-
-
Notifications
You must be signed in to change notification settings - Fork 56.4k
Tengine lite update #18323
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. Weβll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Tengine lite update #18323
Conversation
3rdparty/libtengine/tengine.cmake
Outdated
| SET (TENGINE_TOOLCHIN_FLAG "-march=armv7-a") | ||
| elseif(CMAKE_SYSTEM_PROCESSOR STREQUAL aarch64) ## AARCH64 | ||
| SET(CONFIG_ARCH_ARM64 ON) | ||
| SET (TENGINE_TOOLCHIN_FLAG "-march=armv8-a") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
TOOLCHIN
typo
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done.
modules/dnn/src/dnn.cpp
Outdated
| LayerData &ld = it->second; | ||
| Ptr<Layer> layerCur = ld.layerInstance; | ||
|
|
||
| if (!strcmp(ld.type.c_str(),"Convolution")) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why Layer::finalize() doesn't work here?
(or just the first run of Layer::forward(), check integration of convolutionOp from OpenCL backend)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Move to Layer::forward . please checkout again
3rdparty/libtengine/tengine.cmake
Outdated
| SET (TENGINE_TOOLCHAIN_FLAG "-march=armv7-a") | ||
| elseif(CMAKE_SYSTEM_PROCESSOR STREQUAL aarch64) ## AARCH64 | ||
| SET(CONFIG_ARCH_ARM64 ON) | ||
| SET (TENGINE_TOOLCHAIN_FLAG "-march=armv8-a") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
remove space before bracket.
modules/dnn/CMakeLists.txt
Outdated
| if(EXISTS ${TENGINE_LIBRARIES}) | ||
| list(APPEND libs ${TENGINE_LIBRARIES}) | ||
| if(EXISTS ${TENGINE_LIBRARIES}) | ||
| list(APPEND libs -Wl,--whole-archive ${TENGINE_LIBRARIES} -Wl,--no-whole-archive) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
if(EXISTS ${TENGINE_LIBRARIES})
Don't work with list as variable value.
Broken indentation.
No difference between both code paths. ocv_add_dependencies() is misused (required opencv modules) and does nothing.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done.
3rdparty/libtengine/tengine.cmake
Outdated
| SET(OCV_TENGINE_URL "https://github.com/OAID/Tengine/archive/") #url2 | ||
| SET(tengine_md5sum f51ca8f3963faeeff3f019a6f6edc206) #md5sum2 | ||
| #SET(OCV_TENGINE_URL "https://github.com/OAID/Tengine/archive/") #url2 | ||
| SET(OCV_TENGINE_URL "https://github.com/liqi-c/Tengine/archive/") #url2. Tmp url |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You don't need that for testing. Upstream URL should work fine with commits from fork.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done.
Tengine lite update * update tengine * Modify for arm32 build. * format optimization * add teng_ befor some tengine api * update graph_t to teng_graph_t * update graph_t to teng_graph_t * Code structure optimization * optimization * optimization * remove space * update tengine url Co-authored-by: liqi <[email protected]>
dnn: cleanup of tengine backend #24122 π Cleanup for OpenCV 5.0. Tengine backend is added for convolution layer speedup on ARM CPUs, but it is not maintained and the convolution layer on our default backend has reached similar performance to that of Tengine. Tengine backend related PRs: - #16724 - #18323 ### Pull Request Readiness Checklist See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request - [x] I agree to contribute to the project under Apache 2 License. - [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV - [x] The PR is proposed to the proper branch - [x] There is a reference to the original bug report and related work - [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable Patch to opencv_extra has the same branch name. - [x] The feature is well documented and sample code can be built with the project CMake
dnn: cleanup of tengine backend opencv#24122 π Cleanup for OpenCV 5.0. Tengine backend is added for convolution layer speedup on ARM CPUs, but it is not maintained and the convolution layer on our default backend has reached similar performance to that of Tengine. Tengine backend related PRs: - opencv#16724 - opencv#18323 ### Pull Request Readiness Checklist See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request - [x] I agree to contribute to the project under Apache 2 License. - [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV - [x] The PR is proposed to the proper branch - [x] There is a reference to the original bug report and related work - [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable Patch to opencv_extra has the same branch name. - [x] The feature is well documented and sample code can be built with the project CMake
dnn: cleanup of tengine backend opencv#24122 π Cleanup for OpenCV 5.0. Tengine backend is added for convolution layer speedup on ARM CPUs, but it is not maintained and the convolution layer on our default backend has reached similar performance to that of Tengine. Tengine backend related PRs: - opencv#16724 - opencv#18323 ### Pull Request Readiness Checklist See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request - [x] I agree to contribute to the project under Apache 2 License. - [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV - [x] The PR is proposed to the proper branch - [x] There is a reference to the original bug report and related work - [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable Patch to opencv_extra has the same branch name. - [x] The feature is well documented and sample code can be built with the project CMake
Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
Patch to opencv_extra has the same branch name.