Re: SDAREL output sections

[Geoff Keating <geoffk at geoffk dot org>]

> Cc: binutils@sourceware.cygnus.com
> From: Momchil Velikov <velco@fadata.bg>
> Date: 19 Feb 2001 00:50:26 +0200
> 
> >>>>> "Geoff" == Geoff Keating <geoffk@geoffk.org> writes:
> 
> >> Cc: binutils@sourceware.cygnus.com
> >> From: Momchil Velikov <velco@fadata.bg>
> >> Date: 18 Feb 2001 23:50:47 +0200
> 
> >> Hmm, sorry, obviously I didn't get it after comments that perfectly
> >> working programs are rejected by the linker for the sake of catching
> >> gcc errors.
> 
> Geoff> Perhaps the following analogy might help: lots of perfectly clear
> Geoff> programs are rejected by the compiler for the sake of catching
> Geoff> programmer errors.
> :-)
> 
> Geoff> (It's not just gcc errors.  GCC doesn't trip this test if used
> Geoff> properly.  The problem is that occasionally people forget to use it
> Geoff> properly.)
> 
> Well, then, how about changing the checks to test the input section
> instead of the output one and letting the linker report relocation
> overflows if the input section is inappropriately output ?

Because this would prevent people from overriding the test by folding,
say, the .data section into .sdata.

-- 
- Geoffrey Keating <geoffk@geoffk.org>