reg = VCPU_REGS_RAX;
while (reg <= VCPU_REGS_RDI) {
@@ -1663,8 +1722,18 @@ static void emulate_pusha(struct x86_emulate_ctxt *ctxt,
(c->src.val = old_esp) : (c->src.val = c->regs[reg]);
emulate_push(ctxt, ops);
+
+ rc = writeback(ctxt, ops);
+ if (rc != X86EMUL_CONTINUE)
+ return rc;
+
++reg;
}
+
+ /* Disable writeback. */
+ c->dst.type = OP_NONE;
+
+ return rc;
}
static int emulate_popa(struct x86_emulate_ctxt *ctxt,
@@ -1817,64 +1886,6 @@ static int emulate_ret_far(struct x86_emulate_ctxt *ctxt,
return rc;
}
-static inline int writeback(struct x86_emulate_ctxt *ctxt,
- struct x86_emulate_ops *ops)
-{
- int rc;
- struct decode_cache *c = &ctxt->decode;
- u32 err;
-
- switch (c->dst.type) {
- case OP_REG:
- /* The 4-byte case *is* correct:
- * in 64-bit mode we zero-extend.
- */
- switch (c->dst.bytes) {
- case 1:
- *(u8 *)c->dst.ptr = (u8)c->dst.val;
- break;
- case 2:
- *(u16 *)c->dst.ptr = (u16)c->dst.val;
- break;
- case 4:
- *c->dst.ptr = (u32)c->dst.val;
- break; /* 64b: zero-ext */
- case 8:
- *c->dst.ptr = c->dst.val;
- break;
- }
- break;
- case OP_MEM:
- if (c->lock_prefix)
- rc = ops->cmpxchg_emulated(
- (unsigned long)c->dst.ptr,
- &c->dst.orig_val,
- &c->dst.val,
- c->dst.bytes,
- &err,
- ctxt->vcpu);
- else
- rc = ops->write_emulated(
- (unsigned long)c->dst.ptr,
- &c->dst.val,
- c->dst.bytes,
- &err,
- ctxt->vcpu);
- if (rc == X86EMUL_PROPAGATE_FAULT)
- emulate_pf(ctxt,
- (unsigned long)c->dst.ptr, err);
- if (rc != X86EMUL_CONTINUE)
- return rc;
- break;
- case OP_NONE:
- /* no writeback */
- break;
- default:
- break;
- }
- return X86EMUL_CONTINUE;
-}
-
static inline void
setup_syscalls_segments(struct x86_emulate_ctxt *ctxt,
struct x86_emulate_ops *ops, struct desc_struct *cs,
@@ -2689,7 +2700,9 @@ special_insn:
goto done;
break;
case 0x60: /* pusha */
- emulate_pusha(ctxt, ops);
+ rc = emulate_pusha(ctxt, ops);
+ if (rc != X86EMUL_CONTINUE)
+ goto done;
break;
case 0x61: /* popa */
rc = emulate_popa(ctxt, ops);
--
1.7.0.4
^ permalink raw reply related [flat|nested] 12+ messages in thread* [PATCH] test: add test for pusha and popa instructions
2010-06-15 1:03 ` [PATCHv2] " Wei Yongjun
@ 2010-06-15 2:10 ` Wei Yongjun
2010-06-20 8:29 ` Avi Kivity
2010-06-15 11:00 ` [PATCHv2] KVM: x86 emulator: fix pusha instruction emulation Avi Kivity
1 sibling, 1 reply; 12+ messages in thread
From: Wei Yongjun @ 2010-06-15 2:10 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti; +Cc: kvm
This patch add test for pusha and popa instructions.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
---
kvm/test/x86/realmode.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 46 insertions(+), 0 deletions(-)
diff --git a/kvm/test/x86/realmode.c b/kvm/test/x86/realmode.c
index 70a1e05..bd79348 100644
--- a/kvm/test/x86/realmode.c
+++ b/kvm/test/x86/realmode.c
@@ -820,12 +820,58 @@ void test_null(void)
print_serial("null test: PASS\n");
}
+void test_pusha_popa()
+{
+ struct regs inregs = { .eax = 0, .ebx = 1, .ecx = 2, .edx = 3, .esi = 4, .edi = 5, .ebp = 6, .esp = 7}, outregs;
+
+ MK_INSN(pusha, "pusha\n\t"
