/* Optimized 32-bit memset implementation for POWER6.
Copyright (C) 1997-2017 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
. */
#include
/* void * [r3] memset (void *s [r3], int c [r4], size_t n [r5]));
Returns 's'.
The memset is done in three sizes: byte (8 bits), word (32 bits),
cache line (1024 bits). There is a special case for setting cache lines
to 0, to take advantage of the dcbz instruction. */
.machine power6
EALIGN (memset, 7, 0)
CALL_MCOUNT
#define rTMP r0
#define rRTN r3 /* Initial value of 1st argument. */
#define rMEMP0 r3 /* Original value of 1st arg. */
#define rCHR r4 /* Char to set in each byte. */
#define rLEN r5 /* Length of region to set. */
#define rMEMP r6 /* Address at which we are storing. */
#define rALIGN r7 /* Number of bytes we are setting now (when aligning). */
#define rMEMP2 r8
#define rNEG64 r8 /* Constant -64 for clearing with dcbz. */
#define rMEMP3 r9 /* Alt mem pointer. */
L(_memset):
/* Take care of case for size <= 4. */
cmplwi cr1, rLEN, 4
andi. rALIGN, rMEMP0, 3
mr rMEMP, rMEMP0
ble- cr1, L(small)
/* Align to word boundary. */
cmplwi cr5, rLEN, 31
insrwi rCHR, rCHR, 8, 16 /* Replicate byte to halfword. */
beq+ L(aligned)
mtcrf 0x01, rMEMP0
subfic rALIGN, rALIGN, 4
add rMEMP, rMEMP, rALIGN
sub rLEN, rLEN, rALIGN
bf+ 31, L(g0)
stb rCHR, 0(rMEMP0)
bt 30, L(aligned)
L(g0):
sth rCHR, -2(rMEMP)
.align 4
/* Handle the case of size < 31. */
L(aligned):
mtcrf 0x01, rLEN
insrwi rCHR, rCHR, 16, 0 /* Replicate halfword to word. */
ble cr5, L(medium)
/* Align to 32-byte boundary. */
andi. rALIGN, rMEMP, 0x1C
subfic rALIGN, rALIGN, 0x20
beq L(caligned)
mtcrf 0x01, rALIGN
add rMEMP, rMEMP, rALIGN
sub rLEN, rLEN, rALIGN
cmplwi cr1, rALIGN, 0x10
mr rMEMP2, rMEMP
bf 28, L(a1)
stw rCHR, -4(rMEMP2)
stwu rCHR, -8(rMEMP2)
nop
L(a1): blt cr1, L(a2)
stw rCHR, -4(rMEMP2)
stw rCHR, -8(rMEMP2)
stw rCHR, -12(rMEMP2)
stwu rCHR, -16(rMEMP2)
L(a2): bf 29, L(caligned)
stw rCHR, -4(rMEMP2)
.align 3
/* Now aligned to a 32 byte boundary. */
L(caligned):
cmplwi cr1, rCHR, 0
clrrwi. rALIGN, rLEN, 5
mtcrf 0x01, rLEN
beq cr1, L(zloopstart) /* Special case for clearing memory using dcbz. */
L(nondcbz):
beq L(medium) /* We may not actually get to do a full line. */
nop
/* Storing a non-zero "c" value. We are aligned at a sector (32-byte)
boundary may not be at cache line (128-byte) boundary. */
L(nzloopstart):
/* memset in 32-byte chunks until we get to a cache line boundary.
