Skip to content

[AUTOGENERATED] develop_IFU_20260116 #2911

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

Closed
pragupta wants to merge 1,985 commits into from
Closed

[AUTOGENERATED] develop_IFU_20260116 #2911

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1985 commits
Select commit Hold shift + click to select a range
ab67020
Revert "[Inductor] Add fma lowerings for addcmul ops (#170531)"
pytorchmergebot Jan 12, 2026
7363a1c
Revert "[coor-slicing] _select_split_tensor (#169551)"
pytorchmergebot Jan 12, 2026
732ad91
Revert "[coor-slicing] extract _get_mesh_tensor_from_full_mesh (#1695…
pytorchmergebot Jan 12, 2026
b9fefeb
Revert "Change _compute_local_shape_and_global_offset API to be backw…
pytorchmergebot Jan 12, 2026
01c52be
Revert "[coor-slicing] Factor out DeviceMesh._compute_coordinates_fro…
pytorchmergebot Jan 12, 2026
1779748
[coor-slicing] Factor out DeviceMesh._compute_coordinates_from_mesh (…
aorenste Jan 9, 2026
876ee1e
[coor-slicing] extract _get_mesh_tensor_from_full_mesh (#169550)
aorenste Jan 9, 2026
c8298bb
[coor-slicing] _select_split_tensor (#169551)
aorenste Jan 9, 2026
d2237eb
Change _compute_local_shape_and_global_offset API to be backward-comp…
aorenste Jan 11, 2026
9eb1538
Revert "Support mix BigInt and Number for MemoryViz (#170645)"
pytorchmergebot Jan 12, 2026
1c3e695
Revert "[xpu][feature] Introduce a new API torch.xpu._dumpp_snapshot …
pytorchmergebot Jan 12, 2026
3202274
Small doc suggestion for intermediate hooks (#172023)
ezyang Jan 8, 2026
c5e1ef7
Automatically wrap autograd.grad with allow_in_graph (#169493)
tugsbayasgalan Jan 12, 2026
f2c5f9f
Support tensor.backward in dynamo (#169415)
tugsbayasgalan Jan 12, 2026
1415441
Revert "Support tensor.backward in dynamo (#169415)"
pytorchmergebot Jan 12, 2026
bde2ea1
[BE][Ez]: Add slots to treespec dataclasses (#172172)
Skylion007 Jan 12, 2026
282cbb7
[CUDA][cuBLAS][cuBLASLt] Actually enable unified cuBLASLt workspace b…
eqy Jan 12, 2026
3392106
[MPS] Cholesky inverse (#172186)
Isalia20 Jan 12, 2026
1ba65ac
[MPS][BE] Add checks for MacOS-26 (#172229)
malfet Jan 12, 2026
361dbb5
[ONNX] Change warning to debug log for missing annotations (#172247)
justinchuby Jan 12, 2026
491d607
Don't treat `torch.use_deterministic_algorithms` as a context manager…
guilhermeleobas Jan 7, 2026
80fb791
[ROCm] Increase the timeout value for ops microbenchmark (#172158)
huydhn Jan 12, 2026
8f775e9
Fix `MaxUnpool` crash (#169359)
parsshar-RH Jan 12, 2026
b37f123
Skip device mesh device setup when using fake backend (#171830)
bobrenjc93 Jan 12, 2026
29bcc35
[BE]: Update frozen metautils dataclasses to use slots (#172173)
Skylion007 Jan 12, 2026
79b05ac
remove assert in library/cuda/ao (#170803)
albanD Jan 12, 2026
8333c5b
remove plain asserts in subclasses (#171778)
albanD Jan 12, 2026
5a08a3b
remove assert in nnapi backend (#172095)
albanD Jan 12, 2026
f31ece3
Remove assert in ao/pruning and testing internal (#172096)
albanD Jan 12, 2026
84c8798
Remove assert for export files (#172124)
albanD Jan 12, 2026
5dd5454
