Date: Sat, 25 Aug 2001 19:33:35 +0300 From: Valentin Nechayev <netch@iv.nn.kiev.ua> To: Peter Pentchev <roam@ringlet.net> Cc: freebsd-hackers@FreeBSD.ORG Subject: Re: function calls/rets in assembly Message-ID: <20010825193335.C761@iv.nn.kiev.ua> In-Reply-To: <20010825160302.A559@ringworld.oblivion.bg>; from roam@ringlet.net on Sat, Aug 25, 2001 at 04:03:02PM %2B0300 References: <20010824110805.C88259@dragon.nuxi.com> <XFMail.010824113645.jhb@FreeBSD.org> <20010825154427.B761@iv.nn.kiev.ua> <20010825160302.A559@ringworld.oblivion.bg>
next in thread | previous in thread | raw e-mail | index | archive | help
Sat, Aug 25, 2001 at 16:03:02, roam (Peter Pentchev) wrote about "Re: function calls/rets in assembly": > I wonder if a mentioning of -mpreferred-stack-boundary should be > added to tuning(7).. This will be quite strange idea. Tuning which reduces 2-4 times stack size of userland application... hm... corporations spend years and millions $$ to reach such effect... should man page say "first at all, add the right option to /etc/make.conf and rebuild the whole userland, because gcc is written by morons and we are too lazy or stupid to remove their ugly crap"? Matt Dillon is right that the best variant is to exclude this brain-damaged option at all. If gcc team wants to implement proper alignment to work with SSE and other high-specialized stuff, they should learn commands for bitwise AND, and use only where really needed. /netch To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010825193335.C761>