Skip to content
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.

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

[microTVM] Fix host-driven AOT memory workspaces #13807

Merged
merged 5 commits into from
Jan 26, 2023
Merged

[microTVM] Fix host-driven AOT memory workspaces #13807

merged 5 commits into from
Jan 26, 2023

Conversation

alanmacd
Copy link
Contributor

@alanmacd alanmacd commented Jan 18, 2023

When using host-driven AOT with memory pools enabled, the workspace and constant memory were not properly supported. In order for them to work properly, the _run function (typically tvmgen_default_run()) needed to be called instead of tvmgen_default___tvm_main__() in order to properly setup the memory workspace pointers.

fixes #13777

@tvm-bot
Copy link
Collaborator

tvm-bot commented Jan 18, 2023

Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews from Reviewers by @-ing them in a comment.

Generated by tvm-bot

@alanmacd alanmacd changed the title Fix host-driven AOT memory workspaces [microTVM] Fix host-driven AOT memory workspaces Jan 18, 2023
@mehrdadh mehrdadh self-assigned this Jan 19, 2023
@leandron
Copy link
Contributor

Cc @ashutosh-arm @d-smirnov

@mehrdadh
Copy link
Member

Overall LGTM! I'll wait for @d-smirnov before merging it. Thanks @alanmacd!!

@mehrdadh
Copy link
Member

@alanmacd @d-smirnov thanks! this PR is merged.

@mehrdadh mehrdadh merged commit 4ddb876 into apache:main Jan 26, 2023
@alanmacd alanmacd deleted the fix-host-aot-mem-pools branch January 26, 2023 23:57
fzi-peccia pushed a commit to fzi-peccia/tvm that referenced this pull request Mar 27, 2023
When using host-driven AOT with memory pools enabled, the workspace and constant memory were not properly supported. In order for them to work properly, the _run function (typically tvmgen_default_run()) needed to be called instead of tvmgen_default___tvm_main__() in order to properly setup the memory workspace pointers.

fixes apache#13777
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[Bug][microTVM]Enabling USMP with VWW model has error
6 participants