Remove assert for some hops (#172137)
albanD Jan 12, 2026
e67d44d
[BE] Remove libstdcxx conda-forge hack from Docker builds (#172260)
huydhn Jan 12, 2026
8e5f493
[DTensor] Automatically set release_notes label (#172121)
wconstab Jan 9, 2026
c9e8e31
[vllm hash update] update the pinned vllm hash (#172175)
pytorchupdatebot Jan 12, 2026
62347f0
Part 2: LocalTensor test flatten and unflatten roundtrip. (#170675)
skpark-rh Jan 12, 2026
efcf252
Add _disable_torch_fn_metadata_mode option to make_fx and aot_export_…
ezyang Jan 11, 2026
b81ce8f
Make invoke_subgraph work with FakeTensorMode without ShapeEnv (#172090)
ezyang Jan 11, 2026
7285782
[DTensor] reduce test size due to timeout (#172255)
zpcore Jan 12, 2026
aac2f11
Fix for test/distributed/test_device_mesh.py::TestDeviceMeshGetItem::…
aorenste Jan 11, 2026
2074acf
[dynamo] Fix property setter on nested MutableMapping objects (#172001)
vmoens Jan 12, 2026
74b6a0e
[BE][Ez]: Modernize symbolic shape dataclasses (#172115)
Skylion007 Jan 12, 2026
c6dd7ee
[ROCm][CI] Update AlbertForMaskedLM expected accuracy to pass (#172281)
ethanwee1 Jan 12, 2026
ee6e5f0
[ROCm] Optimize single-block TopK with warp-level compaction (#171940)
apakbin Jan 12, 2026
1cd056e
[MPS] Compile shaders using latest OS capabilities (#172230)
malfet Jan 12, 2026
1c83214
[SymmMem] Add multimem support for NCCL and NVSHMEM (#172185)
kwen2501 Jan 12, 2026
2eed603
[xpu][fix] Fix XPU CI failures introduce by community. (#171498)
etaf Jan 10, 2026
600999e
[Fix] Fix broken size mismatch check in __init__ (always False) (#171…
lingebeng Jan 12, 2026
f4c66d9
Adds support for complex parameter model in DataParallel (#170185)
arkadip-maitra Jan 13, 2026
b6cba46
docker build disable ucc cpp warning (#172279)
jeffdaily Jan 13, 2026
19301b3
[xpu][feature] Introduce a new API torch.xpu._dumpp_snapshot (#170186)
guangyey Jan 12, 2026
42b5803
Optimize GB_REGISTRY lint by deferring get_source_segment call (#172226)
ezyang Jan 12, 2026
3bdf66d
[CUDA][Green Context] Expose green context streams (#171116)
eqy Jan 13, 2026
7061deb
Support tensor.backward in dynamo (#169415)
tugsbayasgalan Jan 12, 2026
23e9a50
[Vulkan] Fix logic error in flip dim check (#170982)
lingebeng Jan 13, 2026
0b7204f
[precompile] e2e cross precompilation (#171600)
bobrenjc93 Jan 12, 2026
dbc19fe
[HOP][Print]Add testing for the dynamo and AOTAutograph graphs (#171767)
fxdawnn Jan 12, 2026
7668bc1
[Inductor][NVGEMM] Add support for kernels with workspaces (#172280)
NikhilAPatel Jan 12, 2026
15d46a5
Clean up existing pyrefly errors on main (#172251)
maggiemoss Jan 13, 2026
b1f19d8
Remove stale @skipIfTorchDynamo for closed issue #115653 (#171937)
FIM43-Redeye Jan 13, 2026
6bdd8c9
Skip `test_gds_fails_in_ci` when not run on CI (#169122)
Flamefire Jan 13, 2026
985a399
Support mix BigInt and Number for MemoryViz (#170645)
guangyey Jan 12, 2026
888a4a0
[ez] fix type annotation on RegionalOutputCode.__call__ (#172239)
bobrenjc93 Jan 13, 2026
c230252
Fix torch.nn.Buffer in compile (#172222)
parsshar-RH Jan 13, 2026
16859e0
[MPS] remove cholesky inverse redundant test (#172264)
Isalia20 Jan 13, 2026
4fb682a
[Misc][BE] Finish type coverage for `torch/_subclasses` and turn on s…
Lucaskabela Jan 13, 2026
5594090
Revert "Fix torch.nn.Buffer in compile (#172222)"
pytorchmergebot Jan 13, 2026
f22ffda
[pallas backend] Use TMA for automatic OOB masking on Mosaic GPU (#17…
oulgen Jan 13, 2026
4e5be54
[pallas backend] Remove Triton backend deadcode (#172307)
oulgen Jan 13, 2026
4a3dc6f
Revert "Clean up existing pyrefly errors on main (#172251)"
pytorchmergebot Jan 13, 2026
1aa99d6
Revert "[ROCm] SDPA fix mem fault when dropout is enabled (#154864)"
pytorchmergebot Jan 13, 2026
e5e6b08
Revert "[ROCm] Optimize single-block TopK with warp-level compaction …
pytorchmergebot Jan 13, 2026
4652482
Revert "[SymmMem] Add multimem support for NCCL and NVSHMEM (#172185)"
pytorchmergebot Jan 13, 2026
6075f32
[ROCm][CI] skip TestFlexAttentionCUDA test_autocast (#172290)
jeffdaily Jan 13, 2026
37a8c7d
Change NamedTupleVariable implementation to subclass UserDefinedTuple…
morrison-turnansky Jan 13, 2026
7440dfe
[Reland] Migrate FBGEMM GenAI usage to MSLK, add MSLK submodule (#172…
cthi Jan 12, 2026
41f2dc8
[Reland] FBGEMM_GENAI -> MSLK (#172224)
cthi Jan 12, 2026
4e43a0b
Fix constant folding crash on 0-d complex tensors (#172181)
aorenste Jan 12, 2026
36ae126
Add docs preview for cpp docs built with ciflow/nightly (doxygen-enab…
mikaylagawarecki Jan 12, 2026
830aba8
Fix stable cpp docs rendering (#171957)
mikaylagawarecki Jan 12, 2026
a22552f
Add Joe Spisak to Core maintainers list (#172355)
albanD Jan 13, 2026
1e7ceca
Part 3: Split LocalTensor rank and world tests. (#170814)
skpark-rh Jan 13, 2026
2ff314b
[nativert] Refactor LaunchParams to use target-specific subclasses (#…
dancgr Jan 13, 2026
baacaa9
Fixes #155822 test:remove slowtest tag from compare cpu (#170165)
yingluAMD Jan 13, 2026
179865b
Skip SAC ILP tests when pulp package is not installed (#171975)
subinz1 Jan 13, 2026
a5e311d
[BE]: Update cutlass submodule to 4.3.5 (#172117)
Skylion007 Jan 13, 2026
d05e0f0
[ROCm] Enable StaticTritonLauncher for ROCm (#166492)
chinmaydk99 Jan 13, 2026
f6bb15e
[dynamo] Add LazyConstantVariable (#169282)
williamwen42 Jan 13, 2026
30ed891
[DTensor] Handle out= ops in single-dim expander (#172276)
wconstab Jan 13, 2026
5cfc15c
[BE][Ez]: Misc enablement of slots on frozen dataclasses (#172346)
Skylion007 Jan 13, 2026
0b8e2b8
[Doc] Use the correct cap for referencing PyTorch (#171893)
spzala Jan 13, 2026
53ccb94
[DTensor] Fix _StridedShard to Replicate padding issue (#170914)
zpcore Jan 13, 2026
a09c340
[DTensor] Fix for incorrect Tensor Meta Population in `expand_to_full…
sanketpurandare Jan 13, 2026
4c2c83b
[vLLM] Retry vLLM tests with a delay (#172323)
huydhn Jan 14, 2026
4301818
[SymmMem] Back symm_mem.emtpy() with implicit pool (#172292)
kwen2501 Jan 13, 2026
c20ab72
[ROCm] remove gfx940 gfx941 (#172369)
lakhinderwalia Jan 14, 2026
643db17
[Inductor] Finely overlap template fusion with async_compile (#170565)
PaulZhang12 Jan 14, 2026
