Commit b8ac64a5 authored by Zhou Peng's avatar Zhou Peng Committed by Brad Fitzpatrick

all: this big patch remove whitespace from assembly files

Don't worry, this patch just remove trailing whitespace from
assembly files, and does not touch any logical changes.

Change-Id: Ia724ac0b1abf8bc1e41454bdc79289ef317c165d
Reviewed-on: https://go-review.googlesource.com/c/113595Reviewed-by: 's avatarBrad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
parent 119fafca
......@@ -70,7 +70,7 @@ label:
// LTYPEM spec6 { outcode(int($1), &$2); }
MOVL AX, BX
MOVL $4, BX
// LTYPEI spec7 { outcode(int($1), &$2); }
IMULL AX
IMULL $4, CX
......
......@@ -550,7 +550,7 @@ label1:
// ftsqrt BF, FRB
FTSQRT F2,$7
// FCFID
// FCFID
// FCFIDS
FCFID F2,F3
......
......@@ -122,7 +122,7 @@ loop:
ROUND2(DX,AX,BX,CX, 7,0xfcefa3f8, 9);
ROUND2(CX,DX,AX,BX,12,0x676f02d9,14);
ROUND2(BX,CX,DX,AX, 0,0x8d2a4c8a,20);
MOVL (5*4)(SI), DI
MOVL CX, BP
......
......@@ -99,7 +99,7 @@ loop:
ROUND2(DX,AX,BX,CX, 7,0xfcefa3f8, 9);
ROUND2(CX,DX,AX,BX,12,0x676f02d9,14);
ROUND2(BX,CX,DX,AX, 0,0x8d2a4c8a,20);
MOVL (5*4)(SI), R8
MOVL CX, R9
......@@ -144,7 +144,7 @@ loop:
ROLL $shift, a; \
XORL c, R9; \
ADDL b, a
ROUND4(AX,BX,CX,DX, 7,0xf4292244, 6);
ROUND4(DX,AX,BX,CX,14,0x432aff97,10);
ROUND4(CX,DX,AX,BX, 5,0xab9423a7,15);
......
......@@ -103,7 +103,7 @@ loop:
ROUND2(DX,AX,BX,CX, 7,0xfcefa3f8, 9);
ROUND2(CX,DX,AX,BX,12,0x676f02d9,14);
ROUND2(BX,CX,DX,AX, 0,0x8d2a4c8a,20);
MOVL (5*4)(SI), R8
MOVL CX, R9
......@@ -148,7 +148,7 @@ loop:
ROLL $shift, a; \
XORL c, R9; \
ADDL b, a
ROUND4(AX,BX,CX,DX, 7,0xf4292244, 6);
ROUND4(DX,AX,BX,CX,14,0x432aff97,10);
ROUND4(CX,DX,AX,BX, 5,0xab9423a7,15);
......
......@@ -105,7 +105,7 @@ TEXT ·block(SB),NOSPLIT,$92-16
MOVL p_len+8(FP), DX
SHRL $6, DX
SHLL $6, DX
LEAL (SI)(DX*1), DI
MOVL (0*4)(BP), AX
MOVL (1*4)(BP), BX
......@@ -148,7 +148,7 @@ loop:
ROUND1x(DX, BP, AX, BX, CX, 17)
ROUND1x(CX, DX, BP, AX, BX, 18)
ROUND1x(BX, CX, DX, BP, AX, 19)
ROUND2(AX, BX, CX, DX, BP, 20)
ROUND2(BP, AX, BX, CX, DX, 21)
ROUND2(DX, BP, AX, BX, CX, 22)
......@@ -169,7 +169,7 @@ loop:
ROUND2(DX, BP, AX, BX, CX, 37)
ROUND2(CX, DX, BP, AX, BX, 38)
ROUND2(BX, CX, DX, BP, AX, 39)
ROUND3(AX, BX, CX, DX, BP, 40)
ROUND3(BP, AX, BX, CX, DX, 41)
ROUND3(DX, BP, AX, BX, CX, 42)
......@@ -190,7 +190,7 @@ loop:
ROUND3(DX, BP, AX, BX, CX, 57)
ROUND3(CX, DX, BP, AX, BX, 58)
ROUND3(BX, CX, DX, BP, AX, 59)
ROUND4(AX, BX, CX, DX, BP, 60)
ROUND4(BP, AX, BX, CX, DX, 61)
ROUND4(DX, BP, AX, BX, CX, 62)
......
......@@ -60,7 +60,7 @@
MOVL b, R9; \
ANDL c, R9; \
ORL R8, R9
#define FUNC4 FUNC2
#define MIX(a, b, c, d, e, const) \
......@@ -102,7 +102,7 @@ TEXT ·blockAMD64(SB),NOSPLIT,$64-32
MOVQ p_len+16(FP), DX
SHRQ $6, DX
SHLQ $6, DX
LEAQ (SI)(DX*1), DI
MOVL (0*4)(BP), AX
MOVL (1*4)(BP), BX
......@@ -141,7 +141,7 @@ loop:
ROUND1x(DX, BP, AX, BX, CX, 17)
ROUND1x(CX, DX, BP, AX, BX, 18)
ROUND1x(BX, CX, DX, BP, AX, 19)
ROUND2(AX, BX, CX, DX, BP, 20)
ROUND2(BP, AX, BX, CX, DX, 21)
ROUND2(DX, BP, AX, BX, CX, 22)
......@@ -162,7 +162,7 @@ loop:
ROUND2(DX, BP, AX, BX, CX, 37)
ROUND2(CX, DX, BP, AX, BX, 38)
ROUND2(BX, CX, DX, BP, AX, 39)
ROUND3(AX, BX, CX, DX, BP, 40)
ROUND3(BP, AX, BX, CX, DX, 41)
ROUND3(DX, BP, AX, BX, CX, 42)
......@@ -183,7 +183,7 @@ loop:
ROUND3(DX, BP, AX, BX, CX, 57)
ROUND3(CX, DX, BP, AX, BX, 58)
ROUND3(BX, CX, DX, BP, AX, 59)
ROUND4(AX, BX, CX, DX, BP, 60)
ROUND4(BP, AX, BX, CX, DX, 61)
ROUND4(DX, BP, AX, BX, CX, 62)
......
