Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 05 Jul 2000 20:12:16 +0900
From:      haro@tk.kubota.co.jp (Munehiro Matsuda)
To:        mark@grondar.za
Cc:        current@FreeBSD.ORG
Subject:   Re: randomdev and APM. Not working together? 
Message-ID:  <20000705201216R.haro@tk.kubota.co.jp>
In-Reply-To: <20000705145916I.haro@tk.kubota.co.jp>
References:  <20000701104544S.haro@tk.kubota.co.jp> <200007010649.IAA13685@grimreaper.grondar.za> <20000705145916I.haro@tk.kubota.co.jp>

next in thread | previous in thread | raw e-mail | index | archive | help
From: haro@tk.kubota.co.jp (Munehiro Matsuda)
Date: Wed, 05 Jul 2000 14:59:16 +0900
::::> If I use randomdev for ssh, then APM seem to have stopped working.
::::
::::This is wierd; I am running this randomdev on my Libretto, (I wrote
::::it there), and it and APM are both fine.
::::
::::> Am I doing something wrong?
::::
::::No - you are doing everything right.
::::
::::Can you do any debugging on this? Are you able perhaps to trace the
::::kernel and see where it fails?
::
::Ok. I'll see what I can do.
::
::BTW, I've taken a diff of dmesg output between randomdev loaded and
::unloaded. As you can see, (1)outcome of "Calibrating clock" differs
::and (2)device attaching order is some what different. 
::
::I'll let you know, if I can find anthing.

  [This mail is Bcc'ed to acpi-jp@jp.FreeBSD.org]

I think I found the problem, and being able to work around it. 
But, I don't know how to realy fix it.

Here's summery of what was happening...

1. APM fails if randomdev_load="YES" in loader.conf and ACPI configured.
 Some drivers' probe & attach order will change, if randomdev module
was loaded using loader.conf.
 I have ACPI test driver (as of early May) and APM driver in my kernel,
that was affected by this change.

 1)Randomdev NOT Loaded: Probe & attach will happen in the order -
   (1)nxp0, (2)apm0, (3)acpi0 .....That's OK

 2)Randomdev Loaded: Probe & attach will happen in the order -
   (1)acpi0, (2)apm0, (3)nxp0 .....NG

   Because acpi0 access the BIOS first, APM support in BIOS seems to
   get turned off. :-(

If there's any way to statically set the probe & attach order, I would
like to know.

2. Work around was to include RANDMDEV in config file.
 I disabled it, because it was panicing on shutdown. 
 That being fixed in rev1.5 of randmdev.c, I have a working system, again.

Thanks you,
  Haro
=------------------------------------------------------------------------------
           _ _    Munehiro (haro) Matsuda
 -|- /_\  |_|_|   Business Incubation Dept., Kubota Corp.
 /|\ |_|  |_|_|   1-3 Nihonbashi-Muromachi 3-Chome
                  Chuo-ku Tokyo 103-8310, Japan
                  Tel: +81-3-3245-3318  Fax: +81-3-3245-3315
                  Email: haro@kubota.co.jp


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?20000705201216R.haro>