summaryrefslogtreecommitdiffstats
path: root/postproc/swscale.c
Commit message (Collapse)AuthorAgeFilesLines
* print more info if -vmichael2001-12-061-4/+201
| | | | | | | | | use new horizontal mmx scaler instead of old x86asm if mmx2 cant be used (FAST_BILINEAR only) fixed overflow in init function ... using double precission fp now :) using C scaler for the last 1-2 lines if there is a chance to write over the end of the dst array git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3353 b3059339-0415-0410-9bf9-f77b7e298cf2
* vertical cubic/linear scalingmichael2001-12-061-22/+55
| | | | | | | | prints some info (MMX, dstbpp, ...) mainloop rewritten git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3345 b3059339-0415-0410-9bf9-f77b7e298cf2
* horizontal up/downscale linear & cubicmichael2001-12-021-2/+17
| | | | | | | -sws command line option (0 -> fast_bilinear, 1->bilinear, 2->bicubic) git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3273 b3059339-0415-0410-9bf9-f77b7e298cf2
* swscaler cleanupmichael2001-11-291-17/+13
| | | | | | | | green line at bottom bugfix green lines in yuv2yuv scaler bugfix git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3210 b3059339-0415-0410-9bf9-f77b7e298cf2
* runtime cpu detection optional (compiles faster)michael2001-11-271-1/+44
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3153 b3059339-0415-0410-9bf9-f77b7e298cf2
* oopsmichael2001-11-261-1/+0
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3137 b3059339-0415-0410-9bf9-f77b7e298cf2
* runtime cpu detectionmichael2001-11-261-1652/+69
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@3127 b3059339-0415-0410-9bf9-f77b7e298cf2
* .balign or we´ll align by 64kb on some architecturesmichael2001-11-101-6/+6
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2801 b3059339-0415-0410-9bf9-f77b7e298cf2
* .align 16michael2001-11-101-0/+6
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2800 b3059339-0415-0410-9bf9-f77b7e298cf2
* gcc does optimize writes to non volatile variables away if it didnt know ↵michael2001-11-061-4/+4
| | | | | | that they were read in between git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2751 b3059339-0415-0410-9bf9-f77b7e298cf2
* better ditheringmichael2001-11-061-42/+43
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2749 b3059339-0415-0410-9bf9-f77b7e298cf2
* faster bgr24 outputmichael2001-11-061-3/+114
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2731 b3059339-0415-0410-9bf9-f77b7e298cf2
* only 6 registers usedmichael2001-11-051-7/+10
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2729 b3059339-0415-0410-9bf9-f77b7e298cf2
* green line fix for dstw%8!=0michael2001-11-041-4/+11
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2681 b3059339-0415-0410-9bf9-f77b7e298cf2
* fixed warningsmichael2001-11-041-27/+48
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2672 b3059339-0415-0410-9bf9-f77b7e298cf2
* faster bgr15/16michael2001-11-041-50/+28
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2670 b3059339-0415-0410-9bf9-f77b7e298cf2
* right green line bugfix for width not %8==0 (untested -vo vesa doesnt work)michael2001-11-021-1/+12
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2639 b3059339-0415-0410-9bf9-f77b7e298cf2
* c speedupmichael2001-10-311-17/+19
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2586 b3059339-0415-0410-9bf9-f77b7e298cf2
* 15/16bit in C speedupmichael2001-10-311-30/+50
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2585 b3059339-0415-0410-9bf9-f77b7e298cf2
* c optimizationsmichael2001-10-301-24/+75
| | | | | | | bugfix git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2577 b3059339-0415-0410-9bf9-f77b7e298cf2
* c optimizationsmichael2001-10-301-17/+68
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2576 b3059339-0415-0410-9bf9-f77b7e298cf2
* c optimizations (array is faster than pointer) (16bpp variants tested and 2% ↵michael2001-10-301-18/+9
| | | | | | faster) git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2573 b3059339-0415-0410-9bf9-f77b7e298cf2
* interpolate chrominance for every second line in the 1:1 vertical scale functionmichael2001-10-301-41/+153
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2570 b3059339-0415-0410-9bf9-f77b7e298cf2
* bugfixesmichael2001-10-301-7/+9
| | | | | | | small speedup for the non x86 stuff git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2567 b3059339-0415-0410-9bf9-f77b7e298cf2
* fixed a warningmichael2001-10-291-0/+1
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2541 b3059339-0415-0410-9bf9-f77b7e298cf2
* patch from Martin Decky <deckm1am@ss1000.ms.mff.cuni.cz> applied and ↵michael2001-10-291-2/+5
| | | | | | unnecassery "memory" removed git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2535 b3059339-0415-0410-9bf9-f77b7e298cf2
* tell the c compiler that the memory changedmichael2001-10-281-0/+8
