lkml.org 
[lkml]   [2011]   [Dec]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH] usb: musb: fix pm_runtime mismatch
From
On Fri, Dec 16, 2011 at 1:01 AM, Felipe Balbi <balbi@ti.com> wrote:
> On Fri, Dec 16, 2011 at 12:42:14AM +0200, Felipe Contreras wrote:
>> In musb_init_controller() there's a pm_runtime_put(), but there's no
>> pm_runtime_get(), which creates a mismatch that causes the driver to
>> sleep when it shouldn't.
>>
>> This was introduced in 7acc619, but it wasn't triggered until 18a2689
>> was merged to Linus' branch at point 6899608.
>
> you need to add the commit description (whatever was the mail's subject)
> here too. And you should put in Cc the author or those commits too,
> otherwise we can't poke into their brains to understand what they were
> thinking when they originally wrote those patches.

True, but code is code, and even if you don't know what he was
thinking, it's clear there's something wrong.

>> However, it seems most of the time this is used in a way that keeps the
>> counter above 0, so nobody noticed. Also, it seems to depend on the
>> configuration used.
>>
>> I found the problem by loading isp1704_charger before any usb gadgets:
>> http://article.gmane.org/gmane.linux.kernel/1226122
>>
>> All versions after 2.6.39 are affected.
>>
>> Cc: stable@vger.kernel.org
>> Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
>> ---
>>  drivers/usb/musb/musb_core.c |    2 --
>>  1 files changed, 0 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c
>> index b63ab15..920f04e 100644
>> --- a/drivers/usb/musb/musb_core.c
>> +++ b/drivers/usb/musb/musb_core.c
>> @@ -2012,8 +2012,6 @@ musb_init_controller(struct device *dev, int nIrq, void __iomem *ctrl)
>>       if (status < 0)
>>               goto fail3;
>>
>> -     pm_runtime_put(musb->controller);
>
> To me the real fix would be add the missing pm_runtime_get_sync(). On
> probe() we're actually accessing MUSB's address space which needs it's
> clocks turned on. I guess it's only working now by chance, probably
> because glue layer calls pm_runtime_get_sync() to access it's own
> address space and that uses the same clocks.

Are you sure it's "musb-hdrc", and not "musb-omap2430" the one
accessing the relevant address-space? From the runtime_pm
documentation it looks like only the probe function should deal with
this.

If "musb-hdrc" was truly accessing these registers, then I would get
the same failure because the clocks are turned off, but I don't...

Cheers.

--
Felipe Contreras
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2011-12-16 00:33    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site