432eb39
ci: Use buildx directly (#172243)
seemethere Jan 12, 2026
9f5d6ec
[ROCm] add hipsparseSpSV and hipsparseSpSM support for triangular sol…
jsandham Jan 14, 2026
7b9af66
[xpu][feature] Enable BHSD layout and add deterministic check for SDP…
LuFinch Jan 14, 2026
b4b094f
[Inductor] Add fma lowerings for addcmul ops (#170531)
mlazos Jan 13, 2026
5cb1604
[Inductor] Ensure combo kernels respect FMA settings (#170532)
mlazos Jan 13, 2026
93a0a6c
[Perf] Remove redundant opt stream (#172191)
yewentao256 Jan 14, 2026
1f99adb
Re-enable test_sampled_addmm_zero_sized (#172381)
pearu Jan 13, 2026
e0c18ff
[dynamo] Support type inspection on unrealized LazyConstantVariables …
williamwen42 Jan 13, 2026
6ee2a19
[dynamo] Support constant ops on unrealized LazyConstantVariables (#1…
williamwen42 Jan 13, 2026
7dcda7b
[dynamo] Support dict assignment with lazy constant keys (#170644)
williamwen42 Jan 13, 2026
643e8ca
Revert "[ROCm] add hipsparseSpSV and hipsparseSpSM support for triang…
pytorchmergebot Jan 14, 2026
3b9640f
[xpu][test]port binary_ufunc test file for Intel GPU (#171238)
wincent8 Jan 14, 2026
b1c6501
ci: Give root the ability to run apt (#172244)
seemethere Jan 12, 2026
e93a344
Update AlbertForMaskedLM graph_breaks from 0 to 4 (#172397)
ethanwee1 Jan 14, 2026
94dd9c5
Fix MPS Inductor for tanh implementation (#172406)
mergennachin Jan 14, 2026
14a9e61
[CI] Add IoU-based accuracy checking for inductor tests segmentation …
jeanschmidt Jan 14, 2026
5600529
Elide score-mod for flex-flash when not needed (#172380)
drisspg Jan 13, 2026
b3d5041
Upgrade usort (#172262)
maggiemoss Jan 14, 2026
da156a2
[vllm hash update] update the pinned vllm hash (#172291)
pytorchupdatebot Jan 14, 2026
a6dc064
[ROCm] Use HIPCachingAllocator for CK argument and workspace buffers …
jagadish-amd Jan 14, 2026
4d20e89
Add b300 to cpp codegen (#172263)
drisspg Jan 14, 2026
172da50
[pytorch][PR] [Durin] Bump Kineto Submodule to fix JSON parsing error…
firoz1905 Jan 14, 2026
f00c5fe
[DTensor] insert Replicate at the begining for matmul single dim (#17…
weifengpy Jan 14, 2026
46924b2
Revert "[dynamo] Support dict assignment with lazy constant keys (#17…
pytorchmergebot Jan 14, 2026
ed935ff
[SymmMem] Add multimem support for NCCL and NVSHMEM (#172185)
kwen2501 Jan 13, 2026
6b83dfd
[SymmMem] Use initializer for devComm requirement (#172400)
kwen2501 Jan 14, 2026
8103dba
[torchcomms] opaque objects - autograd support (#172360)
dolpm Jan 13, 2026
b321605
[xpu][fix] Skip test_fx_memory_profiler_augmentation when using expan…
guangyey Jan 14, 2026
0e227bf
[CI] Run unit tests for XPU without timeout (#171968)
chuanqi129 Jan 14, 2026
65a34a1
Revert "[dynamo] Support constant ops on unrealized LazyConstantVaria…
pytorchmergebot Jan 14, 2026
18f68ed
Revert "[dynamo] Support type inspection on unrealized LazyConstantVa…
pytorchmergebot Jan 14, 2026
e09f7ac
[inductor][overlap] fix circular extra deps; handle artifactos of col…
IvanKobzarev Jan 14, 2026
1b0747d
linalg: sort eigenvalues in eig/eigvals comparison tests (#171717)
