GitLab

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

It doesn't look like this functionality is used. Probably was a thing in
ONNX that became deprecated when they adopted NumPy-like broadcasting.

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

change to shape::operator== makes it so this will work the same

Local Threads of multiples 32 were introduced in #1348
But LocalThreads that are not multiple of 64 are causing correctness issues.

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

Upgraded Dockerfiles and fixed tidy issues to make Ubuntu 20.04 and ROCm 5.3.0 the default 

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_ref_multibroadcast

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

updated GPU pad to now use JIT version.
added range functions for JIT kernels.

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast

Fixes an observed regression error on certain Frozen Protobuf models due to PR 1280

use_dynamic_same_auto_pad was removed from convolution, but the driver models still retain the fields. This PR regenerates the files so that they are compatible again.