source/libvpx/vp9/common/x86/vp9_idct_intrin_sse2.c - Issue 592203002: libvpx: Pull from upstream

Keyboard Shortcuts

	File
u :	up to issue
j / k :	jump to file after / before current file
J / K :	jump to next file with a comment after / before current file
	Side-by-side diff
i :	toggle intra-line diffs
e :	expand all comments
c :	collapse all comments
s :	toggle showing all comments
n / p :	next / previous diff chunk or comment
N / P :	next / previous comment
<Up> / <Down> :	next / previous line

	Issue
u :	up to list of issues
j / k :	jump to patch after / before current patch
o / <Enter> :	open current patch in side-by-side view
i :	open current patch in unified diff view

	Issue List
j / k :	jump to issue after / before current issue
o / <Enter> :	open current issue

Unified Diff: source/libvpx/vp9/common/x86/vp9_idct_intrin_sse2.c

Issue 592203002: libvpx: Pull from upstream (Closed) Base URL: svn://svn.chromium.org/chrome/trunk/deps/third_party/libvpx/

Patch Set: Created 6 years, 3 months ago

Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.

Jump to:

View side-by-side diff with in-line comments

Download patch

« no previous file with comments | « source/libvpx/vp9/common/x86/vp9_high_subpixel_bilinear_sse2.asm ('k') | source/libvpx/vp9/common/x86/vp9_loopfilter_intrin_sse2.c » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Hide Comments ('s')

Index: source/libvpx/vp9/common/x86/vp9_idct_intrin_sse2.c

===================================================================

--- source/libvpx/vp9/common/x86/vp9_idct_intrin_sse2.c (revision 292072)

+++ source/libvpx/vp9/common/x86/vp9_idct_intrin_sse2.c (working copy)

@@ -3573,6 +3573,7 @@

int stride) {

const __m128i rounding = _mm_set1_epi32(DCT_CONST_ROUNDING);

const __m128i final_rounding = _mm_set1_epi16(1<<5);

+ const __m128i zero = _mm_setzero_si128();

// idct constants for each stage

const __m128i stg1_0 = pair_set_epi16(cospi_31_64, -cospi_1_64);

@@ -3635,7 +3636,6 @@

stp2_30, stp2_31;

__m128i tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;

int i, j, i32;

- int zero_flag[2];

for (i = 0; i < 4; i++) {

i32 = (i << 5);

@@ -3710,13 +3710,7 @@

zero_idx[13] = _mm_or_si128(zero_idx[10], zero_idx[11]);

zero_idx[14] = _mm_or_si128(zero_idx[12], zero_idx[13]);

- zero_idx[0] = _mm_unpackhi_epi64(zero_idx[14], zero_idx[14]);

- zero_idx[1] = _mm_or_si128(zero_idx[0], zero_idx[14]);

- zero_idx[2] = _mm_srli_epi64(zero_idx[1], 32);

- zero_flag[0] = _mm_cvtsi128_si32(zero_idx[1]);

- zero_flag[1] = _mm_cvtsi128_si32(zero_idx[2]);

- if (!zero_flag[0] && !zero_flag[1]) {

+ if (_mm_movemask_epi8(_mm_cmpeq_epi32(zero_idx[14], zero)) == 0xFFFF) {

col[i32 + 0] = _mm_setzero_si128();

col[i32 + 1] = _mm_setzero_si128();

col[i32 + 2] = _mm_setzero_si128();

@@ -3795,7 +3789,6 @@

col[i32 + 31] = _mm_sub_epi16(stp1_0, stp1_31);

}

for (i = 0; i < 4; i++) {

- const __m128i zero = _mm_setzero_si128();

// Second 1-D idct

j = i << 3;