elvin-n commented on PR #11505: URL: https://github.com/apache/tvm/pull/11505#issuecomment-1142463429
> Any thoughts about using a library like https://github.com/pytorch/cpuinfo to determine this information vs rolling our own? Thanks you for the pointing of the library. As for the question - on the one hand my patch fixes only Intel flow while AMD might still work wrong. And from this point of view it's better to reuse some solution which covers more platforms. On the other hand for runtime - it's better to have as small solution as it can and it is unknown how lightweight pytorch/cpuinfo is. Even taking into account ability to be compiled into static library. I would move forward with my patch and if we want to switch to pytorch/cpuinfo, we can do this later -- 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]