If rLEN is less than the distance to the next cache-line boundary use
cacheAligned1 code to finish the tail. */
cmplwi cr1,rLEN,128
andi. rTMP,rMEMP,127
blt cr1,L(cacheAligned1)
addi rMEMP3,rMEMP,32
beq L(nzCacheAligned)
addi rLEN,rLEN,-32
stw rCHR,0(rMEMP)
stw rCHR,4(rMEMP)
stw rCHR,8(rMEMP)
stw rCHR,12(rMEMP)
stw rCHR,16(rMEMP)
stw rCHR,20(rMEMP)
addi rMEMP,rMEMP,32
andi. rTMP,rMEMP3,127
stw rCHR,-8(rMEMP3)
stw rCHR,-4(rMEMP3)
beq L(nzCacheAligned)
addi rLEN,rLEN,-32
stw rCHR,0(rMEMP3)
stw rCHR,4(rMEMP3)
addi rMEMP,rMEMP,32
stw rCHR,8(rMEMP3)
stw rCHR,12(rMEMP3)
andi. rTMP,rMEMP,127
stw rCHR,16(rMEMP3)
stw rCHR,20(rMEMP3)
stw rCHR,24(rMEMP3)
stw rCHR,28(rMEMP3)
beq L(nzCacheAligned)
addi rLEN,rLEN,-32
/* At this point we can overrun the store queue (pipe reject) so it is
time to slow things down. The store queue can merge two adjacent
stores into a single L1/L2 op, but the L2 is clocked at 1/2 the CPU.
So we add "group ending nops" to guarantee that we dispatch only two
stores every other cycle. */
ori r1,r1,0
ori r1,r1,0
stw rCHR,32(rMEMP3)
stw rCHR,36(rMEMP3)
addi rMEMP,rMEMP,32
cmplwi cr1,rLEN,128
ori r1,r1,0
ori r1,r1,0
stw rCHR,40(rMEMP3)
stw rCHR,44(rMEMP3)
ori r1,r1,0
ori r1,r1,0
stw rCHR,48(rMEMP3)
stw rCHR,52(rMEMP3)
ori r1,r1,0
ori r1,r1,0
stw rCHR,56(rMEMP3)
stw rCHR,60(rMEMP3)
blt cr1,L(cacheAligned1)
b L(nzCacheAligned)
/* Now we are aligned to the cache line and can use dcbtst. */
.align 5
L(nzCacheAligned):
cmplwi cr1,rLEN,128
cmplwi cr6,rLEN,256
blt cr1,L(cacheAligned1)
blt cr6,L(nzCacheAligned128)
.align 4
L(nzCacheAligned128):
nop
addi rMEMP3,rMEMP,64
stw rCHR,0(rMEMP)
stw rCHR,4(rMEMP)
stw rCHR,8(rMEMP)
stw rCHR,12(rMEMP)
stw rCHR,16(rMEMP)
stw rCHR,20(rMEMP)
stw rCHR,24(rMEMP)
stw rCHR,28(rMEMP)
stw rCHR,32(rMEMP)
stw rCHR,36(rMEMP)
stw rCHR,40(rMEMP)
stw rCHR,44(rMEMP)
stw rCHR,48(rMEMP)
stw rCHR,52(rMEMP)
stw rCHR,56(rMEMP)
stw rCHR,60(rMEMP)
addi rMEMP,rMEMP3,64
addi rLEN,rLEN,-128
/* At this point we can overrun the store queue (pipe reject) so it is
time to slow things down. The store queue can merge two adjacent
stores into a single L1/L2 op, but the L2 is clocked at 1/2 the CPU.
So we add "group ending nops" to guarantee that we dispatch only one
store per cycle. */
stw rCHR,0(rMEMP3)
ori r1,r1,0
stw rCHR,4(rMEMP3)
ori r1,r1,0
stw rCHR,8(rMEMP3)
ori r1,r1,0
stw rCHR,12(rMEMP3)
ori r1,r1,0
stw rCHR,16(rMEMP3)
ori r1,r1,0
stw rCHR,20(rMEMP3)
ori r1,r1,0
stw rCHR,24(rMEMP3)
ori r1,r1,0
stw rCHR,28(rMEMP3)
ori r1,r1,0
stw rCHR,32(rMEMP3)
ori r1,r1,0
stw rCHR,36(rMEMP3)
ori r1,r1,0
stw rCHR,40(rMEMP3)
ori r1,r1,0
stw rCHR,44(rMEMP3)
ori r1,r1,0
stw rCHR,48(rMEMP3)
ori r1,r1,0
stw rCHR,52(rMEMP3)
ori r1,r1,0
stw rCHR,56(rMEMP3)
ori r1,r1,0
stw rCHR,60(rMEMP3)
blt cr6,L(cacheAligned1)
#if IS_IN (libc)
lfd 0,-128(rMEMP)
#endif
b L(nzCacheAligned256)
.align 5
L(nzCacheAligned256):
cmplwi cr1,rLEN,256
addi rMEMP3,rMEMP,64
#if !IS_IN (libc)
/* When we are not in libc we should use only GPRs to avoid the FPU lock
interrupt. */
stw rCHR,0(rMEMP)
stw rCHR,4(rMEMP)
stw rCHR,8(rMEMP)
stw rCHR,12(rMEMP)
stw rCHR,16(rMEMP)
stw rCHR,20(rMEMP)
stw rCHR,24(rMEMP)
stw rCHR,28(rMEMP)
stw rCHR,32(rMEMP)
stw rCHR,36(rMEMP)
stw rCHR,40(rMEMP)
stw rCHR,44(rMEMP)
stw rCHR,48(rMEMP)
stw rCHR,52(rMEMP)
stw rCHR,56(rMEMP)
stw rCHR,60(rMEMP)
addi rMEMP,rMEMP3,64
addi rLEN,rLEN,-128
stw rCHR,0(rMEMP3)
stw rCHR,4(rMEMP3)
stw rCHR,8(rMEMP3)
stw rCHR,12(rMEMP3)
stw rCHR,16(rMEMP3)
stw rCHR,20(rMEMP3)
stw rCHR,24(rMEMP3)
stw rCHR,28(rMEMP3)
stw rCHR,32(rMEMP3)
stw rCHR,36(rMEMP3)
stw rCHR,40(rMEMP3)
stw rCHR,44(rMEMP3)
stw rCHR,48(rMEMP3)
stw rCHR,52(rMEMP3)
stw rCHR,56(rMEMP3)
stw rCHR,60(rMEMP3)
#else
/* We are in libc and this is a long memset so we can use FPRs and can afford
occasional FPU locked interrupts. */
stfd 0,0(rMEMP)
stfd 0,8(rMEMP)
stfd 0,16(rMEMP)
stfd 0,24(rMEMP)
stfd 0,32(rMEMP)
stfd 0,40(rMEMP)
stfd 0,48(rMEMP)
stfd 0,56(rMEMP)
addi rMEMP,rMEMP3,64
addi rLEN,rLEN,-128
stfd 0,0(rMEMP3)
stfd 0,8(rMEMP3)
stfd 0,16(rMEMP3)
stfd 0,24(rMEMP3)
stfd 0,32(rMEMP3)
stfd 0,40(rMEMP3)
stfd 0,48(rMEMP3)
stfd 0,56(rMEMP3)
#endif
bge cr1,L(nzCacheAligned256)
dcbtst 0,rMEMP
b L(cacheAligned1)
.align 4
/* Storing a zero "c" value. We are aligned at a sector (32-byte)
boundary but may not be at cache line (128-byte) boundary. If the
remaining length spans a full cache line we can use the Data cache
block zero instruction. */
L(zloopstart):
/* memset in 32-byte chunks until we get to a cache line boundary.
