[fpc-devel] Registers allocation technique

Lag Programming lagprogramming at aim.com
Tue Sep 30 18:37:40 CEST 2014


   I've sent an email to the list that didn't appeared so I've reformulated my partial(or complete lack) of understanding the subject.
   Please reply if you agree or disagree with the following two statements for x86_64 target:
1) "Unless a function calls another function, the function result register will always be in rax. Calling for inlined functions doesn't change this behaviour as it's not a real function call."
2) "Operating with rax,rbx,rcx,rdx,rbp,rsp,rsi,rdi registers should be considered faster or as fast as operating with r8,r9,r10,r11,r12,r13,r14,r15." This refers to movq, incq, cmpq...whatever.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freepascal.org/pipermail/fpc-devel/attachments/20140930/75ff4f52/attachment.html>


More information about the fpc-devel mailing list