Skip to content

Fix lint #10089

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.

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
wants to merge 4,782 commits into from
Closed

Fix lint #10089

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4782 commits
Select commit Hold shift + click to select a range
12240cf
add sample input for new core op elu.out (#9577)
swolchok Mar 26, 2025
4873681
Fix -Wglobal-constructors/-Wshadow for executor_runner targets (#9509)
swolchok Mar 26, 2025
b89328a
Depend on extension/threadpool, not thread_parallel_interface, in buc…
swolchok Mar 26, 2025
118f0a4
elementwise_util: s/common/compute/ almost everywhere and deprecate S…
swolchok Mar 26, 2025
f8c1014
Deprecate non-internal elementwise_util APIs (#9621)
swolchok Mar 26, 2025
7d614a5
Update pinned optimum-executorch commit hash in the CI (#9653)
guangy10 Mar 26, 2025
80992a9
Remove tombstone messages proactively
bigfootjon Mar 26, 2025
f5bbad1
Upgrade arm runner for phi4-mini (#9658)
mergennachin Mar 26, 2025
0fddf9d
Test CoreML support when building wheels (#9647)
jathu Mar 26, 2025
e9822f9
[BE][ET-VK] update permute to use layout gen
pytorchbot Mar 26, 2025
039ab24
[BE][ET-VK] update clone to use layout gen
pytorchbot Mar 26, 2025
f52f9a0
[BE][ET-VK] update full to use layout gen
pytorchbot Mar 26, 2025
99cb7bc
[BE][ET-VK] update max_pool2d to use layout gen
pytorchbot Mar 26, 2025
9fc101f
[ET-VK][ez] Fix 8 bit linear compute shader dispatch
pytorchbot Mar 26, 2025
dfa7796
[Executorch][llama] Change runner to decouple prompt length from sequ…
pytorchbot Mar 26, 2025
f151357
[ExecuTorch] Arm Ethos: Do not depend on `torch.testing._internal `
pytorchbot Mar 26, 2025
9176fd7
Initialize Tensor with native TensorPtr.
shoumikhin Mar 27, 2025
3cac686
Use `llama_inputs` config option only if available (#9667)
kirklandsign Mar 27, 2025
53ab24c
Tensor properties to get shape, etc.
shoumikhin Mar 27, 2025
d32517b
Tensor constructor to create with a raw pointer w/o copying.
shoumikhin Mar 27, 2025
2e08780
[ET-VK][benchmarking][ez] Don't perform copies when benchmarking
pytorchbot Mar 27, 2025
d5898b7
[ET-VK][ez] Fix printing shader IRs
pytorchbot Mar 27, 2025
93c3b2f
Overloads for the bytesNoCopy Tensor constrcutor.
shoumikhin Mar 27, 2025
1ea101e
Tensor helpers to get the size of a data type and number of elements …
shoumikhin Mar 27, 2025
7030b33
Tensor constructor to create with a raw pointer by copying the data.
shoumikhin Mar 27, 2025
dab0d3c
Overloads for the bytes Tensor constrcutor.
shoumikhin Mar 27, 2025
2865652
Tensor accessorts to get raw data buffer.
shoumikhin Mar 27, 2025
0b1a840
Tensor constructor to create with an existing Tensor.
shoumikhin Mar 27, 2025
3f277f4
Make Tensor conform to NSCopying protocol.
kirklandsign Mar 27, 2025
32b9449
Add Tensor resizing API.
kirklandsign Mar 27, 2025
79743c9
Add Tensor equality API. (#9680)
kirklandsign Mar 27, 2025
1bdee96
Value default initializer. (#9681)
kirklandsign Mar 27, 2025
e02c136
Create Value with Tensor. (#9682)
kirklandsign Mar 27, 2025
311c6cb
Create Module with a filepath and load mode.
kirklandsign Mar 27, 2025
77fd564
Add Module load API. (#9684)
kirklandsign Mar 27, 2025
1d93152
Add Module loadMethod API. (#9685)
kirklandsign Mar 27, 2025
750b1f1
Add Module methodName API. (#9686)
kirklandsign Mar 27, 2025
80138b8
Add Module execute API. (#9687)
kirklandsign Mar 27, 2025
ae74ef9
Overloads for Module execute API. (#9688)
kirklandsign Mar 27, 2025
fa67bde
Add Module forward API with overloads. (#9689)
kirklandsign Mar 27, 2025
05696f8
Tensor constructor to create with an existing NSData. (#9690)
kirklandsign Mar 27, 2025
dd3122d
Overloads for data Tensor constructor. (#9691)
kirklandsign Mar 27, 2025
4076f93
Tensor constructor to create with an array of scalars. (#9692)
kirklandsign Mar 27, 2025
d2eeff4
Overloads for scalars constructor. (#9693)
kirklandsign Mar 27, 2025
cdf9cc5
More tests for scalars constructor. (#9694)
kirklandsign Mar 27, 2025
29bc273
Tensor constructor to create with a single scalar. (#9695)
kirklandsign Mar 27, 2025
08ce3ed
Overloads for scalar constructor. (#9696)
kirklandsign Mar 27, 2025
45d7c30
More tests for scalar constructor. (#9697)
kirklandsign Mar 27, 2025
9cfaeac
Add more types to create a Value from. (#9698)
kirklandsign Mar 27, 2025
f873ada
