This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: [PATCH] allow easier overriding of ELF_DYNAMIC_INTERPRETER
- From: Nick Clifton <nickc at cambridge dot redhat dot com>
- To: obrien at FreeBSD dot org
- Cc: GNU Binutils mailing list <binutils at sources dot redhat dot com>
- Date: 17 Feb 2002 08:53:45 +0000
- Subject: Re: [PATCH] allow easier overriding of ELF_DYNAMIC_INTERPRETER
- References: <20020125234343.A66746@dragon.nuxi.com><20020126101345.GA24173@rembrandt.csv.ica.uni-stuttgart.de><20020128114017.A70744@dragon.nuxi.com><20020128122518.V8364@dr-evil.shagadelic.org><20020128132808.B72046@dragon.nuxi.com><m3ofitwrtx.fsf@north-pole.nickc.cambridge.redhat.com><20020213075528.C5788@dragon.nuxi.com><m3u1slunks.fsf@north-pole.nickc.cambridge.redhat.com><20020217000937.A60590@dragon.nuxi.com>
Hi David,
> This is the basic patch I have to do this. In general, does this
> look OK?
Yes, but I am a little concerned. How is a configure time option
going to be able to override the explicit inclusion of one of these
headers ?
> #include "elf/cris.h"
> +#include "elf32-cris-dyn.h"
Maybe it would be better to have a configure-created "tconfig.h"
header file (ala gcc) which then includes things like the elf/cris.h
and elf32-cris-dyn.h ?
> Are there issues with the length of the header file names?
Ideally it would be nice to have file names that are unique within the
first 8 characters in order to avoid problems with FAT16 based file
systems. So maybe you could leave off the "elf32-" component of the
file name and just have "dyn-arm.h" "dyn-cris.h" etc. After all there
is no real need to say "elf32" for this particular definition.
Cheers
Nick