Skip to content

Pull requests: ROCm/pytorch

New pull request New
42 Open 1,688 Closed
42 Open 1,688 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Build Pytorch extensions with amdclang++
#1961 opened Mar 14, 2025 by akashveramd Loading…
1
[release/2.5] [Navi] fix test_cublas_addmm_size_10000_cuda_float16, skip test_pointwise_op_fusion_post_grad
#1957 opened Mar 13, 2025 by dnikolaev-amd • Draft
1
[release/2.6] Add gfx120x in supported archs for hipBLASLt
#1955 opened Mar 13, 2025 by stsokolo Loading…
2
[release/2.5] Add gfx110x and gfx120x in supported archs for hipBLASLt
#1954 opened Mar 13, 2025 by stsokolo Loading…
2
Fix TORCH_CHECK for head dimensions
#1953 opened Mar 11, 2025 by xinyazhang Loading…
2
testing: correct PLATFORM_SUPPORTS_FLASH_ATTENTION/MEM_EFF_ATTENTION
#1951 opened Mar 10, 2025 by xinyazhang Loading…
2
Revert "[ROCm] Improvements to non-vectorized elementwise kernels (#1…
#1944 opened Mar 6, 2025 by BLOrange-AMD Loading…
4
Support heterogeneous tensor types for vectorized elementwise kernel.
#1940 opened Mar 3, 2025 by carlobertolli Loading…
15
[Release/2.6] Pin requirements
#1928 opened Feb 26, 2025 by ethanwee1 Loading…
10
Enable input vectorization in ewk for input tensors with heterogeneou…
#1906 opened Feb 15, 2025 by carlobertolli Loading…
18
Enable load-compute-store interleaving for unrolled elementwise kernel.
#1886 opened Feb 6, 2025 by carlobertolli • Draft
21
Diable L2_cahce_size display on ROCm platforms
#1876 opened Jan 31, 2025 by tcgu-amd Loading…
21
[Do NOT MERGE] [release/2.5] Enable tf32 testing on test_nn
#1859 opened Jan 27, 2025 by jagadish-amd Loading…
28
[ROCm] Eliminate the need for divisions in layernorm for default vector size.
#1850 opened Jan 22, 2025 by doru1004 Loading…
8
[AUTOGENERATED] [rocm6.3_internal_testing] remove xfail from 'batch_norm_with_update'
#1825 opened Jan 9, 2025 by rocm-mici • Draft
4
[ROCm][WIP] Improve performance of casted elementwise add operations
#1805 opened Dec 20, 2024 by doru1004 Loading…
7
Wrap vec size 8 with USE_ROCM
#1795 opened Dec 16, 2024 by oraluben Loading…
10
[AUTOGENERATED] [release/2.5] Cherry-pick PR-1767
#1794 opened Dec 13, 2024 by rocm-mici • Draft
44
Improve performance of reduce sum for 3D shapes
#1785 opened Dec 12, 2024 by doru1004 Loading…
17
[release/2.5] AMDSMI/layernorm cherry picks
#1764 opened Dec 4, 2024 by jataylo Loading…
44
[release/2.5] Fixed string comparison in test_cpp_wrapper_hipify
#1760 opened Nov 29, 2024 by iupaikov-amd Loading…
45
[release/2.5] Enabled force_shape_pad for test_pad_mm and test_slice_mm_bandwidth_computation
#1755 opened Nov 28, 2024 by iupaikov-amd Loading…
47
[ROCm] Improve softmax performance.
#1740 opened Nov 21, 2024 by doru1004 Loading…
40
[release/2.5] Skipped some inductor tests for no hipcc rocm environments
#1697 opened Nov 13, 2024 by iupaikov-amd • Draft
43
MIOpen bf16 batchnorm enablement
#1665 opened Oct 31, 2024 by jithunnair-amd • Draft
13
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.