Handle more Value types in Module. (#9699)
kirklandsign Mar 27, 2025
50f57fd
Add Value equality API. (#9700)
kirklandsign Mar 27, 2025
404b186
Use different bundle to lookup resources for SwiftPM.
shoumikhin Mar 27, 2025
48c0b19
Module API to execute with a single Tensor.
shoumikhin Mar 27, 2025
07266f9
Arm backend: Add tosa_spec info to .tosa files (#9392)
oscarandersson8218 Mar 27, 2025
6332238
Support channels_last format in portable upsample kernels
GregoryComer Mar 27, 2025
6939136
Arm backend: Only describe pytest and not python unittest in README.m…
zingo Mar 27, 2025
c048ea2
Arm backend: Add missing __init__.py to passes (#9710)
oscarandersson8218 Mar 27, 2025
976fe48
Add method variants in dynamic shim to use core tensor/evalue/result …
GregoryComer Mar 27, 2025
5098808
Remove exception fall back on checkpoint loading (#9660)
jackzhxng Mar 27, 2025
95702dc
[ET-VK] Cache compiled SPIR-V and only recompile when changed (#9701)
pytorchbot Mar 27, 2025
03c5cf7
Arm backend: Use gitlab and bump version of ethos-u-core-driver to 25…
zingo Mar 27, 2025
771588a
Arm backend: Support scalar_tensor op (#9711)
Erik-Lundell Mar 27, 2025
8ee280b
Arm backend: Add MobileNet v3 testcase (#9223)
tom-arm Mar 27, 2025
01f5599
set -x in build_apple_frameworks.sh (#9326)
swolchok Mar 27, 2025
a4925e4
Move ExecutorchRuntimeBridge into fb dir.
shoumikhin Mar 27, 2025
ef8204c
[ET-VK][ez] Small fix to SPIR-V compilation artifact caching (#9720)
pytorchbot Mar 27, 2025
565b2b4
[ET-VK] Adding repeat support to add_copy_packed_dim_offset_node func…
pytorchbot Mar 27, 2025
879b94f
convert to .md and clean up content on main docs page (#9663)
byjlw Mar 27, 2025
895efcb
Support alpha in scalar add/sub cases
mcremon-meta Mar 27, 2025
b0c0fda
Arm backend: Clean up shift support (#9573)
Erik-Lundell Mar 27, 2025
5531a0e
Revert "Depend on extension/threadpool, not thread_parallel_interface…
swolchok Mar 27, 2025
ec1cd04
Add mixed dtype check for XNNPACK partitioner (#9612)
keyprocedure Mar 27, 2025
8b948e8
Add a convenient constructor
larryliu0820 Mar 28, 2025
399a255
Pin cmake version < 4.0.0 (#9732)
larryliu0820 Mar 28, 2025
2d01dfc
Improve android related docs
kirklandsign Mar 28, 2025
d72ef5b
Arm backend: Make download of ethos-u dependencies optional (#9734)
tobbeebbot Mar 28, 2025
f174d55
Fix CoreML pybinding module
jathu Mar 28, 2025
7fc176d
Arm backend: Add support to eq.Scalar (#9715)
fumchin Mar 28, 2025
65ebabb
Deprioritize top level -O2 in CMAKE_CXX_FLAGS_RELEASE (#9394)
AdrianLundell Mar 28, 2025
6bc3e34
[Release 0.6] update version.txt (#9744)
metascroy Mar 28, 2025
ce74f8e
Openvino backend for Executorch to enable inference on Intel CPUs, GP…
ynimmaga Mar 28, 2025
ebe8522
Improve ET_CHECK_OR_RETURN_FALSE error messages in kernels/portable (…
swolchok Mar 28, 2025
7d35c68
[Bug Fix] Fix padding when running in NHWC (#9729)
mcr229 Mar 28, 2025
eee2bf1
Add bf16 support to unary_ufunc_realh
manuelcandales Mar 29, 2025
eda319e
Remove old tokenizer/ directory in ExecuTorch
larryliu0820 Mar 29, 2025
2aa7748
Update mimi export test
limintang Mar 29, 2025
5da974a
Introduce missing APIs to lower ExportedProgram objects directly
mcremon-meta Mar 31, 2025
eef0010
Qualcomm AI Engine Direct - Replace private variable with function (#…
shewu-quic Mar 31, 2025
69cc7fa
Arm backend: Check memory allocation on target (#9735)
zingo Mar 31, 2025
bad2fa9
Arm backend: Refactor any, bitwise, logical tests (#9499)
YufengShi-dudu Mar 31, 2025
7fd589d
Support non-default dim orders in pybindings + aten bridge (#9736)
GregoryComer Mar 31, 2025
cc231cf
Serialize NamedDataStoreOutput into PTD. (#9758)
pytorchbot Mar 31, 2025
11cf2c8
Add ET_PRI_* macros for Tensor methods that differ with ATen (#9768)
swolchok Mar 31, 2025
48de264
[Build] use timespec_get to get current time (#5252)
python3kgae Mar 31, 2025
97bca05
Fix link in using-executorch-building-from-source.md (#9756)
kirklandsign Mar 31, 2025
8ca1507
Fix QNN CI job (#9774)
larryliu0820 Mar 31, 2025
ea2029d
Update pybinding README (#9770)
jathu Mar 31, 2025
16e5901
Turn on XNNPACK by default for wheels (#9773)
jathu Mar 31, 2025
a5994ac
[XNNPACK] Serialize weights as fp16 rather than fp32 (#9753)
mcr229 Mar 31, 2025
2972388
Update docs and pins (#9781)
shoumikhin Mar 31, 2025
be92fb4
Add im2row per tensor overload
mcremon-meta Apr 1, 2025
f95d42f
