summaryrefslogtreecommitdiffstats
path: root/libavcodec/cabac.h
Commit message (Expand)AuthorAgeFilesLines
...
* Fix PIC compilation, some defines were under #ifdef !PIC but usedBernhard Rosenkränzer2006-10-191-3/+3
* unused variableMichael Niedermayer2006-10-191-1/+0
* optimize sign decoding code in decode_residual()Michael Niedermayer2006-10-191-0/+81
* Kill a warning with MSVCJindřich Makovička2006-10-181-2/+3
* x86 asm version of the decode significance loop (not 8x8) of decode_residual(...Michael Niedermayer2006-10-171-0/+165
* cosmetic (%%eax->%0)Michael Niedermayer2006-10-171-10/+10
* Fix crash with illegal instruction, cmov is available on 686 and later only.Diego Biurrun2006-10-161-1/+1
* Expand some #endif comments.Diego Biurrun2006-10-161-2/+2
* fix !CMOV_IS_FAST case (iam not really happy with the fix but i didnt come up...Michael Niedermayer2006-10-161-1/+2
* 10lMichael Niedermayer2006-10-151-1/+1
* this code will not work with PIC as it needs 7 registers and gcc doesnt suppo...Michael Niedermayer2006-10-151-1/+1
* shift CABACContext.range right, this reduces the number of shifts needed in g...Michael Niedermayer2006-10-151-46/+48
* dehack *ps_state indexing in the branchless decoderMichael Niedermayer2006-10-131-2/+3
* add "memory" to the clobber list we change memory so we need it, this also fi...Michael Niedermayer2006-10-121-2/+2
* prevent "mb level" get_cabac() calls from being inlined (3% faster decode_mb_...Michael Niedermayer2006-10-121-1/+9
* adds some useful comments after some of the #else, #elseif,Guillaume Poirier2006-10-121-8/+8
* Rename ABS macro to FFABS.Diego Biurrun2006-10-111-1/+1
* slightly faster on P3 slightly slower on athlon and probably faster on P4Michael Niedermayer2006-10-111-0/+10
* moving lps state transition code a little up in the branched asm code (1% fas...Michael Niedermayer2006-10-111-2/+2
* write cabac low and range variables as early as possible to prevent stalls fr...Michael Niedermayer2006-10-111-3/+4
* use ecx instead of cl (no speed change on P3 but might avoid partial register...Michael Niedermayer2006-10-111-4/+4
* make state transition tables global as they are constant and the code is slig...Michael Niedermayer2006-10-111-19/+14
* 10lMichael Niedermayer2006-10-111-2/+2
* make lps_range a global table its constant anyway (saves 1 addition for acces...Michael Niedermayer2006-10-111-15/+11
* enable CMOV_IS_FAST as its faster or equal speed on every cpu (duron, athlon,...Michael Niedermayer2006-10-111-0/+1
* BRANCHLESS_CABAD --> BRANCHLESS_CABAC_DECODERDiego Biurrun2006-10-101-3/+3
* moving another bit&1 out, this is as fast as with it in there, but it makes m...Michael Niedermayer2006-10-101-2/+1
* move the &1 out of the asm so gcc can optimize it away in inlined cases (yes ...Michael Niedermayer2006-10-101-1/+1
* replace a few and/sub/... by cmovMichael Niedermayer2006-10-101-1/+10
* reading 8bit mem into a 8bit register needs 2 uops on P4, 8bit->32bit with ze...Michael Niedermayer2006-10-091-7/+7
* on the P4 inc needs twice as much time a addMichael Niedermayer2006-10-091-1/+1
* 10lMichael Niedermayer2006-10-091-4/+0
* reverse remainder of the failed attempt to optimize *state=c->mps_state[s]Michael Niedermayer2006-10-091-8/+8
* x86 branchless cabac decoderMichael Niedermayer2006-10-091-1/+68
* optimize branchless C CABAC decoderMichael Niedermayer2006-10-091-3/+4
* move outcommented START/STOP_TIMER to a hopefully better place for benchmarki...Michael Niedermayer2006-10-091-2/+0
* drop failed attempt to optimize *state= c->mps_state[s];Michael Niedermayer2006-10-091-17/+1
* 10l bugfix for some disabled codeMichael Niedermayer2006-10-091-2/+2
* first try of a handwritten get_cabac() for x86, this is 10-20% faster on P3 d...Michael Niedermayer2006-10-091-0/+85
* remove bytestream_end checks, seems to work fine without them and the bitstre...Michael Niedermayer2006-10-091-2/+0
* decrease ff_h264_norm_shift[] sizeMichael Niedermayer2006-10-091-3/+3
* cleanupMichael Niedermayer2006-10-081-4/+0
* branchless renormalization (1% faster get_cabac) old branchless renormalizati...Michael Niedermayer2006-10-081-10/+11
* 5% faster get_cabac()Michael Niedermayer2006-10-081-4/+20
* disable benchmarking codeMichael Niedermayer2006-10-071-3/+3
* renorm_cabac_decoder_once START/STOP_TIMER scores for athlonMichael Niedermayer2006-10-071-6/+6
* refill cabac variables in 16bit steps, 3% faster get_cabac()Michael Niedermayer2006-10-071-6/+6
* Change license headers to say 'FFmpeg' instead of 'this program/this library'Diego Biurrun2006-10-071-4/+6
* () 10lMichael Niedermayer2006-10-071-1/+1
* several x86 renorm_cabac_decoder_once optimizationsMichael Niedermayer2006-10-071-1/+63
OpenPOWER on IntegriCloud