1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980818283848586878889909192939495969798991001011021031041051061071081091101111121131141151161171181191201211221231241251261271281291301311321331341351361371381391401411421431441451461471481491501511521531541551561571581591601611621631641651661671681691701711721731741751761771781791801811821831841851861871881891901911921931941951961971981992002012022032042052062072082092102112122132142152162172182192202212222232242252262272282292302312322332342352362372382392402412422432442452462472482492502512522532542552562572582592602612622632642652662672682692702712722732742752762772782792802812822832842852862872882892902912922932942952962972982993003013023033043053063073083093103113123133143153163173183193203213223233243253263273283293303313323333343353363373383393403413423433443453463473483493503513523533543553563573583593603613623633643653663673683693703713723733743753763773783793803813823833843853863873883893903913923933943953963973983994004014024034044054064074084094104114124134144154164174184194204214224234244254264274284294304314324334344354364374384394404414424434444454464474484494504514524534544554564574584594604614624634644654664674684694704714724734744754764774784794804814824834844854864874884894904914924934944954964974984995005015025035045055065075085095105115125135145155165175185195205215225235245255265275285295305315325335345355365375385395405415425435445455465475485495505515525535545555565575585595605615625635645655665675685695705715725735745755765775785795805815825835845855865875885895905915925935945955965975985996006016026036046056066076086096106116126136146156166176186196206216226236246256266276286296306316326336346356366376386396406416426436446456466476486496506516526536546556566576586596606616626636646656666676686696706716726736746756766776786796806816826836846856866876886896906916926936946956966976986997007017027037047057067077087097107117127137147157167177187197207217227237247257267277287297307317327337347357367377387397407417427437447457467477487497507517527537547557567577587597607617627637647657667677687697707717727737747757767777787797807817827837847857867877887897907917927937947957967977987998008018028038048058068078088098108118128138148158168178188198208218228238248258268278288298308318328338348358368378388398408418428438448458468478488498508518528538548558568578588598608618628638648658668678688698708718728738748758768778788798808818828838848858868878888898908918928938948958968978988999009019029039049059069079089099109119129139149159169179189199209219229239249259269279289299309319329339349359369379389399409419429439449459469479489499509519529539549559569579589599609619629639649659669679689699709719729739749759769779789799809819829839849859869879889899909919929939949959969979989991000100110021003100410051006100710081009101010111012101310141015101610171018101910201021102210231024102510261027102810291030103110321033103410351036103710381039104010411042104310441045104610471048104910501051105210531054105510561057105810591060106110621063106410651066106710681069107010711072107310741075107610771078107910801081108210831084108510861087108810891090109110921093109410951096109710981099110011011102110311041105110611071108110911101111111211131114111511161117111811191120112111221123112411251126112711281129113011311132113311341135113611371138113911401141114211431144114511461147114811491150115111521153115411551156115711581159116011611162116311641165116611671168116911701171117211731174117511761177117811791180118111821183118411851186118711881189119011911192119311941195119611971198119912001201120212031204120512061207120812091210121112121213121412151216121712181219122012211222122312241225122612271228122912301231123212331234123512361237123812391240124112421243124412451246124712481249125012511252125312541255125612571258125912601261126212631264126512661267126812691270127112721273127412751276127712781279128012811282128312841285128612871288128912901291129212931294129512961297129812991300130113021303130413051306130713081309131013111312131313141315131613171318131913201321132213231324132513261327132813291330133113321333133413351336133713381339134013411342134313441345134613471348134913501351135213531354135513561357135813591360136113621363136413651366136713681369137013711372137313741375137613771378137913801381138213831384138513861387138813891390139113921393139413951396139713981399140014011402140314041405140614071408140914101411141214131414141514161417141814191420142114221423142414251426142714281429143014311432143314341435143614371438143914401441144214431444144514461447144814491450145114521453145414551456145714581459146014611462146314641465146614671468146914701471147214731474147514761477147814791480148114821483148414851486148714881489149014911492149314941495149614971498149915001501150215031504150515061507150815091510151115121513151415151516151715181519152015211522152315241525152615271528152915301531153215331534153515361537153815391540154115421543154415451546154715481549155015511552155315541555155615571558155915601561156215631564156515661567156815691570157115721573157415751576157715781579158015811582158315841585158615871588158915901591159215931594159515961597159815991600160116021603160416051606160716081609161016111612161316141615161616171618161916201621162216231624162516261627162816291630163116321633163416351636163716381639164016411642164316441645164616471648164916501651165216531654165516561657165816591660166116621663166416651666166716681669167016711672167316741675167616771678167916801681168216831684168516861687168816891690169116921693169416951696169716981699170017011702170317041705170617071708170917101711171217131714171517161717171817191720172117221723172417251726172717281729173017311732173317341735173617371738173917401741174217431744174517461747174817491750175117521753175417551756175717581759176017611762176317641765176617671768176917701771177217731774177517761777177817791780178117821783178417851786178717881789179017911792179317941795179617971798179918001801180218031804180518061807180818091810181118121813181418151816181718181819182018211822182318241825182618271828182918301831183218331834183518361837183818391840184118421843184418451846184718481849185018511852185318541855185618571858185918601861186218631864186518661867186818691870187118721873187418751876187718781879188018811882188318841885188618871888188918901891189218931894189518961897189818991900190119021903190419051906190719081909191019111912191319141915191619171918191919201921192219231924192519261927192819291930193119321933193419351936193719381939194019411942194319441945194619471948194919501951195219531954195519561957195819591960196119621963196419651966196719681969197019711972197319741975197619771978197919801981198219831984198519861987198819891990199119921993199419951996199719981999200020012002200320042005200620072008200920102011201220132014201520162017201820192020202120222023202420252026202720282029203020312032203320342035203620372038203920402041204220432044204520462047204820492050205120522053205420552056205720582059206020612062206320642065206620672068206920702071207220732074207520762077207820792080208120822083208420852086208720882089209020912092209320942095209620972098209921002101210221032104210521062107210821092110211121122113211421152116211721182119212021212122212321242125212621272128212921302131213221332134213521362137213821392140214121422143214421452146214721482149215021512152215321542155215621572158215921602161216221632164216521662167216821692170217121722173217421752176217721782179218021812182218321842185218621872188218921902191219221932194219521962197219821992200220122022203220422052206220722082209221022112212221322142215221622172218221922202221222222232224222522262227222822292230223122322233223422352236223722382239224022412242224322442245224622472248224922502251225222532254225522562257225822592260226122622263226422652266226722682269227022712272227322742275227622772278227922802281228222832284228522862287228822892290229122922293229422952296229722982299230023012302230323042305230623072308230923102311231223132314231523162317231823192320232123222323232423252326232723282329233023312332233323342335233623372338233923402341234223432344234523462347234823492350235123522353235423552356235723582359236023612362236323642365236623672368236923702371237223732374237523762377237823792380238123822383238423852386238723882389239023912392239323942395239623972398239924002401240224032404240524062407240824092410241124122413241424152416241724182419242024212422242324242425242624272428242924302431243224332434243524362437243824392440244124422443244424452446244724482449245024512452245324542455245624572458245924602461246224632464246524662467246824692470247124722473247424752476247724782479248024812482248324842485248624872488248924902491249224932494249524962497249824992500250125022503250425052506250725082509251025112512251325142515251625172518251925202521252225232524252525262527252825292530253125322533253425352536253725382539254025412542254325442545254625472548254925502551255225532554255525562557255825592560256125622563256425652566256725682569257025712572257325742575257625772578257925802581258225832584258525862587258825892590259125922593259425952596259725982599260026012602260326042605260626072608 |
- /*
- * Copyright (C) 2001-2003 Michael Niedermayer <michaelni@gmx.at>
- *
- * This file is part of FFmpeg.
- *
- * FFmpeg is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation; either
- * version 2.1 of the License, or (at your option) any later version.
- *
- * FFmpeg is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with FFmpeg; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- */
- /*
- supported Input formats: YV12, I420/IYUV, YUY2, UYVY, BGR32, BGR32_1, BGR24, BGR16, BGR15, RGB32, RGB32_1, RGB24, Y8/Y800, YVU9/IF09, PAL8
- supported output formats: YV12, I420/IYUV, YUY2, UYVY, {BGR,RGB}{1,4,8,15,16,24,32}, Y8/Y800, YVU9/IF09
- {BGR,RGB}{1,4,8,15,16} support dithering
- unscaled special converters (YV12=I420=IYUV, Y800=Y8)
- YV12 -> {BGR,RGB}{1,4,8,12,15,16,24,32}
- x -> x
- YUV9 -> YV12
- YUV9/YV12 -> Y800
- Y800 -> YUV9/YV12
- BGR24 -> BGR32 & RGB24 -> RGB32
- BGR32 -> BGR24 & RGB32 -> RGB24
- BGR15 -> BGR16
- */
- /*
- tested special converters (most are tested actually, but I did not write it down ...)
- YV12 -> BGR12/BGR16
- YV12 -> YV12
- BGR15 -> BGR16
- BGR16 -> BGR16
- YVU9 -> YV12
- untested special converters
- YV12/I420 -> BGR15/BGR24/BGR32 (it is the yuv2rgb stuff, so it should be OK)
- YV12/I420 -> YV12/I420
- YUY2/BGR15/BGR24/BGR32/RGB24/RGB32 -> same format
- BGR24 -> BGR32 & RGB24 -> RGB32
- BGR32 -> BGR24 & RGB32 -> RGB24
- BGR24 -> YV12
- */
- #include <inttypes.h>
- #include <string.h>
- #include <math.h>
- #include <stdio.h>
- #include "config.h"
- #include <assert.h>
- #include "swscale.h"
- #include "swscale_internal.h"
- #include "rgb2rgb.h"
- #include "libavutil/avassert.h"
- #include "libavutil/intreadwrite.h"
- #include "libavutil/cpu.h"
- #include "libavutil/avutil.h"
- #include "libavutil/mathematics.h"
- #include "libavutil/bswap.h"
- #include "libavutil/pixdesc.h"
- #define RGB2YUV_SHIFT 15
- #define BY ( (int)(0.114*219/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define BV (-(int)(0.081*224/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define BU ( (int)(0.500*224/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define GY ( (int)(0.587*219/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define GV (-(int)(0.419*224/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define GU (-(int)(0.331*224/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define RY ( (int)(0.299*219/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define RV ( (int)(0.500*224/255*(1<<RGB2YUV_SHIFT)+0.5))
- #define RU (-(int)(0.169*224/255*(1<<RGB2YUV_SHIFT)+0.5))
- static const double rgb2yuv_table[8][9]={
- {0.7152, 0.0722, 0.2126, -0.386, 0.5, -0.115, -0.454, -0.046, 0.5}, //ITU709
- {0.7152, 0.0722, 0.2126, -0.386, 0.5, -0.115, -0.454, -0.046, 0.5}, //ITU709