Support llama3 training
dpalmasan Apr 1, 2025
d6bc799
fix rsqrt test case
zonglinpeng Apr 1, 2025
de7c9cf
Move replacement of conv to linear/fc to opt level 2
mcremon-meta Apr 1, 2025
88bf6d6
Bump nightly version in xcode projects (#9790)
shoumikhin Apr 1, 2025
7cc25b1
Add a util to print tag easily
larryliu0820 Apr 1, 2025
357ed9d
Error handling log intermediate output delegate (#9761)
dillondesilva Apr 1, 2025
dc93fde
Arm backend: update docs to new template (#9712)
Erik-Lundell Apr 1, 2025
5cc98bc
Arm backend: Use dbg_fail when node visitors raise exceptions (#9391)
oscarandersson8218 Apr 1, 2025
feb3fcd
Fix test_mimi
larryliu0820 Apr 1, 2025
3e62f9e
Arm backend: Make passes preserve and update node metadata (#9362)
oscarandersson8218 Apr 1, 2025
a5e326a
Add support for torch.pow in the Arm backend (#9309)
martinlsm Apr 1, 2025
77c35f5
Arm Backend: temp fix for flaky eq op test (#9794)
fumchin Apr 1, 2025
bc3d437
Arm backend: Change _is_ok_for_quantization to support output check (…
Sebastian-Larsson Apr 1, 2025
97ad3c6
Complex Support: diagonal_copy
manuelcandales Apr 1, 2025
5f01843
Update Codegen.cmake for improved Python package path handling (#6730)
AkiSakurai Apr 1, 2025
bcf4b46
Arm backend: Add additional unsupported checks to Ethos-U55 backend (…
Erik-Lundell Apr 1, 2025
655531f
[ET-VK] Store weights transposed for int8 linear (#9803)
pytorchbot Apr 1, 2025
97bb055
[ET-VK] Efficient tiled int8 matmul (#9804)
pytorchbot Apr 1, 2025
354c57e
Add try-before-set tests for DataSink in DebugEvent and LogDelegateIn…
keyprocedure Apr 1, 2025
d023a77
[ExecuTorch][Llama] Change runner to enable chunked prefill (#9805)
pytorchbot Apr 1, 2025
b66c319
Try upgrading to CoreML Tools 8.2 (#9799)
mergennachin Apr 1, 2025
bd5dc27
[ET-VK] Simplify lane offset copy logic in copy_packed_dim_offset sha…
pytorchbot Apr 1, 2025
1379e9f
[ET-VK] Adding all tensor packing support for repeat op. (#9813)
pytorchbot Apr 1, 2025
19950b3
[ET-VK] Moving repeat functionality from copy_packed_dim_offset into …
pytorchbot Apr 1, 2025
d0bf832
Change training pybind extension_module linkage to static (#9809)
larryliu0820 Apr 1, 2025
150cbe1
Allow passing in calibration data to `convert_pt2`
mcremon-meta Apr 1, 2025
a787cbd
fix qnn export
cccclai Apr 2, 2025
ade2e3c
Add cadence.where_Scalar op
mcremon-meta Apr 2, 2025
5fcb6b8
[Executorch][custo ops] Add prototype defs for custom op (#9815)
kirklandsign Apr 2, 2025
108fa3b
executorch
gmagogsfm Apr 2, 2025
4987f0b
Fix Fake bench data type (#9731)
yangw-dev Apr 2, 2025
376d66a
Arm backend: Fixes related to PyTorch 2.8.0 updates (#9834)
SaoirseARM Apr 2, 2025
1572381
Arm backend: Disable test case test_w2l_arm.py::test_w2l_u85_BI (#9835)
martinlsm Apr 2, 2025
7191604
Make Android Module thread-safe and prevent destruction during inference
GregoryComer Apr 2, 2025
7d36e41
Add README.md for the NXP eIQ Neutron Backend (#9718)
robert-kalmar Apr 2, 2025
40443a9
Make PyTorch headers available in optimized_portable_kernels, define …
swolchok Apr 2, 2025
1e64fa2
A short term fix for editable mode install failed to import root leve…
larryliu0820 Apr 2, 2025
cc73d03
introduce DelegateDebugIntId
Gasoonjia Apr 2, 2025
753a88e
Update android docs (#9788)
kirklandsign Apr 2, 2025
2f408dd
Qualcomm AI Engine Direct - Add QNN support for to_edge_transform_and…
shewu-quic Apr 2, 2025
fb1faaf
Support multiple prompts in the runner
cccclai Apr 3, 2025
8606725
Update static attention IO manager to use "smart mask" style update
sxu Apr 3, 2025
3600d4f
Arm backend: Size Test for Cortex-M (#9569)
digantdesai Apr 3, 2025
fe0aee3
Arm backend: Improve pre-push hook (#9873)
Sebastian-Larsson Apr 3, 2025
b5bc1e6
Arm backend: Convert assert to throw TypeError in arm_pass_utils (#9866)
Sebastian-Larsson Apr 3, 2025
9e7b469
Arm backend: Add where.self (#9869)
Sebastian-Larsson Apr 3, 2025
096fa1e
[ET-VK] Adding all tensor packing support for native layer norm. (#9870)
pytorchbot Apr 3, 2025
ce36360
[ET-VK] Adding round op support. (#9871)
pytorchbot Apr 3, 2025
4f748fe
[ET-VK] Replace Uniform buffers with push constants for native layer …
pytorchbot Apr 3, 2025
0844c38
Arm backend: Limit number of build jobs (#9874)
mansnils Apr 3, 2025
ab3b3f0
Fix scalar type logging from #9751 (#9845)
swolchok Apr 3, 2025
a884e92
Arm backend: Convert assert to throw ValueError in op_sigmoid (#9875)
