uxlfoundation/oneDNN

Bad perf for matmul of ba tensors 13 days ago

help wanted platform:cpu-x64

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

Support for long integers (aka int64, s64, long)? 2 months ago

enhancement help wanted

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

unify the data type description in the docs. mixed description like `s8` and `int8` is NOT good. 2 months ago

help wanted documentation

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

Test test_bfloat16_plus_float fails on s390x 5 months ago

help wanted sighting platform:cpu-s390x

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

Large kernels can generate xbyak_aarch64 exceptions 10 months ago

help wanted sighting platform:cpu-aarch64

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

BWDW JIT 256 error reproducible through gtests, but not through benchdnn 10 months ago

help wanted sighting platform:cpu-aarch64 component:tests

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

aarch64: brgconv segfaults with log post-op for some shapes 10 months ago

bug help wanted platform:cpu-aarch64

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

Temporary fix nightly crash for large l_pad failing test 11 months ago

bug help wanted platform:cpu-aarch64

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

[PPC64] '-mmma' requiring instructions not guarded by __MMA__ 11 months ago

bug help wanted platform:cpu-ppc64

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

[RV64] test_benchdnn_modeC_matmul_ci_cpu fails with dropout attribute Description 11 months ago

bug help wanted platform:cpu-rv64

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

How to switch different doc version? 12 months ago

enhancement help wanted documentation

uxlfoundation/oneDNN

4.0K

oneAPI Deep Neural Network Library (oneDNN)

C++

#aarch64#amx#avx512#bfloat16#cpp#deep-learning#deep-neural-networks#library#oneapi#onednn#openmp#performance#sycl#tbb#vnni#x64#x86-64#xe-architecture

Open Issues Need Help