......@@ -55,7 +55,7 @@
MOVL b, R9; \
ANDL c, R9; \
ORL R8, R9
#define FUNC4 FUNC2
#define MIX(a, b, c, d, e, const) \
......@@ -97,7 +97,7 @@ TEXT ·block(SB),NOSPLIT,$64-16
MOVL p_len+8(FP), DX
SHRQ $6, DX
SHLQ $6, DX
LEAQ (SI)(DX*1), DI
MOVL (0*4)(R14), AX
MOVL (1*4)(R14), BX
......@@ -131,7 +131,7 @@ loop:
ROUND1x(DX, BP, AX, BX, CX, 17)
ROUND1x(CX, DX, BP, AX, BX, 18)
ROUND1x(BX, CX, DX, BP, AX, 19)
ROUND2(AX, BX, CX, DX, BP, 20)
ROUND2(BP, AX, BX, CX, DX, 21)
ROUND2(DX, BP, AX, BX, CX, 22)
......@@ -152,7 +152,7 @@ loop:
ROUND2(DX, BP, AX, BX, CX, 37)
ROUND2(CX, DX, BP, AX, BX, 38)
ROUND2(BX, CX, DX, BP, AX, 39)
ROUND3(AX, BX, CX, DX, BP, 40)
ROUND3(BP, AX, BX, CX, DX, 41)
ROUND3(DX, BP, AX, BX, CX, 42)
......@@ -173,7 +173,7 @@ loop:
ROUND3(DX, BP, AX, BX, CX, 57)
ROUND3(CX, DX, BP, AX, BX, 58)
ROUND3(BX, CX, DX, BP, AX, 59)
ROUND4(AX, BX, CX, DX, BP, 60)
ROUND4(BP, AX, BX, CX, DX, 61)
ROUND4(DX, BP, AX, BX, CX, 62)
......
......@@ -63,10 +63,10 @@
ORR Rt1<<24, Rt0, Rt0 ; \
MOVW.P Rt0, 4(Rw) ; \
ADD Rt0, Re, Re
// tmp := w[(i-3)&0xf] ^ w[(i-8)&0xf] ^ w[(i-14)&0xf] ^ w[(i)&0xf]
// w[i&0xf] = tmp<<1 | tmp>>(32-1)
// e += w[i&0xf]
// e += w[i&0xf]
#define SHUFFLE(Re) \
MOVW (-16*4)(Rw), Rt0 ; \
MOVW (-14*4)(Rw), Rt1 ; \
......@@ -167,7 +167,7 @@ loop1: ROUND1(Ra, Rb, Rc, Rd, Re)
ROUND1x(Rd, Re, Ra, Rb, Rc)
ROUND1x(Rc, Rd, Re, Ra, Rb)
ROUND1x(Rb, Rc, Rd, Re, Ra)
MOVW $0x6ED9EBA1, Rconst
MOVW $4, Rctr
loop2: ROUND2(Ra, Rb, Rc, Rd, Re)
......@@ -177,7 +177,7 @@ loop2: ROUND2(Ra, Rb, Rc, Rd, Re)
ROUND2(Rb, Rc, Rd, Re, Ra)
SUB.S $1, Rctr
BNE loop2
MOVW $0x8F1BBCDC, Rconst
MOVW $4, Rctr
loop3: ROUND3(Ra, Rb, Rc, Rd, Re)
......@@ -187,7 +187,7 @@ loop3: ROUND3(Ra, Rb, Rc, Rd, Re)
ROUND3(Rb, Rc, Rd, Re, Ra)
SUB.S $1, Rctr
BNE loop3
MOVW $0xCA62C1D6, Rconst
MOVW $4, Rctr
loop4: ROUND4(Ra, Rb, Rc, Rd, Re)
......
......@@ -63,7 +63,7 @@ loop:
ADDQ $16, DI
SUBQ $16, R8
JMP loop
diff64:
ADDQ $48, SI
ADDQ $48, DI
......
......@@ -62,7 +62,7 @@ loop:
ADDQ $16, DI
SUBQ $16, R8
JMP loop
// AX = bit mask of differences
diff16:
BSFQ AX, BX // index of first byte that differs
......
......@@ -79,7 +79,7 @@ TEXT memeqbody<>(SB),NOSPLIT,$0-0
JB bigloop
CMPB internal∕cpu·X86+const_offsetX86HasAVX2(SB), $1
JE hugeloop_avx2
// 64 bytes at a time using xmm registers
hugeloop:
CMPQ BX, $64
......
......@@ -80,7 +80,7 @@ TEXT memeqbody<>(SB),NOSPLIT,$0-0
CMPQ BX, $8
JB small
// 64 bytes at a time using xmm registers
hugeloop:
CMPQ BX, $64
......
......@@ -9,7 +9,7 @@
TEXT ·Equal(SB),NOSPLIT,$0-25
MOVW a_len+4(FP), R1
MOVW b_len+16(FP), R3
CMP R1, R3 // unequal lengths are not equal
B.NE notequal
......
......@@ -183,7 +183,7 @@ L9: MOVL AX, DX // w = w1
SHRL CX, DX:AX // w>>s | w1<<ŝ
MOVL DX, (DI)(BX*4) // z[i] = w>>s | w1<<ŝ
ADDL $1, BX // i++
E9: CMPL BX, BP
JL L9 // i < n-1
......
......@@ -324,10 +324,10 @@ TEXT ·mulAddVWW(SB),NOSPLIT,$0
MOVQ r+56(FP), CX // c = r
MOVQ z_len+8(FP), R11
MOVQ $0, BX // i = 0
CMPQ R11, $4
JL E5
U5: // i+4 <= n
// regular loop body unrolled 4x
MOVQ (0*8)(R8)(BX*8), AX
......@@ -355,7 +355,7 @@ U5: // i+4 <= n
MOVQ AX, (3*8)(R10)(BX*8)
MOVQ DX, CX
ADDQ $4, BX // i += 4
LEAQ 4(BX), DX
CMPQ DX, R11
JLE U5
......
