diff options
Diffstat (limited to 'third_party/aom/aom_dsp/mips/idct4x4_msa.c')
-rw-r--r-- | third_party/aom/aom_dsp/mips/idct4x4_msa.c | 99 |
1 files changed, 0 insertions, 99 deletions
diff --git a/third_party/aom/aom_dsp/mips/idct4x4_msa.c b/third_party/aom/aom_dsp/mips/idct4x4_msa.c deleted file mode 100644 index 274818baa..000000000 --- a/third_party/aom/aom_dsp/mips/idct4x4_msa.c +++ /dev/null @@ -1,99 +0,0 @@ -/* - * Copyright (c) 2016, Alliance for Open Media. All rights reserved - * - * This source code is subject to the terms of the BSD 2 Clause License and - * the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License - * was not distributed with this source code in the LICENSE file, you can - * obtain it at www.aomedia.org/license/software. If the Alliance for Open - * Media Patent License 1.0 was not distributed with this source code in the - * PATENTS file, you can obtain it at www.aomedia.org/license/patent. - */ - -#include "aom_dsp/mips/inv_txfm_msa.h" - -void aom_iwht4x4_16_add_msa(const int16_t *input, uint8_t *dst, - int32_t dst_stride) { - v8i16 in0, in1, in2, in3; - v4i32 in0_r, in1_r, in2_r, in3_r, in4_r; - - /* load vector elements of 4x4 block */ - LD4x4_SH(input, in0, in2, in3, in1); - TRANSPOSE4x4_SH_SH(in0, in2, in3, in1, in0, in2, in3, in1); - UNPCK_R_SH_SW(in0, in0_r); - UNPCK_R_SH_SW(in2, in2_r); - UNPCK_R_SH_SW(in3, in3_r); - UNPCK_R_SH_SW(in1, in1_r); - SRA_4V(in0_r, in1_r, in2_r, in3_r, UNIT_QUANT_SHIFT); - - in0_r += in2_r; - in3_r -= in1_r; - in4_r = (in0_r - in3_r) >> 1; - in1_r = in4_r - in1_r; - in2_r = in4_r - in2_r; - in0_r -= in1_r; - in3_r += in2_r; - - TRANSPOSE4x4_SW_SW(in0_r, in1_r, in2_r, in3_r, in0_r, in1_r, in2_r, in3_r); - - in0_r += in1_r; - in2_r -= in3_r; - in4_r = (in0_r - in2_r) >> 1; - in3_r = in4_r - in3_r; - in1_r = in4_r - in1_r; - in0_r -= in3_r; - in2_r += in1_r; - - PCKEV_H4_SH(in0_r, in0_r, in1_r, in1_r, in2_r, in2_r, in3_r, in3_r, in0, in1, - in2, in3); - ADDBLK_ST4x4_UB(in0, in3, in1, in2, dst, dst_stride); -} - -void aom_iwht4x4_1_add_msa(const int16_t *input, uint8_t *dst, - int32_t dst_stride) { - int16_t a1, e1; - v8i16 in1, in0 = { 0 }; - - a1 = input[0] >> UNIT_QUANT_SHIFT; - e1 = a1 >> 1; - a1 -= e1; - - in0 = __msa_insert_h(in0, 0, a1); - in0 = __msa_insert_h(in0, 1, e1); - in0 = __msa_insert_h(in0, 2, e1); - in0 = __msa_insert_h(in0, 3, e1); - - in1 = in0 >> 1; - in0 -= in1; - - ADDBLK_ST4x4_UB(in0, in1, in1, in1, dst, dst_stride); -} - -void aom_idct4x4_16_add_msa(const int16_t *input, uint8_t *dst, - int32_t dst_stride) { - v8i16 in0, in1, in2, in3; - - /* load vector elements of 4x4 block */ - LD4x4_SH(input, in0, in1, in2, in3); - /* rows */ - TRANSPOSE4x4_SH_SH(in0, in1, in2, in3, in0, in1, in2, in3); - AOM_IDCT4x4(in0, in1, in2, in3, in0, in1, in2, in3); - /* columns */ - TRANSPOSE4x4_SH_SH(in0, in1, in2, in3, in0, in1, in2, in3); - AOM_IDCT4x4(in0, in1, in2, in3, in0, in1, in2, in3); - /* rounding (add 2^3, divide by 2^4) */ - SRARI_H4_SH(in0, in1, in2, in3, 4); - ADDBLK_ST4x4_UB(in0, in1, in2, in3, dst, dst_stride); -} - -void aom_idct4x4_1_add_msa(const int16_t *input, uint8_t *dst, - int32_t dst_stride) { - int16_t out; - v8i16 vec; - - out = ROUND_POWER_OF_TWO((input[0] * cospi_16_64), DCT_CONST_BITS); - out = ROUND_POWER_OF_TWO((out * cospi_16_64), DCT_CONST_BITS); - out = ROUND_POWER_OF_TWO(out, 4); - vec = __msa_fill_h(out); - - ADDBLK_ST4x4_UB(vec, vec, vec, vec, dst, dst_stride); -} |