Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
226 commits
Select commit Hold shift + click to select a range
c798024
Add NPU device; Revert num_of_threads to 1 to be default
sspintel Nov 3, 2023
630d57d
Merge branch 'master' into openvino-ep-5.2-npu
saurabhkale17 Nov 10, 2023
119e86e
SDXL demo: Add Option to disable refiner (#18455)
tianleiwu Nov 16, 2023
999752a
[WebNN EP] Support GreaterOrEqual and LessOrEqual ops (#18411)
Honry Nov 16, 2023
b291b20
[JS/Web]Added uniforms support to Slice op. (#18422)
satyajandhyala Nov 16, 2023
3588fba
[TensorRT EP] Fix memory leak for cudnn/cublas (#18467)
chilo-ms Nov 16, 2023
b6b9aff
Allow empty shapes and do not validate them for inputs/outputs (#18442)
yuslepukhin Nov 16, 2023
e7a524f
Update to allow large models to be checked for mobile support. (#18357)
skottmckay Nov 16, 2023
6a4e448
[QNN EP] Support Qnn MatMul with 2 dynamic inputs which are uint16 qu…
HectorSVC Nov 16, 2023
adb56df
Aciddelgado/gqa local (#18375)
aciddelgado Nov 16, 2023
f17b6af
[TensorRT EP] Fix bug for no nodes in subgraph at GetCapability (#18449)
chilo-ms Nov 17, 2023
d73073d
remove full protobuf requirement for tensorrt ep (#18413)
jywu-msft Nov 17, 2023
5eb5056
Always run emsdk_env.sh before build.py, even when ccache is disabled…
snnn Nov 17, 2023
1a29460
rope support 4D input tensor (#18454)
kailums Nov 17, 2023
a5537f2
[WebNN Ep] Slice's axes and steps inputs should be constant initializ…
Honry Nov 17, 2023
fac3e33
[js/web] JSEP Attention & MultiHeadAttention (#17742)
dakenf Nov 17, 2023
41f9379
Update NDK version to 26.1.10909125 (#18493)
snnn Nov 17, 2023
cbb85b4
[CoreML] Adapt to `MLMultiArray.dataPointer` deprecation (#17726)
NickLucche Nov 18, 2023
0233329
Removed all the deprecated python training code and related tests and…
askhade Nov 18, 2023
34c5424
[js] update a few packages (#18499)
fs-eire Nov 18, 2023
9364c05
Update web-ci.yml: remove depth=1 (#18500)
snnn Nov 18, 2023
53917a3
Move up members in Lite Custom Op hierarchy for possible memleaks. (#…
RandySheriffH Nov 18, 2023
97cc40d
Add fusion patterns for conformer-transducer model (#18461)
apsonawane Nov 19, 2023
dc9ab4f
Update setup.py: replace libcudart.so.12.0 with libcudart.so.12 (#18501)
snnn Nov 20, 2023
3bcc137
Tiny change to trigger the update of DORT's CI image (#18507)
wschin Nov 20, 2023
d97fc18
Create a new Python Package pipeline for CUDA 12 (#18348)
jchen351 Nov 20, 2023
1af0681
Bfloat16 support for MatMulBnb4, Training support bitsandbytes>=0.41.…
jambayk Nov 20, 2023
1dd9bf5
Remove setup_env_azure.bat (#18482)
jchen351 Nov 20, 2023
247ce21
[js] optimize eslint config (#18460)
fs-eire Nov 20, 2023
cc54202
Create edges with arg positons correctly accounting for non-existing …
yuslepukhin Nov 20, 2023
abdf8b7
[js/webgpu] Optimize broadcast binary. (#18185)
qjia7 Nov 21, 2023
c7fd930
[js/web] unify resolve rules for "Clip" (#18527)
fs-eire Nov 21, 2023
a608c00
fix past-kv in general LLM exporter (#18529)
wejoncy Nov 21, 2023
29a409a
Add missing flags DISABLE_FLOAT8_TYPES in GemmFloat8 custom operator …
xadupre Nov 21, 2023
2a01622
Hide NPU Adapter selection behind macro (#18515)
smk2007 Nov 21, 2023
680a526
Training packaging pipeline for cuda12 (#18524)
ajindal1 Nov 21, 2023
81a763a
Make TensorShapeVector to use InlinedVector<Int64_t> to reduce on tem…
yuslepukhin Nov 21, 2023
ac8598a
[js/webgpu] enable f16 for concat (#18528)
qjia7 Nov 21, 2023
d455b0f
[js/web] use Chrome in CI for npm tests (#18522)
fs-eire Nov 22, 2023
62da3b1
SDXL Latent Consistency Model (LCM) optimization (#18526)
tianleiwu Nov 22, 2023
7c57305
[QDQ Optimizer] Fix logic that drops Q/DQ ops from QDQ split node gro…
adrianlizarraga Nov 22, 2023
3bc9efc
[ORTModule] Adjust Attention Patterns for Efficient Attention ATen Fa…
Lafi7e Nov 22, 2023
89723c8
[WebNN EP] Mark and fallback unsupported op for WebNN CPU backend (#1…
Honry Nov 22, 2023
32fabb5
Fix opset version of the optimizer in function generate_artifacts (#1…
xadupre Nov 22, 2023
3f0ebd6
Fix opset import in GemmFloat8 python unit tests (#18489)
xadupre Nov 22, 2023
1c555c5
[JS/Web] Resize & BiasSplitGelu fp16 support (#18536)
dakenf Nov 22, 2023
841f7ed
[[JS/Web]Added uniform to Expand op. (#18558)
satyajandhyala Nov 22, 2023
42c6799
Update transpose optimization to be more QDQ aware (#18444)
skottmckay Nov 22, 2023
fa10694
[js/webgpu] Refactor matmul conv to support uniforms for matmul (#18452)
axinging Nov 22, 2023
64dacc2
[js/webgpu] Add BatchNormalization Op (#18468)
qjia7 Nov 22, 2023
43a5147
Memory optimization refactor and refinement (#17481)
pengwa Nov 23, 2023
1c79897
[QNN EP] Support LpNormalization (#18561)
adrianlizarraga Nov 23, 2023
6f3c1f9
[ROCm] Update ck for GemmFloat8 (#18487)
cloudhan Nov 23, 2023
62f00ad
[CoreML] Add Softmax and Split op support (#18358)
YUNQIUGUO Nov 23, 2023
b9c935f
[ROCm] Some fixes in tunable (#18575)
mindest Nov 24, 2023
7b2aefa
undo hipify of __half to rocblas_half (#18573)
TedThemistokleous Nov 24, 2023
2f60833
Setup default python formatter for new python plugin (#18563)
cloudhan Nov 24, 2023
a2fd8a6
[QNN EP] Return INVALID_GRAPH if failed to load from context binary (…
HectorSVC Nov 25, 2023
dd355e3
[js/web/training] Added parameters methods (#18250)
carzh Nov 27, 2023
b9fd9c5
remove dead code in openvino EP (#18457)
jslhcl Nov 27, 2023
fc8631e
[js/web] Fix conv2dMatmul errors due to #18452 (#18562)
qjia7 Nov 28, 2023
3f42fba
deleted the unused random_device variables because they caused a warn…
galran Nov 28, 2023
94a6020
Improve parallelization of TfIdfVectorizer, Reduce memory consumption…
xadupre Nov 28, 2023
3ea27c2
Create a new Nuget Package pipeline for CUDA 12 (#18135)
jchen351 Nov 28, 2023
a6d8726
Update ADO windows image to custom image (#18598)
mszhanyi Nov 28, 2023
0b7048e
Update winml to use #cores - #soc cores by Default as the number of i…
smk2007 Nov 28, 2023
8d5ecc4
[Quantization] Fix scale/zero-point for 16-bit QDQ Softmax (#18589)
adrianlizarraga Nov 28, 2023
05046e5
Adding unit test for sm80 prepack (#18514)
chenfucn Nov 28, 2023
288b80d
Add MacOS build to ORT C Pod (#18550)
YUNQIUGUO Nov 28, 2023
e24733c
fix the Olive CI pipeline failure on Windows (#18464)
guotuofeng Nov 28, 2023
a49f31b
Remove drop-nuget artifact from all pipelines (#18592)
jchen351 Nov 28, 2023
50e6235
[js/web] allow ShaderHelper to use internal (non-I/O) variables (#18525)
fs-eire Nov 28, 2023
f13380f
Support LoRA and Control Net in Stable Diffusion demo (#18593)
tianleiwu Nov 28, 2023
e833d22
Change QNN EP Profiling logs to output to CSV (#18201)
ivberg Nov 29, 2023
14a3434
Fix Objective-C static analysis build (#18606)
edgchen1 Nov 29, 2023
38b640c
[WebNN EP] Re-implement Unsqueeze, Squeeze, Flatten with WebNN's resh…
Honry Nov 29, 2023
6820930
Replace all Azure-Pipelines-EO-Windows2022-aiinfrat to Onnxruntime-Wi…
mszhanyi Nov 29, 2023
d2dfbf4
Add float16 type support to SplitToSequence and make code type indepe…
yuslepukhin Nov 29, 2023
483c490
Refine error checks in onnxruntime/core/providers/coreml/model/model.…
edgchen1 Nov 29, 2023
7335760
[JS/Web] Add uniforms to Einsum (#18531)
satyajandhyala Nov 29, 2023
227dcb3
[js/webgpu] Log the key and program info for artifact (#18365)
Nov 30, 2023
c20488c
skip_infer for SkipGroupNorm in SymbolicShapeInference (#18630)
jambayk Nov 30, 2023
5c67a00
Revert "remove full protobuf requirement for tensorrt ep" (#18626)
jywu-msft Nov 30, 2023
e1d1033
[ORTModule] Remove Unused Arguments from Generated Triton Code (#18636)
Lafi7e Nov 30, 2023
148495e
[ORTModule] Use Default Topo-order for GraphViewer (#18410)
Lafi7e Nov 30, 2023
1b5675f
Update post-merge-jobs.yml: increase timeout value for the Ios job (#…
snnn Nov 30, 2023
23a91c8
Fix warning C4003 in ORT python binding code (#18612)
snnn Nov 30, 2023
e7f64f4
[js/web] fix ESLint by excluding generated .js from tsconfig.json (#1…