Sebastian-Larsson Apr 3, 2025
3fc1a97
RMSNorm support - Executorch
ThomasJannaud Apr 3, 2025
8ef1f69
Create release.yml (#9884)
metascroy Apr 3, 2025
e9c2315
forward fix
cccclai Apr 3, 2025
7e487c2
Support QK norm in static attention
sxu Apr 4, 2025
4717459
Update test build instructions in kernels readme
manuelcandales Apr 4, 2025
a5967fd
Arm backend: Add expected failure for test_torch_functions.py (#9898)
SaoirseARM Apr 4, 2025
0efdb91
Arm backend: Add ERF operator (#9836)
maddun01 Apr 4, 2025
b7f35c9
Arm backend: Add info message to assertions in fold_qdq_with_annotate…
Sebastian-Larsson Apr 4, 2025
a4002e8
Arm backend: Convert assert to throw ValueError in op_tanh (#9896)
Sebastian-Larsson Apr 4, 2025
1231aeb
Arm backend: Convert assert to throw TypeError in op_add (#9897)
Sebastian-Larsson Apr 4, 2025
b367540
Arm backend: Align MobileNetV2 with MobileNetV3 unittest (#9900)
oscarandersson8218 Apr 4, 2025
d48d7a9
Arm backend: Remove use of TOSA_DBG_VERBOSE (#9901)
oscarandersson8218 Apr 4, 2025
cfca108
Arm backend: Add support for Leaky ReLU (#9903)
gggekov Apr 4, 2025
2b4e525
Arm backend: Handle memory mode corectly in test_model.py (#9902)
zingo Apr 4, 2025
c70c208
Arm backend: Convert assert to throw ValueError in tosa_backend (#9905)
Sebastian-Larsson Apr 4, 2025
dd9746e
Added hardtanh operator
ckmadhira Apr 4, 2025
bd98b5e
[Executorch][sdpa] Refactor sdpa into impl and op (#9909)
pytorchbot Apr 4, 2025
ab0558b
Fix build gating on optimized_portable_kernels (#9838)
swolchok Apr 4, 2025
7538d9b
set INTERFACE_LINK_LIBARIES for extension_threadpool (#9839)
swolchok Apr 4, 2025
a40ed4c
[Executorch][sdpa] Add accidentaly removed flash attentiona args chec…
pytorchbot Apr 4, 2025
356d4f9
[Executorch][SDPA] Remove slice creation (#9911)
pytorchbot Apr 4, 2025
95d38c4
[Executorch][sdpa] Setup the structure to enable quantized gemms for …
pytorchbot Apr 4, 2025
bf3e0af
Arm backend: Skip instead of returning in Llama test (#9906)
martinlsm Apr 4, 2025
9eea1f1
Bump PyTorch pin to 2025-03-25 (#9578)
swolchok Apr 4, 2025
5dcd3e1
Arm backend: Reduce arm_executor_runner binary size (#9907)
AdrianLundell Apr 4, 2025
730a4d8
Add optimized ELU implementation (#9521)
swolchok Apr 4, 2025
d6e14fc
Arm backend: Add ethos_u_minimal_example jupyter notebook (#9543)
AdrianLundell Apr 4, 2025
91c9417
[ExecuTorch][to_backend] add AllNodePartitioner (#9822)
mcr229 Apr 4, 2025
a55367b
Expose symbols on macos in the xplat pytorch stack
stepanhruda Apr 4, 2025
56c8dc2
etdumpfilter impl
Gasoonjia Apr 4, 2025
b7eee0c
Reapply "Depend on extension/threadpool, not thread_parallel_interfac…
kirklandsign Apr 5, 2025
6adff9c
Support slice ops with default start
pssrawat Apr 6, 2025
302cb06
Arm backend: Add pytest.mark.flaky on U85 tests in test_mm.py (#9926)
martinlsm Apr 7, 2025
4224d73
Arm backend: Convert assert to throw ValueError in op_exp (#9929)
Sebastian-Larsson Apr 7, 2025
491096e
Arm backend: Convert assert to raise ValueError for comparison operat…
Sebastian-Larsson Apr 7, 2025
709560f
Qualcomm AI Engine Direct - Fix mobilebert finetune script (#9927)
shewu-quic Apr 7, 2025
2d56897
aten.leakyrelu.default Op registery
hossein1387 Apr 7, 2025
5e9e9d1
Update release.yaml (#9891)
metascroy Apr 7, 2025
789799b
Add code structure and a few other links to CONTRIBUTING.md (#9793)
larryliu0820 Apr 7, 2025
1f5ca0c
Add a path to use quantized gemm from torchao in sdpa
kimishpatel Apr 7, 2025
68c84cc
per_channel_group can't be dynamic
mcr229 Apr 7, 2025
52b4483
Update Executorch ops registration for rms_norm (#9920)
Vysarat Apr 7, 2025
0ebb4ec
[Easy] Fix numbering typo in Llama README docs (#9936)
Jack-Khuu Apr 7, 2025
031459f
[ET-VK] Improve packing format for int4 linear operator + misc improv…
SS-JIA Apr 7, 2025
1bef520
[ET-VK][ez] Make squeeze insertion requirements more strict (#9950)
SS-JIA Apr 7, 2025
2cce2db
[ET-VK][ez] Allow logit linear layer to be lowered to Vulkan (#9951)
SS-JIA Apr 7, 2025
aefffda
Expose L4 ops to ExecuTorch Client and add MWA to ExecuTorch Client
derekxu Apr 8, 2025
db21501
introducing filter to etdumpgen
Gasoonjia Apr 8, 2025
1e97232
Arm backend: Add pre-push checks for op tests (#9899)
AdrianLundell Apr 8, 2025
a67b6b8
Fix naming convention in quantizer
mcremon-meta Apr 8, 2025
0ee74ff
Minor fixes on Intro How It Works (#9939)
mergennachin Apr 8, 2025
5c2b693
executorch/backends/xnnpack/test/ops
gmagogsfm Apr 8, 2025
1facfa9
Allow emitting mutable buffer names in schema
JacobSzwejbka Apr 8, 2025
87aa699
Update test_ios_ci.sh to use the app from executorch-examples repo (#…
shoumikhin Apr 8, 2025
3dc7639
android-release-artifacts.yml allow upload AAR to maven (#9954)
kirklandsign Apr 8, 2025
f9c3f63
[Release 0.6] update torchao pin (#9947)
metascroy Apr 8, 2025
a238d6a
Update test_ios.sh (#9968)
shoumikhin Apr 8, 2025
8d80185
Update demo-apps-ios.md (#9970)
shoumikhin Apr 8, 2025
040ff36
Getting Started, compare against reference eager model for verificati…
mergennachin Apr 8, 2025
7e8acb9
Add python version in building from source page (#9975)
mergennachin Apr 8, 2025
e6c2907
Delete examples/demo-apps/apple_ios/ExecuTorchDemo directory (#9976)
shoumikhin Apr 8, 2025
46ce593
Add op_amax support (#9955)
cccclai Apr 9, 2025
dea14b1
Modify executorch_module_static and executorch_tensor spelling proble…
zxc503 Apr 9, 2025
f28b5db
Just build AAR in place
kirklandsign Apr 9, 2025
c2e3e17
[ez][release blocker fix] Insert `linalg_vector_norm` into decomp tab…
SS-JIA Apr 9, 2025
9b7a878
Add maven version in main in Getting Started page (#9980)
mergennachin Apr 9, 2025
3a940da
[doc] Fix tokenizer related documentation (#10000)
larryliu0820 Apr 9, 2025
20c7047
Fix tokenizer convert in xnnpack_README.md (#10003)
larryliu0820 Apr 9, 2025
a1af1ff
using-executorch-android.md: Use an easier example (#10008)
kirklandsign Apr 9, 2025
9b697ae
[Doc] Update getting-started-architecture.md (#10027)
iseeyuan Apr 9, 2025
4bd7798
Typo updates for README and Contributions (#10025)
Jack-Khuu Apr 9, 2025
1d43b3b
Qualcomm AI Engine Direct - Add submodule quant config setting (#9355)
chunit-quic Apr 10, 2025
7e8eb2b
Fix LLM getting-started.md (#10028)
kirklandsign Apr 10, 2025
0bc52fe
Fix deprecated unittest asserts in executorch
itamaro Apr 10, 2025
ad6f5ee
Add a namespace for ATen mode
larryliu0820 Apr 10, 2025
c9c5481
remove reduntant log_delegate_intermediate_logging_helper call
Gasoonjia Apr 10, 2025
f56a65e
Arm backend: Set all 16/32 bit sigmoid tests to flaky (#9993)
martinlsm Apr 10, 2025
2fa749c
Arm backend: Raise atol in MobileNetV3 unit tests (#9959)
martinlsm Apr 10, 2025
3ec2a7c
Add examples for CPP demo app (#10022)
mergennachin Apr 10, 2025
f2a08da
[arm][ez] Add `xfail` for `norm` tests (#10043)
SS-JIA Apr 10, 2025
6c63cc9
Add link to executorch-examples in module/c++ doc (#10026)
lucylq Apr 10, 2025
fcf431f
Arm Backend: Add Tiny Stories llama test case for TOSA BI (#9996)
ArmRyan Apr 10, 2025
b8b3819
[Executorch][SDPA] Refactor + Make quantized sdpa handle sequence at …
pytorchbot Apr 10, 2025
8d6aa35
[Executorch][llama] Renamed quantized_kv_cache to custom_kv_cache (#1…
pytorchbot Apr 10, 2025
40beade
[Executorch][llama] Enable quantized sdpa (#10062)
pytorchbot Apr 10, 2025
ebe0c81
Fix up some Docs (#10038)
mcr229 Apr 10, 2025
5e4f045
Qualcomm AI Engine Direct - oss model enablement (EfficientSAM) (#9266)
DannyYuyang-quic Apr 10, 2025
009efdc
memory_planning algos take the specs as inputs instead of calculating…
JacobSzwejbka Apr 10, 2025
060cda3
Update kernels readme
manuelcandales Apr 10, 2025
2ed44ce
[#9971] Gracefully error out in ETDump for get_flatbuffer_scalar_type…
pytorchbot Apr 10, 2025
d42676f
Llama exported model from executorch-community (#10064)
mergennachin Apr 10, 2025
7f3b382
Qualcomm AI Engine Direct - Support tile op for different I/O rank (#…
DannyYuyang-quic Apr 10, 2025
cea9d1c
llama doc update (#10082)
lucylq Apr 10, 2025
c738625
Update using-executorch-android.md (#10084)
kirklandsign Apr 10, 2025
57ae20e
Update LlamaDemo AAR build docs (#10086)
kirklandsign Apr 10, 2025
c925fe0
Improve android instrumentation test experience and add docs (#9989)
kirklandsign Apr 10, 2025
5e11bd9
fix lint
cccclai Apr 10, 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
  •  
  •  
  •  
The diff you're trying to view is too large. We only load the first 3000 changed files.
File renamed without changes.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
Empty file.
25 changes: 22 additions & 3 deletions .buckconfig
Original file line number Diff line number Diff line change
Expand Up @@ -8,18 +8,37 @@
root = .
prelude = third-party/prelude
shim = shim
shim_et = shim_et

[repository_aliases]
bazel_skylib = shim
config = prelude
ovr_config = prelude
toolchains = shim
fbcode = shim
toolchains = shim_et
fbcode = shim_et
fbcode_macros = shim
fbsource = shim
fbsource = shim_et
buck = shim
gh_facebook_buck2_shims_meta = shim

[cxx]
cxxflags = -g -std=c++17

[parser]
target_platform_detector_spec = target:root//...->prelude//platforms:default target:shim//...->prelude//platforms:default

# Limit the number of files that the buck daemon needs to monitor. If every
# submodule is cloned recursively, some system can fail to build with "OS file
# watch limit reached".
[project]
ignore = \
.git, \
**/.git, \
cmake-out, \
pip-out

[buck2]
restarter=true

[oss]
folly_cxx_tests = False
22 changes: 22 additions & 0 deletions .ci/docker/build.sh
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,10 @@ case "${IMAGE_NAME}" in
LINTRUNNER=""
CLANG_VERSION=12
;;
executorch-ubuntu-22.04-gcc11-aarch64)
LINTRUNNER=""
GCC_VERSION=11
;;
executorch-ubuntu-22.04-linter)
LINTRUNNER=yes
CLANG_VERSION=12
Expand All @@ -37,6 +41,21 @@ case "${IMAGE_NAME}" in
ARM_SDK=yes
CLANG_VERSION=12
;;
executorch-ubuntu-22.04-qnn-sdk)
QNN_SDK=yes
CLANG_VERSION=12
;;
executorch-ubuntu-22.04-mediatek-sdk)
MEDIATEK_SDK=yes
CLANG_VERSION=12
ANDROID_NDK_VERSION=r27b
;;
executorch-ubuntu-22.04-clang12-android)
LINTRUNNER=""
CLANG_VERSION=12
# From https://developer.android.com/ndk/downloads
ANDROID_NDK_VERSION=r27b
;;
*)
echo "Invalid image name ${IMAGE_NAME}"
exit 1
Expand Down Expand Up @@ -66,6 +85,9 @@ docker build \
--build-arg "LINTRUNNER=${LINTRUNNER:-}" \
--build-arg "BUILD_DOCS=${BUILD_DOCS}" \
--build-arg "ARM_SDK=${ARM_SDK:-}" \
--build-arg "QNN_SDK=${QNN_SDK:-}" \
--build-arg "MEDIATEK_SDK=${MEDIATEK_SDK:-}" \
--build-arg "ANDROID_NDK_VERSION=${ANDROID_NDK_VERSION:-}" \
-f "${OS}"/Dockerfile \
"$@" \
.
2 changes: 1 addition & 1 deletion .ci/docker/ci_commit_pins/buck2.txt
Original file line number Diff line number Diff line change
@@ -1 +1 @@
2024-02-15
2024-12-16
2 changes: 1 addition & 1 deletion .ci/docker/ci_commit_pins/pytorch.txt
Original file line number Diff line number Diff line change
@@ -1 +1 @@
f5b99976adcbb01fd71bd0a39ea15bdac6c9e48a
59d5cf083b4f860dea76fe8936076177f9367f10
82 changes: 82 additions & 0 deletions .ci/docker/common/install_android.sh
Original file line number Diff line number Diff line change
@@ -0,0 +1,82 @@
#!/bin/bash
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree.

set -ex

# Double check if the NDK version is set
[ -n "${ANDROID_NDK_VERSION}" ]

install_prerequiresites() {
OS=$(grep -oP '(?<=^ID=).+' /etc/os-release | tr -d '"')
case "$OS" in
amzn)
# https://docs.aws.amazon.com/corretto/latest/corretto-17-ug/amazon-linux-install.html
yum install -y java-17-amazon-corretto \
ca-certificates \
ant
;;
*)
apt-get update

# NB: Need OpenJDK 17 at the minimum
apt-get install -y --no-install-recommends \
openjdk-17-jdk \
ca-certificates-java \
ant

# Cleanup package manager
apt-get autoclean && apt-get clean
rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
;;
esac
}

install_ndk() {
NDK_INSTALLATION_DIR=/opt/ndk
rm -rf "${NDK_INSTALLATION_DIR}" && mkdir -p "${NDK_INSTALLATION_DIR}"

pushd /tmp
# The NDK installation is cached on ossci-android S3 bucket
curl -Os --retry 3 "https://ossci-android.s3.amazonaws.com/android-ndk-${ANDROID_NDK_VERSION}-linux.zip"
unzip -qo "android-ndk-${ANDROID_NDK_VERSION}-linux.zip"

# Print the content for manual verification
ls -lah "android-ndk-${ANDROID_NDK_VERSION}"
mv "android-ndk-${ANDROID_NDK_VERSION}"/* "${NDK_INSTALLATION_DIR}"

popd
}

install_cmdtools() {
CMDTOOLS_FILENAME=commandlinetools-linux-11076708_latest.zip

pushd /tmp
# The file is cached on ossci-android S3 bucket
curl -Os --retry 3 "https://ossci-android.s3.us-west-1.amazonaws.com/${CMDTOOLS_FILENAME}"
unzip -qo "${CMDTOOLS_FILENAME}" -d /opt

ls -lah /opt/cmdline-tools/bin
popd
}

install_sdk() {
SDK_INSTALLATION_DIR=/opt/android/sdk
rm -rf "${SDK_INSTALLATION_DIR}" && mkdir -p "${SDK_INSTALLATION_DIR}"

# These are the tools needed to build Android apps
yes | /opt/cmdline-tools/bin/sdkmanager --sdk_root="${SDK_INSTALLATION_DIR}" --install "platforms;android-34"
yes | /opt/cmdline-tools/bin/sdkmanager --sdk_root="${SDK_INSTALLATION_DIR}" --install "build-tools;33.0.1"
yes | /opt/cmdline-tools/bin/sdkmanager --sdk_root="${SDK_INSTALLATION_DIR}" --install "build-tools;35.0.0"
# And some more tools for future emulator tests
yes | /opt/cmdline-tools/bin/sdkmanager --sdk_root="${SDK_INSTALLATION_DIR}" --install "platform-tools"
yes | /opt/cmdline-tools/bin/sdkmanager --sdk_root="${SDK_INSTALLATION_DIR}" --install "tools"
}

install_prerequiresites
install_ndk
install_cmdtools
install_sdk
8 changes: 7 additions & 1 deletion .ci/docker/common/install_base.sh
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,13 @@ install_ubuntu() {
unzip \
gdb \
rsync \
libssl-dev
libssl-dev \
zip

# These libraries are needed by TorchVision
apt-get install -y --no-install-recommends \
libjpeg-dev \
libpng-dev

# Cleanup package manager
apt-get autoclean && apt-get clean
Expand Down
52 changes: 45 additions & 7 deletions .ci/docker/common/install_cache.sh
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,26 @@ set -ex
# shellcheck source=/dev/null
source "$(dirname "${BASH_SOURCE[0]}")/utils.sh"

install_ubuntu() {
echo "Preparing to build sccache from source"
apt-get update
# libssl-dev will not work as it is upgraded to libssl3 in Ubuntu-22.04.
# Instead use lib and headers from OpenSSL1.1 installed in `install_openssl.sh``
apt-get install -y cargo
echo "Checking out sccache repo"
git clone https://github.com/mozilla/sccache -b v0.8.2

cd sccache
echo "Building sccache"
cargo build --release
cp target/release/sccache /opt/cache/bin
echo "Cleaning up"
cd ..
rm -rf sccache
apt-get remove -y cargo rustc
apt-get autoclean && apt-get clean
}

install_binary() {
echo "Downloading sccache binary from S3 repo"
curl --retry 3 https://s3.amazonaws.com/ossci-linux/sccache -o /opt/cache/bin/sccache
Expand All @@ -22,15 +42,33 @@ mkdir -p /opt/cache/bin
sed -e 's|PATH="\(.*\)"|PATH="/opt/cache/bin:\1"|g' -i /etc/environment
export PATH="/opt/cache/bin:$PATH"

# NB: Install the pre-built binary from S3 as building from source
# https://github.com/pytorch/sccache has started failing mysteriously
# in which sccache server couldn't start with the following error:
# sccache: error: Invalid argument (os error 22)
install_binary
install_ubuntu

function write_sccache_stub() {
BINARY=$1
printf "#!/bin/sh\nif [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then\n exec sccache %s \"\$@\"\nelse\n exec %s \"\$@\"\nfi" "$(which "${BINARY}")" "$(which "${BINARY}")" > "/opt/cache/bin/${BINARY}"
if [ $1 == "gcc" ]; then
# Do not call sccache recursively when dumping preprocessor argument
# For some reason it's very important for the first cached nvcc invocation
cat >"/opt/cache/bin/$1" <<EOF
#!/bin/sh
if [ "\$1" = "-E" ] || [ "\$2" = "-E" ]; then
exec $(which $1) "\$@"
elif [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then
exec sccache $(which $1) "\$@"
else
exec $(which $1) "\$@"
fi
EOF
else
cat >"/opt/cache/bin/$1" <<EOF
#!/bin/sh
if [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then
exec sccache $(which $1) "\$@"
else
exec $(which $1) "\$@"
fi
EOF
fi
chmod a+x "/opt/cache/bin/${BINARY}"
}

Expand All @@ -44,7 +82,7 @@ init_sccache() {

# NB: This function is adopted from PyTorch core at
# https://github.com/pytorch/pytorch/blob/main/.ci/pytorch/common-build.sh
as_ci_user sccache --stop-server > /dev/null 2>&1 || true
as_ci_user sccache --stop-server >/dev/null 2>&1 || true
rm -f "${SCCACHE_ERROR_LOG}" || true

# Clear sccache stats before using it
Expand Down
2 changes: 1 addition & 1 deletion .ci/docker/common/install_clang.sh
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ install_ubuntu() {
apt-get install -y --no-install-recommends clang-"$CLANG_VERSION"
apt-get install -y --no-install-recommends llvm-"$CLANG_VERSION"
# Also require LLD linker from llvm and libomp to build PyTorch from source
apt-get install -y lld "libomp-${CLANG_VERSION}-dev"
apt-get install -y lld "libomp-${CLANG_VERSION}-dev" "libc++-${CLANG_VERSION}-dev"

# Use update-alternatives to make this version the default
update-alternatives --install /usr/bin/clang clang /usr/bin/clang-"$CLANG_VERSION" 50
Expand Down
23 changes: 18 additions & 5 deletions .ci/docker/common/install_conda.sh
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,9 @@ source "$(dirname "${BASH_SOURCE[0]}")/utils.sh"
install_miniconda() {
BASE_URL="https://repo.anaconda.com/miniconda"
CONDA_FILE="Miniconda3-py${PYTHON_VERSION//./}_${MINICONDA_VERSION}-Linux-x86_64.sh"
if [[ $(uname -m) == "aarch64" ]]; then
CONDA_FILE="Miniconda3-py${PYTHON_VERSION//./}_${MINICONDA_VERSION}-Linux-aarch64.sh"
fi

mkdir -p /opt/conda
chown ci-user:ci-user /opt/conda
Expand All @@ -31,8 +34,16 @@ install_miniconda() {

install_python() {
pushd /opt/conda
# Install the correct Python version
# Install the selected Python version for CI jobs
as_ci_user conda create -n "py_${PYTHON_VERSION}" -y --file /opt/conda/conda-env-ci.txt python="${PYTHON_VERSION}"

# From https://github.com/pytorch/pytorch/blob/main/.ci/docker/common/install_conda.sh
if [[ $(uname -m) == "aarch64" ]]; then
conda_install "openblas==0.3.29=*openmp*" -c conda-forge
else
conda_install mkl=2022.1.0 mkl-include=2022.1.0
fi

popd
}

Expand All @@ -44,14 +55,16 @@ install_pip_dependencies() {
}

fix_conda_ubuntu_libstdcxx() {
cat /etc/issue
# WARNING: This is a HACK from PyTorch core to be able to build PyTorch on 22.04.
# The issue still exists with the latest conda 23.10.0-1 at the time of writing
# (2023/11/16).
# Specifically, ubuntu-20+ all comes lib libstdc++ newer than 3.30+, but anaconda
# is stuck with 3.29. So, remove libstdc++6.so.3.29 as installed by
# https://anaconda.org/anaconda/libstdcxx-ng/files?version=11.2.0
#
# PyTorch sev: https://github.com/pytorch/pytorch/issues/105248
# Ref: https://github.com/pytorch/pytorch/blob/main/.ci/docker/common/install_conda.sh
if grep -e "[12][82].04.[623]" /etc/issue >/dev/null; then
rm "/opt/conda/envs/py_${PYTHON_VERSION}/lib/libstdc++.so.6"
if grep -e "2[02].04." /etc/issue >/dev/null; then
rm /opt/conda/envs/py_${PYTHON_VERSION}/lib/libstdc++.so*
fi
}

Expand Down
12 changes: 12 additions & 0 deletions .ci/docker/common/install_java.sh
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
#!/bin/bash
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree.

set -ex

apt-get update

apt-get install -y --no-install-recommends openjdk-17-jdk
4 changes: 4 additions & 0 deletions .ci/docker/common/install_linter.sh
Original file line number Diff line number Diff line change
Expand Up @@ -13,3 +13,7 @@ source "$(dirname "${BASH_SOURCE[0]}")/utils.sh"
# NB: Install all linter dependencies, the caching of lintrunner init could be
# done after Executorch becomes public
pip_install -r requirements-lintrunner.txt

# Install google-java-format
curl -L --retry 3 https://github.com/google/google-java-format/releases/download/v1.23.0/google-java-format_linux-x86-64 > /opt/google-java-format
chmod +x /opt/google-java-format
2 changes: 1 addition & 1 deletion .ci/docker/common/install_pytorch.sh
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ install_pytorch_and_domains() {

chown -R ci-user .

export _GLIBCXX_USE_CXX11_ABI=0
export _GLIBCXX_USE_CXX11_ABI=1
# Then build and install PyTorch
conda_run python setup.py bdist_wheel
pip_install "$(echo dist/*.whl)"
Expand Down
5 changes: 4 additions & 1 deletion .ci/docker/conda-env-ci.txt
Original file line number Diff line number Diff line change
@@ -1,2 +1,5 @@
cmake=3.22.1
cmake=3.26.4
ninja=1.10.2
libuv
llvm-openmp
pkg-config
7 changes: 4 additions & 3 deletions .ci/docker/requirements-ci.txt
Original file line number Diff line number Diff line change
@@ -1,13 +1,14 @@
mpmath==1.3.0
numpy==1.25.2
numpy>=2.0.0; python_version >= '3.10'
PyYAML==6.0.1
ruamel.yaml==0.17.32
sympy==1.12
timm==0.6.13
tomli==2.0.1
torchsr==1.0.4
transformers==4.36.0
transformers==4.47.1
zstd==1.5.5.1
pandas>=2.2.2; python_version >= '3.10'
pytest==7.2.0
pytest-cov==4.1.0
expecttest==0.1.6
Expand All @@ -20,7 +21,7 @@ sphinx-gallery==0.14.0
breathe==4.34.0
exhale==0.2.3
docutils==0.16
matplotlib==3.7.2
matplotlib>=3.9.4
# PyTorch Theme
-e git+https://github.com/pytorch/pytorch_sphinx_theme.git#egg=pytorch_sphinx_theme
myst-parser==0.18.1
Expand Down
Loading