| Message ID | 20220705194107.3404910-1-goldstein.w.n@gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [v6] x86: Add support for building {w}memmove{_chk} with explicit ISA level | expand |
On Tue, Jul 5, 2022 at 12:41 PM Noah Goldstein <goldstein.w.n@gmail.com> wrote: > > 1. Refactor files so that all implementations are in the multiarch > directory > - Moved the implementation portion of memmove sse2 from memmove.S > to multiarch/memmove-sse2.S > > - The non-multiarch file now only includes one of the > implementations in the multiarch directory based on the compiled > ISA level (only used for non-multiarch builds. Otherwise we go > through the ifunc selector). > > 2. Add ISA level build guards to different implementations. > - I.e memmove-avx2-unaligned-erms.S which is ISA level 3 will only > build if compiled ISA level <= 3. Otherwise there is no reason > to include it as we will always use one of the ISA level 4 > implementations (memmove-evex-unaligned-erms.S). > > 3. Add new multiarch/rtld-memmove.S that just include the > non-multiarch memmove.S which will in turn select the best > implementation based on the compiled ISA level. > > 4. Refactor the ifunc selector and ifunc implementation list to use > the ISA level aware wrapper macros that allow functions below the > compiled ISA level (with a guranteed replacement) to be skipped. > > Tested with and without multiarch on x86_64 for ISA levels: > {generic, x86-64-v2, x86-64-v3, x86-64-v4} > > And m32 with and without multiarch. > isa raising memmove > --- > sysdeps/x86_64/memmove.S | 32 +- > sysdeps/x86_64/multiarch/ifunc-impl-list.c | 467 ++++++++++-------- > sysdeps/x86_64/multiarch/ifunc-memmove.h | 47 +- > .../multiarch/memmove-avx-unaligned-erms.S | 15 +- > .../multiarch/memmove-avx512-no-vzeroupper.S | 3 +- > .../multiarch/memmove-avx512-unaligned-erms.S | 11 +- > .../multiarch/memmove-evex-unaligned-erms.S | 16 +- > .../x86_64/multiarch/memmove-shlib-compat.h | 26 + > .../multiarch/memmove-sse2-unaligned-erms.S | 36 +- > sysdeps/x86_64/multiarch/memmove-ssse3.S | 4 +- > sysdeps/x86_64/multiarch/rtld-memmove.S | 18 + > 11 files changed, 403 insertions(+), 272 deletions(-) > create mode 100644 sysdeps/x86_64/multiarch/memmove-shlib-compat.h > create mode 100644 sysdeps/x86_64/multiarch/rtld-memmove.S > > diff --git a/sysdeps/x86_64/memmove.S b/sysdeps/x86_64/memmove.S > index 78e8d974d9..19527690eb 100644 > --- a/sysdeps/x86_64/memmove.S > +++ b/sysdeps/x86_64/memmove.S > @@ -16,17 +16,6 @@ > License along with the GNU C Library; if not, see > <https://www.gnu.org/licenses/>. */ > > -#include <sysdep.h> > - > -#define VEC_SIZE 16 > -#define VEC(i) xmm##i > -#define PREFETCHNT prefetchnta > -#define VMOVNT movntdq > -/* Use movups and movaps for smaller code sizes. */ > -#define VMOVU movups > -#define VMOVA movaps > -#define MOV_SIZE 3 > -#define SECTION(p) p > > #ifdef USE_MULTIARCH > # if !IS_IN (libc) > @@ -42,12 +31,18 @@ > #if !defined USE_MULTIARCH || !IS_IN (libc) > # define MEMPCPY_SYMBOL(p,s) __mempcpy > #endif > -#ifndef MEMMOVE_SYMBOL > -# define MEMMOVE_CHK_SYMBOL(p,s) p > -# define MEMMOVE_SYMBOL(p,s) memmove > -#endif > > -#include "multiarch/memmove-vec-unaligned-erms.S" > +#define MEMMOVE_CHK_SYMBOL(p,s) p > +#define MEMMOVE_SYMBOL(p,s) memmove > + > + > +#define DEFAULT_IMPL_V1 "multiarch/memmove-sse2-unaligned-erms.S" > +#define DEFAULT_IMPL_V3 "multiarch/memmove-avx-unaligned-erms.S" > +#define DEFAULT_IMPL_V4 "multiarch/memmove-evex-unaligned-erms.S" > + > +#include "isa-default-impl.h" > + > +weak_alias (__mempcpy, mempcpy) > > #ifndef USE_MULTIARCH > libc_hidden_builtin_def (memmove) > @@ -59,13 +54,10 @@ libc_hidden_def (__mempcpy) > weak_alias (__mempcpy, mempcpy) > libc_hidden_builtin_def (mempcpy) > > + > # if defined SHARED && IS_IN (libc) > # undef memcpy > # include <shlib-compat.h> > versioned_symbol (libc, __memcpy, memcpy, GLIBC_2_14); > - > -# if SHLIB_COMPAT (libc, GLIBC_2_2_5, GLIBC_2_14) > -compat_symbol (libc, memmove, memcpy, GLIBC_2_2_5); > -# endif > # endif > #endif > diff --git a/sysdeps/x86_64/multiarch/ifunc-impl-list.c b/sysdeps/x86_64/multiarch/ifunc-impl-list.c > index b84acfead2..7858aa316f 100644 > --- a/sysdeps/x86_64/multiarch/ifunc-impl-list.c > +++ b/sysdeps/x86_64/multiarch/ifunc-impl-list.c > @@ -101,84 +101,96 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, > #ifdef SHARED > /* Support sysdeps/x86_64/multiarch/memmove_chk.c. */ > IFUNC_IMPL (i, name, __memmove_chk, > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX512F), > - __memmove_chk_avx512_no_vzeroupper) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_chk_avx512_unaligned) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_chk_avx512_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX), > - __memmove_chk_avx_unaligned) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX), > - __memmove_chk_avx_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memmove_chk_avx_unaligned_rtm) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memmove_chk_avx_unaligned_erms_rtm) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_chk_evex_unaligned) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_chk_evex_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, > - CPU_FEATURE_USABLE (SSSE3), > - __memmove_chk_ssse3) > IFUNC_IMPL_ADD (array, i, __memmove_chk, 1, > - __memmove_chk_sse2_unaligned) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, 1, > - __memmove_chk_sse2_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memmove_chk, 1, > - __memmove_chk_erms)) > + __memmove_chk_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX512F), > + __memmove_chk_avx512_no_vzeroupper) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_chk_avx512_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_chk_avx512_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_chk_evex_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_chk_evex_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX), > + __memmove_chk_avx_unaligned) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (AVX), > + __memmove_chk_avx_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memmove_chk_avx_unaligned_rtm) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memmove_chk_avx_unaligned_erms_rtm) > + /* By V3 we assume fast aligned copy. */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, __memmove_chk, > + CPU_FEATURE_USABLE (SSSE3), > + __memmove_chk_ssse3) > + /* ISA V2 wrapper for SSE2 implementation because the SSE2 > + implementation is also used at ISA level 2 (SSSE3 is too > + optimized around aligned copy to be better as general > + purpose memmove). */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, __memmove_chk, 1, > + __memmove_chk_sse2_unaligned) > + X86_IFUNC_IMPL_ADD_V2 (array, i, __memmove_chk, 1, > + __memmove_chk_sse2_unaligned_erms)) > #endif > > /* Support sysdeps/x86_64/multiarch/memmove.c. */ > IFUNC_IMPL (i, name, memmove, > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX), > - __memmove_avx_unaligned) > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX), > - __memmove_avx_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memmove, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memmove_avx_unaligned_rtm) > - IFUNC_IMPL_ADD (array, i, memmove, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memmove_avx_unaligned_erms_rtm) > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_evex_unaligned) > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_evex_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX512F), > - __memmove_avx512_no_vzeroupper) > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_avx512_unaligned) > - IFUNC_IMPL_ADD (array, i, memmove, > - CPU_FEATURE_USABLE (AVX512VL), > - __memmove_avx512_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memmove, CPU_FEATURE_USABLE (SSSE3), > - __memmove_ssse3) > - IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_erms) > IFUNC_IMPL_ADD (array, i, memmove, 1, > - __memmove_sse2_unaligned) > - IFUNC_IMPL_ADD (array, i, memmove, 1, > - __memmove_sse2_unaligned_erms)) > + __memmove_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX512F), > + __memmove_avx512_no_vzeroupper) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_avx512_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_avx512_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_evex_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX512VL), > + __memmove_evex_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX), > + __memmove_avx_unaligned) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, > + CPU_FEATURE_USABLE (AVX), > + __memmove_avx_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memmove_avx_unaligned_rtm) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memmove_avx_unaligned_erms_rtm) > + /* By V3 we assume fast aligned copy. */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, memmove, > + CPU_FEATURE_USABLE (SSSE3), > + __memmove_ssse3) > + /* ISA V2 wrapper for SSE2 implementation because the SSE2 > + implementation is also used at ISA level 2 (SSSE3 is too > + optimized around aligned copy to be better as general > + purpose memmove). */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, memmove, 1, > + __memmove_sse2_unaligned) > + X86_IFUNC_IMPL_ADD_V2 (array, i, memmove, 1, > + __memmove_sse2_unaligned_erms)) > > /* Support sysdeps/x86_64/multiarch/memrchr.c. */ > IFUNC_IMPL (i, name, memrchr, > @@ -832,165 +844,190 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, > #ifdef SHARED > /* Support sysdeps/x86_64/multiarch/memcpy_chk.c. */ > IFUNC_IMPL (i, name, __memcpy_chk, > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX512F), > - __memcpy_chk_avx512_no_vzeroupper) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_chk_avx512_unaligned) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_chk_avx512_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX), > - __memcpy_chk_avx_unaligned) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX), > - __memcpy_chk_avx_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memcpy_chk_avx_unaligned_rtm) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memcpy_chk_avx_unaligned_erms_rtm) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_chk_evex_unaligned) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_chk_evex_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, > - CPU_FEATURE_USABLE (SSSE3), > - __memcpy_chk_ssse3) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, 1, > - __memcpy_chk_sse2_unaligned) > - IFUNC_IMPL_ADD (array, i, __memcpy_chk, 1, > - __memcpy_chk_sse2_unaligned_erms) > IFUNC_IMPL_ADD (array, i, __memcpy_chk, 1, > - __memcpy_chk_erms)) > + __memcpy_chk_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX512F), > + __memcpy_chk_avx512_no_vzeroupper) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_chk_avx512_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_chk_avx512_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_chk_evex_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_chk_evex_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX), > + __memcpy_chk_avx_unaligned) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (AVX), > + __memcpy_chk_avx_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memcpy_chk_avx_unaligned_rtm) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memcpy_chk_avx_unaligned_erms_rtm) > + /* By V3 we assume fast aligned copy. */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, __memcpy_chk, > + CPU_FEATURE_USABLE (SSSE3), > + __memcpy_chk_ssse3) > + /* ISA V2 wrapper for SSE2 implementation because the SSE2 > + implementation is also used at ISA level 2 (SSSE3 is too > + optimized around aligned copy to be better as general > + purpose memmove). */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, __memcpy_chk, 1, > + __memcpy_chk_sse2_unaligned) > + X86_IFUNC_IMPL_ADD_V2 (array, i, __memcpy_chk, 1, > + __memcpy_chk_sse2_unaligned_erms)) > #endif > > /* Support sysdeps/x86_64/multiarch/memcpy.c. */ > IFUNC_IMPL (i, name, memcpy, > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX), > - __memcpy_avx_unaligned) > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX), > - __memcpy_avx_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memcpy, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memcpy_avx_unaligned_rtm) > - IFUNC_IMPL_ADD (array, i, memcpy, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __memcpy_avx_unaligned_erms_rtm) > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_evex_unaligned) > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_evex_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memcpy, CPU_FEATURE_USABLE (SSSE3), > - __memcpy_ssse3) > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX512F), > - __memcpy_avx512_no_vzeroupper) > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_avx512_unaligned) > - IFUNC_IMPL_ADD (array, i, memcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __memcpy_avx512_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_sse2_unaligned) > IFUNC_IMPL_ADD (array, i, memcpy, 1, > - __memcpy_sse2_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_erms)) > + __memcpy_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX512F), > + __memcpy_avx512_no_vzeroupper) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_avx512_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_avx512_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_evex_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __memcpy_evex_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX), > + __memcpy_avx_unaligned) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, > + CPU_FEATURE_USABLE (AVX), > + __memcpy_avx_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memcpy_avx_unaligned_rtm) > + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __memcpy_avx_unaligned_erms_rtm) > + /* By V3 we assume fast aligned copy. */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, memcpy, > + CPU_FEATURE_USABLE (SSSE3), > + __memcpy_ssse3) > + /* ISA V2 wrapper for SSE2 implementation because the SSE2 > + implementation is also used at ISA level 2 (SSSE3 is too > + optimized around aligned copy to be better as general > + purpose memmove). */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, memcpy, 1, > + __memcpy_sse2_unaligned) > + X86_IFUNC_IMPL_ADD_V2 (array, i, memcpy, 1, > + __memcpy_sse2_unaligned_erms)) > > #ifdef SHARED > /* Support sysdeps/x86_64/multiarch/mempcpy_chk.c. */ > IFUNC_IMPL (i, name, __mempcpy_chk, > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX512F), > - __mempcpy_chk_avx512_no_vzeroupper) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_chk_avx512_unaligned) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_chk_avx512_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX), > - __mempcpy_chk_avx_unaligned) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX), > - __mempcpy_chk_avx_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __mempcpy_chk_avx_unaligned_rtm) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __mempcpy_chk_avx_unaligned_erms_rtm) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_chk_evex_unaligned) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_chk_evex_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, > - CPU_FEATURE_USABLE (SSSE3), > - __mempcpy_chk_ssse3) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, 1, > - __mempcpy_chk_sse2_unaligned) > IFUNC_IMPL_ADD (array, i, __mempcpy_chk, 1, > - __mempcpy_chk_sse2_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, 1, > - __mempcpy_chk_erms)) > + __mempcpy_chk_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX512F), > + __mempcpy_chk_avx512_no_vzeroupper) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_chk_avx512_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_chk_avx512_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_chk_evex_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_chk_evex_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX), > + __mempcpy_chk_avx_unaligned) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (AVX), > + __mempcpy_chk_avx_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __mempcpy_chk_avx_unaligned_rtm) > + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __mempcpy_chk_avx_unaligned_erms_rtm) > + /* By V3 we assume fast aligned copy. */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, __mempcpy_chk, > + CPU_FEATURE_USABLE (SSSE3), > + __mempcpy_chk_ssse3) > + /* ISA V2 wrapper for SSE2 implementation because the SSE2 > + implementation is also used at ISA level 2 (SSSE3 is too > + optimized around aligned copy to be better as general > + purpose memmove). */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, __mempcpy_chk, 1, > + __mempcpy_chk_sse2_unaligned) > + X86_IFUNC_IMPL_ADD_V2 (array, i, __mempcpy_chk, 1, > + __mempcpy_chk_sse2_unaligned_erms)) > #endif > > /* Support sysdeps/x86_64/multiarch/mempcpy.c. */ > IFUNC_IMPL (i, name, mempcpy, > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX512F), > - __mempcpy_avx512_no_vzeroupper) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_avx512_unaligned) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_avx512_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX), > - __mempcpy_avx_unaligned) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX), > - __mempcpy_avx_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __mempcpy_avx_unaligned_rtm) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - (CPU_FEATURE_USABLE (AVX) > - && CPU_FEATURE_USABLE (RTM)), > - __mempcpy_avx_unaligned_erms_rtm) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_evex_unaligned) > - IFUNC_IMPL_ADD (array, i, mempcpy, > - CPU_FEATURE_USABLE (AVX512VL), > - __mempcpy_evex_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, mempcpy, CPU_FEATURE_USABLE (SSSE3), > - __mempcpy_ssse3) > - IFUNC_IMPL_ADD (array, i, mempcpy, 1, > - __mempcpy_sse2_unaligned) > IFUNC_IMPL_ADD (array, i, mempcpy, 1, > - __mempcpy_sse2_unaligned_erms) > - IFUNC_IMPL_ADD (array, i, mempcpy, 1, __mempcpy_erms)) > + __mempcpy_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX512F), > + __mempcpy_avx512_no_vzeroupper) > + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_avx512_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_avx512_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_evex_unaligned) > + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX512VL), > + __mempcpy_evex_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX), > + __mempcpy_avx_unaligned) > + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, > + CPU_FEATURE_USABLE (AVX), > + __mempcpy_avx_unaligned_erms) > + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __mempcpy_avx_unaligned_rtm) > + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, > + (CPU_FEATURE_USABLE (AVX) > + && CPU_FEATURE_USABLE (RTM)), > + __mempcpy_avx_unaligned_erms_rtm) > + /* By V3 we assume fast aligned copy. */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, mempcpy, > + CPU_FEATURE_USABLE (SSSE3), > + __mempcpy_ssse3) > + /* ISA V2 wrapper for SSE2 implementation because the SSE2 > + implementation is also used at ISA level 2 (SSSE3 is too > + optimized around aligned copy to be better as general > + purpose memmove). */ > + X86_IFUNC_IMPL_ADD_V2 (array, i, mempcpy, 1, > + __mempcpy_sse2_unaligned) > + X86_IFUNC_IMPL_ADD_V2 (array, i, mempcpy, 1, > + __mempcpy_sse2_unaligned_erms)) > > /* Support sysdeps/x86_64/multiarch/strncmp.c. */ > IFUNC_IMPL (i, name, strncmp, > diff --git a/sysdeps/x86_64/multiarch/ifunc-memmove.h b/sysdeps/x86_64/multiarch/ifunc-memmove.h > index fb01fbb301..1643d32887 100644 > --- a/sysdeps/x86_64/multiarch/ifunc-memmove.h > +++ b/sysdeps/x86_64/multiarch/ifunc-memmove.h > @@ -20,11 +20,19 @@ > #include <init-arch.h> > > extern __typeof (REDIRECT_NAME) OPTIMIZE (erms) attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned) > + > +extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned) > attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned_erms) > +extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned_erms) > attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (ssse3) attribute_hidden; > +extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_no_vzeroupper) > + attribute_hidden; > + > +extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned) > + attribute_hidden; > +extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned_erms) > + attribute_hidden; > + > extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned) attribute_hidden; > extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned_erms) > attribute_hidden; > @@ -32,30 +40,27 @@ extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned_rtm) > attribute_hidden; > extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned_erms_rtm) > attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned) > - attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned_erms) > - attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned) > - attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned_erms) > + > +extern __typeof (REDIRECT_NAME) OPTIMIZE (ssse3) attribute_hidden; > + > +extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned) > attribute_hidden; > -extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_no_vzeroupper) > +extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned_erms) > attribute_hidden; > > static inline void * > IFUNC_SELECTOR (void) > { > - const struct cpu_features* cpu_features = __get_cpu_features (); > + const struct cpu_features *cpu_features = __get_cpu_features (); > > if (CPU_FEATURES_ARCH_P (cpu_features, Prefer_ERMS) > || CPU_FEATURES_ARCH_P (cpu_features, Prefer_FSRM)) > return OPTIMIZE (erms); > > - if (CPU_FEATURE_USABLE_P (cpu_features, AVX512F) > + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, AVX512F) > && !CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_AVX512)) > { > - if (CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) > + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) > { > if (CPU_FEATURE_USABLE_P (cpu_features, ERMS)) > return OPTIMIZE (avx512_unaligned_erms); > @@ -66,9 +71,10 @@ IFUNC_SELECTOR (void) > return OPTIMIZE (avx512_no_vzeroupper); > } > > - if (CPU_FEATURES_ARCH_P (cpu_features, AVX_Fast_Unaligned_Load)) > + if (X86_ISA_CPU_FEATURES_ARCH_P (cpu_features, > + AVX_Fast_Unaligned_Load, )) > { > - if (CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) > + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) > { > if (CPU_FEATURE_USABLE_P (cpu_features, ERMS)) > return OPTIMIZE (evex_unaligned_erms); > @@ -84,7 +90,8 @@ IFUNC_SELECTOR (void) > return OPTIMIZE (avx_unaligned_rtm); > } > > - if (!CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_VZEROUPPER)) > + if (X86_ISA_CPU_FEATURES_ARCH_P (cpu_features, > + Prefer_No_VZEROUPPER, !)) > { > if (CPU_FEATURE_USABLE_P (cpu_features, ERMS)) > return OPTIMIZE (avx_unaligned_erms); > @@ -93,7 +100,11 @@ IFUNC_SELECTOR (void) > } > } > > - if (CPU_FEATURE_USABLE_P (cpu_features, SSSE3) > + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, SSSE3) > + /* Leave this as runtime check. The SSSE3 is optimized almost > + exclusively for avoiding unaligned memory access during the > + copy and by and large is not better than the sse2 > + implementation as a general purpose memmove. */ > && !CPU_FEATURES_ARCH_P (cpu_features, Fast_Unaligned_Copy)) > { > return OPTIMIZE (ssse3); > diff --git a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S > index 975ae6c051..a14b155667 100644 > --- a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S > +++ b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S > @@ -1,12 +1,23 @@ > -#if IS_IN (libc) > +#include <isa-level.h> > + > +#if ISA_SHOULD_BUILD (3) > + > # define VEC_SIZE 32 > # define VEC(i) ymm##i > # define VMOVNT vmovntdq > # define VMOVU vmovdqu > # define VMOVA vmovdqa > # define MOV_SIZE 4 > + > # define SECTION(p) p##.avx > -# define MEMMOVE_SYMBOL(p,s) p##_avx_##s > + > +# ifndef MEMMOVE_SYMBOL > +# define MEMMOVE_SYMBOL(p,s) p##_avx_##s > +# endif > > # include "memmove-vec-unaligned-erms.S" > + > +# if MINIMUM_X86_ISA_LEVEL == 3 > +# include "memmove-shlib-compat.h" > +# endif > #endif > diff --git a/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S b/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S > index 42d15a142a..9c090d368b 100644 > --- a/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S > +++ b/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S > @@ -17,8 +17,9 @@ > <https://www.gnu.org/licenses/>. */ > > #include <sysdep.h> > +#include <isa-level.h> > > -#if IS_IN (libc) > +#if ISA_SHOULD_BUILD (4) > > # include "asm-syntax.h" > > diff --git a/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S > index 0fa7126830..8d1568a7ba 100644 > --- a/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S > +++ b/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S > @@ -1,4 +1,7 @@ > -#if IS_IN (libc) > +#include <isa-level.h> > + > +#if ISA_SHOULD_BUILD (4) > + > # define VEC_SIZE 64 > # define XMM0 xmm16 > # define XMM1 xmm17 > @@ -26,8 +29,12 @@ > # define VMOVA vmovdqa64 > # define VZEROUPPER > # define MOV_SIZE 6 > + > # define SECTION(p) p##.evex512 > -# define MEMMOVE_SYMBOL(p,s) p##_avx512_##s > + > +# ifndef MEMMOVE_SYMBOL > +# define MEMMOVE_SYMBOL(p,s) p##_avx512_##s > +# endif > > # include "memmove-vec-unaligned-erms.S" > #endif > diff --git a/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S > index 88715441fe..2373017358 100644 > --- a/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S > +++ b/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S > @@ -1,4 +1,7 @@ > -#if IS_IN (libc) > +#include <isa-level.h> > + > +#if ISA_SHOULD_BUILD (4) > + > # define VEC_SIZE 32 > # define XMM0 xmm16 > # define XMM1 xmm17 > @@ -26,8 +29,17 @@ > # define VMOVA vmovdqa64 > # define VZEROUPPER > # define MOV_SIZE 6 > + > # define SECTION(p) p##.evex > -# define MEMMOVE_SYMBOL(p,s) p##_evex_##s > + > +# ifndef MEMMOVE_SYMBOL > +# define MEMMOVE_SYMBOL(p,s) p##_evex_##s > +# endif > > # include "memmove-vec-unaligned-erms.S" > + > + > +# if MINIMUM_X86_ISA_LEVEL == 4 > +# include "memmove-shlib-compat.h" > +# endif > #endif > diff --git a/sysdeps/x86_64/multiarch/memmove-shlib-compat.h b/sysdeps/x86_64/multiarch/memmove-shlib-compat.h > new file mode 100644 > index 0000000000..c0793d6eef > --- /dev/null > +++ b/sysdeps/x86_64/multiarch/memmove-shlib-compat.h > @@ -0,0 +1,26 @@ > +/* Copyright (C) 2016-2022 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 > + <https://www.gnu.org/licenses/>. */ > + > +#if defined SHARED && IS_IN(libc) > +# include <shlib-compat.h> > +# if SHLIB_COMPAT(libc, GLIBC_2_2_5, GLIBC_2_14) > +/* Use __memmove_{isa_level}_unaligned to support overlapping > + addresses. */ > +compat_symbol (libc, MEMMOVE_SYMBOL (__memmove, unaligned), memcpy, > + GLIBC_2_2_5); > +# endif > +#endif > diff --git a/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S > index 09e7c1d6cd..422a079902 100644 > --- a/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S > +++ b/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S > @@ -16,18 +16,32 @@ > License along with the GNU C Library; if not, see > <https://www.gnu.org/licenses/>. */ > > -#if IS_IN (libc) > -# define MEMMOVE_SYMBOL(p,s) p##_sse2_##s > -#else > -weak_alias (__mempcpy, mempcpy) > -#endif > +#include <isa-level.h> > + > +/* MINIMUM_X86_ISA_LEVEL <= 2 because there is no V2 implementation > + so we need this to build for ISA V2 builds. */ > +#if ISA_SHOULD_BUILD (2) > + > +# include <sysdep.h> > + > +# define VEC_SIZE 16 > +# define VEC(i) xmm##i > +# define PREFETCHNT prefetchnta > +# define VMOVNT movntdq > +/* Use movups and movaps for smaller code sizes. */ > +# define VMOVU movups > +# define VMOVA movaps > +# define MOV_SIZE 3 > + > +# define SECTION(p) p > + > +# ifndef MEMMOVE_SYMBOL > +# define MEMMOVE_SYMBOL(p,s) p##_sse2_##s > +# endif > > -#include <sysdeps/x86_64/memmove.S> > +# include "multiarch/memmove-vec-unaligned-erms.S" > > -#if defined SHARED && IS_IN (libc) > -# include <shlib-compat.h> > -# if SHLIB_COMPAT (libc, GLIBC_2_2_5, GLIBC_2_14) > -/* Use __memmove_sse2_unaligned to support overlapping addresses. */ > -compat_symbol (libc, __memmove_sse2_unaligned, memcpy, GLIBC_2_2_5); > +# if MINIMUM_X86_ISA_LEVEL <= 2 > +# include "memmove-shlib-compat.h" > # endif > #endif > diff --git a/sysdeps/x86_64/multiarch/memmove-ssse3.S b/sysdeps/x86_64/multiarch/memmove-ssse3.S > index a88fde4a8f..57599752c7 100644 > --- a/sysdeps/x86_64/multiarch/memmove-ssse3.S > +++ b/sysdeps/x86_64/multiarch/memmove-ssse3.S > @@ -18,7 +18,9 @@ > <https://www.gnu.org/licenses/>. */ > > > -#if IS_IN (libc) > +#include <isa-level.h> > + > +#if ISA_SHOULD_BUILD (2) > > # include <sysdep.h> > # ifndef MEMMOVE > diff --git a/sysdeps/x86_64/multiarch/rtld-memmove.S b/sysdeps/x86_64/multiarch/rtld-memmove.S > new file mode 100644 > index 0000000000..1f3ad6433b > --- /dev/null > +++ b/sysdeps/x86_64/multiarch/rtld-memmove.S > @@ -0,0 +1,18 @@ > +/* Copyright (C) 2022 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 > + <https://www.gnu.org/licenses/>. */ > + > +#include "../memmove.S" > -- > 2.34.1 > LGTM. Thanks.
diff --git a/sysdeps/x86_64/memmove.S b/sysdeps/x86_64/memmove.S index 78e8d974d9..19527690eb 100644 --- a/sysdeps/x86_64/memmove.S +++ b/sysdeps/x86_64/memmove.S @@ -16,17 +16,6 @@ License along with the GNU C Library; if not, see <https://www.gnu.org/licenses/>. */ -#include <sysdep.h> - -#define VEC_SIZE 16 -#define VEC(i) xmm##i -#define PREFETCHNT prefetchnta -#define VMOVNT movntdq -/* Use movups and movaps for smaller code sizes. */ -#define VMOVU movups -#define VMOVA movaps -#define MOV_SIZE 3 -#define SECTION(p) p #ifdef USE_MULTIARCH # if !IS_IN (libc) @@ -42,12 +31,18 @@ #if !defined USE_MULTIARCH || !IS_IN (libc) # define MEMPCPY_SYMBOL(p,s) __mempcpy #endif -#ifndef MEMMOVE_SYMBOL -# define MEMMOVE_CHK_SYMBOL(p,s) p -# define MEMMOVE_SYMBOL(p,s) memmove -#endif -#include "multiarch/memmove-vec-unaligned-erms.S" +#define MEMMOVE_CHK_SYMBOL(p,s) p +#define MEMMOVE_SYMBOL(p,s) memmove + + +#define DEFAULT_IMPL_V1 "multiarch/memmove-sse2-unaligned-erms.S" +#define DEFAULT_IMPL_V3 "multiarch/memmove-avx-unaligned-erms.S" +#define DEFAULT_IMPL_V4 "multiarch/memmove-evex-unaligned-erms.S" + +#include "isa-default-impl.h" + +weak_alias (__mempcpy, mempcpy) #ifndef USE_MULTIARCH libc_hidden_builtin_def (memmove) @@ -59,13 +54,10 @@ libc_hidden_def (__mempcpy) weak_alias (__mempcpy, mempcpy) libc_hidden_builtin_def (mempcpy) + # if defined SHARED && IS_IN (libc) # undef memcpy # include <shlib-compat.h> versioned_symbol (libc, __memcpy, memcpy, GLIBC_2_14); - -# if SHLIB_COMPAT (libc, GLIBC_2_2_5, GLIBC_2_14) -compat_symbol (libc, memmove, memcpy, GLIBC_2_2_5); -# endif # endif #endif diff --git a/sysdeps/x86_64/multiarch/ifunc-impl-list.c b/sysdeps/x86_64/multiarch/ifunc-impl-list.c index b84acfead2..7858aa316f 100644 --- a/sysdeps/x86_64/multiarch/ifunc-impl-list.c +++ b/sysdeps/x86_64/multiarch/ifunc-impl-list.c @@ -101,84 +101,96 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, #ifdef SHARED /* Support sysdeps/x86_64/multiarch/memmove_chk.c. */ IFUNC_IMPL (i, name, __memmove_chk, - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX512F), - __memmove_chk_avx512_no_vzeroupper) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_chk_avx512_unaligned) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_chk_avx512_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX), - __memmove_chk_avx_unaligned) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX), - __memmove_chk_avx_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memmove_chk_avx_unaligned_rtm) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memmove_chk_avx_unaligned_erms_rtm) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_chk_evex_unaligned) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_chk_evex_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memmove_chk, - CPU_FEATURE_USABLE (SSSE3), - __memmove_chk_ssse3) IFUNC_IMPL_ADD (array, i, __memmove_chk, 1, - __memmove_chk_sse2_unaligned) - IFUNC_IMPL_ADD (array, i, __memmove_chk, 1, - __memmove_chk_sse2_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memmove_chk, 1, - __memmove_chk_erms)) + __memmove_chk_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX512F), + __memmove_chk_avx512_no_vzeroupper) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_chk_avx512_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_chk_avx512_unaligned_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_chk_evex_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_chk_evex_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX), + __memmove_chk_avx_unaligned) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (AVX), + __memmove_chk_avx_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memmove_chk_avx_unaligned_rtm) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memmove_chk, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memmove_chk_avx_unaligned_erms_rtm) + /* By V3 we assume fast aligned copy. */ + X86_IFUNC_IMPL_ADD_V2 (array, i, __memmove_chk, + CPU_FEATURE_USABLE (SSSE3), + __memmove_chk_ssse3) + /* ISA V2 wrapper for SSE2 implementation because the SSE2 + implementation is also used at ISA level 2 (SSSE3 is too + optimized around aligned copy to be better as general + purpose memmove). */ + X86_IFUNC_IMPL_ADD_V2 (array, i, __memmove_chk, 1, + __memmove_chk_sse2_unaligned) + X86_IFUNC_IMPL_ADD_V2 (array, i, __memmove_chk, 1, + __memmove_chk_sse2_unaligned_erms)) #endif /* Support sysdeps/x86_64/multiarch/memmove.c. */ IFUNC_IMPL (i, name, memmove, - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX), - __memmove_avx_unaligned) - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX), - __memmove_avx_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memmove, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memmove_avx_unaligned_rtm) - IFUNC_IMPL_ADD (array, i, memmove, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memmove_avx_unaligned_erms_rtm) - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_evex_unaligned) - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_evex_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX512F), - __memmove_avx512_no_vzeroupper) - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_avx512_unaligned) - IFUNC_IMPL_ADD (array, i, memmove, - CPU_FEATURE_USABLE (AVX512VL), - __memmove_avx512_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memmove, CPU_FEATURE_USABLE (SSSE3), - __memmove_ssse3) - IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_erms) IFUNC_IMPL_ADD (array, i, memmove, 1, - __memmove_sse2_unaligned) - IFUNC_IMPL_ADD (array, i, memmove, 1, - __memmove_sse2_unaligned_erms)) + __memmove_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, + CPU_FEATURE_USABLE (AVX512F), + __memmove_avx512_no_vzeroupper) + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_avx512_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_avx512_unaligned_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_evex_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, memmove, + CPU_FEATURE_USABLE (AVX512VL), + __memmove_evex_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, + CPU_FEATURE_USABLE (AVX), + __memmove_avx_unaligned) + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, + CPU_FEATURE_USABLE (AVX), + __memmove_avx_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memmove_avx_unaligned_rtm) + X86_IFUNC_IMPL_ADD_V3 (array, i, memmove, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memmove_avx_unaligned_erms_rtm) + /* By V3 we assume fast aligned copy. */ + X86_IFUNC_IMPL_ADD_V2 (array, i, memmove, + CPU_FEATURE_USABLE (SSSE3), + __memmove_ssse3) + /* ISA V2 wrapper for SSE2 implementation because the SSE2 + implementation is also used at ISA level 2 (SSSE3 is too + optimized around aligned copy to be better as general + purpose memmove). */ + X86_IFUNC_IMPL_ADD_V2 (array, i, memmove, 1, + __memmove_sse2_unaligned) + X86_IFUNC_IMPL_ADD_V2 (array, i, memmove, 1, + __memmove_sse2_unaligned_erms)) /* Support sysdeps/x86_64/multiarch/memrchr.c. */ IFUNC_IMPL (i, name, memrchr, @@ -832,165 +844,190 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array, #ifdef SHARED /* Support sysdeps/x86_64/multiarch/memcpy_chk.c. */ IFUNC_IMPL (i, name, __memcpy_chk, - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX512F), - __memcpy_chk_avx512_no_vzeroupper) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_chk_avx512_unaligned) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_chk_avx512_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX), - __memcpy_chk_avx_unaligned) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX), - __memcpy_chk_avx_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memcpy_chk_avx_unaligned_rtm) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memcpy_chk_avx_unaligned_erms_rtm) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_chk_evex_unaligned) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_chk_evex_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, - CPU_FEATURE_USABLE (SSSE3), - __memcpy_chk_ssse3) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, 1, - __memcpy_chk_sse2_unaligned) - IFUNC_IMPL_ADD (array, i, __memcpy_chk, 1, - __memcpy_chk_sse2_unaligned_erms) IFUNC_IMPL_ADD (array, i, __memcpy_chk, 1, - __memcpy_chk_erms)) + __memcpy_chk_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX512F), + __memcpy_chk_avx512_no_vzeroupper) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_chk_avx512_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_chk_avx512_unaligned_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_chk_evex_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_chk_evex_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX), + __memcpy_chk_avx_unaligned) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (AVX), + __memcpy_chk_avx_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memcpy_chk_avx_unaligned_rtm) + X86_IFUNC_IMPL_ADD_V3 (array, i, __memcpy_chk, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memcpy_chk_avx_unaligned_erms_rtm) + /* By V3 we assume fast aligned copy. */ + X86_IFUNC_IMPL_ADD_V2 (array, i, __memcpy_chk, + CPU_FEATURE_USABLE (SSSE3), + __memcpy_chk_ssse3) + /* ISA V2 wrapper for SSE2 implementation because the SSE2 + implementation is also used at ISA level 2 (SSSE3 is too + optimized around aligned copy to be better as general + purpose memmove). */ + X86_IFUNC_IMPL_ADD_V2 (array, i, __memcpy_chk, 1, + __memcpy_chk_sse2_unaligned) + X86_IFUNC_IMPL_ADD_V2 (array, i, __memcpy_chk, 1, + __memcpy_chk_sse2_unaligned_erms)) #endif /* Support sysdeps/x86_64/multiarch/memcpy.c. */ IFUNC_IMPL (i, name, memcpy, - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX), - __memcpy_avx_unaligned) - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX), - __memcpy_avx_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memcpy, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memcpy_avx_unaligned_rtm) - IFUNC_IMPL_ADD (array, i, memcpy, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __memcpy_avx_unaligned_erms_rtm) - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_evex_unaligned) - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_evex_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memcpy, CPU_FEATURE_USABLE (SSSE3), - __memcpy_ssse3) - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX512F), - __memcpy_avx512_no_vzeroupper) - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_avx512_unaligned) - IFUNC_IMPL_ADD (array, i, memcpy, - CPU_FEATURE_USABLE (AVX512VL), - __memcpy_avx512_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_sse2_unaligned) IFUNC_IMPL_ADD (array, i, memcpy, 1, - __memcpy_sse2_unaligned_erms) - IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_erms)) + __memcpy_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX512F), + __memcpy_avx512_no_vzeroupper) + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_avx512_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_avx512_unaligned_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_evex_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX512VL), + __memcpy_evex_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX), + __memcpy_avx_unaligned) + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, + CPU_FEATURE_USABLE (AVX), + __memcpy_avx_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memcpy_avx_unaligned_rtm) + X86_IFUNC_IMPL_ADD_V3 (array, i, memcpy, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __memcpy_avx_unaligned_erms_rtm) + /* By V3 we assume fast aligned copy. */ + X86_IFUNC_IMPL_ADD_V2 (array, i, memcpy, + CPU_FEATURE_USABLE (SSSE3), + __memcpy_ssse3) + /* ISA V2 wrapper for SSE2 implementation because the SSE2 + implementation is also used at ISA level 2 (SSSE3 is too + optimized around aligned copy to be better as general + purpose memmove). */ + X86_IFUNC_IMPL_ADD_V2 (array, i, memcpy, 1, + __memcpy_sse2_unaligned) + X86_IFUNC_IMPL_ADD_V2 (array, i, memcpy, 1, + __memcpy_sse2_unaligned_erms)) #ifdef SHARED /* Support sysdeps/x86_64/multiarch/mempcpy_chk.c. */ IFUNC_IMPL (i, name, __mempcpy_chk, - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX512F), - __mempcpy_chk_avx512_no_vzeroupper) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_chk_avx512_unaligned) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_chk_avx512_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX), - __mempcpy_chk_avx_unaligned) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX), - __mempcpy_chk_avx_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __mempcpy_chk_avx_unaligned_rtm) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __mempcpy_chk_avx_unaligned_erms_rtm) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_chk_evex_unaligned) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_chk_evex_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, - CPU_FEATURE_USABLE (SSSE3), - __mempcpy_chk_ssse3) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, 1, - __mempcpy_chk_sse2_unaligned) IFUNC_IMPL_ADD (array, i, __mempcpy_chk, 1, - __mempcpy_chk_sse2_unaligned_erms) - IFUNC_IMPL_ADD (array, i, __mempcpy_chk, 1, - __mempcpy_chk_erms)) + __mempcpy_chk_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX512F), + __mempcpy_chk_avx512_no_vzeroupper) + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_chk_avx512_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_chk_avx512_unaligned_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_chk_evex_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_chk_evex_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX), + __mempcpy_chk_avx_unaligned) + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (AVX), + __mempcpy_chk_avx_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __mempcpy_chk_avx_unaligned_rtm) + X86_IFUNC_IMPL_ADD_V3 (array, i, __mempcpy_chk, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __mempcpy_chk_avx_unaligned_erms_rtm) + /* By V3 we assume fast aligned copy. */ + X86_IFUNC_IMPL_ADD_V2 (array, i, __mempcpy_chk, + CPU_FEATURE_USABLE (SSSE3), + __mempcpy_chk_ssse3) + /* ISA V2 wrapper for SSE2 implementation because the SSE2 + implementation is also used at ISA level 2 (SSSE3 is too + optimized around aligned copy to be better as general + purpose memmove). */ + X86_IFUNC_IMPL_ADD_V2 (array, i, __mempcpy_chk, 1, + __mempcpy_chk_sse2_unaligned) + X86_IFUNC_IMPL_ADD_V2 (array, i, __mempcpy_chk, 1, + __mempcpy_chk_sse2_unaligned_erms)) #endif /* Support sysdeps/x86_64/multiarch/mempcpy.c. */ IFUNC_IMPL (i, name, mempcpy, - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX512F), - __mempcpy_avx512_no_vzeroupper) - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_avx512_unaligned) - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_avx512_unaligned_erms) - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX), - __mempcpy_avx_unaligned) - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX), - __mempcpy_avx_unaligned_erms) - IFUNC_IMPL_ADD (array, i, mempcpy, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __mempcpy_avx_unaligned_rtm) - IFUNC_IMPL_ADD (array, i, mempcpy, - (CPU_FEATURE_USABLE (AVX) - && CPU_FEATURE_USABLE (RTM)), - __mempcpy_avx_unaligned_erms_rtm) - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_evex_unaligned) - IFUNC_IMPL_ADD (array, i, mempcpy, - CPU_FEATURE_USABLE (AVX512VL), - __mempcpy_evex_unaligned_erms) - IFUNC_IMPL_ADD (array, i, mempcpy, CPU_FEATURE_USABLE (SSSE3), - __mempcpy_ssse3) - IFUNC_IMPL_ADD (array, i, mempcpy, 1, - __mempcpy_sse2_unaligned) IFUNC_IMPL_ADD (array, i, mempcpy, 1, - __mempcpy_sse2_unaligned_erms) - IFUNC_IMPL_ADD (array, i, mempcpy, 1, __mempcpy_erms)) + __mempcpy_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX512F), + __mempcpy_avx512_no_vzeroupper) + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_avx512_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_avx512_unaligned_erms) + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_evex_unaligned) + X86_IFUNC_IMPL_ADD_V4 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX512VL), + __mempcpy_evex_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX), + __mempcpy_avx_unaligned) + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, + CPU_FEATURE_USABLE (AVX), + __mempcpy_avx_unaligned_erms) + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __mempcpy_avx_unaligned_rtm) + X86_IFUNC_IMPL_ADD_V3 (array, i, mempcpy, + (CPU_FEATURE_USABLE (AVX) + && CPU_FEATURE_USABLE (RTM)), + __mempcpy_avx_unaligned_erms_rtm) + /* By V3 we assume fast aligned copy. */ + X86_IFUNC_IMPL_ADD_V2 (array, i, mempcpy, + CPU_FEATURE_USABLE (SSSE3), + __mempcpy_ssse3) + /* ISA V2 wrapper for SSE2 implementation because the SSE2 + implementation is also used at ISA level 2 (SSSE3 is too + optimized around aligned copy to be better as general + purpose memmove). */ + X86_IFUNC_IMPL_ADD_V2 (array, i, mempcpy, 1, + __mempcpy_sse2_unaligned) + X86_IFUNC_IMPL_ADD_V2 (array, i, mempcpy, 1, + __mempcpy_sse2_unaligned_erms)) /* Support sysdeps/x86_64/multiarch/strncmp.c. */ IFUNC_IMPL (i, name, strncmp, diff --git a/sysdeps/x86_64/multiarch/ifunc-memmove.h b/sysdeps/x86_64/multiarch/ifunc-memmove.h index fb01fbb301..1643d32887 100644 --- a/sysdeps/x86_64/multiarch/ifunc-memmove.h +++ b/sysdeps/x86_64/multiarch/ifunc-memmove.h @@ -20,11 +20,19 @@ #include <init-arch.h> extern __typeof (REDIRECT_NAME) OPTIMIZE (erms) attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned) + +extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned) attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned_erms) +extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned_erms) attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (ssse3) attribute_hidden; +extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_no_vzeroupper) + attribute_hidden; + +extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned) + attribute_hidden; +extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned_erms) + attribute_hidden; + extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned) attribute_hidden; extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned_erms) attribute_hidden; @@ -32,30 +40,27 @@ extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned_rtm) attribute_hidden; extern __typeof (REDIRECT_NAME) OPTIMIZE (avx_unaligned_erms_rtm) attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned) - attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (evex_unaligned_erms) - attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned) - attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_unaligned_erms) + +extern __typeof (REDIRECT_NAME) OPTIMIZE (ssse3) attribute_hidden; + +extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned) attribute_hidden; -extern __typeof (REDIRECT_NAME) OPTIMIZE (avx512_no_vzeroupper) +extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2_unaligned_erms) attribute_hidden; static inline void * IFUNC_SELECTOR (void) { - const struct cpu_features* cpu_features = __get_cpu_features (); + const struct cpu_features *cpu_features = __get_cpu_features (); if (CPU_FEATURES_ARCH_P (cpu_features, Prefer_ERMS) || CPU_FEATURES_ARCH_P (cpu_features, Prefer_FSRM)) return OPTIMIZE (erms); - if (CPU_FEATURE_USABLE_P (cpu_features, AVX512F) + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, AVX512F) && !CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_AVX512)) { - if (CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) { if (CPU_FEATURE_USABLE_P (cpu_features, ERMS)) return OPTIMIZE (avx512_unaligned_erms); @@ -66,9 +71,10 @@ IFUNC_SELECTOR (void) return OPTIMIZE (avx512_no_vzeroupper); } - if (CPU_FEATURES_ARCH_P (cpu_features, AVX_Fast_Unaligned_Load)) + if (X86_ISA_CPU_FEATURES_ARCH_P (cpu_features, + AVX_Fast_Unaligned_Load, )) { - if (CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, AVX512VL)) { if (CPU_FEATURE_USABLE_P (cpu_features, ERMS)) return OPTIMIZE (evex_unaligned_erms); @@ -84,7 +90,8 @@ IFUNC_SELECTOR (void) return OPTIMIZE (avx_unaligned_rtm); } - if (!CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_VZEROUPPER)) + if (X86_ISA_CPU_FEATURES_ARCH_P (cpu_features, + Prefer_No_VZEROUPPER, !)) { if (CPU_FEATURE_USABLE_P (cpu_features, ERMS)) return OPTIMIZE (avx_unaligned_erms); @@ -93,7 +100,11 @@ IFUNC_SELECTOR (void) } } - if (CPU_FEATURE_USABLE_P (cpu_features, SSSE3) + if (X86_ISA_CPU_FEATURE_USABLE_P (cpu_features, SSSE3) + /* Leave this as runtime check. The SSSE3 is optimized almost + exclusively for avoiding unaligned memory access during the + copy and by and large is not better than the sse2 + implementation as a general purpose memmove. */ && !CPU_FEATURES_ARCH_P (cpu_features, Fast_Unaligned_Copy)) { return OPTIMIZE (ssse3); diff --git a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S index 975ae6c051..a14b155667 100644 --- a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S @@ -1,12 +1,23 @@ -#if IS_IN (libc) +#include <isa-level.h> + +#if ISA_SHOULD_BUILD (3) + # define VEC_SIZE 32 # define VEC(i) ymm##i # define VMOVNT vmovntdq # define VMOVU vmovdqu # define VMOVA vmovdqa # define MOV_SIZE 4 + # define SECTION(p) p##.avx -# define MEMMOVE_SYMBOL(p,s) p##_avx_##s + +# ifndef MEMMOVE_SYMBOL +# define MEMMOVE_SYMBOL(p,s) p##_avx_##s +# endif # include "memmove-vec-unaligned-erms.S" + +# if MINIMUM_X86_ISA_LEVEL == 3 +# include "memmove-shlib-compat.h" +# endif #endif diff --git a/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S b/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S index 42d15a142a..9c090d368b 100644 --- a/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S +++ b/sysdeps/x86_64/multiarch/memmove-avx512-no-vzeroupper.S @@ -17,8 +17,9 @@ <https://www.gnu.org/licenses/>. */ #include <sysdep.h> +#include <isa-level.h> -#if IS_IN (libc) +#if ISA_SHOULD_BUILD (4) # include "asm-syntax.h" diff --git a/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S index 0fa7126830..8d1568a7ba 100644 --- a/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S @@ -1,4 +1,7 @@ -#if IS_IN (libc) +#include <isa-level.h> + +#if ISA_SHOULD_BUILD (4) + # define VEC_SIZE 64 # define XMM0 xmm16 # define XMM1 xmm17 @@ -26,8 +29,12 @@ # define VMOVA vmovdqa64 # define VZEROUPPER # define MOV_SIZE 6 + # define SECTION(p) p##.evex512 -# define MEMMOVE_SYMBOL(p,s) p##_avx512_##s + +# ifndef MEMMOVE_SYMBOL +# define MEMMOVE_SYMBOL(p,s) p##_avx512_##s +# endif # include "memmove-vec-unaligned-erms.S" #endif diff --git a/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S index 88715441fe..2373017358 100644 --- a/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S @@ -1,4 +1,7 @@ -#if IS_IN (libc) +#include <isa-level.h> + +#if ISA_SHOULD_BUILD (4) + # define VEC_SIZE 32 # define XMM0 xmm16 # define XMM1 xmm17 @@ -26,8 +29,17 @@ # define VMOVA vmovdqa64 # define VZEROUPPER # define MOV_SIZE 6 + # define SECTION(p) p##.evex -# define MEMMOVE_SYMBOL(p,s) p##_evex_##s + +# ifndef MEMMOVE_SYMBOL +# define MEMMOVE_SYMBOL(p,s) p##_evex_##s +# endif # include "memmove-vec-unaligned-erms.S" + + +# if MINIMUM_X86_ISA_LEVEL == 4 +# include "memmove-shlib-compat.h" +# endif #endif diff --git a/sysdeps/x86_64/multiarch/memmove-shlib-compat.h b/sysdeps/x86_64/multiarch/memmove-shlib-compat.h new file mode 100644 index 0000000000..c0793d6eef --- /dev/null +++ b/sysdeps/x86_64/multiarch/memmove-shlib-compat.h @@ -0,0 +1,26 @@ +/* Copyright (C) 2016-2022 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 + <https://www.gnu.org/licenses/>. */ + +#if defined SHARED && IS_IN(libc) +# include <shlib-compat.h> +# if SHLIB_COMPAT(libc, GLIBC_2_2_5, GLIBC_2_14) +/* Use __memmove_{isa_level}_unaligned to support overlapping + addresses. */ +compat_symbol (libc, MEMMOVE_SYMBOL (__memmove, unaligned), memcpy, + GLIBC_2_2_5); +# endif +#endif diff --git a/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S index 09e7c1d6cd..422a079902 100644 --- a/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S @@ -16,18 +16,32 @@ License along with the GNU C Library; if not, see <https://www.gnu.org/licenses/>. */ -#if IS_IN (libc) -# define MEMMOVE_SYMBOL(p,s) p##_sse2_##s -#else -weak_alias (__mempcpy, mempcpy) -#endif +#include <isa-level.h> + +/* MINIMUM_X86_ISA_LEVEL <= 2 because there is no V2 implementation + so we need this to build for ISA V2 builds. */ +#if ISA_SHOULD_BUILD (2) + +# include <sysdep.h> + +# define VEC_SIZE 16 +# define VEC(i) xmm##i +# define PREFETCHNT prefetchnta +# define VMOVNT movntdq +/* Use movups and movaps for smaller code sizes. */ +# define VMOVU movups +# define VMOVA movaps +# define MOV_SIZE 3 + +# define SECTION(p) p + +# ifndef MEMMOVE_SYMBOL +# define MEMMOVE_SYMBOL(p,s) p##_sse2_##s +# endif -#include <sysdeps/x86_64/memmove.S> +# include "multiarch/memmove-vec-unaligned-erms.S" -#if defined SHARED && IS_IN (libc) -# include <shlib-compat.h> -# if SHLIB_COMPAT (libc, GLIBC_2_2_5, GLIBC_2_14) -/* Use __memmove_sse2_unaligned to support overlapping addresses. */ -compat_symbol (libc, __memmove_sse2_unaligned, memcpy, GLIBC_2_2_5); +# if MINIMUM_X86_ISA_LEVEL <= 2 +# include "memmove-shlib-compat.h" # endif #endif diff --git a/sysdeps/x86_64/multiarch/memmove-ssse3.S b/sysdeps/x86_64/multiarch/memmove-ssse3.S index a88fde4a8f..57599752c7 100644 --- a/sysdeps/x86_64/multiarch/memmove-ssse3.S +++ b/sysdeps/x86_64/multiarch/memmove-ssse3.S @@ -18,7 +18,9 @@ <https://www.gnu.org/licenses/>. */ -#if IS_IN (libc) +#include <isa-level.h> + +#if ISA_SHOULD_BUILD (2) # include <sysdep.h> # ifndef MEMMOVE diff --git a/sysdeps/x86_64/multiarch/rtld-memmove.S b/sysdeps/x86_64/multiarch/rtld-memmove.S new file mode 100644 index 0000000000..1f3ad6433b --- /dev/null +++ b/sysdeps/x86_64/multiarch/rtld-memmove.S @@ -0,0 +1,18 @@ +/* Copyright (C) 2022 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 + <https://www.gnu.org/licenses/>. */ + +#include "../memmove.S"
1. Refactor files so that all implementations are in the multiarch directory - Moved the implementation portion of memmove sse2 from memmove.S to multiarch/memmove-sse2.S - The non-multiarch file now only includes one of the implementations in the multiarch directory based on the compiled ISA level (only used for non-multiarch builds. Otherwise we go through the ifunc selector). 2. Add ISA level build guards to different implementations. - I.e memmove-avx2-unaligned-erms.S which is ISA level 3 will only build if compiled ISA level <= 3. Otherwise there is no reason to include it as we will always use one of the ISA level 4 implementations (memmove-evex-unaligned-erms.S). 3. Add new multiarch/rtld-memmove.S that just include the non-multiarch memmove.S which will in turn select the best implementation based on the compiled ISA level. 4. Refactor the ifunc selector and ifunc implementation list to use the ISA level aware wrapper macros that allow functions below the compiled ISA level (with a guranteed replacement) to be skipped. Tested with and without multiarch on x86_64 for ISA levels: {generic, x86-64-v2, x86-64-v3, x86-64-v4} And m32 with and without multiarch. isa raising memmove --- sysdeps/x86_64/memmove.S | 32 +- sysdeps/x86_64/multiarch/ifunc-impl-list.c | 467 ++++++++++-------- sysdeps/x86_64/multiarch/ifunc-memmove.h | 47 +- .../multiarch/memmove-avx-unaligned-erms.S | 15 +- .../multiarch/memmove-avx512-no-vzeroupper.S | 3 +- .../multiarch/memmove-avx512-unaligned-erms.S | 11 +- .../multiarch/memmove-evex-unaligned-erms.S | 16 +- .../x86_64/multiarch/memmove-shlib-compat.h | 26 + .../multiarch/memmove-sse2-unaligned-erms.S | 36 +- sysdeps/x86_64/multiarch/memmove-ssse3.S | 4 +- sysdeps/x86_64/multiarch/rtld-memmove.S | 18 + 11 files changed, 403 insertions(+), 272 deletions(-) create mode 100644 sysdeps/x86_64/multiarch/memmove-shlib-compat.h create mode 100644 sysdeps/x86_64/multiarch/rtld-memmove.S