mbs-octoml opened a new pull request #9421: URL: https://github.com/apache/tvm/pull/9421
The VM ManifestAlloc pass was allocating constants in a few places I forgot to tag with on_device for the host/cpu. As a result the runtime would (silently) do the x-device copy, which destroys perf. To make this easier to spot in the future added a 'constants' property to the VM Executable to dump the shape & device for all VM constants. This is CORE-102 in OctoML JIRA. (Special thanks to @mbrookhart & @tkonolige for finding this.) -- 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]
