davemgreen wrote: Usually when new ISD nodes are added they are expanded for all types, so that every backend will get at least working code even if it is not optimal. The targets can then come along and override the defaults for the types they are interested in, to get better results.
For tan I would expect most vector types would want to scalarize, so marking them as expand would make sense. If more types than are necessary get marked as Expand that shouldn't be an issue, it looks like we already do that for a number of other nodes. https://github.com/llvm/llvm-project/pull/94559 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
