diff mbox

[6/8] KVM: x86 emulator: convert DIV/IDIV to fastop

Message ID 1360402311-19904-7-git-send-email-avi.kivity@gmail.com (mailing list archive)
State New, archived
Headers show

Commit Message

Avi Kivity Feb. 9, 2013, 9:31 a.m. UTC 
Since DIV and IDIV can generate exceptions, we need an additional output
parameter indicating whether an execption has occured.  To avoid increasing
register pressure on i386, we use %rsi, which is already allocated for
the fastop code pointer.

Signed-off-by: Avi Kivity <avi.kivity@gmail.com>
---
 arch/x86/kvm/emulate.c | 51 +++++++++++++++++++++++++-------------------------
 1 file changed, 26 insertions(+), 25 deletions(-)

Comments

Gleb Natapov Feb. 10, 2013, 12:26 p.m. UTC | #1 
On Sat, Feb 09, 2013 at 11:31:49AM +0200, Avi Kivity wrote:
> Since DIV and IDIV can generate exceptions, we need an additional output
> parameter indicating whether an execption has occured.  To avoid increasing
> register pressure on i386, we use %rsi, which is already allocated for
> the fastop code pointer.
> 
> Signed-off-by: Avi Kivity <avi.kivity@gmail.com>
> ---
>  arch/x86/kvm/emulate.c | 51 +++++++++++++++++++++++++-------------------------
>  1 file changed, 26 insertions(+), 25 deletions(-)
> 
> diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
> index 0f0c15e..89f56bb 100644
> --- a/arch/x86/kvm/emulate.c
> +++ b/arch/x86/kvm/emulate.c
> @@ -180,6 +180,7 @@
>   * src:    rdx        (in/out)
>   * src2:   rcx        (in)
>   * flags:  rflags     (in/out)
> + * ex:     rsi        (in:nonzero, out:zero if exception)
in:fastop pointer, or "code pointer" may be?

