Re: [PATCH v2 04/10] mfd: cros_ec: Use a zero-length array for command data

[Javier Martinez Canillas <javier.martinez@collabora.co.uk>]

Hello Heiko and Alexandru,

On 05/11/2015 11:47 PM, Heiko Stuebner wrote:
>> Heiko, Alexandru,
>> 
>> I don't have access to my Chromebooks right now but could you please test
>> the following patch [0] to see if that is the problem?
>> 
>> It seems I got it working on my test just out of luck since it happened that
>> version was initialized to 0 due the data it was on the stack.
>> 
>> Tomorrow when I've access to the Chromebooks I'll do more extensive testing
>> and see if I can reproduce the issue and if my assumption is correct.
>> 
>> Best regards,
>> Javier
>> 
>> [0]
>> diff --git a/drivers/input/keyboard/cros_ec_keyb.c
>> b/drivers/input/keyboard/cros_ec_keyb.c index ef3ba20f4560..4f233e248a4d
>> 100644
>> --- a/drivers/input/keyboard/cros_ec_keyb.c
>> +++ b/drivers/input/keyboard/cros_ec_keyb.c
>> @@ -151,6 +151,7 @@ static int cros_ec_keyb_get_state(struct cros_ec_keyb
>> *ckdev, uint8_t *kb_state) int ret;
>>  	struct cros_ec_command *msg = (struct cros_ec_command *)kb_state;
>> 
>> +	msg->version = 0;
>>  	msg->command = EC_CMD_MKBP_STATE;
>>  	msg->insize = ckdev->cols;
>>  	msg->outsize = 0;
> 
> on my rk3288-pinky this seems to fix the issue ... keyboard works out of the 
> box again.
>

Thank you both for testing the fix and again sorry for the silly mistake...

I'll wait a couple of days to see if I get more feedback and re-spin a v3
of the series with this fix patch squashed.
 
> 
> Heiko
>

Best regards,
Javier