Date: Tue, 28 Apr 1998 00:24:44 -0500 (CDT) From: Mail Receiver <rmail@ittc.ukans.edu> To: aic7xxx@FreeBSD.ORG Subject: Re: Duration of Blocked Interrupts Message-ID: <Pine.LNX.3.95.980428001618.5895A-100000@hegel.ittc.ukans.edu> In-Reply-To: <354549A5.3F52E18F@dialnet.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 27 Apr 1998, Doug Ledford wrote: > What version of the aic7xxx driver were you testing? I wouldn't be suprised > if the latest versions can do exactly this. There are things that can be > done to get around this problem in 2.0.x, but there won't be much we can do > about it at all in 2.1.x and probably 2.2.x due to the spin lock semantics > and requirements. I would have to consider the consequences of changing > things around in the 2.1.x driver before I would recommend the changes I'm > thinking of. Sorry about not mentioning that in the first message... These particular tests were done using the driver that came with 2.1.75. Would it be worth instrumenting a 2.0.x (say 33, for instance) kernel and checking the duration of blocking there? We're not tied to any particular kernel version (although somewhat recent kernels would be preferable...) For our particular purposes, we also need access to ATM drivers but those can be found in both the recent 2.1.x as well as the 2.0.x kernels. Part of our system gives us access to much finer resolution timers as well, so were not bound to just doing things on jiffy boundaries (we can schedule events with effectively microsecond accuracy). We'd be willing to add a kernel thread using polling or something if that might help for our particular purposes. Thanks, Robert To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe aic7xxx" 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.3.95.980428001618.5895A-100000>