>   *
>   * Moreover, they are all exactly FASTOP_SIZE bytes long, so functions for
>   * different operand sizes can be reached by calculation, rather than a jump
> @@ -467,7 +468,10 @@ static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *));
>  #define FOPNOP() FOP_ALIGN FOP_RET
>  
>  #define FOP1E(op,  dst) \
> -	FOP_ALIGN #op " %" #dst " \n\t" FOP_RET
> +	FOP_ALIGN "10: " #op " %" #dst " \n\t" FOP_RET
> +
> +#define FOP1EEX(op,  dst) \
> +	FOP1E(op, dst) _ASM_EXTABLE(10b, kvm_fastop_exception)
>  
>  #define FASTOP1(op) \
>  	FOP_START(op) \
> @@ -486,6 +490,15 @@ static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *));
>  	ON64(FOP1E(op, rcx)) \
>  	FOP_END
>  
> +/* 1-operand, using src2 (for MUL/DIV r/m), with exceptions */
> +#define FASTOP1SRC2EX(op, name) \
> +	FOP_START(name) \
> +	FOP1EEX(op, cl) \
> +	FOP1EEX(op, cx) \
> +	FOP1EEX(op, ecx) \
> +	ON64(FOP1EEX(op, rcx)) \
> +	FOP_END
> +
>  #define FOP2E(op,  dst, src)	   \
>  	FOP_ALIGN #op " %" #src ", %" #dst " \n\t" FOP_RET
>  
> @@ -530,6 +543,9 @@ static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *));
>  /* Special case for SETcc - 1 instruction per cc */
>  #define FOP_SETCC(op) ".align 4; " #op " %al; ret \n\t"
>  
> +asm(".global kvm_fastop_exception \n"
> +    "kvm_fastop_exception: xor %esi, %esi; ret");
> +
>  FOP_START(setcc)
>  FOP_SETCC(seto)
>  FOP_SETCC(setno)
> @@ -1001,6 +1017,8 @@ FASTOP2(test);
>  
>  FASTOP1SRC2(mul, mul_ex);
>  FASTOP1SRC2(imul, imul_ex);
> +FASTOP1SRC2EX(div, div_ex);
> +FASTOP1SRC2EX(idiv, idiv_ex);
>  
>  FASTOP3WCL(shld);
>  FASTOP3WCL(shrd);
> @@ -2116,26 +2134,6 @@ static int em_jmp_far(struct x86_emulate_ctxt *ctxt)
>  	return X86EMUL_CONTINUE;
>  }
>  
> -static int em_div_ex(struct x86_emulate_ctxt *ctxt)
> -{
> -	u8 de = 0;
> -
> -	emulate_1op_rax_rdx(ctxt, "div", de);
> -	if (de)
> -		return emulate_de(ctxt);
> -	return X86EMUL_CONTINUE;
> -}
> -
> -static int em_idiv_ex(struct x86_emulate_ctxt *ctxt)
> -{
> -	u8 de = 0;
> -
> -	emulate_1op_rax_rdx(ctxt, "idiv", de);
> -	if (de)
> -		return emulate_de(ctxt);
> -	return X86EMUL_CONTINUE;
> -}
> -
>  static int em_grp45(struct x86_emulate_ctxt *ctxt)
>  {
>  	int rc = X86EMUL_CONTINUE;
> @@ -3700,8 +3698,8 @@ static const struct opcode group3[] = {
>  	F(DstMem | SrcNone | Lock, em_neg),
>  	F(DstXacc | Src2Mem, em_mul_ex),
>  	F(DstXacc | Src2Mem, em_imul_ex),
> -	I(DstXacc | Src2Mem, em_div_ex),
> -	I(DstXacc | Src2Mem, em_idiv_ex),
> +	F(DstXacc | Src2Mem, em_div_ex),
> +	F(DstXacc | Src2Mem, em_idiv_ex),
>  };
>  
>  static const struct opcode group4[] = {
> @@ -4518,9 +4516,12 @@ static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *))
>  	if (!(ctxt->d & ByteOp))
>  		fop += __ffs(ctxt->dst.bytes) * FASTOP_SIZE;
>  	asm("push %[flags]; popf; call *%[fastop]; pushf; pop %[flags]\n"
> -	    : "+a"(ctxt->dst.val), "+d"(ctxt->src.val), [flags]"+D"(flags)
> -	: "c"(ctxt->src2.val), [fastop]"S"(fop));
> +	    : "+a"(ctxt->dst.val), "+d"(ctxt->src.val), [flags]"+D"(flags),
> +	      [fastop]"+S"(fop)
> +	    : "c"(ctxt->src2.val));
>  	ctxt->eflags = (ctxt->eflags & ~EFLAGS_MASK) | (flags & EFLAGS_MASK);
> +	if (!fop)
Looks like this magic deserves at least a short comment :)

> +		return emulate_de(ctxt);
>  	return X86EMUL_CONTINUE;
>  }
>  
> -- 
> 1.8.1.2

--
			Gleb.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 0f0c15e..89f56bb 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -180,6 +180,7 @@ 
  * src:    rdx        (in/out)
  * src2:   rcx        (in)
  * flags:  rflags     (in/out)
+ * ex:     rsi        (in:nonzero, out:zero if exception)
  *
  * Moreover, they are all exactly FASTOP_SIZE bytes long, so functions for
  * different operand sizes can be reached by calculation, rather than a jump
@@ -467,7 +468,10 @@  static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *));
 #define FOPNOP() FOP_ALIGN FOP_RET
 
 #define FOP1E(op,  dst) \
-	FOP_ALIGN #op " %" #dst " \n\t" FOP_RET
+	FOP_ALIGN "10: " #op " %" #dst " \n\t" FOP_RET
+
+#define FOP1EEX(op,  dst) \
+	FOP1E(op, dst) _ASM_EXTABLE(10b, kvm_fastop_exception)
 
 #define FASTOP1(op) \
 	FOP_START(op) \
@@ -486,6 +490,15 @@  static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *));
 	ON64(FOP1E(op, rcx)) \
 	FOP_END
 
+/* 1-operand, using src2 (for MUL/DIV r/m), with exceptions */
+#define FASTOP1SRC2EX(op, name) \
+	FOP_START(name) \
+	FOP1EEX(op, cl) \
+	FOP1EEX(op, cx) \
+	FOP1EEX(op, ecx) \
+	ON64(FOP1EEX(op, rcx)) \
+	FOP_END
+
 #define FOP2E(op,  dst, src)	   \
 	FOP_ALIGN #op " %" #src ", %" #dst " \n\t" FOP_RET
 
@@ -530,6 +543,9 @@  static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *));
 /* Special case for SETcc - 1 instruction per cc */
 #define FOP_SETCC(op) ".align 4; " #op " %al; ret \n\t"
 
+asm(".global kvm_fastop_exception \n"
+    "kvm_fastop_exception: xor %esi, %esi; ret");
+
 FOP_START(setcc)
 FOP_SETCC(seto)
 FOP_SETCC(setno)
@@ -1001,6 +1017,8 @@  FASTOP2(test);
 
 FASTOP1SRC2(mul, mul_ex);
 FASTOP1SRC2(imul, imul_ex);
+FASTOP1SRC2EX(div, div_ex);
+FASTOP1SRC2EX(idiv, idiv_ex);
 
 FASTOP3WCL(shld);
 FASTOP3WCL(shrd);
@@ -2116,26 +2134,6 @@  static int em_jmp_far(struct x86_emulate_ctxt *ctxt)
 	return X86EMUL_CONTINUE;
 }
 
-static int em_div_ex(struct x86_emulate_ctxt *ctxt)
-{
-	u8 de = 0;
-
-	emulate_1op_rax_rdx(ctxt, "div", de);
-	if (de)
-		return emulate_de(ctxt);
-	return X86EMUL_CONTINUE;
-}
-
-static int em_idiv_ex(struct x86_emulate_ctxt *ctxt)
-{
-	u8 de = 0;
-
-	emulate_1op_rax_rdx(ctxt, "idiv", de);
-	if (de)
-		return emulate_de(ctxt);
-	return X86EMUL_CONTINUE;
-}
-
 static int em_grp45(struct x86_emulate_ctxt *ctxt)
 {
 	int rc = X86EMUL_CONTINUE;
@@ -3700,8 +3698,8 @@  static const struct opcode group3[] = {
 	F(DstMem | SrcNone | Lock, em_neg),
 	F(DstXacc | Src2Mem, em_mul_ex),
 	F(DstXacc | Src2Mem, em_imul_ex),
-	I(DstXacc | Src2Mem, em_div_ex),
-	I(DstXacc | Src2Mem, em_idiv_ex),
+	F(DstXacc | Src2Mem, em_div_ex),
+	F(DstXacc | Src2Mem, em_idiv_ex),
 };
 
 static const struct opcode group4[] = {
@@ -4518,9 +4516,12 @@  static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *))
 	if (!(ctxt->d & ByteOp))
 		fop += __ffs(ctxt->dst.bytes) * FASTOP_SIZE;
 	asm("push %[flags]; popf; call *%[fastop]; pushf; pop %[flags]\n"
-	    : "+a"(ctxt->dst.val), "+d"(ctxt->src.val), [flags]"+D"(flags)
-	: "c"(ctxt->src2.val), [fastop]"S"(fop));
+	    : "+a"(ctxt->dst.val), "+d"(ctxt->src.val), [flags]"+D"(flags),
+	      [fastop]"+S"(fop)
+	    : "c"(ctxt->src2.val));
 	ctxt->eflags = (ctxt->eflags & ~EFLAGS_MASK) | (flags & EFLAGS_MASK);
+	if (!fop)
+		return emulate_de(ctxt);
 	return X86EMUL_CONTINUE;
 }