Date: Fri, 23 Aug 1996 15:44:21 -0700 From: "Justin T. Gibbs" <gibbs@freefall.freebsd.org> To: Stefan Esser <se@zpr.uni-koeln.de> Cc: "Michael L. VanLoon -- HeadCandy.com" <michaelv@HeadCandy.com>, "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com>, jas@flyingfox.com (Jim Shankland), freebsd-hardware@freebsd.org Subject: Re: SCSI again: Asus SC200 vs. Adaptec 2940 Message-ID: <199608232244.PAA21647@freefall.freebsd.org> In-Reply-To: Your message of "Sat, 24 Aug 1996 00:14:39 %2B0200." <199608232214.AAA22908@x14.mi.uni-koeln.de>
next in thread | previous in thread | raw e-mail | index | archive | help
>Well, anyway: I know that you can't accept being kept back >for too long, and I appreciate the changes you have prepared >and hope they'll be intergrated soon. And I know there are >dependencies ... Well, I haven't been held back too much. I know have the generic SCSI layer performing all tagged-queueing processing. Basically the lower level drivers read the "tag_type" field out of the scsi_xfer structure and if non-zero send it to the device. I've also implemented bowrite which converts sync writes to ordered async writes giving around a 10% improvement on some of the ad-hoc benchmarks I've tried. Its not as fast as mounting async (that does delayed writes), but I'll take a safe 10% improvement any day. >Regards, STefan -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations ===========================================
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199608232244.PAA21647>