]> git.proxmox.com Git - mirror_edk2.git/blob - EdkCompatibilityPkg/Foundation/Library/CompilerStub/X64/memcpySSE2.asm
Update the copyright notice format
[mirror_edk2.git] / EdkCompatibilityPkg / Foundation / Library / CompilerStub / X64 / memcpySSE2.asm
1 ;------------------------------------------------------------------------------
2 ;
3 ; Copyright (c) 2007, Intel Corporation. All rights reserved.<BR>
4 ; This program and the accompanying materials
5 ; are licensed and made available under the terms and conditions of the BSD License
6 ; which accompanies this distribution. The full text of the license may be found at
7 ; http://opensource.org/licenses/bsd-license.php
8 ;
9 ; THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
10 ; WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
11 ;
12 ; Module Name:
13 ;
14 ; CopyMem.asm
15 ;
16 ; Abstract:
17 ;
18 ; memcpy function
19 ;
20 ; Notes:
21 ;
22 ;------------------------------------------------------------------------------
23
24 .code
25
26 ;------------------------------------------------------------------------------
27 ; VOID *
28 ; memcpy (
29 ; OUT VOID *DestinationBuffer,
30 ; IN CONST VOID *SourceBuffer,
31 ; IN UINTN Length
32 ; );
33 ;------------------------------------------------------------------------------
34 memcpy PROC USES rsi rdi
35 mov rax, rcx ; rax <- Destination as return value
36 cmp rdx, rcx ; if Source == Destination, do nothing
37 je @CopyMemDone
38 cmp r8, 0 ; if Count == 0, do nothing
39 je @CopyMemDone
40 mov rsi, rdx ; rsi <- Source
41 mov rdi, rcx ; rdi <- Destination
42 lea r9, [rsi + r8 - 1] ; r9 <- Last byte of Source
43 cmp rsi, rdi
44 jae @F ; Copy forward if Source > Destination
45 cmp r9, rdi ; Overlapped?
46 jae @CopyBackward ; Copy backward if overlapped
47 @@:
48 xor rcx, rcx
49 sub rcx, rdi ; rcx <- -rdi
50 and rcx, 15 ; rcx + rsi should be 16 bytes aligned
51 jz @F ; skip if rcx == 0
52 cmp rcx, r8
53 cmova rcx, r8
54 sub r8, rcx
55 rep movsb
56 @@:
57 mov rcx, r8
58 and r8, 15
59 shr rcx, 4 ; rcx <- # of DQwords to copy
60 jz @CopyBytes
61 @@:
62 movdqu xmm0, [rsi] ; rsi may not be 16-byte aligned
63 movdqa [rdi], xmm0 ; rdi should be 16-byte aligned
64 add rsi, 16
65 add rdi, 16
66 loop @B
67 jmp @CopyBytes ; copy remaining bytes
68 @CopyBackward:
69 mov rsi, r9 ; rsi <- Last byte of Source
70 lea rdi, [rdi + r8 - 1] ; rdi <- Last byte of Destination
71 std
72 @CopyBytes:
73 mov rcx, r8
74 rep movsb
75 cld
76 @CopyMemDone:
77 ret
78 memcpy ENDP
79
80 END