Subject: Re: "pmap_unwire: wiring ... didn't change!"
To: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
From: Markus W Kilbinger <kilbi@rad.rwth-aachen.de>
List: port-mips
Date: 02/26/2005 22:05:16
>>>>> "Izumi" == Izumi Tsutsui <tsutsui@ceres.dti.ne.jp> writes:

    Izumi> I'm also using another patch as workaround for virtual
    Izumi> alias problem. Could you try this one?

I tried this patch on my qube2 in combination with your pci related
patch in

  http://mail-index.netbsd.org/port-cobalt/2004/11/08/0000.html

and Chuck Silvers' <chuq@chuq.com> pmap patch in

  http://mail-index.netbsd.org/port-cobalt/2005/02/12/0002.html

with no real change: The pure read activity seems quite reliable now
(absolutely no 'cmp' or 'sha1' failures within the last hours), but
stress test write activity (multiple 'cp file1 fileN' with large
files) still can and does produce data corruptions.

Markus.