-#
-# ConvertAsm.py: Automatically generated from SetMem.asm
-#
-#------------------------------------------------------------------------------
-#
-# Copyright (c) 2006, Intel Corporation
-# All rights reserved. This program and the accompanying materials
-# are licensed and made available under the terms and conditions of the BSD License
-# which accompanies this distribution. The full text of the license may be found at
-# http://opensource.org/licenses/bsd-license.php
-#
-# THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
-# WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
-#
-# Module Name:
-#
-# SetMem.S
-#
-# Abstract:
-#
-# SetMem function
-#
-# Notes:
-#
-#------------------------------------------------------------------------------
-
-
-#------------------------------------------------------------------------------
-# VOID *
-# EFIAPI
-# InternalMemSetMem (
-# IN VOID *Buffer,
-# IN UINTN Count,
-# IN UINT8 Value
-# )
-#------------------------------------------------------------------------------
-.intel_syntax noprefix
-.globl ASM_PFX(InternalMemSetMem)
-ASM_PFX(InternalMemSetMem):
- push rdi
- mov rdi, rcx # rdi <- Buffer
- mov al, r8b # al <- Value
- mov r9, rdi # r9 <- Buffer as return value
- xor rcx, rcx
- sub rcx, rdi
- and rcx, 15 # rcx + rdi aligns on 16-byte boundary
- jz L0
- cmp rcx, rdx
- cmova rcx, rdx
- sub rdx, rcx
- rep stosb
-L0:
- mov rcx, rdx
- and rdx, 15
- shr rcx, 4
- jz L_SetBytes
- mov ah, al # ax <- Value repeats twice
- movdqa [rsp + 0x10], xmm0 # save xmm0
- movd xmm0, eax # xmm0[0..16] <- Value repeats twice
- pshuflw xmm0, xmm0, 0 # xmm0[0..63] <- Value repeats 8 times
- movlhps xmm0, xmm0 # xmm0 <- Value repeats 16 times
-L1:
- movntdq [rdi], xmm0 # rdi should be 16-byte aligned
- add rdi, 16
- loop L1
- mfence
- movdqa xmm0, [rsp + 0x10] # restore xmm0
-L_SetBytes:
- mov ecx, edx # high 32 bits of rcx are always zero
- rep stosb
- mov rax, r9 # rax <- Return value
- pop rdi
- ret
+#\r
+# ConvertAsm.py: Automatically generated from SetMem.asm\r
+#\r
+#------------------------------------------------------------------------------\r
+#\r
+# Copyright (c) 2006 - 2009, Intel Corporation. All rights reserved.<BR>\r
+# This program and the accompanying materials\r
+# are licensed and made available under the terms and conditions of the BSD License\r
+# which accompanies this distribution. The full text of the license may be found at\r
+# http://opensource.org/licenses/bsd-license.php\r
+#\r
+# THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,\r
+# WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.\r
+#\r
+# Module Name:\r
+#\r
+# SetMem.S\r
+#\r
+# Abstract:\r
+#\r
+# SetMem function\r
+#\r
+# Notes:\r
+#\r
+#------------------------------------------------------------------------------\r
+\r
+\r
+#------------------------------------------------------------------------------\r
+# VOID *\r
+# EFIAPI\r
+# InternalMemSetMem (\r
+# IN VOID *Buffer,\r
+# IN UINTN Count,\r
+# IN UINT8 Value\r
+# )\r
+#------------------------------------------------------------------------------\r
+ASM_GLOBAL ASM_PFX(InternalMemSetMem)\r
+ASM_PFX(InternalMemSetMem):\r
+ pushq %rdi\r
+ movq %rcx, %rdi # rdi <- Buffer\r
+ movb %r8b, %al # al <- Value\r
+ movq %rdi, %r9 # r9 <- Buffer as return value\r
+ xorq %rcx, %rcx\r
+ subq %rdi, %rcx\r
+ andq $15, %rcx # rcx + rdi aligns on 16-byte boundary\r
+ jz L0\r
+ cmpq %rdx, %rcx\r
+ cmova %rdx, %rcx\r
+ subq %rcx, %rdx\r
+ rep stosb\r
+L0:\r
+ movq %rdx, %rcx\r
+ andq $15, %rdx\r
+ shrq $4, %rcx\r
+ jz L_SetBytes\r
+ movb %al, %ah # ax <- Value repeats twice\r
+ movdqa %xmm0, 0x10(%rsp) # save xmm0\r
+ movd %eax, %xmm0 # xmm0[0..16] <- Value repeats twice\r
+ pshuflw $0, %xmm0, %xmm0 # xmm0[0..63] <- Value repeats 8 times\r
+ movlhps %xmm0, %xmm0 # xmm0 <- Value repeats 16 times\r
+L1:\r
+ movntdq %xmm0, (%rdi) # rdi should be 16-byte aligned\r
+ add $16, %rdi\r
+ loop L1\r
+ mfence\r
+ movdqa 0x10(%rsp), %xmm0 # restore xmm0\r
+L_SetBytes:\r
+ movl %edx, %ecx # high 32 bits of rcx are always zero\r
+ rep stosb\r
+ movq %r9, %rax # rax <- Return value\r
+ popq %rdi\r
+ ret\r