diff options
author | Matt A. Tobin <mattatobin@localhost.localdomain> | 2018-02-02 04:16:08 -0500 |
---|---|---|
committer | Matt A. Tobin <mattatobin@localhost.localdomain> | 2018-02-02 04:16:08 -0500 |
commit | 5f8de423f190bbb79a62f804151bc24824fa32d8 (patch) | |
tree | 10027f336435511475e392454359edea8e25895d /media/libvpx/vp9/common/arm/neon/vp9_idct8x8_1_add_neon.c | |
parent | 49ee0794b5d912db1f95dce6eb52d781dc210db5 (diff) | |
download | UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.gz UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.lz UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.xz UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.zip |
Add m-esr52 at 52.6.0
Diffstat (limited to 'media/libvpx/vp9/common/arm/neon/vp9_idct8x8_1_add_neon.c')
-rw-r--r-- | media/libvpx/vp9/common/arm/neon/vp9_idct8x8_1_add_neon.c | 64 |
1 files changed, 64 insertions, 0 deletions
diff --git a/media/libvpx/vp9/common/arm/neon/vp9_idct8x8_1_add_neon.c b/media/libvpx/vp9/common/arm/neon/vp9_idct8x8_1_add_neon.c new file mode 100644 index 000000000..5369697c7 --- /dev/null +++ b/media/libvpx/vp9/common/arm/neon/vp9_idct8x8_1_add_neon.c @@ -0,0 +1,64 @@ +/* + * Copyright (c) 2014 The WebM project authors. All Rights Reserved. + * + * Use of this source code is governed by a BSD-style license + * that can be found in the LICENSE file in the root of the source + * tree. An additional intellectual property rights grant can be found + * in the file PATENTS. All contributing project authors may + * be found in the AUTHORS file in the root of the source tree. + */ + +#include <arm_neon.h> + +#include "vpx_ports/mem.h" +#include "vp9/common/vp9_idct.h" + +void vp9_idct8x8_1_add_neon( + int16_t *input, + uint8_t *dest, + int dest_stride) { + uint8x8_t d2u8, d3u8, d30u8, d31u8; + uint64x1_t d2u64, d3u64, d4u64, d5u64; + uint16x8_t q0u16, q9u16, q10u16, q11u16, q12u16; + int16x8_t q0s16; + uint8_t *d1, *d2; + int16_t i, a1, cospi_16_64 = 11585; + int16_t out = dct_const_round_shift(input[0] * cospi_16_64); + out = dct_const_round_shift(out * cospi_16_64); + a1 = ROUND_POWER_OF_TWO(out, 5); + + q0s16 = vdupq_n_s16(a1); + q0u16 = vreinterpretq_u16_s16(q0s16); + + d1 = d2 = dest; + for (i = 0; i < 2; i++) { + d2u64 = vld1_u64((const uint64_t *)d1); + d1 += dest_stride; + d3u64 = vld1_u64((const uint64_t *)d1); + d1 += dest_stride; + d4u64 = vld1_u64((const uint64_t *)d1); + d1 += dest_stride; + d5u64 = vld1_u64((const uint64_t *)d1); + d1 += dest_stride; + + q9u16 = vaddw_u8(q0u16, vreinterpret_u8_u64(d2u64)); + q10u16 = vaddw_u8(q0u16, vreinterpret_u8_u64(d3u64)); + q11u16 = vaddw_u8(q0u16, vreinterpret_u8_u64(d4u64)); + q12u16 = vaddw_u8(q0u16, vreinterpret_u8_u64(d5u64)); + + d2u8 = vqmovun_s16(vreinterpretq_s16_u16(q9u16)); + d3u8 = vqmovun_s16(vreinterpretq_s16_u16(q10u16)); + d30u8 = vqmovun_s16(vreinterpretq_s16_u16(q11u16)); + d31u8 = vqmovun_s16(vreinterpretq_s16_u16(q12u16)); + + vst1_u64((uint64_t *)d2, vreinterpret_u64_u8(d2u8)); + d2 += dest_stride; + vst1_u64((uint64_t *)d2, vreinterpret_u64_u8(d3u8)); + d2 += dest_stride; + vst1_u64((uint64_t *)d2, vreinterpret_u64_u8(d30u8)); + d2 += dest_stride; + vst1_u64((uint64_t *)d2, vreinterpret_u64_u8(d31u8)); + d2 += dest_stride; + } + return; +} |