History

quantize avx: copy 32x32 implementation

Ensure avx and ssse3 stay in sync by testing them against each other.

Johann committed 9 years ago

Commit f60d1dc1 parent 1787e7d5 files · +223 −512Browse files

Showing the changes to test (1 of 5 files). Show all

Unified · Split

test/vp9_quantize_test.cc+8−10
@@ -392,16 +392,14 @@ INSTANTIATE_TEST_CASE_P( · 391 unmodified lines
392392 // TODO(johannkoenig): AVX optimizations do not yet pass the 32x32 test or
393393 // highbitdepth configurations.
394394 #if HAVE_AVX && !CONFIG_VP9_HIGHBITDEPTH
395−INSTANTIATE_TEST_CASE_P(AVX, VP9QuantizeTest,
396− ::testing::Values(make_tuple(&vpx_quantize_b_avx,
397− &vpx_quantize_b_c,
398− VPX_BITS_8, 16)));
399−#if ARCH_X86_64
400−INSTANTIATE_TEST_CASE_P(DISABLED_AVX, VP9QuantizeTest,
401− ::testing::Values(make_tuple(&vpx_quantize_b_32x32_avx,
402− &vpx_quantize_b_32x32_c,
403− VPX_BITS_8, 32)));
404−#endif // ARCH_X86_64
395+INSTANTIATE_TEST_CASE_P(
396+ AVX, VP9QuantizeTest,
397+ ::testing::Values(make_tuple(&vpx_quantize_b_avx, &vpx_quantize_b_c,
398+ VPX_BITS_8, 16),
399+ // Even though SSSE3 and AVX do not match the reference
400+ // code, we can keep them in sync with each other.
401+ make_tuple(&vpx_quantize_b_32x32_avx,
402+ &vpx_quantize_b_32x32_ssse3, VPX_BITS_8, 32)));
405403 #endif // HAVE_AVX && !CONFIG_VP9_HIGHBITDEPTH
406404
407405 // TODO(webm:1448): dqcoeff is not handled correctly in HBD builds.

Keyboard shortcuts

General

Show keyboard shortcuts
?
Focus the current search
/

Issues and pull requests

Open filters
f

Issues, pull requests and Inbox

Select next item
j↓
Select previous item
k↑
Open selected item
Enter
Dismiss filters or clear selection
Esc

Repository code

Go to file
t