mirror of
git://sourceware.org/git/glibc.git
synced 2025-01-18 12:16:13 +08:00
Fix whitespace issues.
This commit is contained in:
parent
057edf90e0
commit
d6ac9329b3
@ -43,16 +43,16 @@
|
|||||||
.align 7
|
.align 7
|
||||||
|
|
||||||
EALIGN (BP_SYM (memcpy), 5, 0)
|
EALIGN (BP_SYM (memcpy), 5, 0)
|
||||||
CALL_MCOUNT
|
CALL_MCOUNT
|
||||||
|
|
||||||
dcbt 0,r4 /* Prefetch ONE SRC cacheline */
|
dcbt 0,r4 /* Prefetch ONE SRC cacheline */
|
||||||
cmplwi cr1,r5,16 /* is size < 16 ? */
|
cmplwi cr1,r5,16 /* is size < 16 ? */
|
||||||
mr r6,r3
|
mr r6,r3
|
||||||
blt+ cr1,.Lshortcopy
|
blt+ cr1,.Lshortcopy
|
||||||
|
|
||||||
.Lbigcopy:
|
.Lbigcopy:
|
||||||
neg r8,r3 /* LS 3 bits = # bytes to 8-byte dest bdry */
|
neg r8,r3 /* LS 3 bits = # bytes to 8-byte dest bdry */
|
||||||
clrlwi r8,r8,32-4 /* aling to 16byte boundary */
|
clrlwi r8,r8,32-4 /* aling to 16byte boundary */
|
||||||
sub r7,r4,r3
|
sub r7,r4,r3
|
||||||
cmplwi cr0,r8,0
|
cmplwi cr0,r8,0
|
||||||
beq+ .Ldst_aligned
|
beq+ .Ldst_aligned
|
||||||
@ -112,8 +112,8 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.LprefetchSRC:
|
.LprefetchSRC:
|
||||||
dcbt r12,r4
|
dcbt r12,r4
|
||||||
addi r12,r12,128
|
addi r12,r12,128
|
||||||
bdnz .LprefetchSRC
|
bdnz .LprefetchSRC
|
||||||
|
|
||||||
.Lnocacheprefetch:
|
.Lnocacheprefetch:
|
||||||
mtctr r7
|
mtctr r7
|
||||||
@ -122,7 +122,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
beq cr6,.Lcachelinealigned
|
beq cr6,.Lcachelinealigned
|
||||||
|
|
||||||
.Laligntocacheline:
|
.Laligntocacheline:
|
||||||
lfd fp9,0x08(r4)
|
lfd fp9,0x08(r4)
|
||||||
lfdu fp10,0x10(r4)
|
lfdu fp10,0x10(r4)
|
||||||
stfd fp9,0x08(r6)
|
stfd fp9,0x08(r6)
|
||||||
stfdu fp10,0x10(r6)
|
stfdu fp10,0x10(r6)
|
||||||
@ -131,10 +131,10 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.Lcachelinealigned: /* copy while cache lines */
|
.Lcachelinealigned: /* copy while cache lines */
|
||||||
|
|
||||||
blt- cr1,.Llessthancacheline /* size <128 */
|
blt- cr1,.Llessthancacheline /* size <128 */
|
||||||
|
|
||||||
.Louterloop:
|
.Louterloop:
|
||||||
cmpwi r11,0
|
cmpwi r11,0
|
||||||
mtctr r11
|
mtctr r11
|
||||||
beq- .Lendloop
|
beq- .Lendloop
|
||||||
|
|
||||||
@ -142,7 +142,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.align 4
|
.align 4
|
||||||
/* Copy whole cachelines, optimized by prefetching SRC cacheline */
|
/* Copy whole cachelines, optimized by prefetching SRC cacheline */
|
||||||
.Lloop: /* Copy aligned body */
|
.Lloop: /* Copy aligned body */
|