johannesz-codes Jan 14, 2026
88ba998
Fix segfault when RNN batch_sizes tensor is on CUDA (#172066) (#172182)
aorenste Jan 13, 2026
64af7a2
Update Rough 2026-2027 Release dates (#171832)
atalman Jan 14, 2026
ee562d9
Revert "[SymmMem] Back symm_mem.emtpy() with implicit pool (#172292)"
pytorchmergebot Jan 14, 2026
75fd1a2
Bump filelock from 3.20.1 to 3.20.3 in /.ci/docker (#172359)
dependabot[bot] Jan 14, 2026
9bbc5b2
Improve error message for missing CH_KEY_ID (#172449)
ezyang Jan 14, 2026
08601dc
[LocalTensor] support misc sym ops (#172268)
pianpwk Jan 14, 2026
3e1aa56
[BE][MPSInductor] Use `::precise::` flavors of trigonometic functions…
malfet Jan 14, 2026
e14a2f0
Skip broken BW compat test in our fa3 smoke tests (#172452)
janeyx99 Jan 14, 2026
f71f226
[DTensor] Added conversion from Replicate to _StridedShard (#171337)
zpcore Jan 14, 2026
017d37f
Update tlparse instructions in pt2 bug report template (#172392)
yushangdi Jan 14, 2026
9748dd9
Update memory error message documentation link and variable name for …
M-Moein-M Jan 14, 2026
8fb088f
Revert "[BE][Ez]: Misc enablement of slots on frozen dataclasses (#17…
pytorchmergebot Jan 14, 2026
4c78f40
[BE] Remove opentelemetry submodule (#172458)
malfet Jan 14, 2026
5e37bf8
ROCm: Fixing test_truediv_base_not_bitwise_equivalent. (#172498)
umechand-amd Jan 14, 2026
c22397c
Redirect setup (#172035)
sekyondaMeta Jan 14, 2026
7add056
remove is_called_from_pytest weirdness in opaque obj test (#172484)
dolpm Jan 14, 2026
ac02de9
Remove MB < PP check for Gpipe (#171462)
H-Huang Jan 14, 2026
5f360e5
Revert "[BE] Remove opentelemetry submodule (#172458)"
pytorchmergebot Jan 14, 2026
0218bcd
[scan][doc] remove doesn't support autograd from doc (#172148)
ydwu4 Jan 9, 2026
b7035de
Add validation for out_channels in quantized ConvTranspose modules (#…
dsashidh Jan 14, 2026
eafa453
[DTensor] reduce test size due to timeout (#172486)
mori360 Jan 15, 2026
faf9d0b
[debug mode] Support dispatching into subgraphs in DebugMode for Invo…
yushangdi Jan 15, 2026
6e9ed21
[easy] dont invoke dce when in temporary, unordered state (#172488)
eellison Jan 14, 2026
2fb65b5
[DTensor] single_dim fix symint + _create_expanded_strategy (#172421)
wconstab Jan 14, 2026
d8fbe78
opaque objects - handle class attributes (#172413)
dolpm Jan 14, 2026
48503ff
[ROCm] add hipsparseSpSV and hipsparseSpSM support for triangular sol…
jsandham Jan 15, 2026
319d3ae
[xpu][test] Port distributed checkpoint test cases on Intel GPU (#168…
zxd1997066 Jan 15, 2026
66d99db
[xpu][test] Port distributed elastic test cases to Intel GPU (#168923)
zxd1997066 Jan 15, 2026
dc50420
[DTensor] benchmark for misc dispatch paths (#171847)
pianpwk Jan 15, 2026
234a24a
[oapque obj] Clean up classes properly (#172503)
angelayi Jan 15, 2026
4f32dd7
[MPS] Add OpInfo skips and dtypes for MPS ops (4/N) (#171952)
kurtamohler Jan 12, 2026
647c081
[MPS] Add OpInfo skips and dtypes for MPS ops (5/N) (#171953)
kurtamohler Jan 12, 2026
2d31067
[MPS] Add OpInfo skips and dtypes for MPS ops (6/N) (#172269)