If rLEN is less than the distance to the next cache-line boundary use
cacheAligned1 code to finish the tail. */
cmplwi cr1,rLEN,128
beq L(medium)
L(getCacheAligned):
andi. rTMP,rMEMP,127
blt cr1,L(cacheAligned1)
addi rMEMP3,rMEMP,32
beq L(cacheAligned)
addi rLEN,rLEN,-32
stw rCHR,0(rMEMP)
stw rCHR,4(rMEMP)
stw rCHR,8(rMEMP)
stw rCHR,12(rMEMP)
stw rCHR,16(rMEMP)
stw rCHR,20(rMEMP)
addi rMEMP,rMEMP,32
andi. rTMP,rMEMP3,127
stw rCHR,-8(rMEMP3)
stw rCHR,-4(rMEMP3)
L(getCacheAligned2):
beq L(cacheAligned)
addi rLEN,rLEN,-32
addi rMEMP,rMEMP,32
stw rCHR,0(rMEMP3)
stw rCHR,4(rMEMP3)
stw rCHR,8(rMEMP3)
stw rCHR,12(rMEMP3)
andi. rTMP,rMEMP,127
nop
stw rCHR,16(rMEMP3)
stw rCHR,20(rMEMP3)
stw rCHR,24(rMEMP3)
stw rCHR,28(rMEMP3)
L(getCacheAligned3):
beq L(cacheAligned)
/* At this point we can overrun the store queue (pipe reject) so it is
time to slow things down. The store queue can merge two adjacent
stores into a single L1/L2 op, but the L2 is clocked at 1/2 the CPU.
So we add "group ending nops" to guarantee that we dispatch only two
stores every other cycle. */
addi rLEN,rLEN,-32
ori r1,r1,0
ori r1,r1,0
stw rCHR,32(rMEMP3)
stw rCHR,36(rMEMP3)
addi rMEMP,rMEMP,32
cmplwi cr1,rLEN,128
ori r1,r1,0
stw rCHR,40(rMEMP3)
stw rCHR,44(rMEMP3)
cmplwi cr6,rLEN,256
li rMEMP2,128
ori r1,r1,0
stw rCHR,48(rMEMP3)
stw rCHR,52(rMEMP3)
ori r1,r1,0
ori r1,r1,0
stw rCHR,56(rMEMP3)
stw rCHR,60(rMEMP3)
blt cr1,L(cacheAligned1)
blt cr6,L(cacheAligned128)
b L(cacheAlignedx)
/* Now we are aligned to the cache line and can use dcbz. */
.align 4
L(cacheAligned):
cmplwi cr1,rLEN,128
cmplwi cr6,rLEN,256
blt cr1,L(cacheAligned1)
li rMEMP2,128
L(cacheAlignedx):
cmplwi cr5,rLEN,640
blt cr6,L(cacheAligned128)
bgt cr5,L(cacheAligned512)
cmplwi cr6,rLEN,512
dcbz 0,rMEMP
cmplwi cr1,rLEN,384
dcbz rMEMP2,rMEMP
addi rMEMP,rMEMP,256
addi rLEN,rLEN,-256
blt cr1,L(cacheAligned1)
blt cr6,L(cacheAligned128)
b L(cacheAligned256)
.align 5
/* A simple loop for the longer (>640 bytes) lengths. This form limits
the branch miss-predicted to exactly 1 at loop exit.*/
L(cacheAligned512):
cmplwi cr1,rLEN,128
blt cr1,L(cacheAligned1)
dcbz 0,rMEMP
addi rLEN,rLEN,-128
addi rMEMP,rMEMP,128
b L(cacheAligned512)
.align 5
L(cacheAligned256):
cmplwi cr6,rLEN,512
dcbz 0,rMEMP
cmplwi cr1,rLEN,384
dcbz rMEMP2,rMEMP
addi rMEMP,rMEMP,256
addi rLEN,rLEN,-256
bge cr6,L(cacheAligned256)
blt cr1,L(cacheAligned1)
.align 4
L(cacheAligned128):
dcbz 0,rMEMP
