Error in the llm executorch profiling example

[sheetalarkadam]

📚 The doc issue

I am following the instructions here
I am getting an error in etdump generation

I have made the following changes as mentioned in the blog

//main.cpp
#include <executorch/devtools/etdump/etdump_flatcc.h>
ETDumpGen* etdump_gen = static_cast<ETDumpGen*>(model.event_tracer());

ET_LOG(Info, "ETDump size: %zu blocks", etdump_gen->get_num_blocks());
etdump_result result = etdump_gen->get_etdump_data();
if (result.buf != nullptr && result.size > 0) {
    // On a device with a file system, users can just write it to a file.
    FILE* f = fopen("etdump.etdp", "w+");
    fwrite((uint8_t*)result.buf, 1, result.size, f);
    fclose(f);
    free(result.buf);
}

#CMakeLists.txt

option(EXECUTORCH_ENABLE_EVENT_TRACER "" ON)
option(EXECUTORCH_BUILD_DEVTOOLS "" ON)

target_link_libraries(
  nanogpt_runner
  PRIVATE executorch
          etdump # Provides the EventTracer class
)
target_compile_options(executorch PUBLIC -DET_EVENT_TRACER_ENABLED)
target_compile_options(portable_ops_lib PUBLIC -DET_EVENT_TRACER_ENABLED)

cmake --build cmake-out -j10 errors out with
[100%] Built target portable_kernels gmake[1]: Leaving directory 'et-nanogpt/cmake-out' gmake: *** [Makefile:136: all] Error 2
Line 136 in Makefile is $(MAKE) $(MAKESILENT) -f CMakeFiles/Makefile2 all. Doesn't say anything else.

Also, can somebody also explain the line "(Note that this time we build in release mode to get around a flatccrt build limitation.)".

Suggest a potential alternative/fix

No response

cc @mergennachin @byjlw @Olivia-liu @Jack-Khuu @cccclai @helunwencser @dvorjackz

[mcr229]

@tarun292 @Olivia-liu would either of you be able to take a look at this?

[digantdesai]

@Olivia-liu any update?

[Gasoonjia]

hi i'm taking over devtool stuffs. Is the bug solved? @digantdesai @mcr229 @sheetalarkadam