arsenm wrote: amdlibm seems to use full vector alignment since it uses __mm types. For the current usage it's technically fine, we're not depending on these alignments being as optimistic as possible
https://github.com/llvm/llvm-project/pull/237486 _______________________________________________ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
