Re: [yocto] do_package error after building meta-tensorflow and libcamera-apps does not detect tensorflow-lite
Gyorgy Sarvari <[email protected]>
| Newsgroups | org.yoctoproject.lists.yocto |
|---|---|
| Message-ID | <[email protected]> |
Have you applied all the mentioned changes? The first 2 should be fixed by the tflite patch[1] (which packages these missing headers). For the 3rd point there is an extra patch in the github PR[2], which addresses it. I haven't seen the last error, but the first three at least should be gone... [1]: https://patchwork.yoctoproject.org/project/yocto/patch/[email protected]/ [2]: https://github.com/agherzan/meta-raspberrypi/pull/1522/files#diff-c47f7714624569c0978292d85655edf001a084d2c2b830147b52647eccba602b On 9/15/25 18:38, ashish shrikhande6 via Lists.Yoctoproject.Org wrote: > After Integrating tflite with libcamera I am getting some compilation > errors: > 1. > | In file included from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tf_stage.cpp:7: > | > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tf_stage.hpp:16:10: > fatal error: tensorflow/lite/builtin_op_data.h: No such file or directory > | 16 | #include "tensorflow/lite/builtin_op_data.h" > | | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > | compilation terminated. > > 2. > > | In file included from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/schema/schema_generated.h:18, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/api/op_resolver.h:29, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/subgraph.h:36, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/async/async_subgraph.h:27, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/async/async_signature_runner.h:24, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/interpreter.h:48, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/interpreter.h:21, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tf_stage.hpp:18, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/object_classify_tf_stage.cpp:8: > | > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/compiler/mlir/lite/schema/schema_generated.h:21:10: > fatal error: flatbuffers/flatbuffers.h: No such file or directory > | 21 | #include "flatbuffers/flatbuffers.h" > | | ^~~~~~~~~~~~~~~~~~~~~~~~~~~ > | compilation terminated. > > 3. > > | In file included from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/segmentation_tf_stage.cpp:8: > | > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/segmentation.hpp:15:82: > error: 'uint8_t' was not declared in this scope > | 15 | Segmentation(int w, int h, std::vector<std::string> > l, const std::vector<uint8_t> &s) > | | > ^~~~~~~ > | > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/segmentation.hpp:12:1: > note: 'uint8_t' is defined in header '<cstdint>'; this is probably > fixable by adding '#include <cstdint>' > > 4. > > | In file included from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/c/builtin_op_data.h:24, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/builtin_op_data.h:20, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tf_stage.hpp:16, > | from > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/object_detect_tf_stage.cpp:9: > | > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/c/common.h:125:7: > error: ISO C++ forbids flexible array member 'data' [-Werror=pedantic] > | 125 | int data[]; > | | ^~~~ > | > ../sources/libcamera-apps-1.9.0+git/post_processing_stages/tensorflow/lite/core/c/common.h:169:9: > error: ISO C++ forbids flexible array member 'data' [-Werror=pedantic] > | 169 | float data[]; > | | ^~~~ > | cc1plus: all warnings being treated as errors >