Nanmur commented on PR #20412: URL: https://github.com/apache/tvm/pull/20412#issuecomment-5980086599
@tlopex The requested dtype handling has been implemented in the latest revision (`b7ec429156`). Both `div` and `truncmod` now reject only `kDLBool`, preserving the previous behavior for BF16, FP8, and other dtype codes. Regression coverage was added for `bfloat16`, `float8_e4m3fn`, and `float8_e5m2` with both operators. The previous CPU failure was unrelated to this PR: it came from obsolete `T.Buffer` syntax in a WebGPU test and has since been fixed upstream by #20538. A fresh Jenkins run is now pending. The current `CHANGES_REQUESTED` status is still attached to the earlier revision. Could you please re-review the latest revision when convenient? -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