||||||
dcbt r12,r4 /* PREFETCH SOURCE some cache lines ahead */
|
dcbt r12,r4 /* PREFETCH SOURCE some cache lines ahead */
|
||||||
lfd fp9, 0x08(r4)
|
lfd fp9, 0x08(r4)
|
||||||
dcbz r11,r6
|
dcbz r11,r6
|
||||||
@ -186,7 +186,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
beq- .Lendloop2
|
beq- .Lendloop2
|
||||||
mtctr r10
|
mtctr r10
|
||||||
|
|
||||||
.Lloop2: /* Copy aligned body */
|
.Lloop2: /* Copy aligned body */
|
||||||
lfd fp9, 0x08(r4)
|
lfd fp9, 0x08(r4)
|
||||||
lfd fp10, 0x10(r4)
|
lfd fp10, 0x10(r4)
|
||||||
lfd fp11, 0x18(r4)
|
lfd fp11, 0x18(r4)
|
||||||
@ -206,7 +206,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
mtctr r7
|
mtctr r7
|
||||||
|
|
||||||
.Lcopy_remaining:
|
.Lcopy_remaining:
|
||||||
lfd fp9,0x08(r4)
|
lfd fp9,0x08(r4)
|
||||||
lfdu fp10,0x10(r4)
|
lfdu fp10,0x10(r4)
|
||||||
stfd fp9,0x08(r6)
|
stfd fp9,0x08(r6)
|
||||||
stfdu fp10,0x10(r6)
|
stfdu fp10,0x10(r6)
|
||||||
@ -214,7 +214,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.Ldo_lt16: /* less than 16 ? */
|
.Ldo_lt16: /* less than 16 ? */
|
||||||
cmplwi cr0,r5,0 /* copy remaining bytes (0-15) */
|
cmplwi cr0,r5,0 /* copy remaining bytes (0-15) */
|
||||||
beqlr+ /* no rest to copy */
|
beqlr+ /* no rest to copy */
|
||||||
addi r4,r4,8
|
addi r4,r4,8
|
||||||
addi r6,r6,8
|
addi r6,r6,8
|
||||||
|
|
||||||
|
@ -43,16 +43,16 @@
|
|||||||
.align 7
|
.align 7
|
||||||
|
|
||||||
EALIGN (BP_SYM (memcpy), 5, 0)
|
EALIGN (BP_SYM (memcpy), 5, 0)
|
||||||
CALL_MCOUNT 3
|
CALL_MCOUNT 3
|
||||||
|
|
||||||
dcbt 0,r4 /* Prefetch ONE SRC cacheline */
|
dcbt 0,r4 /* Prefetch ONE SRC cacheline */
|
||||||
cmpldi cr1,r5,16 /* is size < 16 ? */
|
cmpldi cr1,r5,16 /* is size < 16 ? */
|
||||||
mr r6,r3
|
mr r6,r3
|
||||||
blt+ cr1,.Lshortcopy
|
blt+ cr1,.Lshortcopy
|
||||||
|
|
||||||
.Lbigcopy:
|
.Lbigcopy:
|
||||||
neg r8,r3 /* LS 3 bits = # bytes to 8-byte dest bdry */
|
neg r8,r3 /* LS 3 bits = # bytes to 8-byte dest bdry */
|
||||||
clrldi r8,r8,64-4 /* aling to 16byte boundary */
|
clrldi r8,r8,64-4 /* aling to 16byte boundary */
|
||||||
sub r7,r4,r3
|
sub r7,r4,r3
|
||||||
cmpldi cr0,r8,0
|
cmpldi cr0,r8,0
|
||||||
beq+ .Ldst_aligned
|
beq+ .Ldst_aligned
|
||||||
@ -112,8 +112,8 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.LprefetchSRC:
|
.LprefetchSRC:
|
||||||
dcbt r12,r4
|
dcbt r12,r4
|
||||||
addi r12,r12,128
|
addi r12,r12,128
|
||||||
bdnz .LprefetchSRC
|
bdnz .LprefetchSRC
|
||||||
|
|
||||||
.Lnocacheprefetch:
|
.Lnocacheprefetch:
|
||||||
mtctr r7
|
mtctr r7
|
||||||
@ -122,7 +122,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
beq cr6,.Lcachelinealigned
|
beq cr6,.Lcachelinealigned
|
||||||
|
|
||||||
.Laligntocacheline:
|
.Laligntocacheline:
|
||||||
ld r9,0x08(r4)
|
ld r9,0x08(r4)
|
||||||
ldu r7,0x10(r4)
|
ldu r7,0x10(r4)
|
||||||
std r9,0x08(r6)
|
std r9,0x08(r6)
|
||||||
stdu r7,0x10(r6)
|
stdu r7,0x10(r6)
|
||||||
@ -131,10 +131,10 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.Lcachelinealigned: /* copy while cache lines */
|
.Lcachelinealigned: /* copy while cache lines */
|
||||||
|
|
||||||
blt- cr1,.Llessthancacheline /* size <128 */
|
blt- cr1,.Llessthancacheline /* size <128 */
|
||||||
|
|
||||||
.Louterloop:
|
.Louterloop:
|
||||||
cmpdi r11,0
|
cmpdi r11,0
|
||||||
mtctr r11
|
mtctr r11
|
||||||
beq- .Lendloop
|
beq- .Lendloop
|
||||||
|
|
||||||
@ -142,7 +142,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.align 4
|
.align 4
|
||||||
/* Copy whole cachelines, optimized by prefetching SRC cacheline */
|
/* Copy whole cachelines, optimized by prefetching SRC cacheline */
|
||||||
.Lloop: /* Copy aligned body */
|
.Lloop: /* Copy aligned body */
|
||||||
dcbt r12,r4 /* PREFETCH SOURCE some cache lines ahead */
|
dcbt r12,r4 /* PREFETCH SOURCE some cache lines ahead */
|
||||||
ld r9, 0x08(r4)
|
ld r9, 0x08(r4)
|
||||||
dcbz r11,r6
|
dcbz r11,r6
|
||||||
@ -186,7 +186,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
beq- .Lendloop2
|
beq- .Lendloop2
|
||||||
mtctr r10
|
mtctr r10
|
||||||
|
|
||||||
.Lloop2: /* Copy aligned body */
|
.Lloop2: /* Copy aligned body */
|
||||||
ld r9, 0x08(r4)
|
ld r9, 0x08(r4)
|
||||||
ld r7, 0x10(r4)
|
ld r7, 0x10(r4)
|
||||||
ld r8, 0x18(r4)
|
ld r8, 0x18(r4)
|
||||||
@ -206,7 +206,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
mtctr r7
|
mtctr r7
|
||||||
|
|
||||||
.Lcopy_remaining:
|
.Lcopy_remaining:
|
||||||
ld r8,0x08(r4)
|
ld r8,0x08(r4)
|
||||||
ldu r7,0x10(r4)
|
ldu r7,0x10(r4)
|
||||||
std r8,0x08(r6)
|
std r8,0x08(r6)
|
||||||
stdu r7,0x10(r6)
|
stdu r7,0x10(r6)
|
||||||
@ -214,7 +214,7 @@ EALIGN (BP_SYM (memcpy), 5, 0)
|
|||||||
|
|
||||||
.Ldo_lt16: /* less than 16 ? */
|
.Ldo_lt16: /* less than 16 ? */
|
||||||
cmpldi cr0,r5,0 /* copy remaining bytes (0-15) */
|
cmpldi cr0,r5,0 /* copy remaining bytes (0-15) */
|
||||||
beqlr+ /* no rest to copy */
|
beqlr+ /* no rest to copy */
|
||||||
addi r4,r4,8
|
addi r4,r4,8
|
||||||
addi r6,r6,8
|
addi r6,r6,8
|
||||||
|
|
||||||
|
Loading…
Reference in New Issue
Block a user