| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
POWER9 DD2.1 and earlier has an issue where some cache inhibited
vector load traps to the kernel, causing a performance degradation. To
handle this in memcpy and memmove, lvx/stvx is used for aligned
addresses instead of lxvd2x/stxvd2x.
Reference: https://patchwork.ozlabs.org/patch/814059/
* sysdeps/powerpc/powerpc64/power7/memcpy.S: Replace
lxvd2x/stxvd2x with lvx/stvx.
* sysdeps/powerpc/powerpc64/power7/memmove.S: Likewise.
Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.vnet.ibm.com>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
sys/cdefs.h has a macro __ptr_t, which a few places in glibc use
instead of void *. void * is a well-understood standard type for that
purpose and in a post-C89 context there is no need for a macro for it;
this patch changes those places to use void * directly instead.
Unlike __long_double_t, __ptr_t is widely used outside glibc (or at
least has many hits on codesearch.debian.net). I don't know how many
of those uses would break if sys/cdefs.h ceased to define the macro,
but there's enough risk that this patch leaves the definition and just
removes the uses within glibc; removal of the definition can be
considered separately if desired.
Tested for x86_64, and with build-many-glibcs.py.
* malloc/mcheck.c (old_free_hook): Use void * instead of __ptr_t.
(old_malloc_hook): Likewise.
(old_memalign_hook): Likewise.
(old_realloc_hook): Likewise.
(struct hdr): Likewise.
(flood): Likewise.
(freehook): Likewise.
(mallochook): Likewise.
(memalignhook): Likewise.
(reallochook): Likewise.
(mprobe): Likewise.
* malloc/mtrace.c (mallwatch): Likewise.
(tr_old_free_hook): Likewise.
(tr_old_malloc_hook): Likewise.
(tr_old_realloc_hook): Likewise.
(tr_old_memalign_hook): Likewise.
(tr_where): Likewise.
(lock_and_info): Likewise.
(tr_freehook): Likewise.
(tr_mallochook): Likewise.
(tr_reallochook): Likewise.
(tr_memalignhook): Likewise.
* misc/err.h [!__GNUC_VA_LIST] (__gnuc_va_list): Likewise.
* misc/mmap.c (__mmap): Likewise.
* misc/mmap64.c (__mmap64): Likewise.
* misc/mprotect.c (__mprotect): Likewise.
* misc/msync.c (msync): Likewise.
* misc/munmap.c (__munmap): Likewise.
* posix/posix_madvise.c (posix_madvise): Likewise.
* socket/send.c (__send): Likewise.
* socket/sendto.c (__sendto): Likewise.
* socket/setsockopt.c (__setsockopt): Likewise.
* string/memcmp.c (__ptr_t): Remove macro.
(MEMCMP): Use void * instead of ptr_t.
* string/memrchr.c (__ptr_t): Remove macro.
(__memrchr): Use void * instead of ptr_t.
* sysdeps/mach/hurd/dl-sysdep.c (__mmap): Likewise.
* sysdeps/mach/hurd/mmap.c (__mmap): Likewise.
* sysdeps/mach/hurd/mmap64.c (__mmap64): Likewise.
* sysdeps/mach/mprotect.c (__mprotect): Likewise.
* sysdeps/mach/msync.c (msync): Likewise.
* sysdeps/mach/munmap.c (__munmap): Likewise.
* sysdeps/mips/bits/setjmp.h (struct __jmp_buf_internal_tag):
Likewise.
* sysdeps/posix/getcwd.c (__getcwd): Likewise.
* sysdeps/powerpc/powerpc32/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc32/power4/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc32/power4/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc32/power6/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc32/power6/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc32/power7/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc32/power7/mempcpy.S (__mempcpy):
Likewise.
* sysdeps/powerpc/powerpc32/power7/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc64/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc64/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc64/power4/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc64/power4/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc64/power6/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc64/power6/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc64/power7/memcpy.S (memcpy): Likewise.
* sysdeps/powerpc/powerpc64/power7/mempcpy.S (__mempcpy):
Likewise.
* sysdeps/powerpc/powerpc64/power7/memset.S (memset): Likewise.
* sysdeps/powerpc/powerpc64/power8/memset.S (memset): Likewise.
* sysdeps/tile/memcmp.c (__ptr_t): Remove macro.
(MEMCMP): Use void * instead of ptr_t.
* sysdeps/unix/sysv/linux/alpha/oldglob.c (old_glob_t): Likewise.
* sysdeps/unix/sysv/linux/mmap.c (__mmap): Likewise.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Clean up the IFUNC implementations for powerpc in order to remove
unneeded macro definitions.
Tested on ppc64le with and without --disable-multi-arch flag.
* sysdeps/powerpc/powerpc64/multiarch/memcpy-a2.S: Define the
implementation-specific function name and remove unneeded
macros definition.
* sysdeps/powerpc/powerpc64/multiarch/memcpy-cell.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/memcpy-power4.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/memcpy-power6.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/memcpy-power7.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/memcpy-ppc64.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/mempcpy-power7.S: Likewise.
* sysdeps/powerpc/powerpc64/a2/memcpy.S: Set a default function
name if not defined and pass as parameter to macros accordingly.
* sysdeps/powerpc/powerpc64/cell/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power4/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power6/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/mempcpy.S: Likewise.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
This patch changes power7 memcpy to use VSX instructions only when
memory is aligned to quardword. It is to avoid unaligned kernel traps
on non-cacheable memory (for instance, memory-mapped I/O).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
http://sourceware.org/ml/libc-alpha/2013-08/msg00103.html
LIttle-endian support for memcpy. I spent some time cleaning up the
64-bit power7 memcpy, in order to avoid the extra alignment traps
power7 takes for little-endian. It probably would have been better
to copy the linux kernel version of memcpy.
* sysdeps/powerpc/powerpc32/power4/memcpy.S: Add little endian support.
* sysdeps/powerpc/powerpc32/power6/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc32/power7/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc32/power7/mempcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power4/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power6/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/memcpy.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/mempcpy.S: Likewise. Make better
use of regs. Use power7 mtocrf. Tidy function tails.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|