File-copy from v4.4.100

This is the result of 'cp' from a linux-stable tree with the 'v4.4.100'
tag checked out (commit 26d6298789e695c9f627ce49a7bbd2286405798a) on
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

Please refer to that tree for all history prior to this point.

Change-Id: I8a9ee2aea93cd29c52c847d0ce33091a73ae6afe
diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S
new file mode 100644
index 0000000..2661fad
--- /dev/null
+++ b/arch/x86/lib/memset_64.S
@@ -0,0 +1,138 @@
+/* Copyright 2002 Andi Kleen, SuSE Labs */
+
+#include <linux/linkage.h>
+#include <asm/cpufeature.h>
+#include <asm/alternative-asm.h>
+
+.weak memset
+
+/*
+ * ISO C memset - set a memory block to a byte value. This function uses fast
+ * string to get better performance than the original function. The code is
+ * simpler and shorter than the orignal function as well.
+ *
+ * rdi   destination
+ * rsi   value (char)
+ * rdx   count (bytes)
+ *
+ * rax   original destination
+ */
+ENTRY(memset)
+ENTRY(__memset)
+	/*
+	 * Some CPUs support enhanced REP MOVSB/STOSB feature. It is recommended
+	 * to use it when possible. If not available, use fast string instructions.
+	 *
+	 * Otherwise, use original memset function.
+	 */
+	ALTERNATIVE_2 "jmp memset_orig", "", X86_FEATURE_REP_GOOD, \
+		      "jmp memset_erms", X86_FEATURE_ERMS
+
+	movq %rdi,%r9
+	movq %rdx,%rcx
+	andl $7,%edx
+	shrq $3,%rcx
+	/* expand byte value  */
+	movzbl %sil,%esi
+	movabs $0x0101010101010101,%rax
+	imulq %rsi,%rax
+	rep stosq
+	movl %edx,%ecx
+	rep stosb
+	movq %r9,%rax
+	ret
+ENDPROC(memset)
+ENDPROC(__memset)
+
+/*
+ * ISO C memset - set a memory block to a byte value. This function uses
+ * enhanced rep stosb to override the fast string function.
+ * The code is simpler and shorter than the fast string function as well.
+ *
+ * rdi   destination
+ * rsi   value (char)
+ * rdx   count (bytes)
+ *
+ * rax   original destination
+ */
+ENTRY(memset_erms)
+	movq %rdi,%r9
+	movb %sil,%al
+	movq %rdx,%rcx
+	rep stosb
+	movq %r9,%rax
+	ret
+ENDPROC(memset_erms)
+
+ENTRY(memset_orig)
+	movq %rdi,%r10
+
+	/* expand byte value  */
+	movzbl %sil,%ecx
+	movabs $0x0101010101010101,%rax
+	imulq  %rcx,%rax
+
+	/* align dst */
+	movl  %edi,%r9d
+	andl  $7,%r9d
+	jnz  .Lbad_alignment
+.Lafter_bad_alignment:
+
+	movq  %rdx,%rcx
+	shrq  $6,%rcx
+	jz	 .Lhandle_tail
+
+	.p2align 4
+.Lloop_64:
+	decq  %rcx
+	movq  %rax,(%rdi)
+	movq  %rax,8(%rdi)
+	movq  %rax,16(%rdi)
+	movq  %rax,24(%rdi)
+	movq  %rax,32(%rdi)
+	movq  %rax,40(%rdi)
+	movq  %rax,48(%rdi)
+	movq  %rax,56(%rdi)
+	leaq  64(%rdi),%rdi
+	jnz    .Lloop_64
+
+	/* Handle tail in loops. The loops should be faster than hard
+	   to predict jump tables. */
+	.p2align 4
+.Lhandle_tail:
+	movl	%edx,%ecx
+	andl    $63&(~7),%ecx
+	jz 		.Lhandle_7
+	shrl	$3,%ecx
+	.p2align 4
+.Lloop_8:
+	decl   %ecx
+	movq  %rax,(%rdi)
+	leaq  8(%rdi),%rdi
+	jnz    .Lloop_8
+
+.Lhandle_7:
+	andl	$7,%edx
+	jz      .Lende
+	.p2align 4
+.Lloop_1:
+	decl    %edx
+	movb 	%al,(%rdi)
+	leaq	1(%rdi),%rdi
+	jnz     .Lloop_1
+
+.Lende:
+	movq	%r10,%rax
+	ret
+
+.Lbad_alignment:
+	cmpq $7,%rdx
+	jbe	.Lhandle_7
+	movq %rax,(%rdi)	/* unaligned store */
+	movq $8,%r8
+	subq %r9,%r8
+	addq %r8,%rdi
+	subq %r8,%rdx
+	jmp .Lafter_bad_alignment
+.Lfinal:
+ENDPROC(memset_orig)