ArmPkg/Library/BaseMemoryLibVstm/Arm/CopyMem.S

   1 #------------------------------------------------------------------------------
   2 #
   3 # CopyMem() worker for ARM
   4 #
   5 # This file started out as C code that did 64 bit moves if the buffer was
   6 # 32-bit aligned, else it does a byte copy. It also does a byte copy for
   7 # any trailing bytes. Update using VSTM/SLDM to do 128 byte copies.
   8 #
   9 # Copyright (c) 2008 - 2010, Apple Inc. All rights reserved.<BR>
  10 # This program and the accompanying materials
  11 # are licensed and made available under the terms and conditions of the BSD License
  12 # which accompanies this distribution.  The full text of the license may be found at
  13 # http://opensource.org/licenses/bsd-license.php
  14 #
  15 # THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
  16 # WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
  17 #
  18 #------------------------------------------------------------------------------
  19
  20 /**
  21   Copy Length bytes from Source to Destination. Overlap is OK.
  22
  23   This implementation
  24
  25   @param  Destination Target of copy
  26   @param  Source      Place to copy from
  27   @param  Length      Number of bytes to copy
  28
  29   @return Destination
  30
  31
  32 VOID *
  33 EFIAPI
  34 InternalMemCopyMem (
  35   OUT     VOID                      *DestinationBuffer,
  36   IN      CONST VOID                *SourceBuffer,
  37   IN      UINTN                     Length
  38   )
  39 **/
  40 .text
  41 .align 2
  42 .globl ASM_PFX(InternalMemCopyMem)
  43 INTERWORK_FUNC(InternalMemCopyMem)
  44
  45 ASM_PFX(InternalMemCopyMem):
  46         stmfd   sp!, {r4, r9, lr}
  47         tst     r0, #3
  48         mov     r4, r0
  49         mov     r9, r0
  50         mov     ip, r2
  51         mov     lr, r1
  52         movne   r0, #0
  53         bne     L4
  54         tst     r1, #3
  55         movne   r3, #0
  56         moveq   r3, #1
  57         cmp     r2, #127
  58         movls   r0, #0
  59         andhi   r0, r3, #1
  60 L4:
  61         cmp     r4, r1
  62         bcc     L26
  63         bls     L7
  64         rsb     r3, r1, r4
  65         cmp     ip, r3
  66         bcc     L26
  67         cmp     ip, #0
  68         beq     L7
  69         add     r9, r4, ip
  70         add     lr, ip, r1
  71         b       L16
  72 L29:
  73         sub     ip, ip, #8
  74         cmp     ip, #7
  75         ldrd    r2, [lr, #-8]!
  76         movls   r0, #0
  77         cmp     ip, #0
  78         strd    r2, [r9, #-8]!
  79         beq     L7
  80 L16:
  81         cmp     r0, #0
  82         bne     L29
  83         sub     r3, lr, #1
  84         sub     ip, ip, #1
  85         ldrb    r3, [r3, #0]
  86         sub     r2, r9, #1
  87         cmp     ip, #0
  88         sub     r9, r9, #1
  89         sub     lr, lr, #1
  90         strb    r3, [r2, #0]
  91         bne     L16
  92         b   L7
  93 L11:
  94         ldrb    r3, [lr], #1
  95         sub     ip, ip, #1
  96         strb    r3, [r9], #1
  97 L26:
  98         cmp     ip, #0
  99         beq     L7
 100 L30:
 101         cmp     r0, #0
 102         beq     L11
 103         sub     ip, ip, #128          // 32
 104         cmp     ip, #127              // 31
 105         vldm     lr!, {d0-d15}
 106         movls   r0, #0
 107         cmp     ip, #0
 108         vstm  r9!, {d0-d15}
 109         bne     L30
 110 L7:
 111   dsb
 112   mov   r0, r4
 113         ldmfd   sp!, {r4, r9, pc}
 114
 115