projects / platform / upstream / libvpx.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1239be9) | patch
quantize: replace highbd versions
authorJohann <johannkoenig@google.com>
Wed, 30 Mar 2022 05:57:46 +0000 (14:57 +0900)
committerJohann Koenig <johannkoenig@google.com>
Thu, 31 Mar 2022 00:43:52 +0000 (00:43 +0000)
commit2200039d33c49a9f7a5c438656df143755b022c4
tree92759b7e41336f04e41833c78e5fae4f5e171954tree | snapshot
parent1239be9e5faf3f7c1603548ed669af35639b74f8commit | diff
quantize: replace highbd versions

The optimized quantize functions were already built to handle
highbd values. The only difference is the clamping. All highbd
functions expand to 32bits when running in highbd mode.

Removes vpx_highbd_quantize_32x32_sse2 as it is slower than the
C version in the worst case.

Bug: webm:1586
Change-Id: I49bf8a6a2041f78450bf43a4f655c67656b0f8d9
test/vp9_quantize_test.cc diff | blob | history
vp9/encoder/vp9_encodemb.c diff | blob | history
vp9/encoder/vp9_quantize.c diff | blob | history
vpx_dsp/quantize.h diff | blob | history
vpx_dsp/vpx_dsp.mk diff | blob | history
vpx_dsp/vpx_dsp_rtcd_defs.pl diff | blob | history
vpx_dsp/x86/highbd_quantize_intrin_sse2.c [deleted file] blob | history
Domain: Multimedia / Codec;