Skip to content

Issues: openxla/xla

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

194 Open 185 Closed
194 Open 185 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Explicilty interface for AcquireExternalReference and WaitExternalReference
#19845 opened Nov 26, 2024 by yliu120
1
//xla/service/cpu:vectorized_reduce_with_no_vector_registers_test fails on Apple Silicon
#19830 opened Nov 26, 2024 by majnemer
@d0k
2
//xla/tests:complex_unary_op_test_cpu fails on macOS Apple Silicon
#19824 opened Nov 26, 2024 by majnemer
@pearu
3
Pallas/Triton gives incorrect result on Nvidia H100 GPU
#19780 opened Nov 25, 2024 by jaro-sevcik
1
CPU vs GPU inconsistency with at[].set ops under jit
#19716 opened Nov 22, 2024 by vfdev-5
@NaiyerRizz
[ROCm] failed to legalize operation 'math.exp' for exponential op with bf16 dtype
#19700 opened Nov 22, 2024 by hugomano
3
[GPU bug] Memcpy local p2p leads to numerical issues (mem access problems).
#19555 opened Nov 20, 2024 by yliu120
@Tixxx
1
stablehlo.cbrt crashes with complex numbers when StableHLO spec supports it
#19482 opened Nov 19, 2024 by mofeing
2
The tiles of some instructions in Triton fusion are not powers of 2. How did this issue arise?
#19478 opened Nov 19, 2024 by yulong-chen551
1
is set access to the previous GPUs at cudamallocasync_allocator.cc cause allocateRaw slower?
#19431 opened Nov 18, 2024 by zjjott
pjrt_c_api_cpu_plugin - MacOS amd and arm64 compilation
#19152 opened Nov 7, 2024 by dmarro89
4
[cpu] LLVM error during compilation of bf16 convert on aarch64 CPU Related to XLA on CPU
#19105 opened Nov 6, 2024 by hawkinsp
@ezhulenev @penpornk
1
PJRT build failed with absl on x86
#18938 opened Nov 1, 2024 by willhuang1114
1
docs/tf2xla/tfcompile.md is outdated
#18748 opened Oct 25, 2024 by LucasChollet
XLA:CPU performance regression with the min alignment changed from 16 to 64
#18611 opened Oct 22, 2024 by snadampal
@Adam-Banas
5
jax_cpu_enable_async_dispatch is degrading the inference performance on x86 and arm64 cpu backend
#18608 opened Oct 22, 2024 by snadampal
How to run onednn convolution test on CPU
#18586 opened Oct 22, 2024 by akote123
Unexpected slow-down with @jit on simple functions that only use element-wise operations and jnp.roll() on CPUs
#18478 opened Oct 18, 2024 by pmocz
@NaiyerRizz
Unexpected speedup from wrapping function call in trivial jax.lax.cond statement
#18440 opened Oct 17, 2024 by cgiovanetti
1
Deserialization of executables fails on non-zero ranks when deserializing single-device executable
#18286 opened Oct 14, 2024 by jaro-sevcik
3
1
error: call of overloaded 'TileAssignment(<brace-enclosed initializer list>)' is ambiguous with gcc 10
#18140 opened Oct 10, 2024 by elistevens
1
Inadequate memory consumption when using HSDP without gradient accumulation
#18090 opened Oct 9, 2024 by qGentry
2
SPMD: broadcast of replicated tensor is mark sharded
#17913 opened Oct 4, 2024 by fhaolinaws
@NaiyerRizz
1
[XLA:GPU] Regression in FP8 matmul scaling fusion
#17887 opened Oct 3, 2024 by balancap
@mooskagh
AVX512 quantization (cast from float to uint8) returns wrong results
#17800 opened Oct 1, 2024 by Flamefire
ProTip! Follow long discussions with comments:>50.