From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dan Carpenter Subject: Re: [PATCH] platform/x86: huawei-wmi: make validation stricter in huawei_wmi_battery_set() Date: Tue, 15 Oct 2019 15:46:23 +0300 Message-ID: <20191015124623.GF21344@kadam> References: <20191015083837.GA29104@mwanda> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Ayman Bagabas Cc: Darren Hart , Andy Shevchenko , Takashi Iwai , Mattias Jacobsson <2pi@mok.nu>, platform-driver-x86@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-janitors@vger.kernel.org List-Id: platform-driver-x86.vger.kernel.org On Tue, Oct 15, 2019 at 08:21:59AM -0400, Ayman Bagabas wrote: > Hi Dan > > On Tue, Oct 15, 2019, 4:39 AM Dan Carpenter > wrote: > > > I don't think it makes sense for "end" to be negative or for even for it > > to be less than "start". That also means that "start" can't be more > > than 100 which is good. > > > > While this makes sense, you run into issues where you cannot set "start" > before "end" and vice versa. > > Take this scenario, you have start=70 and end=90, now you want to set these > to start=40 and end=60, you would have to set "start" first before you can > change the value of "end" otherwise you will run into EINVAL. Now imagine > you wanna go the opposite direction, you would have to set "end" before you > can change "start". > I think having a little wiggle room is fine for such scenarios. > I haven't tested this code... What you're describing sounds really very weird to me, but I will accept that you know more about your use cases than I do. My other concern is that right now you can set start > 100 or end < 0. regards, dan carpenter