......@@ -123,7 +123,7 @@ TEXT ·shlVU(SB),NOSPLIT,$0
MOVW z_len+4(FP), R5
TEQ $0, R5
BEQ X7
MOVW z+0(FP), R1
MOVW x+12(FP), R2
ADD R5<<2, R2, R2
......@@ -135,7 +135,7 @@ TEXT ·shlVU(SB),NOSPLIT,$0
MOVW $32, R4
SUB R3, R4
MOVW $0, R7
MOVW.W -4(R2), R6
MOVW R6<<R3, R7
MOVW R6>>R4, R6
......
......@@ -54,7 +54,7 @@ TEXT ·divWW(SB),NOSPLIT,$0
TEXT ·addVV(SB),NOSPLIT,$0
MOVD addvectorfacility+0x00(SB),R1
BR (R1)
TEXT ·addVV_check(SB),NOSPLIT, $0
MOVB ·hasVX(SB), R1
CMPBEQ R1, $1, vectorimpl // vectorfacility = 1, vector supported
......@@ -89,7 +89,7 @@ TEXT ·addVV_vec(SB),NOSPLIT,$0
BLT v1
SUB $12, R3 // n -= 16
BLT A1 // if n < 0 goto A1
MOVD R8, R5
MOVD R9, R6
MOVD R2, R7
......@@ -291,7 +291,7 @@ E1n: NEG R4, R4
TEXT ·subVV(SB),NOSPLIT,$0
MOVD subvectorfacility+0x00(SB),R1
BR (R1)
TEXT ·subVV_check(SB),NOSPLIT,$0
MOVB ·hasVX(SB), R1
CMPBEQ R1, $1, vectorimpl // vectorfacility = 1, vector supported
......@@ -321,7 +321,7 @@ TEXT ·subVV_vec(SB),NOSPLIT,$0
MOVD $0, R4 // c = 0
MOVD $0, R0 // make sure it's zero
MOVD $0, R10 // i = 0
// s/JL/JMP/ below to disable the unrolled loop
SUB $4, R3 // n -= 4
BLT v1 // if n < 0 goto v1
......@@ -413,7 +413,7 @@ UU1: VLM 0(R5), V1, V4 // 64-bytes into V1..V8
A1: ADD $12, R3 // n += 16
BLT v1 // if n < 0 goto v1
U1: // n >= 0
// regular loop body unrolled 4x
MOVD 0(R8)(R10*1), R5
......@@ -532,7 +532,7 @@ E1: NEG R4, R4
TEXT ·addVW(SB),NOSPLIT,$0
MOVD addwvectorfacility+0x00(SB),R1
BR (R1)
TEXT ·addVW_check(SB),NOSPLIT,$0
MOVB ·hasVX(SB), R1
CMPBEQ R1, $1, vectorimpl // vectorfacility = 1, vector supported
......@@ -742,7 +742,7 @@ E4: MOVD R4, c+56(FP) // return c
TEXT ·subVW(SB),NOSPLIT,$0
MOVD subwvectorfacility+0x00(SB),R1
BR (R1)
TEXT ·subVW_check(SB),NOSPLIT,$0
MOVB ·hasVX(SB), R1
CMPBEQ R1, $1, vectorimpl // vectorfacility = 1, vector supported
......
......@@ -8,7 +8,7 @@
TEXT ·Expm1(SB),NOSPLIT,$0
FLDLN2 // F0=log(2) = 1/log2(e) ~ 0.693147
FMOVD x+0(FP), F0 // F0=x, F1=1/log2(e)
FABS // F0=|x|, F1=1/log2(e)
FABS // F0=|x|, F1=1/log2(e)
FUCOMPP F0, F1 // compare F0 to F1
FSTSW AX
SAHF
......@@ -36,7 +36,7 @@ use_exp:
FSCALE // F0=e**x, F1=int(x*log2(e))
FMOVDP F0, F1 // F0=e**x
FLD1 // F0=1, F1=e**x
FSUBDP F0, F1 // F0=e**x-1
FSUBDP F0, F1 // F0=e**x-1
FMOVDP F0, ret+8(FP)
RET
not_finite:
......
......@@ -24,7 +24,7 @@ TEXT ·Cos(SB),NOSPLIT,$0
FCOS // F0=cos(reduced_x)
FMOVDP F0, ret+8(FP)
RET
// func Sin(x float64) float64
TEXT ·Sin(SB),NOSPLIT,$0
FMOVD x+0(FP), F0 // F0=x
......
......@@ -4,7 +4,7 @@
#include "textflag.h"
// func Sqrt(x float64) float64
// func Sqrt(x float64) float64
TEXT ·Sqrt(SB),NOSPLIT,$0
FMOVD x+0(FP),F0
FSQRT
......
......@@ -4,7 +4,7 @@
#include "textflag.h"
// func Sqrt(x float64) float64
// func Sqrt(x float64) float64
TEXT ·Sqrt(SB),NOSPLIT,$0
MOVB runtime·goarm(SB), R11
CMP $5, R11
......
......@@ -107,7 +107,7 @@ TEXT runtime·rt0_go(SB),NOSPLIT,$0
MOVL BX, g_stackguard1(BP)
MOVL BX, (g_stack+stack_lo)(BP)
MOVL SP, (g_stack+stack_hi)(BP)
// find out information about the processor we're on
#ifdef GOOS_nacl // NaCl doesn't like PUSHFL/POPFL
JMP has_cpuid
......@@ -827,7 +827,7 @@ havem:
MOVL (g_sched+gobuf_sp)(SI), SP
MOVL 0(SP), AX
MOVL AX, (g_sched+gobuf_sp)(SI)
// If the m on entry was nil, we called needm above to borrow an m
// for the duration of the call. Since the call is over, return it with dropm.
CMPL DX, $0
......@@ -942,7 +942,7 @@ TEXT runtime·aeshashbody(SB),NOSPLIT,$0-0
CMPL BX, $64
JBE aes33to64
JMP aes65plus
aes0to15:
TESTL BX, BX
JE aes0
......@@ -957,7 +957,7 @@ aes0to15:
ADDL BX, BX
PAND masks<>(SB)(BX*8), X1
final1:
final1:
AESENC X0, X1 // scramble input, xor in seed
AESENC X1, X1 // scramble combo 2 times
AESENC X1, X1
......@@ -987,7 +987,7 @@ aes17to32:
// make second starting seed
PXOR runtime·aeskeysched+16(SB), X1
AESENC X1, X1
// load data to be hashed
MOVOU (AX), X2
MOVOU -16(AX)(BX*1), X3
......@@ -1015,22 +1015,22 @@ aes33to64:
AESENC X1, X1
AESENC X2, X2
AESENC X3, X3
MOVOU (AX), X4
MOVOU 16(AX), X5
MOVOU -32(AX)(BX*1), X6
MOVOU -16(AX)(BX*1), X7
AESENC X0, X4
AESENC X1, X5
AESENC X2, X6
AESENC X3, X7
AESENC X4, X4
AESENC X5, X5
AESENC X6, X6
AESENC X7, X7
AESENC X4, X4
AESENC X5, X5
AESENC X6, X6
......@@ -1052,7 +1052,7 @@ aes65plus:
AESENC X1, X1
AESENC X2, X2
AESENC X3, X3
// start with last (possibly overlapping) block
MOVOU -64(AX)(BX*1), X4
MOVOU -48(AX)(BX*1), X5
......@@ -1068,7 +1068,7 @@ aes65plus:
// compute number of remaining 64-byte blocks
DECL BX
SHRL $6, BX
aesloop:
// scramble state, xor in a block
MOVOU (AX), X0
......@@ -1095,7 +1095,7 @@ aesloop:
AESENC X5, X5
AESENC X6, X6
AESENC X7, X7
AESENC X4, X4
AESENC X5, X5
AESENC X6, X6
......@@ -1132,77 +1132,77 @@ DATA masks<>+0x00(SB)/4, $0x00000000
DATA masks<>+0x04(SB)/4, $0x00000000
DATA masks<>+0x08(SB)/4, $0x00000000
DATA masks<>+0x0c(SB)/4, $0x00000000
DATA masks<>+0x10(SB)/4, $0x000000ff
DATA masks<>+0x14(SB)/4, $0x00000000
DATA masks<>+0x18(SB)/4, $0x00000000
DATA masks<>+0x1c(SB)/4, $0x00000000
DATA masks<>+0x20(SB)/4, $0x0000ffff
DATA masks<>+0x24(SB)/4, $0x00000000
DATA masks<>+0x28(SB)/4, $0x00000000
DATA masks<>+0x2c(SB)/4, $0x00000000
DATA masks<>+0x30(SB)/4, $0x00ffffff
DATA masks<>+0x34(SB)/4, $0x00000000
DATA masks<>+0x38(SB)/4, $0x00000000
DATA masks<>+0x3c(SB)/4, $0x00000000
DATA masks<>+0x40(SB)/4, $0xffffffff
DATA masks<>+0x44(SB)/4, $0x00000000
DATA masks<>+0x48(SB)/4, $0x00000000
DATA masks<>+0x4c(SB)/4, $0x00000000
DATA masks<>+0x50(SB)/4, $0xffffffff
DATA masks<>+0x54(SB)/4, $0x000000ff
DATA masks<>+0x58(SB)/4, $0x00000000
DATA masks<>+0x5c(SB)/4, $0x00000000
DATA masks<>+0x60(SB)/4, $0xffffffff
DATA masks<>+0x64(SB)/4, $0x0000ffff
DATA masks<>+0x68(SB)/4, $0x00000000
DATA masks<>+0x6c(SB)/4, $0x00000000
DATA masks<>+0x70(SB)/4, $0xffffffff
DATA masks<>+0x74(SB)/4, $0x00ffffff
DATA masks<>+0x78(SB)/4, $0x00000000
DATA masks<>+0x7c(SB)/4, $0x00000000
DATA masks<>+0x80(SB)/4, $0xffffffff
DATA masks<>+0x84(SB)/4, $0xffffffff
DATA masks<>+0x88(SB)/4, $0x00000000
DATA masks<>+0x8c(SB)/4, $0x00000000
DATA masks<>+0x90(SB)/4, $0xffffffff
DATA masks<>+0x94(SB)/4, $0xffffffff
DATA masks<>+0x98(SB)/4, $0x000000ff
DATA masks<>+0x9c(SB)/4, $0x00000000
DATA masks<>+0xa0(SB)/4, $0xffffffff
DATA masks<>+0xa4(SB)/4, $0xffffffff
DATA masks<>+0xa8(SB)/4, $0x0000ffff
DATA masks<>+0xac(SB)/4, $0x00000000
DATA masks<>+0xb0(SB)/4, $0xffffffff
DATA masks<>+0xb4(SB)/4, $0xffffffff
DATA masks<>+0xb8(SB)/4, $0x00ffffff
DATA masks<>+0xbc(SB)/4, $0x00000000
DATA masks<>+0xc0(SB)/4, $0xffffffff
DATA masks<>+0xc4(SB)/4, $0xffffffff
DATA masks<>+0xc8(SB)/4, $0xffffffff
DATA masks<>+0xcc(SB)/4, $0x00000000
DATA masks<>+0xd0(SB)/4, $0xffffffff
DATA masks<>+0xd4(SB)/4, $0xffffffff
DATA masks<>+0xd8(SB)/4, $0xffffffff
DATA masks<>+0xdc(SB)/4, $0x000000ff
DATA masks<>+0xe0(SB)/4, $0xffffffff
DATA masks<>+0xe4(SB)/4, $0xffffffff
DATA masks<>+0xe8(SB)/4, $0xffffffff
DATA masks<>+0xec(SB)/4, $0x0000ffff
DATA masks<>+0xf0(SB)/4, $0xffffffff