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2522 b3059339-0415-0410-9bf9-f77b7e298cf2
* 3dnow prefetch & sfencemichael2001-10-281-26/+11
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2521 b3059339-0415-0410-9bf9-f77b7e298cf2
* yv12 to yv12 scalermichael2001-10-281-21/+40
| | | | | | | someone who knows a bit about vo_odivx could add support for it ... git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2520 b3059339-0415-0410-9bf9-f77b7e298cf2
* yuv2rgb bugfixmichael2001-10-281-31/+31
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2504 b3059339-0415-0410-9bf9-f77b7e298cf2
* a few warning fixes (missing #include's)pl2001-10-251-0/+1
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2477 b3059339-0415-0410-9bf9-f77b7e298cf2
* downscalemichael2001-10-251-279/+326
| | | | | | | fixed some warnings git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2470 b3059339-0415-0410-9bf9-f77b7e298cf2
* BGR24 bugfixmichael2001-10-211-5/+5
| | | | | | | BGR15 & BGR24 tested git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2327 b3059339-0415-0410-9bf9-f77b7e298cf2
* fixed alignment (static variables where sometimes not 8-byte aligned)michael2001-10-201-289/+888
| | | | | | | | | | added half uv interpolation support added prefetch BGR15 support in MMX (untested) (so BGR15,16,24,32 are supported) special unscaled height version (not much faster but it doesnt interpolate uv vertically) git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2317 b3059339-0415-0410-9bf9-f77b7e298cf2
* first line bug fixed (happend if the image was scaled down or 1:1)michael2001-10-191-2/+7
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2298 b3059339-0415-0410-9bf9-f77b7e298cf2
* horizontal scaling bugs fixed, should be mostly bugfree nowmichael2001-10-191-25/+28
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2280 b3059339-0415-0410-9bf9-f77b7e298cf2
* xinc scaled by 16 instead of 8arpi2001-10-191-2/+2
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2275 b3059339-0415-0410-9bf9-f77b7e298cf2
* horizontal lines bugfixmichael2001-10-191-4/+4
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2272 b3059339-0415-0410-9bf9-f77b7e298cf2
* vertical lines bugfixmichael2001-10-191-7/+12
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2271 b3059339-0415-0410-9bf9-f77b7e298cf2
* (C) fixedarpi2001-10-191-1/+1
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2270 b3059339-0415-0410-9bf9-f77b7e298cf2
* fixed a bug which caused horizontal linesmichael2001-10-191-2/+2
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2268 b3059339-0415-0410-9bf9-f77b7e298cf2
* increased precission of s_xinc s_xinc2 (needed for the mmx2 bugfix)michael2001-10-181-206/+271
| | | | | | | | | | | | | | moved mmx variables to top to avoid alignment issues mmx2 code should work fine now if and only if the input width is %16=0 and the output width is %32=0 reordered some code (5% faster with a simply -benchmark) first line bug fixed (i hope i didnt introduce any new bugs with that ...) changed a lot of the vertical scale setup code, i hope i fixed something and didnt mess it up :) a few known bugs left (rightmost line is wrong) MMX2 code will only be used for upscaling & acceptable width´s 16bit dithering can be disabled git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2265 b3059339-0415-0410-9bf9-f77b7e298cf2
* 24bpp support (untested)michael2001-10-171-2/+52
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2238 b3059339-0415-0410-9bf9-f77b7e298cf2
* added (f)emmsarpi2001-10-171-0/+7
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2237 b3059339-0415-0410-9bf9-f77b7e298cf2
* bpp changed from bytespp to bitspparpi2001-10-171-8/+8
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2236 b3059339-0415-0410-9bf9-f77b7e298cf2
* fixed out of regs "bug"michael2001-10-171-6/+9
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2234 b3059339-0415-0410-9bf9-f77b7e298cf2
* MMX & MMX2 optimizations (MMX2 is buggy and commented out)michael2001-10-171-36/+563
| | | | | | | | | 32, 24, 16, 15 bit support in C (only 32 & 16 tested) 32, 16 bit support in mmx (16 with dithering) ranges of some variables changed so mmx likes them more git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2233 b3059339-0415-0410-9bf9-f77b7e298cf2
* fix rgb/bgr for 15/16bppatmos42001-10-171-4/+4
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2231 b3059339-0415-0410-9bf9-f77b7e298cf2
* 16bpp code for atmosarpi2001-10-161-0/+10
| | | | git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2230 b3059339-0415-0410-9bf9-f77b7e298cf2
* software yv12->rgb scaler - separated from fsdgaarpi2001-10-151-0/+124
git-svn-id: svn://svn.mplayerhq.hu/mplayer/trunk@2217 b3059339-0415-0410-9bf9-f77b7e298cf2
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-08 11:34:21 +0000