Date: Sun, 17 Jan 1999 17:41:12 +0200 From: Mark Murray <mark@grondar.za> To: "Andrey A. Chernov" <ache@nagual.pp.ru> Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/gnu/usr.bin/perl/libperl config.SH-aout.i386 config.SH-elf.alpha config.SH-elf.i386 Message-ID: <199901171541.RAA05521@greenpeace.grondar.za> In-Reply-To: Your message of " Sun, 17 Jan 1999 18:35:30 %2B0300." <19990117183530.A1460@nagual.pp.ru> References: <199901170953.BAA10010@freefall.freebsd.org> <19990117162723.A58648@nagual.pp.ru> <199901171503.RAA87113@greenpeace.grondar.za> <19990117183530.A1460@nagual.pp.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
"Andrey A. Chernov" wrote: > On Sun, Jan 17, 1999 at 05:03:07PM +0200, Mark Murray wrote: > > I fully understand that the current state-off-affairs does not work as it > > should. I reverted it because is broke "make world" or "make release" for > > some folks. > > What the breakage is exactly? Of course I test my change before commiting > it, but by simple perl rebuilding from clean state, not from 'make world'. In some folks' "make world", during the build of dynaloader, stdio.h was not found. *Boom* > > I intent to keep it there as a reminder to fix that line. > > The line works fine for libperl. > > As result perl *.so modules not optimized at all causing performance > problems :-( It was the reason why I change it... I agree this is a problem. :-) Any perl-build changes must be tested through _at_least_ a "make world", and I am coming to believe that it should also be run through a "make release" (I am prepared to admit that making this a requirement is on the difficult side, but if "make release" fails, then the commit gets tossed). M -- Mark Murray Join the anti-SPAM movement: http://www.cauce.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199901171541.RAA05521>