kurtamohler Jan 12, 2026
8dfd796
[MPS] Add OpInfo skips and dtypes for MPS ops (7/N) (#172270)
kurtamohler Jan 12, 2026
b592255
[dynamo] use closure cell name in recompilation reasons (#172403)
ydwu4 Jan 14, 2026
e6978b9
[opaque obj] Fix call_method on tensor subclasses (#172265)
angelayi Jan 14, 2026
e040897
Revert "[dynamo] Fix property setter on nested MutableMapping objects…
pytorchmergebot Jan 15, 2026
bbef3b9
[Fix] fix fully_shard arg typehint inconsistency (#171574)
nil0x9 Jan 15, 2026
a7ac482
[user-streams] Assign streams to epilogue copies (#168368)
mlazos Jan 15, 2026
7870277
Revert "[SymmMem] Add multimem support for NCCL and NVSHMEM (#172185)"
pytorchmergebot Jan 15, 2026
5952510
[vision hash update] update the pinned vision hash (#171842)
pytorchupdatebot Jan 15, 2026
56f2281
explicit dynamic shapes maps namings (#172405)
laithsakka Jan 14, 2026
3d2bedd
[dcp][hf] Write metadata file for Consolidate hf safetensors file on …
ankitageorge Jan 15, 2026
158613f
align cov operator with numpy behavior (#170390)
kkomma Jan 15, 2026
d23d914
Fix cudnn compile bug (#172528)
drisspg Jan 15, 2026
011e373
Fix the expected number of imports in `test_cutedsl_template.py` (#17…
crcrpar Jan 15, 2026
7559a34
[xpu][test] Enable XPU triton template related UT that not supported …
etaf Jan 15, 2026
232f25e
DTensor Ops: Made aten.div.* linearity similar to aten.mul.* (#172514)
vishal9-team Jan 15, 2026
19d8ad0
remove override hint from size_hint, atomically_size_hint, and symbol…
laithsakka Jan 14, 2026
2be90b1
[inductor] add tlparse to overlap_scheduling (#172502)
IvanKobzarev Jan 14, 2026
1857f59
Update scaled_mm opinfo to generate appropriately scaled sample input…
kundaMwiza Jan 15, 2026
60a9d86
[inductor][TMA] Use revert_sort_idx rather than sort_idx (#172009)
kundaMwiza Jan 15, 2026
7f884f0
env variable for disabling dynamic shapes (#172334)
AmdSampsa Jan 15, 2026
56ddebf
Revert "[xpu][test] Port distributed checkpoint test cases on Intel G…
pytorchmergebot Jan 15, 2026
62abff3
[fx] Don't delete keys from previous metadata (#172317)
angelayi Jan 15, 2026
5ab3874
[ROCm] Enable hipBLASLt on gfx1103 (#172180)
LuXuxue Jan 15, 2026
818aabe
Revert "[Inductor] Ensure combo kernels respect FMA settings (#170532)"
pytorchmergebot Jan 15, 2026
6922a53
Revert "[Inductor] Add fma lowerings for addcmul ops (#170531)"
pytorchmergebot Jan 15, 2026
227bab2
[Allocator] Make large segment size configurable (#172056)
lakshayg Jan 15, 2026
26bebf3
Print the source when assert fails (#172489)
ezyang Jan 14, 2026
34dfe13
[Inductor] Add env var for native matmul (#172492)
jananisriram Jan 15, 2026
528909e
Revert "[fx] Don't delete keys from previous metadata (#172317)"
pytorchmergebot Jan 15, 2026
12acf1e
[dynamo, nested graph breaks] fix nested WithExitFunctionVariable cod…
williamwen42 Jan 15, 2026
4bcb839
[dynamo, nested graph breaks] fix nested graph breaks with error_on_g…
williamwen42 Jan 15, 2026
17d32fd
[dynamo, nested graph breaks] ensure upper ctx managers are active in…
williamwen42 Jan 15, 2026
9f24e80
