x86-64: Fix memset() to support sizes of 4Gb and above

While currently there doesn't appear to be any reachable in-tree
case where such large memory blocks may be passed to memset()
(alloc_bootmem() being the primary non-reachable one, as it gets
called with suitably large sizes in FLATMEM configurations), we
have recently hit the problem a second time in our Xen kernels.

Rather than working around it a second time, prevent others from
falling into the same trap by fixing this long standing
limitation.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/4F05D992020000780006AA09@nat28.tlf.novell.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

arch/x86/lib/memset_64.S

@@ -19,16 +19,15 @@
 	.section .altinstr_replacement, "ax", @progbits
 .Lmemset_c:
 	movq %rdi,%r9
-	movl %edx,%r8d
-	andl $7,%r8d
-	movl %edx,%ecx
-	shrl $3,%ecx
+	movq %rdx,%rcx
+	andl $7,%edx
+	shrq $3,%rcx
 	/* expand byte value  */
 	movzbl %sil,%esi
 	movabs $0x0101010101010101,%rax
-	mulq %rsi		/* with rax, clobbers rdx */
+	imulq %rsi,%rax
 	rep stosq
-	movl %r8d,%ecx
+	movl %edx,%ecx
 	rep stosb
 	movq %r9,%rax
 	ret
@@ -50,7 +49,7 @@
 .Lmemset_c_e:
 	movq %rdi,%r9
 	movb %sil,%al
-	movl %edx,%ecx
+	movq %rdx,%rcx
 	rep stosb
 	movq %r9,%rax
 	ret
@@ -61,12 +60,11 @@ ENTRY(memset)
 ENTRY(__memset)
 	CFI_STARTPROC
 	movq %rdi,%r10
-	movq %rdx,%r11
 
 	/* expand byte value  */
 	movzbl %sil,%ecx
 	movabs $0x0101010101010101,%rax
-	mul    %rcx		/* with rax, clobbers rdx */
+	imulq  %rcx,%rax
 
 	/* align dst */
 	movl  %edi,%r9d
@@ -75,13 +73,13 @@ ENTRY(__memset)
 	CFI_REMEMBER_STATE
 .Lafter_bad_alignment:
 
-	movl %r11d,%ecx
-	shrl $6,%ecx
+	movq  %rdx,%rcx
+	shrq  $6,%rcx
 	jz	 .Lhandle_tail
 
 	.p2align 4
 .Lloop_64:
-	decl   %ecx
+	decq  %rcx
 	movq  %rax,(%rdi)
 	movq  %rax,8(%rdi)
 	movq  %rax,16(%rdi)
@@ -97,7 +95,7 @@ ENTRY(__memset)
 	   to predict jump tables. */
 	.p2align 4
 .Lhandle_tail:
-	movl	%r11d,%ecx
+	movl	%edx,%ecx
 	andl    $63&(~7),%ecx
 	jz 		.Lhandle_7
 	shrl	$3,%ecx
@@ -109,12 +107,11 @@ ENTRY(__memset)
 	jnz    .Lloop_8
 
 .Lhandle_7:
-	movl	%r11d,%ecx
-	andl	$7,%ecx
+	andl	$7,%edx
 	jz      .Lende
 	.p2align 4
 .Lloop_1:
-	decl    %ecx
+	decl    %edx
 	movb 	%al,(%rdi)
 	leaq	1(%rdi),%rdi
 	jnz     .Lloop_1
@@ -125,13 +122,13 @@ ENTRY(__memset)
 
 	CFI_RESTORE_STATE
 .Lbad_alignment:
-	cmpq $7,%r11
+	cmpq $7,%rdx
 	jbe	.Lhandle_7
 	movq %rax,(%rdi)	/* unaligned store */
 	movq $8,%r8
 	subq %r9,%r8
 	addq %r8,%rdi
-	subq %r8,%r11
+	subq %r8,%rdx
 	jmp .Lafter_bad_alignment
 .Lfinal:
 	CFI_ENDPROC