- {0.587 , 0.114 , 0.299 , -0.331, 0.5, -0.169, -0.419, -0.081, 0.5}, //DEFAULT / ITU601 / ITU624 / SMPTE 170M
- {0.587 , 0.114 , 0.299 , -0.331, 0.5, -0.169, -0.419, -0.081, 0.5}, //DEFAULT / ITU601 / ITU624 / SMPTE 170M
- {0.59 , 0.11 , 0.30 , -0.331, 0.5, -0.169, -0.421, -0.079, 0.5}, //FCC
- {0.587 , 0.114 , 0.299 , -0.331, 0.5, -0.169, -0.419, -0.081, 0.5}, //DEFAULT / ITU601 / ITU624 / SMPTE 170M
- {0.587 , 0.114 , 0.299 , -0.331, 0.5, -0.169, -0.419, -0.081, 0.5}, //DEFAULT / ITU601 / ITU624 / SMPTE 170M
- {0.701 , 0.087 , 0.212 , -0.384, 0.5, -0.116, -0.445, -0.055, 0.5}, //SMPTE 240M
- };
- /*
- NOTES
- Special versions: fast Y 1:1 scaling (no interpolation in y direction)
- TODO
- more intelligent misalignment avoidance for the horizontal scaler
- write special vertical cubic upscale version
- optimize C code (YV12 / minmax)
- add support for packed pixel YUV input & output
- add support for Y8 output
- optimize BGR24 & BGR32
- add BGR4 output support
- write special BGR->BGR scaler
- */
- DECLARE_ALIGNED(8, static const uint8_t, dither_2x2_4)[2][8]={
- { 1, 3, 1, 3, 1, 3, 1, 3, },
- { 2, 0, 2, 0, 2, 0, 2, 0, },
- };
- DECLARE_ALIGNED(8, static const uint8_t, dither_2x2_8)[2][8]={
- { 6, 2, 6, 2, 6, 2, 6, 2, },
- { 0, 4, 0, 4, 0, 4, 0, 4, },
- };
- DECLARE_ALIGNED(8, const uint8_t, dither_4x4_16)[4][8]={
- { 8, 4, 11, 7, 8, 4, 11, 7, },
- { 2, 14, 1, 13, 2, 14, 1, 13, },
- { 10, 6, 9, 5, 10, 6, 9, 5, },
- { 0, 12, 3, 15, 0, 12, 3, 15, },
- };
- DECLARE_ALIGNED(8, const uint8_t, dither_8x8_32)[8][8]={
- { 17, 9, 23, 15, 16, 8, 22, 14, },
- { 5, 29, 3, 27, 4, 28, 2, 26, },
- { 21, 13, 19, 11, 20, 12, 18, 10, },
- { 0, 24, 6, 30, 1, 25, 7, 31, },
- { 16, 8, 22, 14, 17, 9, 23, 15, },
- { 4, 28, 2, 26, 5, 29, 3, 27, },
- { 20, 12, 18, 10, 21, 13, 19, 11, },
- { 1, 25, 7, 31, 0, 24, 6, 30, },
- };
- DECLARE_ALIGNED(8, const uint8_t, dither_8x8_73)[8][8]={
- { 0, 55, 14, 68, 3, 58, 17, 72, },
- { 37, 18, 50, 32, 40, 22, 54, 35, },
- { 9, 64, 5, 59, 13, 67, 8, 63, },
- { 46, 27, 41, 23, 49, 31, 44, 26, },
- { 2, 57, 16, 71, 1, 56, 15, 70, },
- { 39, 21, 52, 34, 38, 19, 51, 33, },
- { 11, 66, 7, 62, 10, 65, 6, 60, },
- { 48, 30, 43, 25, 47, 29, 42, 24, },
- };
- #if 1
- DECLARE_ALIGNED(8, const uint8_t, dither_8x8_220)[8][8]={
- {117, 62, 158, 103, 113, 58, 155, 100, },
- { 34, 199, 21, 186, 31, 196, 17, 182, },
- {144, 89, 131, 76, 141, 86, 127, 72, },
- { 0, 165, 41, 206, 10, 175, 52, 217, },
- {110, 55, 151, 96, 120, 65, 162, 107, },
- { 28, 193, 14, 179, 38, 203, 24, 189, },
- {138, 83, 124, 69, 148, 93, 134, 79, },
- { 7, 172, 48, 213, 3, 168, 45, 210, },
- };
- #elif 1
- // tries to correct a gamma of 1.5
- DECLARE_ALIGNED(8, const uint8_t, dither_8x8_220)[8][8]={
- { 0, 143, 18, 200, 2, 156, 25, 215, },
- { 78, 28, 125, 64, 89, 36, 138, 74, },
- { 10, 180, 3, 161, 16, 195, 8, 175, },
- {109, 51, 93, 38, 121, 60, 105, 47, },
- { 1, 152, 23, 210, 0, 147, 20, 205, },
- { 85, 33, 134, 71, 81, 30, 130, 67, },
- { 14, 190, 6, 171, 12, 185, 5, 166, },
- {117, 57, 101, 44, 113, 54, 97, 41, },
- };
- #elif 1
- // tries to correct a gamma of 2.0
- DECLARE_ALIGNED(8, const uint8_t, dither_8x8_220)[8][8]={
- { 0, 124, 8, 193, 0, 140, 12, 213, },
- { 55, 14, 104, 42, 66, 19, 119, 52, },
- { 3, 168, 1, 145, 6, 187, 3, 162, },
- { 86, 31, 70, 21, 99, 39, 82, 28, },
- { 0, 134, 11, 206, 0, 129, 9, 200, },
- { 62, 17, 114, 48, 58, 16, 109, 45, },
- { 5, 181, 2, 157, 4, 175, 1, 151, },
- { 95, 36, 78, 26, 90, 34, 74, 24, },
- };
- #else
- // tries to correct a gamma of 2.5
- DECLARE_ALIGNED(8, const uint8_t, dither_8x8_220)[8][8]={
- { 0, 107, 3, 187, 0, 125, 6, 212, },
- { 39, 7, 86, 28, 49, 11, 102, 36, },
- { 1, 158, 0, 131, 3, 180, 1, 151, },
- { 68, 19, 52, 12, 81, 25, 64, 17, },
- { 0, 119, 5, 203, 0, 113, 4, 195, },
- { 45, 9, 96, 33, 42, 8, 91, 30, },
- { 2, 172, 1, 144, 2, 165, 0, 137, },
- { 77, 23, 60, 15, 72, 21, 56, 14, },
- };
- #endif
- DECLARE_ALIGNED(8, const uint8_t, dithers)[8][8][8]={
- {
- { 0, 1, 0, 1, 0, 1, 0, 1,},
- { 1, 0, 1, 0, 1, 0, 1, 0,},
- { 0, 1, 0, 1, 0, 1, 0, 1,},
- { 1, 0, 1, 0, 1, 0, 1, 0,},
- { 0, 1, 0, 1, 0, 1, 0, 1,},
- { 1, 0, 1, 0, 1, 0, 1, 0,},
- { 0, 1, 0, 1, 0, 1, 0, 1,},
- { 1, 0, 1, 0, 1, 0, 1, 0,},
- },{
- { 1, 2, 1, 2, 1, 2, 1, 2,},
- { 3, 0, 3, 0, 3, 0, 3, 0,},
- { 1, 2, 1, 2, 1, 2, 1, 2,},
- { 3, 0, 3, 0, 3, 0, 3, 0,},
- { 1, 2, 1, 2, 1, 2, 1, 2,},
- { 3, 0, 3, 0, 3, 0, 3, 0,},
- { 1, 2, 1, 2, 1, 2, 1, 2,},
- { 3, 0, 3, 0, 3, 0, 3, 0,},
- },{
- { 2, 4, 3, 5, 2, 4, 3, 5,},
- { 6, 0, 7, 1, 6, 0, 7, 1,},
- { 3, 5, 2, 4, 3, 5, 2, 4,},
- { 7, 1, 6, 0, 7, 1, 6, 0,},
- { 2, 4, 3, 5, 2, 4, 3, 5,},
- { 6, 0, 7, 1, 6, 0, 7, 1,},
- { 3, 5, 2, 4, 3, 5, 2, 4,},
- { 7, 1, 6, 0, 7, 1, 6, 0,},
- },{
- { 4, 8, 7, 11, 4, 8, 7, 11,},
- { 12, 0, 15, 3, 12, 0, 15, 3,},
- { 6, 10, 5, 9, 6, 10, 5, 9,},
- { 14, 2, 13, 1, 14, 2, 13, 1,},
- { 4, 8, 7, 11, 4, 8, 7, 11,},
- { 12, 0, 15, 3, 12, 0, 15, 3,},
- { 6, 10, 5, 9, 6, 10, 5, 9,},
- { 14, 2, 13, 1, 14, 2, 13, 1,},
- },{
- { 9, 17, 15, 23, 8, 16, 14, 22,},
- { 25, 1, 31, 7, 24, 0, 30, 6,},
- { 13, 21, 11, 19, 12, 20, 10, 18,},
- { 29, 5, 27, 3, 28, 4, 26, 2,},
- { 8, 16, 14, 22, 9, 17, 15, 23,},
- { 24, 0, 30, 6, 25, 1, 31, 7,},
- { 12, 20, 10, 18, 13, 21, 11, 19,},
- { 28, 4, 26, 2, 29, 5, 27, 3,},
- },{
- { 18, 34, 30, 46, 17, 33, 29, 45,},
- { 50, 2, 62, 14, 49, 1, 61, 13,},
- { 26, 42, 22, 38, 25, 41, 21, 37,},
- { 58, 10, 54, 6, 57, 9, 53, 5,},
- { 16, 32, 28, 44, 19, 35, 31, 47,},
- { 48, 0, 60, 12, 51, 3, 63, 15,},
- { 24, 40, 20, 36, 27, 43, 23, 39,},
- { 56, 8, 52, 4, 59, 11, 55, 7,},
- },{
- { 18, 34, 30, 46, 17, 33, 29, 45,},
- { 50, 2, 62, 14, 49, 1, 61, 13,},
- { 26, 42, 22, 38, 25, 41, 21, 37,},
- { 58, 10, 54, 6, 57, 9, 53, 5,},
- { 16, 32, 28, 44, 19, 35, 31, 47,},
- { 48, 0, 60, 12, 51, 3, 63, 15,},
- { 24, 40, 20, 36, 27, 43, 23, 39,},
- { 56, 8, 52, 4, 59, 11, 55, 7,},
- },{
- { 36, 68, 60, 92, 34, 66, 58, 90,},
- { 100, 4,124, 28, 98, 2,122, 26,},
- { 52, 84, 44, 76, 50, 82, 42, 74,},
- { 116, 20,108, 12,114, 18,106, 10,},
- { 32, 64, 56, 88, 38, 70, 62, 94,},
- { 96, 0,120, 24,102, 6,126, 30,},
- { 48, 80, 40, 72, 54, 86, 46, 78,},
- { 112, 16,104, 8,118, 22,110, 14,},
- }};
- static const uint8_t flat64[8]={64,64,64,64,64,64,64,64};
- const uint16_t dither_scale[15][16]={
- { 2, 3, 3, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,},
- { 2, 3, 7, 7, 13, 13, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,},
- { 3, 3, 4, 15, 15, 29, 57, 57, 57, 113, 113, 113, 113, 113, 113, 113,},
- { 3, 4, 4, 5, 31, 31, 61, 121, 241, 241, 241, 241, 481, 481, 481, 481,},
- { 3, 4, 5, 5, 6, 63, 63, 125, 249, 497, 993, 993, 993, 993, 993, 1985,},
- { 3, 5, 6, 6, 6, 7, 127, 127, 253, 505, 1009, 2017, 4033, 4033, 4033, 4033,},
- { 3, 5, 6, 7, 7, 7, 8, 255, 255, 509, 1017, 2033, 4065, 8129,16257,16257,},
- { 3, 5, 6, 8, 8, 8, 8, 9, 511, 511, 1021, 2041, 4081, 8161,16321,32641,},
- { 3, 5, 7, 8, 9, 9, 9, 9, 10, 1023, 1023, 2045, 4089, 8177,16353,32705,},
- { 3, 5, 7, 8, 10, 10, 10, 10, 10, 11, 2047, 2047, 4093, 8185,16369,32737,},
- { 3, 5, 7, 8, 10, 11, 11, 11, 11, 11, 12, 4095, 4095, 8189,16377,32753,},
- { 3, 5, 7, 9, 10, 12, 12, 12, 12, 12, 12, 13, 8191, 8191,16381,32761,},
- { 3, 5, 7, 9, 10, 12, 13, 13, 13, 13, 13, 13, 14,16383,16383,32765,},
- { 3, 5, 7, 9, 10, 12, 14, 14, 14, 14, 14, 14, 14, 15,32767,32767,},
- { 3, 5, 7, 9, 11, 12, 14, 15, 15, 15, 15, 15, 15, 15, 16,65535,},
- };
- static av_always_inline void
- yuv2yuvX16_c_template(const int16_t *lumFilter, const int16_t **lumSrc,
- int lumFilterSize, const int16_t *chrFilter,
- const int16_t **chrUSrc, const int16_t **chrVSrc,
- int chrFilterSize, const int16_t **alpSrc,
- uint16_t *dest, uint16_t *uDest, uint16_t *vDest,
- uint16_t *aDest, int dstW, int chrDstW,
- int big_endian, int output_bits)
- {
- //FIXME Optimize (just quickly written not optimized..)
- int i;
- int shift = 11 + 16 - output_bits;
- #define output_pixel(pos, val) \
- if (big_endian) { \
- if (output_bits == 16) { \
- AV_WB16(pos, av_clip_uint16(val >> shift)); \
- } else { \
- AV_WB16(pos, av_clip_uintp2(val >> shift, output_bits)); \
- } \
- } else { \
- if (output_bits == 16) { \
- AV_WL16(pos, av_clip_uint16(val >> shift)); \
- } else { \
- AV_WL16(pos, av_clip_uintp2(val >> shift, output_bits)); \
- } \
- }
- for (i = 0; i < dstW; i++) {
- int val = 1 << (26-output_bits);
- int j;
- for (j = 0; j < lumFilterSize; j++)
- val += lumSrc[j][i] * lumFilter[j];
- output_pixel(&dest[i], val);
- }
- if (uDest) {
- for (i = 0; i < chrDstW; i++) {
- int u = 1 << (26-output_bits);
- int v = 1 << (26-output_bits);
- int j;
- for (j = 0; j < chrFilterSize; j++) {
- u += chrUSrc[j][i] * chrFilter[j];
- v += chrVSrc[j][i] * chrFilter[j];
- }
- output_pixel(&uDest[i], u);
- output_pixel(&vDest[i], v);
- }
- }
- if (CONFIG_SWSCALE_ALPHA && aDest) {
- for (i = 0; i < dstW; i++) {
- int val = 1 << (26-output_bits);
- int j;
- for (j = 0; j < lumFilterSize; j++)
- val += alpSrc[j][i] * lumFilter[j];
- output_pixel(&aDest[i], val);
- }
- }
- #undef output_pixel
- }
- #define yuv2NBPS(bits, BE_LE, is_be) \
- static void yuv2yuvX ## bits ## BE_LE ## _c(SwsContext *c, const int16_t *lumFilter, \
- const int16_t **lumSrc, int lumFilterSize, \
- const int16_t *chrFilter, const int16_t **chrUSrc, \
- const int16_t **chrVSrc, \
- int chrFilterSize, const int16_t **alpSrc, \
- uint8_t *_dest, uint8_t *_uDest, uint8_t *_vDest, \
- uint8_t *_aDest, int dstW, int chrDstW) \
- { \
- uint16_t *dest = (uint16_t *) _dest, *uDest = (uint16_t *) _uDest, \
- *vDest = (uint16_t *) _vDest, *aDest = (uint16_t *) _aDest; \
- yuv2yuvX16_c_template(lumFilter, lumSrc, lumFilterSize, \
- chrFilter, chrUSrc, chrVSrc, chrFilterSize, \
- alpSrc, \
- dest, uDest, vDest, aDest, \
- dstW, chrDstW, is_be, bits); \
- }
- yuv2NBPS( 9, BE, 1);
- yuv2NBPS( 9, LE, 0);
- yuv2NBPS(10, BE, 1);
- yuv2NBPS(10, LE, 0);
- yuv2NBPS(16, BE, 1);
- yuv2NBPS(16, LE, 0);
- static void yuv2yuvX_c(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc,
- int chrFilterSize, const int16_t **alpSrc,
- uint8_t *dest, uint8_t *uDest, uint8_t *vDest,
- uint8_t *aDest, int dstW, int chrDstW,
- const uint8_t *lumDither, const uint8_t *chrDither)
- {
- //FIXME Optimize (just quickly written not optimized..)
- int i;
- for (i=0; i<dstW; i++) {
- int val = lumDither[i&7] << 12;
- int j;
- for (j=0; j<lumFilterSize; j++)
- val += lumSrc[j][i] * lumFilter[j];
- dest[i]= av_clip_uint8(val>>19);
- }
- if (uDest)
- for (i=0; i<chrDstW; i++) {
- int u = chrDither[i&7] << 12;
- int v = chrDither[(i+3)&7] << 12;
- int j;
- for (j=0; j<chrFilterSize; j++) {
- u += chrUSrc[j][i] * chrFilter[j];
- v += chrVSrc[j][i] * chrFilter[j];
- }
- uDest[i]= av_clip_uint8(u>>19);
- vDest[i]= av_clip_uint8(v>>19);
- }
- if (CONFIG_SWSCALE_ALPHA && aDest)
- for (i=0; i<dstW; i++) {
- int val = lumDither[i&7] << 12;
- int j;
- for (j=0; j<lumFilterSize; j++)
- val += alpSrc[j][i] * lumFilter[j];
- aDest[i]= av_clip_uint8(val>>19);
- }
- }
- static void yuv2yuv1_c(SwsContext *c, const int16_t *lumSrc,
- const int16_t *chrUSrc, const int16_t *chrVSrc,
- const int16_t *alpSrc,
- uint8_t *dest, uint8_t *uDest, uint8_t *vDest,
- uint8_t *aDest, int dstW, int chrDstW, const uint8_t *lumDither, const uint8_t *chrDither)
- {
- int i;
- for (i=0; i<dstW; i++) {
- int val= (lumSrc[i]+lumDither[i&7])>>7;
- dest[i]= av_clip_uint8(val);
- }
- if (uDest)
- for (i=0; i<chrDstW; i++) {
- int u=(chrUSrc[i]+chrDither[i&7])>>7;
- int v=(chrVSrc[i]+chrDither[(i+3)&7])>>7;
- uDest[i]= av_clip_uint8(u);
- vDest[i]= av_clip_uint8(v);
- }
- if (CONFIG_SWSCALE_ALPHA && aDest)
- for (i=0; i<dstW; i++) {
- int val= (alpSrc[i]+lumDither[i&7])>>7;
- aDest[i]= av_clip_uint8(val);
- }
- }
- static void yuv2nv12X_c(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, uint8_t *uDest,
- uint8_t *vDest, uint8_t *aDest,
- int dstW, int chrDstW,
- const uint8_t *lumDither, const uint8_t *chrDither)
- {
- enum PixelFormat dstFormat = c->dstFormat;
- //FIXME Optimize (just quickly written not optimized..)
