Date: Sat, 9 Aug 2003 18:20:26 -0600 From: Tillman <tillman@seekingfire.com> To: sparc64@freebsd.org Subject: Re: ponderous 'make world' times post GCC 3.3... Message-ID: <20030809182026.W16062@seekingfire.com> In-Reply-To: <20030809235533.GA56513@dragon.nuxi.com>; from obrien@freebsd.org on Sat, Aug 09, 2003 at 04:55:33PM -0700 References: <20030807062536.GA68747@dragon.nuxi.com> <p052106c7bb59ce43912c@[128.113.24.47]> <20030809084019.GA4704@rot13.obsecurity.org> <20030809235533.GA56513@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Aug 09, 2003 at 04:55:33PM -0700, David O'Brien wrote: > On Sat, Aug 09, 2003 at 01:40:19AM -0700, Kris Kennaway wrote: > > This suggests that something might have been pessimized with the gcc > > 3.3 code generation on sparc. i.e. when gcc 3.3 is built with gcc 3.2 > > (your build 2, which used the installed 3.2 compiler to build the 3.3 > > compiler that was then used for the remainder of the world build), it > > performs well (your build 2), but when gcc 3.3 is built with gcc 3.3 > > it performs slowly. > > > > I wonder if the gcc 3.3 compiler specs were set up incorrectly for > > sparc. > > I can think of nothing in the sparc MD compiler specs that could account > for this. Mostly things there are in the "work / doesn't work" category. Is there some kind of profiling that I could perform during a buildworld that might be helpful? -T -- Humans hardly ever learn from the experience of others. They learn - when they do, which isn't often - on their own, the hard way. - Robert Heinlein
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030809182026.W16062>