Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Avoid processing LPC coeffs beyond the given order in NEON optimizations | Felicia Lim | 2019-07-08 | 1 | -7/+15 |
| | |||||
* | Silence compiler warnings | Mark Harris | 2018-07-21 | 1 | -5/+5 |
| | | | | | | | clang -Wcast-align warnings with ambisonics enabled clang -Wnull-pointer-arithmetic warnings in test_opus_api.c gcc -Wimplicit-fallthrough warnings on arm msvc warning C4244 in celt_encoder.c with fixed point | ||||
* | Don't use MAY_HAVE_NEON in arm_silk_map.c. | Jonathan Lennox | 2017-06-02 | 1 | -16/+16 |
| | | | | | | | | | It's unnecessary, and isn't defined correctly on floating-point. This makes us correctly use Neon functions (in floating-point mode) on platforms where Neon is detected by RTCD. Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org> | ||||
* | Optimize silk_biquad_alt_stride2() for ARM NEON | Linfeng Zhang | 2017-05-17 | 3 | -0/+238 |
| | | | | | | | | The optimization is bit exact with C function. Change-Id: Ifb8f04b19f2d576e79ce5dcfa7e0fc374d71d6c8 Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca> | ||||
* | Optimize silk_warped_autocorrelation_FIX() for ARM NEON | Linfeng Zhang | 2017-04-13 | 1 | -0/+19 |
| | | | | | | | | | | | The optimization is bit exact with C function. This optimization speeds up fixed-point SILK encoder on NEON about 5% to 8%. (Tested on Acer Chromebook, ARMv7 Processor rev 3 (v7l).) Change-Id: I582f6f3585b7946149e16a2ad3084ebc0ae79a4f Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca> | ||||
* | Optimize silk_LPC_inverse_pred_gain() for ARM NEON | Linfeng Zhang | 2017-02-14 | 3 | -0/+348 |
| | | | | | | | | The optimization is bit exact with C function. Change-Id: Ib3bdc26a5a4ebe02e7f24be85104e8e9a2a9a738 Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca> | ||||
* | fix ARM build w/--disable-intrinsics --enable-asm | James Zern | 2017-02-04 | 1 | -1/+1 |
| | | | | | | | | | and rtcd disabled (CFLAGS=-mfpu=neon) broken since: cfdaf365 Optimize silk_NSQ_del_dec() for ARM NEON Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca> | ||||
* | Optimize silk_NSQ_del_dec() for ARM NEON | Linfeng Zhang | 2017-01-30 | 3 | -0/+1248 |
| | | | | | | | | | | | | | | | | The optimization is bit exact with C function. This optimization speeds up SILK encoder on NEON as following. Fixed-point: Complexity 0-5: 0% Complexity 6-7: 6% Complexity 8-9: 10% Complexity 10: 8% Got similar results on floating-point. Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca> | ||||
* | Avoids undefined behaviour in ARM-optimized code | Jean-Marc Valin | 2017-01-25 | 3 | -20/+35 |
| | | | | Casting to unsigned to avoid shifting negative values left. | ||||
* | Add Aarch64 intrinsics for saturated add/subtract. | Jonathan Lennox | 2016-07-06 | 1 | -0/+39 |
| | | | | Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org> | ||||
* | Add Neon intrinsics for Silk noise shape feedback loop. | Jonathan Lennox | 2016-07-06 | 3 | -0/+114 |
| | | | | Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org> | ||||
* | Add Neon intrinsics for Silk noise shape quantization. | Jonathan Lennox | 2016-07-06 | 2 | -0/+166 |
| | | | | Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org> | ||||
* | Replace "inline" with OPUS_INLINE. | Gregory Maxwell | 2013-10-28 | 4 | -19/+19 |
| | | | | | | | | | Newer versions of MSVC are unhappy with the strategy of the build environment redefining "inline" (even though they don't support the actual keyword). Instead we define OPUS_INLINE to the right thing in opus_defines.h. This is the same approach we use for restrict. | ||||
