summaryrefslogtreecommitdiffstats
path: root/libavcodec/aarch64
Commit message (Collapse)AuthorAgeFilesLines
* avcodec/dca: add new decoder based on libdcadecfoo862016-01-311-2/+2
|
* avcodec/dca: remove old decoderfoo862016-01-313-152/+2
| | | | | Remove all files and functions which are not going to be reused, and disable all functions and FATE tests temporarily which will be.
* avcodec/synth_filter: split off remaining code from dcadec filesJames Almer2016-01-253-22/+49
| | | | Signed-off-by: James Almer <jamrial@gmail.com>
* Merge commit '2008f76054906e9ff6bf744800af0e5a5bfe61be'Hendrik Leppkes2016-01-022-67/+0
|\ | | | | | | | | | | | | * commit '2008f76054906e9ff6bf744800af0e5a5bfe61be': dca: remove unused decode_hf function and quant_d tables Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
| * dca: remove unused decode_hf function and quant_d tablesAlexandra Hájková2015-12-242-67/+0
| | | | | | | | | | They were superseded with their integer equivalents. Rename integer decode_hf to decode_hf.
| * arm64: fix inverted register order in transpose_4x4HJanne Grunau2015-12-212-4/+4
| | | | | | | | | | | | Fix related register order issue in ff_h264_idct_add_neon. Found-by: zjh8890 <243186085@qq.com>
* | Merge commit 'a0fc780a2093784e8664f88205ee1b215e109cee'Hendrik Leppkes2016-01-023-0/+121
|\ \ | |/ | | | | | | | | | | * commit 'a0fc780a2093784e8664f88205ee1b215e109cee': arm64: int32_to_float_fmul neon asm Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
| * arm64: int32_to_float_fmul neon asmJanne Grunau2015-12-143-0/+121
| | | | | | | | | | | | | | | | | | | | 3% faster dts decoding on a cortex-a57. cortex-a57 cortex-a53 int32_to_float_fmul_array8_c: 1270.9 4475.6 int32_to_float_fmul_array8_neon: 328.6 569.2 int32_to_float_fmul_scalar_c: 928.5 4119.6 int32_to_float_fmul_scalar_neon: 309.1 524.1
* | Merge commit '705f5e5e155f6f280a360af220fc5b30cfcee702'Hendrik Leppkes2016-01-024-1/+140
|\ \ | |/ | | | | | | | | | | * commit '705f5e5e155f6f280a360af220fc5b30cfcee702': arm64: port synth_filter_float_neon from arm Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
| * arm64: port synth_filter_float_neon from armJanne Grunau2015-12-144-1/+140
| | | | | | | | | | | | | | | | | | | | | | | | | | | | ~25% faster dts decoding overall. The checkasm CPU cycles numbers are not that useful since synth_filter_float() calls FFTContext.imdct_half(). cortex-a57 cortex-a53 synth_filter_float_c: 1866.2 3490.9 synth_filter_float_neon: 915.0 1531.5 With fftc.imdct_half forced to imdct_half_neon: cortex-a57 cortex-a53 synth_filter_float_c: 1718.4 3025.3 synth_filter_float_neon: 926.2 1530.1
* | Merge commit 'c33c1fa8af2b2e82418a06901b6ad17b3d61b73e'Hendrik Leppkes2016-01-023-0/+222
|\ \ | |/ | | | | | | | | | | * commit 'c33c1fa8af2b2e82418a06901b6ad17b3d61b73e': arm64: convert dcadsp neon asm from arm Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
| * arm64: convert dcadsp neon asm from armJanne Grunau2015-12-143-0/+222
| | | | | | | | | | | | | | | | | | | | | | | | ~2% faster dts decoding overall. cortex-a57 cortex-a53 dca_decode_hf_c: 474.8 1659.9 dca_decode_hf_neon: 225.2 301.1 dca_lfe_fir0_c: 913.2 1537.7 dca_lfe_fir0_neon: 286.8 451.9 dca_lfe_fir1_c: 848.7 1711.5 dca_lfe_fir1_neon: 387.1 506.4
* | avcodec/arm64: fix inverted register order in transpose_4x4HJanne Grunau2015-12-192-4/+4
| | | | | | | | | | | | | | | | Fix related register order issue in ff_h264_idct_add_neon. Found-by: zjh8890 <243186085@qq.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* | Revert "avcodec/aarch64/neon.S: Update neon.s for transpose_4x4H"Michael Niedermayer2015-12-171-2/+2
| | | | | | | | | | | | The change was not correct and broke H264 This reverts commit cd83f899c94f691b045697d12efa21f83eb2329f.
* | avcodec/aarch64/neon.S: Update neon.s for transpose_4x4Hzjh88902015-12-121-2/+2
| | | | | | | | | | The transpose_4x4H is wrong which cost me much time to find this bug. The orders of r2 and r3 are wrong, this bug waste me much time while I make aarch64 arm instruction which used the function.
* | Merge commit 'f56d8d8dd72b1ab52aa814c5a0fccabf8040ef68'Michael Niedermayer2015-07-213-0/+456
|\ \ | |/ | | | | | | | | | | | | | | | | * commit 'f56d8d8dd72b1ab52aa814c5a0fccabf8040ef68': h264: aarch64: intra prediction optimisations Conflicts: libavcodec/h264pred.c Merged-by: Michael Niedermayer <michael@niedermayer.cc>
| * h264: aarch64: intra prediction optimisationsJanne Grunau2015-07-203-0/+456
| |
| * arm64: constify src in h264qpel dsp function definitionsJanne Grunau2015-06-241-64/+64
| |
* | Merge commit '3d5d46233cd81f78138a6d7418d480af04d3f6c8'Michael Niedermayer2015-02-023-11/+12
|\ \ | |/ | | | | | | | | | | | | | | | | | | * commit '3d5d46233cd81f78138a6d7418d480af04d3f6c8': opus: Factor out imdct15 into a standalone component Conflicts: configure libavcodec/opus_celt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * opus: Factor out imdct15 into a standalone componentDiego Biurrun2015-02-023-11/+12
| | | | | | | | It will be reused by the AAC decoder.
* | lavc/aarch64: Do not use the neon horizontal chroma loop filter for H.264 4:2:2.Carl Eugen Hoyos2015-01-311-0/+1
| |
* | Merge commit '780cd20b00a69e26bbfffbb8eec16fbe999ea793'Michael Niedermayer2014-12-092-28/+27
|\ \ | |/ | | | | | | | | | | * commit '780cd20b00a69e26bbfffbb8eec16fbe999ea793': aarch64: Use .data.rel.ro for const data with relocations Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: Use .data.rel.ro for const data with relocationsMartin Storsjö2014-12-092-28/+27
| | | | | | | | | | | | | | This reverts commit c00365b46d464ce47716315c1801818d811bdb9a in addition to using a different section. Signed-off-by: Martin Storsjö <martin@martin.st>
* | Merge commit 'c00365b46d464ce47716315c1801818d811bdb9a'Michael Niedermayer2014-11-162-25/+26
|\ \ | |/ | | | | | | | | | | * commit 'c00365b46d464ce47716315c1801818d811bdb9a': aarch64: Make the function pointer tables position independent Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: Make the function pointer tables position independentMartin Storsjö2014-11-162-25/+26
| | | | | | | | | | | | | | This allows running the code on android, where 64 bit binaries with text relocations aren't allowed to be loaded. Signed-off-by: Martin Storsjö <martin@martin.st>
* | avcodec/aarch64/h264qpel_init_aarch64: mark src as constMichael Niedermayer2014-08-301-64/+64
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | Merge commit 'ac6b95dbc0b53b3ea461bd5e5e7f7f31d2983733'Michael Niedermayer2014-08-043-6/+6
|\ \ | |/ | | | | | | | | | | * commit 'ac6b95dbc0b53b3ea461bd5e5e7f7f31d2983733': aarch64: add ',' between assembler macro arguments where missing Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: add ',' between assembler macro arguments where missingJanne Grunau2014-08-043-6/+6
| | | | | | | | | | | | | | llvm's integrated assembler does not accept spaces as macro argument delimiter when targeting darwin. Using a explicit delimiter is a good idea in principle since it makes case like 'macro 4 -2' vs 'macro 4 - 2' clear.
* | Merge commit 'f23d26a6864128001b03876b0b92fffe131f2060'Michael Niedermayer2014-06-231-4/+55
|\ \ | |/ | | | | | | | | | | * commit 'f23d26a6864128001b03876b0b92fffe131f2060': h264: avoid using uninitialized memory in NEON chroma mc Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * h264: avoid using uninitialized memory in NEON chroma mcJanne Grunau2014-06-231-4/+55
| | | | | | | | | | Adapt commit 982b596ea6640bfe218a31f6c3fc542d9fe61c31 for the arm and aarch64 NEON asm. 5-10% faster on Cortex-A9.
* | Merge commit 'd3f5b94762fb803c0f3b29f9ad6c5eaa813998ba'Michael Niedermayer2014-05-154-0/+724
|\ \ | |/ | | | | | | | | | | * commit 'd3f5b94762fb803c0f3b29f9ad6c5eaa813998ba': aarch64: opus NEON iMDCT and FFT Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: opus NEON iMDCT and FFTJanne Grunau2014-05-154-0/+724
| | | | | | | | | | Opus celt decoding 11% faster and the iMDCT over 2.5 times faster on Apple's A7.
