Commit d19d222d authored by Scott LaVarnway's avatar Scott LaVarnway

Added vp9_fdct8x8_neon(), vp9_fdct8x8_1_neon()

On a Nexus 7, vpxenc (in realtime mode, speed -12)
reported a performance improvement of ~3.7%.

Change-Id: I428c72c40df82c6d537955e320a8debf99343004
parent 4ba92dc5
......@@ -337,7 +337,7 @@ INSTANTIATE_TEST_CASE_P(
INSTANTIATE_TEST_CASE_P(
NEON, FwdTrans8x8DCT,
::testing::Values(
make_tuple(&vp9_fdct8x8_c, &vp9_idct8x8_64_add_neon, 0)));
make_tuple(&vp9_fdct8x8_neon, &vp9_idct8x8_64_add_neon, 0)));
INSTANTIATE_TEST_CASE_P(
DISABLED_NEON, FwdTrans8x8HT,
::testing::Values(
......
......@@ -757,10 +757,10 @@ add_proto qw/void vp9_fdct4x4/, "const int16_t *input, int16_t *output, int stri
specialize qw/vp9_fdct4x4 sse2/;
add_proto qw/void vp9_fdct8x8_1/, "const int16_t *input, int16_t *output, int stride";
specialize qw/vp9_fdct8x8_1 sse2/;
specialize qw/vp9_fdct8x8_1 sse2 neon/;
add_proto qw/void vp9_fdct8x8/, "const int16_t *input, int16_t *output, int stride";
specialize qw/vp9_fdct8x8 sse2/, "$ssse3_x86_64";
specialize qw/vp9_fdct8x8 sse2 neon/, "$ssse3_x86_64";
add_proto qw/void vp9_fdct16x16_1/, "const int16_t *input, int16_t *output, int stride";
specialize qw/vp9_fdct16x16_1 sse2/;
......
This diff is collapsed.
......@@ -130,5 +130,6 @@ VP9_CX_SRCS-$(HAVE_AVX2) += encoder/x86/vp9_error_intrin_avx2.c
VP9_CX_SRCS-$(HAVE_AVX2) += encoder/x86/vp9_variance_avx2.c
VP9_CX_SRCS-$(HAVE_NEON) += encoder/arm/neon/vp9_sad_neon.c
VP9_CX_SRCS-$(HAVE_NEON) += encoder/arm/neon/vp9_dct_neon.c
VP9_CX_SRCS-yes := $(filter-out $(VP9_CX_SRCS_REMOVE-yes),$(VP9_CX_SRCS-yes))
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment