Skip to content

[AUTOGENERATED] develop_IFU_20251114 #2806

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

Open
wants to merge 511 commits into
base: develop
Choose a base branch
from
Open

[AUTOGENERATED] develop_IFU_20251114 #2806

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
511 commits
Select commit Hold shift + click to select a range
0384104
Update pythoncapi_compat.h to 11cb80f2652cb2fe5231bf60b9dd98c83a4e25f…
colesbury Nov 9, 2025
06aa3ef
Move types from typing_extensions to typing (#167185)
cyyever Nov 9, 2025
9cf623a
Update inductor-unittest.yml (#167417)
oulgen Nov 9, 2025
e7c1905
Fix test_fsdp_logging (#167312)
Flamefire Nov 9, 2025
5135ace
Enable ruff UP035 rule (#167307)
cyyever Nov 9, 2025
14a845a
[2/N] Use Python 3.10 typing (#167167)
cyyever Nov 9, 2025
b91a2ab
[2/N] Use context managers (#167404)
cyyever Nov 9, 2025
afb0145
Separately handle null data_ptr storages when creating unique ID (#1…
ngimel Nov 9, 2025
a4c7856
[Inductor][Grouped Gemm] Add Blackwell CuTeDSL Kernel (#167340)
NikhilAPatel Nov 10, 2025
abf31db
Introduce a new API torch.accelerator.get_memory_info (#156812)
guangyey Nov 7, 2025
fe6615e
Swap pallas test shard to 12.8 (#167428)
oulgen Nov 9, 2025
2c78080
Register functorch XPU/HPU dispatch keys (#167095)
Silv3S Nov 10, 2025
a058bbd
[xpu][test] Enable profiler test for XPU (#165423)
chuanqi129 Nov 10, 2025
e545ba2
[DTensor] Fix Conv behavior for replicate stategy (#167402)
malfet Nov 8, 2025
50af6f3
[MPS] erfinv for sparse mps (#166711)
Isalia20 Nov 10, 2025
47db552
[MPS] sparse sparse mm (#167013)
Isalia20 Nov 10, 2025
3cfbf98
[xpu][feature] Add XPU support on torch.accelerator.get_memory_info (…
guangyey Nov 7, 2025
52e744d
[DTensor] Support convert StridedShard to shard order and vice versa …
zpcore Nov 10, 2025
74aec83
[xla hash update] update the pinned xla hash (#167452)
pytorchupdatebot Nov 10, 2025
c28475d
Update slow tests (#166844)
pytorchupdatebot Nov 10, 2025
59307ca
[BE] adding documentation (#167334)
liangel-02 Nov 10, 2025
31ccd8f
[AOTI] Fix a mixed-device bug for scatter_add (#167341)
desertfire Nov 7, 2025
2fcf41d
Add the ruff rule and skip everything for now (#167360)
albanD Nov 10, 2025
f6a79b2
[inductor] Wrap pallas_call in jax.jit (#167441)
jansel Nov 10, 2025
3966b5a
[BE] Fix out-of-bounds index_put in test_mps.py (#167444)
malfet Nov 10, 2025
3ea829a
Fix torch.cond HOP device in inductor (#167354)
yushangdi Nov 10, 2025
a4437d7
Add some labeler rules that used to be in the autolabel bot (#167330)
clee2000 Nov 10, 2025
04a85b4
[compile-on-one-rank] Step 1: DeviceId (#166680)
aorenste Nov 9, 2025
9491830
move subgraph_has_impure_ops from `node.is_impure` into const_fold t…
jazlyn5 Nov 10, 2025
86130aa
Fix flaky memory profiler test [2] (#167268)
yushangdi Nov 10, 2025
cdc8460
Use c7i.2xlarge for H100 build (#167466)
zxiiro Nov 10, 2025
6cf21fa
Fix -ffunction-sections, -fdata-sections not being added on aarch64. …
robert-hardwick Nov 5, 2025
0c2f206
Typo fix - baddbmm_strategy (#166963)
stmcgovern Nov 10, 2025
3e4faca
[torch.export] Refactor placeholder_naming_pass to reduce CCN (#166600)
malaybag Nov 10, 2025
5320ca3
[inductor, 3.14] fix itertools.product pickle error in test_cpu_repro…
williamwen42 Nov 8, 2025
ad7db36
[inductor, 3.14] catch pickle.PicklingError exceptions (#167383)
