Date: Mon, 22 Aug 2005 09:40:02 +0200 From: Divacky Roman <xdivac02@stud.fit.vutbr.cz> To: Kris Kennaway <kris@obsecurity.org> Cc: current@freebsd.org Subject: Re: [PATCH]: fdalloc optimization Message-ID: <20050822074002.GA31176@stud.fit.vutbr.cz> In-Reply-To: <20050821175406.GA36164@xor.obsecurity.org> References: <20050821084546.GA77111@stud.fit.vutbr.cz> <20050821175406.GA36164@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Aug 21, 2005 at 01:54:06PM -0400, Kris Kennaway wrote: > On Sun, Aug 21, 2005 at 10:45:46AM +0200, Divacky Roman wrote: > > hi, > > > > I made a patch for fdalloc which could speed it up a bit, its stored under > > kern/85176 or http://hysteria.sk/~neologism/kern_descrip.patch > > That's all nice and everything, but *does* it speed it up a bit? :-) this is my buildkernel test: old kernel: 1. 250.560u 23.428s 4:47.09 95.4% 3243+2002k 1911+977io 235pf+0w 2. 249.645u 25.571s 4:38.72 98.7% 3240+1998k 89+961io 3pf+0w 3. 249.976u 25.356s 4:38.56 98.8% 3238+1997k 87+974io 3pf+0w new kernel: 1. forgot to prepend time command ;( 2. 250.281u 24.268s 4:38.49 98.5% 3245+1998k 92+996io 3pf+0w 3. 249.279u 25.390s 4:38.08 98.7% 3240+1997k 83+980io 3pf+0w ie. its never slower but I think this is insignificant. on the other hand - buildkernel is not that fdalloc sensitive. dont know how to measure it ;( roman
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050822074002.GA31176>