+ "pop %edi\n\t"
+ "pop %esi\n\t"
+ "pop %ebp\n\t"
+ "pop %eax\n\t"
+ "pop %ebx\n\t"
+ "pop %edx\n\t"
+ "pop %ecx\n\t"
+ "pop %esp\n\t"
+ "xchg %esp, %eax\n\t"
+ );
+
+ MK_INSN(popa, "push %eax\n\t"
+ "push %ecx\n\t"
+ "push %edx\n\t"
+ "push %ebx\n\t"
+ "push %esp\n\t"
+ "push %ebp\n\t"
+ "push %esi\n\t"
+ "push %edi\n\t"
+ "popa\n\t"
+ );
+
+ exec_in_big_real_mode(&inregs, &outregs,
+ insn_pusha,
+ insn_pusha_end - insn_pusha);
+
+ if (!regs_equal(&inregs, &outregs, 0))
+ print_serial("Pusha/Popa Test1: FAIL\n");
+ else
+ print_serial("Pusha/Popa Test1: PASS\n");
+
+ exec_in_big_real_mode(&inregs, &outregs,
+ insn_popa,
+ insn_popa_end - insn_popa);
+ if (!regs_equal(&inregs, &outregs, 0))
+ print_serial("Pusha/Popa Test2: FAIL\n");
+ else
+ print_serial("Pusha/Popa Test2: PASS\n");
+}
+
void realmode_start(void)
{
test_null();
test_shld();
test_push_pop();
+ test_pusha_popa();
test_mov_imm();
test_cmp_imm();
test_add_imm();
--
1.7.0.4
^ permalink raw reply related [flat|nested] 12+ messages in thread* Re: [PATCH] test: add test for pusha and popa instructions
2010-06-15 2:10 ` [PATCH] test: add test for pusha and popa instructions Wei Yongjun
@ 2010-06-20 8:29 ` Avi Kivity
2010-06-21 2:48 ` Wei Yongjun
0 siblings, 1 reply; 12+ messages in thread
From: Avi Kivity @ 2010-06-20 8:29 UTC (permalink / raw)
To: Wei Yongjun; +Cc: Marcelo Tosatti, kvm
On 06/15/2010 05:10 AM, Wei Yongjun wrote:
> This patch add test for pusha and popa instructions.
>
>
Did you test the test? These tests require
'emulate_invalid_guest_state=1' and to run on Intel to actually test
anything. You can check with ftrace whether kvm actually emulated
pusha/popa.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] test: add test for pusha and popa instructions
2010-06-20 8:29 ` Avi Kivity
@ 2010-06-21 2:48 ` Wei Yongjun
2010-06-21 13:15 ` Avi Kivity
0 siblings, 1 reply; 12+ messages in thread
From: Wei Yongjun @ 2010-06-21 2:48 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
> On 06/15/2010 05:10 AM, Wei Yongjun wrote:
>
>> This patch add test for pusha and popa instructions.
>>
>>
>>
> Did you test the test? These tests require
> 'emulate_invalid_guest_state=1' and to run on Intel to actually test
> anything. You can check with ftrace whether kvm actually emulated
> pusha/popa.
>
Yes, I did test this case, and checked kvm actually emulated
pusha/popa by add printk to kernel source.
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] test: add test for pusha and popa instructions
2010-06-21 2:48 ` Wei Yongjun
@ 2010-06-21 13:15 ` Avi Kivity
0 siblings, 0 replies; 12+ messages in thread
From: Avi Kivity @ 2010-06-21 13:15 UTC (permalink / raw)
To: Wei Yongjun; +Cc: Marcelo Tosatti, kvm
On 06/21/2010 05:48 AM, Wei Yongjun wrote:
>>
>> Did you test the test? These tests require
>> 'emulate_invalid_guest_state=1' and to run on Intel to actually test
>> anything. You can check with ftrace whether kvm actually emulated
>> pusha/popa.
>>
>>
> Yes, I did test this case, and checked kvm actually emulated
> pusha/popa by add printk to kernel source.