addi rMEMP,rMEMP,128
addi rLEN,rLEN,-128
.align 4
L(cacheAligned1):
cmplwi cr1,rLEN,32
blt cr1,L(handletail32)
addi rMEMP3,rMEMP,32
addi rLEN,rLEN,-32
stw rCHR,0(rMEMP)
stw rCHR,4(rMEMP)
stw rCHR,8(rMEMP)
stw rCHR,12(rMEMP)
stw rCHR,16(rMEMP)
stw rCHR,20(rMEMP)
addi rMEMP,rMEMP,32
cmplwi cr1,rLEN,32
stw rCHR,-8(rMEMP3)
stw rCHR,-4(rMEMP3)
L(cacheAligned2):
blt cr1,L(handletail32)
addi rLEN,rLEN,-32
stw rCHR,0(rMEMP3)
stw rCHR,4(rMEMP3)
stw rCHR,8(rMEMP3)
stw rCHR,12(rMEMP3)
addi rMEMP,rMEMP,32
cmplwi cr1,rLEN,32
stw rCHR,16(rMEMP3)
stw rCHR,20(rMEMP3)
stw rCHR,24(rMEMP3)
stw rCHR,28(rMEMP3)
nop
L(cacheAligned3):
blt cr1,L(handletail32)
/* At this point we can overrun the store queue (pipe reject) so it is
time to slow things down. The store queue can merge two adjacent
stores into a single L1/L2 op, but the L2 is clocked at 1/2 the CPU.
So we add "group ending nops" to guarantee that we dispatch only two
stores every other cycle. */
ori r1,r1,0
ori r1,r1,0
addi rMEMP,rMEMP,32
addi rLEN,rLEN,-32
ori r1,r1,0
ori r1,r1,0
stw rCHR,32(rMEMP3)
stw rCHR,36(rMEMP3)
ori r1,r1,0
ori r1,r1,0
stw rCHR,40(rMEMP3)
stw rCHR,44(rMEMP3)
ori r1,r1,0
ori r1,r1,0
stw rCHR,48(rMEMP3)
stw rCHR,52(rMEMP3)
ori r1,r1,0
ori r1,r1,0
stw rCHR,56(rMEMP3)
stw rCHR,60(rMEMP3)
/* We are here because the length or remainder (rLEN) is less than the
cache line/sector size and does not justify aggressive loop unrolling.
So set up the preconditions for L(medium) and go there. */
.align 3
L(handletail32):
cmplwi cr1,rLEN,0
beqlr cr1
b L(medium)
.align 4
L(small):
/* Memset of 4 bytes or less. */
cmplwi cr5, rLEN, 1
cmplwi cr1, rLEN, 3
bltlr cr5
stb rCHR, 0(rMEMP)
beqlr cr5
stb rCHR, 1(rMEMP)
bltlr cr1
stb rCHR, 2(rMEMP)
beqlr cr1
stb rCHR, 3(rMEMP)
blr
/* Memset of 0-31 bytes. */
.align 5
L(medium):
cmplwi cr1, rLEN, 16
L(medium_tail2):
add rMEMP, rMEMP, rLEN
L(medium_tail):
bt- 31, L(medium_31t)
bt- 30, L(medium_30t)
L(medium_30f):
bt 29, L(medium_29t)
L(medium_29f):
bge cr1, L(medium_27t)
bflr 28
stw rCHR, -4(rMEMP)
stw rCHR, -8(rMEMP)
blr
L(medium_31t):
stbu rCHR, -1(rMEMP)
bf- 30, L(medium_30f)
L(medium_30t):
sthu rCHR, -2(rMEMP)
bf- 29, L(medium_29f)
L(medium_29t):
stwu rCHR, -4(rMEMP)
blt cr1, L(medium_27f)
L(medium_27t):
stw rCHR, -4(rMEMP)
stw rCHR, -8(rMEMP)
stw rCHR, -12(rMEMP)
stwu rCHR, -16(rMEMP)
L(medium_27f):
bflr 28
L(medium_28t):
stw rCHR, -4(rMEMP)
stw rCHR, -8(rMEMP)
blr
END (memset)
libc_hidden_builtin_def (memset)