Lunderberg opened a new pull request, #14918: URL: https://github.com/apache/tvm/pull/14918
This PR refactors `SplitHostDevice` into three separate transformations. Previously, `SplitHostDevice` would replace device regions with a `builtin::tvm_call_packed()` node to replace the extracted region. After this PR, this process is performed in three separate steps. 1. `AnnotateDeviceRegion`: Annotate the regions that should be executed on another target. 2. `SplitHostDevice`: Extract the annotated region into an independent PrimFunc, with a `GlobalVar` to represent the call from into the new subroutine. 3. `LowerDeviceKernelLaunch`: For any subroutine call where the caller and callee are on different devices, replace with a device kernel launch. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
