Date: Wed, 3 Mar 1999 22:08:17 -0800 (PST) From: Brian Beattie <beattie@aracnet.com> To: mi@aldan.algebra.com Cc: current@FreeBSD.ORG Subject: Re: aic0 and CAM Message-ID: <Pine.LNX.3.96.990303220428.3574A-100000@shell2.aracnet.com> In-Reply-To: <199903032112.QAA19717@misha.cisco.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 3 Mar 1999, Mikhail Teterin wrote: > > > It has pretty horrible performance problems on 2.2.8-STABLE, even when > > > using DMA: > > > > [...] > > > 4.4%Sys 91.9%Intr 3.7%User 0.0%Nice 0.0%Idl 4244 inact 204 pci irq9 > > > | | | | | | | | | | 5288 cache 105 aic0 irq11 > > > ==++++++++++++++++++++++++++++++++++++++++++++++ 248 free 104 wdc0 irq14 > > > [...] > > > I would not use this type of card for disk if I could avoid it, and > > probably not for tape either, I have a scanner I plan to use with it. > > Yes, tape is the only target on this bus here... From reading the 2.2.8's > driver's source I see the DMA is not "yet supported", although the card > supports it and lets me choose the drq. Still, the amount of CPU it takes > to process an interrupt (CyrixInstead 8x86 at over 200MHz) does not look > right.... > Unless it is a really slow tape drive, I'd get a better adapter. In any case, As I said much of the protocol must be implemented by the CPU, most of this happens in the interrupt routine. Brian Beattie | The only problem with beattie@aracnet.com | winning the rat race ... www.aracnet.com/~beattie | in the end you're still a rat To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.3.96.990303220428.3574A-100000>