summaryrefslogtreecommitdiffstats
path: root/arch/x86/lib/memcpy_64.S
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/lib/memcpy_64.S')
-rw-r--r--arch/x86/lib/memcpy_64.S44
1 files changed, 20 insertions, 24 deletions
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
index efbf2a0ecde..1c273be7c97 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
@@ -27,9 +27,8 @@
.section .altinstr_replacement, "ax", @progbits
.Lmemcpy_c:
movq %rdi, %rax
-
- movl %edx, %ecx
- shrl $3, %ecx
+ movq %rdx, %rcx
+ shrq $3, %rcx
andl $7, %edx
rep movsq
movl %edx, %ecx
@@ -48,8 +47,7 @@
.section .altinstr_replacement, "ax", @progbits
.Lmemcpy_c_e:
movq %rdi, %rax
-
- movl %edx, %ecx
+ movq %rdx, %rcx
rep movsb
ret
.Lmemcpy_e_e:
@@ -60,10 +58,7 @@ ENTRY(memcpy)
CFI_STARTPROC
movq %rdi, %rax
- /*
- * Use 32bit CMP here to avoid long NOP padding.
- */
- cmp $0x20, %edx
+ cmpq $0x20, %rdx
jb .Lhandle_tail
/*
@@ -72,7 +67,7 @@ ENTRY(memcpy)
*/
cmp %dil, %sil
jl .Lcopy_backward
- subl $0x20, %edx
+ subq $0x20, %rdx
.Lcopy_forward_loop:
subq $0x20, %rdx
@@ -91,7 +86,7 @@ ENTRY(memcpy)
movq %r11, 3*8(%rdi)
leaq 4*8(%rdi), %rdi
jae .Lcopy_forward_loop
- addq $0x20, %rdx
+ addl $0x20, %edx
jmp .Lhandle_tail
.Lcopy_backward:
@@ -123,11 +118,11 @@ ENTRY(memcpy)
/*
* Calculate copy position to head.
*/
- addq $0x20, %rdx
+ addl $0x20, %edx
subq %rdx, %rsi
subq %rdx, %rdi
.Lhandle_tail:
- cmpq $16, %rdx
+ cmpl $16, %edx
jb .Lless_16bytes
/*
@@ -144,7 +139,7 @@ ENTRY(memcpy)
retq
.p2align 4
.Lless_16bytes:
- cmpq $8, %rdx
+ cmpl $8, %edx
jb .Lless_8bytes
/*
* Move data from 8 bytes to 15 bytes.
@@ -156,7 +151,7 @@ ENTRY(memcpy)
retq
.p2align 4
.Lless_8bytes:
- cmpq $4, %rdx
+ cmpl $4, %edx
jb .Lless_3bytes
/*
@@ -169,18 +164,19 @@ ENTRY(memcpy)
retq
.p2align 4
.Lless_3bytes:
- cmpl $0, %edx
- je .Lend
+ subl $1, %edx
+ jb .Lend
/*
* Move data from 1 bytes to 3 bytes.
*/
-.Lloop_1:
- movb (%rsi), %r8b
- movb %r8b, (%rdi)
- incq %rdi
- incq %rsi
- decl %edx
- jnz .Lloop_1
+ movzbl (%rsi), %ecx
+ jz .Lstore_1byte
+ movzbq 1(%rsi), %r8
+ movzbq (%rsi, %rdx), %r9
+ movb %r8b, 1(%rdi)
+ movb %r9b, (%rdi, %rdx)
+.Lstore_1byte:
+ movb %cl, (%rdi)
.Lend:
retq