Shreesh Adiga 26f2f03e0d swscale/x86/rgb2rgb: optimize AVX2 version of uyvytoyuv422 ... Currently the AVX2 version of uyvytoyuv422 in the SIMD loop does the following: 4 vinsertq to have interleaving of the vector lanes during load from memory. 4 vperm2i128 inside 4 RSHIFT_COPY calls to achieve the desired layout. This patch replaces the above 8 instructions with 2 vpermq and 2 vpermd with a vector register similar to AVX512ICL version. Observed the following numbers on various microarchitectures: On AMD Zen3 laptop: Before: uyvytoyuv422_c: 51979.7 ( 1.00x) uyvytoyuv422_sse2: 5410.5 ( 9.61x) uyvytoyuv422_avx: 4642.7 (11.20x) uyvytoyuv422_avx2: 4249.0 (12.23x) After: uyvytoyuv422_c: 51659.8 ( 1.00x) uyvytoyuv422_sse2: 5420.8 ( 9.53x) uyvytoyuv422_avx: 4651.2 (11.11x) uyvytoyuv422_avx2: 3953.8 (13.07x) On Intel Macbook Pro 2019: Before: uyvytoyuv422_c: 185014.4 ( 1.00x) uyvytoyuv422_sse2: 22800.4 ( 8.11x) uyvytoyuv422_avx: 19796.9 ( 9.35x) uyvytoyuv422_avx2: 13141.9 (14.08x) After: uyvytoyuv422_c: 185093.4 ( 1.00x) uyvytoyuv422_sse2: 22795.4 ( 8.12x) uyvytoyuv422_avx: 19791.9 ( 9.35x) uyvytoyuv422_avx2: 12043.1 (15.37x) On AMD Zen4 desktop: Before: uyvytoyuv422_c: 29105.0 ( 1.00x) uyvytoyuv422_sse2: 3888.0 ( 7.49x) uyvytoyuv422_avx: 3374.2 ( 8.63x) uyvytoyuv422_avx2: 2649.8 (10.98x) uyvytoyuv422_avx512icl: 1615.0 (18.02x) After: uyvytoyuv422_c: 29093.4 ( 1.00x) uyvytoyuv422_sse2: 3874.4 ( 7.51x) uyvytoyuv422_avx: 3371.6 ( 8.63x) uyvytoyuv422_avx2: 2174.6 (13.38x) uyvytoyuv422_avx512icl: 1625.1 (17.90x) Signed-off-by: Shreesh Adiga <16567adigashreesh@gmail.com>		2025-03-23 15:25:48 +00:00
..
aarch64	swscale: aarch64: Simplify the assignment of lumToYV12	2025-03-10 14:03:58 +02:00
arm	libswscale/arm/swscale_unscaled: Fix function prototype	2025-03-02 01:10:38 +02:00
loongarch	loongarch: fixes fate-checkasm-sw_rgb failure	2025-01-15 01:27:36 +01:00
ppc	avutil/libm: Only include intfloat.h when needed	2025-03-22 03:35:28 +01:00
riscv	swscale/range_convert: saturate output instead of limiting input	2024-12-05 21:10:29 +01:00
tests	swscale/tests/swscale: Fix potential buffer overflow	2025-03-21 04:30:09 +01:00
x86	swscale/x86/rgb2rgb: optimize AVX2 version of uyvytoyuv422	2025-03-23 15:25:48 +00:00
alphablend.c	swscale/internal: group user-facing options together	2024-11-21 12:49:56 +01:00
bayer_template.c	swscale/internal: constify SwsFunc	2024-10-07 19:51:34 +02:00
cms.c	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
cms.h	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
csputils.c	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
csputils.h	swscale/csputils: add internal colorspace math helpers	2024-12-23 12:33:43 +01:00
format.c	avutil/csp: Improve enum range comparisons	2025-03-21 04:30:09 +01:00
format.h	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
gamma.c	swscale: rename SwsContext to SwsInternal	2024-10-24 22:50:00 +02:00
graph.c	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
graph.h	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
half2float.c	swscale/input: add rgbaf16 input support	2022-08-19 22:09:36 +02:00
hscale.c	swscale/range_convert: fix mpeg ranges in yuv range conversion for non-8-bit pixel formats	2024-12-05 21:10:29 +01:00
hscale_fast_bilinear.c	swscale: rename SwsContext to SwsInternal	2024-10-24 22:50:00 +02:00
input.c	avutil/libm: Only include intfloat.h when needed	2025-03-22 03:35:28 +01:00
libswscale.v	build: Change structure of the linker version script templates	2016-05-29 16:43:11 +02:00
log2_tab.c	lsws: duplicate ff_log2_tab	2014-08-12 20:52:21 +02:00
lut3d.c	swscale/cms,graph,lut3d: Use ff_-prefix, don't export internal functions	2025-01-12 15:41:39 +01:00
lut3d.h	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
Makefile	avutil: only duplicate hal2float and float2half in shared builds	2025-03-18 17:21:23 -03:00
options.c	swscale/options: add -sws_dither none alias	2024-12-23 12:47:10 +01:00
output.c	avutil/libm: Only include intfloat.h when needed	2025-03-22 03:35:28 +01:00
rgb2rgb.c	swscale/swscale_unscaled: add unscaled x2rgb10le to packed RGB	2024-11-06 17:34:32 -03:00
rgb2rgb.h	swscale/swscale_unscaled: add unscaled x2rgb10le to packed RGB	2024-11-06 17:34:32 -03:00
rgb2rgb_template.c	swscale/swscale_unscaled: add unscaled conversion for AYUV/VUYA/UYVA	2024-11-02 15:01:31 -03:00
slice.c	swscale/slice: fix init of 32 bpc planes	2024-12-16 12:21:55 +01:00
swscale.c	swscale: fix gray -> grayf32 SIGFPE	2025-03-17 11:40:05 +01:00
swscale.h	swscale: add ICC intent enum and option	2024-12-23 12:33:43 +01:00
swscale_internal.h	swscale: use 16-bit intermediate precision for RGB/XYZ conversion	2024-12-26 20:31:36 +01:00
swscale_unscaled.c	swscale/swscale_unscaled: avoid nv12 <-> nv21 bug	2025-03-17 11:40:05 +01:00
swscaleres.rc	Add Windows resource file support for shared libraries	2013-12-05 23:42:07 +01:00
utils.c	swscale/utils: split off format code into new file	2025-03-14 19:50:44 +01:00
version.c	lib*/version: Use static_assert for static asserts	2024-03-31 00:08:42 +01:00
version.h	swscale/output: add support for NV20	2025-03-19 09:34:05 -03:00
version_major.h	libs: bump major version for all libraries	2024-03-07 11:29:43 -03:00
vscale.c	swscale/internal: group user-facing options together	2024-11-21 12:49:56 +01:00
yuv2rgb.c	swscale/internal: group user-facing options together	2024-11-21 12:49:56 +01:00