This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Indirect C-function calls not sibcall-optimised?


Could someone point out to me why it is not possible in gcc 3.1 (Don't know 
about newer CVS-versions.) to use sibcall-optimisation with indirect 
function calls?  Maybe I am missing something obvious, but I can't 
understand the rationale behind it.

I'm aware of the typical sibcall constraints and I know how to bypass or 
work-around most of them, but I'd like to understand what consequences this 
could have with other than my little test-programs.

Currently, I am busy trying to extend the sibcall mechanism to support a 
wider range of optimisation-cases, or come up with a more general approach 
to the "tail-call problem".

Thanks in advance,
Andi.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]