movq %rbx,(%rsp)
movq %r12,1*8(%rsp)
movq %r13,2*8(%rsp)
-
+
movl $(4096/64)-5,%ecx
.p2align 4
-.Loop64:
+.Loop64:
dec %rcx
movq (%rsi), %rax
movl $5,%ecx
.p2align 4
-.Loop2:
+.Loop2:
decl %ecx
movq (%rsi), %rax
movq %r10, 40 (%rdi)
movq %r11, 48 (%rdi)
movq %r12, 56 (%rdi)
-
- leaq 64(%rdi),%rdi
- leaq 64(%rsi),%rsi
-
- jnz .Loop2
-
+
+ leaq 64(%rdi),%rdi
+ leaq 64(%rsi),%rsi
+
+ jnz .Loop2
+
movq (%rsp),%rbx
movq 1*8(%rsp),%r12
movq 2*8(%rsp),%r13
addq $3*8,%rsp
ret
-
- /* C stepping K8 run faster using the string copy instructions.
+
+ /* Some CPUs run faster using the string copy instructions.
It is also a lot simpler. Use this when possible */
-#include <asm/cpufeature.h>
-
+#include <asm/cpufeature.h>
+
.section .altinstructions,"a"
.align 8
.quad copy_page
.quad copy_page_c
- .byte X86_FEATURE_K8_C
+ .byte X86_FEATURE_REP_GOOD
.byte copy_page_c_end-copy_page_c
.byte copy_page_c_end-copy_page_c
.previous