1 #------------------------------------------------------------------------------ 2 # 3 # Copyright (c) 2016, Linaro Ltd. All rights reserved.<BR> 4 # 5 # This program and the accompanying materials are licensed and made available 6 # under the terms and conditions of the BSD License which accompanies this 7 # distribution. The full text of the license may be found at 8 # http://opensource.org/licenses/bsd-license.php 9 # 10 # THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS, 11 # WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED. 12 # 13 #------------------------------------------------------------------------------ 14 15 .text 16 .thumb 17 .syntax unified 18 .align 5 19 ASM_GLOBAL ASM_PFX(InternalMemSetMem16) 20 ASM_PFX(InternalMemSetMem16): 21 uxth r2, r2 22 lsl r1, r1, #1 23 orr r2, r2, r2, lsl #16 24 b 0f 25 26 ASM_GLOBAL ASM_PFX(InternalMemSetMem32) 27 ASM_PFX(InternalMemSetMem32): 28 lsl r1, r1, #2 29 b 0f 30 31 ASM_GLOBAL ASM_PFX(InternalMemSetMem64) 32 ASM_PFX(InternalMemSetMem64): 33 lsl r1, r1, #3 34 b 1f 35 36 .align 5 37 ASM_GLOBAL ASM_PFX(InternalMemSetMem) 38 ASM_PFX(InternalMemSetMem): 39 uxtb r2, r2 40 orr r2, r2, r2, lsl #8 41 orr r2, r2, r2, lsl #16 42 b 0f 43 44 ASM_GLOBAL ASM_PFX(InternalMemZeroMem) 45 ASM_PFX(InternalMemZeroMem): 46 movs r2, #0 47 0: mov r3, r2 48 49 1: push {r4, lr} 50 cmp r1, #16 // fewer than 16 bytes of input? 51 add r1, r1, r0 // r1 := dst + length 52 add lr, r0, #16 53 blt 2f 54 bic lr, lr, #15 // align output pointer 55 56 str r2, [r0] // potentially unaligned store of 4 bytes 57 str r3, [r0, #4] // potentially unaligned store of 4 bytes 58 str r2, [r0, #8] // potentially unaligned store of 4 bytes 59 str r3, [r0, #12] // potentially unaligned store of 4 bytes 60 beq 1f 61 62 0: add lr, lr, #16 // advance the output pointer by 16 bytes 63 subs r4, r1, lr // past the output? 64 blt 3f // break out of the loop 65 strd r2, r3, [lr, #-16] // aligned store of 16 bytes 66 strd r2, r3, [lr, #-8] 67 bne 0b // goto beginning of loop 68 1: pop {r4, pc} 69 70 2: subs r4, r1, lr 71 3: adds r4, r4, #16 72 subs r1, r1, #8 73 cmp r4, #4 // between 4 and 15 bytes? 74 blt 4f 75 cmp r4, #8 // between 8 and 15 bytes? 76 sub r4, lr, #16 77 str r2, [r4] // overlapping store of 4 + (4 + 4) + 4 bytes 78 it gt 79 strgt.n r3, [r4, #4] 80 it gt 81 strgt.n r2, [r1] 82 str r3, [r1, #4] 83 pop {r4, pc} 84 85 4: cmp r4, #2 // 2 or 3 bytes? 86 strb r2, [lr, #-16] // store 1 byte 87 it ge 88 strhge.n r2, [r1, #6] // store 2 bytes 89 pop {r4, pc} 90
