[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <46C1F4F0.8010709@linpro.no>
Date: Tue, 14 Aug 2007 20:31:12 +0200
From: Tore Anderson <tore@...pro.no>
To: Matthew Wilcox <matthew@....cx>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
linux-scsi@...r.kernel.org, James.Smart@...lex.Com
Subject: Re: Linux 2.6.23-rc3
* Matthew Wilcox
> You didn't seem to include your .config, but I assume you're using
> async scanning.
I knew I forgot something. You're right about async scanning being
enabled (due to it being enabled in Ubuntu's 2.6.22 .config, on which
I based mine). I'm attaching it in case it's still interesting to you.
> Could you try including this patch:
>
> http://marc.info/?l=linux-scsi&m=118289275414202&w=2
I can confirm that this patch solves my problem without any side
effects (as far as I can tell).
When I unloaded and inserted the lpfc module in a loop with
2.6.23-rc3, I don't have to wait long for the -EEXIST error to appear
during insertion, and it was always followed by some form of badness
during the subsequent insertion (unable to handle kernel paging
request, unable to handle kernel NULL pointer dereference, general
protection fault). It never made it past five iterations, and I tried
it about ten times.
A kernel with your patch applied is doing much better, it's now on its
123rd iteration with no problems so far.
Regards
--
Tore Anderson
View attachment "config-2.6.23-rc3a" of type "text/plain" (74692 bytes)
Powered by blists - more mailing lists