>
I see, thanks. Anyway Marcelo already applied it.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCHv2] KVM: x86 emulator: fix pusha instruction emulation
2010-06-15 1:03 ` [PATCHv2] " Wei Yongjun
2010-06-15 2:10 ` [PATCH] test: add test for pusha and popa instructions Wei Yongjun
@ 2010-06-15 11:00 ` Avi Kivity
1 sibling, 0 replies; 12+ messages in thread
From: Avi Kivity @ 2010-06-15 11:00 UTC (permalink / raw)
To: Wei Yongjun; +Cc: Marcelo Tosatti, kvm
On 06/15/2010 04:03 AM, Wei Yongjun wrote:
> emulate pusha instruction only writeback the last
> EDI register, but the other registers which need
> to be writeback is ignored. This patch fixed it.
>
>
Ouch.
>
> -static void emulate_pusha(struct x86_emulate_ctxt *ctxt,
> +static int emulate_pusha(struct x86_emulate_ctxt *ctxt,
> struct x86_emulate_ops *ops)
> {
> struct decode_cache *c = &ctxt->decode;
> unsigned long old_esp = c->regs[VCPU_REGS_RSP];
> + int rc = X86EMUL_CONTINUE;
> int reg = VCPU_REGS_RAX;
>
> while (reg <= VCPU_REGS_RDI) {
> @@ -1663,8 +1722,18 @@ static void emulate_pusha(struct x86_emulate_ctxt *ctxt,
> (c->src.val = old_esp) : (c->src.val = c->regs[reg]);
>
> emulate_push(ctxt, ops);
> +
> + rc = writeback(ctxt, ops);
> + if (rc != X86EMUL_CONTINUE)
> + return rc;
> +
>
We might have partial completion if we fault on the one the second or
later register. Still, it's an improvement over what we have now.
> ++reg;
> }
> +
> + /* Disable writeback. */
> + c->dst.type = OP_NONE;
> +
> + return rc;
> }
>
>
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] KVM: x86 emulator: fix pusha instruction emulation
2010-06-15 1:00 [PATCH] KVM: x86 emulator: fix pusha instruction emulation Wei Yongjun
2010-06-15 1:03 ` [PATCHv2] " Wei Yongjun
@ 2010-06-16 14:09 ` Marcelo Tosatti
2010-06-17 9:33 ` [PATCH] KVM: x86 emulator: fix group3 instruction decoding Wei Yongjun
1 sibling, 1 reply; 12+ messages in thread
From: Marcelo Tosatti @ 2010-06-16 14:09 UTC (permalink / raw)
To: Wei Yongjun; +Cc: Avi Kivity, kvm
On Tue, Jun 15, 2010 at 09:00:16AM +0800, Wei Yongjun wrote:
> emulate pusha instruction only writeback the last
> EDI register, but the other registers which need
> to be writeback is ignored. This patch fixed it.
>
> ---
> arch/x86/kvm/emulate.c | 133 ++++++++++++++++++++++++++----------------------
> 1 files changed, 73 insertions(+), 60 deletions(-)
Applied, thanks.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH] KVM: x86 emulator: fix group3 instruction decoding
2010-06-16 14:09 ` [PATCH] " Marcelo Tosatti
@ 2010-06-17 9:33 ` Wei Yongjun
2010-06-17 12:00 ` Avi Kivity
0 siblings, 1 reply; 12+ messages in thread
From: Wei Yongjun @ 2010-06-17 9:33 UTC (permalink / raw)
To: Marcelo Tosatti; +Cc: Avi Kivity, kvm
Group 3 instruction with ModRM reg field as 001 is
defined as test instruction under AMD arch, and
emulate_grp3() is ready for emulate it, so fix the
decoding.
static inline int emulate_grp3(...)
{
...
switch (c->modrm_reg) {
case 0 ... 1: /* test */
emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags);
...
}
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
---
arch/x86/kvm/emulate.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index c990db0..abb8cec 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -336,11 +336,11 @@ static u32 group_table[] = {
[Group1A*8] =
DstMem | SrcNone | ModRM | Mov | Stack, 0, 0, 0, 0, 0, 0, 0,
[Group3_Byte*8] =
- ByteOp | SrcImm | DstMem | ModRM, 0,
+ ByteOp | SrcImm | DstMem | ModRM, ByteOp | SrcImm | DstMem | ModRM,
ByteOp | DstMem | SrcNone | ModRM, ByteOp | DstMem | SrcNone | ModRM,
0, 0, 0, 0,
[Group3*8] =
- DstMem | SrcImm | ModRM, 0,
+ DstMem | SrcImm | ModRM, DstMem | SrcImm | ModRM,
DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM,
0, 0, 0, 0,
[Group4*8] =
--
1.7.0.4
^ permalink raw reply related [flat|nested] 12+ messages in thread* Re: [PATCH] KVM: x86 emulator: fix group3 instruction decoding
2010-06-17 9:33 ` [PATCH] KVM: x86 emulator: fix group3 instruction decoding Wei Yongjun
@ 2010-06-17 12:00 ` Avi Kivity
2010-06-18 0:51 ` Wei Yongjun
0 siblings, 1 reply; 12+ messages in thread
From: Avi Kivity @ 2010-06-17 12:00 UTC (permalink / raw)
To: Wei Yongjun; +Cc: Marcelo Tosatti, kvm
On 06/17/2010 12:33 PM, Wei Yongjun wrote:
> Group 3 instruction with ModRM reg field as 001 is
> defined as test instruction under AMD arch, and
> emulate_grp3() is ready for emulate it, so fix the
> decoding.
>
>
Strange but true.
Did you encounter any situation which actually needed this, or did you
find this by code inspection?
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] KVM: x86 emulator: fix group3 instruction decoding
2010-06-17 12:00 ` Avi Kivity
@ 2010-06-18 0:51 ` Wei Yongjun
2010-06-20 8:24 ` Avi Kivity
0 siblings, 1 reply; 12+ messages in thread
From: Wei Yongjun @ 2010-06-18 0:51 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
> On 06/17/2010 12:33 PM, Wei Yongjun wrote:
>> Group 3 instruction with ModRM reg field as 001 is
>> defined as test instruction under AMD arch, and
>> emulate_grp3() is ready for emulate it, so fix the
>> decoding.
>>
>>
>
> Strange but true.
>
> Did you encounter any situation which actually needed this, or did you
> find this by code inspection?
Just found it by source review, looks strange, so...
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] KVM: x86 emulator: fix group3 instruction decoding
2010-06-18 0:51 ` Wei Yongjun
@ 2010-06-20 8:24 ` Avi Kivity
0 siblings, 0 replies; 12+ messages in thread
From: Avi Kivity @ 2010-06-20 8:24 UTC (permalink / raw)
To: Wei Yongjun; +Cc: Marcelo Tosatti, kvm
On 06/18/2010 03:51 AM, Wei Yongjun wrote:
>
>> On 06/17/2010 12:33 PM, Wei Yongjun wrote:
>>
>>> Group 3 instruction with ModRM reg field as 001 is
>>> defined as test instruction under AMD arch, and
>>> emulate_grp3() is ready for emulate it, so fix the
>>> decoding.
>>>
>>>
>>>
>> Strange but true.
>>
>> Did you encounter any situation which actually needed this, or did you
>> find this by code inspection?
>>
> Just found it by source review, looks strange, so...
>
Well, the emulator usually puts me to sleep, so good catch. Applied the
patch, thanks.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2010-06-21 13:15 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-06-15 1:00 [PATCH] KVM: x86 emulator: fix pusha instruction emulation Wei Yongjun
2010-06-15 1:03 ` [PATCHv2] " Wei Yongjun
2010-06-15 2:10 ` [PATCH] test: add test for pusha and popa instructions Wei Yongjun
2010-06-20 8:29 ` Avi Kivity
2010-06-21 2:48 ` Wei Yongjun
2010-06-21 13:15 ` Avi Kivity
2010-06-15 11:00 ` [PATCHv2] KVM: x86 emulator: fix pusha instruction emulation Avi Kivity
2010-06-16 14:09 ` [PATCH] " Marcelo Tosatti
2010-06-17 9:33 ` [PATCH] KVM: x86 emulator: fix group3 instruction decoding Wei Yongjun
2010-06-17 12:00 ` Avi Kivity
2010-06-18 0:51 ` Wei Yongjun
2010-06-20 8:24 ` Avi Kivity
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox