x86: Add support for building {w}memmove{_chk} with explicit ISA level

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

11 files changed, 403 insertions, 272 deletions

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,