fs-eire Nov 30, 2023
c5ea154
Eliminate intermediate string conversion buffer. (#18608)
yuslepukhin Nov 30, 2023
b1e749e
[js/webgpu] Add program name into webgpuProfiling info (#18640)
qjia7 Nov 30, 2023
4025bd8
[WebNN EP] Fix bug of padding in Op ConvTranspose (#18577)
Nov 30, 2023
efee9ab
Reduce downloads in Nuget-Java pipeline to reduce connection exceptio…
mszhanyi Nov 30, 2023
6781b6c
[js/webgpu] add bool type for Expand/Gather (#18615)
qjia7 Nov 30, 2023
73a2eb8
Fixed bug in Flatten's axis (#18645)
Honry Dec 1, 2023
73d9b03
[js/webgpu] Add multidimensional(>4) uniform support (#18546)
axinging Dec 1, 2023
c7732a7
[WebNN EP] Fixed bug in op checking (#18638)
Honry Dec 1, 2023
9c9e6ad
Add SDXL Turbo to demo (#18627)
tianleiwu Dec 1, 2023
ccfea55
[QNN EP] Enable QNN HTP VTCM size setting (#18653)
HectorSVC Dec 1, 2023
182c525
Support MatMulBnb4 in PaddingElimination (#18646)
guyang3532 Dec 1, 2023
d698422
Update the template files to correct stage to fix the python cuda 12 …
jchen351 Dec 1, 2023
05a9c95
[DNNL] add Arm Compute Library (ACL) backend for dnnl execution provi…
snadampal Dec 1, 2023
fcea2cb
[Dort] Run type promotion pass to resolve dtype discrepancy (#18516)
BowenBao Dec 1, 2023
b22f49f
Fix unit tests failures in build with contrib ops disabled (#18659)
edgchen1 Dec 1, 2023
a353805
Fix Windows TVM CI workflow (#18667)
edgchen1 Dec 1, 2023
9c45fe4
Fix macos xcframework test stage codesign info (#18649)
YUNQIUGUO Dec 1, 2023
eaaf270
Remove EnvSetupScript parameter from win-ci.yml (#18662)
snnn Dec 1, 2023
92ee664
[js/webgpu] Fix shader errors in indicesGet/Set when rank > 4 (#18661)
qjia7 Dec 1, 2023
2f8b86b
Fix typo in the TensorShape (#17813)
Dec 2, 2023
81dde5c
Merge branch 'master' into openvino-ep-5.2-npu
saurabhkale17 Nov 10, 2023
5004f61
Merge branch 'openvino-ep-5.2-npu' of https://github.com/intel/onnxru…
sspintel Dec 3, 2023
d9b1406
Add support for LayerNormalization Op; NPU to go through ReadModel ->…
Dec 3, 2023
d1b9995
Fix an issue with provider options getting overwritten
sspintel Dec 4, 2023
814127b
Add device_precision access for UnsupportedOpModes
sspintel Dec 4, 2023
a5b2291
[Transformer Optimization]Return model directly for unknown model typ…
trajepl Dec 4, 2023
5353adc
[js/webgpu] Use the naive convTranspose when in/out channels are both…
qjia7 Dec 4, 2023
c02a386
[js/web/training] Implemented runEvalStep & runOptimizerStep (#18259)
carzh Dec 4, 2023
d514a96
Remove "Python Checks" pipeline status from readme as that pipeline n…
edgchen1 Dec 4, 2023
01b5c78
Add SD-Turbo and refine diffusion demo (#18694)
tianleiwu Dec 5, 2023
e066fca
[Quantization] Tensor quant overrides and QNN EP quantization configu…
adrianlizarraga Dec 5, 2023
2b3050b
Zhijxu/fix toposort (#18705)
zhijxu-MS Dec 5, 2023
c14fae9
add SAVE_TEST_GRAPH macro (#18696)
rui-ren Dec 5, 2023
10c5475
[JS/Web] Added CumSum operator to JSEP (#18637)
satyajandhyala Dec 5, 2023
f949e05
[js/webgpu] Support uniforms for pool (#18656)
axinging Dec 5, 2023
7081600
[JS/Web] AddedUniforms in GatherElements. (#18670)
satyajandhyala Dec 5, 2023
07aabcc
Set cuda device before create cuda stream for IOBinding case (#18583)
jslhcl Dec 5, 2023
9aa7284
fix lint error (#18708)
guschmue Dec 5, 2023
4bfa844
Skip module clone for preparing large model export (#18663)
pengwa Dec 5, 2023
c9e558c
Adding common python test requirements.txt (#18698)
jchen351 Dec 5, 2023
871c529
Mistral Optimization & Benchmarking Support (#18225)
petermcaughan Dec 5, 2023
c012e41
MoE with Expert Slicing (#18565)
wangyems Dec 6, 2023
559bd52
[QNN EP] Update QNN SDK to version 2.17.0 (#18684)
adrianlizarraga Dec 6, 2023
9768a72
[QNN EP] Fix a bug that can't create context binary if the model has …
HectorSVC Dec 6, 2023
c4b8120
Rename op elementwiseIf to where (#18657)
Honry Dec 6, 2023
7762f3f
[NNAPI EP] Add NNAPI Split (#18702)
YUNQIUGUO Dec 6, 2023
9479ba5
Build onnxruntime.dll as arm64x (#18633)
moyo1997 Dec 7, 2023
e603e78
Enforce If condition size == 1 (#18733)
yuslepukhin Dec 7, 2023
49470f0
Add benchmark script for control net (#18717)
tianleiwu Dec 7, 2023
3d8af6e
[WebNN EP] Skip split initializer (#18729)
Honry Dec 7, 2023
e469de6
Re-enable Sign op int64 test for QNN CPU test (#18734)
HectorSVC Dec 7, 2023
0625e94
Fix an issue that shared global_context across subgraphs
sspintel Dec 7, 2023
a045be3
use EO pool for windows web_cpu stage (#18737)
mszhanyi Dec 7, 2023
4abec97
[mlas] add loongarch lsx and lasx optimize code (#17937)
junchao-loongson Dec 7, 2023
efbef5f
[js/webgpu] allow to specify callback for profiling data (#18732)
fs-eire Dec 7, 2023
305db31
fix build aar error in Zip-Nuget-Java-Nodejs Packaging pipeline (#18745)
YUNQIUGUO Dec 7, 2023
bf33919
Update absl and gtest to fix an ARM64EC build error (#18735)
snnn Dec 7, 2023
7ed48a2
Objective-C API updates (#18738)
edgchen1 Dec 8, 2023
e8f33b5
[WebNN EP] Don't covert all inputs except the 0th input for Resize (#…
Honry Dec 8, 2023
e6740aa
Fix lintrunner issues
sspintel Dec 8, 2023
44b5843
Fix gemm_float8 build failure on CUDA 11.3-11.7 (#18760)
pengwa Dec 8, 2023
c7799d7
Build fixes for Windows ARM32 desktop build (#18752)
snnn Dec 8, 2023
2f93d97
Add cuda visible devices for Mistral benchmark (#18764)
ajindal1 Dec 9, 2023
d41dd77
Extend API page on the python documentation (#18762)
xadupre Dec 9, 2023
de32bae
[ROCm] Add GemmFloat8 (#18488)
cloudhan Dec 11, 2023
8d64122
Fix GQA shape inference (#18723)
PatriceVignola Dec 11, 2023
ee28be3
Merge pull request #332 from intel/sp/fix-global-context-bug
sspintel Dec 11, 2023
16df837
Update transformers package to fix the security issue (#18730)
askhade Dec 11, 2023
bfa5eb4
Adding a new pipeline for pubilshing cuda 12 nuget packages (#18713)
jchen351 Dec 11, 2023
ce1fed6
Adding a new pipeline for publishing to Python Cuda 12 packages. (#18…
jchen351 Dec 11, 2023
68c832d
Fix buffer overrun in 4b dequant cuda (#18780)
chenfucn Dec 11, 2023
ccf3b20
Allow layer-wise recompute (#18566)
pengwa Dec 12, 2023
eb03032
[js/web/training] lazyResetGrad implementation (#18711)
carzh Dec 12, 2023
a85ef65
Log out ORT session options (#16259)
ivberg Dec 12, 2023
b4be9e1
[js/webgpu] Fix shader compilation errors in cumsum (#18779)
qjia7 Dec 12, 2023
d673e39
[JS/WebGPU] Added uniforms to Tile and Where Ops (#18768)
satyajandhyala Dec 12, 2023
6530061
[PowerPC] Type casting the output operand of vec_xst. (#18057)
BODAPATIMAHESH Dec 12, 2023
81796a3
[QNN EP Quantization] Add fusion preprocessing to QNN quantization (#…
adrianlizarraga Dec 12, 2023
0ca8454
[JS/Web] Added uniforms to Reduce, Resize and Split Ops. (#18727)
satyajandhyala Dec 12, 2023
3940ef2
[ROCm] Refactor to hide ck layout (Row/Col) from ORT interface (#18777)
cloudhan Dec 13, 2023
dbe886a
Disable test_bert_result_with_layerwise_recompute (#18800)
pengwa Dec 13, 2023
1ad6eb1
Add DynamicQuantizeLinear as supported OP (#18798)
TedThemistokleous Dec 13, 2023
0a7c8ab
Enable OV CPU fallback for NPU compilation failures
preetha-intel Dec 4, 2023
b30e721
[js/webgpu] Provide a naive vectorized matmul algorithm (#18758)
qjia7 Dec 13, 2023
44054e7
Move NuGet nightly package publishing job to a separated pipeline (#1…
snnn Dec 13, 2023
17eaf9b
Fix a build warning in SparseTensor code for 32-bit build configs (#1…
snnn Dec 13, 2023
487abcd
Update gradient ops tests (#18783)
askhade Dec 13, 2023
f3fa045
Enable MacOS build in ORT Objc Pod (#18786)
YUNQIUGUO Dec 13, 2023
0723dcb
OpenVINO Execution Provider with 2023.2 support (#18596)
sspintel Dec 13, 2023
7047d13
Update windowsai-steps.yml: enable "/profile" linker flag (#18022)
snnn Dec 14, 2023
7dade5d
Readd basetargets in Microsoft.ML.OnnxRuntime.csproj (#18789)
mszhanyi Dec 14, 2023
9275dcd
Add NPU device in supported list of openvino devices
preetha-intel Dec 13, 2023
ba7bc96
Handle dynamic shapes fallback for NPU to OV CPU
preetha-intel Dec 11, 2023
d3e6168
