]> granicus.if.org Git - libvpx/commit
Adds SSE2 support for interpolation filter for width 4 and 8
authorchiyotsai <chiyotsai@google.com>
Tue, 16 Oct 2018 22:45:05 +0000 (15:45 -0700)
committerchiyotsai <chiyotsai@google.com>
Wed, 17 Oct 2018 20:29:13 +0000 (13:29 -0700)
commit71b4e0bdedefea29dd27ebaae20ffb8703ee4703
tree3c9dd776b0590d79ff9c96f34cd5f5bc4e9e5a57
parent62830c53a644f5feaa49431b39c85093d2e387fc
Adds SSE2 support for interpolation filter for width 4 and 8

Performance:
The chart below shows the speed relative to baseline
(baseline_time/new_time)
_____| 4X4 | 8X8 |16X16|64X64|
2 DIM|1.889|1.780|1.811|1.963|
 HORZ|2.266|1.834|1.617|1.595|
 VERI|2.043|2.190|2.373|2.485|

Change-Id: Ic4262222db78f013b94a8c61b46efb8520722927
vpx_dsp/x86/convolve.h
vpx_dsp/x86/vpx_asm_stubs.c
vpx_dsp/x86/vpx_subpixel_4t_intrin_sse2.c
vpx_dsp/x86/vpx_subpixel_8t_intrin_avx2.c
vpx_dsp/x86/vpx_subpixel_8t_intrin_ssse3.c