DATA masks<>+0xf4(SB)/4, $0xffffffff
DATA masks<>+0xf8(SB)/4, $0xffffffff
......@@ -1217,77 +1217,77 @@ DATA shifts<>+0x00(SB)/4, $0x00000000
DATA shifts<>+0x04(SB)/4, $0x00000000
DATA shifts<>+0x08(SB)/4, $0x00000000
DATA shifts<>+0x0c(SB)/4, $0x00000000
DATA shifts<>+0x10(SB)/4, $0xffffff0f
DATA shifts<>+0x14(SB)/4, $0xffffffff
DATA shifts<>+0x18(SB)/4, $0xffffffff
DATA shifts<>+0x1c(SB)/4, $0xffffffff
DATA shifts<>+0x20(SB)/4, $0xffff0f0e
DATA shifts<>+0x24(SB)/4, $0xffffffff
DATA shifts<>+0x28(SB)/4, $0xffffffff
DATA shifts<>+0x2c(SB)/4, $0xffffffff
DATA shifts<>+0x30(SB)/4, $0xff0f0e0d
DATA shifts<>+0x34(SB)/4, $0xffffffff
DATA shifts<>+0x38(SB)/4, $0xffffffff
DATA shifts<>+0x3c(SB)/4, $0xffffffff
DATA shifts<>+0x40(SB)/4, $0x0f0e0d0c
DATA shifts<>+0x44(SB)/4, $0xffffffff
DATA shifts<>+0x48(SB)/4, $0xffffffff
DATA shifts<>+0x4c(SB)/4, $0xffffffff
DATA shifts<>+0x50(SB)/4, $0x0e0d0c0b
DATA shifts<>+0x54(SB)/4, $0xffffff0f
DATA shifts<>+0x58(SB)/4, $0xffffffff
DATA shifts<>+0x5c(SB)/4, $0xffffffff
DATA shifts<>+0x60(SB)/4, $0x0d0c0b0a
DATA shifts<>+0x64(SB)/4, $0xffff0f0e
DATA shifts<>+0x68(SB)/4, $0xffffffff
DATA shifts<>+0x6c(SB)/4, $0xffffffff
DATA shifts<>+0x70(SB)/4, $0x0c0b0a09
DATA shifts<>+0x74(SB)/4, $0xff0f0e0d
DATA shifts<>+0x78(SB)/4, $0xffffffff
DATA shifts<>+0x7c(SB)/4, $0xffffffff
DATA shifts<>+0x80(SB)/4, $0x0b0a0908
DATA shifts<>+0x84(SB)/4, $0x0f0e0d0c
DATA shifts<>+0x88(SB)/4, $0xffffffff
DATA shifts<>+0x8c(SB)/4, $0xffffffff
DATA shifts<>+0x90(SB)/4, $0x0a090807
DATA shifts<>+0x94(SB)/4, $0x0e0d0c0b
DATA shifts<>+0x98(SB)/4, $0xffffff0f
DATA shifts<>+0x9c(SB)/4, $0xffffffff
DATA shifts<>+0xa0(SB)/4, $0x09080706
DATA shifts<>+0xa4(SB)/4, $0x0d0c0b0a
DATA shifts<>+0xa8(SB)/4, $0xffff0f0e
DATA shifts<>+0xac(SB)/4, $0xffffffff
DATA shifts<>+0xb0(SB)/4, $0x08070605
DATA shifts<>+0xb4(SB)/4, $0x0c0b0a09
DATA shifts<>+0xb8(SB)/4, $0xff0f0e0d
DATA shifts<>+0xbc(SB)/4, $0xffffffff
DATA shifts<>+0xc0(SB)/4, $0x07060504
DATA shifts<>+0xc4(SB)/4, $0x0b0a0908
DATA shifts<>+0xc8(SB)/4, $0x0f0e0d0c
DATA shifts<>+0xcc(SB)/4, $0xffffffff
DATA shifts<>+0xd0(SB)/4, $0x06050403
DATA shifts<>+0xd4(SB)/4, $0x0a090807
DATA shifts<>+0xd8(SB)/4, $0x0e0d0c0b
DATA shifts<>+0xdc(SB)/4, $0xffffff0f
DATA shifts<>+0xe0(SB)/4, $0x05040302
DATA shifts<>+0xe4(SB)/4, $0x09080706
DATA shifts<>+0xe8(SB)/4, $0x0d0c0b0a
DATA shifts<>+0xec(SB)/4, $0xffff0f0e
DATA shifts<>+0xf0(SB)/4, $0x04030201
DATA shifts<>+0xf4(SB)/4, $0x08070605
DATA shifts<>+0xf8(SB)/4, $0x0c0b0a09
......
......@@ -92,7 +92,7 @@ TEXT runtime·rt0_go(SB),NOSPLIT,$0
ANDQ $~15, SP
MOVQ AX, 16(SP)
MOVQ BX, 24(SP)
// create istack out of the given (operating system) stack.
// _cgo_init may update stackguard.
MOVQ $runtime·g0(SB), DI
......@@ -273,7 +273,7 @@ TEXT runtime·gogo(SB), NOSPLIT, $16-8
// to keep running g.
TEXT runtime·mcall(SB), NOSPLIT, $0-8
MOVQ fn+0(FP), DI
get_tls(CX)
MOVQ g(CX), AX // save state in g->sched
MOVQ 0(SP), BX // caller's PC
......@@ -617,7 +617,7 @@ TEXT ·asmcgocall(SB),NOSPLIT,$0-20
MOVQ m_gsignal(R8), SI
CMPQ SI, DI
JEQ nosave
// Switch to system stack.
MOVQ m_g0(R8), SI
CALL gosave<>(SB)
......@@ -717,7 +717,7 @@ needm:
get_tls(CX)
MOVQ g(CX), BX
MOVQ g_m(BX), BX
// Set m->sched.sp = SP, so that if a panic happens
// during the function we are about to execute, it will
// have a valid SP to run on the g0 stack.
......@@ -801,7 +801,7 @@ havem:
MOVQ (g_sched+gobuf_sp)(SI), SP
MOVQ 0(SP), AX
MOVQ AX, (g_sched+gobuf_sp)(SI)
// If the m on entry was nil, we called needm above to borrow an m
// for the duration of the call. Since the call is over, return it with dropm.
