Cause of Suspend2 resume failures on Toshiba laptops

Cause of Suspend2 resume failures on Toshiba laptops

[Bernard Blackham]

[-- Attachment #1: Type: text/plain, Size: 1508 bytes --]

Hi all,

As some of you will know, some laptops (notably almost all recent
Toshibas) have issues resuming with Software Suspend 2, resulting in
a hang during "Reading caches", or a stream of "Unable to queue
event ..." messages. If you try and resume enough times,
occasionally it will make it to the end of resuming and the system
will be perfectly usable.

The attached patch more or less reverts the change in 2.6.9-rc1 that
began this all. If you have a Toshiba or other laptop that's
affected by this, I ask you to see if it helps your problems. (It
hacks around the problem rather than fixing it, but I'm not entirely
sure what functionality it disables).

To the ACPI developers:

I'm not entirely sure what this is breaking. What uses these GPEs?
It seems the hangs are caused by one of these GPEs firing upon
resume (though I'm not sure why it causes a hang - perhaps related
to the low memory conditions during a suspend/resume cycle?)

The comment above the line I changed mentions that the GPE enable
mask is updated, but not the HW GPE itself. Whereabouts are these
later enabled?

Are the HW GPE enable masks stored in memory that might be replaced
when resuming, with their values from suspend? If so, could/should
GPEs be disabled before suspending so that they aren't enabled at an
unfortunate moment on resume?

Apologies if none of the above makes much sense - it doesn't make
all that much sense to me yet :)

Thanks,

Bernard.

-- 
 Bernard Blackham <bernard at blackham dot com dot au>

[-- Attachment #2: toshiba-acpi-hack.diff --]
[-- Type: text/plain, Size: 433 bytes --]

--- linux-2.6.9-rc1/drivers/acpi/events/evgpeblk.c.orig	2005-03-05 01:24:29.000000000 +0800
+++ linux-2.6.9-rc1/drivers/acpi/events/evgpeblk.c	2005-03-05 01:25:28.000000000 +0800
@@ -314,7 +314,7 @@
 
 	/* Update enable mask, but don't enable the HW GPE as of yet */
 
-	status = acpi_ev_enable_gpe (gpe_event_info, FALSE);
+	status = AE_OK;
 
 	ACPI_DEBUG_PRINT ((ACPI_DB_LOAD,
 		"Registered GPE method %s as GPE number 0x%.2X\n",

Re: Cause of Suspend2 resume failures on Toshiba laptops

[Dumitru Ciobarcianu]

În data de Du, 06-03-2005 la 17:46 +0100, mhf@berlios.de a scris:
> I am afraid the only effect of the patch on my Toshiba L2
> is that the RTC stops working, thus the rtc.ko module can
> not be inserted (no such device).

Please revert the first patch and use the second and the 3rd.
toshiba-acpi-hack-2.diff and toshiba-acpi-hack-3.diff. 
(if you already did that i'll just shut up and sit in a corner :)

-- 
Cioby

Re: Cause of Suspend2 resume failures on Toshiba laptops

[mhf]

On Sunday 06 March 2005 18:34, Dumitru Ciobarcianu wrote:
> =C3=8En data de Du, 06-03-2005 la 17:46 +0100, mhf@berlios.de a=20
scris:
> > I am afraid the only effect of the patch on my Toshiba
> > L2 is that the RTC stops working, thus the rtc.ko
> > module can not be inserted (no such device).
>
> Please revert the first patch and use the second and the
> 3rd. toshiba-acpi-hack-2.diff and
> toshiba-acpi-hack-3.diff. (if you already did that i'll
> just shut up and sit in a corner :)

With 2.6.11 and swsusp-2.1.8:

I have tried a patch Bernard sent me and which improves
the situation considerably. I am now able to resume in
my test setup which is root on tempfs linked to suspended=20
reiser partition. However running of the same reiser=20
partition directly (the normal way) does hang after=20
copying kernel back, but now _without_ the GPE message.=20
Suppose it is a driver issue.  I'll figure it out and=20
report back=20

Am optimistic that the GPE issue is actually fixed :-)

The RTC problem (in the test setup) was related to a messed
up/unhandled udev installation and was _not_  due to the=20
first version of the patch. RTC OK with devfs and after
cleanup of dev directory. Got to work on testing with
udev...

OT: does 2.6 toshiba ACPI driver still require a patch to=20
work with toshset-1.66 (I had also to patch 2.4.29 so it
looks like it)

	Regards
	Michael

Problems with Swsusp2

[Timo Vuorela]

Hi,

I have succesfully been able to suspend but sometimes something goes terribly 
wrong: when I boot my it runs some checks on my harddisk and doesn't boot, 
says I have to run reiserfschk --rebuild-tree to correct errors. What I'm 
doing wrong? Has it anything to do with booting sometimes with noresume2 
option if I don't want to continue my resumed session?

Thanks for help,

-- 
Timo Vuorela
timo.a.vuorela-A0cwMDoKKKE@public.gmane.org


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: [ACPI] Problems with Swsusp2

[Nigel Cunningham]

Hi Timo.

The correctly list for Suspend2 is suspend2-users@lists.suspend2.net.
You might like to subscribe and remove the ACPI list from any reply.

On Mon, 2005-03-07 at 17:04, Timo Vuorela wrote: 
> Hi,
> 
> I have succesfully been able to suspend but sometimes something goes terribly 
> wrong: when I boot my it runs some checks on my harddisk and doesn't boot, 
> says I have to run reiserfschk --rebuild-tree to correct errors. What I'm 
> doing wrong? Has it anything to do with booting sometimes with noresume2 
> option if I don't want to continue my resumed session?

This may happen if you suspend, then boot with another kernel, and
resume with the original one. Unfortunately there's no way for us to
warn you when this happens - we just have to wait for more distributions
to include suspend-to-disk support.

Regards,

Nigel
-- 
Nigel Cunningham
Software Engineer, Canberra, Australia
http://www.cyclades.com
Bus: +61 (2) 6291 9554; Hme: +61 (2) 6292 8028;  Mob: +61 (417) 100 574

Maintainer of Suspend2 Kernel Patches http://softwaresuspend.berlios.de