Welcome to mirror list, hosted at ThFree Co, Russian Federation.

gitlab.xiph.org/xiph/opus.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2023-12-15use opus_(re)alloc and opus_free for dnn and DRED related functionsopus-ng-fix-allocMichael Klingbeil
2023-12-14handle extensions in opus_repacketizer_out_range_implMichael Klingbeil
2023-12-06add extensions of the first frame of a multiframe packetMichael Klingbeil
2023-12-06Fix RESYNTH bit rotJean-Marc Valin
2023-11-30use vec_avx.h for MSVC buildsMichael Klingbeil
2023-11-30don't redefine _mm_loadu_si32 on MSVCMichael Klingbeil
2023-11-30Defining __SSEx__ macros when needed for MSVCJean-Marc Valin
2023-11-29fix autogen.bat model downloadMichael Klingbeil
2023-11-29Add a script to shrink the DNN modelsJean-Marc Valin
Removes float debug weights, as well as useless spaces
2023-11-29Fix Windows pathJean-Marc Valin
2023-11-29Fix model download path for windowsJean-Marc Valin
2023-11-29Opus github ci filesJean-Marc Valin
Use OPUS_DRED instead of NEURAL_FEC
2023-11-29Add dotprod support to mesonJean-Marc Valin
Also default to disabling dnn float debugging
2023-11-29Trying to fix/update meson buildJean-Marc Valin
Still don't quite know what I'm doing
2023-11-28Oops, fix the fixed-point buildJean-Marc Valin
2023-11-28Trying to use fma instructions when possibleJean-Marc Valin
Compilers sometimes replace vmlaq*() with fmul+fadd instead of fmla. Trying to use vfmaq*() instead when possible.
2023-11-28FARGAN model updateJean-Marc Valin
Finished adversarial training on 800k model. Also, move weights to a new location.
2023-11-28Fixes for ARMv7/AArch32Jean-Marc Valin
1) Enable asm/intrinsics even for floating-point 2) Make sure ARMv8 asimd enables EDSP/MEDIA/Neon 3) Add dotp architecture to rtcd table since AArch *can* have dotp
2023-11-28Enabling DNN optimizations for ARMv7Jean-Marc Valin
Adds RTCD tables for compute_activation() and compute_conv2d()
2023-11-28Only force auto-vectorization for GCC >= 5.1Jean-Marc Valin
2023-11-28Force vectorization for DNN primitivesJean-Marc Valin
Avoids having to write intrinsics for simple loops
2023-11-27Enable floating-point approximations by defaultJean-Marc Valin
Enabling only on platforms that have been tested just in case we run into a non-IEEE754 platform where they would break.
2023-11-27Fix ARMv7 optimizations for DNN codeJean-Marc Valin
2023-11-26First step towards DNN optimization for ARMv7 NeonJean-Marc Valin
Still missing some intrinsics
2023-11-26Fix potential read out of bounds in farganJean-Marc Valin
2023-11-25Adding dotprod instruction to ARM rtcdJean-Marc Valin
Used for DNN matrix multiplies
2023-11-25Speed up cross-correlation normalizationJean-Marc Valin
2023-11-25Use arch-specific celt_inner_prod() for featuresJean-Marc Valin
2023-11-25Optimize biquad() to reduce dependency chainsJean-Marc Valin
2023-11-24Remove process_single_frame()Jean-Marc Valin
Code moved to compute_frame_features()
2023-11-24Remove feature writing (fwrite()) from libopusJean-Marc Valin
2023-11-22Using the same condition for enabling rtcdJean-Marc Valin
for cmake, force PRESEUME_SSE4_1 on PRESUME_AVX2
2023-11-22Trying to fix CMake buildJean-Marc Valin
aka banging on it until it builds on my machine. Further improvements welcome
2023-11-21Add rtcd for silk_inner_product_FLP()Jean-Marc Valin
2023-11-21Start enabling AVX2 silk_inner_product_FLP()Jean-Marc Valin
Not yet with rtcd
2023-11-21Avoids AVX2 optimizations being disabledJean-Marc Valin
2023-11-21Use SILK VBR when using CBR with DREDJean-Marc Valin
DRED will absorb the bitrate variation
2023-11-21Misc fixes on previous patchJean-Marc Valin
Fixes warnings, undefined behaviour, and check-asm failure
2023-11-21Optimize NSQ_del_dec() for AVX2Victor Ding
The optimization is bit-exact with C function. This optimization speeds up SILK encoder (floating point) as following: AMD Zen: Complexity 0-5 : 0% Complexity 6-7 : 3 - 7% Complexity 8-10: 8 - 15% Intel Skylake: Complexity 0-5 : 0% Complexity 6-7 : 14 - 18% Complexity 8-10: 17 - 22% Adapted by Jean-Marc Valin
2023-11-21AVX2 version of silk_inner_product_FLP()Jean-Marc Valin
Not hooked up
2023-11-21Remove AVX pitch code for fixed-pointJean-Marc Valin
2023-11-21Speeding up transient_analysis()Jean-Marc Valin
Reducing dependency chains
2023-11-20Make sure weights files are marked as modifiedJean-Marc Valin
2023-11-18Speed up silk_warped_autocorrelation_FLP()Jean-Marc Valin
Reducing the dependency chain between tmp1 and tmp2 at the cost of an extra multiply.
2023-11-18Add rtcd support for celt_pitch_xcorr_avx2()Jean-Marc Valin
2023-11-18Fix non-RTCD case when SSE is not assumed presentJean-Marc Valin
Should never occur on amd64, but it could on 32-bit x86
2023-11-18Use celt_pitch_xcorr_avx2() when guaranteedJean-Marc Valin
No RTCD yet
2023-11-17Adding RTCD for compute_conv2d()Jean-Marc Valin
2023-11-16FARGAN model updateJean-Marc Valin
2023-11-16Smaller version of farganJean-Marc Valin
800k parameters, 600 MFLOPS, with a receptive field of 3 feature vectors
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-03 04:03:36 +0300