- int i;
- for (i=0; i<dstW; i++) {
- int val = lumDither[i&7]<<12;
- int j;
- for (j=0; j<lumFilterSize; j++)
- val += lumSrc[j][i] * lumFilter[j];
- dest[i]= av_clip_uint8(val>>19);
- }
- if (!uDest)
- return;
- if (dstFormat == PIX_FMT_NV12)
- for (i=0; i<chrDstW; i++) {
- int u = chrDither[i&7]<<12;
- int v = chrDither[(i+3)&7]<<12;
- int j;
- for (j=0; j<chrFilterSize; j++) {
- u += chrUSrc[j][i] * chrFilter[j];
- v += chrVSrc[j][i] * chrFilter[j];
- }
- uDest[2*i]= av_clip_uint8(u>>19);
- uDest[2*i+1]= av_clip_uint8(v>>19);
- }
- else
- for (i=0; i<chrDstW; i++) {
- int u = chrDither[i&7]<<12;
- int v = chrDither[(i+3)&7]<<12;
- int j;
- for (j=0; j<chrFilterSize; j++) {
- u += chrUSrc[j][i] * chrFilter[j];
- v += chrVSrc[j][i] * chrFilter[j];
- }
- uDest[2*i]= av_clip_uint8(v>>19);
- uDest[2*i+1]= av_clip_uint8(u>>19);
- }
- }
- #define output_pixel(pos, val) \
- if (target == PIX_FMT_GRAY16BE) { \
- AV_WB16(pos, val); \
- } else { \
- AV_WL16(pos, val); \
- }
- static av_always_inline void
- yuv2gray16_X_c_template(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, int dstW,
- int y, enum PixelFormat target)
- {
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- int j;
- int Y1 = 1 << 18;
- int Y2 = 1 << 18;
- const int i2 = 2 * i;
- for (j = 0; j < lumFilterSize; j++) {
- Y1 += lumSrc[j][i2] * lumFilter[j];
- Y2 += lumSrc[j][i2+1] * lumFilter[j];
- }
- Y1 >>= 11;
- Y2 >>= 11;
- if ((Y1 | Y2) & 0x10000) {
- Y1 = av_clip_uint16(Y1);
- Y2 = av_clip_uint16(Y2);
- }
- output_pixel(&dest[2 * i2 + 0], Y1);
- output_pixel(&dest[2 * i2 + 2], Y2);
- }
- }
- static av_always_inline void
- yuv2gray16_2_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *buf1, const uint16_t *ubuf0,
- const uint16_t *ubuf1, const uint16_t *vbuf0,
- const uint16_t *vbuf1, const uint16_t *abuf0,
- const uint16_t *abuf1, uint8_t *dest, int dstW,
- int yalpha, int uvalpha, int y,
- enum PixelFormat target)
- {
- int yalpha1 = 4095 - yalpha; \
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- const int i2 = 2 * i;
- int Y1 = (buf0[i2 ] * yalpha1 + buf1[i2 ] * yalpha) >> 11;
- int Y2 = (buf0[i2+1] * yalpha1 + buf1[i2+1] * yalpha) >> 11;
- output_pixel(&dest[2 * i2 + 0], Y1);
- output_pixel(&dest[2 * i2 + 2], Y2);
- }
- }
- static av_always_inline void
- yuv2gray16_1_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *ubuf0, const uint16_t *ubuf1,
- const uint16_t *vbuf0, const uint16_t *vbuf1,
- const uint16_t *abuf0, uint8_t *dest, int dstW,
- int uvalpha, enum PixelFormat dstFormat,
- int flags, int y, enum PixelFormat target)
- {
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- const int i2 = 2 * i;
- int Y1 = buf0[i2 ] << 1;
- int Y2 = buf0[i2+1] << 1;
- output_pixel(&dest[2 * i2 + 0], Y1);
- output_pixel(&dest[2 * i2 + 2], Y2);
- }
- }
- #undef output_pixel
- #define YUV2PACKEDWRAPPER(name, base, ext, fmt) \
- static void name ## ext ## _X_c(SwsContext *c, const int16_t *lumFilter, \
- const int16_t **lumSrc, int lumFilterSize, \
- const int16_t *chrFilter, const int16_t **chrUSrc, \
- const int16_t **chrVSrc, int chrFilterSize, \
- const int16_t **alpSrc, uint8_t *dest, int dstW, \
- int y) \
- { \
- name ## base ## _X_c_template(c, lumFilter, lumSrc, lumFilterSize, \
- chrFilter, chrUSrc, chrVSrc, chrFilterSize, \
- alpSrc, dest, dstW, y, fmt); \
- } \
- \
- static void name ## ext ## _2_c(SwsContext *c, const uint16_t *buf0, \
- const uint16_t *buf1, const uint16_t *ubuf0, \
- const uint16_t *ubuf1, const uint16_t *vbuf0, \
- const uint16_t *vbuf1, const uint16_t *abuf0, \
- const uint16_t *abuf1, uint8_t *dest, int dstW, \
- int yalpha, int uvalpha, int y) \
- { \
- name ## base ## _2_c_template(c, buf0, buf1, ubuf0, ubuf1, \
- vbuf0, vbuf1, abuf0, abuf1, \
- dest, dstW, yalpha, uvalpha, y, fmt); \
- } \
- \
- static void name ## ext ## _1_c(SwsContext *c, const uint16_t *buf0, \
- const uint16_t *ubuf0, const uint16_t *ubuf1, \
- const uint16_t *vbuf0, const uint16_t *vbuf1, \
- const uint16_t *abuf0, uint8_t *dest, int dstW, \
- int uvalpha, enum PixelFormat dstFormat, \
- int flags, int y) \
- { \
- name ## base ## _1_c_template(c, buf0, ubuf0, ubuf1, vbuf0, \
- vbuf1, abuf0, dest, dstW, uvalpha, \
- dstFormat, flags, y, fmt); \
- }
- YUV2PACKEDWRAPPER(yuv2gray16,, LE, PIX_FMT_GRAY16LE);
- YUV2PACKEDWRAPPER(yuv2gray16,, BE, PIX_FMT_GRAY16BE);
- #define output_pixel(pos, acc) \
- if (target == PIX_FMT_MONOBLACK) { \
- pos = acc; \
- } else { \
- pos = ~acc; \
- }
- static av_always_inline void
- yuv2mono_X_c_template(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, int dstW,
- int y, enum PixelFormat target)
- {
- const uint8_t * const d128=dither_8x8_220[y&7];
- uint8_t *g = c->table_gU[128] + c->table_gV[128];
- int i;
- int acc = 0;
- for (i = 0; i < dstW - 1; i += 2) {
- int j;
- int Y1 = 1 << 18;
- int Y2 = 1 << 18;
- for (j = 0; j < lumFilterSize; j++) {
- Y1 += lumSrc[j][i] * lumFilter[j];
- Y2 += lumSrc[j][i+1] * lumFilter[j];
- }
- Y1 >>= 19;
- Y2 >>= 19;
- if ((Y1 | Y2) & 0x100) {
- Y1 = av_clip_uint8(Y1);
- Y2 = av_clip_uint8(Y2);
- }
- acc += acc + g[Y1 + d128[(i + 0) & 7]];
- acc += acc + g[Y2 + d128[(i + 1) & 7]];
- if ((i & 7) == 6) {
- output_pixel(*dest++, acc);
- }
- }
- }
- static av_always_inline void
- yuv2mono_2_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *buf1, const uint16_t *ubuf0,
- const uint16_t *ubuf1, const uint16_t *vbuf0,
- const uint16_t *vbuf1, const uint16_t *abuf0,
- const uint16_t *abuf1, uint8_t *dest, int dstW,
- int yalpha, int uvalpha, int y,
- enum PixelFormat target)
- {
- const uint8_t * const d128 = dither_8x8_220[y & 7];
- uint8_t *g = c->table_gU[128] + c->table_gV[128];
- int yalpha1 = 4095 - yalpha;
- int i;
- for (i = 0; i < dstW - 7; i += 8) {
- int acc = g[((buf0[i ] * yalpha1 + buf1[i ] * yalpha) >> 19) + d128[0]];
- acc += acc + g[((buf0[i + 1] * yalpha1 + buf1[i + 1] * yalpha) >> 19) + d128[1]];
- acc += acc + g[((buf0[i + 2] * yalpha1 + buf1[i + 2] * yalpha) >> 19) + d128[2]];
- acc += acc + g[((buf0[i + 3] * yalpha1 + buf1[i + 3] * yalpha) >> 19) + d128[3]];
- acc += acc + g[((buf0[i + 4] * yalpha1 + buf1[i + 4] * yalpha) >> 19) + d128[4]];
- acc += acc + g[((buf0[i + 5] * yalpha1 + buf1[i + 5] * yalpha) >> 19) + d128[5]];
- acc += acc + g[((buf0[i + 6] * yalpha1 + buf1[i + 6] * yalpha) >> 19) + d128[6]];
- acc += acc + g[((buf0[i + 7] * yalpha1 + buf1[i + 7] * yalpha) >> 19) + d128[7]];
- output_pixel(*dest++, acc);
- }
- }
- static av_always_inline void
- yuv2mono_1_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *ubuf0, const uint16_t *ubuf1,
- const uint16_t *vbuf0, const uint16_t *vbuf1,
- const uint16_t *abuf0, uint8_t *dest, int dstW,
- int uvalpha, enum PixelFormat dstFormat,
- int flags, int y, enum PixelFormat target)
- {
- const uint8_t * const d128 = dither_8x8_220[y & 7];
- uint8_t *g = c->table_gU[128] + c->table_gV[128];
- int i;
- for (i = 0; i < dstW - 7; i += 8) {
- int acc = g[(buf0[i ] >> 7) + d128[0]];
- acc += acc + g[(buf0[i + 1] >> 7) + d128[1]];
- acc += acc + g[(buf0[i + 2] >> 7) + d128[2]];
- acc += acc + g[(buf0[i + 3] >> 7) + d128[3]];
- acc += acc + g[(buf0[i + 4] >> 7) + d128[4]];
- acc += acc + g[(buf0[i + 5] >> 7) + d128[5]];
- acc += acc + g[(buf0[i + 6] >> 7) + d128[6]];
- acc += acc + g[(buf0[i + 7] >> 7) + d128[7]];
- output_pixel(*dest++, acc);
- }
- }
- #undef output_pixel
- YUV2PACKEDWRAPPER(yuv2mono,, white, PIX_FMT_MONOWHITE);
- YUV2PACKEDWRAPPER(yuv2mono,, black, PIX_FMT_MONOBLACK);
- #define output_pixels(pos, Y1, U, Y2, V) \
- if (target == PIX_FMT_YUYV422) { \
- dest[pos + 0] = Y1; \
- dest[pos + 1] = U; \
- dest[pos + 2] = Y2; \
- dest[pos + 3] = V; \
- } else { \
- dest[pos + 0] = U; \
- dest[pos + 1] = Y1; \
- dest[pos + 2] = V; \
- dest[pos + 3] = Y2; \
- }
- static av_always_inline void
- yuv2422_X_c_template(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, int dstW,
- int y, enum PixelFormat target)
- {
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- int j;
- int Y1 = 1 << 18;
- int Y2 = 1 << 18;
- int U = 1 << 18;
- int V = 1 << 18;
- for (j = 0; j < lumFilterSize; j++) {
- Y1 += lumSrc[j][i * 2] * lumFilter[j];
- Y2 += lumSrc[j][i * 2 + 1] * lumFilter[j];
- }
- for (j = 0; j < chrFilterSize; j++) {
- U += chrUSrc[j][i] * chrFilter[j];
- V += chrVSrc[j][i] * chrFilter[j];
- }
- Y1 >>= 19;
- Y2 >>= 19;
- U >>= 19;
- V >>= 19;
- if ((Y1 | Y2 | U | V) & 0x100) {
- Y1 = av_clip_uint8(Y1);
- Y2 = av_clip_uint8(Y2);
- U = av_clip_uint8(U);
- V = av_clip_uint8(V);
- }
- output_pixels(4*i, Y1, U, Y2, V);
- }
- }
- static av_always_inline void
- yuv2422_2_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *buf1, const uint16_t *ubuf0,
- const uint16_t *ubuf1, const uint16_t *vbuf0,
- const uint16_t *vbuf1, const uint16_t *abuf0,
- const uint16_t *abuf1, uint8_t *dest, int dstW,
- int yalpha, int uvalpha, int y,
- enum PixelFormat target)
- {
- int yalpha1 = 4095 - yalpha;
- int uvalpha1 = 4095 - uvalpha;
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = (buf0[i * 2] * yalpha1 + buf1[i * 2] * yalpha) >> 19;
- int Y2 = (buf0[i * 2 + 1] * yalpha1 + buf1[i * 2 + 1] * yalpha) >> 19;
- int U = (ubuf0[i] * uvalpha1 + ubuf1[i] * uvalpha) >> 19;
- int V = (vbuf0[i] * uvalpha1 + vbuf1[i] * uvalpha) >> 19;
- output_pixels(i * 4, Y1, U, Y2, V);
- }
- }
- static av_always_inline void
- yuv2422_1_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *ubuf0, const uint16_t *ubuf1,
- const uint16_t *vbuf0, const uint16_t *vbuf1,
- const uint16_t *abuf0, uint8_t *dest, int dstW,
- int uvalpha, enum PixelFormat dstFormat,
- int flags, int y, enum PixelFormat target)
- {
- int i;
- if (uvalpha < 2048) {
- for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = buf0[i * 2] >> 7;
- int Y2 = buf0[i * 2 + 1] >> 7;
- int U = ubuf1[i] >> 7;
- int V = vbuf1[i] >> 7;
- output_pixels(i * 4, Y1, U, Y2, V);
- }
- } else {
- for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = buf0[i * 2] >> 7;
- int Y2 = buf0[i * 2 + 1] >> 7;
- int U = (ubuf0[i] + ubuf1[i]) >> 8;
- int V = (vbuf0[i] + vbuf1[i]) >> 8;
- output_pixels(i * 4, Y1, U, Y2, V);
- }
- }
- }
- #undef output_pixels
- YUV2PACKEDWRAPPER(yuv2, 422, yuyv422, PIX_FMT_YUYV422);
- YUV2PACKEDWRAPPER(yuv2, 422, uyvy422, PIX_FMT_UYVY422);
- #define r_b ((target == PIX_FMT_RGB48LE || target == PIX_FMT_RGB48BE) ? r : b)
- #define b_r ((target == PIX_FMT_RGB48LE || target == PIX_FMT_RGB48BE) ? b : r)
- static av_always_inline void
- yuv2rgb48_X_c_template(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, int dstW,
- int y, enum PixelFormat target)
- {
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- int j;
- int Y1 = 1 << 18;
- int Y2 = 1 << 18;
- int U = 1 << 18;
- int V = 1 << 18;
- const uint8_t *r, *g, *b;
- for (j = 0; j < lumFilterSize; j++) {
- Y1 += lumSrc[j][i * 2] * lumFilter[j];
- Y2 += lumSrc[j][i * 2 + 1] * lumFilter[j];
- }
- for (j = 0; j < chrFilterSize; j++) {
- U += chrUSrc[j][i] * chrFilter[j];
- V += chrVSrc[j][i] * chrFilter[j];
- }
- Y1 >>= 19;
- Y2 >>= 19;
- U >>= 19;
- V >>= 19;
- if ((Y1 | Y2 | U | V) & 0x100) {
- Y1 = av_clip_uint8(Y1);
- Y2 = av_clip_uint8(Y2);
- U = av_clip_uint8(U);
- V = av_clip_uint8(V);
- }
- /* FIXME fix tables so that clipping is not needed and then use _NOCLIP*/
- r = (const uint8_t *) c->table_rV[V];
- g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]);
- b = (const uint8_t *) c->table_bU[U];
- dest[ 0] = dest[ 1] = r_b[Y1];
- dest[ 2] = dest[ 3] = g[Y1];
- dest[ 4] = dest[ 5] = b_r[Y1];
- dest[ 6] = dest[ 7] = r_b[Y2];
- dest[ 8] = dest[ 9] = g[Y2];
- dest[10] = dest[11] = b_r[Y2];
- dest += 12;
- }
- }
- static av_always_inline void
- yuv2rgb48_2_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *buf1, const uint16_t *ubuf0,
- const uint16_t *ubuf1, const uint16_t *vbuf0,
- const uint16_t *vbuf1, const uint16_t *abuf0,
- const uint16_t *abuf1, uint8_t *dest, int dstW,
- int yalpha, int uvalpha, int y,
- enum PixelFormat target)
- {
- int yalpha1 = 4095 - yalpha;
- int uvalpha1 = 4095 - uvalpha;
- int i;
- for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = (buf0[i * 2] * yalpha1 + buf1[i * 2] * yalpha) >> 19;
- int Y2 = (buf0[i * 2 + 1] * yalpha1 + buf1[i * 2 + 1] * yalpha) >> 19;
- int U = (ubuf0[i] * uvalpha1 + ubuf1[i] * uvalpha) >> 19;
- int V = (vbuf0[i] * uvalpha1 + vbuf1[i] * uvalpha) >> 19;
- const uint8_t *r = (const uint8_t *) c->table_rV[V],
- *g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]),
- *b = (const uint8_t *) c->table_bU[U];
- dest[ 0] = dest[ 1] = r_b[Y1];
- dest[ 2] = dest[ 3] = g[Y1];
- dest[ 4] = dest[ 5] = b_r[Y1];
- dest[ 6] = dest[ 7] = r_b[Y2];
- dest[ 8] = dest[ 9] = g[Y2];
- dest[10] = dest[11] = b_r[Y2];
- dest += 12;
- }
- }
- static av_always_inline void
- yuv2rgb48_1_c_template(SwsContext *c, const uint16_t *buf0,
- const uint16_t *ubuf0, const uint16_t *ubuf1,
- const uint16_t *vbuf0, const uint16_t *vbuf1,
- const uint16_t *abuf0, uint8_t *dest, int dstW,
- int uvalpha, enum PixelFormat dstFormat,
- int flags, int y, enum PixelFormat target)
- {
- int i;
- if (uvalpha < 2048) {
- for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = buf0[i * 2] >> 7;
- int Y2 = buf0[i * 2 + 1] >> 7;
- int U = ubuf1[i] >> 7;
- int V = vbuf1[i] >> 7;
- const uint8_t *r = (const uint8_t *) c->table_rV[V],
- *g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]),
- *b = (const uint8_t *) c->table_bU[U];
- dest[ 0] = dest[ 1] = r_b[Y1];
- dest[ 2] = dest[ 3] = g[Y1];
- dest[ 4] = dest[ 5] = b_r[Y1];
- dest[ 6] = dest[ 7] = r_b[Y2];
- dest[ 8] = dest[ 9] = g[Y2];
- dest[10] = dest[11] = b_r[Y2];
- dest += 12;
- }
- } else {
- for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = buf0[i * 2] >> 7;
- int Y2 = buf0[i * 2 + 1] >> 7;
- int U = (ubuf0[i] + ubuf1[i]) >> 8;
- int V = (vbuf0[i] + vbuf1[i]) >> 8;
- const uint8_t *r = (const uint8_t *) c->table_rV[V],
- *g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]),
- *b = (const uint8_t *) c->table_bU[U];
- dest[ 0] = dest[ 1] = r_b[Y1];
- dest[ 2] = dest[ 3] = g[Y1];
- dest[ 4] = dest[ 5] = b_r[Y1];
- dest[ 6] = dest[ 7] = r_b[Y2];
- dest[ 8] = dest[ 9] = g[Y2];
- dest[10] = dest[11] = b_r[Y2];
- dest += 12;
- }
- }
- }
- #undef r_b
- #undef b_r
- YUV2PACKEDWRAPPER(yuv2, rgb48, rgb48be, PIX_FMT_RGB48BE);
- //YUV2PACKEDWRAPPER(yuv2, rgb48, rgb48le, PIX_FMT_RGB48LE);
- YUV2PACKEDWRAPPER(yuv2, rgb48, bgr48be, PIX_FMT_BGR48BE);
- //YUV2PACKEDWRAPPER(yuv2, rgb48, bgr48le, PIX_FMT_BGR48LE);
- #define YSCALE_YUV_2_RGBX_C(type,alpha) \
- for (i=0; i<(dstW>>1); i++) {\
- int j;\
- int Y1 = 1<<18;\
- int Y2 = 1<<18;\
- int U = 1<<18;\
- int V = 1<<18;\
- int av_unused A1, A2;\
- type av_unused *r, *b, *g;\
- const int i2= 2*i;\
- \
- for (j=0; j<lumFilterSize; j++) {\
- Y1 += lumSrc[j][i2] * lumFilter[j];\
- Y2 += lumSrc[j][i2+1] * lumFilter[j];\
- }\
- for (j=0; j<chrFilterSize; j++) {\
- U += chrUSrc[j][i] * chrFilter[j];\
- V += chrVSrc[j][i] * chrFilter[j];\
- }\
- Y1>>=19;\
- Y2>>=19;\
- U >>=19;\
- V >>=19;\
- if ((Y1|Y2|U|V)&0x100) {\
- Y1 = av_clip_uint8(Y1); \
- Y2 = av_clip_uint8(Y2); \
- U = av_clip_uint8(U); \
- V = av_clip_uint8(V); \
- }\
- if (alpha) {\
- A1 = 1<<18;\
- A2 = 1<<18;\
- for (j=0; j<lumFilterSize; j++) {\
- A1 += alpSrc[j][i2 ] * lumFilter[j];\
- A2 += alpSrc[j][i2+1] * lumFilter[j];\
- }\
- A1>>=19;\
- A2>>=19;\
- if ((A1|A2)&0x100) {\
- A1 = av_clip_uint8(A1); \
- A2 = av_clip_uint8(A2); \
- }\
- }\
- /* FIXME fix tables so that clipping is not needed and then use _NOCLIP*/\
- r = (type *)c->table_rV[V]; \
- g = (type *)(c->table_gU[U] + c->table_gV[V]); \
- b = (type *)c->table_bU[U];
- #define YSCALE_YUV_2_RGBX_FULL_C(rnd,alpha) \
- for (i=0; i<dstW; i++) {\
- int j;\
- int Y = 1<<9;\
- int U = (1<<9)-(128<<19);\
- int V = (1<<9)-(128<<19);\
- int av_unused A;\
- int R,G,B;\
- \
- for (j=0; j<lumFilterSize; j++) {\
- Y += lumSrc[j][i ] * lumFilter[j];\
- }\
- for (j=0; j<chrFilterSize; j++) {\
- U += chrUSrc[j][i] * chrFilter[j];\
- V += chrVSrc[j][i] * chrFilter[j];\
- }\
- Y >>=10;\
- U >>=10;\
- V >>=10;\
- if (alpha) {\
- A = rnd>>3;\
- for (j=0; j<lumFilterSize; j++)\
- A += alpSrc[j][i ] * lumFilter[j];\
- A >>=19;\
- if (A&0x100)\
- A = av_clip_uint8(A);\
- }\
- Y-= c->yuv2rgb_y_offset;\
- Y*= c->yuv2rgb_y_coeff;\
- Y+= rnd;\
- R= Y + V*c->yuv2rgb_v2r_coeff;\
- G= Y + V*c->yuv2rgb_v2g_coeff + U*c->yuv2rgb_u2g_coeff;\
- B= Y + U*c->yuv2rgb_u2b_coeff;\
- if ((R|G|B)&(0xC0000000)) {\
- R = av_clip_uintp2(R, 30); \
- G = av_clip_uintp2(G, 30); \
- B = av_clip_uintp2(B, 30); \
- }
- #define YSCALE_YUV_2_RGB2_C(type,alpha) \
- for (i=0; i<(dstW>>1); i++) { \
- const int i2= 2*i; \
- int Y1= (buf0[i2 ]*yalpha1+buf1[i2 ]*yalpha)>>19; \
- int Y2= (buf0[i2+1]*yalpha1+buf1[i2+1]*yalpha)>>19; \
- int U= (ubuf0[i]*uvalpha1+ubuf1[i]*uvalpha)>>19; \
- int V= (vbuf0[i]*uvalpha1+vbuf1[i]*uvalpha)>>19; \
- type av_unused *r, *b, *g; \
- int av_unused A1, A2; \
- if (alpha) {\
- A1= (abuf0[i2 ]*yalpha1+abuf1[i2 ]*yalpha)>>19; \
- A2= (abuf0[i2+1]*yalpha1+abuf1[i2+1]*yalpha)>>19; \
- }\
- r = (type *)c->table_rV[V];\
- g = (type *)(c->table_gU[U] + c->table_gV[V]);\
- b = (type *)c->table_bU[U];
- #define YSCALE_YUV_2_RGB1_C(type,alpha) \
- for (i=0; i<(dstW>>1); i++) {\
- const int i2= 2*i;\
- int Y1= buf0[i2 ]>>7;\
- int Y2= buf0[i2+1]>>7;\
- int U= (ubuf1[i])>>7;\
- int V= (vbuf1[i])>>7;\
- type av_unused *r, *b, *g;\
- int av_unused A1, A2;\
- if (alpha) {\
- A1= abuf0[i2 ]>>7;\
- A2= abuf0[i2+1]>>7;\
- }\
- r = (type *)c->table_rV[V];\
- g = (type *)(c->table_gU[U] + c->table_gV[V]);\
- b = (type *)c->table_bU[U];
- #define YSCALE_YUV_2_RGB1B_C(type,alpha) \
- for (i=0; i<(dstW>>1); i++) {\
- const int i2= 2*i;\
- int Y1= buf0[i2 ]>>7;\
- int Y2= buf0[i2+1]>>7;\
- int U= (ubuf0[i] + ubuf1[i])>>8;\
- int V= (vbuf0[i] + vbuf1[i])>>8;\
- type av_unused *r, *b, *g;\
- int av_unused A1, A2;\
- if (alpha) {\
- A1= abuf0[i2 ]>>7;\
- A2= abuf0[i2+1]>>7;\
- }\
- r = (type *)c->table_rV[V];\
- g = (type *)(c->table_gU[U] + c->table_gV[V]);\
- b = (type *)c->table_bU[U];
- #define YSCALE_YUV_2_ANYRGB_C(func)\
- switch(c->dstFormat) {\
- case PIX_FMT_RGBA:\
- case PIX_FMT_BGRA:\
- if (CONFIG_SMALL) {\
- int needAlpha = CONFIG_SWSCALE_ALPHA && c->alpPixBuf;\
- func(uint32_t,needAlpha)\
- ((uint32_t*)dest)[i2+0]= r[Y1] + g[Y1] + b[Y1] + (needAlpha ? (A1<<24) : 0);\
- ((uint32_t*)dest)[i2+1]= r[Y2] + g[Y2] + b[Y2] + (needAlpha ? (A2<<24) : 0);\
- }\
- } else {\
- if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf) {\
- func(uint32_t,1)\
- ((uint32_t*)dest)[i2+0]= r[Y1] + g[Y1] + b[Y1] + (A1<<24);\
- ((uint32_t*)dest)[i2+1]= r[Y2] + g[Y2] + b[Y2] + (A2<<24);\
- }\
- } else {\
- func(uint32_t,0)\
- ((uint32_t*)dest)[i2+0]= r[Y1] + g[Y1] + b[Y1];\
- ((uint32_t*)dest)[i2+1]= r[Y2] + g[Y2] + b[Y2];\
- }\
- }\
- }\
- break;\
- case PIX_FMT_ARGB:\
- case PIX_FMT_ABGR:\
- if (CONFIG_SMALL) {\
- int needAlpha = CONFIG_SWSCALE_ALPHA && c->alpPixBuf;\
- func(uint32_t,needAlpha)\
- ((uint32_t*)dest)[i2+0]= r[Y1] + g[Y1] + b[Y1] + (needAlpha ? A1 : 0);\
- ((uint32_t*)dest)[i2+1]= r[Y2] + g[Y2] + b[Y2] + (needAlpha ? A2 : 0);\
- }\
- } else {\
- if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf) {\
- func(uint32_t,1)\
- ((uint32_t*)dest)[i2+0]= r[Y1] + g[Y1] + b[Y1] + A1;\
- ((uint32_t*)dest)[i2+1]= r[Y2] + g[Y2] + b[Y2] + A2;\
- }\
- } else {\
- func(uint32_t,0)\
- ((uint32_t*)dest)[i2+0]= r[Y1] + g[Y1] + b[Y1];\
- ((uint32_t*)dest)[i2+1]= r[Y2] + g[Y2] + b[Y2];\
- }\
- }\
- } \
- break;\
- case PIX_FMT_RGB24:\
- func(uint8_t,0)\
- ((uint8_t*)dest)[0]= r[Y1];\
- ((uint8_t*)dest)[1]= g[Y1];\
- ((uint8_t*)dest)[2]= b[Y1];\
- ((uint8_t*)dest)[3]= r[Y2];\
- ((uint8_t*)dest)[4]= g[Y2];\
- ((uint8_t*)dest)[5]= b[Y2];\
- dest+=6;\
- }\
- break;\
- case PIX_FMT_BGR24:\
- func(uint8_t,0)\
- ((uint8_t*)dest)[0]= b[Y1];\
- ((uint8_t*)dest)[1]= g[Y1];\
- ((uint8_t*)dest)[2]= r[Y1];\
- ((uint8_t*)dest)[3]= b[Y2];\
- ((uint8_t*)dest)[4]= g[Y2];\
- ((uint8_t*)dest)[5]= r[Y2];\
- dest+=6;\
- }\
- break;\
- case PIX_FMT_RGB565:\
- case PIX_FMT_BGR565:\
- {\
- const int dr1= dither_2x2_8[y&1 ][0];\
- const int dg1= dither_2x2_4[y&1 ][0];\
- const int db1= dither_2x2_8[(y&1)^1][0];\
- const int dr2= dither_2x2_8[y&1 ][1];\
- const int dg2= dither_2x2_4[y&1 ][1];\
- const int db2= dither_2x2_8[(y&1)^1][1];\
- func(uint16_t,0)\
- ((uint16_t*)dest)[i2+0]= r[Y1+dr1] + g[Y1+dg1] + b[Y1+db1];\
- ((uint16_t*)dest)[i2+1]= r[Y2+dr2] + g[Y2+dg2] + b[Y2+db2];\
- }\
- }\
- break;\
- case PIX_FMT_RGB555:\
- case PIX_FMT_BGR555:\
- {\
- const int dr1= dither_2x2_8[y&1 ][0];\
- const int dg1= dither_2x2_8[y&1 ][1];\
- const int db1= dither_2x2_8[(y&1)^1][0];\
- const int dr2= dither_2x2_8[y&1 ][1];\
- const int dg2= dither_2x2_8[y&1 ][0];\
- const int db2= dither_2x2_8[(y&1)^1][1];\
- func(uint16_t,0)\
