OLD | NEW |
1 /* | 1 /* |
2 * Copyright 2011 The LibYuv Project Authors. All rights reserved. | 2 * Copyright 2011 The LibYuv 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 1280 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
1291 "vst1.8 {d0}, [%1]! \n" // store 8 pixels Y. | 1291 "vst1.8 {d0}, [%1]! \n" // store 8 pixels Y. |
1292 "bgt 1b \n" | 1292 "bgt 1b \n" |
1293 : "+r"(src_argb), // %0 | 1293 : "+r"(src_argb), // %0 |
1294 "+r"(dst_y), // %1 | 1294 "+r"(dst_y), // %1 |
1295 "+r"(width) // %2 | 1295 "+r"(width) // %2 |
1296 : | 1296 : |
1297 : "cc", "memory", "q0", "q1", "q2", "q12", "q13" | 1297 : "cc", "memory", "q0", "q1", "q2", "q12", "q13" |
1298 ); | 1298 ); |
1299 } | 1299 } |
1300 | 1300 |
| 1301 void ARGBExtractAlphaRow_NEON(const uint8* src_argb, uint8* dst_a, int width) { |
| 1302 asm volatile ( |
| 1303 "1: \n" |
| 1304 MEMACCESS(0) |
| 1305 "vld4.8 {d0, d1, d2, d3}, [%0]! \n" // load row 8 pixels |
| 1306 "subs %2, %2, #8 \n" // 8 processed per loop |
| 1307 MEMACCESS(1) |
| 1308 "vst1.8 {d3}, [%1]! \n" // store 8 A's. |
| 1309 "bgt 1b \n" |
| 1310 : "+r"(src_argb), // %0 |
| 1311 "+r"(dst_a), // %1 |
| 1312 "+r"(width) // %2 |
| 1313 : |
| 1314 : "cc", "memory", "d0", "d1", "d2", "d3" // Clobber List |
| 1315 ); |
| 1316 } |
| 1317 |
1301 void ARGBToYJRow_NEON(const uint8* src_argb, uint8* dst_y, int width) { | 1318 void ARGBToYJRow_NEON(const uint8* src_argb, uint8* dst_y, int width) { |
1302 asm volatile ( | 1319 asm volatile ( |
1303 "vmov.u8 d24, #15 \n" // B * 0.11400 coefficient | 1320 "vmov.u8 d24, #15 \n" // B * 0.11400 coefficient |
1304 "vmov.u8 d25, #75 \n" // G * 0.58700 coefficient | 1321 "vmov.u8 d25, #75 \n" // G * 0.58700 coefficient |
1305 "vmov.u8 d26, #38 \n" // R * 0.29900 coefficient | 1322 "vmov.u8 d26, #38 \n" // R * 0.29900 coefficient |
1306 "1: \n" | 1323 "1: \n" |
1307 MEMACCESS(0) | 1324 MEMACCESS(0) |
1308 "vld4.8 {d0, d1, d2, d3}, [%0]! \n" // load 8 ARGB pixels. | 1325 "vld4.8 {d0, d1, d2, d3}, [%0]! \n" // load 8 ARGB pixels. |
1309 "subs %2, %2, #8 \n" // 8 processed per loop. | 1326 "subs %2, %2, #8 \n" // 8 processed per loop. |
1310 "vmull.u8 q2, d0, d24 \n" // B | 1327 "vmull.u8 q2, d0, d24 \n" // B |
(...skipping 1508 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
2819 "r"(6) // %5 | 2836 "r"(6) // %5 |
2820 : "cc", "memory", "q0", "q1" // Clobber List | 2837 : "cc", "memory", "q0", "q1" // Clobber List |
2821 ); | 2838 ); |
2822 } | 2839 } |
2823 #endif // defined(__ARM_NEON__) && !defined(__aarch64__) | 2840 #endif // defined(__ARM_NEON__) && !defined(__aarch64__) |
2824 | 2841 |
2825 #ifdef __cplusplus | 2842 #ifdef __cplusplus |
2826 } // extern "C" | 2843 } // extern "C" |
2827 } // namespace libyuv | 2844 } // namespace libyuv |
2828 #endif | 2845 #endif |
OLD | NEW |