This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: How inlined are inlined functions?
- To: mark at codesourcery dot com (Mark Mitchell)
- Subject: Re: How inlined are inlined functions?
- From: Joe Buck <jbuck at synopsys dot COM>
- Date: Fri, 3 Dec 99 12:11:42 PST
- Cc: oliva at lsd dot ic dot unicamp dot br, jimb at cygnus dot com, egcs at sourceware dot cygnus dot com
> >>>>> "Alexandre" == Alexandre Oliva <oliva@lsd.ic.unicamp.br> writes:
>
> Alexandre> break? Shouldn't such functions only be inlined when
> Alexandre> they're explicitly declared inline?
>
> The new code only inlines functions that are marked `inline'. It
> could be extended to what -O3 does now, but in that case, we would
> have to be careful about some of these things.
You are more of an expert on the standard than I am, but it seems that
the issue of trashing of automatic variables shouldn't be ignored even
if the keyword "inline" appears.