- ((uint16_t*)dest)[i2+0]= r[Y1+dr1] + g[Y1+dg1] + b[Y1+db1];\
- ((uint16_t*)dest)[i2+1]= r[Y2+dr2] + g[Y2+dg2] + b[Y2+db2];\
- }\
- }\
- break;\
- case PIX_FMT_RGB444:\
- case PIX_FMT_BGR444:\
- {\
- const int dr1= dither_4x4_16[y&3 ][0];\
- const int dg1= dither_4x4_16[y&3 ][1];\
- const int db1= dither_4x4_16[(y&3)^3][0];\
- const int dr2= dither_4x4_16[y&3 ][1];\
- const int dg2= dither_4x4_16[y&3 ][0];\
- const int db2= dither_4x4_16[(y&3)^3][1];\
- func(uint16_t,0)\
- ((uint16_t*)dest)[i2+0]= r[Y1+dr1] + g[Y1+dg1] + b[Y1+db1];\
- ((uint16_t*)dest)[i2+1]= r[Y2+dr2] + g[Y2+dg2] + b[Y2+db2];\
- }\
- }\
- break;\
- case PIX_FMT_RGB8:\
- case PIX_FMT_BGR8:\
- {\
- const uint8_t * const d64= dither_8x8_73[y&7];\
- const uint8_t * const d32= dither_8x8_32[y&7];\
- func(uint8_t,0)\
- ((uint8_t*)dest)[i2+0]= r[Y1+d32[(i2+0)&7]] + g[Y1+d32[(i2+0)&7]] + b[Y1+d64[(i2+0)&7]];\
- ((uint8_t*)dest)[i2+1]= r[Y2+d32[(i2+1)&7]] + g[Y2+d32[(i2+1)&7]] + b[Y2+d64[(i2+1)&7]];\
- }\
- }\
- break;\
- case PIX_FMT_RGB4:\
- case PIX_FMT_BGR4:\
- {\
- const uint8_t * const d64= dither_8x8_73 [y&7];\
- const uint8_t * const d128=dither_8x8_220[y&7];\
- func(uint8_t,0)\
- ((uint8_t*)dest)[i]= r[Y1+d128[(i2+0)&7]] + g[Y1+d64[(i2+0)&7]] + b[Y1+d128[(i2+0)&7]]\
- + ((r[Y2+d128[(i2+1)&7]] + g[Y2+d64[(i2+1)&7]] + b[Y2+d128[(i2+1)&7]])<<4);\
- }\
- }\
- break;\
- case PIX_FMT_RGB4_BYTE:\
- case PIX_FMT_BGR4_BYTE:\
- {\
- const uint8_t * const d64= dither_8x8_73 [y&7];\
- const uint8_t * const d128=dither_8x8_220[y&7];\
- func(uint8_t,0)\
- ((uint8_t*)dest)[i2+0]= r[Y1+d128[(i2+0)&7]] + g[Y1+d64[(i2+0)&7]] + b[Y1+d128[(i2+0)&7]];\
- ((uint8_t*)dest)[i2+1]= r[Y2+d128[(i2+1)&7]] + g[Y2+d64[(i2+1)&7]] + b[Y2+d128[(i2+1)&7]];\
- }\
- }\
- break;\
- }
- static void yuv2packedX_c(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, int dstW, int y)
- {
- int i;
- YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGBX_C)
- }
- static void yuv2rgbX_c_full(SwsContext *c, const int16_t *lumFilter,
- const int16_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int16_t **chrUSrc,
- const int16_t **chrVSrc, int chrFilterSize,
- const int16_t **alpSrc, uint8_t *dest, int dstW, int y)
- {
- int i;
- int step= c->dstFormatBpp/8;
- int aidx= 3;
- switch(c->dstFormat) {
- case PIX_FMT_ARGB:
- dest++;
- aidx= 0;
- case PIX_FMT_RGB24:
- aidx--;
- case PIX_FMT_RGBA:
- if (CONFIG_SMALL) {
- int needAlpha = CONFIG_SWSCALE_ALPHA && c->alpPixBuf;
- YSCALE_YUV_2_RGBX_FULL_C(1<<21, needAlpha)
- dest[aidx]= needAlpha ? A : 255;
- dest[0]= R>>22;
- dest[1]= G>>22;
- dest[2]= B>>22;
- dest+= step;
- }
- } else {
- if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf) {
- YSCALE_YUV_2_RGBX_FULL_C(1<<21, 1)
- dest[aidx]= A;
- dest[0]= R>>22;
- dest[1]= G>>22;
- dest[2]= B>>22;
- dest+= step;
- }
- } else {
- YSCALE_YUV_2_RGBX_FULL_C(1<<21, 0)
- dest[aidx]= 255;
- dest[0]= R>>22;
- dest[1]= G>>22;
- dest[2]= B>>22;
- dest+= step;
- }
- }
- }
- break;
- case PIX_FMT_ABGR:
- dest++;
- aidx= 0;
- case PIX_FMT_BGR24:
- aidx--;
- case PIX_FMT_BGRA:
- if (CONFIG_SMALL) {
- int needAlpha = CONFIG_SWSCALE_ALPHA && c->alpPixBuf;
- YSCALE_YUV_2_RGBX_FULL_C(1<<21, needAlpha)
- dest[aidx]= needAlpha ? A : 255;
- dest[0]= B>>22;
- dest[1]= G>>22;
- dest[2]= R>>22;
- dest+= step;
- }
- } else {
- if (CONFIG_SWSCALE_ALPHA && c->alpPixBuf) {
- YSCALE_YUV_2_RGBX_FULL_C(1<<21, 1)
- dest[aidx]= A;
- dest[0]= B>>22;
- dest[1]= G>>22;
- dest[2]= R>>22;
- dest+= step;
- }
- } else {
- YSCALE_YUV_2_RGBX_FULL_C(1<<21, 0)
- dest[aidx]= 255;
- dest[0]= B>>22;
- dest[1]= G>>22;
- dest[2]= R>>22;
- dest+= step;
- }
- }
- }
- break;
- default:
- assert(0);
- }
- }
- /**
- * vertical bilinear scale YV12 to RGB
- */
- static void yuv2packed2_c(SwsContext *c, const uint16_t *buf0,
- const uint16_t *buf1, const uint16_t *ubuf0,
- const uint16_t *ubuf1, const uint16_t *vbuf0,
- const uint16_t *vbuf1, const uint16_t *abuf0,
- const uint16_t *abuf1, uint8_t *dest, int dstW,
- int yalpha, int uvalpha, int y)
- {
- int yalpha1=4095- yalpha;
- int uvalpha1=4095-uvalpha;
- int i;
- YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGB2_C)
- }
- /**
- * YV12 to RGB without scaling or interpolating
- */
- static void yuv2packed1_c(SwsContext *c, const uint16_t *buf0,
- const uint16_t *ubuf0, const uint16_t *ubuf1,
- const uint16_t *vbuf0, const uint16_t *vbuf1,
- const uint16_t *abuf0, uint8_t *dest, int dstW,
- int uvalpha, enum PixelFormat dstFormat,
- int flags, int y)
- {
- int i;
- if (uvalpha < 2048) {
- YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGB1_C)
- } else {
- YSCALE_YUV_2_ANYRGB_C(YSCALE_YUV_2_RGB1B_C)
- }
- }
- static av_always_inline void fillPlane(uint8_t* plane, int stride,
- int width, int height,
- int y, uint8_t val)
- {
- int i;
- uint8_t *ptr = plane + stride*y;
- for (i=0; i<height; i++) {
- memset(ptr, val, width);
- ptr += stride;
- }
- }
- #define input_pixel(pos) (isBE(origin) ? AV_RB16(pos) : AV_RL16(pos))
- #define r ((origin == PIX_FMT_BGR48BE || origin == PIX_FMT_BGR48LE) ? b_r : r_b)
- #define b ((origin == PIX_FMT_BGR48BE || origin == PIX_FMT_BGR48LE) ? r_b : b_r)
- static av_always_inline void
- rgb48ToY_c_template(int16_t *dst, const uint16_t *src, int width,
- enum PixelFormat origin)
- {
- int i;
- for (i = 0; i < width; i++) {
- int r_b = input_pixel(&src[i*3+0]);
- int g = input_pixel(&src[i*3+1]);
- int b_r = input_pixel(&src[i*3+2]);
- dst[i] = (RY*r + GY*g + BY*b + (32<<(RGB2YUV_SHIFT-1+8)) + (1<<(RGB2YUV_SHIFT-7+8))) >> (RGB2YUV_SHIFT-6+8);
- }
- }
- static av_always_inline void
- rgb48ToUV_c_template(int16_t *dstU, int16_t *dstV,
- const uint16_t *src1, const uint16_t *src2,
- int width, enum PixelFormat origin)
- {
- int i;
- assert(src1==src2);
- for (i = 0; i < width; i++) {
- int r_b = input_pixel(&src1[i*3+0]);
- int g = input_pixel(&src1[i*3+1]);
- int b_r = input_pixel(&src1[i*3+2]);
- dstU[i] = (RU*r + GU*g + BU*b + (256<<(RGB2YUV_SHIFT-1+8)) + (1<<(RGB2YUV_SHIFT-7+8))) >> (RGB2YUV_SHIFT-6+8);
- dstV[i] = (RV*r + GV*g + BV*b + (256<<(RGB2YUV_SHIFT-1+8)) + (1<<(RGB2YUV_SHIFT-7+8))) >> (RGB2YUV_SHIFT-6+8);
- }
- }
- static av_always_inline void
- rgb48ToUV_half_c_template(int16_t *dstU, int16_t *dstV,
- const uint16_t *src1, const uint16_t *src2,
- int width, enum PixelFormat origin)
- {
- int i;
- assert(src1==src2);
- for (i = 0; i < width; i++) {
- int r_b = (input_pixel(&src1[6*i + 0])) + (input_pixel(&src1[6*i + 3]));
- int g = (input_pixel(&src1[6*i + 1])) + (input_pixel(&src1[6*i + 4]));
- int b_r = (input_pixel(&src1[6*i + 2])) + (input_pixel(&src1[6*i + 5]));
- dstU[i]= (RU*r + GU*g + BU*b + (256U<<(RGB2YUV_SHIFT+8)) + (1<<(RGB2YUV_SHIFT-6+8))) >> (RGB2YUV_SHIFT-5+8);
- dstV[i]= (RV*r + GV*g + BV*b + (256U<<(RGB2YUV_SHIFT+8)) + (1<<(RGB2YUV_SHIFT-6+8))) >> (RGB2YUV_SHIFT-5+8);
- }
- }
- #undef r
- #undef b
- #undef input_pixel
- #define rgb48funcs(pattern, BE_LE, origin) \
- static void pattern ## 48 ## BE_LE ## ToY_c(uint8_t *dst, const uint8_t *src, \
- int width, uint32_t *unused) \
- { \
- rgb48ToY_c_template(dst, src, width, origin); \
- } \
- \
- static void pattern ## 48 ## BE_LE ## ToUV_c(uint8_t *dstU, uint8_t *dstV, \
- const uint8_t *src1, const uint8_t *src2, \
- int width, uint32_t *unused) \
- { \
- rgb48ToUV_c_template(dstU, dstV, src1, src2, width, origin); \
- } \
- \
- static void pattern ## 48 ## BE_LE ## ToUV_half_c(uint8_t *dstU, uint8_t *dstV, \
- const uint8_t *src1, const uint8_t *src2, \
- int width, uint32_t *unused) \
- { \
- rgb48ToUV_half_c_template(dstU, dstV, src1, src2, width, origin); \
- }
- rgb48funcs(rgb, LE, PIX_FMT_RGB48LE);
- rgb48funcs(rgb, BE, PIX_FMT_RGB48BE);
- rgb48funcs(bgr, LE, PIX_FMT_BGR48LE);
- rgb48funcs(bgr, BE, PIX_FMT_BGR48BE);
- #define input_pixel(i) ((origin == PIX_FMT_RGBA || origin == PIX_FMT_BGRA || \
- origin == PIX_FMT_ARGB || origin == PIX_FMT_ABGR) ? AV_RN32A(&src[(i)*4]) : \
- (isBE(origin) ? AV_RB16(&src[(i)*2]) : AV_RL16(&src[(i)*2])))
- static av_always_inline void
- rgb16_32ToY_c_template(int16_t *dst, const uint8_t *src,
- int width, enum PixelFormat origin,
- int shr, int shg, int shb, int shp,
- int maskr, int maskg, int maskb,
- int rsh, int gsh, int bsh, int S)
- {
- const int ry = RY << rsh, gy = GY << gsh, by = BY << bsh,
- rnd = (32<<((S)-1)) + (1<<(S-7));
- int i;
- for (i = 0; i < width; i++) {
- int px = input_pixel(i) >> shp;
- int b = (px & maskb) >> shb;
- int g = (px & maskg) >> shg;
- int r = (px & maskr) >> shr;
- dst[i] = (ry * r + gy * g + by * b + rnd) >> ((S)-6);
- }
- }
- static av_always_inline void
- rgb16_32ToUV_c_template(int16_t *dstU, int16_t *dstV,
- const uint8_t *src, int width,
- enum PixelFormat origin,
- int shr, int shg, int shb, int shp,
- int maskr, int maskg, int maskb,
- int rsh, int gsh, int bsh, int S)
- {
- const int ru = RU << rsh, gu = GU << gsh, bu = BU << bsh,
- rv = RV << rsh, gv = GV << gsh, bv = BV << bsh,
- rnd = (256<<((S)-1)) + (1<<(S-7));
- int i;
- for (i = 0; i < width; i++) {
- int px = input_pixel(i) >> shp;
- int b = (px & maskb) >> shb;
- int g = (px & maskg) >> shg;
- int r = (px & maskr) >> shr;
- dstU[i] = (ru * r + gu * g + bu * b + rnd) >> ((S)-6);
- dstV[i] = (rv * r + gv * g + bv * b + rnd) >> ((S)-6);
- }
- }
- static av_always_inline void
- rgb16_32ToUV_half_c_template(int16_t *dstU, int16_t *dstV,
- const uint8_t *src, int width,
- enum PixelFormat origin,
- int shr, int shg, int shb, int shp,
- int maskr, int maskg, int maskb,
- int rsh, int gsh, int bsh, int S)
- {
- const int ru = RU << rsh, gu = GU << gsh, bu = BU << bsh,
- rv = RV << rsh, gv = GV << gsh, bv = BV << bsh,
- rnd = (256U<<(S)) + (1<<(S-6)), maskgx = ~(maskr | maskb);
- int i;
- maskr |= maskr << 1; maskb |= maskb << 1; maskg |= maskg << 1;
- for (i = 0; i < width; i++) {
- int px0 = input_pixel(2 * i + 0) >> shp;
- int px1 = input_pixel(2 * i + 1) >> shp;
- int b, r, g = (px0 & maskgx) + (px1 & maskgx);
- int rb = px0 + px1 - g;
- b = (rb & maskb) >> shb;
- if (shp || origin == PIX_FMT_BGR565LE || origin == PIX_FMT_BGR565BE ||
- origin == PIX_FMT_RGB565LE || origin == PIX_FMT_RGB565BE) {
- g >>= shg;
- } else {
- g = (g & maskg) >> shg;
- }
- r = (rb & maskr) >> shr;
- dstU[i] = (ru * r + gu * g + bu * b + (unsigned)rnd) >> ((S)-6+1);
- dstV[i] = (rv * r + gv * g + bv * b + (unsigned)rnd) >> ((S)-6+1);
- }
- }
- #undef input_pixel
- #define rgb16_32_wrapper(fmt, name, shr, shg, shb, shp, maskr, \
