General purpose GPU compute framework built on Vulkan to support 1000s of cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends). Blazing fast, mobile-enabled, asynchronous and optimized for advanced GPU data processing usecases. Backed by the Linux Foundation.
@MiroPalmu I have added this PR with a few fixes from the work you've done, really great push, the contribution is very much appreciated. This should now work as expected, two tests have been added - namely copying with the OpTensorCopy, and copying through an Algorithm.
Extra in this PR:
Ensuring SPDLOG is used by default in build config
Ensuring spdlog is set up as sync by default, but allow feature flag (otherwise debugging harder on segfault)
MiroPalmu
commented
1 year ago
Addresses part of #14 as continuation from #304.
@MiroPalmu I have added this PR with a few fixes from the work you've done, really great push, the contribution is very much appreciated. This should now work as expected, two tests have been added - namely copying with the OpTensorCopy, and copying through an Algorithm.
Extra in this PR: