From: "Andrew Crabtree" <andrewc AT rosemail DOT rose DOT hp DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: COFF obsolete, let's port to Elf for version 3
Date: Tue, 19 May 1998 14:03:58 -0700
Organization: Hewlett-Packard, Roseville
Lines: 28
Message-ID: <6jss3q$s86$1@rosenews.rose.hp.com>
References: <m0ybP3H-000S3GC AT inti DOT gov DOT ar> <6jpugg$luo$1 AT rosenews DOT rose DOT hp DOT com> <LTV1OBA+LeY1Ews4 AT talula DOT demon DOT co DOT uk>
NNTP-Posting-Host: ros51675cra.rose.hp.com
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk


Shawn Hargreaves wrote in message ...
>I know that Linux doesn't use the underscore prefix with ELF objects,
Correct.  Nor Sys5 or hp or sun.

>but surely this is just a convention of what the compiler has been
>configured to output
I believe it should be possible to confiure gcc to output for elf and to
prepend underscores but ...

> and not a specific requirement of the object format?
It is :(   I was told this on the gcc developers list some time ago when I
reported problems related to libgcc-test.c having to do with pre-pending
underscores., and just now (before replying) verified it in the elf 1.1
standard.

1-17 "Note - External C symbols have the same names in C and object files'
symbol tables."

>If ELF really doesn't allow these underscores, I think that would be a
>very strong reason to continue using COFF.
Definitely.  I hadn't even thought of the implication that all (most)
assembly would have to change.

Andy