Remove NPU operator from static mapping
preetha-intel Dec 14, 2023
95193cb
Set NDK version in Linux CPU Minimal Build E2E CI Pipeline (#18810)
snnn Dec 14, 2023
7386e21
Replace some ORT_ENFORCE with ORT_THROW_IF_ERROR (#18812)
snnn Dec 14, 2023
afe5cdc
[TensorRT EP] Switch to enqueueV3 with support DDS output (copy versi…
chilo-ms Dec 14, 2023
b129f42
Fix test model URL issue (#18823)
snnn Dec 14, 2023
1db1c75
[WebNN EP] WebNN only supports 4-D input and weight for Conv/ConvTran…
Honry Dec 14, 2023
6d5ee4d
[WebNN EP] Use explicit padding (#18688)
Dec 14, 2023
b42d4b8
[VitisAI] 1. api compatbile 2. dynamic load onnx (#18470)
BoarQing Dec 14, 2023
cbad4fe
Update absl and googletest (#18827)
snnn Dec 15, 2023
5eda79b
Improve perf for stage3 training (#18099)
pengwa Dec 15, 2023
fc9ecb5
Add Windows ARM build jobs to post merge pipeline (#18832)
snnn Dec 15, 2023
d795fc6
FIX: Our cmake script didn't check googletest's hash (#18826)
snnn Dec 15, 2023
d111eed
[WebNN EP] Change axis to axes for argMax/argMin (#18838)
peishenyan Dec 15, 2023
81ad1e6
[js/webgpu] Fix typo of outputShapes in profiling message (#18837)
Dec 15, 2023
89168b8
Fix CI error: The workflow is not valid. .github/workflows/rust-ci.y…
wirthual Dec 15, 2023
f52668c
Disable mlas unit test in ARM64EC build (#18747)
snnn Dec 15, 2023
4bbed4c
[js/webgpu] Fix f16 errors in unary (#18839)
qjia7 Dec 15, 2023
8f7b89b
[js/webgpu] Optimize NCHW layout for InstanceNormalization (#18123)
qjia7 Dec 15, 2023
2952cf8
Access map by iterator to silence sanity check. (#18835)
RandySheriffH Dec 15, 2023
50cbcf9
Build function bodies according to the imported global opset. (#18833)
yuslepukhin Dec 15, 2023
ad476d5
Change Nuget packaging pipeline's build TRT job to download CUDA SDK …
snnn Dec 16, 2023
9426bd5
[TensorRT EP] Update deprecated TRT api (#18834)
yf711 Dec 18, 2023
ea6186e
Update stale.yml to correct close-issue-message (#18849)
sophies927 Dec 18, 2023
3ff4a4c
Bump actions/stale from 8.0.0 to 9.0.0 (#18774)
dependabot[bot] Dec 18, 2023
63b47ce
[REACT NATIVE] Bugfix -> casing Podfile (#18861)
frankbaele Dec 19, 2023
6d7519e
Adding new pipeline for python cuda testing (#18718)
jchen351 Dec 19, 2023
f2a7ca0
Merge branch 'openvino-ep-5.2-npu' into ort_ovep_1.17_npu
sspintel Dec 19, 2023
0c7d93a
Remove static mapping of LayerNorm op for the NPU; Remove unused MLAS…
sspintel Dec 19, 2023
26bcf8d
Add support for UINT16 DTYPE in initializers, NPU, and CPU devices
sspintel Dec 18, 2023
d4224e1
Merge branch 'master' into sp/model_psa
Dec 18, 2023
4e5bcd3
Temporarily disable model domain check as it is yet to be supported b…
sspintel Dec 19, 2023
4dff154
Fix nightly pipeline failure (#18867)
askhade Dec 19, 2023
5f00bc9
Integrate high-performance x64 gemm library to MLAS (#17669)
luoyu-intel Dec 19, 2023
32fcf73
Implement dft(20) (#17821)
liqunfu Dec 19, 2023
98510fb
[JS/WebGPU] fix an error in Clip (#18799)
satyajandhyala Dec 19, 2023
ffa6602
[js/node] support manually dispose session (#18655)
fs-eire Dec 20, 2023
535a240
Update Nuget publishing jobs (#18851)
snnn Dec 20, 2023
666fcbd
Add LeakyRelu to list of NNAPI operators (#18880)
skottmckay Dec 20, 2023
07d4118
Merge branch 'master' into sp/model_psa
sspintel Dec 20, 2023
a601718
Allow overriding NPU compiler type through an environmental variable
sspintel Jan 11, 2024
38193f2
Fix a mistake in OpenVINO 2023.2 build flag
sspintel Jan 2, 2024
e94fd7b
Remove unused parameter op_map
sspintel Jan 11, 2024
122f26b
Merge branch 'sp/model_psa' into ort_ovep_1.17_npu
sspintel Jan 11, 2024
6319fe7
Add pow to no dimension supported list
saurabhkale17 Jan 12, 2024
9da42a1
Remove deprecated model domain check
sspintel Jan 11, 2024
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
  •  
  •  
  •  
2 changes: 1 addition & 1 deletion .github/workflows/rust-ci.yml
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,7 @@ jobs:
name: Download prebuilt ONNX Runtime archive from build.rs
runs-on: ubuntu-latest
env:
ORT_RUST_STRATEGY=download
ORT_RUST_STRATEGY: download
steps:
- uses: actions/checkout@v4
- uses: ./.github/actions/rust-toolchain-setup
Expand Down
4 changes: 2 additions & 2 deletions .github/workflows/stale.yml
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ jobs:
issues: write
pull-requests: write
steps:
- uses: actions/stale@v8.0.0
- uses: actions/stale@v9.0.0
with:
# Comma separated list of labels that can be assigned to issues to exclude them from being marked as stale
exempt-issue-labels: contributions welcome, feature request, regression
Expand All @@ -29,7 +29,7 @@ jobs:
# Label you want to apply to issues that have been inactive for the amount of time specified by days-before-issue-stale
stale-issue-label: "stale"
# Comment that you want to add to issues that are labeled by the actions/stale action
stale-issue-message: "This issue has been automatically marked as stale due to inactivity and will be closed in 7 days if no further activity occurs. If further support is needed, please provide an update and/or more details."
stale-issue-message: "This issue has been automatically marked as stale due to inactivity and will be closed in 30 days if no further activity occurs. If further support is needed, please provide an update and/or more details."
# Comment that you want to add to issues that are closed by the actions/stale action
close-issue-message: "This issue has been automatically closed due to inactivity. Please reactivate if further support is needed."
# If you never want this action to label PRs, set this value to -1
Expand Down
7 changes: 2 additions & 5 deletions .github/workflows/windows.yml
Original file line number Diff line number Diff line change
Expand Up @@ -49,13 +49,10 @@ jobs:
- uses: actions/checkout@v4
with:
submodules: true
- uses: actions/setup-python@v4
with:
python-version: '3.8.x'
architecture: 'x64'
- uses: conda-incubator/setup-miniconda@v2
with:
activate-environment: ""
activate-environment: "ort_build"
python-version: 3.8
- name: 'Install LLVM-Dev'
shell: pwsh
run: |
Expand Down
2 changes: 1 addition & 1 deletion .pipelines/windowsai-steps.yml
Original file line number Diff line number Diff line change
Expand Up @@ -84,7 +84,7 @@ jobs:
7z x cmake-3.26.3-windows-x86_64.zip
set PYTHONHOME=$(Build.BinariesDirectory)\${{ parameters.PythonPackageName }}.3.9.7\tools
set PYTHONPATH=$(Build.BinariesDirectory)\${{ parameters.PythonPackageName }}.3.9.7\tools
$(Build.BinariesDirectory)\${{ parameters.PythonPackageName }}.3.9.7\tools\python.exe "$(Build.SourcesDirectory)\tools\ci_build\build.py" --build_dir $(Build.BinariesDirectory) --build_shared_lib --enable_onnx_tests --ms_experimental --use_dml --use_winml --cmake_generator "Visual Studio 17 2022" --update --config RelWithDebInfo --enable_lto --use_telemetry --disable_rtti --enable_wcos $(BuildFlags) --cmake_extra_defines CMAKE_SYSTEM_VERSION=10.0.19041.0 --cmake_path $(Build.BinariesDirectory)\cmake-3.26.3-windows-x86_64\bin\cmake.exe --ctest_path $(Build.BinariesDirectory)\cmake-3.26.3-windows-x86_64\bin\ctest.exe
$(Build.BinariesDirectory)\${{ parameters.PythonPackageName }}.3.9.7\tools\python.exe "$(Build.SourcesDirectory)\tools\ci_build\build.py" --build_dir $(Build.BinariesDirectory) --build_shared_lib --enable_onnx_tests --ms_experimental --use_dml --use_winml --cmake_generator "Visual Studio 17 2022" --update --config RelWithDebInfo --enable_lto --use_telemetry --disable_rtti --enable_wcos $(BuildFlags) --cmake_extra_defines "CMAKE_EXE_LINKER_FLAGS_RELWITHDEBINFO=/PROFILE" "CMAKE_SHARED_LINKER_FLAGS_RELWITHDEBINFO=/PROFILE" CMAKE_SYSTEM_VERSION=10.0.19041.0 --cmake_path $(Build.BinariesDirectory)\cmake-3.26.3-windows-x86_64\bin\cmake.exe --ctest_path $(Build.BinariesDirectory)\cmake-3.26.3-windows-x86_64\bin\ctest.exe
workingDirectory: '$(Build.BinariesDirectory)'
displayName: 'Generate cmake config'

Expand Down
1 change: 1 addition & 0 deletions .vscode/settings.json
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@
"editor.codeActionsOnSave": {
"source.organizeImports": true
},
"editor.defaultFormatter": "ms-python.black-formatter"
},
// Enable Python linting and Pylance type checking
"python.analysis.typeCheckingMode": "basic",
Expand Down
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@
|Android|[![Build Status](https://dev.azure.com/onnxruntime/onnxruntime/_apis/build/status/Android%20CI%20Pipeline?label=Android)](https://dev.azure.com/onnxruntime/onnxruntime/_build/latest?definitionId=53)||
|iOS|[![Build Status](https://dev.azure.com/onnxruntime/onnxruntime/_apis/build/status/iOS%20CI%20Pipeline?label=iOS)](https://dev.azure.com/onnxruntime/onnxruntime/_build/latest?definitionId=134)||
|Web|[![Build Status](https://dev.azure.com/onnxruntime/onnxruntime/_apis/build/status/ONNX%20Runtime%20Web%20CI%20Pipeline?label=Web)](https://dev.azure.com/onnxruntime/onnxruntime/_build/latest?definitionId=161)||
|Other|[![Build Status](https://dev.azure.com/onnxruntime/onnxruntime/_apis/build/status/onnxruntime-binary-size-checks-ci-pipeline?repoName=microsoft%2Fonnxruntime&label=Binary+Size+Check)](https://dev.azure.com/onnxruntime/onnxruntime/_build/latest?definitionId=187&repoName=microsoft%2Fonnxruntime)<br>[![Build Status](https://dev.azure.com/onnxruntime/onnxruntime/_apis/build/status/onnxruntime-python-checks-ci-pipeline?label=Python+Checks)](https://dev.azure.com/onnxruntime/onnxruntime/_build/latest?definitionId=164)||
|Other|[![Build Status](https://dev.azure.com/onnxruntime/onnxruntime/_apis/build/status/onnxruntime-binary-size-checks-ci-pipeline?repoName=microsoft%2Fonnxruntime&label=Binary+Size+Check)](https://dev.azure.com/onnxruntime/onnxruntime/_build/latest?definitionId=187&repoName=microsoft%2Fonnxruntime)||

## Third-party Pipeline Status

Expand Down
12 changes: 12 additions & 0 deletions build_arm64x.bat
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
:: Copyright (c) Microsoft Corporation. All rights reserved.
:: Licensed under the MIT License.

@echo off

setlocal
set PATH=C:\Program Files\Git\usr\bin;%PATH%
set LINK_REPRO_NAME=/mylink.rsp

rem Requires a Python install to be available in your PATH
python "%~dp0\tools\ci_build\build.py" --arm64 --buildasx --build_dir "%~dp0\build\arm64-x" %*
python "%~dp0\tools\ci_build\build.py" --arm64ec --buildasx --build_dir "%~dp0\build\arm64ec-x" %*
6 changes: 3 additions & 3 deletions cgmanifests/generated/cgmanifest.json
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@
"component": {
"type": "git",
"git": {
"commitHash": "29bf8085f3bf17b84d30e34b3d7ff8248fda404e",
"commitHash": "dcd5bd5fd593e31465af3d9ef291d26c646b0a4f",
"repositoryUrl": "https://github.com/abseil/abseil-cpp.git"
},
"comments": "abseil_cpp"
Expand Down Expand Up @@ -126,7 +126,7 @@
"component": {
"type": "git",
"git": {
"commitHash": "f8d7d77c06936315286eb55f8de22cd23c188571",
"commitHash": "530d5c8c84abd2a46f38583ee817743c9b3a42b4",
"repositoryUrl": "https://github.com/google/googletest.git"
},
"comments": "googletest"
Expand Down Expand Up @@ -316,7 +316,7 @@
"component": {
"type": "git",
"git": {
"commitHash": "a4f72a314a85732ed67d5aa8d1088d207a7e0e61",
"commitHash": "5356c4a943a35e74d7cdc69486afcb8703b9a59a",
"repositoryUrl": "https://github.com/ROCmSoftwarePlatform/composable_kernel.git"
},
"comments": "composable_kernel"
Expand Down
44 changes: 43 additions & 1 deletion cmake/CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -87,6 +87,7 @@ option(onnxruntime_USE_QNN "Build with QNN support" OFF)
option(onnxruntime_USE_SNPE "Build with SNPE support" OFF)
option(onnxruntime_USE_RKNPU "Build with RKNPU support" OFF)
option(onnxruntime_USE_DNNL "Build with DNNL support" OFF)
option(onnxruntime_USE_JBLAS "Build MLAS with JBLAS support" ON)
option(onnxruntime_USE_JSEP "Build with JavaScript implemented kernels support" OFF)
option(onnxruntime_BUILD_UNIT_TESTS "Build ONNXRuntime unit tests" ON)
option(onnxruntime_BUILD_CSHARP "Build C# library" OFF)
Expand Down Expand Up @@ -1166,6 +1167,17 @@ if (onnxruntime_USE_DNNL)
add_compile_definitions(DNNL_OPENMP)
endif()

set(USE_JBLAS FALSE)
if (onnxruntime_USE_JBLAS AND NOT onnxruntime_MINIMAL_BUILD)
if ("${CMAKE_CXX_COMPILER_ID}" STREQUAL "GNU" AND onnxruntime_target_platform STREQUAL "x86_64")
add_compile_definitions(MLAS_JBLAS)
set(USE_JBLAS TRUE)
elseif ("${CMAKE_CXX_COMPILER_ID}" STREQUAL "MSVC" AND onnxruntime_target_platform STREQUAL "x64")
add_compile_definitions(MLAS_JBLAS)
set(USE_JBLAS TRUE)
endif()
endif()

# TVM EP
if (onnxruntime_USE_TVM)
if (NOT TARGET tvm)
Expand Down Expand Up @@ -1269,7 +1281,7 @@ if (onnxruntime_USE_OPENVINO)
add_definitions(-DOPENVINO_2023_1=1)
elseif ($ENV{INTEL_OPENVINO_DIR} MATCHES "2023.2")
set(OPENVINO_VERSION "2023.2")
add_definitions(-DOPENVINO_2023_1=1)
add_definitions(-DOPENVINO_2023_2=1)
elseif ($ENV{INTEL_OPENVINO_DIR} MATCHES "openvino")
set(OPENVINO_VERSION "2023.2")
add_definitions(-DOPENVINO_2023_2=1)
Expand All @@ -1293,6 +1305,14 @@ if (onnxruntime_USE_OPENVINO)
add_definitions(-DOPENVINO_CONFIG_CPU_FP16=1)
endif()

if (onnxruntime_USE_OPENVINO_NPU_FP16)
add_definitions(-DOPENVINO_CONFIG_NPU_FP16=1)
endif()

if (onnxruntime_USE_OPENVINO_NPU_U8)
add_definitions(-DOPENVINO_CONFIG_NPU_U8=1)
endif()

if (onnxruntime_USE_OPENVINO_GPU_FP32_NP)
add_definitions(-DOPENVINO_CONFIG_GPU_FP32=1)
add_definitions(-DOPENVINO_DISABLE_GRAPH_PARTITION=1)
Expand All @@ -1313,6 +1333,16 @@ if (onnxruntime_USE_OPENVINO)
add_definitions(-DOPENVINO_DISABLE_GRAPH_PARTITION=1)
endif()

if (onnxruntime_USE_OPENVINO_NPU_FP16_NP)
add_definitions(-DOPENVINO_CONFIG_NPU_FP16=1)
add_definitions(-DOPENVINO_DISABLE_GRAPH_PARTITION=1)
endif()

if (onnxruntime_USE_OPENVINO_NPU_U8_NP)
add_definitions(-DOPENVINO_CONFIG_NPU_U8=1)
add_definitions(-DOPENVINO_DISABLE_GRAPH_PARTITION=1)
endif()

if (onnxruntime_USE_OPENVINO_HETERO)
add_definitions(-DOPENVINO_CONFIG_HETERO=1)
add_definitions(-DDEVICE_NAME="${onnxruntime_USE_OPENVINO_DEVICE}")
Expand Down Expand Up @@ -1584,6 +1614,13 @@ set(VERSION_STRING "Internal Build" CACHE STRING "String representation of
if (WIN32)
list(APPEND onnxruntime_EXTERNAL_LIBRARIES ${SYS_PATH_LIB})
list(APPEND onnxruntime_EXTERNAL_LIBRARIES debug Dbghelp)
# In a onecore build the umbrella libs already contains references to the APIs in advapi32, so in onecore build we do not need to link to advapi32
# In a non-onecore build, usually we also do not need to link to advapi32 because VC++ by default should have provide everything we need, except when the build target is Windows ARM32.
# In the future we will add a build option to allow users disabling all API uses from advapi32 because some Windows environments do not have these APIs. For example, some Windows do not have
# Windows Registry so we cannot query Registry values.
if(onnxruntime_target_platform STREQUAL "ARM" AND CMAKE_CXX_STANDARD_LIBRARIES MATCHES kernel32.lib)
list(APPEND onnxruntime_EXTERNAL_LIBRARIES advapi32)
endif()
else()
list(APPEND onnxruntime_EXTERNAL_LIBRARIES nsync::nsync_cpp)
list(APPEND onnxruntime_EXTERNAL_LIBRARIES ${ICONV_LIB} ${CMAKE_DL_LIBS} Threads::Threads)
Expand Down Expand Up @@ -1773,3 +1810,8 @@ if(TARGET onnxruntime)
"${PROJECT_BINARY_DIR}/${PROJECT_NAME}ConfigVersion.cmake"
DESTINATION "${CMAKE_INSTALL_LIBDIR}/cmake/${PROJECT_NAME}")
endif()

if(DEFINED BUILD_AS_ARM64X)
set(ARM64X_TARGETS onnxruntime)
include("${CMAKE_SOURCE_DIR}/arm64x.cmake")
endif()
33 changes: 33 additions & 0 deletions cmake/arm64x.cmake
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
set(arm64ReproDir "${CMAKE_SOURCE_DIR}/repros")

if("${BUILD_AS_ARM64X}" STREQUAL "ARM64")
foreach (n ${ARM64X_TARGETS})
add_custom_target(mkdirs_${n} ALL COMMAND cmd /c (if exist \"${arm64ReproDir}/${n}_temp/\" rmdir /s /q \"${arm64ReproDir}/${n}_temp\") && mkdir \"${arm64ReproDir}/${n}_temp\" )
add_dependencies(${n} mkdirs_${n})
target_link_options(${n} PRIVATE "/LINKREPRO:${arm64ReproDir}/${n}_temp")
add_custom_target(${n}_checkRepro ALL COMMAND cmd /c if exist \"${n}_temp/*.obj\" if exist \"${n}\" rmdir /s /q \"${n}\" 2>nul && if not exist \"${n}\" ren \"${n}_temp\" \"${n}\" DEPENDS ${n}
WORKING_DIRECTORY ${arm64ReproDir})
endforeach()


elseif("${BUILD_AS_ARM64X}" STREQUAL "ARM64EC")
foreach (n ${ARM64X_TARGETS})
set(ARM64_LIBS)
set(ARM64_OBJS)
set(ARM64_DEF)

file(GLOB ARM64_OBJS "${arm64ReproDir}/${n}/*.obj")
file(GLOB ARM64_DEF "${arm64ReproDir}/${n}/*.def")
file(GLOB ARM64_LIBS "${arm64ReproDir}/${n}/*.LIB")

if(NOT "${ARM64_DEF}" STREQUAL "")
set(ARM64_DEF "/defArm64Native:${ARM64_DEF}")
endif()
target_sources(${n} PRIVATE ${ARM64_OBJS})
target_link_options(${n} PRIVATE /machine:arm64x "${ARM64_DEF}")

if(NOT "${ARM64_LIBS}" STREQUAL "")
target_link_libraries(${n} PUBLIC ${ARM64_LIBS})
endif()
endforeach()
endif()
6 changes: 3 additions & 3 deletions cmake/deps.txt
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
# NOTE: You must run deps_update_and_upload.py and generate_cgmanifest.py when ready to test your changes in a CI.
# See https://microsoft.sharepoint.com/teams/ONNX2/_layouts/OneNote.aspx?id=%2Fteams%2FONNX2%2FShared%20Documents%2FNotebooks%2FONNX%20Ecosystem%20Team%20Notebook&wd=target%28Development.one%7C63D3AB47-51D1-4A62-9965-66882234BD44%2FAdd%20or%20update%20a%20dependency%20in%20deps.txt%7C0E9ED71D-89D5-40FA-B05F-C0123289C591%2F%29
#
abseil_cpp;https://github.com/abseil/abseil-cpp/archive/refs/tags/20230802.0.zip;04271dfbfac59269b6939e1e9d5faf0d18a7ba91
abseil_cpp;https://github.com/abseil/abseil-cpp/archive/dcd5bd5fd593e31465af3d9ef291d26c646b0a4f.zip;6cc204586014e189f5c0fe3274f83162fa7c700c
cxxopts;https://github.com/jarro2783/cxxopts/archive/3c73d91c0b04e2b59462f0a741be8c07024c1bc0.zip;6c6ca7f8480b26c8d00476e0e24b7184717fe4f0
date;https://github.com/HowardHinnant/date/archive/refs/tags/v3.0.1.zip;2dac0c81dc54ebdd8f8d073a75c053b04b56e159
dlpack;https://github.com/dmlc/dlpack/archive/refs/tags/v0.6.zip;4d565dd2e5b31321e5549591d78aa7f377173445
Expand All @@ -27,7 +27,7 @@ fp16;https://github.com/Maratyszcza/FP16/archive/0a92994d729ff76a58f692d3028ca1b
fxdiv;https://github.com/Maratyszcza/FXdiv/archive/63058eff77e11aa15bf531df5dd34395ec3017c8.zip;a5658f4036402dbca7cebee32be57fb8149811e1
google_benchmark;https://github.com/google/benchmark/archive/refs/tags/v1.7.0.zip;e97c368b176e8614e3f1bf13dd9abcf6a7ad9908
google_nsync;https://github.com/google/nsync/archive/refs/tags/1.26.0.zip;5e7c00ef6bf5b787386fc040067903ec774e2752
googletest;https://github.com/google/googletest/archive/refs/tags/v1.14.0.zip;0ac421f2ec11af38b0fff0f1992184032731a8bc
googletest;https://github.com/google/googletest/archive/530d5c8c84abd2a46f38583ee817743c9b3a42b4.zip;5e3a61db2aa975cfd0f97ba92c818744e7fa7034
googlexnnpack;https://github.com/google/XNNPACK/archive/0da379fc4808f9601faef392352018c741c0f297.zip;663883491e380b628e0a5b162b5f2658032fae73
json;https://github.com/nlohmann/json/archive/refs/tags/v3.10.5.zip;f257f8dc27c5b8c085dc887b40cddd18ae1f725c
microsoft_gsl;https://github.com/microsoft/GSL/archive/refs/tags/v4.0.0.zip;cf368104cd22a87b4dd0c80228919bb2df3e2a14
Expand All @@ -54,4 +54,4 @@ tensorboard;https://github.com/tensorflow/tensorboard/archive/373eb09e4c5d2b3cc2
cutlass;https://github.com/NVIDIA/cutlass/archive/refs/tags/v3.1.0.zip;757f90a795034a89d4f48a79d1f009f7a04c8dee
utf8_range;https://github.com/protocolbuffers/utf8_range/archive/72c943dea2b9240cd09efde15191e144bc7c7d38.zip;9925739c9debc0efa2adcb194d371a35b6a03156
extensions;https://github.com/microsoft/onnxruntime-extensions/archive/94142d8391c9791ec71c38336436319a2d4ac7a0.zip;4365ac5140338b4cb75a39944a4be276e3829b3c
composable_kernel;https://github.com/ROCmSoftwarePlatform/composable_kernel/archive/a4f72a314a85732ed67d5aa8d1088d207a7e0e61.zip;f57357ab6d300e207a632d034ebc8aa036a090d9
composable_kernel;https://github.com/ROCmSoftwarePlatform/composable_kernel/archive/5356c4a943a35e74d7cdc69486afcb8703b9a59a.zip;522382c2af437e09124287e5879ab64af5b2e299
1 change: 0 additions & 1 deletion cmake/external/abseil-cpp.natvis
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,6 @@
<Intrinsic Name="_capacity" Expression="_commonfields().capacity_"/>
<Intrinsic Name="_control" Expression="_commonfields().control_"/>
<Intrinsic Name="_slots" Expression="(slot_type*)(_commonfields().slots_)"/>
<DisplayString Condition="_size() == 0">empty</DisplayString>
<DisplayString IncludeView="noparens">size={ _size() }</DisplayString>
<DisplayString ExcludeView="noparens">size=({_size()})</DisplayString>
<Expand>
Expand Down
12 changes: 11 additions & 1 deletion cmake/external/dnnl.cmake
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,16 @@ elseif(onnxruntime_USE_DNNL AND onnxruntime_DNNL_GPU_RUNTIME STREQUAL "ocl" AND
set(DNNL_GPU_CMAKE_ARGS "-DDNNL_GPU_RUNTIME=OCL " "-DOPENCLROOT=${onnxruntime_DNNL_OPENCL_ROOT}")
endif()

if(onnxruntime_USE_DNNL AND onnxruntime_DNNL_AARCH64_RUNTIME STREQUAL "acl" AND onnxruntime_DNNL_ACL_ROOT STREQUAL "")
message(FATAL_ERROR "--dnnl_acl_root required")
elseif(onnxruntime_USE_DNNL AND onnxruntime_DNNL_AARCH64_RUNTIME STREQUAL "" AND NOT (onnxruntime_DNNL_ACL_ROOT STREQUAL ""))
message(FATAL_ERROR "--dnnl_aarch64_runtime required")
elseif(onnxruntime_USE_DNNL AND onnxruntime_DNNL_AARCH64_RUNTIME STREQUAL "acl" AND NOT (onnxruntime_DNNL_ACL_ROOT STREQUAL ""))
file(TO_CMAKE_PATH ${onnxruntime_DNNL_ACL_ROOT} onnxruntime_DNNL_ACL_ROOT)
set(ACL_INCLUDE_DIR ${onnxruntime_DNNL_ACL_ROOT}/arm_compute)
set(DNNL_AARCH64_CMAKE_ARGS "-DDNNL_AARCH64_USE_ACL=ON")
endif()

if (onnxruntime_USE_DNNL)
set(DNNL_SOURCE ${CMAKE_CURRENT_BINARY_DIR}/dnnl/src/dnnl/src)
set(DNNL_INSTALL ${CMAKE_CURRENT_BINARY_DIR}/dnnl/install)
Expand All @@ -51,7 +61,7 @@ if (onnxruntime_USE_DNNL)
GIT_TAG ${DNNL_TAG}
# PATCH_COMMAND ${MKLDNN_PATCH_DISCARD_COMMAND} COMMAND ${DNNL_PATCH_COMMAND}
SOURCE_DIR ${DNNL_SOURCE}
CMAKE_ARGS -DDNNL_BUILD_TESTS=OFF -DDNNL_ENABLE_CONCURRENT_EXEC=ON -DDNNL_BUILD_EXAMPLES=OFF -DCMAKE_BUILD_TYPE=${CMAKE_BUILD_TYPE} -DCMAKE_INSTALL_PREFIX=${DNNL_INSTALL} ${DNNL_GPU_CMAKE_ARGS}
CMAKE_ARGS -DDNNL_BUILD_TESTS=OFF -DDNNL_ENABLE_CONCURRENT_EXEC=ON -DDNNL_BUILD_EXAMPLES=OFF -DCMAKE_BUILD_TYPE=${CMAKE_BUILD_TYPE} -DCMAKE_INSTALL_PREFIX=${DNNL_INSTALL} ${DNNL_GPU_CMAKE_ARGS} ${DNNL_AARCH64_CMAKE_ARGS}
)
link_directories(${DNNL_LIB_DIR})
endif()
8 changes: 4 additions & 4 deletions cmake/external/onnxruntime_external_deps.cmake
Original file line number Diff line number Diff line change
Expand Up @@ -47,8 +47,8 @@ if (onnxruntime_BUILD_UNIT_TESTS)
FetchContent_Declare(
googletest
URL ${DEP_URL_googletest}
FIND_PACKAGE_ARGS 1.14.0...<2.0.0 NAMES GTest
URL_HASH SHA1=${DEP_SHA1_googletest}
FIND_PACKAGE_ARGS 1.14.0...<2.0.0 NAMES GTest
)
endif()

Expand Down Expand Up @@ -124,7 +124,7 @@ if(CMAKE_CROSSCOMPILING AND NOT ONNX_CUSTOM_PROTOC_EXECUTABLE)
if(protoc_binary_SOURCE_DIR)
message("Use prebuilt protoc")
set(ONNX_CUSTOM_PROTOC_EXECUTABLE ${protoc_binary_SOURCE_DIR}/bin/protoc.exe)
set(PROTOC_EXECUTABLE ${ONNX_CUSTOM_PROTOC_EXECUTABLE})
set(PROTOC_EXECUTABLE ${ONNX_CUSTOM_PROTOC_EXECUTABLE})
endif()
elseif(CMAKE_HOST_SYSTEM_NAME STREQUAL "Linux")
if(CMAKE_HOST_SYSTEM_PROCESSOR MATCHES "^(x86_64|amd64)$")
Expand All @@ -140,7 +140,7 @@ if(CMAKE_CROSSCOMPILING AND NOT ONNX_CUSTOM_PROTOC_EXECUTABLE)
if(protoc_binary_SOURCE_DIR)
message("Use prebuilt protoc")
set(ONNX_CUSTOM_PROTOC_EXECUTABLE ${protoc_binary_SOURCE_DIR}/bin/protoc)
set(PROTOC_EXECUTABLE ${ONNX_CUSTOM_PROTOC_EXECUTABLE})
set(PROTOC_EXECUTABLE ${ONNX_CUSTOM_PROTOC_EXECUTABLE})
endif()
elseif ((CMAKE_SYSTEM_NAME STREQUAL "Emscripten" OR CMAKE_SYSTEM_NAME STREQUAL "Android" OR CMAKE_SYSTEM_NAME STREQUAL "iOS") AND CMAKE_HOST_SYSTEM_NAME STREQUAL "Darwin")
FetchContent_Declare(protoc_binary URL ${DEP_URL_protoc_mac_universal} URL_HASH SHA1=${DEP_SHA1_protoc_mac_universal})
Expand Down Expand Up @@ -281,7 +281,7 @@ if ((CPUINFO_SUPPORTED OR onnxruntime_USE_XNNPACK) AND NOT ANDROID)
pytorch_clog
URL ${DEP_URL_pytorch_cpuinfo}
URL_HASH SHA1=${DEP_SHA1_pytorch_cpuinfo}
SOURCE_SUBDIR deps/clog
SOURCE_SUBDIR deps/clog
)
set(ONNXRUNTIME_CLOG_PROJ pytorch_clog)
set(ONNXRUNTIME_CLOG_TARGET_NAME clog)
Expand Down
6 changes: 1 addition & 5 deletions cmake/onnxruntime.cmake
Original file line number Diff line number Diff line change
Expand Up @@ -282,11 +282,7 @@ endif()

# Assemble the Apple static framework (iOS and macOS)
if(onnxruntime_BUILD_APPLE_FRAMEWORK)
if(${CMAKE_SYSTEM_NAME} STREQUAL "iOS")
set(STATIC_FRAMEWORK_OUTPUT_DIR ${CMAKE_CURRENT_BINARY_DIR}/${CMAKE_BUILD_TYPE}-${CMAKE_OSX_SYSROOT})
else() # macOS
set(STATIC_FRAMEWORK_OUTPUT_DIR ${CMAKE_CURRENT_BINARY_DIR})
endif()
set(STATIC_FRAMEWORK_OUTPUT_DIR ${CMAKE_CURRENT_BINARY_DIR}/${CMAKE_BUILD_TYPE}-${CMAKE_OSX_SYSROOT})

# Setup the various directories required. Remove any existing ones so we start with a clean directory.
set(STATIC_LIB_DIR ${CMAKE_CURRENT_BINARY_DIR}/static_libraries)
Expand Down
Loading