- maskg, maskb, rsh, gsh, bsh, S) \
- static void name ## ToY_c(uint8_t *dst, const uint8_t *src, \
- int width, uint32_t *unused) \
- { \
- rgb16_32ToY_c_template(dst, src, width, fmt, shr, shg, shb, shp, \
- maskr, maskg, maskb, rsh, gsh, bsh, S); \
- } \
- \
- static void name ## ToUV_c(uint8_t *dstU, uint8_t *dstV, \
- const uint8_t *src, const uint8_t *dummy, \
- int width, uint32_t *unused) \
- { \
- rgb16_32ToUV_c_template(dstU, dstV, src, width, fmt, shr, shg, shb, shp, \
- maskr, maskg, maskb, rsh, gsh, bsh, S); \
- } \
- \
- static void name ## ToUV_half_c(uint8_t *dstU, uint8_t *dstV, \
- const uint8_t *src, const uint8_t *dummy, \
- int width, uint32_t *unused) \
- { \
- rgb16_32ToUV_half_c_template(dstU, dstV, src, width, fmt, shr, shg, shb, shp, \
- maskr, maskg, maskb, rsh, gsh, bsh, S); \
- }
- rgb16_32_wrapper(PIX_FMT_BGR32, bgr32, 16, 0, 0, 0, 0xFF0000, 0xFF00, 0x00FF, 8, 0, 8, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_BGR32_1, bgr321, 16, 0, 0, 8, 0xFF0000, 0xFF00, 0x00FF, 8, 0, 8, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_RGB32, rgb32, 0, 0, 16, 0, 0x00FF, 0xFF00, 0xFF0000, 8, 0, 8, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_RGB32_1, rgb321, 0, 0, 16, 8, 0x00FF, 0xFF00, 0xFF0000, 8, 0, 8, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_BGR565LE, bgr16le, 0, 0, 0, 0, 0x001F, 0x07E0, 0xF800, 11, 5, 0, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_BGR555LE, bgr15le, 0, 0, 0, 0, 0x001F, 0x03E0, 0x7C00, 10, 5, 0, RGB2YUV_SHIFT+7);
- rgb16_32_wrapper(PIX_FMT_RGB565LE, rgb16le, 0, 0, 0, 0, 0xF800, 0x07E0, 0x001F, 0, 5, 11, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_RGB555LE, rgb15le, 0, 0, 0, 0, 0x7C00, 0x03E0, 0x001F, 0, 5, 10, RGB2YUV_SHIFT+7);
- rgb16_32_wrapper(PIX_FMT_BGR565BE, bgr16be, 0, 0, 0, 0, 0x001F, 0x07E0, 0xF800, 11, 5, 0, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_BGR555BE, bgr15be, 0, 0, 0, 0, 0x001F, 0x03E0, 0x7C00, 10, 5, 0, RGB2YUV_SHIFT+7);
- rgb16_32_wrapper(PIX_FMT_RGB565BE, rgb16be, 0, 0, 0, 0, 0xF800, 0x07E0, 0x001F, 0, 5, 11, RGB2YUV_SHIFT+8);
- rgb16_32_wrapper(PIX_FMT_RGB555BE, rgb15be, 0, 0, 0, 0, 0x7C00, 0x03E0, 0x001F, 0, 5, 10, RGB2YUV_SHIFT+7);
- static void abgrToA_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- dst[i]= src[4*i]<<6;
- }
- }
- static void rgbaToA_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- dst[i]= src[4*i+3]<<6;
- }
- }
- static void palToA_c(int16_t *dst, const uint8_t *src, int width, uint32_t *pal)
- {
- int i;
- for (i=0; i<width; i++) {
- int d= src[i];
- dst[i]= (pal[d] >> 24)<<6;
- }
- }
- static void palToY_c(int16_t *dst, const uint8_t *src, long width, uint32_t *pal)
- {
- int i;
- for (i=0; i<width; i++) {
- int d= src[i];
- dst[i]= (pal[d] & 0xFF)<<6;
- }
- }
- static void palToUV_c(uint16_t *dstU, int16_t *dstV,
- const uint8_t *src1, const uint8_t *src2,
- int width, uint32_t *pal)
- {
- int i;
- assert(src1 == src2);
- for (i=0; i<width; i++) {
- int p= pal[src1[i]];
- dstU[i]= (uint8_t)(p>> 8)<<6;
- dstV[i]= (uint8_t)(p>>16)<<6;
- }
- }
- static void monowhite2Y_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
- {
- int i, j;
- for (i=0; i<width/8; i++) {
- int d= ~src[i];
- for(j=0; j<8; j++)
- dst[8*i+j]= ((d>>(7-j))&1)*16383;
- }
- if(width&7){
- int d= ~src[i];
- for(j=0; j<(width&7); j++)
- dst[8*i+j]= ((d>>(7-j))&1)*16383;
- }
- }
- static void monoblack2Y_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
- {
- int i, j;
- for (i=0; i<width/8; i++) {
- int d= src[i];
- for(j=0; j<8; j++)
- dst[8*i+j]= ((d>>(7-j))&1)*16383;
- }
- if(width&7){
- int d= src[i];
- for(j=0; j<(width&7); j++)
- dst[8*i+j]= ((d>>(7-j))&1)*16383;
- }
- }
- //FIXME yuy2* can read up to 7 samples too much
- static void yuy2ToY_c(uint8_t *dst, const uint8_t *src, int width,
- uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++)
- dst[i]= src[2*i];
- }
- static void yuy2ToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- dstU[i]= src1[4*i + 1];
- dstV[i]= src1[4*i + 3];
- }
- assert(src1 == src2);
- }
- static void LEToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- dstU[i]= src1[2*i + 1];
- dstV[i]= src2[2*i + 1];
- }
- }
- /* This is almost identical to the previous, end exists only because
- * yuy2ToY/UV)(dst, src+1, ...) would have 100% unaligned accesses. */
- static void uyvyToY_c(uint8_t *dst, const uint8_t *src, int width,
- uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++)
- dst[i]= src[2*i+1];
- }
- static void uyvyToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- dstU[i]= src1[4*i + 0];
- dstV[i]= src1[4*i + 2];
- }
- assert(src1 == src2);
- }
- static void BEToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- dstU[i]= src1[2*i];
- dstV[i]= src2[2*i];
- }
- }
- static av_always_inline void nvXXtoUV_c(uint8_t *dst1, uint8_t *dst2,
- const uint8_t *src, int width)
- {
- int i;
- for (i = 0; i < width; i++) {
- dst1[i] = src[2*i+0];
- dst2[i] = src[2*i+1];
- }
- }
- static void nv12ToUV_c(uint8_t *dstU, uint8_t *dstV,
- const uint8_t *src1, const uint8_t *src2,
- int width, uint32_t *unused)
- {
- nvXXtoUV_c(dstU, dstV, src1, width);
- }
- static void nv21ToUV_c(uint8_t *dstU, uint8_t *dstV,
- const uint8_t *src1, const uint8_t *src2,
- int width, uint32_t *unused)
- {
- nvXXtoUV_c(dstV, dstU, src1, width);
- }
- #define input_pixel(pos) (isBE(origin) ? AV_RB16(pos) : AV_RL16(pos))
- // FIXME Maybe dither instead.
- static av_always_inline void
- yuv9_OR_10ToUV_c_template(uint8_t *dstU, uint8_t *dstV,
- const uint8_t *_srcU, const uint8_t *_srcV,
- int width, enum PixelFormat origin, int depth)
- {
- int i;
- const uint16_t *srcU = (const uint16_t *) _srcU;
- const uint16_t *srcV = (const uint16_t *) _srcV;
- for (i = 0; i < width; i++) {
- dstU[i] = input_pixel(&srcU[i]) >> (depth - 8);
- dstV[i] = input_pixel(&srcV[i]) >> (depth - 8);
- }
- }
- static av_always_inline void
- yuv9_or_10ToY_c_template(uint8_t *dstY, const uint8_t *_srcY,
- int width, enum PixelFormat origin, int depth)
- {
- int i;
- const uint16_t *srcY = (const uint16_t*)_srcY;
- for (i = 0; i < width; i++)
- dstY[i] = input_pixel(&srcY[i]) >> (depth - 8);
- }
- #undef input_pixel
- #define YUV_NBPS(depth, BE_LE, origin) \
- static void BE_LE ## depth ## ToUV_c(uint8_t *dstU, uint8_t *dstV, \
- const uint8_t *srcU, const uint8_t *srcV, \
- int width, uint32_t *unused) \
- { \
- yuv9_OR_10ToUV_c_template(dstU, dstV, srcU, srcV, width, origin, depth); \
- } \
- static void BE_LE ## depth ## ToY_c(uint8_t *dstY, const uint8_t *srcY, \
- int width, uint32_t *unused) \
- { \
- yuv9_or_10ToY_c_template(dstY, srcY, width, origin, depth); \
- }
- YUV_NBPS( 9, LE, PIX_FMT_YUV420P9LE);
- YUV_NBPS( 9, BE, PIX_FMT_YUV420P9BE);
- YUV_NBPS(10, LE, PIX_FMT_YUV420P10LE);
- YUV_NBPS(10, BE, PIX_FMT_YUV420P10BE);
- static void bgr24ToY_c(int16_t *dst, const uint8_t *src,
- int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- int b= src[i*3+0];
- int g= src[i*3+1];
- int r= src[i*3+2];
- dst[i]= ((RY*r + GY*g + BY*b + (32<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6));
- }
- }
- static void bgr24ToUV_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- int b= src1[3*i + 0];
- int g= src1[3*i + 1];
- int r= src1[3*i + 2];
- dstU[i]= (RU*r + GU*g + BU*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
- dstV[i]= (RV*r + GV*g + BV*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
- }
- assert(src1 == src2);
- }
- static void bgr24ToUV_half_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- int b= src1[6*i + 0] + src1[6*i + 3];
- int g= src1[6*i + 1] + src1[6*i + 4];
- int r= src1[6*i + 2] + src1[6*i + 5];
- dstU[i]= (RU*r + GU*g + BU*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
- dstV[i]= (RV*r + GV*g + BV*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
- }
- assert(src1 == src2);
- }
- static void rgb24ToY_c(int16_t *dst, const uint8_t *src, int width,
- uint32_t *unused)
- {
- int i;
- for (i=0; i<width; i++) {
- int r= src[i*3+0];
- int g= src[i*3+1];
- int b= src[i*3+2];
- dst[i]= ((RY*r + GY*g + BY*b + (32<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6));
- }
- }
- static void rgb24ToUV_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- assert(src1==src2);
- for (i=0; i<width; i++) {
- int r= src1[3*i + 0];
- int g= src1[3*i + 1];
- int b= src1[3*i + 2];
- dstU[i]= (RU*r + GU*g + BU*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
- dstV[i]= (RV*r + GV*g + BV*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
- }
- }
- static void rgb24ToUV_half_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
- {
- int i;
- assert(src1==src2);
- for (i=0; i<width; i++) {
- int r= src1[6*i + 0] + src1[6*i + 3];
- int g= src1[6*i + 1] + src1[6*i + 4];
- int b= src1[6*i + 2] + src1[6*i + 5];
- dstU[i]= (RU*r + GU*g + BU*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
- dstV[i]= (RV*r + GV*g + BV*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
- }
- }
- // bilinear / bicubic scaling
- static void hScale_c(int16_t *dst, int dstW, const uint8_t *src,
- int srcW, int xInc,
- const int16_t *filter, const int16_t *filterPos,
- int filterSize)
- {
- int i;
- for (i=0; i<dstW; i++) {
- int j;
- int srcPos= filterPos[i];
- int val=0;
- for (j=0; j<filterSize; j++) {
- val += ((int)src[srcPos + j])*filter[filterSize*i + j];
- }
- //filter += hFilterSize;
- dst[i] = FFMIN(val>>7, (1<<15)-1); // the cubic equation does overflow ...
- //dst[i] = val>>7;
- }
- }
- static inline void hScale16_c(int16_t *dst, int dstW, const uint16_t *src, int srcW, int xInc,
- const int16_t *filter, const int16_t *filterPos, long filterSize, int shift)
- {
- int i, j;
- for (i=0; i<dstW; i++) {
- int srcPos= filterPos[i];
- int val=0;
- for (j=0; j<filterSize; j++) {
- val += ((int)src[srcPos + j])*filter[filterSize*i + j];
- }
- dst[i] = FFMIN(val>>shift, (1<<15)-1); // the cubic equation does overflow ...
- }
- }
- static inline void hScale16X_c(int16_t *dst, int dstW, const uint16_t *src, int srcW, int xInc,
- const int16_t *filter, const int16_t *filterPos, long filterSize, int shift)
- {
- int i, j;
- for (i=0; i<dstW; i++) {
- int srcPos= filterPos[i];
- int val=0;
- for (j=0; j<filterSize; j++) {
- val += ((int)av_bswap16(src[srcPos + j]))*filter[filterSize*i + j];
- }
- dst[i] = FFMIN(val>>shift, (1<<15)-1); // the cubic equation does overflow ...