* | Merge commit '9aa4592076d4dbb29d1198b0e258f9f85c0c00b5'Michael Niedermayer2014-05-131-1/+1
|\ \ | |/ | | | | | | | | | | * commit '9aa4592076d4dbb29d1198b0e258f9f85c0c00b5': aarch64: assembler in clang-3.4 ignores the division by two Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: assembler in clang-3.4 ignores the division by twoJanne Grunau2014-05-131-1/+1
| | | | | | | | Values are positive powers of two, so just replace it with right shift.
* | Merge commit '3956a5e0ea46ed7e27ca888fe11c47986ad99261'Michael Niedermayer2014-04-223-0/+119
|\ \ | |/ | | | | | | | | | | * commit '3956a5e0ea46ed7e27ca888fe11c47986ad99261': aarch64: NEON vorbis_inverse_coupling Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: NEON vorbis_inverse_couplingJanne Grunau2014-04-223-0/+119
| | | | | | | | | | From the ARMv7 NEON version. 16 times faster as the C version, overall more than 12% faster vorbis decoding on Apple's A7.
* | Merge commit '8f9fe6ae3461ce270bce6b7083fda5ec314cdad4'Michael Niedermayer2014-04-223-0/+267
|\ \ | |/ | | | | | | | | | | * commit '8f9fe6ae3461ce270bce6b7083fda5ec314cdad4': aarch64: NEON fixed/floating point MPADSP apply_window Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: NEON fixed/floating point MPADSP apply_windowJanne Grunau2014-04-223-0/+267
| | | | | | | | | | 30%/25% (fixed/float) faster mp3 decoding on Apple's A7. The floating point decoder is approximately 7% faster.
* | Merge commit 'ee2bc5974fe64fd214f52574400ae01c85f4b855'Michael Niedermayer2014-04-223-0/+334
|\ \ | |/ | | | | | | | | | | * commit 'ee2bc5974fe64fd214f52574400ae01c85f4b855': aarch64: NEON float (i)MDCT Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: NEON float (i)MDCTJanne Grunau2014-04-223-0/+334
| | | | | | | | Approximately as fast as the ARM NEON version on Apple's A7.
* | Merge commit '650c4300d94aa9398ff1dd4f454bf39eaa285f62'Michael Niedermayer2014-04-223-0/+481
|\ \ | |/ | | | | | | | | | | * commit '650c4300d94aa9398ff1dd4f454bf39eaa285f62': aarch64: NEON float FFT Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: NEON float FFTJanne Grunau2014-04-223-0/+481
| | | | | | | | Approximately as fast as the ARM NEON version on Apple's A7.
* | Merge commit 'd3789eeeed3423bd1ca9dc40030a2f7a21ea5332'Michael Niedermayer2014-04-073-0/+63
|\ \ | |/ | | | | | | | | | | * commit 'd3789eeeed3423bd1ca9dc40030a2f7a21ea5332': aarch64: implement videodsp.prefetch Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: implement videodsp.prefetchJanne Grunau2014-04-063-0/+63
| | | | | | | | 8% faster h264 decoding on Apple A7.
* | Merge commit '0e083d7e43805db1a978cb57bfa25fda62e8ff18'Michael Niedermayer2014-03-201-0/+1
|\ \ | |/ | | | | | | | | | | | | | | | | | | * commit '0e083d7e43805db1a978cb57bfa25fda62e8ff18': build: Group general components separate from de/encoders in arch Makefiles Conflicts: libavcodec/arm/Makefile libavcodec/x86/Makefile Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * build: Group general components separate from de/encoders in arch MakefilesDiego Biurrun2014-03-201-0/+1
| | | | | | | | This is in line with how the top-level libavcodec Makefile is structured.
* | Merge commit 'dfe224f377be3e45758c69d881ca7874b82d647a'Michael Niedermayer2014-03-091-0/+104
|\ \ | |/ | | | | | | | | | | * commit 'dfe224f377be3e45758c69d881ca7874b82d647a': aarch64: get_cabac inline asm Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: get_cabac inline asmJanne Grunau2014-03-091-0/+104
| | | | | | | | | | | | | | Based on the x86 branchless get_cabac asm. get_cabac_noinline() gets approximately 20% faster (no cycle counts available) compared to clang from Xcode 5.1 beta5. More than 6% faster overall. A part of the overall speedup might be explained by additional inlining of get_cabac().
* | Merge commit '9c029f67ca82147ddfa83a1546ee1e109e11fbd4'Michael Niedermayer2014-02-201-8/+8
|\ \ | |/ | | | | | | | | | | * commit '9c029f67ca82147ddfa83a1546ee1e109e11fbd4': aarch64: use EXTERN_ASM consistently for exported symbols Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * aarch64: use EXTERN_ASM consistently for exported symbolsJanne Grunau2014-02-201-8/+8
| | | | | | | | Based on e3fec3f095ab5ea08ee662942d98526aaf5e3635 for arm.
OpenPOWER on IntegriCloud