cloud-mxd opened a new pull request #6452: URL: https://github.com/apache/incubator-tvm/pull/6452
I'm using tvm to convert pytorch model, but raise error like this: _"intel_ mkl error: parameter 2 was incorrect on entry to cblas_sgemm." After using gdb debug, I found that the problem was that the inline function was redefined, and mkl incorrectly used the BooleanToTranspose function defined in cublas. This was mainly because the inline function was not marked with static or was not included in the anonymous namespace. Please help review whether this change is reasonable. Thanks ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
