| OLD | NEW |
|---|
| 1 ; | 1 ; |
| 2 ; Copyright (c) 2011 The WebM project authors. All Rights Reserved. | 2 ; Copyright (c) 2011 The WebM project authors. All Rights Reserved. |
| 3 ; | 3 ; |
| 4 ; Use of this source code is governed by a BSD-style license | 4 ; Use of this source code is governed by a BSD-style license |
| 5 ; that can be found in the LICENSE file in the root of the source | 5 ; that can be found in the LICENSE file in the root of the source |
| 6 ; tree. An additional intellectual property rights grant can be found | 6 ; tree. An additional intellectual property rights grant can be found |
| 7 ; in the file PATENTS. All contributing project authors may | 7 ; in the file PATENTS. All contributing project authors may |
| 8 ; be found in the AUTHORS file in the root of the source tree. | 8 ; be found in the AUTHORS file in the root of the source tree. |
| 9 ; | 9 ; |
| 10 | 10 |
| (...skipping 41 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 52 usub8 r6, r5, r4 ; calculate difference with reversed operands | 52 usub8 r6, r5, r4 ; calculate difference with reversed operands |
| 53 pld [r2, r3, lsl #1] | 53 pld [r2, r3, lsl #1] |
| 54 sel r6, r6, lr ; select bytes with negative difference | 54 sel r6, r6, lr ; select bytes with negative difference |
| 55 | 55 |
| 56 ; calculate partial sums | 56 ; calculate partial sums |
| 57 usad8 r4, r7, lr ; calculate sum of positive differences | 57 usad8 r4, r7, lr ; calculate sum of positive differences |
| 58 usad8 r5, r6, lr ; calculate sum of negative differences | 58 usad8 r5, r6, lr ; calculate sum of negative differences |
| 59 orr r6, r6, r7 ; differences of all 4 pixels | 59 orr r6, r6, r7 ; differences of all 4 pixels |
| 60 ; calculate total sum | 60 ; calculate total sum |
| 61 adds r8, r8, r4 ; add positive differences to sum | 61 adds r8, r8, r4 ; add positive differences to sum |
| 62 subs r8, r8, r5 ; substract negative differences from sum | 62 subs r8, r8, r5 ; subtract negative differences from sum |
| 63 | 63 |
| 64 ; calculate sse | 64 ; calculate sse |
| 65 uxtb16 r5, r6 ; byte (two pixels) to halfwords | 65 uxtb16 r5, r6 ; byte (two pixels) to halfwords |
| 66 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords | 66 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords |
| 67 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) | 67 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) |
| 68 | 68 |
| 69 ; 2nd 4 pixels | 69 ; 2nd 4 pixels |
| 70 ldr r4, [r0, #4] ; load 4 src pixels | 70 ldr r4, [r0, #4] ; load 4 src pixels |
| 71 ldr r6, [r9, #4] ; load 4 src pixels from next row | 71 ldr r6, [r9, #4] ; load 4 src pixels from next row |
| 72 ldr r5, [r2, #4] ; load 4 ref pixels | 72 ldr r5, [r2, #4] ; load 4 ref pixels |
| (...skipping 10 matching lines...) Expand all Loading... |
| 83 usub8 r6, r5, r4 ; calculate difference with reversed operands | 83 usub8 r6, r5, r4 ; calculate difference with reversed operands |
| 84 sel r6, r6, lr ; select bytes with negative difference | 84 sel r6, r6, lr ; select bytes with negative difference |
| 85 | 85 |
| 86 ; calculate partial sums | 86 ; calculate partial sums |
| 87 usad8 r4, r7, lr ; calculate sum of positive differences | 87 usad8 r4, r7, lr ; calculate sum of positive differences |
| 88 usad8 r5, r6, lr ; calculate sum of negative differences | 88 usad8 r5, r6, lr ; calculate sum of negative differences |
| 89 orr r6, r6, r7 ; differences of all 4 pixels | 89 orr r6, r6, r7 ; differences of all 4 pixels |
| 90 | 90 |
| 91 ; calculate total sum | 91 ; calculate total sum |
| 92 add r8, r8, r4 ; add positive differences to sum | 92 add r8, r8, r4 ; add positive differences to sum |
| 93 sub r8, r8, r5 ; substract negative differences from sum | 93 sub r8, r8, r5 ; subtract negative differences from sum |
| 94 | 94 |
| 95 ; calculate sse | 95 ; calculate sse |
| 96 uxtb16 r5, r6 ; byte (two pixels) to halfwords | 96 uxtb16 r5, r6 ; byte (two pixels) to halfwords |
| 97 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords | 97 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords |
| 98 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) | 98 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) |
| 99 | 99 |
| 100 ; 3rd 4 pixels | 100 ; 3rd 4 pixels |
| 101 ldr r4, [r0, #8] ; load 4 src pixels | 101 ldr r4, [r0, #8] ; load 4 src pixels |
| 102 ldr r6, [r9, #8] ; load 4 src pixels from next row | 102 ldr r6, [r9, #8] ; load 4 src pixels from next row |
| 103 ldr r5, [r2, #8] ; load 4 ref pixels | 103 ldr r5, [r2, #8] ; load 4 ref pixels |
| (...skipping 10 matching lines...) Expand all Loading... |
| 114 usub8 r6, r5, r4 ; calculate difference with reversed operands | 114 usub8 r6, r5, r4 ; calculate difference with reversed operands |
| 115 sel r6, r6, lr ; select bytes with negative difference | 115 sel r6, r6, lr ; select bytes with negative difference |
| 116 | 116 |
| 117 ; calculate partial sums | 117 ; calculate partial sums |
| 118 usad8 r4, r7, lr ; calculate sum of positive differences | 118 usad8 r4, r7, lr ; calculate sum of positive differences |
| 119 usad8 r5, r6, lr ; calculate sum of negative differences | 119 usad8 r5, r6, lr ; calculate sum of negative differences |
| 120 orr r6, r6, r7 ; differences of all 4 pixels | 120 orr r6, r6, r7 ; differences of all 4 pixels |
| 121 | 121 |
| 122 ; calculate total sum | 122 ; calculate total sum |
| 123 add r8, r8, r4 ; add positive differences to sum | 123 add r8, r8, r4 ; add positive differences to sum |
| 124 sub r8, r8, r5 ; substract negative differences from sum | 124 sub r8, r8, r5 ; subtract negative differences from sum |
| 125 | 125 |
| 126 ; calculate sse | 126 ; calculate sse |
| 127 uxtb16 r5, r6 ; byte (two pixels) to halfwords | 127 uxtb16 r5, r6 ; byte (two pixels) to halfwords |
| 128 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords | 128 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords |
| 129 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) | 129 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) |
| 130 | 130 |
| 131 ; 4th 4 pixels | 131 ; 4th 4 pixels |
| 132 ldr r4, [r0, #12] ; load 4 src pixels | 132 ldr r4, [r0, #12] ; load 4 src pixels |
| 133 ldr r6, [r9, #12] ; load 4 src pixels from next row | 133 ldr r6, [r9, #12] ; load 4 src pixels from next row |
| 134 ldr r5, [r2, #12] ; load 4 ref pixels | 134 ldr r5, [r2, #12] ; load 4 ref pixels |
| (...skipping 12 matching lines...) Expand all Loading... |
| 147 add r2, r2, r3 ; set dst_ptr to next row | 147 add r2, r2, r3 ; set dst_ptr to next row |
| 148 sel r6, r6, lr ; select bytes with negative difference | 148 sel r6, r6, lr ; select bytes with negative difference |
| 149 | 149 |
| 150 ; calculate partial sums | 150 ; calculate partial sums |
| 151 usad8 r4, r7, lr ; calculate sum of positive differences | 151 usad8 r4, r7, lr ; calculate sum of positive differences |
| 152 usad8 r5, r6, lr ; calculate sum of negative differences | 152 usad8 r5, r6, lr ; calculate sum of negative differences |
| 153 orr r6, r6, r7 ; differences of all 4 pixels | 153 orr r6, r6, r7 ; differences of all 4 pixels |
| 154 | 154 |
| 155 ; calculate total sum | 155 ; calculate total sum |
| 156 add r8, r8, r4 ; add positive differences to sum | 156 add r8, r8, r4 ; add positive differences to sum |
| 157 sub r8, r8, r5 ; substract negative differences from sum | 157 sub r8, r8, r5 ; subtract negative differences from sum |
| 158 | 158 |
| 159 ; calculate sse | 159 ; calculate sse |
| 160 uxtb16 r5, r6 ; byte (two pixels) to halfwords | 160 uxtb16 r5, r6 ; byte (two pixels) to halfwords |
| 161 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords | 161 uxtb16 r7, r6, ror #8 ; another two pixels to halfwords |
| 162 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) | 162 smlad r11, r5, r5, r11 ; dual signed multiply, add and accumulate (1) |
| 163 smlad r11, r7, r7, r11 ; dual signed multiply, add and accumulate (2) | 163 smlad r11, r7, r7, r11 ; dual signed multiply, add and accumulate (2) |
| 164 | 164 |
| 165 | 165 |
| 166 subs r12, r12, #1 | 166 subs r12, r12, #1 |
| 167 | 167 |
| 168 bne loop | 168 bne loop |
| 169 | 169 |
| 170 ; return stuff | 170 ; return stuff |
| 171 ldr r6, [sp, #40] ; get address of sse | 171 ldr r6, [sp, #40] ; get address of sse |
| 172 mul r0, r8, r8 ; sum * sum | 172 mul r0, r8, r8 ; sum * sum |
| 173 str r11, [r6] ; store sse | 173 str r11, [r6] ; store sse |
| 174 sub r0, r11, r0, lsr #8 ; return (sse - ((sum * sum) >> 8)) | 174 sub r0, r11, r0, lsr #8 ; return (sse - ((sum * sum) >> 8)) |
| 175 | 175 |
| 176 ldmfd sp!, {r4-r12, pc} | 176 ldmfd sp!, {r4-r12, pc} |
| 177 | 177 |
| 178 ENDP | 178 ENDP |
| 179 | 179 |
| 180 c80808080 | 180 c80808080 |
| 181 DCD 0x80808080 | 181 DCD 0x80808080 |
| 182 | 182 |
| 183 END | 183 END |
| 184 | 184 |
| OLD | NEW |
|---|
Chromium Code Reviews
Issue 181493009:
libvpx: Pull from upstream (Closed)
Base URL: svn://svn.chromium.org/chrome/trunk/deps/third_party/libvpx/