~mattst88/pixman - Unnamed repository; edit this file to name it for gitweb.

Branch	Commit message	Author	Age
avx2	try doing a single load and a single store	Matt Turner	5 years
bilinear-7bit	mmx: optimize bilinear function when using 7-bit precision	Matt Turner	12 years
load-store	wip commit to remove load8888/store8888 from combine funcs	root	12 years
loongson	mmx: optimize over_8888_0565 main loop	Matt Turner	12 years
loongson-wip2	fix up on x86	Matt Turner	12 years
loongson3	mmx	Matt Turner	6 years
master	MIPS: DSPr2: Added over_n_8888_8888_ca and over_n_8888_0565_ca fast paths.	Nemanja Lukic	12 years
mmx-fetchers	mmx: add x8f8g8b8 fetcher	Matt Turner	12 years
mmx-nearest	mmx: faster bilinear interpolation (get rid of XOR instruction)	Matt Turner	10 years
no-xor-bilinear	mmx: faster bilinear interpolation (get rid of XOR instruction)	Matt Turner	11 years
[...]


Age	Commit message	Author	Files	Lines
2019-04-01	try doing a single load and a single storeavx2	Matt Turner	1	-14/+11
2019-04-01	wip	Matt Turner	2	-23/+47
2019-03-30	avx2: Use _mm_maddubs_epi16 which isn't available in SSE2	Matt Turner	1	-7/+3
2019-03-30	avx2: Use _mm_set1_*	Matt Turner	1	-7/+6
2019-03-30	avx2: Only support bilinear scaling for BILINEAR_INTERPOLATION_BITS < 8	Matt Turner	1	-25/+13
2019-03-30	avx2: Copy SSE2 bilinear src_8888_8888 implementation	Matt Turner	1	-0/+131
2019-03-30	avx2: Add add_8888_8888	Matt Turner	1	-0/+54
2019-03-30	Add AVX2 implementation	Matt Turner	5	-0/+120
2019-03-27	void function should not return a value	Niveditha Rau	1	-6/+24
2019-03-27	Windows: Support building with SHELL=cmd.exe	Simon Richter	1	-2/+14
[...]

Clone
git://people.freedesktop.org/~mattst88/pixman
ssh://people.freedesktop.org/~mattst88/pixman