CMPQ R8, $0
......@@ -953,7 +953,7 @@ aes17to32:
// make second starting seed
PXOR runtime·aeskeysched+16(SB), X1
AESENC X1, X1
// load data to be hashed
MOVOU (AX), X2
MOVOU -16(AX)(CX*1), X3
......@@ -985,7 +985,7 @@ aes33to64:
AESENC X1, X1
AESENC X2, X2
AESENC X3, X3
MOVOU (AX), X4
MOVOU 16(AX), X5
MOVOU -32(AX)(CX*1), X6
......@@ -995,17 +995,17 @@ aes33to64:
PXOR X1, X5
PXOR X2, X6
PXOR X3, X7
AESENC X4, X4
AESENC X5, X5
AESENC X6, X6
AESENC X7, X7
AESENC X4, X4
AESENC X5, X5
AESENC X6, X6
AESENC X7, X7
AESENC X4, X4
AESENC X5, X5
AESENC X6, X6
......@@ -1121,7 +1121,7 @@ aes129plus:
AESENC X5, X5
AESENC X6, X6
AESENC X7, X7
// start with last (possibly overlapping) block
MOVOU -128(AX)(CX*1), X8
MOVOU -112(AX)(CX*1), X9
......@@ -1141,11 +1141,11 @@ aes129plus:
PXOR X5, X13
PXOR X6, X14
PXOR X7, X15
// compute number of remaining 128-byte blocks
DECQ CX
SHRQ $7, CX
aesloop:
// scramble state
AESENC X8, X8
......
......@@ -18,7 +18,7 @@ TEXT runtime·rt0_go(SB),NOSPLIT,$0
MOVL AX, 16(SP)
MOVL BX, 24(SP)
// create istack out of the given (operating system) stack.
MOVL $runtime·g0(SB), DI
LEAL (-64*1024+104)(SP), BX
......@@ -150,7 +150,7 @@ TEXT runtime·gogo(SB), NOSPLIT, $8-4
// to keep running g.
TEXT runtime·mcall(SB), NOSPLIT, $0-4
MOVL fn+0(FP), DI
get_tls(CX)
MOVL g(CX), AX // save state in g->sched
MOVL 0(SP), BX // caller's PC
......
......@@ -514,7 +514,7 @@ TEXT callRet<>(SB), NOSPLIT, $16-0
MOVW R1, 12(R13)
MOVW R2, 16(R13)
BL runtime·reflectcallmove(SB)
RET
RET
CALLFN(·call16, 16)
CALLFN(·call32, 32)
......@@ -673,7 +673,7 @@ TEXT runtime·cgocallback(SB),NOSPLIT,$16-16
// See cgocall.go for more details.
TEXT ·cgocallback_gofunc(SB),NOSPLIT,$8-16
NO_LOCAL_POINTERS
// Load m and g from thread-local storage.
MOVB runtime·iscgo(SB), R0
CMP $0, R0
......@@ -855,12 +855,12 @@ TEXT _cgo_topofstack(SB),NOSPLIT,$8
// callee-save in the gcc calling convention, so save them here.
MOVW R11, saveR11-4(SP)
MOVW g, saveG-8(SP)
BL runtime·load_g(SB)
MOVW g_m(g), R0
MOVW m_curg(R0), R0
MOVW (g_stack+stack_hi)(R0), R0
MOVW saveG-8(SP), g
MOVW saveR11-4(SP), R11
RET
......
......@@ -36,9 +36,9 @@ TEXT crosscall2(SB),NOSPLIT,$0x110-0 /* also need to save xmm6 - xmm15 */
MOVQ DX, 0x0(SP) /* arg */
MOVQ R8, 0x8(SP) /* argsize (includes padding) */
MOVQ R9, 0x10(SP) /* ctxt */
CALL CX /* fn */
MOVQ 0x48(SP), DI
MOVQ 0x50(SP), SI
MOVUPS 0x60(SP), X6
......@@ -64,5 +64,5 @@ TEXT crosscall2(SB),NOSPLIT,$0x110-0 /* also need to save xmm6 - xmm15 */
MOVQ 0x30(SP), R13
MOVQ 0x38(SP), R14
MOVQ 0x40(SP), R15
RET
......@@ -8,7 +8,7 @@
// func crosscall2(fn func(a unsafe.Pointer, n int32, ctxt uintptr), a unsafe.Pointer, n int32, ctxt uintptr)
// Saves C callee-saved registers and calls fn with three arguments.
TEXT crosscall2(SB),NOSPLIT|NOFRAME,$0
/*
/*
* We still need to save all callee save register as before, and then
* push 3 args for fn (R1, R2, R3).
* Also note that at procedure entry in gc world, 4(R13) will be the
......
......@@ -37,7 +37,7 @@ ongothread:
// Build a 32-byte stack frame for us for this call.
// Saved LR (none available) is at the bottom,
// then the PC argument for setsigsegv,
// then the PC argument for setsigsegv,
// then a copy of the LR for us to restore.
MOVD.W $0, -32(RSP)
MOVD R1, 8(RSP)
......
......@@ -34,7 +34,7 @@ back:
ADDL BX, DI
ADDL BX, SI
STD
MOVL BX, CX
SHRL $2, CX
ANDL $3, BX
......
......@@ -138,7 +138,7 @@ _f32loop:
CMP TMP, TS
BHS _f4tail
MOVM.IA.W (FROM), [R1-R8]
MOVM.IA.W (FROM), [R1-R8]
MOVM.IA.W [R1-R8], (TS)
B _f32loop
......
......@@ -56,7 +56,7 @@ tail:
/*
* forward copy loop
*/
forward:
forward:
MOVL BX, CX
SHRL $2, CX
ANDL $3, BX
......
......@@ -73,7 +73,7 @@ back:
ADDQ BX, CX
CMPQ CX, DI
JLS forward
/*
* whole thing backwards has
* adjusted addresses
......
