Re: [ECOS] redefined symbols on weak binding

[Jonathan Larmour <jlarmour at redhat dot com>]

luca.spinacci@marconicomms.com wrote:
> 
> I have to link an archive of power-pc-eabi gcc compiled files with diab compiled
> binaries and link under the diab environment.
> The linker returns warnings on several symbol redefinitions (...not making an
> elf).
> Actually all the redefined symbols have a WEAK bind tag in their symbol table (I
> expected that a symbol with a weak bind tag wasn't a reason of problems even if
> no globl bind tag was found at link time).
> Then I tried to understand more in weak bind tagging: I created two trivial
> classes and their files...

This question is probably more appropriate for the
binutils@sources.redhat.com list. In my opinion, having two weak symbols
should not result in a conflict, but I don't know what any standards say
about it. Obviously GNU ld doesn't treat it as an error. The binutils list
should know.

Jifl
-- 
Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
Maybe this world is another planet's Hell -Aldous Huxley || Opinions==mine