Skip to content

Qualcomm AI Engine Direct -- update documents #4590

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.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
facebook-github-bot merged 6 commits into from
Aug 9, 2024
Merged

Qualcomm AI Engine Direct -- update documents #4590

facebook-github-bot merged 6 commits into from
Aug 9, 2024

Conversation

@chiwwang
Copy link
Contributor

@chiwwang chiwwang commented Aug 8, 2024

  1. Updated QNN download links.
  2. Unified setup.md and build-run-qualcomm-ai-engine-direct-backend.md
  3. Fixed build commands in build-run-qualcomm-ai-engine-direct-backend.md

chiwwang added 2 commits August 8, 2024 11:40
Qualcomm AI Engine Direct -- update documents
caa484f 
1. Updated QNN download links.
2. Unified setup.md and build-run-qualcomm-ai-engine-direct-backend.md
3. Fixed build commands in
   build-run-qualcomm-ai-engine-direct-backend.md
Merge remote-tracking branch 'fork/main' into update_qc_doc
46fa7ed
@pytorch-bot
Copy link

pytorch-bot bot commented Aug 8, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/4590

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

✅ No Failures

As of commit dbb452f with merge base 2405f74 (image):
💚 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 facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Aug 8, 2024
@kirklandsign kirklandsign requested a review from cccclai August 8, 2024 04:01
cccclai
cccclai approved these changes Aug 8, 2024
View reviewed changes
Copy link
Contributor

@cccclai cccclai left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for updating the documents! Just some minor comments. Should we add a session for x86 simulator? I find it super helpful. And possibly a model supported list.

docs/source/build-run-qualcomm-ai-engine-direct-backend.md
Comment on lines -119 to -120
cp schema/program.fbs exir/_serialize/program.fbs
cp schema/scalar_type.fbs exir/_serialize/scalar_type.fbs
Copy link
Contributor

@cccclai cccclai Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are we good with this now?

Copy link
Contributor Author

@chiwwang chiwwang Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No. But this action is more related to AOT compilation, so I moved there:
https://github.com/pytorch/executorch/pull/4590/files#diff-8bac58bd3ddef654a20549948dfea5d27f76a8ee29544328f65819cc0ccd7bcfR205

docs/source/build-run-qualcomm-ai-engine-direct-backend.md Outdated
Comment on lines 129 to 130
mkdir build_x86_64
cd build_x86_64
Copy link
Contributor

@cccclai cccclai Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

let's add them to git ignore?

Copy link
Contributor Author

@chiwwang chiwwang Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm fine with that... but how about using the name already in .gitignore?

cmake-out/
cmake-android-out/
cmake-out-android/

I can change build_x86_64 to cmake-out and build_android to cmake-android-out. It seems more consistent with other backends.

Copy link
Contributor

@cccclai cccclai Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That sounds good

docs/source/build-run-qualcomm-ai-engine-direct-backend.md Outdated
-DPYTHON_EXECUTABLE=python3 \
-DEXECUTORCH_SEPARATE_FLATCC_HOST_PROJECT=OFF

cmake --build $PWD -t "PyQnnManagerAdaptor" "PyQnnWrapperAdaptor" -j8
Copy link
Contributor

@cccclai cccclai Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Any specific reason we use -j8 here and -j16 in the following cmake command?

Copy link
Contributor Author

@chiwwang chiwwang Aug 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No. I will fix this to avoid confusion.

@chiwwang
Copy link
Contributor Author

chiwwang commented Aug 8, 2024

Thanks for updating the documents! Just some minor comments. Should we add a session for x86 simulator? I find it super helpful. And possibly a model supported list.

Thank you for the promot review!
I will do this in days and notify you.

@chiwwang
Copy link
Contributor Author

chiwwang commented Aug 8, 2024 •
edited
Loading

Turning to draft. I will open the PR again after

  1. address feedback
  2. add x86 emulator section
  3. mention support model list (basically, those in examples/qualcomm)

@chiwwang chiwwang marked this pull request as draft August 8, 2024 04:54
@chiwwang chiwwang marked this pull request as ready for review August 9, 2024 05:00
@chiwwang
Copy link
Contributor Author

chiwwang commented Aug 9, 2024

Hi @cccclai ,
I addressed the comments. Please take a look if you happen to have some time :)

@winskuo-quic winskuo-quic mentioned this pull request Aug 9, 2024
Merged
@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Aug 9, 2024

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

@facebook-github-bot facebook-github-bot merged commit 3f9b39e into pytorch:main Aug 9, 2024
@cccclai cccclai mentioned this pull request Aug 9, 2024
Merged
@guangy10
Copy link
Contributor

guangy10 commented Aug 9, 2024 •
edited
Loading

I ran into issues with the updated guide:
./backends/qualcomm/scripts/build.sh --release

/data/users/guangyang/executorch/examples/qualcomm/llama2/qaihub_runner/runner.cpp:21:10: fatal error: 'executorch/examples/models/llama2/runner/util.h' file not found
   21 | #include <executorch/examples/models/llama2/runner/util.h>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.

Filed an issue #4655. cc: @cccclai

@cccclai
Copy link
Contributor

cccclai commented Aug 10, 2024

Looks like cause by #4588

@chiwwang chiwwang deleted the update_qc_doc branch August 12, 2024 06:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@cccclai cccclai cccclai approved these changes

Assignees

No one assigned

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@chiwwang @facebook-github-bot @guangy10 @cccclai