williamwen42 Nov 8, 2025
17e70ae
[dynamo, 3.14] enable dynamo in 3.14 (#167384)
williamwen42 Nov 8, 2025
cf63b21
[3.14, dataloader] handle forkserver default mp start method in 3.14 …
williamwen42 Nov 8, 2025
fe0bb7c
[export, 3.14] handle patching methods with functools.partial correct…
williamwen42 Nov 8, 2025
2751b1d
Support repr on user defined objects (#167372)
tugsbayasgalan Nov 10, 2025
bb37483
Use c7i.2xlarge for B200 build (#167078)
zxiiro Nov 10, 2025
6ca8cc6
Rework PyObject preservation (#166342)
colesbury Nov 10, 2025
cda7604
[ez] Remove spammy deprecation log (#167470)
SherlockNoMad Nov 10, 2025
7da82b8
Add torch::stable::Device (#166579)
mikaylagawarecki Nov 10, 2025
6f01824
Add stable::Tensor.device() (#166694)
mikaylagawarecki Nov 10, 2025
232baa3
Redo add parallel_for to torch/csrc/stable (#166695)
mikaylagawarecki Nov 10, 2025
69ab1f9
Add shim for at::get_num_threads (#167362)
mikaylagawarecki Nov 10, 2025
47da714
[inductor][determinism] type errors + use odc to dump imc on exit (#1…
nmacchioni Nov 10, 2025
de77336
Support AC in default partitioner when functionalization is enabled (…
soulitzer Nov 7, 2025
8ef4099
Revert "Add min/max support for barebones uint types (#166813)"
pytorchmergebot Nov 10, 2025
4c3721f
allow sym_stride, and sym_size lowering in inductor to return ints (…
laithsakka Nov 7, 2025
9d9e7c7
[Pytorch] Extend OSS conversion benchmarks (#167099)
Nicoshev Nov 10, 2025
e3d6896
[MTIAGraph][Pytorch][3/n] Implement mtia_graph python wrapper in pyto…
andyanwang Nov 10, 2025
5a85b6e
Migrate TypeTraits, TypeList, Metaprogramming to torch:: headeronly (…
janeyx99 Nov 10, 2025
f8d408d
[CP] Correctly compile create_cp_block_mask (#167153)
fegin Nov 10, 2025
f633119
Enable Doc builds for: Minor Releases RCs. Minor and Patch Releases f…
atalman Nov 11, 2025
8d5ccee
[torchbench][optimus] Add backend optimus (#167357)
xuzhao9 Nov 11, 2025
b2be4d2
[DTensor] Make ExplicitRedistributeContext strict/non-strict mode (#1…
wconstab Nov 10, 2025
c3bc56c
[xpu][fix] Format XPU c10 and aten code (#167298)
guangyey Nov 10, 2025
406f294
Revert "Rework PyObject preservation (#166342)"
pytorchmergebot Nov 11, 2025
7a48db0
Revert "[export, 3.14] handle patching methods with functools.partial…
pytorchmergebot Nov 11, 2025
619f329
Revert "[3.14, dataloader] handle forkserver default mp start method …
pytorchmergebot Nov 11, 2025
a14452b
Revert "[dynamo, 3.14] enable dynamo in 3.14 (#167384)"
pytorchmergebot Nov 11, 2025
b5142f7
Revert "[inductor, 3.14] catch pickle.PicklingError exceptions (#1673…
pytorchmergebot Nov 11, 2025
f0fa39a
Revert "[inductor, 3.14] fix itertools.product pickle error in test_c…
pytorchmergebot Nov 11, 2025
b06453c
Make PT2 compile backprop through custom op without autograd key a ha…
ezyang Nov 1, 2025
2ca0b3f
[simplefsdp] fix autobucketing pass that takes comm op as input (#167…
ruisizhang123 Nov 11, 2025
5676de1
[PT2] Supply index to fake tensors on mtia device (#167457)
blaine-rister Nov 11, 2025
94f6f79
[3/N] Use Python 3.10 typing (#167431)
cyyever Nov 11, 2025
52a6b5a
Expose `THPVariable_Wrap()` with a type argument (#167488)
colesbury Nov 11, 2025
f9caae4
[MPS] Move `dispatch_sync_with_rethrow` to MPSStream (#167445)
malfet Nov 11, 2025
cf71c53
