Refactor Memory Allocator from Arm Executor Runner

[BujSet]

Summary

Some applications desire using more minimal memory allocators when leveraging the existing the exist arm_executor_runner.cpp, e.g. here. This change decouples a custom memory allocator wrapper from the executor runner so that apps can choose different implementations as needed.

Test plan

Ran the steps of the Arm EthosU tutorial to ensure that existing flow still works. I.e. running:

./examples/arm/setup.sh --i-agree-to-the-contained-eula
source ...
./examples/arm/run.sh --model_name=add --no_quantize --target=ethos-u55-128

still produces the following output:

I [executorch:arm_executor_runner.cpp:762 main()] Model executed successfully.
I [executorch:arm_executor_runner.cpp:768 main()] model_pte_program_size:     2064 bytes.
I [executorch:arm_executor_runner.cpp:769 main()] model_pte_loaded_size:      2064 bytes.
I [executorch:arm_executor_runner.cpp:783 main()] method_allocator_used:     320 / 62914560  free: 62914240 ( used: 0 % )
I [executorch:arm_executor_runner.cpp:790 main()] method_allocator_planned:  64 bytes
I [executorch:arm_executor_runner.cpp:794 main()] method_allocator_loaded:   232 bytes
I [executorch:arm_executor_runner.cpp:798 main()] method_allocator_input:    24 bytes
I [executorch:arm_executor_runner.cpp:799 main()] method_allocator_executor: 0 bytes
I [executorch:arm_executor_runner.cpp:802 main()] peak_temp_allocator:       64 / 2097152 free: 2097152 ( used: 0 % ) 
I [executorch:arm_executor_runner.cpp:812 main()] 1 outputs:
Output[0][0]: (int) 2
Output[0][1]: (int) 2
Output[0][2]: (int) 2
Output[0][3]: (int) 2
Output[0][4]: (int) 2
I [executorch:arm_executor_runner.cpp:952 main()] Program complete, exiting.
I [executorch:arm_executor_runner.cpp:956 main()] ♦
Info: /OSCI/SystemC: Simulation stopped by user.
[warning ][main@0][3440 ns] Simulation stopped by user
[backends/arm/scripts/run_fvp.sh] Simulation complete, 0
Checking for problems in log:
No problems found!
+ set +x

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/12353

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 4 Unrelated Failures

As of commit 579f693 with merge base f909aaa ():

NEW FAILURE - The following job has failed:

• pull / test-llava-runner-linux / linux-job (gh)
  RuntimeError: Command docker exec -t 69beadf4cbca18ff4870fe7a75ff83a79aa684a6281bba9cb126a9c6cf08def6 /exec failed with exit code 139

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• pull / unittest / linux / linux-job (gh) (trunk failure)
  devtools/inspector/tests/inspector_utils_test.py::TestInspectorUtils::test_equip_debug_handle_to_export_program_success
• pull / unittest / macos / macos-job (gh) (trunk failure)
  devtools/inspector/tests/inspector_utils_test.py::TestInspectorUtils::test_equip_debug_handle_to_export_program_success
• pull / unittest-editable / linux / linux-job (gh) (trunk failure)
  devtools/inspector/tests/inspector_utils_test.py::TestInspectorUtils::test_equip_debug_handle_to_export_program_success
• pull / unittest-editable / macos / macos-job (gh) (trunk failure)
  devtools/inspector/tests/inspector_utils_test.py::TestInspectorUtils::test_equip_debug_handle_to_export_program_success

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[BujSet]

@pytorchbot label "release notes: none"

[zingo]

Nice work, good idea to split it out so it can be usefull for more runners. We will probably have a few runner examples going forward. For examples for different OS/targets and I think we could do a better splitup between testing / semihosting and a "minimal" good example, as it's now much of the #ifdef stuff gets in the way if a good readable example.