* | Move ARM asm into its own directories. | Timothy B. Terriberry | 2013-05-21 | 4 | -0/+424 |
| | |||||
* | Moved all SILK source code to the silk/ directory | Jean-Marc Valin | 2011-04-28 | 1 | -6/+0 |
| | |||||
* | Relying on SILK for the switching decisions | Jean-Marc Valin | 2011-04-28 | 1 | -5/+5 |
| | |||||
* | SILK makefile update | Jean-Marc Valin | 2011-04-27 | 1 | -5/+5 |
| | |||||
* | Changing the SILK bandwidth only when there's no speech | Jean-Marc Valin | 2011-04-27 | 1 | -29/+5 |
| | |||||
* | SILK update | Koen Vos | 2011-04-27 | 1 | -5/+29 |
| | |||||
* | Project files update | Jean-Marc Valin | 2011-03-18 | 1 | -8/+5 |
| | |||||
* | DTX fix | Jean-Marc Valin | 2011-03-17 | 1 | -5/+8 |
| | |||||
* | Support for glitchles mode switching | Jean-Marc Valin | 2011-03-07 | 1 | -5/+5 |
| | | | | | | Uses a 5ms redundant CELT frame embedded into the SILK or hybrid packet to handle the switching. It's still possible to use the PLC-based method when no redundant packet is included. | ||||
* | SILK bugfix | Jean-Marc Valin | 2011-03-03 | 1 | -5/+5 |
| | |||||
* | SILK update | Jean-Marc Valin | 2011-03-02 | 1 | -17/+5 |
| | |||||
* | SILK/CELT update | Jean-Marc Valin | 2011-02-28 | 1 | -10/+10 |
| | |||||
* | SILK update | Jean-Marc Valin | 2011-02-17 | 1 | -5/+17 |
| | |||||
* | fix make dist | Jean-Marc Valin | 2011-02-15 | 1 | -5/+5 |
| | |||||
* | missing files | Jean-Marc Valin | 2011-02-14 | 1 | -8/+5 |
| | |||||
* | SILK and CELT updates | Jean-Marc Valin | 2011-02-14 | 1 | -5/+8 |
| | |||||
* | Update for in-band FEC | Koen Vos | 2011-02-14 | 1 | -9/+5 |
| | |||||
* | Update Opus range coder due to CELT refactoring. | Timothy B. Terriberry | 2011-02-03 | 1 | -5/+9 |
| | | | | | The byte buffer is now part of the range coder struct itself, and rangeenc.c and rangedec.c have gone away. | ||||
* | Build fixes | Jean-Marc Valin | 2011-02-03 | 1 | -5/+5 |
| | |||||
* | Testing the range coder final state | Koen Vos | 2011-02-03 | 1 | -5/+5 |
| | |||||
* | SILK update | Jean-Marc Valin | 2011-02-02 | 1 | -5/+5 |
| | |||||
* | SILK update | Jean-Marc Valin | 2011-02-02 | 1 | -5/+5 |
| | |||||
* | SILK update (fixing segfault) and MSVS fix | Jean-Marc Valin | 2011-02-02 | 1 | -5/+5 |
| | |||||
* | SILK update | Jean-Marc Valin | 2011-02-02 | 1 | -5/+5 |
| | |||||
* | Disabling debug | Jean-Marc Valin | 2011-01-31 | 1 | -5/+5 |
| | |||||
* | Adding missing SILK files to the build | Jean-Marc Valin | 2010-11-13 | 1 | -5/+5 |
| | |||||
* | Updated to follow changes in SILK API | Jean-Marc Valin | 2010-11-12 | 1 | -5/+5 |
| | |||||
* | SILK build update | Jean-Marc Valin | 2010-07-08 | 1 | -5/+5 |
| | |||||
* | Build update | Jean-Marc Valin | 2010-07-08 | 1 | -5/+5 |
| | |||||
* | autotools-based build system | Jean-Marc Valin | 2010-07-08 | 1 | -5/+5 |
| | |||||
* | Adds a _ctl() control API to the encoder and decoder | Jean-Marc Valin | 2010-07-06 | 1 | -5/+5 |
| | |||||
* | Fixing a high bit-rate bug in SILK -- now scales well above 64 kb/s | Jean-Marc Valin | 2010-07-05 | 1 | -5/+5 |
| | |||||
* | silk update (ec_enc_done()) | Jean-Marc Valin | 2010-07-04 | 1 | -5/+5 |
| | |||||
* | SILK update (interface for using the CELT range coder) | Jean-Marc Valin | 2010-07-04 | 1 | -5/+5 |
| | |||||
* | gitignore files | Jean-Marc Valin | 2010-07-03 | 1 | -5/+5 |
| | |||||
* | SILK update | Jean-Marc Valin | 2010-06-30 | 1 | -5/+5 |
| | |||||
* | Updated SILK version | Jean-Marc Valin | 2010-06-30 | 1 | -4/+5 |
| |