......@@ -12,7 +12,7 @@ TEXT _rt0_ppc64le_linux_lib(SB),NOSPLIT,$-8
MOVW CR, R0 // Save CR in caller's frame
MOVD R0, 8(R1)
MOVDU R1, -320(R1) // Allocate frame.
// Preserve callee-save registers.
MOVD R14, 24(R1)
MOVD R15, 32(R1)
......
......@@ -11,7 +11,7 @@
// 8(DI) - argc
// 12(DI) - argv, then 0, then envv, then 0, then auxv
// NaCl entry here is almost the same, except that there
// is no saved caller PC, so 0(FP) is -8(FP) and so on.
// is no saved caller PC, so 0(FP) is -8(FP) and so on.
TEXT _rt0_amd64p32_nacl(SB),NOSPLIT,$16
MOVL DI, 0(SP)
CALL runtime·nacl_sysinfo(SB)
......
......@@ -9,7 +9,7 @@
#include "go_asm.h"
#include "go_tls.h"
#include "textflag.h"
TEXT runtime·sys_umtx_sleep(SB),NOSPLIT,$0
MOVQ addr+0(FP), DI // arg 1 - ptr
MOVL val+8(FP), SI // arg 2 - value
......
......@@ -9,7 +9,7 @@
#include "go_asm.h"
#include "go_tls.h"
#include "textflag.h"
TEXT runtime·sys_umtx_op(SB),NOSPLIT,$-4
MOVL $454, AX
INT $0x80
......@@ -39,7 +39,7 @@ TEXT runtime·thr_start(SB),NOSPLIT,$0
POPAL
get_tls(CX)
MOVL BX, g(CX)
MOVL AX, g_m(BX)
CALL runtime·stackcheck(SB) // smashes AX
CALL runtime·mstart(SB)
......
......@@ -608,7 +608,7 @@ TEXT runtime·settls(SB),NOSPLIT,$32
// Same as in sys_darwin_386.s:/ugliness, different constant.
// DI currently holds m->tls, which must be fs:0x1d0.
// See cgo/gcc_android_amd64.c for the derivation of the constant.
SUBQ $0x1d0, DI // In android, the tls base
SUBQ $0x1d0, DI // In android, the tls base
#else
ADDQ $8, DI // ELF wants to use -8(FS)
#endif
......
......@@ -266,7 +266,7 @@ TEXT runtime·nacl_clock_gettime(SB),NOSPLIT,$8
NACL_SYSCALL(SYS_clock_gettime)
MOVL AX, ret+8(FP)
RET
TEXT runtime·nanotime(SB),NOSPLIT,$20
MOVL $0, 0(SP) // real time clock
LEAL 8(SP), AX
......@@ -308,12 +308,12 @@ TEXT runtime·sigtramp(SB),NOSPLIT,$0
// save g
MOVL DI, 20(SP)
// g = m->gsignal
MOVL g_m(DI), BX
MOVL m_gsignal(BX), BX
MOVL BX, g(CX)
// copy arguments for sighandler
MOVL $11, 0(SP) // signal
MOVL $0, 4(SP) // siginfo
......@@ -356,7 +356,7 @@ ret:
// Today those registers are just PC and SP, but in case additional registers
// are relevant in the future (for example DX is the Go func context register)
// we restore as many registers as possible.
//
//
// We smash BP, because that's what the linker smashes during RET.
//
LEAL ctxt+4(FP), BP
......
......@@ -334,13 +334,13 @@ TEXT runtime·sigtramp(SB),NOSPLIT,$80
// check that g exists
get_tls(CX)
MOVL g(CX), DI
CMPL DI, $0
JEQ nog
// save g
MOVL DI, 20(SP)
// g = m->gsignal
MOVL g_m(DI), BX
MOVL m_gsignal(BX), BX
......
......@@ -23,7 +23,7 @@ TEXT runtime·lwp_create(SB),NOSPLIT,$0
RET
TEXT runtime·lwp_tramp(SB),NOSPLIT,$0
// Set FS to point at m->tls.
LEAQ m_tls(R8), DI
CALL runtime·settls(SB)
......
......@@ -295,7 +295,7 @@ TEXT runtime·tfork(SB),NOSPLIT,$12
CALL runtime·settls(SB)
POPL AX
POPAL
// Now segment is established. Initialize m, g.
get_tls(AX)
MOVL DX, g(AX)
......
......@@ -126,7 +126,7 @@ TEXT runtime·noted(SB),NOSPLIT,$0
INT $64
MOVL AX, ret+4(FP)
RET
TEXT runtime·plan9_semrelease(SB),NOSPLIT,$0
MOVL $38, AX
INT $64
......
......@@ -123,7 +123,7 @@ TEXT runtime·noted(SB),NOSPLIT,$0
SYSCALL
MOVL AX, ret+8(FP)
RET
TEXT runtime·plan9_semrelease(SB),NOSPLIT,$0
MOVQ $38, BP
SYSCALL
......
......@@ -63,9 +63,9 @@ TEXT runtime·pipe1(SB),NOSPLIT,$0
// Call a library function with SysV calling conventions.
// The called function can take a maximum of 6 INTEGER class arguments,
// see
// see
// Michael Matz, Jan Hubicka, Andreas Jaeger, and Mark Mitchell
// System V Application Binary Interface
// System V Application Binary Interface
// AMD64 Architecture Processor Supplement
// section 3.2.3.
//
......@@ -119,7 +119,7 @@ skipargs:
MOVL 0(AX), AX
MOVQ AX, libcall_err(DI)
skiperrno2:
skiperrno2:
RET
// uint32 tstart_sysvicall(M *newm);
......@@ -186,7 +186,7 @@ allgood:
// Save m->libcall and m->scratch. We need to do this because we
// might get interrupted by a signal in runtime·asmcgocall.
// save m->libcall
// save m->libcall
MOVQ g_m(R10), BP
LEAQ m_libcall(BP), R11
MOVQ libcall_fn(R11), R10
......
