pogo59 wrote: > `optnone` is a very strange attribute; it's a directive to the compiler with > no clear semantics.
In an ideal world, an `optnone` function would be compiled as-if the entire compilation unit had `-O0`. At `-O0`, Clang attaches `optnone` to every function, implying that this equivalence is the exact intent. Clang does this so that `-O0 -flto` will not lose the `-O0` semantic during LTO. When we devised `optnone`, there was an argument that this goal was not practical, and so the description could not imply that it was in fact equivalent to `-O0`. Thus, the description is fuzzy. Perhaps that was a mistake, and we should revise the description to include the goal, even if the goal is not 100% attainable. With that goal in mind, having `optnone` and `-O0` be deliberately different here makes no sense. https://github.com/llvm/llvm-project/pull/85605 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
