This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: analysis: arch=i686, -fPIC won't play nice on OSR5
- To: Robert Lipe <robertl at dgii dot com>
- Subject: Re: analysis: arch=i686, -fPIC won't play nice on OSR5
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Tue, 30 Jun 1998 23:42:02 -0600
- cc: Ulrich Drepper <drepper at cygnus dot com>, egcs-patches at cygnus dot com
- Reply-To: law at cygnus dot com
In message <19980630223944.A2325@dgii.com>you write:
> [ moved to egcs-patches. Discussion was on egcs. ]
>
> > > If this sounds like the right solution, let me know and I'll do the
> > > patch thing.
> >
> > It's correct that the internal labels need no type information. If
> > your patch simply avoids adding the .type line it is correct.
>
> I've run this code through the testsuite on every i386 host/target that
> I can (i586 and i686, linux and openserver, elf, elf + pic, coff) and
> don't think I have have any reason to distrust it so far.
>
> Unfortunately, I have completely too many build trees right now
> so I cannot say with absolute certainty that it hasn't caused any
> regressions.
I would indicate in a comment in that code which assembler couldn't
handle the type information for local labels.
That way if later we find an assembler that can't handle calls to
local labels we'll know why the label isn't declared as a fucntion
with type information and such.
Consider it approved after adding the comment.
jeff