The thing is that i do not know what to blame. 
I consider this change necessary to cope with misbehaving hardware.

The hardware affected by this is the ones that
1: Does not interrupt on state change.
2: Reports state change late.

Because there is no way of knowing if the hardware is fawlty then the polling 
cannot be disabled.
I think that there might be more advanced ways to determine if the polling can 
be disabled or not,
one might imagine that if there has been any interrupt from the battery at all 
then disabling the polling 
would be ok. Also one migth consider some other means to know if the AC is 
connected or not but as
i stated in a previous email we are within a battery context function and it 
has no knowledge of the AC supply
at all. 

I have tried to find any information on the battery drivers to see if they are 
responsible for not reporting 
discharging fast enough but i do not know where to look to be honest. 

This bug has beeen present since 2009 and affects acer laptops and some other 
models. I think it depends on 
how much the hardware filters the charge current it reports.

Mikael


> Date: Mon, 11 Jul 2011 09:15:26 +0100
> Subject: Re: [PATCH] Patch to solve issues with some hardware that causes     
> battery to never update
> From: hughsi...@gmail.com
> To: j_m_hj...@hotmail.com
> CC: devkit-devel@lists.freedesktop.org
> 
> On 10 July 2011 21:34, Mikael Hjelm <j_m_hj...@hotmail.com> wrote:
> > -          not if we've guessed the state to be fully charged */
> > -       supply->priv->enable_poll = (state !=
> > UP_DEVICE_STATE_FULLY_CHARGED);
> > +          not if we've guessed the state to be fully charged. This
> > function
> > +          has been disabled since some battery hardware/drivers are very
> > +          late to report discharging and does not interrupt on state change
> > +       */
> > +       /*supply->priv->enable_poll = (state !=
> > UP_DEVICE_STATE_FULLY_CHARGED);
> > +        */
> > +
> 
> You've just commented out the function. Surely finding the bug would
> be the right thing to do?
> 
> Richard.
> _______________________________________________
> devkit-devel mailing list
> devkit-devel@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/devkit-devel
                                          
_______________________________________________
devkit-devel mailing list
devkit-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/devkit-devel

Reply via email to