all: change the old assembly style AX:CX to CX, AX

Assembly files with "/vendor/" or "testdata" in their paths were ignored.

Change-Id: I3882ff07eb4426abb9f8ee96f82dff73c81cd61f
GitHub-Last-Rev: 51ae8c324d72a12a059272fcf8568e670bfaf21b
GitHub-Pull-Request: golang/go#31166
Reviewed-on: https://go-review.googlesource.com/c/go/+/170197Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

src/crypto/elliptic/p256_asm_amd64.s

@@ -2300,10 +2300,10 @@ TEXT ·p256PointDoubleAsm(SB),NOSPLIT,$256-48
 	CMOVQEQ t3, acc7
 	ANDQ t0, mul0
 
-	SHRQ $1, acc4:acc5
-	SHRQ $1, acc5:acc6
-	SHRQ $1, acc6:acc7
-	SHRQ $1, acc7:mul0
+	SHRQ $1, acc5, acc4
+	SHRQ $1, acc6, acc5
+	SHRQ $1, acc7, acc6
+	SHRQ $1, mul0, acc7
 	ST (y)
 	/////////////////////////
 	LDacc (x)

src/math/big/arith_386.s

@@ -136,7 +136,7 @@ TEXT ·shlVU(SB),NOSPLIT,$0
 	MOVL s+24(FP), CX
 	MOVL (SI)(BX*4), AX	// w1 = x[n-1]
 	MOVL $0, DX
-	SHLL CX, DX:AX		// w1>>ŝ
+	SHLL CX, AX, DX		// w1>>ŝ
 	MOVL DX, c+28(FP)
 
 	CMPL BX, $0
@@ -145,7 +145,7 @@ TEXT ·shlVU(SB),NOSPLIT,$0
 	// i > 0
 L8:	MOVL AX, DX		// w = w1
 	MOVL -4(SI)(BX*4), AX	// w1 = x[i-1]
-	SHLL CX, DX:AX		// w<<s | w1>>ŝ
+	SHLL CX, AX, DX		// w<<s | w1>>ŝ
 	MOVL DX, (DI)(BX*4)	// z[i] = w<<s | w1>>ŝ
 	SUBL $1, BX		// i--
 	JG L8			// i > 0
@@ -171,7 +171,7 @@ TEXT ·shrVU(SB),NOSPLIT,$0
 	MOVL s+24(FP), CX
 	MOVL (SI), AX		// w1 = x[0]
 	MOVL $0, DX
-	SHRL CX, DX:AX		// w1<<ŝ
+	SHRL CX, AX, DX		// w1<<ŝ
 	MOVL DX, c+28(FP)
 
 	MOVL $0, BX		// i = 0
@@ -180,7 +180,7 @@ TEXT ·shrVU(SB),NOSPLIT,$0
 	// i < n-1
 L9:	MOVL AX, DX		// w = w1
 	MOVL 4(SI)(BX*4), AX	// w1 = x[i+1]
-	SHRL CX, DX:AX		// w>>s | w1<<ŝ
+	SHRL CX, AX, DX		// w>>s | w1<<ŝ
 	MOVL DX, (DI)(BX*4)	// z[i] = w>>s | w1<<ŝ
 	ADDL $1, BX		// i++
 

src/math/big/arith_amd64.s

@@ -264,7 +264,7 @@ TEXT ·shlVU(SB),NOSPLIT,$0
 	MOVQ s+48(FP), CX
 	MOVQ (R8)(BX*8), AX	// w1 = x[n-1]
 	MOVQ $0, DX
-	SHLQ CX, DX:AX		// w1>>ŝ
+	SHLQ CX, AX, DX		// w1>>ŝ
 	MOVQ DX, c+56(FP)
 
 	CMPQ BX, $0
@@ -273,7 +273,7 @@ TEXT ·shlVU(SB),NOSPLIT,$0
 	// i > 0
 L8:	MOVQ AX, DX		// w = w1
 	MOVQ -8(R8)(BX*8), AX	// w1 = x[i-1]
-	SHLQ CX, DX:AX		// w<<s | w1>>ŝ
+	SHLQ CX, AX, DX		// w<<s | w1>>ŝ
 	MOVQ DX, (R10)(BX*8)	// z[i] = w<<s | w1>>ŝ
 	SUBQ $1, BX		// i--
 	JG L8			// i > 0
@@ -299,7 +299,7 @@ TEXT ·shrVU(SB),NOSPLIT,$0
 	MOVQ s+48(FP), CX
 	MOVQ (R8), AX		// w1 = x[0]
 	MOVQ $0, DX
-	SHRQ CX, DX:AX		// w1<<ŝ
+	SHRQ CX, AX, DX		// w1<<ŝ
 	MOVQ DX, c+56(FP)
 
 	MOVQ $0, BX		// i = 0
@@ -308,7 +308,7 @@ TEXT ·shrVU(SB),NOSPLIT,$0
 	// i < n-1
 L9:	MOVQ AX, DX		// w = w1
 	MOVQ 8(R8)(BX*8), AX	// w1 = x[i+1]
-	SHRQ CX, DX:AX		// w>>s | w1<<ŝ
+	SHRQ CX, AX, DX		// w>>s | w1<<ŝ
 	MOVQ DX, (R10)(BX*8)	// z[i] = w>>s | w1<<ŝ
 	ADDQ $1, BX		// i++