linux-kernel - Re: 2.6.26.[1-3] + x61 tablet + x6ultrabase: no resume after undocking

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <48BBC8FC.1020809@dbservice.com>
Date:	Mon, 01 Sep 2008 12:50:36 +0200
From:	Tomas Carnecky <tom@...ervice.com>
To:	Zhang Rui <rui.zhang@...el.com>
CC:	Steven King <sfking@...dc.com>, linux-kernel@...r.kernel.org,
	linux-acpi@...r.kernel.org, ibm-acpi-devel@...ts.sourceforge.net
Subject: Re: 2.6.26.[1-3] + x61 tablet + x6ultrabase: no resume after	undocking

Zhang Rui wrote:
> On Sat, 2008-08-30 at 15:35 -0700, Steven King wrote:
>> While 2.6.26 made my x61t work with the x6ultrabase, I find that once I have 
>> docked the x61t, I can only resume after a suspend2ram if the x61t is docked.  
>> That is:
>>
>> Never have been docked: suspend & resume works fine.
>> Docked: suspend & resume works fine.
>> Docked, undocked, re-docked: suspend & resume works fine.
>>
>> Docked, then undocked: never finishes resuming (the disk seems to spin up and 
>> the sleep led goes off, but the system doesnt seem to get any further).
>>
>> Any suggestions?
>>
> I have an acpidump of a X61, and it may also send notifications in _STA
> method.
> Please please try the following patch and see if it helps.
> and it would be great if you can attach the acpidump of your laptop.

It has always been a mystery to me how the docking station works. It
never worked for me correctly, in some situations, depending on how I
docked/undocked the notebook, it would hang during resume. I brought the
problem up in linux-acpi after I bought my X61 and the X6 UltraBase, but
no patch resulted so I simply got used to shutting down my notebook
instead of putting it to sleep.
There is one question though that was never answered: Is there any
interaction needed from the userspace when docking/undocking? Something
that the acpi daemon would have to do in response to a dock/undock
event? I'm thinking about
 $ echo 1 > /sys/bus/scsi/devices/3\:0\:0\:0/rescan
or something like that. To tell the kernel that some devices may have
disappeared and that it shouldn't try to bring them up when resuming
from sleep. I would expect that ACPI would take care of that (it
certainly removes all USB devices and detaches ata4 (which is the bay in
the docking station).

Anyway, I tried your patch (on top of 2.6.26.3). If I start up the
notebook in the dock, take it out, put it to sleep, then it hangs during
resume (all I see on the screen is the cursor in the top left corner).

That behavior is consistent with my earlier experience, that is: If the
notebook was _ever_ running in the docking station, it won't resume
outside of it. No matter whether I start it outside the dock and then
put it into the dock, or start it up in the dock - once the notebook has
been docked, suspend/resume outside of the dock won't work anymore.
My wild uneducated guess is that the kernel doesn't remove/disable some
devices when undocking (the bay device?) and when it tries to resume
without them present it will hang.

Attached is my acpidump, I hope it helps.

tom

Download attachment "acpi-dump" of type "application/octet-stream" (63905 bytes)