- }
- }
- //FIXME all pal and rgb srcFormats could do this convertion as well
- //FIXME all scalers more complex than bilinear could do half of this transform
- static void chrRangeToJpeg_c(int16_t *dstU, int16_t *dstV, int width)
- {
- int i;
- for (i = 0; i < width; i++) {
- dstU[i] = (FFMIN(dstU[i],30775)*4663 - 9289992)>>12; //-264
- dstV[i] = (FFMIN(dstV[i],30775)*4663 - 9289992)>>12; //-264
- }
- }
- static void chrRangeFromJpeg_c(int16_t *dstU, int16_t *dstV, int width)
- {
- int i;
- for (i = 0; i < width; i++) {
- dstU[i] = (dstU[i]*1799 + 4081085)>>11; //1469
- dstV[i] = (dstV[i]*1799 + 4081085)>>11; //1469
- }
- }
- static void lumRangeToJpeg_c(int16_t *dst, int width)
- {
- int i;
- for (i = 0; i < width; i++)
- dst[i] = (FFMIN(dst[i],30189)*19077 - 39057361)>>14;
- }
- static void lumRangeFromJpeg_c(int16_t *dst, int width)
- {
- int i;
- for (i = 0; i < width; i++)
- dst[i] = (dst[i]*14071 + 33561947)>>14;
- }
- static void hyscale_fast_c(SwsContext *c, int16_t *dst, int dstWidth,
- const uint8_t *src, int srcW, int xInc)
- {
- int i;
- unsigned int xpos=0;
- for (i=0;i<dstWidth;i++) {
- register unsigned int xx=xpos>>16;
- register unsigned int xalpha=(xpos&0xFFFF)>>9;
- dst[i]= (src[xx]<<7) + (src[xx+1] - src[xx])*xalpha;
- xpos+=xInc;
- }
- for (i=dstWidth-1; (i*xInc)>>16 >=srcW-1; i--)
- dst[i] = src[srcW-1]*128;
- }
- // *** horizontal scale Y line to temp buffer
- static av_always_inline void hyscale(SwsContext *c, uint16_t *dst, int dstWidth,
- const uint8_t *src, int srcW, int xInc,
- const int16_t *hLumFilter,
- const int16_t *hLumFilterPos, int hLumFilterSize,
- uint8_t *formatConvBuffer,
- uint32_t *pal, int isAlpha)
- {
- void (*toYV12)(uint8_t *, const uint8_t *, int, uint32_t *) = isAlpha ? c->alpToYV12 : c->lumToYV12;
- void (*convertRange)(int16_t *, int) = isAlpha ? NULL : c->lumConvertRange;
- if (toYV12) {
- toYV12(formatConvBuffer, src, srcW, pal);
- src= formatConvBuffer;
- }
- if (c->hScale16) {
- int shift= isAnyRGB(c->srcFormat) || c->srcFormat==PIX_FMT_PAL8 ? 13 : av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1;
- c->hScale16(dst, dstWidth, (const uint16_t*)src, srcW, xInc, hLumFilter, hLumFilterPos, hLumFilterSize, shift);
- } else if (!c->hyscale_fast) {
- c->hScale(dst, dstWidth, src, srcW, xInc, hLumFilter, hLumFilterPos, hLumFilterSize);
- } else { // fast bilinear upscale / crap downscale
- c->hyscale_fast(c, dst, dstWidth, src, srcW, xInc);
- }
- if (convertRange)
- convertRange(dst, dstWidth);
- }
- static void hcscale_fast_c(SwsContext *c, int16_t *dst1, int16_t *dst2,
- int dstWidth, const uint8_t *src1,
- const uint8_t *src2, int srcW, int xInc)
- {
- int i;
- unsigned int xpos=0;
- for (i=0;i<dstWidth;i++) {
- register unsigned int xx=xpos>>16;
- register unsigned int xalpha=(xpos&0xFFFF)>>9;
- dst1[i]=(src1[xx]*(xalpha^127)+src1[xx+1]*xalpha);
- dst2[i]=(src2[xx]*(xalpha^127)+src2[xx+1]*xalpha);
- xpos+=xInc;
- }
- for (i=dstWidth-1; (i*xInc)>>16 >=srcW-1; i--) {
- dst1[i] = src1[srcW-1]*128;
- dst2[i] = src2[srcW-1]*128;
- }
- }
- static av_always_inline void hcscale(SwsContext *c, uint16_t *dst1, uint16_t *dst2, int dstWidth,
- const uint8_t *src1, const uint8_t *src2,
- int srcW, int xInc, const int16_t *hChrFilter,
- const int16_t *hChrFilterPos, int hChrFilterSize,
- uint8_t *formatConvBuffer, uint32_t *pal)
- {
- if (c->chrToYV12) {
- uint8_t *buf2 = formatConvBuffer + FFALIGN(srcW*2+78, 16);
- c->chrToYV12(formatConvBuffer, buf2, src1, src2, srcW, pal);
- src1= formatConvBuffer;
- src2= buf2;
- }
- if (c->hScale16) {
- int shift= isAnyRGB(c->srcFormat) || c->srcFormat==PIX_FMT_PAL8 ? 13 : av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1;
- c->hScale16(dst1, dstWidth, (const uint16_t*)src1, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize, shift);
- c->hScale16(dst2, dstWidth, (const uint16_t*)src2, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize, shift);
- } else if (!c->hcscale_fast) {
- c->hScale(dst1, dstWidth, src1, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize);
- c->hScale(dst2, dstWidth, src2, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize);
- } else { // fast bilinear upscale / crap downscale
- c->hcscale_fast(c, dst1, dst2, dstWidth, src1, src2, srcW, xInc);
- }
- if (c->chrConvertRange)
- c->chrConvertRange(dst1, dst2, dstWidth);
- }
- static av_always_inline void
- find_c_packed_planar_out_funcs(SwsContext *c,
- yuv2planar1_fn *yuv2yuv1, yuv2planarX_fn *yuv2yuvX,
- yuv2packed1_fn *yuv2packed1, yuv2packed2_fn *yuv2packed2,
- yuv2packedX_fn *yuv2packedX)
- {
- enum PixelFormat dstFormat = c->dstFormat;
- if (dstFormat == PIX_FMT_NV12 || dstFormat == PIX_FMT_NV21) {
- *yuv2yuvX = yuv2nv12X_c;
- } else if (is16BPS(dstFormat)) {
- *yuv2yuvX = isBE(dstFormat) ? yuv2yuvX16BE_c : yuv2yuvX16LE_c;
- } else if (is9_OR_10BPS(dstFormat)) {
- if (av_pix_fmt_descriptors[dstFormat].comp[0].depth_minus1 == 8) {
- *yuv2yuvX = isBE(dstFormat) ? yuv2yuvX9BE_c : yuv2yuvX9LE_c;
- } else {
- *yuv2yuvX = isBE(dstFormat) ? yuv2yuvX10BE_c : yuv2yuvX10LE_c;
- }
- } else {
- *yuv2yuv1 = yuv2yuv1_c;
- *yuv2yuvX = yuv2yuvX_c;
- }
- if(c->flags & SWS_FULL_CHR_H_INT) {
- *yuv2packedX = yuv2rgbX_c_full;
- } else {
- switch (dstFormat) {
- case PIX_FMT_GRAY16BE:
- *yuv2packed1 = yuv2gray16BE_1_c;
- *yuv2packed2 = yuv2gray16BE_2_c;
- *yuv2packedX = yuv2gray16BE_X_c;
- break;
- case PIX_FMT_GRAY16LE:
- *yuv2packed1 = yuv2gray16LE_1_c;
- *yuv2packed2 = yuv2gray16LE_2_c;
- *yuv2packedX = yuv2gray16LE_X_c;
- break;
- case PIX_FMT_MONOWHITE:
- *yuv2packed1 = yuv2monowhite_1_c;
- *yuv2packed2 = yuv2monowhite_2_c;
- *yuv2packedX = yuv2monowhite_X_c;
- break;
- case PIX_FMT_MONOBLACK:
- *yuv2packed1 = yuv2monoblack_1_c;
- *yuv2packed2 = yuv2monoblack_2_c;
- *yuv2packedX = yuv2monoblack_X_c;
- break;
- case PIX_FMT_YUYV422:
- *yuv2packed1 = yuv2yuyv422_1_c;
- *yuv2packed2 = yuv2yuyv422_2_c;
- *yuv2packedX = yuv2yuyv422_X_c;
- break;
- case PIX_FMT_UYVY422:
- *yuv2packed1 = yuv2uyvy422_1_c;
- *yuv2packed2 = yuv2uyvy422_2_c;
- *yuv2packedX = yuv2uyvy422_X_c;
- break;
- case PIX_FMT_RGB48LE:
- //*yuv2packed1 = yuv2rgb48le_1_c;
- //*yuv2packed2 = yuv2rgb48le_2_c;
- //*yuv2packedX = yuv2rgb48le_X_c;
- //break;
- case PIX_FMT_RGB48BE:
- *yuv2packed1 = yuv2rgb48be_1_c;
- *yuv2packed2 = yuv2rgb48be_2_c;
- *yuv2packedX = yuv2rgb48be_X_c;
- break;
- case PIX_FMT_BGR48LE:
- //*yuv2packed1 = yuv2bgr48le_1_c;
- //*yuv2packed2 = yuv2bgr48le_2_c;
- //*yuv2packedX = yuv2bgr48le_X_c;
- //break;
- case PIX_FMT_BGR48BE:
- *yuv2packed1 = yuv2bgr48be_1_c;
- *yuv2packed2 = yuv2bgr48be_2_c;
- *yuv2packedX = yuv2bgr48be_X_c;
- break;
- default:
- *yuv2packed1 = yuv2packed1_c;
- *yuv2packed2 = yuv2packed2_c;
- *yuv2packedX = yuv2packedX_c;
- break;
- }
- }
- }
- #define DEBUG_SWSCALE_BUFFERS 0
- #define DEBUG_BUFFERS(...) if (DEBUG_SWSCALE_BUFFERS) av_log(c, AV_LOG_DEBUG, __VA_ARGS__)
- static int swScale(SwsContext *c, const uint8_t* src[],
- int srcStride[], int srcSliceY,
- int srcSliceH, uint8_t* dst[], int dstStride[])
- {
- /* load a few things into local vars to make the code more readable? and faster */
- const int srcW= c->srcW;
- const int dstW= c->dstW;
- const int dstH= c->dstH;
- const int chrDstW= c->chrDstW;
- const int chrSrcW= c->chrSrcW;
- const int lumXInc= c->lumXInc;
- const int chrXInc= c->chrXInc;
- const enum PixelFormat dstFormat= c->dstFormat;
- const int flags= c->flags;
- int16_t *vLumFilterPos= c->vLumFilterPos;
- int16_t *vChrFilterPos= c->vChrFilterPos;
- int16_t *hLumFilterPos= c->hLumFilterPos;
- int16_t *hChrFilterPos= c->hChrFilterPos;
- int16_t *vLumFilter= c->vLumFilter;
- int16_t *vChrFilter= c->vChrFilter;
- int16_t *hLumFilter= c->hLumFilter;
- int16_t *hChrFilter= c->hChrFilter;
- int32_t *lumMmxFilter= c->lumMmxFilter;
- int32_t *chrMmxFilter= c->chrMmxFilter;
- int32_t av_unused *alpMmxFilter= c->alpMmxFilter;
- const int vLumFilterSize= c->vLumFilterSize;
- const int vChrFilterSize= c->vChrFilterSize;
- const int hLumFilterSize= c->hLumFilterSize;
- const int hChrFilterSize= c->hChrFilterSize;
- int16_t **lumPixBuf= c->lumPixBuf;
- int16_t **chrUPixBuf= c->chrUPixBuf;
- int16_t **chrVPixBuf= c->chrVPixBuf;
- int16_t **alpPixBuf= c->alpPixBuf;
- const int vLumBufSize= c->vLumBufSize;
- const int vChrBufSize= c->vChrBufSize;
- uint8_t *formatConvBuffer= c->formatConvBuffer;
- const int chrSrcSliceY= srcSliceY >> c->chrSrcVSubSample;
- const int chrSrcSliceH= -((-srcSliceH) >> c->chrSrcVSubSample);
- int lastDstY;
- uint32_t *pal=c->pal_yuv;
- int should_dither= isNBPS(c->srcFormat) || is16BPS(c->srcFormat);
- yuv2planar1_fn yuv2yuv1 = c->yuv2yuv1;
- yuv2planarX_fn yuv2yuvX = c->yuv2yuvX;
- yuv2packed1_fn yuv2packed1 = c->yuv2packed1;
- yuv2packed2_fn yuv2packed2 = c->yuv2packed2;
- yuv2packedX_fn yuv2packedX = c->yuv2packedX;
- /* vars which will change and which we need to store back in the context */
- int dstY= c->dstY;
- int lumBufIndex= c->lumBufIndex;
- int chrBufIndex= c->chrBufIndex;
- int lastInLumBuf= c->lastInLumBuf;
- int lastInChrBuf= c->lastInChrBuf;
- if (isPacked(c->srcFormat)) {
- src[0]=
- src[1]=
- src[2]=
- src[3]= src[0];
- srcStride[0]=
- srcStride[1]=
- srcStride[2]=
- srcStride[3]= srcStride[0];
- }
- srcStride[1]<<= c->vChrDrop;
- srcStride[2]<<= c->vChrDrop;
- DEBUG_BUFFERS("swScale() %p[%d] %p[%d] %p[%d] %p[%d] -> %p[%d] %p[%d] %p[%d] %p[%d]\n",
- src[0], srcStride[0], src[1], srcStride[1], src[2], srcStride[2], src[3], srcStride[3],
- dst[0], dstStride[0], dst[1], dstStride[1], dst[2], dstStride[2], dst[3], dstStride[3]);
- DEBUG_BUFFERS("srcSliceY: %d srcSliceH: %d dstY: %d dstH: %d\n",
- srcSliceY, srcSliceH, dstY, dstH);
- DEBUG_BUFFERS("vLumFilterSize: %d vLumBufSize: %d vChrFilterSize: %d vChrBufSize: %d\n",
- vLumFilterSize, vLumBufSize, vChrFilterSize, vChrBufSize);
- if (dstStride[0]%8 !=0 || dstStride[1]%8 !=0 || dstStride[2]%8 !=0 || dstStride[3]%8 != 0) {
- static int warnedAlready=0; //FIXME move this into the context perhaps
- if (flags & SWS_PRINT_INFO && !warnedAlready) {
- av_log(c, AV_LOG_WARNING, "Warning: dstStride is not aligned!\n"
- " ->cannot do aligned memory accesses anymore\n");
- warnedAlready=1;
- }
- }
- /* Note the user might start scaling the picture in the middle so this
- will not get executed. This is not really intended but works
- currently, so people might do it. */
- if (srcSliceY ==0) {
- lumBufIndex=-1;
- chrBufIndex=-1;
- dstY=0;
- lastInLumBuf= -1;
- lastInChrBuf= -1;
- }
- lastDstY= dstY;
- for (;dstY < dstH; dstY++) {
- unsigned char *dest =dst[0]+dstStride[0]*dstY;
- const int chrDstY= dstY>>c->chrDstVSubSample;
- unsigned char *uDest=dst[1]+dstStride[1]*chrDstY;
- unsigned char *vDest=dst[2]+dstStride[2]*chrDstY;
- unsigned char *aDest=(CONFIG_SWSCALE_ALPHA && alpPixBuf) ? dst[3]+dstStride[3]*dstY : NULL;
- const uint8_t *lumDither= should_dither ? dithers[7][dstY &7] : flat64;
- const uint8_t *chrDither= should_dither ? dithers[7][chrDstY&7] : flat64;
- const int firstLumSrcY= vLumFilterPos[dstY]; //First line needed as input
- const int firstLumSrcY2= vLumFilterPos[FFMIN(dstY | ((1<<c->chrDstVSubSample) - 1), dstH-1)];
- const int firstChrSrcY= vChrFilterPos[chrDstY]; //First line needed as input
- int lastLumSrcY= firstLumSrcY + vLumFilterSize -1; // Last line needed as input
- int lastLumSrcY2=firstLumSrcY2+ vLumFilterSize -1; // Last line needed as input
- int lastChrSrcY= firstChrSrcY + vChrFilterSize -1; // Last line needed as input
- int enough_lines;
- //handle holes (FAST_BILINEAR & weird filters)
- if (firstLumSrcY > lastInLumBuf) lastInLumBuf= firstLumSrcY-1;
- if (firstChrSrcY > lastInChrBuf) lastInChrBuf= firstChrSrcY-1;
- assert(firstLumSrcY >= lastInLumBuf - vLumBufSize + 1);
- assert(firstChrSrcY >= lastInChrBuf - vChrBufSize + 1);
- DEBUG_BUFFERS("dstY: %d\n", dstY);
- DEBUG_BUFFERS("\tfirstLumSrcY: %d lastLumSrcY: %d lastInLumBuf: %d\n",
- firstLumSrcY, lastLumSrcY, lastInLumBuf);
- DEBUG_BUFFERS("\tfirstChrSrcY: %d lastChrSrcY: %d lastInChrBuf: %d\n",
- firstChrSrcY, lastChrSrcY, lastInChrBuf);
