Skip to content

Issues: intel/torch-xpu-ops

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

109 Open 253 Closed
109 Open 253 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

Investigate whether pad mm is useful on XPU enhancement New feature or request
#1129 opened Nov 29, 2024 by jianyizh
@jianyizh
add xpu support to align with cuda for inductor sdpa fusion enhancement New feature or request
#1128 opened Nov 29, 2024 by jianyizh
@jianyizh
feature gap in sparsity
#1125 opened Nov 28, 2024 by daisyden PT2.7
@xytintel
Precision issues depend on Compiler
#1124 opened Nov 28, 2024 by daisyden 2.8
[Driver] [Ubuntu 24.10] Command Hang after driver installation bug Something isn't working client os: Ubuntu
#1122 opened Nov 27, 2024 by Stonepia
SYCL: Runtime: Roll back to original usage of sycl::get_kernel_bundle
#1121 opened Nov 27, 2024 by fengyuan14 2.8
@fengyuan14
add scaled_mm on XPU
#1120 opened Nov 25, 2024 by yuchengliu1 PT2.7
@yuchengliu1
Abort (core dumped) on Triton tutorial with nightly wheels
#1113 opened Nov 21, 2024 by pbchekin PT2.6
@ratnampa
3
Add RNN related op on XPU xpu-op
#1109 opened Nov 21, 2024 by jianyizh
@xytintel
Basic operator performance optimization on PVC performance
#1108 opened Nov 21, 2024 by xytintel
3 of 4 tasks
@xytintel
[Client] DNNL does not support bf16/f16 backward on the platform with avx2_vnni_2 client
#1094 opened Nov 18, 2024 by Stonepia
@Stonepia
RMSNorm Forward Segmentation fault (core dumped) client triaged
#1092 opened Nov 18, 2024 by faaany
@Stonepia
7
shapes are not equal with FakeTensor
#1078 opened Nov 13, 2024 by daisyden
@chunhuanMeng
2
[Performance] Low performance of vectorized kernels loops_kernel Loops Kernel Backbone
#1077 opened Nov 12, 2024 by cfgfung
@cfgfung
4
AssertionError: "Simulate error" does not match "grad can be implicitly created only for scalar outputs" in test_reentrant_parent_error_on_cpu_xpu ut_triage
#1071 opened Nov 12, 2024 by PenghuiCheng PT2.6
@guangyey
2
Explore grid_sample_2d fp16/bf16 accuracy error
#1061 opened Nov 7, 2024 by xytintel
4 tasks done
@daisyden
SYCL RT: Using recommended shortcut API for kernel specific max work group size.
#1059 opened Nov 7, 2024 by fengyuan14 PT2.7
@majing921201
Clarify branch policy of torch-xpu-ops repo - what's viable/strict branch?
#1025 opened Oct 29, 2024 by dvrogozh
@EikanWang
1
APIs requiring xpu backend support for UT upstream
#1023 opened Oct 29, 2024 by daisyden
@riverliuintel
Performance: Host overhead: Severe host overhead in sycl::get_kernel_bundle. performance
#1016 opened Oct 23, 2024 by fengyuan14
@majing921201 @fengyuan14
KLDivLoss function in pytorch always fallsback to CPU
#1011 opened Oct 23, 2024 by jgtong
@jgtong
3
[PT2.6][UX]Torch initialize and verbose message are clear and comparable with PyTorch GPUs enhancement New feature or request feature
#1009 opened Oct 22, 2024 by riverliuintel PT2.7
[PT2.6] Pip install Torch and all PyPI dependency feature
#1007 opened Oct 22, 2024 by riverliuintel PT2.6 - Feature Freeze
@chuanqi129
1
[PT2.6] LLM customized kernels (GEMM INT4 serves for Torchao) feature
#1005 opened Oct 22, 2024 by riverliuintel PT2.7
@ZhiweiYan-96
[PT2.6][Torch.complie] Performance analysis and optimization feature performance
#1004 opened Oct 22, 2024 by riverliuintel PT2.6 - Feature Freeze
@retonym @jianyizh @weishi-deng
5
ProTip! Exclude everything labeled bug with -label:bug.