[LLVMdev] Removing -tailcallopt?
jyasskin at google.com
Fri Feb 5 21:03:21 PST 2010
Thanks for improving this! Since no good deed can go unpunished, could
you also update the LangRef and codegen documents to describe the new
state of the world? It might also be nice to mention there that tail
calls can be needed for correctness in addition to performance and how
users can guarantee that a call will be transformed, since both if us
have forgotten recently. I can take a stab at the second one if you
don't have time.
On Friday, February 5, 2010, Evan Cheng <evan.cheng at apple.com> wrote:
> Hi all,
> I've added tail call optimization to x86. This is different from what -tailcallopt does, which forces fastcc function to be tail callable. My changes detect opportunities to do tail call without having to change the ABI.
> I've looked at the codegen of -tailcallopt and it doesn't look all that good. Running it as a llcbeta option shows it significantly pessimize code in most cases.
> As far as I can tell only PPC and X86 targets are supporting this option. Does anyone actually using it? I'd prefer to just remove it to clean up the implementation if no one has any objections.
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
More information about the llvm-dev