- // Do we have enough lines in this slice to output the dstY line
- enough_lines = lastLumSrcY2 < srcSliceY + srcSliceH && lastChrSrcY < -((-srcSliceY - srcSliceH)>>c->chrSrcVSubSample);
- if (!enough_lines) {
- lastLumSrcY = srcSliceY + srcSliceH - 1;
- lastChrSrcY = chrSrcSliceY + chrSrcSliceH - 1;
- DEBUG_BUFFERS("buffering slice: lastLumSrcY %d lastChrSrcY %d\n",
- lastLumSrcY, lastChrSrcY);
- }
- //Do horizontal scaling
- while(lastInLumBuf < lastLumSrcY) {
- const uint8_t *src1= src[0]+(lastInLumBuf + 1 - srcSliceY)*srcStride[0];
- const uint8_t *src2= src[3]+(lastInLumBuf + 1 - srcSliceY)*srcStride[3];
- lumBufIndex++;
- assert(lumBufIndex < 2*vLumBufSize);
- assert(lastInLumBuf + 1 - srcSliceY < srcSliceH);
- assert(lastInLumBuf + 1 - srcSliceY >= 0);
- hyscale(c, lumPixBuf[ lumBufIndex ], dstW, src1, srcW, lumXInc,
- hLumFilter, hLumFilterPos, hLumFilterSize,
- formatConvBuffer,
- pal, 0);
- if (CONFIG_SWSCALE_ALPHA && alpPixBuf)
- hyscale(c, alpPixBuf[ lumBufIndex ], dstW, src2, srcW,
- lumXInc, hLumFilter, hLumFilterPos, hLumFilterSize,
- formatConvBuffer,
- pal, 1);
- lastInLumBuf++;
- DEBUG_BUFFERS("\t\tlumBufIndex %d: lastInLumBuf: %d\n",
- lumBufIndex, lastInLumBuf);
- }
- while(lastInChrBuf < lastChrSrcY) {
- const uint8_t *src1= src[1]+(lastInChrBuf + 1 - chrSrcSliceY)*srcStride[1];
- const uint8_t *src2= src[2]+(lastInChrBuf + 1 - chrSrcSliceY)*srcStride[2];
- chrBufIndex++;
- assert(chrBufIndex < 2*vChrBufSize);
- assert(lastInChrBuf + 1 - chrSrcSliceY < (chrSrcSliceH));
- assert(lastInChrBuf + 1 - chrSrcSliceY >= 0);
- //FIXME replace parameters through context struct (some at least)
- if (c->needs_hcscale)
- hcscale(c, chrUPixBuf[chrBufIndex], chrVPixBuf[chrBufIndex],
- chrDstW, src1, src2, chrSrcW, chrXInc,
- hChrFilter, hChrFilterPos, hChrFilterSize,
- formatConvBuffer, pal);
- lastInChrBuf++;
- DEBUG_BUFFERS("\t\tchrBufIndex %d: lastInChrBuf: %d\n",
- chrBufIndex, lastInChrBuf);
- }
- //wrap buf index around to stay inside the ring buffer
- if (lumBufIndex >= vLumBufSize) lumBufIndex-= vLumBufSize;
- if (chrBufIndex >= vChrBufSize) chrBufIndex-= vChrBufSize;
- if (!enough_lines)
- break; //we can't output a dstY line so let's try with the next slice
- #if HAVE_MMX
- updateMMXDitherTables(c, dstY, lumBufIndex, chrBufIndex, lastInLumBuf, lastInChrBuf);
- #endif
- if (dstY >= dstH-2) {
- // hmm looks like we can't use MMX here without overwriting this array's tail
- find_c_packed_planar_out_funcs(c, &yuv2yuv1, &yuv2yuvX,
- &yuv2packed1, &yuv2packed2,
- &yuv2packedX);
- }
- {
- const int16_t **lumSrcPtr= (const int16_t **) lumPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize;
- const int16_t **chrUSrcPtr= (const int16_t **) chrUPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
- const int16_t **chrVSrcPtr= (const int16_t **) chrVPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
- const int16_t **alpSrcPtr= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? (const int16_t **) alpPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize : NULL;
- if (isPlanarYUV(dstFormat) || dstFormat==PIX_FMT_GRAY8) { //YV12 like
- const int chrSkipMask= (1<<c->chrDstVSubSample)-1;
- if ((dstY&chrSkipMask) || isGray(dstFormat)) uDest=vDest= NULL; //FIXME split functions in lumi / chromi
- if (c->yuv2yuv1 && vLumFilterSize == 1 && vChrFilterSize == 1) { // unscaled YV12
- const int16_t *lumBuf = lumSrcPtr[0];
- const int16_t *chrUBuf= chrUSrcPtr[0];
- const int16_t *chrVBuf= chrVSrcPtr[0];
- const int16_t *alpBuf= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? alpSrcPtr[0] : NULL;
- yuv2yuv1(c, lumBuf, chrUBuf, chrVBuf, alpBuf, dest,
- uDest, vDest, aDest, dstW, chrDstW, lumDither, chrDither);
- } else { //General YV12
- yuv2yuvX(c,
- vLumFilter+dstY*vLumFilterSize , lumSrcPtr, vLumFilterSize,
- vChrFilter+chrDstY*vChrFilterSize, chrUSrcPtr,
- chrVSrcPtr, vChrFilterSize,
- alpSrcPtr, dest, uDest, vDest, aDest, dstW, chrDstW, lumDither, chrDither);
- }
- } else {
- assert(lumSrcPtr + vLumFilterSize - 1 < lumPixBuf + vLumBufSize*2);
- assert(chrUSrcPtr + vChrFilterSize - 1 < chrUPixBuf + vChrBufSize*2);
- if (c->yuv2packed1 && vLumFilterSize == 1 && vChrFilterSize == 2) { //unscaled RGB
- int chrAlpha= vChrFilter[2*dstY+1];
- yuv2packed1(c, *lumSrcPtr, *chrUSrcPtr, *(chrUSrcPtr+1),
- *chrVSrcPtr, *(chrVSrcPtr+1),
- alpPixBuf ? *alpSrcPtr : NULL,
- dest, dstW, chrAlpha, dstFormat, flags, dstY);
- } else if (c->yuv2packed2 && vLumFilterSize == 2 && vChrFilterSize == 2) { //bilinear upscale RGB
- int lumAlpha= vLumFilter[2*dstY+1];
- int chrAlpha= vChrFilter[2*dstY+1];
- lumMmxFilter[2]=
- lumMmxFilter[3]= vLumFilter[2*dstY ]*0x10001;
- chrMmxFilter[2]=
- chrMmxFilter[3]= vChrFilter[2*chrDstY]*0x10001;
- yuv2packed2(c, *lumSrcPtr, *(lumSrcPtr+1), *chrUSrcPtr, *(chrUSrcPtr+1),
- *chrVSrcPtr, *(chrVSrcPtr+1),
- alpPixBuf ? *alpSrcPtr : NULL, alpPixBuf ? *(alpSrcPtr+1) : NULL,
- dest, dstW, lumAlpha, chrAlpha, dstY);
- } else { //general RGB
- yuv2packedX(c,
- vLumFilter+dstY*vLumFilterSize, lumSrcPtr, vLumFilterSize,
- vChrFilter+dstY*vChrFilterSize, chrUSrcPtr, chrVSrcPtr, vChrFilterSize,
- alpSrcPtr, dest, dstW, dstY);
- }
- }
- }
- }
- if ((dstFormat == PIX_FMT_YUVA420P) && !alpPixBuf)
- fillPlane(dst[3], dstStride[3], dstW, dstY-lastDstY, lastDstY, 255);
- #if HAVE_MMX2
- if (av_get_cpu_flags() & AV_CPU_FLAG_MMX2)
- __asm__ volatile("sfence":::"memory");
- #endif
- emms_c();
- /* store changed local vars back in the context */
- c->dstY= dstY;
- c->lumBufIndex= lumBufIndex;
- c->chrBufIndex= chrBufIndex;
- c->lastInLumBuf= lastInLumBuf;
- c->lastInChrBuf= lastInChrBuf;
- return dstY - lastDstY;
- }
- static av_cold void sws_init_swScale_c(SwsContext *c)
- {
- enum PixelFormat srcFormat = c->srcFormat;
- find_c_packed_planar_out_funcs(c, &c->yuv2yuv1, &c->yuv2yuvX,
- &c->yuv2packed1, &c->yuv2packed2,
- &c->yuv2packedX);
- c->hScale = hScale_c;
- if (c->flags & SWS_FAST_BILINEAR) {
- c->hyscale_fast = hyscale_fast_c;
- c->hcscale_fast = hcscale_fast_c;
- }
- c->chrToYV12 = NULL;
- switch(srcFormat) {
- case PIX_FMT_YUYV422 : c->chrToYV12 = yuy2ToUV_c; break;
- case PIX_FMT_UYVY422 : c->chrToYV12 = uyvyToUV_c; break;
- case PIX_FMT_NV12 : c->chrToYV12 = nv12ToUV_c; break;
- case PIX_FMT_NV21 : c->chrToYV12 = nv21ToUV_c; break;
- case PIX_FMT_RGB8 :
- case PIX_FMT_BGR8 :
- case PIX_FMT_PAL8 :
- case PIX_FMT_BGR4_BYTE:
- case PIX_FMT_RGB4_BYTE: c->chrToYV12 = palToUV_c; break;
- case PIX_FMT_GRAY16BE :
- case PIX_FMT_YUV444P9BE:
- case PIX_FMT_YUV420P9BE:
- case PIX_FMT_YUV444P10BE:
- case PIX_FMT_YUV422P10BE:
- case PIX_FMT_YUV420P10BE:
- case PIX_FMT_YUV420P16BE:
- case PIX_FMT_YUV422P16BE:
- case PIX_FMT_YUV444P16BE: c->hScale16= HAVE_BIGENDIAN ? hScale16_c : hScale16X_c; break;
- case PIX_FMT_GRAY16LE :
- case PIX_FMT_YUV444P9LE:
- case PIX_FMT_YUV420P9LE:
- case PIX_FMT_YUV422P10LE:
- case PIX_FMT_YUV420P10LE:
- case PIX_FMT_YUV444P10LE:
- case PIX_FMT_YUV420P16LE:
- case PIX_FMT_YUV422P16LE:
- case PIX_FMT_YUV444P16LE: c->hScale16= HAVE_BIGENDIAN ? hScale16X_c : hScale16_c; break;
- }
- if (c->chrSrcHSubSample) {
- switch(srcFormat) {
- case PIX_FMT_RGB48BE : c->chrToYV12 = rgb48BEToUV_half_c; break;
- case PIX_FMT_RGB48LE : c->chrToYV12 = rgb48LEToUV_half_c; break;
- case PIX_FMT_BGR48BE : c->chrToYV12 = bgr48BEToUV_half_c; break;
- case PIX_FMT_BGR48LE : c->chrToYV12 = bgr48LEToUV_half_c; break;
- case PIX_FMT_RGB32 : c->chrToYV12 = bgr32ToUV_half_c; break;
- case PIX_FMT_RGB32_1 : c->chrToYV12 = bgr321ToUV_half_c; break;
- case PIX_FMT_BGR24 : c->chrToYV12 = bgr24ToUV_half_c; break;
- case PIX_FMT_BGR565LE: c->chrToYV12 = bgr16leToUV_half_c; break;
- case PIX_FMT_BGR565BE: c->chrToYV12 = bgr16beToUV_half_c; break;
- case PIX_FMT_BGR555LE: c->chrToYV12 = bgr15leToUV_half_c; break;
- case PIX_FMT_BGR555BE: c->chrToYV12 = bgr15beToUV_half_c; break;
- case PIX_FMT_BGR32 : c->chrToYV12 = rgb32ToUV_half_c; break;
- case PIX_FMT_BGR32_1 : c->chrToYV12 = rgb321ToUV_half_c; break;
- case PIX_FMT_RGB24 : c->chrToYV12 = rgb24ToUV_half_c; break;
- case PIX_FMT_RGB565LE: c->chrToYV12 = rgb16leToUV_half_c; break;
- case PIX_FMT_RGB565BE: c->chrToYV12 = rgb16beToUV_half_c; break;
- case PIX_FMT_RGB555LE: c->chrToYV12 = rgb15leToUV_half_c; break;
- case PIX_FMT_RGB555BE: c->chrToYV12 = rgb15beToUV_half_c; break;
- }
- } else {
- switch(srcFormat) {
- case PIX_FMT_RGB48BE : c->chrToYV12 = rgb48BEToUV_c; break;
- case PIX_FMT_RGB48LE : c->chrToYV12 = rgb48LEToUV_c; break;
- case PIX_FMT_BGR48BE : c->chrToYV12 = bgr48BEToUV_c; break;
- case PIX_FMT_BGR48LE : c->chrToYV12 = bgr48LEToUV_c; break;
- case PIX_FMT_RGB32 : c->chrToYV12 = bgr32ToUV_c; break;
- case PIX_FMT_RGB32_1 : c->chrToYV12 = bgr321ToUV_c; break;
- case PIX_FMT_BGR24 : c->chrToYV12 = bgr24ToUV_c; break;
- case PIX_FMT_BGR565LE: c->chrToYV12 = bgr16leToUV_c; break;
- case PIX_FMT_BGR565BE: c->chrToYV12 = bgr16beToUV_c; break;
- case PIX_FMT_BGR555LE: c->chrToYV12 = bgr15leToUV_c; break;
- case PIX_FMT_BGR555BE: c->chrToYV12 = bgr15beToUV_c; break;
- case PIX_FMT_BGR32 : c->chrToYV12 = rgb32ToUV_c; break;
- case PIX_FMT_BGR32_1 : c->chrToYV12 = rgb321ToUV_c; break;
- case PIX_FMT_RGB24 : c->chrToYV12 = rgb24ToUV_c; break;
- case PIX_FMT_RGB565LE: c->chrToYV12 = rgb16leToUV_c; break;
- case PIX_FMT_RGB565BE: c->chrToYV12 = rgb16beToUV_c; break;
- case PIX_FMT_RGB555LE: c->chrToYV12 = rgb15leToUV_c; break;
- case PIX_FMT_RGB555BE: c->chrToYV12 = rgb15beToUV_c; break;
- }
- }
- c->lumToYV12 = NULL;
- c->alpToYV12 = NULL;
- switch (srcFormat) {
- case PIX_FMT_YUYV422 :
- case PIX_FMT_GRAY8A :
- c->lumToYV12 = yuy2ToY_c; break;
- case PIX_FMT_UYVY422 :
- c->lumToYV12 = uyvyToY_c; break;
- case PIX_FMT_BGR24 : c->lumToYV12 = bgr24ToY_c; break;
- case PIX_FMT_BGR565LE : c->lumToYV12 = bgr16leToY_c; break;
- case PIX_FMT_BGR565BE : c->lumToYV12 = bgr16beToY_c; break;
- case PIX_FMT_BGR555LE : c->lumToYV12 = bgr15leToY_c; break;
- case PIX_FMT_BGR555BE : c->lumToYV12 = bgr15beToY_c; break;
- case PIX_FMT_RGB24 : c->lumToYV12 = rgb24ToY_c; break;
- case PIX_FMT_RGB565LE : c->lumToYV12 = rgb16leToY_c; break;
- case PIX_FMT_RGB565BE : c->lumToYV12 = rgb16beToY_c; break;
- case PIX_FMT_RGB555LE : c->lumToYV12 = rgb15leToY_c; break;
- case PIX_FMT_RGB555BE : c->lumToYV12 = rgb15beToY_c; break;
- case PIX_FMT_RGB8 :
- case PIX_FMT_BGR8 :
- case PIX_FMT_PAL8 :
- case PIX_FMT_BGR4_BYTE:
- case PIX_FMT_RGB4_BYTE: c->lumToYV12 = palToY_c; break;
- case PIX_FMT_MONOBLACK: c->lumToYV12 = monoblack2Y_c; break;
- case PIX_FMT_MONOWHITE: c->lumToYV12 = monowhite2Y_c; break;
- case PIX_FMT_RGB32 : c->lumToYV12 = bgr32ToY_c; break;
- case PIX_FMT_RGB32_1: c->lumToYV12 = bgr321ToY_c; break;
- case PIX_FMT_BGR32 : c->lumToYV12 = rgb32ToY_c; break;
- case PIX_FMT_BGR32_1: c->lumToYV12 = rgb321ToY_c; break;
- case PIX_FMT_RGB48BE: c->lumToYV12 = rgb48BEToY_c; break;
- case PIX_FMT_RGB48LE: c->lumToYV12 = rgb48LEToY_c; break;
- case PIX_FMT_BGR48BE: c->lumToYV12 = bgr48BEToY_c; break;
- case PIX_FMT_BGR48LE: c->lumToYV12 = bgr48LEToY_c; break;
- }
- if (c->alpPixBuf) {
- switch (srcFormat) {
- case PIX_FMT_BGRA:
- case PIX_FMT_RGBA: c->alpToYV12 = rgbaToA_c; break;
- case PIX_FMT_ABGR:
- case PIX_FMT_ARGB: c->alpToYV12 = abgrToA_c; break;
- case PIX_FMT_Y400A: c->alpToYV12 = uyvyToY_c; break;
- case PIX_FMT_PAL8 : c->alpToYV12 = palToA_c; break;
- }
- }
- if(isAnyRGB(c->srcFormat) || c->srcFormat == PIX_FMT_PAL8)
- c->hScale16= hScale16_c;
- if (c->srcRange != c->dstRange && !isAnyRGB(c->dstFormat)) {
- if (c->srcRange) {
- c->lumConvertRange = lumRangeFromJpeg_c;
- c->chrConvertRange = chrRangeFromJpeg_c;
- } else {
- c->lumConvertRange = lumRangeToJpeg_c;
- c->chrConvertRange = chrRangeToJpeg_c;
- }
- }
- if (!(isGray(srcFormat) || isGray(c->dstFormat) ||
- srcFormat == PIX_FMT_MONOBLACK || srcFormat == PIX_FMT_MONOWHITE))
- c->needs_hcscale = 1;
- }
- SwsFunc ff_getSwsFunc(SwsContext *c)
- {
- sws_init_swScale_c(c);
- if (HAVE_MMX)
- ff_sws_init_swScale_mmx(c);
- if (HAVE_ALTIVEC)
- ff_sws_init_swScale_altivec(c);
- return swScale;
- }
|