[dynamo, nested graph breaks] fix codegen error in nested step graph …
williamwen42 Jan 15, 2026
1fda242
[dynamo, nested graph breaks] restore upper contexts for nested step …
williamwen42 Jan 15, 2026
ceccb44
[dynamo, nested graph breaks] add nested graph break microbenchmark (…
williamwen42 Jan 15, 2026
86f2ebf
[MPS] log normal implementation (#172187)
Isalia20 Jan 15, 2026
080120c
[Graph Partition] Improve support for mutation ops (#172483)
BoyuanFeng Jan 15, 2026
0811ce9
[torch][cuda][device_limits] Fixed memory bandwidth calculation formu…
valentinandrei Jan 15, 2026
d3af720
[inductor][3/N] optimize reorder/sink passes compile time - no list m…
IvanKobzarev Dec 9, 2025
48b5311
option to disable ftz in triton (#167696)
v0i0 Jan 15, 2026
e8ac856
[Dynamo] Clearer compile error messages for sparse tensors (#172256)
hinriksnaer Jan 15, 2026
a90b73c
[dynamo] Support type inspection on unrealized LazyConstantVariables …
williamwen42 Jan 15, 2026
9fcc785
Revert "[BE]: Update frozen metautils dataclasses to use slots (#1721…
pytorchmergebot Jan 15, 2026
331ec93
[BE] Remove unnecessary workarounds (#172586)
malfet Jan 15, 2026
511d331
[BE] Remove opentelemetry submodule (#172458)
malfet Jan 15, 2026
11aa017
DTensor Ops: Add linearity support for neg operation (#172563)
vishal9-team Jan 15, 2026
24af852
[dynamo] Add per-graph backend override for debugging/bisecting (#172…
desertfire Jan 14, 2026
83e1da5
[BE] Remove redundant temp vars/functions (#172590)
malfet Jan 16, 2026
067b835
Use existing CSV to preserve entries when the shard doesn't download …
ezyang Jan 16, 2026
853b125
[opaque type] support saving for backward (#171186)
kshitij12345 Jan 16, 2026
da3f2d6
Add non-strict mode for mix order reduction (#171941)
Microve Jan 16, 2026
801a945
[PyTorch] Graph Pickler supports custom ignored metadata field keys (…
nakuliyer Jan 16, 2026
f15f591
[vllm hash update] update the pinned vllm hash (#172518)
pytorchupdatebot Jan 16, 2026
ee26d1c
Fix failing inductor periodic job by tracking wobbles (#172451)
ezyang Jan 16, 2026
d0ea158
Fix torch.nn.Buffer in compile (#172222)
parsshar-RH Jan 16, 2026
0ab064b
[docs] Clarify torch.unique behavior with dim parameter (#171608)
vishalgoyal316 Jan 16, 2026
e6ff4cd
Implement MKLGenerator (#151218)
michalowski-arm Jan 16, 2026
5f15010
[Inductor][NVGEMM] Add BMM support to NVGEMM Inductor backend (#172283)
NikhilAPatel Jan 12, 2026
1cf9ec3
[Inductor][NVGEMM] Filter EFC kernels (#172378)
NikhilAPatel Jan 15, 2026
cfb6625
[Inductor][NVGEMM] Remove guards now handled by new cutlass api versi…
NikhilAPatel Jan 15, 2026
99a7a41
[Inductor][NVGEMM] Enable dynamic shape support (#172391)
NikhilAPatel Jan 15, 2026
5dc99ca
Fix `TestGradCollectives.test_all_reduce` (#172555)
Flamefire Jan 16, 2026
d803917
make `addmm` aware of cublas backend selection: a simple fix for #172…
nikitaved Jan 14, 2026
3b0e35b
[inductor] Use custom triton kernel subclass when available (#167456)
kundaMwiza Jan 16, 2026
7702cf7
[pytorch][WaitCounter] Add WITH_WAIT_COUNTER Macro and add unit tests…
ericcfu Jan 16, 2026
bb6d269
[CD] Exclude Volta from CUDA-12.8 and 12.9 builds (#172598)
malfet Jan 15, 2026
17dbbfb
Harden non strict leak detector some more (#172597)
tugsbayasgalan Jan 15, 2026
6cc9de0
Revert "remove override hint from size_hint, atomically_size_hint, an…
pytorchmergebot Jan 16, 2026
68286ac
Revert "explicit dynamic shapes maps namings (#172405)"
pytorchmergebot Jan 16, 2026
83506e5
Fix libuv build error in windows-arm64-build-test. (#172203)
chinazhangchao Jan 16, 2026
ff8dd55
Revert "[ROCm] Enable StaticTritonLauncher for ROCm (#166492)"
pytorchmergebot Jan 16, 2026
a93aa5a
Revert "make `addmm` aware of cublas backend selection: a simple fix …
pytorchmergebot Jan 16, 2026
f309c1b
Revert "[xpu][feature][static_launcher] 4/4: Enable static triton ker…
pytorchmergebot Jan 16, 2026
681d49c
Revert "[xpu][feature][static_launcher] 3/4: Add xpu static launcher …
pytorchmergebot Jan 16, 2026
a01bb47
Revert "[xpu][static_launcher] 2/4: Refactor static cuda launcher to …
pytorchmergebot Jan 16, 2026
1c3a363
Revert "[inductor] Use custom triton kernel subclass when available (…
pytorchmergebot Jan 16, 2026
1a63385
Revert "option to disable ftz in triton (#167696)"
pytorchmergebot Jan 16, 2026
f796950
CUDA Stream Sanitizer fixes (#172562)
jeffdaily Jan 16, 2026
02513f5
[pytorch][PR] reland D89141291 add fused compute_grad_weight kernel w…
yaoyj11 Jan 16, 2026
3f738f6
[MPSInductor] Fix masked op logic (#170134)
malfet Jan 16, 2026
fedfc4d
[BE] Don't print 12 `triton not found` on import (#172614)
malfet Jan 16, 2026
795fe95
[DOCS][cuDNN][SDPA] Update SDPA context manager docs to show cuDNN ba…
eqy Jan 16, 2026
2eb6de2
Updating CPU code to match CUDA for torch.nn.InstanceNorm2d: Issue#13…
vishalgoyal316 Jan 16, 2026
071bbf6
Revert "[pytorch][PR] reland D89141291 add fused compute_grad_weight …
pytorchmergebot Jan 16, 2026
989c482
[SymmMem] Back symm_mem.emtpy() with implicit pool (#172292)
kwen2501 Jan 13, 2026
f360283
Add warning about inactive development to NJT doc (#172645)
jbschlosser Jan 16, 2026
3f8cf05
Add cudagraph application to bisect (#172461)
eellison Jan 14, 2026
15aad50
Merge remote-tracking branch 'upstream/main' into develop_IFU_20260116
github-actions[bot] Jan 16, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
The diff you're trying to view is too large. We only load the first 3000 changed files.
1 change: 0 additions & 1 deletion .bazelignore
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,3 @@
# We do not use this library in our Bazel build. It contains an
# infinitely recursing symlink that makes Bazel very unhappy.
third_party/ittapi/
third_party/opentelemetry-cpp
19 changes: 0 additions & 19 deletions .ci/aarch64_linux/README.md
View file
Open in desktop

This file was deleted.

53 changes: 0 additions & 53 deletions .ci/aarch64_linux/aarch64_ci_build.sh
View file
Open in desktop

This file was deleted.

21 changes: 0 additions & 21 deletions .ci/aarch64_linux/aarch64_ci_setup.sh
View file
Open in desktop

This file was deleted.

Loading