On Friday 01 January 2010 11:29:41 Joakim Tjernlund wrote: > Mike Frysinger <vap...@gentoo.org> wrote on 01/01/2010 07:18:44: > > yes, but that doesnt mean gcc takes care of inlining all of printf_puts() > > into the puts() and all the new call sites go to puts() > > Sure, gcc might not inline the current code in this case. I guess > you don't want the extra size this would incur or is there some else you > are concerned about?
the extra size > > i dont have any plans on wanting this, and it seems pretty invasive ... > > and easy to introduce new code that breaks PIC people but no one else > > really notices ... > > Yes, it is a bit invasive hence the question if this is acceptable to > u-boot. I have been looking for other ways to do this but there isn't one > sans modifying gcc. You are sort of an gcc guy, what do you think the odds > are > that gcc would add a few new options mainly useful for smaller > embedded progs like u-boot (and possible the kernel too)? i'm not really a gcc guy ... i just sometimes fake it. i honestly dont have an idea here how they would respond. -mike
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
