cor3ntin wrote: > > If split stacks are negatively impacting profiling, debugging, or other > > compiler development tasks, I have to ask, have we considered optimizing > > clang stack usage? > > Strong +1 to this.
I removed a few uses of ParsedAttributes recently, it helped https://github.com/llvm/llvm-project/pull/132021#issue-2931754944 - I'm sure there is a lot more room for improvement I wish we had some way to profile whether our SmallVectors and other SSO optimizations are reasonably dimensioned. We also store references to Sema in a ton of places where, realistically, Sema might as well be a singleton. I wonder if we do that enough to have a measurable impact on stack size. But anyway, we still need a way to grow the stack :) https://github.com/llvm/llvm-project/pull/133173 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
