vinx13 opened a new pull request #9420: URL: https://github.com/apache/tvm/pull/9420
When checking `Store` in `VerifyGPUCode`, check the vector size of the value of store instead of the index. Even though the vectorized index like `int32x8` might exceed max vector size, in CUDA/OpenCL only the vector size of the value matters because only the start index is needed in Codegen. -- 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]
