3.0-rc2: mei breaks suspend

3.0-rc2: mei breaks suspend

[Stefan Seyfried]

Hi,

having the "mei" driver loaded, which happened automatically on my
ThinkPad X200s, successfully prevents any suspend efforts:

[39736.728116] pci_pm_suspend(): mei_pci_suspend+0x0/0xc0 [mei] returns 9999

The code has not changed since 3.0-rc2, so I did not try compiling latest
git.

Jiri Slaby hinted that this code in drivers/staging/mei/wd.c is to blame:

169         ret = wait_event_interruptible_timeout(dev->wait_stop_wd,
170                                         dev->wd_stopped, 10 * HZ);

I am, however, not familiar enough with timers and stuff to immediately
spot the problem.

Removing the module before suspend is a workaround.

This could be considered a regression, as suspend worked very well before
this driver was introduced :-)

Best regards,

	Stefan
-- 
Stefan Seyfried

Linux Consultant & Developer
B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537

Re: 3.0-rc2: mei breaks suspend

[Maarten Lankhorst]

Hi Stefan,

2011/6/16 Stefan Seyfried <stefan.seyfried@googlemail.com>:
> Hi,
>
> having the "mei" driver loaded, which happened automatically on my
> ThinkPad X200s, successfully prevents any suspend efforts:
>
> [39736.728116] pci_pm_suspend(): mei_pci_suspend+0x0/0xc0 [mei] returns 9999
>
> The code has not changed since 3.0-rc2, so I did not try compiling latest
> git.
>
> Jiri Slaby hinted that this code in drivers/staging/mei/wd.c is to blame:
>
> 169         ret = wait_event_interruptible_timeout(dev->wait_stop_wd,
> 170                                         dev->wd_stopped, 10 * HZ);
>
> I am, however, not familiar enough with timers and stuff to immediately
> spot the problem.
>
> Removing the module before suspend is a workaround.
>
> This could be considered a regression, as suspend worked very well before
> this driver was introduced :-)
Does changing that to wait_event_timeout work? My guess is that is
what he meant.

~Maarten

Re: 3.0-rc2: mei breaks suspend

[Maarten Lankhorst]

2011/6/16 Maarten Lankhorst <m.b.lankhorst@gmail.com>:
> Hi Stefan,
>
> 2011/6/16 Stefan Seyfried <stefan.seyfried@googlemail.com>:
>> Hi,
>>
>> having the "mei" driver loaded, which happened automatically on my
>> ThinkPad X200s, successfully prevents any suspend efforts:
>>
>> [39736.728116] pci_pm_suspend(): mei_pci_suspend+0x0/0xc0 [mei] returns 9999
>>
>> The code has not changed since 3.0-rc2, so I did not try compiling latest
>> git.
>>
>> Jiri Slaby hinted that this code in drivers/staging/mei/wd.c is to blame:
>>
>> 169         ret = wait_event_interruptible_timeout(dev->wait_stop_wd,
>> 170                                         dev->wd_stopped, 10 * HZ);
>>
>> I am, however, not familiar enough with timers and stuff to immediately
>> spot the problem.
>>
>> Removing the module before suspend is a workaround.
>>
>> This could be considered a regression, as suspend worked very well before
>> this driver was introduced :-)
Nm, stupid suggestion.

Adding the author to cc, maybe he knows.

Re: 3.0-rc2: mei breaks suspend

[Arkadiusz Miskiewicz]

On Thursday 16 of June 2011, Stefan Seyfried wrote:
> Hi,
> 
> having the "mei" driver loaded, which happened automatically on my
> ThinkPad X200s, successfully prevents any suspend efforts:
> 
> [39736.728116] pci_pm_suspend(): mei_pci_suspend+0x0/0xc0 [mei] returns
> 9999

https://lkml.org/lkml/2011/6/13/148

-- 
Arkadiusz Miśkiewicz        PLD/Linux Team
arekm / maven.pl            http://ftp.pld-linux.org/

Re: 3.0-rc2: mei breaks suspend

[Stefan Seyfried]

Am Thu, 16 Jun 2011 12:24:32 +0200
schrieb Arkadiusz Miskiewicz <a.miskiewicz@gmail.com>:

> On Thursday 16 of June 2011, Stefan Seyfried wrote:
> > [39736.728116] pci_pm_suspend(): mei_pci_suspend+0x0/0xc0 [mei] returns
> > 9999
> 
> https://lkml.org/lkml/2011/6/13/148

Thanks. I should start really reading all LKML mail again ;-P

I pinged Greg again, so that we'll get this for 3.0
-- 
Stefan Seyfried

Linux Consultant & Developer
B1 Systems GmbH
Osterfeldstraße 7 / 85088 Vohburg / http://www.b1-systems.de
GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt,HRB 3537