Home | History | Annotate | Download | only in X64 
      1 #
      2 # ConvertAsm.py: Automatically generated from SetMem.asm
      3 #
      4 #------------------------------------------------------------------------------
      5 #
      6 # Copyright (c) 2006 - 2009, Intel Corporation. All rights reserved.<BR>
      7 # This program and the accompanying materials
      8 # are licensed and made available under the terms and conditions of the BSD License
      9 # which accompanies this distribution.  The full text of the license may be found at
     10 # http://opensource.org/licenses/bsd-license.php.
     11 #
     12 # THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
     13 # WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
     14 #
     15 # Module Name:
     16 #
     17 #   SetMem.S
     18 #
     19 # Abstract:
     20 #
     21 #   SetMem function
     22 #
     23 # Notes:
     24 #
     25 #------------------------------------------------------------------------------
     26 
     27 
     28 #------------------------------------------------------------------------------
     29 #  VOID *
     30 #  EFIAPI
     31 #  InternalMemSetMem (
     32 #    IN VOID   *Buffer,
     33 #    IN UINTN  Count,
     34 #    IN UINT8  Value
     35 #    )
     36 #------------------------------------------------------------------------------
     37 ASM_GLOBAL ASM_PFX(InternalMemSetMem)
     38 ASM_PFX(InternalMemSetMem):
     39     pushq   %rdi
     40     movq    %rcx, %rdi                  # rdi <- Buffer
     41     movb    %r8b, %al                   # al <- Value
     42     movq    %rdi, %r9                   # r9 <- Buffer as return value
     43     xorq    %rcx, %rcx
     44     subq    %rdi, %rcx
     45     andq    $15, %rcx                   # rcx + rdi aligns on 16-byte boundary
     46     jz      L0
     47     cmpq    %rdx, %rcx
     48     cmova   %rdx, %rcx
     49     subq    %rcx, %rdx
     50     rep     stosb
     51 L0:
     52     movq    %rdx, %rcx
     53     andq    $15, %rdx
     54     shrq    $4, %rcx
     55     jz      L_SetBytes
     56     movb    %al, %ah                    # ax <- Value repeats twice
     57     movdqa  %xmm0, 0x10(%rsp)           # save xmm0
     58     movd    %eax, %xmm0                 # xmm0[0..16] <- Value repeats twice
     59     pshuflw $0, %xmm0, %xmm0            # xmm0[0..63] <- Value repeats 8 times
     60     movlhps %xmm0, %xmm0                # xmm0 <- Value repeats 16 times
     61 L1:
     62     movntdq %xmm0, (%rdi)               # rdi should be 16-byte aligned
     63     add     $16, %rdi
     64     loop    L1
     65     mfence
     66     movdqa  0x10(%rsp), %xmm0           # restore xmm0
     67 L_SetBytes:
     68     movl    %edx, %ecx                  # high 32 bits of rcx are always zero
     69     rep     stosb
     70     movq    %r9, %rax                   # rax <- Return value
     71     popq    %rdi
     72     ret
     73