From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 26511 invoked by alias); 11 May 2005 14:45:07 -0000 Mailing-List: contact binutils-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: binutils-owner@sources.redhat.com Received: (qmail 26432 invoked from network); 11 May 2005 14:45:00 -0000 Received: from unknown (HELO omta04sl.mx.bigpond.com) (144.140.93.156) by sourceware.org with SMTP; 11 May 2005 14:45:00 -0000 Received: from grove.modra.org ([144.136.167.90]) by omta04sl.mx.bigpond.com with ESMTP id <20050511144458.QQCD28740.omta04sl.mx.bigpond.com@grove.modra.org> for ; Wed, 11 May 2005 14:44:58 +0000 Received: by bubble.grove.modra.org (Postfix, from userid 500) id 5EE341AA5E5; Thu, 12 May 2005 00:14:58 +0930 (CST) Date: Wed, 11 May 2005 14:59:00 -0000 From: Alan Modra To: binutils@sources.redhat.com Subject: Re: powerpc new PLT and GOT Message-ID: <20050511144458.GB29302@bubble.grove.modra.org> Mail-Followup-To: binutils@sources.redhat.com References: <20050511141249.GA29302@bubble.grove.modra.org> <20050511142209.GA10062@nevyn.them.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050511142209.GA10062@nevyn.them.org> User-Agent: Mutt/1.4i X-SW-Source: 2005-05/txt/msg00395.txt.bz2 On Wed, May 11, 2005 at 10:22:09AM -0400, Daniel Jacobowitz wrote: > I'm not oing to look at the code, but I have one comment on the ABI. > You've got a data section named .plt and a code section named .glink > (well, you note that the text stubs can be anywhere in the text > segment, but that's true on most architectures; I imagine that for now > they're all going in a single .glink section). Yes, they go in a single linker created .glink section, but this section doesn't have it's own output section. Instead, it contributes to .text. > Everyone else seems to call the data section ".got.plt" and the text > section ".plt". GDB even knows about this; it uses the name ".plt" to > help in skipping dynamic linker code. Is there a reason why you can't > use the traditional names? I could, but .plt is traditionally a (somewhat) regular array. .glink doesn't really fit this model since it could be dispersed throughout the text segment, something you might want to do in large programs where the 24-bit powerpc branch offset is limiting. I know I don't support such a .glink at the moment on powerpc (as we do on powerpc64), but I might in the future. In that case you couldn't output .glink to its own output section, so gdb could not use the section name to skip plt call stubs. Then, given that .glink doesn't really fit the traditional .plt, I don't want to use .got.plt because I feel some section ought to be called .plt, simply because that's the traditional name for sections associated with dynamic function linkage. Powerpc64 also has a .plt that just consists of data. You're not the first to suggest .got.plt though. :) -- Alan Modra IBM OzLabs - Linux Technology Centre