generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

partial fdct neon: add 4x4_1

Browse Source 
BUG=webm:1424

Change-Id: Ib0f3cfd6116fc1f5a99acb8bfd76e25b90177ffc
This commit is contained in:
Johann 2017-06-22 18:12:28 -07:00
parent cf75ab6ccd
commit 4959dd3eb3
3 changed files with 46 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
test/dct_partial_test.cc
View File

@ -144,11 +144,13 @@ INSTANTIATE_TEST_CASE_P(
NEON, PartialFdctTest,
::testing::Values(make_tuple(&vpx_fdct8x8_1_neon, 8, VPX_BITS_12),
make_tuple(&vpx_fdct8x8_1_neon, 8, VPX_BITS_10),
make_tuple(&vpx_fdct8x8_1_neon, 8, VPX_BITS_8)));
make_tuple(&vpx_fdct8x8_1_neon, 8, VPX_BITS_8),
make_tuple(&vpx_fdct4x4_1_neon, 4, VPX_BITS_8)));
#else
INSTANTIATE_TEST_CASE_P(NEON, PartialFdctTest,
::testing::Values(make_tuple(&vpx_fdct8x8_1_neon, 8,
VPX_BITS_8)));
INSTANTIATE_TEST_CASE_P(
NEON, PartialFdctTest,
::testing::Values(make_tuple(&vpx_fdct8x8_1_neon, 8, VPX_BITS_8),
make_tuple(&vpx_fdct4x4_1_neon, 4, VPX_BITS_8)));
#endif // CONFIG_VP9_HIGHBITDEPTH
#endif // HAVE_NEON

50
vpx_dsp/arm/fdct_partial_neon.c
View File

@ -12,6 +12,41 @@
#include "./vpx_dsp_rtcd.h"
#include "./vpx_config.h"
#include "vpx_dsp/arm/mem_neon.h"
static INLINE tran_low_t sum_int16x8(const int16x8_t a) {
const int32x4_t b = vpaddlq_s16(a);
const int64x2_t c = vpaddlq_s32(b);
const int32x2_t d = vadd_s32(vreinterpret_s32_s64(vget_low_s64(c)),
vreinterpret_s32_s64(vget_high_s64(c)));
#if CONFIG_VP9_HIGHBITDEPTH
return vget_lane_s32(d, 0);
#else
return vget_lane_s16(vreinterpret_s16_s32(d), 0);
#endif
}
void vpx_fdct4x4_1_neon(const int16_t *input, tran_low_t *output, int stride) {
int16x4_t a0, a1, a2, a3;
int16x8_t b0, b1;
int16x8_t c;
a0 = vld1_s16(input);
input += stride;
a1 = vld1_s16(input);
input += stride;
a2 = vld1_s16(input);
input += stride;
a3 = vld1_s16(input);
b0 = vcombine_s16(a0, a1);
b1 = vcombine_s16(a2, a3);
c = vaddq_s16(b0, b1);
output[0] = sum_int16x8(c) << 1;
output[1] = 0;
}
void vpx_fdct8x8_1_neon(const int16_t *input, tran_low_t *output, int stride) {
int r;
@ -20,16 +55,7 @@ void vpx_fdct8x8_1_neon(const int16_t *input, tran_low_t *output, int stride) {
const int16x8_t input_00 = vld1q_s16(&input[r * stride]);
sum = vaddq_s16(sum, input_00);
}
{
const int32x4_t a = vpaddlq_s16(sum);
const int64x2_t b = vpaddlq_s32(a);
const int32x2_t c = vadd_s32(vreinterpret_s32_s64(vget_low_s64(b)),
vreinterpret_s32_s64(vget_high_s64(b)));
#if CONFIG_VP9_HIGHBITDEPTH
output[0] = vget_lane_s32(c, 0);
#else
output[0] = vget_lane_s16(vreinterpret_s16_s32(c), 0);
#endif
output[1] = 0;
}
output[0] = sum_int16x8(sum);
output[1] = 0;
}

4
vpx_dsp/vpx_dsp_rtcd_defs.pl
View File

@ -487,7 +487,7 @@ if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
specialize qw/vpx_fdct4x4 neon sse2/;
add_proto qw/void vpx_fdct4x4_1/, "const int16_t *input, tran_low_t *output, int stride";
specialize qw/vpx_fdct4x4_1 sse2/;
specialize qw/vpx_fdct4x4_1 sse2 neon/;
add_proto qw/void vpx_fdct8x8/, "const int16_t *input, tran_low_t *output, int stride";
specialize qw/vpx_fdct8x8 neon sse2/;
@ -537,7 +537,7 @@ if (vpx_config("CONFIG_VP9_HIGHBITDEPTH") eq "yes") {
specialize qw/vpx_fdct4x4 neon sse2 msa/;
add_proto qw/void vpx_fdct4x4_1/, "const int16_t *input, tran_low_t *output, int stride";
specialize qw/vpx_fdct4x4_1 sse2/;
specialize qw/vpx_fdct4x4_1 sse2 neon/;
add_proto qw/void vpx_fdct8x8/, "const int16_t *input, tran_low_t *output, int stride";
specialize qw/vpx_fdct8x8 sse2 neon msa/, "$ssse3_x86_64";