Date: Wed, 21 Jan 1998 16:39:59 -0600 (CST) From: Doug Ledford <dledford@dialnet.net> To: Bernhard Rosenkraenzer <linux@bero-online.ml.org> Cc: linux-kernel@vger.rutgers.edu, aic7xxx@freebsd.org Subject: RE: [PATCH] Re: aic7xxx-5.0.1 Message-ID: <XFMail.980121164649.dledford@dialnet.net> In-Reply-To: <Pine.LNX.3.96.980121190238.876A-100000@ufp.in-trier.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On 21-Jan-98 Bernhard Rosenkraenzer wrote: >On Tue, 20 Jan 1998, Doug Ledford wrote: > >> The next version of my patch is now out. First, the prelims: >> >> ftp.dialnet.net:/pub/linux/aic7xxx/aic7xxx-5.0.1.tar.gz >> >> That's about it. Give it a shot and see if it breaks :) > >It is not compatible with recent Linux 2.1.x kernels... Well, I didn't expect it to be really :) At least in my original announcement about 5.0.0 I did remember to point out it was against 2.0.33 and not likely to work in 2.1.x :) >aic7xxx.c uses queue_task_irq_off, which hasn't been around for some time >(anyone remember exactly when it was removed?) Actually, this is one of the things that made me want to actually do a forward port of the code in the near future after the patch stabalizes under 2.0.x. Because of the changes in the 2.1.x scsi mid level code, this entire section of code that used the referenced function shouldn't need to exist any longer. Furthermore, there are now some "strange" <grinning widely> new interrupt semantics under 2.1.80 that I'm going to have to look into and think about. So, it may still require some changes to fundamental design issues when considering 2.1.x (including my plans for using the spin locks instead of cli()/sti() stuff like it does in 2.0.x) But, for my part anyway, that's still future speak :) ---------------------------------- E-Mail: Doug Ledford <dledford@dialnet.net> Date: 21-Jan-98 Time: 16:40:00 ----------------------------------
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.980121164649.dledford>