On Wed, 2024-05-22 at 11:01 -0700, Martin KaFai Lau wrote: [...] > Inlining it would be nice (as Eduard mentioned in another thread). I also wonder > if Eduard's work on the no caller saved registers could help the dynptr kfunc? I > think the dynptr kfunc optimization could be a followup. For the context: https://clang.llvm.org/docs/AttributeReference.html#no-caller-saved-registers Basically the attribute says that compiler does not need to save all r0-r5 registers for some function calls. My changes for LLVM/verifier are not public yet, I'll try to speedup.
