[dev.power64] cmd/5a, cmd/6a, cmd/8a, cmd/9a: make labels function-scoped

I removed support for jumping between functions years ago,
as part of doing the instruction layout for each function separately.

Given that, it makes sense to treat labels as function-scoped.
This lets each function have its own 'loop' label, for example.

Makes the assembly much cleaner and removes the last
reason anyone would reach for the 123(PC) form instead.

Note that this is on the dev.power64 branch, but it changes all
the assemblers. The change will ship in Go 1.5 (perhaps after
being ported into the new assembler).

Came up as part of CL 167730043.

LGTM=r
R=r
CC=austin, dave, golang-codereviews, minux
https://codereview.appspot.com/159670043

1 files changed, 29 insertions, 29 deletions

diff --git a/src/runtime/asm_amd64p32.s b/src/runtime/asm_amd64p32.s
index 28875bc55..de3ef3a23 100644
--- a/src/runtime/asm_amd64p32.s
+++ b/src/runtime/asm_amd64p32.s
@@ -444,11 +444,11 @@ TEXT runtime·cas64(SB), NOSPLIT, $0-25
 	MOVQ	new+16(FP), CX
 	LOCK
 	CMPXCHGQ	CX, 0(BX)
-	JNZ	cas64_fail
+	JNZ	fail
 	MOVL	$1, AX
 	MOVB	AX, ret+24(FP)
 	RET
-cas64_fail:
+fail:
 	MOVL	$0, AX
 	MOVB	AX, ret+24(FP)
 	RET
@@ -834,29 +834,29 @@ TEXT runtime·cmpbytes(SB),NOSPLIT,$0-28
 //   AX = 1/0/-1
 TEXT runtime·cmpbody(SB),NOSPLIT,$0-0
 	CMPQ	SI, DI
-	JEQ	cmp_allsame
+	JEQ	allsame
 	CMPQ	BX, DX
 	MOVQ	DX, R8
 	CMOVQLT	BX, R8 // R8 = min(alen, blen) = # of bytes to compare
 	CMPQ	R8, $8
-	JB	cmp_small
+	JB	small
 
-cmp_loop:
+loop:
 	CMPQ	R8, $16
-	JBE	cmp_0through16
+	JBE	_0through16
 	MOVOU	(SI), X0
 	MOVOU	(DI), X1
 	PCMPEQB X0, X1
 	PMOVMSKB X1, AX
 	XORQ	$0xffff, AX	// convert EQ to NE
-	JNE	cmp_diff16	// branch if at least one byte is not equal
+	JNE	diff16	// branch if at least one byte is not equal
 	ADDQ	$16, SI
 	ADDQ	$16, DI
 	SUBQ	$16, R8
-	JMP	cmp_loop
+	JMP	loop
 	
 	// AX = bit mask of differences
-cmp_diff16:
+diff16:
 	BSFQ	AX, BX	// index of first byte that differs
 	XORQ	AX, AX
 	ADDQ	BX, SI
@@ -868,23 +868,23 @@ cmp_diff16:
 	RET
 
 	// 0 through 16 bytes left, alen>=8, blen>=8
-cmp_0through16:
+_0through16:
 	CMPQ	R8, $8
-	JBE	cmp_0through8
+	JBE	_0through8
 	MOVQ	(SI), AX
 	MOVQ	(DI), CX
 	CMPQ	AX, CX
-	JNE	cmp_diff8
-cmp_0through8:
+	JNE	diff8
+_0through8:
 	ADDQ	R8, SI
 	ADDQ	R8, DI
 	MOVQ	-8(SI), AX
 	MOVQ	-8(DI), CX
 	CMPQ	AX, CX
-	JEQ	cmp_allsame
+	JEQ	allsame
 
 	// AX and CX contain parts of a and b that differ.
-cmp_diff8:
+diff8:
 	BSWAPQ	AX	// reverse order of bytes
 	BSWAPQ	CX
 	XORQ	AX, CX
@@ -895,46 +895,46 @@ cmp_diff8:
 	RET
 
 	// 0-7 bytes in common
-cmp_small:
+small:
 	LEAQ	(R8*8), CX	// bytes left -> bits left
 	NEGQ	CX		//  - bits lift (== 64 - bits left mod 64)
-	JEQ	cmp_allsame
+	JEQ	allsame
 
 	// load bytes of a into high bytes of AX
 	CMPB	SI, $0xf8
-	JA	cmp_si_high
+	JA	si_high
 	MOVQ	(SI), SI
-	JMP	cmp_si_finish
-cmp_si_high:
+	JMP	si_finish
+si_high:
 	ADDQ	R8, SI
 	MOVQ	-8(SI), SI
 	SHRQ	CX, SI
-cmp_si_finish:
+si_finish:
 	SHLQ	CX, SI
 
 	// load bytes of b in to high bytes of BX
 	CMPB	DI, $0xf8
-	JA	cmp_di_high
+	JA	di_high
 	MOVQ	(DI), DI
-	JMP	cmp_di_finish
-cmp_di_high:
+	JMP	di_finish
+di_high:
 	ADDQ	R8, DI
 	MOVQ	-8(DI), DI
 	SHRQ	CX, DI
-cmp_di_finish:
+di_finish:
 	SHLQ	CX, DI
 
 	BSWAPQ	SI	// reverse order of bytes
 	BSWAPQ	DI
 	XORQ	SI, DI	// find bit differences
-	JEQ	cmp_allsame
+	JEQ	allsame
 	BSRQ	DI, CX	// index of highest bit difference
 	SHRQ	CX, SI	// move a's bit to bottom
 	ANDQ	$1, SI	// mask bit
 	LEAQ	-1(SI*2), AX // 1/0 => +1/-1
 	RET
 
-cmp_allsame:
+allsame:
 	XORQ	AX, AX
 	XORQ	CX, CX
 	CMPQ	BX, DX
@@ -969,7 +969,7 @@ TEXT runtime·indexbytebody(SB),NOSPLIT,$0
 	MOVL SI, DI
 
 	CMPL BX, $16
-	JLT indexbyte_small
+	JLT small
 
 	// round up to first 16-byte boundary
 	TESTL $15, SI
@@ -1027,7 +1027,7 @@ failure:
 	RET
 
 // handle for lengths < 16
-indexbyte_small:
+small:
 	MOVL BX, CX
 	REPN; SCASB
 	JZ success