Re: [PATCH] platform/chrome: Fix locking pattern in wilco_ec_mailbox()

[Enric Balletbo i Serra]

Hi Nick,

On 11/3/19 16:58, Nick Crews wrote:
> Before, ec->data_buffer could be written to from multiple
> contexts at the same time. Since the ec is shared data,
> it needs to be inside the mutex as well.
> 

Probably you're missing a fixes tag here.

> Signed-off-by: Nick Crews <ncrews@xxxxxxxxxxxx>
> ---
>  drivers/platform/chrome/wilco_ec/mailbox.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/platform/chrome/wilco_ec/mailbox.c b/drivers/platform/chrome/wilco_ec/mailbox.c
> index f6ff29a11f1a..d6cb96168285 100644
> --- a/drivers/platform/chrome/wilco_ec/mailbox.c
> +++ b/drivers/platform/chrome/wilco_ec/mailbox.c
> @@ -223,12 +223,12 @@ int wilco_ec_mailbox(struct wilco_ec_device *ec, struct wilco_ec_message *msg)
>  		msg->command, msg->type, msg->flags, msg->response_size,
>  		msg->request_size);
>  
> -	/* Prepare request packet */

There is a reason to remove this comment?

> +	mutex_lock(&ec->mailbox_lock);
> +
>  	rq = ec->data_buffer;
>  	wilco_ec_prepare(msg, rq);
> -
> -	mutex_lock(&ec->mailbox_lock);
>  	ret = wilco_ec_transfer(ec, msg, rq);
> +	

Trailing whitespaces here

>  	mutex_unlock(&ec->mailbox_lock);
>  
>  	return ret;
> 

You don't need to resend, I'll add the fixes tag, fix the trailing whitespaces
and queue as a fix for 5.1 once the merge window closes.

Thanks,
 Enric