Fix check_compiler_is_gcc to detect versioned GCC compilers (#167501)
razaaliraza Nov 11, 2025
52231a7
show current env before running lint (#166860)
albanD Nov 11, 2025
db250fa
Revert "Expose `THPVariable_Wrap()` with a type argument (#167488)"
pytorchmergebot Nov 11, 2025
c34b743
[Dynamo] Support for xor (#166065)
jmaczan Nov 11, 2025
53422e6
[MPS] Add mechanism for reporting asserts from kernels (#166615)
malfet Nov 11, 2025
87d17e9
[pallas backend] Implementing Strided/Scatter Access (#167426)
oulgen Nov 11, 2025
7886070
Use stable topological sort in fuse_by_partitions (#167397)
SherlockNoMad Nov 11, 2025
525eb9f
Fix command injection vulnerability in PCH compilation (#167502)
razaaliraza Nov 11, 2025
bd24774
[XPU][Test] Enable XPU tests in inductor/test_analysis.py (#166840)
Stonepia Nov 11, 2025
c297b02
[DTensor] statically_known_true for slice strategy (#166990)
pianpwk Nov 8, 2025
53947ad
[Inductor] optimize the heuristics of sum reduction (#163144)
jiayisunx Nov 4, 2025
daed97a
[Inductor] fix CppTile2DKernel for fp8 datatype (#167451)
jiayisunx Nov 10, 2025
678915d
Update Arm copyright dates in `LICENSE` file (#167529)
puneetmatharu Nov 11, 2025
5f0a563
[pallas backend] implement complex indexing (#167493)
oulgen Nov 11, 2025
3e7a66f
[BugFix][Refactor] fix several instances which use f = open(...) with…
lingebeng Nov 11, 2025
fad48ff
[ROCm][CI] Match workflow names with workflow file names (#167483)
jithunnair-amd Nov 11, 2025
f1a129a
Clarify that crashes/OOB accesses and not security threats (#167519)
malfet Nov 11, 2025
dc00842
[ROCm][CI] trigger magma build with gfx950 for ROCm7.1 (#167390)
jagadish-amd Nov 11, 2025
43041f0
Remove superflous/misplaced `TestFailure` specs (#165989)
Flamefire Nov 11, 2025
db1f3f6
[inductor] Only generate compile-time auto-tuning block in the main g…
desertfire Nov 10, 2025
ffe6cc3
[inductor] Optimize cold compile time when cudagraphs-partition is en…
desertfire Nov 10, 2025
f9b81e2
[ROCm] Disable group gemm CK path when composable kernel (CK) is not…
jagadish-amd Nov 11, 2025
1df723e
[inductor] Fix constant creation (#167398)
angelayi Nov 11, 2025
4945180
Add empty tensor check for `_pad_packed_sequence` (#167521)
malfet Nov 11, 2025
573a79f
[OpenReg] Initialize device stream states for all devices in `initOpe…
KarhouTam Nov 11, 2025
402c465
[ARM] Improve LLM performance & mem usage using int4-bf16 KleidiAI ke…
usamahz Nov 11, 2025
ce8672c
Fix use of TORCH_CHECK in torch/csrc/stable (#167495)
mikaylagawarecki Nov 11, 2025
bd99ae3
[Docs] Add warning that `torch.export.load` uses `pickle` (#167557)
malfet Nov 11, 2025
416421c
fix failure of exporting compiled model with nested dynamic shapes (#…
supercharleszhu Nov 11, 2025
184e2cb
[ROCm][CI] Update docker-cache-mi300.yml to test MI300 CI docker cach…
jithunnair-amd Nov 11, 2025
f44a1dd
Revert "[ROCm][CI] Update docker-cache-mi300.yml to test MI300 CI doc…
pytorchmergebot Nov 11, 2025
b0e0ae9
include `thrust/distance.h` explicitly in cuda sparse softmax (#167436)
crcrpar Nov 11, 2025
b65829b
[DTensor] Log API usage metrics for DTensor and DeviceMesh (#167374)
wconstab Nov 11, 2025
fe841a1
[DeviceMesh] Log DeviceMesh.__init__ usage (#167375)
wconstab Nov 11, 2025
d8ada1e
[user-streams] Allow new events to be created and registered during c…
mlazos Nov 11, 2025
61883a5
[user-streams] Allow new streams to be created and registered during …
mlazos Nov 11, 2025
8def619
[user-streams] wait_stream op (#167512)
mlazos Nov 11, 2025
5603407
Revert "[Inductor] Naive foreach autotune support (#162053)"
pytorchmergebot Nov 11, 2025
3a94466
Cpython test_math.FMATests (#167217)
trichmo Nov 11, 2025
ee387c4
[ROCm][CI] Update docker-cache-mi300.yml to test MI300 CI docker cach…
jithunnair-amd Nov 11, 2025
051f1fe
Revert "[ROCm][CI] Update docker-cache-mi300.yml to test MI300 CI doc…
pytorchmergebot Nov 11, 2025
259ba0e
[DebugMode] record triton kernels, run-to-run determinism checks (#16…
pianpwk Nov 11, 2025
b21856f
Revert "[DebugMode] record triton kernels, run-to-run determinism che…
pytorchmergebot Nov 11, 2025
1f7e434
[ROCm][CI] Add docker-cache-rocm.yml to test MI3xx CI docker caching …
jithunnair-amd Nov 11, 2025
5b6ff81
Revert "[ARM] Improve LLM performance & mem usage using int4-bf16 Kle…
pytorchmergebot Nov 11, 2025
ae71b0e
Fix typo in torch._refs (#167310)
can-gaa-hou Nov 11, 2025
9ae62fc
[ROCm][CI] dynamo benchmarks update ci expected accuracy (#167574)
jeffdaily Nov 11, 2025
514dd96
Remove --no-use-pep517 flag (#167096)
ethanwee1 Nov 11, 2025
2e5233d
Revert "Support AC in default partitioner when functionalization is e…
pytorchmergebot Nov 11, 2025
e8d411e
FSDPMemTracker fix with multihander hooks. (#165662)
skpark-rh Nov 11, 2025
971e6ca
fix sym_size_, sym_stride lowering (#167565)
laithsakka Nov 11, 2025
115016f
[Device Mesh][ez] Clean up unused parameters and duplicate codes (#16…
fduwjj Nov 11, 2025
cdf0a9c
Add FA4 to sdpa (#167348)
drisspg Nov 11, 2025
1debfd4
Revert "Add FA4 to sdpa (#167348)"
pytorchmergebot Nov 12, 2025
6739069
[ROCm][CI] Restrict docker-cache-rocm.yml to main/release branches (#…
jithunnair-amd Nov 12, 2025
91f0c5a
[simplefsdp] add manual bucketing pass (#165487)
ruisizhang123 Nov 12, 2025
1366a2f
[ROCm][CI] Enable uploading of artifacts from docker-builds.yml (#167…
jithunnair-amd Nov 12, 2025
edd365e
[MemTracker] Fix: Remove monkey patching DTensor dispatch (#167580)
sanketpurandare Nov 12, 2025
bc882f8
Support Python 3.14 Lazy Function Annotations on FX graph (#167573)
guilhermeleobas Nov 11, 2025
25e9d81
Revert "Use c7i.2xlarge for B200 build (#167078)"
pytorchmergebot Nov 12, 2025
05c6a06
Add FA4 to sdpa (#167348)
drisspg Nov 12, 2025
3260bf3
[export] stop gap strict export v2 enable and testing. (#167236)
zhxchen17 Nov 12, 2025
7dd5647
[annotation] Skip copying custom meta for gradient accumulation nodes…
yushangdi Nov 12, 2025
5ce4a8b
Revert "fix wrong accuracy_status when exception. (#165731)"
pytorchmergebot Nov 12, 2025
7bb8d8c
[ROCm][CI] Add trunk-rocm-mi300.yml to test new MI3xx CI capacity (#1…
jithunnair-amd Nov 12, 2025
148fd9a
[audio hash update] update the pinned audio hash (#167490)
pytorchupdatebot Nov 12, 2025
bf380fb
[vision hash update] update the pinned vision hash (#167491)
pytorchupdatebot Nov 12, 2025
2a09f6e
[4/N] Use Python 3.10 typing (#167458)
cyyever Nov 12, 2025
9997e85
[DebugMode] record triton kernels, run-to-run determinism checks (#16…
pianpwk Nov 12, 2025
dc8bb52
Inductor Lite Mode (#167115)
BoyuanFeng Nov 12, 2025
d33d125
[inductor] Remove output copy_ for pallas backend in some cases (#167…
jansel Nov 12, 2025
6bf51de
harden backed_size_oblivious and broadcast_shapes (#167232)
laithsakka Nov 12, 2025
780e325
Move XPUEvent to c10 (#158336)
guangyey Nov 12, 2025
4714eb7
Update dynamic_inductor_timm_training.csv (#167609)
zou3519 Nov 12, 2025
4cff8b5
Add option to disable applying side effects in dynamo (#167239)
tugsbayasgalan Nov 11, 2025
4f6aae3
Revert "[MPS] SparseMps mv op (#166708)"
pytorchmergebot Nov 12, 2025
a328326
Revert "[xpu][feature] Add XPU support on torch.accelerator.get_memor…
pytorchmergebot Nov 12, 2025
c5d91d9
Revert "Introduce a new API torch.accelerator.get_memory_info (#156812)"
pytorchmergebot Nov 12, 2025
7557e38
[ROCm] hipSPARSELt support - Update cuda_to_hip_mappings.py (#167335)
rraminen Nov 12, 2025
dd7a45a
[5/N] Use Python 3.10 typing (#167449)
cyyever Nov 12, 2025
bdb3753
Add Tests (#167392)
drisspg Nov 12, 2025
10a1578
Revert "Update Kineto Submodule (#167343)"
pytorchmergebot Nov 12, 2025
ed79693
[ROCm][CI] dynamo benchmark repvgg_a2 is flaky (#167660)
jeffdaily Nov 12, 2025
d105e3a
[dynamo][DebugMode] mask python keys in dispatch_key_set guard checks…
pianpwk Nov 12, 2025
760c901
[torch] Update caffe2/torch/csrc to build under CUDA 13 (#167401)
q10 Nov 12, 2025
bc09a84
Hide all symbols (except stable/headeronly/shim) if TORCH_STABLE_ONLY…
mikaylagawarecki Nov 12, 2025
2ad70c9
[CI] manually gen json for segfaults (#167250)
clee2000 Nov 12, 2025
a95eee6
[user-streams] Add API for accessing current stream given a device (#…
mlazos Nov 12, 2025
a6a0379
[caffe2] Address -Wswitch-default warnings in headers (#167563)
NSProgrammer Nov 12, 2025
74e85c6
Add TORCH_BOX helper for STABLE_TORCH_LIBRARY_IMPL (#167582)
janeyx99 Nov 12, 2025
5f0a5b8
Revert "Use stable topological sort in fuse_by_partitions (#167397)"
pytorchmergebot Nov 12, 2025
1311385
Revert "fix failure of exporting compiled model with nested dynamic s…
pytorchmergebot Nov 12, 2025
2ca428c
[CD] Preload `libnvrtc-builtinso.so` (#167614)
malfet Nov 12, 2025
0184ef2
[inductor][NFC][1/X] extract create_no_valid_choices from AlgorithmSe…
nmacchioni Nov 12, 2025
158e724
[torch] Update caffe2/c10/cuda to build under CUDA 13 (#167534)
q10 Nov 12, 2025
0dac408
MatMal - fix folding logic (#166891)
dstaay-fb Nov 12, 2025
537167a
Fix thread safety in getCurrentCUDABlasHandle and getCUDABlasLtWorksp…
t-ivan-gr Nov 12, 2025
2fa18d1
[export] Codemod more tests to use dynamo_graph_capture_for_export (#…
zhxchen17 Nov 12, 2025
a76dd6b
[MPS] SparseMps mv op (#166708)
Isalia20 Nov 12, 2025
273babe
[precompile] Integrate AOTI as a backend. (#167338)
zhxchen17 Nov 13, 2025
2d73900
[dynamo] speculate_subgraph_with_auto_output_flattening (#167438)
anijain2305 Nov 12, 2025
5f98a03
[dynamo] Make HintsWrapperHigherOrderVariable follow wrap semantics (…
anijain2305 Nov 12, 2025
0c5d5c7
[dynamo][invoke_subgraph] Do not restore side effects on invoke_subgr…
anijain2305 Nov 12, 2025
485f2b6
ProxyTorchDispatchMode: Decomposing missing sympy.SymExpr should hand…
aorenste Nov 12, 2025
35571fe
[effects] Add register_effectful_op (#163284)
angelayi Nov 12, 2025
c9b09a3
[opaque obj] Allow non-effectful scriptobjs (#163714)
angelayi Nov 12, 2025
e3dadb1
[opaque obj] torch.compile support (#163936)
angelayi Nov 12, 2025
19c8678
[opqaue obj] Add attribute support (#167230)
angelayi Nov 12, 2025
8919f69
[Inductor][2/2] Decouple flags for optimization and debug symbols (#1…
bbeckca Nov 13, 2025
8f5f89c
Revert "Fix thread safety in getCurrentCUDABlasHandle and getCUDABlas…
pytorchmergebot Nov 13, 2025
9b68682
[ROCm] Enable several DISABLED issues (#167183)
pragupta Nov 13, 2025
2984331
[inductor][NFC][2/X] extract do_autotuning/autotune/benchmark from Al…
nmacchioni Nov 13, 2025
d9a50bf
[dynamo] [3.14] Support np._CopyMode (#167619)
rtimpe Nov 12, 2025
eeebf9f
[dynamo] [3.14] Update broken numpy test (#167681)
rtimpe Nov 12, 2025
f9851af
Add Attention ops to CI (#165915)
jainapurva Nov 13, 2025
f570e58
Add C++ fast path for `DTensor.__torch_dispatch__` (#167051)
swolchok Nov 13, 2025
480b4ff
Avoid creating Python OpSchema in the DTensor dispatch fast path (#16…
swolchok Nov 13, 2025
2034ca9
extend C++ DTensor fast path to local operator dispatch (#166808)
swolchok Nov 13, 2025
3d801a4
DTensor fast path: port return_and_correct_aliasing and inplace/out c…
swolchok Nov 13, 2025
1a67403
Move MemPool out of c10 and into ATen. (#167506)
galv Nov 11, 2025
782fc3c
[DTensor] Add CPU instruction count benchmark for dispatch (#167394)
wconstab Nov 11, 2025
8f96e7b
Only remove_noop in pre_grad passes if remove_noop is not in the remo…
ShyamalShah3 Nov 13, 2025
8c86ccf
[DebugMode] .show_stack_trace inline (#167589)
pianpwk Nov 13, 2025
2c846bb
[xpu][test]port embedding indexing and native_mha test files for Inte…
wincent8 Nov 13, 2025
ce4f31f
[OpenReg][Feat][Docs] Enrich hook implementation and add focused docu…
KarhouTam Nov 13, 2025
9ae0ece
Introduce a new API torch.accelerator.get_memory_info (#156812)
guangyey Nov 13, 2025
f2d0a47
[xpu][feature] Add XPU support on torch.accelerator.get_memory_info (…
guangyey Nov 13, 2025
4de24bc
[Fix XPU typo] Fix a comment typo of FindSYCLToolkit.cmake (#165884)
luoyu-intel Nov 13, 2025
c940b1f
address DDE in matmul decomp (#166541)
laithsakka Nov 13, 2025
d3ca4a3
[CUDA][64-bit indexing] Handle 64-bit outer dim `cumsum` case (#167326)
eqy Nov 13, 2025
698aa0f
[MPS] sparse_mask_projection (#166260)
Isalia20 Nov 13, 2025
374ee9e
Fix missing thrust includes (#167450)
miscco Nov 13, 2025
7aac506
Revert "[precompile] Integrate AOTI as a backend. (#167338)"
pytorchmergebot Nov 13, 2025
460c7e1
Handle only a Tensor for IntList parsing (#167606)
ezyang Nov 13, 2025
6ea7791
[DebugMode] torch.hash_tensor option (#167486)
pianpwk Nov 13, 2025
b5e0e69
Correctly populate storage offset in DTensor constructor (#167597)
ezyang Nov 12, 2025
e5eb89e
remove allocation of new unbacked symbols during mod eval (#167123)
laithsakka Nov 13, 2025
fadb62f
[PyTorch] fix profiler issue with empty exported trace file (#167601)
TroyGarden Nov 13, 2025
d273422
[CUDA] Large max pool fix (#167427)
Isalia20 Nov 13, 2025
d8384e2
[Inductor] Remove bf16 fallback for atomic_add (#167380)
karthickai Nov 12, 2025
cfb3a6b
[2/N][BugFix][Refactor] fix several instances which use f = open(...)…
lingebeng Nov 13, 2025
38806f3
[inductor, 3.14] fix itertools.product pickle error in test_cpu_repro…
williamwen42 Nov 13, 2025
9ac3fc0
[inductor, 3.14] catch pickle.PicklingError exceptions (#167383)
williamwen42 Nov 13, 2025
23f4f32
[dynamo, 3.14] enable dynamo in 3.14 (#167384)
williamwen42 Nov 13, 2025
4fc6886
[3.14, dataloader] handle forkserver default mp start method in 3.14 …
williamwen42 Nov 13, 2025
940979a
[export, 3.14] handle patching methods with functools.partial correct…
williamwen42 Nov 13, 2025
21f32e4
[dynamo] clean up BaseUserFunctionVariable and LocalGeneratorObjectVa…
williamwen42 Nov 13, 2025
8f00ec3
[dynamo, nested graph breaks] disallow graph breaks in functorch ops,…
williamwen42 Nov 13, 2025
0b3bdb0
[EZ][BE] Remove unnecessary semicolon in Module.cpp (#167756)
malfet Nov 13, 2025
3d06351
[inductor][ez] skip cache for unit test via envvar (#167237)
shunting314 Nov 6, 2025
f79cdc8
[CD] [aarch64] unify the build.sh to build for aarch64 wheel (#166044)
tinglvv Nov 13, 2025
a954242
[MPS] Add Metal complex mm implementation (#167755)
malfet Nov 13, 2025
fe33d7c
Revert "address DDE in matmul decomp (#166541)"
pytorchmergebot Nov 13, 2025
0cd0bd7
address DDE in matmul decomp (#166541)
laithsakka Nov 13, 2025
08de54f
[3.14] Skip failing spherical_bessel_j0 tests (#167691)
rtimpe Nov 13, 2025
532389f
[torchelastic] Add flush option to TailLog (#167169)
cnphil Nov 14, 2025
2ef236e
[3.14, jit] skip jit tests on 3.14+, add jit deprecation warnings to …
williamwen42 Nov 13, 2025
813e5ea
[fx, 3.14] fix assert detection for 3.14 (#167700)
williamwen42 Nov 13, 2025
8cf0bdd
[xpu][fix] Fix conv1d precision error (#162944)
yucai-intel Nov 14, 2025
05bcfcc
[Profiler] Add Documentation for FunctionEvent (#167688)
sraikund16 Nov 14, 2025
96a4c4b
add device generalization support for distributed tests (#165067)
harikodali Nov 14, 2025
79317dc
Fix no source name in backward kernel names; Add flex_attention HOP t…
yushangdi Nov 14, 2025
5e6ac5c
[Pytorch] Improve conversion to bfloat16 on aarch64/NEON (#166958)
Nicoshev Nov 14, 2025
5b1e112
[Dynamo] Imporve-graph-break-skip-logs (#167067)
parsshar-RH Nov 14, 2025
45b2c3d
[OpenReg][Feat][Docs] Enrich OpenReg device management implementation…
KarhouTam Nov 14, 2025
2aba180
Always track _local_scalar_dense output in tensorify_python_scalars. …
laithsakka Nov 14, 2025
5623628
[SymmMem] op to get remote tensors (#167779)
kwen2501 Nov 14, 2025
c78e646
Fix different seq length (#167481)
Microve Nov 14, 2025
50bf1f0
deprecate check_is_size and guard_size_oblivious (#167198)
laithsakka Nov 14, 2025
3522e0c
Revert "Fix different seq length (#167481)"
pytorchmergebot Nov 14, 2025
0e7235e
[xpu][feature] [1/3] add fp8 scaled_mm implementation for XPU (#165978)
Stonepia Nov 14, 2025
e2c6834
Revert "deprecate check_is_size and guard_size_oblivious (#167198)"
pytorchmergebot Nov 14, 2025
f8a2ce3
Fix inplace ops on Partial DTensors to preserve aliasing semantics (#…
RohitRathore1 Nov 14, 2025
226850c
[ATen][CUDA] Add sm_121a flag for RowwiseScaledMM (#167734)
Aidyn-A Nov 14, 2025
b657061
[precompile] Integrate AOTI as a backend. (#167338)
zhxchen17 Nov 14, 2025
bfddfde
Add basic spin config and linting commands (#167226)
zklaus Nov 13, 2025
40e6f09
Re-land "Fix thread safety in getCurrentCUDABlasHandle and getCUDABla…
t-ivan-gr Nov 14, 2025
a3addec
Merge remote-tracking branch 'upstream/main' into develop_IFU_20251114
github-actions[bot] Nov 14, 2025
2903e7a
Fix merge conflicts
Nov 14, 2025
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
  •  
  •  
  •  
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