glibc/sysdeps/powerpc/powerpc64/power7/memchr.S

200 lines
5.2 KiB
ArmAsm
Raw Normal View History

/* Optimized memchr implementation for PowerPC64/POWER7 using cmpb insn.
Copyright (C) 2010-2021 Free Software Foundation, Inc.
Contributed by Luis Machado <luisgpm@br.ibm.com>.
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
Prefer https to http for gnu.org and fsf.org URLs Also, change sources.redhat.com to sourceware.org. This patch was automatically generated by running the following shell script, which uses GNU sed, and which avoids modifying files imported from upstream: sed -ri ' s,(http|ftp)(://(.*\.)?(gnu|fsf|sourceware)\.org($|[^.]|\.[^a-z])),https\2,g s,(http|ftp)(://(.*\.)?)sources\.redhat\.com($|[^.]|\.[^a-z]),https\2sourceware.org\4,g ' \ $(find $(git ls-files) -prune -type f \ ! -name '*.po' \ ! -name 'ChangeLog*' \ ! -path COPYING ! -path COPYING.LIB \ ! -path manual/fdl-1.3.texi ! -path manual/lgpl-2.1.texi \ ! -path manual/texinfo.tex ! -path scripts/config.guess \ ! -path scripts/config.sub ! -path scripts/install-sh \ ! -path scripts/mkinstalldirs ! -path scripts/move-if-change \ ! -path INSTALL ! -path locale/programs/charmap-kw.h \ ! -path po/libc.pot ! -path sysdeps/gnu/errlist.c \ ! '(' -name configure \ -execdir test -f configure.ac -o -f configure.in ';' ')' \ ! '(' -name preconfigure \ -execdir test -f preconfigure.ac ';' ')' \ -print) and then by running 'make dist-prepare' to regenerate files built from the altered files, and then executing the following to cleanup: chmod a+x sysdeps/unix/sysv/linux/riscv/configure # Omit irrelevant whitespace and comment-only changes, # perhaps from a slightly-different Autoconf version. git checkout -f \ sysdeps/csky/configure \ sysdeps/hppa/configure \ sysdeps/riscv/configure \ sysdeps/unix/sysv/linux/csky/configure # Omit changes that caused a pre-commit check to fail like this: # remote: *** error: sysdeps/powerpc/powerpc64/ppc-mcount.S: trailing lines git checkout -f \ sysdeps/powerpc/powerpc64/ppc-mcount.S \ sysdeps/unix/sysv/linux/s390/s390-64/syscall.S # Omit change that caused a pre-commit check to fail like this: # remote: *** error: sysdeps/sparc/sparc64/multiarch/memcpy-ultra3.S: last line does not end in newline git checkout -f sysdeps/sparc/sparc64/multiarch/memcpy-ultra3.S
2019-09-07 05:40:42 +00:00
<https://www.gnu.org/licenses/>. */
#include <sysdep.h>
/* int [r3] memchr (char *s [r3], int byte [r4], int size [r5]) */
#ifndef MEMCHR
# define MEMCHR __memchr
#endif
.machine power7
PowerPC64 ENTRY_TOCLESS A number of functions in the sysdeps/powerpc/powerpc64/ tree don't use or change r2, yet declare a global entry that sets up r2. This patch fixes that problem, and consolidates the ENTRY and EALIGN macros. * sysdeps/powerpc/powerpc64/sysdep.h: Formatting. (NOPS, ENTRY_3): New macros. (ENTRY): Rewrite. (ENTRY_TOCLESS): Define. (EALIGN, EALIGN_W_0, EALIGN_W_1, EALIGN_W_2, EALIGN_W_4, EALIGN_W_5, EALIGN_W_6, EALIGN_W_7, EALIGN_W_8): Delete. * sysdeps/powerpc/powerpc64/a2/memcpy.S: Replace EALIGN with ENTRY. * sysdeps/powerpc/powerpc64/dl-trampoline.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_ceil.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_ceilf.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_floor.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_floorf.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_nearbyint.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_nearbyintf.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_rint.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_rintf.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_round.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_roundf.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_trunc.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_truncf.S: Likewise. * sysdeps/powerpc/powerpc64/memset.S: Likewise. * sysdeps/powerpc/powerpc64/power7/fpu/s_finite.S: Likewise. * sysdeps/powerpc/powerpc64/power7/fpu/s_isinf.S: Likewise. * sysdeps/powerpc/powerpc64/power7/fpu/s_isnan.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strstr.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/e_expf.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_cosf.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_sinf.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strcasestr.S: Likewise. * sysdeps/powerpc/powerpc64/addmul_1.S: Use ENTRY_TOCLESS. * sysdeps/powerpc/powerpc64/cell/memcpy.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_copysign.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_copysignl.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_fabsl.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_isnan.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_llrint.S: Likewise. * sysdeps/powerpc/powerpc64/fpu/s_llrintf.S: Likewise. * sysdeps/powerpc/powerpc64/lshift.S: Likewise. * sysdeps/powerpc/powerpc64/memcpy.S: Likewise. * sysdeps/powerpc/powerpc64/mul_1.S: Likewise. * sysdeps/powerpc/powerpc64/power4/memcmp.S: Likewise. * sysdeps/powerpc/powerpc64/power4/memcpy.S: Likewise. * sysdeps/powerpc/powerpc64/power4/memset.S: Likewise. * sysdeps/powerpc/powerpc64/power4/strncmp.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_ceil.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_ceilf.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_floor.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_floorf.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_llround.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_round.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_roundf.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_trunc.S: Likewise. * sysdeps/powerpc/powerpc64/power5+/fpu/s_truncf.S: Likewise. * sysdeps/powerpc/powerpc64/power5/fpu/s_isnan.S: Likewise. * sysdeps/powerpc/powerpc64/power6/fpu/s_copysign.S: Likewise. * sysdeps/powerpc/powerpc64/power6/fpu/s_isnan.S: Likewise. * sysdeps/powerpc/powerpc64/power6/memcpy.S: Likewise. * sysdeps/powerpc/powerpc64/power6/memset.S: Likewise. * sysdeps/powerpc/powerpc64/power6x/fpu/s_isnan.S: Likewise. * sysdeps/powerpc/powerpc64/power6x/fpu/s_llrint.S: Likewise. * sysdeps/powerpc/powerpc64/power6x/fpu/s_llround.S: Likewise. * sysdeps/powerpc/powerpc64/power7/add_n.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memcmp.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memcpy.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memmove.S: Likewise. * sysdeps/powerpc/powerpc64/power7/mempcpy.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memset.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strcasecmp.S (strcasecmp_l): Likewise. * sysdeps/powerpc/powerpc64/power7/strchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strchrnul.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strcmp.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strlen.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strncmp.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strncpy.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strnlen.S: Likewise. * sysdeps/powerpc/powerpc64/power7/strrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_finite.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_isinf.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_isnan.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_llrint.S: Likewise. * sysdeps/powerpc/powerpc64/power8/fpu/s_llround.S: Likewise. * sysdeps/powerpc/powerpc64/power8/memcmp.S: Likewise. * sysdeps/powerpc/powerpc64/power8/memset.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strchr.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strcmp.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strcpy.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strlen.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strncmp.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strncpy.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strnlen.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power8/strspn.S: Likewise. * sysdeps/powerpc/powerpc64/power9/strcmp.S: Likewise. * sysdeps/powerpc/powerpc64/power9/strncmp.S: Likewise. * sysdeps/powerpc/powerpc64/strchr.S: Likewise. * sysdeps/powerpc/powerpc64/strcmp.S: Likewise. * sysdeps/powerpc/powerpc64/strlen.S: Likewise. * sysdeps/powerpc/powerpc64/strncmp.S: Likewise. * sysdeps/powerpc/powerpc64/ppc-mcount.S: Store LR earlier. Don't add nop when SHARED. * sysdeps/powerpc/powerpc64/start.S: Fix comment. * sysdeps/powerpc/powerpc64/multiarch/strrchr-power8.S (ENTRY): Don't define. (ENTRY_TOCLESS): Define. * sysdeps/powerpc/powerpc32/sysdep.h (ENTRY_TOCLESS): Define. * sysdeps/powerpc/fpu/s_fma.S: Use ENTRY_TOCLESS. * sysdeps/powerpc/fpu/s_fmaf.S: Likewise.
2017-06-14 01:15:50 +00:00
ENTRY_TOCLESS (MEMCHR)
CALL_MCOUNT 3
dcbt 0,r3
clrrdi r8,r3,3
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
insrdi r4,r4,8,48
/* Calculate the last acceptable address and check for possible
addition overflow by using satured math:
r7 = r3 + r5
r7 |= -(r7 < x) */
add r7,r3,r5
subfc r6,r3,r7
subfe r9,r9,r9
extsw r6,r9
or r7,r7,r6
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
insrdi r4,r4,16,32
cmpldi r5,32
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
li r9, -1
rlwinm r6,r3,3,26,28 /* Calculate padding. */
insrdi r4,r4,32,0
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
addi r7,r7,-1
#ifdef __LITTLE_ENDIAN__
sld r9,r9,r6
#else
srd r9,r9,r6
#endif
ble L(small_range)
ld r12,0(r8) /* Load doubleword from memory. */
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4 /* Check for BYTEs in DWORD1. */
and r3,r3,r9
clrldi r5,r7,61 /* Byte count - 1 in last dword. */
clrrdi r7,r7,3 /* Address of last doubleword. */
cmpldi cr7,r3,0 /* Does r3 indicate we got a hit? */
bne cr7,L(done)
mtcrf 0x01,r8
/* Are we now aligned to a quadword boundary? If so, skip to
the main loop. Otherwise, go through the alignment code. */
bt 28,L(loop_setup)
/* Handle DWORD2 of pair. */
ldu r12,8(r8)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4
cmpldi cr7,r3,0
bne cr7,L(done)
L(loop_setup):
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
/* The last dword we want to read in the loop below is the one
containing the last byte of the string, ie. the dword at
(s + size - 1) & ~7, or r7. The first dword read is at
r8 + 8, we read 2 * cnt dwords, so the last dword read will
be at r8 + 8 + 16 * cnt - 8. Solving for cnt gives
cnt = (r7 - r8) / 16 */
sub r6,r7,r8
srdi r6,r6,4 /* Number of loop iterations. */
mtctr r6 /* Setup the counter. */
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
/* Main loop to look for BYTE in the string. Since
it's a small loop (8 instructions), align it to 32-bytes. */
.align 5
L(loop):
/* Load two doublewords, compare and merge in a
single register for speed. This is an attempt
to speed up the byte-checking process for bigger strings. */
ld r12,8(r8)
ldu r11,16(r8)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4
cmpb r9,r11,r4
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
or r6,r9,r3 /* Merge everything in one doubleword. */
cmpldi cr7,r6,0
bne cr7,L(found)
bdnz L(loop)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
/* We may have one more dword to read. */
cmpld r8,r7
beqlr
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
ldu r12,8(r8)
cmpb r3,r12,r4
cmpldi cr6,r3,0
bne cr6,L(done)
blr
.align 4
L(found):
/* OK, one (or both) of the doublewords contains BYTE. Check
the first doubleword and decrement the address in case the first
doubleword really contains BYTE. */
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpldi cr6,r3,0
addi r8,r8,-8
bne cr6,L(done)
/* BYTE must be in the second doubleword. Adjust the address
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
again and move the result of cmpb to r3 so we can calculate the
pointer. */
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
mr r3,r9
addi r8,r8,8
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
/* r3 has the output of the cmpb instruction, that is, it contains
0xff in the same position as BYTE in the original
doubleword from the string. Use that to calculate the pointer.
We need to make sure BYTE is *before* the end of the range. */
L(done):
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
#ifdef __LITTLE_ENDIAN__
addi r0,r3,-1
andc r0,r0,r3
popcntd r0,r0 /* Count trailing zeros. */
#else
cntlzd r0,r3 /* Count leading zeros before the match. */
#endif
cmpld r8,r7 /* Are we on the last dword? */
srdi r0,r0,3 /* Convert leading/trailing zeros to bytes. */
add r3,r8,r0
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpld cr7,r0,r5 /* If on the last dword, check byte offset. */
bnelr
blelr cr7
li r3,0
blr
.align 4
L(null):
li r3,0
blr
/* Deals with size <= 32. */
.align 4
L(small_range):
cmpldi r5,0
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
beq L(null)
ld r12,0(r8) /* Load word from memory. */
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4 /* Check for BYTE in DWORD1. */
and r3,r3,r9
cmpldi cr7,r3,0
clrldi r5,r7,61 /* Byte count - 1 in last dword. */
clrrdi r7,r7,3 /* Address of last doubleword. */
cmpld r8,r7 /* Are we done already? */
bne cr7,L(done)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
beqlr
ldu r12,8(r8)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4
cmpldi cr6,r3,0
cmpld r8,r7
bne cr6,L(done) /* Found something. */
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
beqlr /* Hit end of string (length). */
ldu r12,8(r8)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4
cmpldi cr6,r3,0
cmpld r8,r7
bne cr6,L(done)
beqlr
ldu r12,8(r8)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4
cmpldi cr6,r3,0
cmpld r8,r7
bne cr6,L(done)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
beqlr
ldu r12,8(r8)
PowerPC LE memchr and memrchr http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html Like strnlen, memchr and memrchr had a number of defects fixed by this patch as well as adding little-endian support. The first one I noticed was that the entry to the main loop needlessly checked for "are we done yet?" when we know the size is large enough that we can't be done. The second defect I noticed was that the main loop count was wrong, which in turn meant that the small loop needed to handle an extra word. Thirdly, there is nothing to say that the string can't wrap around zero, except of course that we'd normally hit a segfault on trying to read from address zero. Fixing that simplified a number of places: - /* Are we done already? */ - addi r9,r8,8 - cmpld r9,r7 - bge L(null) becomes + cmpld r8,r7 + beqlr However, the exit gets an extra test because I test for being on the last word then if so whether the byte offset is less than the end. Overall, the change is a win. Lastly, memrchr used the wrong cache hint. * sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with insrdi. Make better use of reg selection to speed exit slightly. Schedule entry path a little better. Remove useless "are we done" checks on entry to main loop. Handle wrapping around zero address. Correct main loop count. Handle single left-over word from main loop inline rather than by using loop_small. Remove extra word case in loop_small caused by wrong loop count. Add little-endian support. * sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper cache hint. * sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise. * sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian support. Avoid rlwimi. * sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
2013-08-17 09:18:36 +00:00
cmpb r3,r12,r4
cmpldi cr6,r3,0
bne cr6,L(done)
blr
END (MEMCHR)
weak_alias (__memchr, memchr)
libc_hidden_builtin_def (memchr)