This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
RE: Problem with linker with binutils-040414
- From: "Dave Korn" <dk at artimi dot com>
- To: <binutils at sources dot redhat dot com>
- Date: Mon, 19 Apr 2004 14:29:04 +0100
- Subject: RE: Problem with linker with binutils-040414
> -----Original Message-----
> From: binutils-owner On Behalf Of Nick Clifton
> Sent: 19 April 2004 14:21
> Hi Guys,
>
> > The [BFD] library should never abort
>
> Alan Modra wrote:
>
> >>Yes, and gcc should never ICE. The situation is the same, I think.
> >>
> >>
> Similar, but
> a) GCC handles its ICEs more gracefully than does BFD.
> (Well unless
> we improve BFD).
> b) GCC is not a library, it is an entire application. It
> is able to
> detect and handle is internal errors in an entirely consistent manner.
>
> BFD is a library, and it should not dictate error handling
> policy to its
> users. Imagine if GCC used the BFD library but could not
> generate its
> ICE messages because the BFD library aborted before giving GCC the
> chance to intervene.
>
> Cheers
> Nick
>
Maybe what bfd really needs is an abstracted exception handling mechanism,
that defaults to calling abort, but that bfd-exception-aware applications
can override to attempt to deal with in their own fashion (even if that's
just issuing a gcc-style ICE message and aborting).
cheers,
DaveK
--
Can't think of a witty .sigline today....