| OLD | NEW |
| 1 /* | 1 /* |
| 2 * Copyright 2014 The LibYuv Project Authors. All rights reserved. | 2 * Copyright 2014 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 2693 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... |
| 2704 : "+r"(src_y0), // %0 | 2704 : "+r"(src_y0), // %0 |
| 2705 "+r"(src_y1), // %1 | 2705 "+r"(src_y1), // %1 |
| 2706 "+r"(dst_sobely), // %2 | 2706 "+r"(dst_sobely), // %2 |
| 2707 "+r"(width) // %3 | 2707 "+r"(width) // %3 |
| 2708 : "r"(1LL), // %4 | 2708 : "r"(1LL), // %4 |
| 2709 "r"(6LL) // %5 | 2709 "r"(6LL) // %5 |
| 2710 : "cc", "memory", "v0", "v1", "v2", "v3" // Clobber List | 2710 : "cc", "memory", "v0", "v1", "v2", "v3" // Clobber List |
| 2711 ); | 2711 ); |
| 2712 } | 2712 } |
| 2713 | 2713 |
| 2714 // Caveat - rounds float to half float whereas scaling version truncates. |
| 2714 void HalfFloat1Row_NEON(const uint16* src, uint16* dst, float, int width) { | 2715 void HalfFloat1Row_NEON(const uint16* src, uint16* dst, float, int width) { |
| 2715 asm volatile ( | 2716 asm volatile ( |
| 2716 "1: \n" | 2717 "1: \n" |
| 2717 MEMACCESS(0) | 2718 MEMACCESS(0) |
| 2718 "ld1 {v1.16b}, [%0], #16 \n" // load 8 shorts | 2719 "ld1 {v1.16b}, [%0], #16 \n" // load 8 shorts |
| 2719 "subs %w2, %w2, #8 \n" // 8 pixels per loop | 2720 "subs %w2, %w2, #8 \n" // 8 pixels per loop |
| 2720 "uxtl v2.4s, v1.4h \n" // 8 int's | 2721 "uxtl v2.4s, v1.4h \n" // 8 int's |
| 2721 "uxtl2 v3.4s, v1.8h \n" | 2722 "uxtl2 v3.4s, v1.8h \n" |
| 2722 "scvtf v2.4s, v2.4s \n" // 8 floats | 2723 "scvtf v2.4s, v2.4s \n" // 8 floats |
| 2723 "scvtf v3.4s, v3.4s \n" | 2724 "scvtf v3.4s, v3.4s \n" |
| 2724 "fcvtn v1.4h, v2.4s \n" // 8 floatsgit | 2725 "fcvtn v1.4h, v2.4s \n" // 8 half floats |
| 2725 "fcvtn2 v1.8h, v3.4s \n" | 2726 "fcvtn2 v1.8h, v3.4s \n" |
| 2726 MEMACCESS(1) | 2727 MEMACCESS(1) |
| 2727 "st1 {v1.16b}, [%1], #16 \n" // store 8 shorts | 2728 "st1 {v1.16b}, [%1], #16 \n" // store 8 shorts |
| 2728 "b.gt 1b \n" | 2729 "b.gt 1b \n" |
| 2729 : "+r"(src), // %0 | 2730 : "+r"(src), // %0 |
| 2730 "+r"(dst), // %1 | 2731 "+r"(dst), // %1 |
| 2731 "+r"(width) // %2 | 2732 "+r"(width) // %2 |
| 2732 : | 2733 : |
| 2733 : "cc", "memory", "v1", "v2", "v3" | 2734 : "cc", "memory", "v1", "v2", "v3" |
| 2734 ); | 2735 ); |
| (...skipping 23 matching lines...) Expand all Loading... |
| 2758 : "cc", "memory", "v1", "v2", "v3" | 2759 : "cc", "memory", "v1", "v2", "v3" |
| 2759 ); | 2760 ); |
| 2760 } | 2761 } |
| 2761 | 2762 |
| 2762 #endif // !defined(LIBYUV_DISABLE_NEON) && defined(__aarch64__) | 2763 #endif // !defined(LIBYUV_DISABLE_NEON) && defined(__aarch64__) |
| 2763 | 2764 |
| 2764 #ifdef __cplusplus | 2765 #ifdef __cplusplus |
| 2765 } // extern "C" | 2766 } // extern "C" |
| 2766 } // namespace libyuv | 2767 } // namespace libyuv |
| 2767 #endif | 2768 #endif |
| OLD | NEW |