X86-64: Remove the previous SSE2/AVX2 memsets

Since the new SSE2/AVX2 memsets are faster than the previous ones, we
can remove the previous SSE2/AVX2 memsets and replace them with the
new ones.  This reduces the size of libc.so by about 900 bytes.

No change in IFUNC selection if SSE2 and AVX2 memsets weren't used
before.  If SSE2 or AVX2 memset was used, the new SSE2 or AVX2 memset
optimized with Enhanced REP STOSB will be used for processors with
ERMS.  The new AVX512 memset will be used for processors with AVX512
which prefer vzeroupper.

	[BZ #19881]
	* sysdeps/x86_64/multiarch/memset-sse2-unaligned-erms.S: Folded
	into ...
	* sysdeps/x86_64/memset.S: This.
	(__bzero): Removed.
	(__memset_tail): Likewise.
	(__memset_chk): Likewise.
	(memset): Likewise.
	(MEMSET_CHK_SYMBOL): New. Define only if MEMSET_SYMBOL isn't
	defined.
	(MEMSET_SYMBOL): Define only if MEMSET_SYMBOL isn't defined.
	* sysdeps/x86_64/multiarch/memset-avx2.S: Removed.
	(__memset_zero_constant_len_parameter): Check SHARED instead of
	PIC.
	* sysdeps/x86_64/multiarch/Makefile (sysdep_routines): Remove
	memset-avx2 and memset-sse2-unaligned-erms.
	* sysdeps/x86_64/multiarch/ifunc-impl-list.c
	(__libc_ifunc_impl_list): Remove __memset_chk_sse2,
	__memset_chk_avx2, __memset_sse2 and __memset_avx2_unaligned.
	* sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
	(__bzero): Enabled.
	* sysdeps/x86_64/multiarch/memset.S (memset): Replace
	__memset_sse2 and __memset_avx2 with __memset_sse2_unaligned
	and __memset_avx2_unaligned.  Use __memset_sse2_unaligned_erms
	or __memset_avx2_unaligned_erms if processor has ERMS.  Support
	__memset_avx512_unaligned_erms and __memset_avx512_unaligned.
	(memset): Removed.
	(__memset_chk): Likewise.
	(MEMSET_SYMBOL): New.
	(libc_hidden_builtin_def): Replace __memset_sse2 with
	__memset_sse2_unaligned.
	* sysdeps/x86_64/multiarch/memset_chk.S (__memset_chk): Replace
	__memset_chk_sse2 and __memset_chk_avx2 with
	__memset_chk_sse2_unaligned and __memset_chk_avx2_unaligned_erms.
	Use __memset_chk_sse2_unaligned_erms or
	__memset_chk_avx2_unaligned_erms if processor has ERMS.  Support
	__memset_chk_avx512_unaligned_erms and
	__memset_chk_avx512_unaligned.

1 files changed, 21 insertions, 102 deletions

diff --git a/sysdeps/x86_64/memset.S b/sysdeps/x86_64/memset.S
index 4cf0da0fb8..62b85c39b8 100644
--- a/sysdeps/x86_64/memset.S
+++ b/sysdeps/x86_64/memset.S
@@ -19,113 +19,32 @@
 
 #include <sysdep.h>
 
-	.text
-#if IS_IN (libc)
-ENTRY(__bzero)
-	movq	%rdi, %rax /* Set return value.  */
-	movq	%rsi, %rdx /* Set n.  */
-	pxor	%xmm0, %xmm0
-	jmp	L(entry_from_bzero)
-END(__bzero)
-weak_alias (__bzero, bzero)
-
-/* Like memset but takes additional parameter with return value.  */
-ENTRY(__memset_tail)
-	movq	%rcx, %rax /* Set return value.  */
-
-	movd	%esi, %xmm0
-	punpcklbw	%xmm0, %xmm0
-	punpcklwd	%xmm0, %xmm0
-	pshufd	$0, %xmm0, %xmm0
-
-	jmp	L(entry_from_bzero)
-END(__memset_tail)
-#endif
-
-#if defined PIC && IS_IN (libc)
-ENTRY_CHK (__memset_chk)
-	cmpq	%rdx, %rcx
-	jb	HIDDEN_JUMPTARGET (__chk_fail)
-END_CHK (__memset_chk)
+#define VEC_SIZE	16
+#define VEC(i)		xmm##i
+/* Don't use movups and movaps since it will get larger nop paddings for
+   alignment.  */
+#define VMOVU		movdqu
+#define VMOVA		movdqa
+
+#define VDUP_TO_VEC0_AND_SET_RETURN(d, r) \
+  movd d, %xmm0; \
+  movq r, %rax; \
+  punpcklbw %xmm0, %xmm0; \
+  punpcklwd %xmm0, %xmm0; \
+  pshufd $0, %xmm0, %xmm0
+
+#define SECTION(p)		p
+
+#ifndef MEMSET_SYMBOL
+# define MEMSET_CHK_SYMBOL(p,s)	p
+# define MEMSET_SYMBOL(p,s)	memset
 #endif
 
-ENTRY (memset)
-	movd	%esi, %xmm0
-	movq	%rdi, %rax
-	punpcklbw	%xmm0, %xmm0
-	punpcklwd	%xmm0, %xmm0
-	pshufd	$0, %xmm0, %xmm0
-L(entry_from_bzero):
-	cmpq	$64, %rdx
-	ja	L(loop_start)
-	cmpq	$16, %rdx
-	jbe	L(less_16_bytes)
-	cmpq	$32, %rdx
-	movdqu	%xmm0, (%rdi)
-	movdqu	%xmm0, -16(%rdi,%rdx)
-	ja	L(between_32_64_bytes)
-L(return):
-	rep
-	ret
-	.p2align 4
-L(between_32_64_bytes):
-	movdqu	%xmm0, 16(%rdi)
-	movdqu	%xmm0, -32(%rdi,%rdx)
-	ret
-	.p2align 4
-L(loop_start):
-	leaq	64(%rdi), %rcx
-	movdqu	%xmm0, (%rdi)
-	andq	$-64, %rcx
-	movdqu	%xmm0, -16(%rdi,%rdx)
-	movdqu	%xmm0, 16(%rdi)
-	movdqu	%xmm0, -32(%rdi,%rdx)
-	movdqu	%xmm0, 32(%rdi)
-	movdqu	%xmm0, -48(%rdi,%rdx)
-	movdqu	%xmm0, 48(%rdi)
-	movdqu	%xmm0, -64(%rdi,%rdx)
-	addq	%rdi, %rdx
-	andq	$-64, %rdx
-	cmpq	%rdx, %rcx
-	je	L(return)
-	.p2align 4
-L(loop):
-	movdqa	%xmm0, (%rcx)
-	movdqa	%xmm0, 16(%rcx)
-	movdqa	%xmm0, 32(%rcx)
-	movdqa	%xmm0, 48(%rcx)
-	addq	$64, %rcx
-	cmpq	%rcx, %rdx
-	jne	L(loop)
-	rep
-	ret
-L(less_16_bytes):
-	movq %xmm0, %rcx
-	testb	$24, %dl
-	jne	L(between8_16bytes)
-	testb	$4, %dl
-	jne	L(between4_7bytes)
-	testb	$1, %dl
-	je	L(odd_byte)
-	movb	%cl, (%rdi)
-L(odd_byte):
-	testb	$2, %dl
-	je	L(return)
-	movw	%cx, -2(%rax,%rdx)
-	ret
-L(between4_7bytes):
-	movl	%ecx, (%rdi)
-	movl	%ecx, -4(%rdi,%rdx)
-	ret
-L(between8_16bytes):
-	movq	%rcx, (%rdi)
-	movq	%rcx, -8(%rdi,%rdx)
-	ret
+#include "multiarch/memset-vec-unaligned-erms.S"
 
-END (memset)
 libc_hidden_builtin_def (memset)
 
-#if defined PIC && IS_IN (libc) && !defined USE_MULTIARCH
+#if defined SHARED && IS_IN (libc) && !defined USE_MULTIARCH
 strong_alias (__memset_chk, __memset_zero_constant_len_parameter)
 	.section .gnu.warning.__memset_zero_constant_len_parameter
 	.string "memset used with constant zero length parameter; this could be due to transposed parameters"