......@@ -494,13 +494,13 @@ wall:
MOVL (_SYSTEM_TIME+time_hi2), DX
CMPL CX, DX
JNE wall
// w = DX:AX
// convert to Unix epoch (but still 100ns units)
#define delta 116444736000000000
SUBL $(delta & 0xFFFFFFFF), AX
SBBL $(delta >> 32), DX
// nano/100 = DX:AX
// split into two decimal halves by div 1e9.
// (decimal point is two spots over from correct place,
......@@ -509,7 +509,7 @@ wall:
DIVL CX
MOVL AX, DI
MOVL DX, SI
// DI = nano/100/1e9 = nano/1e11 = sec/100, DX = SI = nano/100%1e9
// split DX into seconds and nanoseconds by div 1e7 magic multiply.
MOVL DX, AX
......@@ -520,7 +520,7 @@ wall:
IMULL $10000000, DX
MOVL SI, CX
SUBL DX, CX
// DI = sec/100 (still)
// BX = (nano/100%1e9)/1e7 = (nano/1e9)%100 = sec%100
// CX = (nano/100%1e9)%1e7 = (nano%1e9)/100 = nsec/100
......
......@@ -89,7 +89,7 @@ TEXT runtime·badsignal2(SB),NOSPLIT|NOFRAME,$48
MOVQ $0, 32(SP) // overlapped
MOVQ runtime·_WriteFile(SB), AX
CALL AX
RET
// faster get/set last error
......
......@@ -30,7 +30,7 @@
// func runtime·udiv(n, d uint32) (q, r uint32)
// compiler knowns the register usage of this function
// Reference:
// Reference:
// Sloss, Andrew et. al; ARM System Developer's Guide: Designing and Optimizing System Software
// Morgan Kaufmann; 1 edition (April 8, 2004), ISBN 978-1558608740
#define Rq R0 // input d, output q
......
......@@ -42,4 +42,4 @@ ok:
MOVL DX, r2+20(FP)
MOVL $0, err+24(FP)
CALL runtime·exitsyscall(SB)
RET
RET
......@@ -39,4 +39,4 @@ ok:
MOVL DX, r2+20(FP)
MOVL $0, err+24(FP)
CALL runtime·exitsyscall(SB)
RET
RET
......@@ -41,4 +41,4 @@ ok:
MOVW $0, R2
MOVW R2, err+24(FP)
BL runtime·exitsyscall(SB)
RET
RET
......@@ -45,11 +45,11 @@ TEXT ·Syscall(SB),NOSPLIT,$148-32
CALL runtime·gostring(SB)
LEAL str-144(SP), SI
JMP copyresult3
ok3:
CALL runtime·exitsyscall(SB)
LEAL ·emptystring(SB), SI
copyresult3:
LEAL err+24(FP), DI
......@@ -78,7 +78,7 @@ TEXT ·Syscall6(SB),NOSPLIT,$148-44
MOVL $0, r2+32(FP)
CMPL AX, $-1
JNE ok4
LEAL errbuf-128(SP), AX
MOVL AX, sysargs-144(SP)
MOVL $128, sysargs1-140(SP)
......@@ -90,11 +90,11 @@ TEXT ·Syscall6(SB),NOSPLIT,$148-44
CALL runtime·gostring(SB)
LEAL str-144(SP), SI
JMP copyresult4
ok4:
CALL runtime·exitsyscall(SB)
LEAL ·emptystring(SB), SI
copyresult4:
LEAL err+36(FP), DI
......@@ -144,7 +144,7 @@ TEXT ·seek(SB),NOSPLIT,$24-36
NO_LOCAL_POINTERS
LEAL newoffset+20(FP), AX
MOVL AX, placeholder+0(FP)
// copy args down
LEAL placeholder+0(FP), SI
LEAL sysargs-20(SP), DI
......@@ -156,19 +156,19 @@ TEXT ·seek(SB),NOSPLIT,$24-36
MOVSL
MOVL $SYS_SEEK, AX // syscall entry
INT $64
CMPL AX, $-1
JNE ok6
MOVL AX, newoffset_lo+20(FP)
MOVL AX, newoffset_hi+24(FP)
CALL syscall·errstr(SB)
MOVL SP, SI
JMP copyresult6
ok6:
LEAL ·emptystring(SB), SI
copyresult6:
LEAL err+28(FP), DI
......
......@@ -44,11 +44,11 @@ TEXT ·Syscall(SB),NOSPLIT,$168-64
CALL runtime·gostring(SB)
LEAQ str-160(SP), SI
JMP copyresult3
ok3:
CALL runtime·exitsyscall(SB)
LEAQ ·emptystring(SB), SI
copyresult3:
LEAQ err+48(FP), DI
......@@ -77,7 +77,7 @@ TEXT ·Syscall6(SB),NOSPLIT,$168-88
MOVQ $0, r2+64(FP)
CMPL AX, $-1
JNE ok4
LEAQ errbuf-128(SP), AX
MOVQ AX, sysargs-160(SP)
MOVQ $128, sysargs1-152(SP)
......@@ -89,11 +89,11 @@ TEXT ·Syscall6(SB),NOSPLIT,$168-88
CALL runtime·gostring(SB)
LEAQ str-160(SP), SI
JMP copyresult4
ok4:
CALL runtime·exitsyscall(SB)
LEAQ ·emptystring(SB), SI
copyresult4:
LEAQ err+72(FP), DI
......@@ -143,7 +143,7 @@ TEXT ·seek(SB),NOSPLIT,$48-56
NO_LOCAL_POINTERS
LEAQ newoffset+32(FP), AX
MOVQ AX, placeholder+0(FP)
// copy args down
LEAQ placeholder+0(FP), SI
LEAQ sysargs-40(SP), DI
......@@ -155,18 +155,18 @@ TEXT ·seek(SB),NOSPLIT,$48-56
MOVSQ
MOVQ $SYS_SEEK, BP // syscall entry
SYSCALL
CMPL AX, $-1
JNE ok6
MOVQ AX, newoffset+32(FP)
CALL syscall·errstr(SB)
MOVQ SP, SI
JMP copyresult6
ok6:
LEAQ ·emptystring(SB), SI
copyresult6:
LEAQ err+40(FP), DI
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment