[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHp75VfOP0bnkWfAKPRpb5kBHj7ABeR1uhZwYdNFp25JUnO9Yg@mail.gmail.com>
Date: Wed, 19 Apr 2017 19:21:20 +0300
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Carlo Caione <carlo@...one.org>
Cc: "dvhart@...radead.org" <dvhart@...radead.org>,
Andy Shevchenko <andy@...radead.org>,
Platform Driver <platform-driver-x86@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
linux@...lessm.com, Carlo Caione <carlo@...lessm.com>
Subject: Re: [PATCH 1/2] hp-wmi: Fix error value for hp_wmi_tablet_state
On Sun, Apr 9, 2017 at 4:56 PM, Carlo Caione <carlo@...one.org> wrote:
> From: Carlo Caione <carlo@...lessm.com>
>
> hp_wmi_tablet_state() fails to return the correct error code when
> hp_wmi_perform_query() returns the HP WMI query specific error code
> that is a positive value.
> int ret = hp_wmi_perform_query(HPWMI_HARDWARE_QUERY, 0, &state,
> sizeof(state), sizeof(state));
> if (ret)
> - return ret;
> + return -EINVAL;
Shouldn't be something like
if (ret)
return ret < 0 ? ret : -EINVAL;
Looking into the code it looks like it may return all possible values:
0, negative, positive.
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists