Qualcomm AI Engine Direct - Optimize memory usage at runtime

[shewu-quic]

Qnn backend doesn't need processed data after qnn_context_create_from_binary.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7003

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• [DomainsOnly] Jobs fail with GLIBC version not found

✅ No Failures

As of commit f1fe34b with merge base a39ea29 ():
💚 Looks good so far! There are no failures yet. 💚

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

[facebook-github-bot]

@cccclai has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[shewu-quic]

Hi @cccclai,

The PR is to fix the issue of memory usage.
Could you please help take a look?

Thanks,
Hutton

[cccclai]

Thanks! Maybe it also helps to resolve the RAM issue we ran into before for the 8b model on 16GB? I understand you already fixed it in an alternative way.

[shewu-quic]

I think spill fill buffer is still necessary for 8b due to the memory usage in HTP.
This change only reduced PSS not dma buffer.