On Sun, Jan 29, 2012 at 7:18 AM, Dmitry Torokhov
[off-list ref] wrote:

On Sun, Jan 29, 2012 at 06:53:50AM +0100, Javier Martinez Canillas wrote:

quoted

This is patch fixes two bugs in Dmitry's last cleanup patch.

1- The hardware tracking ids are stored in the ids array and the information for
   each contact is obtained calling cyttsp_get_tch() with an index. In the clean-up
   patch the value of the tracking id was used instead of the contact index.

Oops, thank you for fixing that.

quoted

2- i2c_set_clientdata() is called after the generic cyttsp_probe() function and
   this function calls cyttsp_power_on() that sends an ttsp command to the device
   and needs the client data before is set. The fix is to execute cyttsp_power_on
   inside the transport specific probe function (I2C, SPI) after the generic probe
   function is executed and the client data is set.

Not quite happy about this one, how about we pass cyttsp directly to bus
methods instead of relying on drvdata (as in the patch below)?

Thanks.

--
Dmitry

Hi Dmitry,

Yes, you are right that is a much more clever solution to the issue.
Thank you for suggesting that.

I just sent an v2 of the incremental patch to be applied on top of
your last clean-up.

This fixes the index bug stated before and the dev->i2c_client->cyttsp
reference being NULL issue changing the read/write methods signature
as you suggested.

It also fixes an unbalanced IRQ warning since soft_reset() now is
called inside the probe function and removes the wake_up platform
function in enable() as with the incremental v1.

Thanks a lot for your help and best regards,
Javier
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html