shiltian wrote: > I would like to avoid adding additional special properties to AS0, or > defining the flat concept.
How can we add a new specification w/o defining it? > The simple solution is just have a switch over the target architecture in > Attributor. That means in the future when a new (GPU/accelerator) target is added, someone has to know that is one place to add a new case. FWIW, it is not just Attributor. The situation applies to all passes that potentially don't have access to TTI/TM. I don't really like the solution, but that is what I'm doing to move other PRs forward w/o being blocked by this one. https://github.com/llvm/llvm-project/pull/108786 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
