From mint-bounce@lists.fishpool.fi Mon Mar 9 14:18:43 2009 Message-ID: <49B55D06.8070103@atari-source.org> Date: Mon, 09 Mar 2009 14:16:38 -0400 From: Mark Duckworth User-Agent: Thunderbird 2.0.0.19 (Macintosh/20081209) MIME-Version: 1.0 To: Keith Scroggins CC: mint@fishpool.com Subject: Re: [MiNT] Should Setting Stack be added to Binutils? References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-SA-Score: -1.4 X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-to: mint-bounce@lists.fishpool.fi X-original-sender: mduckworth@atari-source.org Precedence: bulk List-help: List-unsubscribe: List-Id: X-List-ID: List-subscribe: List-owner: List-post: Yeah this was discussed and it would be great to be able to add it to LDFLAGS or something but I looked over the binutils source and I got confused *very* quickly. I'm not sure how the --stack setting on i386 is passed through to the BFD. Thanks, Mark Keith Scroggins wrote: > I did a little work on bootstrapping GCC 4.3.3 this weekend, and had > problems with the stack not being set when going to the next stage. > What I was wondering was if we could/should add stack functionality to > Binutils so you could pass a linking flag of something like > --stack=512k to have it set? Any input? > > And, while thinking Binutils, any thoughts on updating the default > Sparemint version with the newer 2.19.1? What all should be tested? > > Here is a snippet of email between myself and Vincent from about an > hour ago: > > >> I actually got GCC 4.3.3 to stage 3 and then failed in AranyM this >> weekend > > That's very good news ! > You have made most of it. > >> (it was problems with c++ lib I think, failed during configure for it). > To be investigated... > >> Anyway, the point is, I am wondering if it is worthwhile to add stack >> setting capability to binutils directly, so it can be set right after >> an exe is linked, and could just be an additional LD FLAG. Maybe >> something along the line of --stack=512k or similar. Basically, when >> doing this locally, without having hacked the Makefiles, need to >> catch when the exe's are made, and if the build dies, and they have >> had stack adjusted, gcc will rebuild the exe once make is restarted. >> >> Did any of that make sense? I was thinking about sending this to the >> MiNTList, but was not really sure if it was a good thought or not.... > > I definitely agree with you. > Some time ago, I posted a message with several solutions to the stack > problem. I have been thinking to that recently, finally the linker > option is the better (i386 targets already have that !). So it would > be probably possible to type "configure LDFLAGS=--stack=512k", or > maybe the same with make. I don't think it is too much difficult to do. > > Go ahead, post this to the MiNT mailing list ! > > Vincent > >