[PATCH] ld/ldlang.c: fatal error on architecture mismatch

[Alan Modra]

On Mon, Apr 30, 2007 at 07:19:37AM -0400, Daniel Jacobowitz wrote:
> On Mon, Apr 30, 2007 at 05:51:36PM +0930, Alan Modra wrote:
> > I'm not strongly opposed to the patch either, since someone wanting to
> > link foreign object files can always turn off the error.
> 
> I think it should be a fatal error by default.  I was going to say
> that we needed a documentation update to go with this patch, but
> although the option is --no-warn-mismatch the documentation says:
>  Normally ld will give an error [snip]

Yeah, OK, patch approved then.

> While we're on the subject of errors, it bugs me that the linker warns
> and auto-selects an address if you don't provide _start or specify an
> entry point.  How do you feel about making that fatal?

It might impact the testsuite.  Otherwise I'm OK with the idea.

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre