On Monday 22 December 2008 06:20:27 Graeme Russ wrote:
On Mon, Dec 22, 2008 at 10:44 PM, Mike Frysinger vapier@gentoo.org wrote:
On Monday 22 December 2008 04:16:33 Graeme Russ wrote:
On Mon, Dec 22, 2008 at 8:05 PM, Mike Frysinger wrote:
On Saturday 13 December 2008 00:26:26 Graeme Russ wrote:
This patch makes all definitions, declarations and usages of weak functions consistent.
a quick glance shows that it breaks things (the ELF and Blackfin stuff certainly appears to be wrong). i'm guessing you focused on style for the RFC part rather than the result actually being correct ... -mike
Thanks for having a look at this. Would it be possible for you to be a bit more specific about 'wrong' if for nothing other that me gaining a better understanding of how it works, and how it breaks
you set the aliases to functions that do not exist
Ack - I can see that for the ELF - the main function needs to be renamed __do_bootelf_exec ()
I cannot see the problem with Blackfin - I will freely admit that Blackfin has been fundamentally changed (any therefore needs thorough testing), but unless the Blackfin toolchain treats weak function linking differently, it _should_ "just still work"(tm)
Maybe I am not seeing the obvious?
sorry, you're right, you added the new function. that said, i dont think marking those functions as "inline" is correct or makes sense. -mike