linux-kernel - Re: [PATCH v4 2/2] media: Add a driver for the ov7251 camera sensor

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 26 Apr 2018 10:16:56 +0300
From:   Sakari Ailus <sakari.ailus@....fi>
To:     Todor Tomov <todor.tomov@...aro.org>
Cc:     mchehab@...nel.org, hverkuil@...all.nl,
        laurent.pinchart@...asonboard.com, linux-media@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 2/2] media: Add a driver for the ov7251 camera sensor

On Thu, Apr 26, 2018 at 10:04:25AM +0300, Todor Tomov wrote:
> Hi Sakari,
> 
> On 26.04.2018 09:50, Sakari Ailus wrote:
> > Hi Todor,
> > 
> > On Wed, Apr 25, 2018 at 07:20:46PM +0300, Todor Tomov wrote:
> > ...
> >> +static int ov7251_write_reg(struct ov7251 *ov7251, u16 reg, u8 val)
> >> +{
> >> +	u8 regbuf[3];
> >> +	int ret;
> >> +
> >> +	regbuf[0] = reg >> 8;
> >> +	regbuf[1] = reg & 0xff;
> >> +	regbuf[2] = val;
> >> +
> >> +	ret = i2c_master_send(ov7251->i2c_client, regbuf, 3);
> >> +	if (ret < 0) {
> >> +		dev_err(ov7251->dev, "%s: write reg error %d: reg=%x, val=%x\n",
> >> +			__func__, ret, reg, val);
> >> +		return ret;
> >> +	}
> >> +
> >> +	return 0;
> > 
> > How about:
> > 
> > 	return ov7251_write_seq_regs(ov7251, reg, &val, 1);
> > 
> > And put the function below ov2751_write_seq_regs().
> 
> I'm not sure... It will calculate message length each time and then check
> that it is not greater than 5, which it is. Seems redundant.
> 
> > 
> >> +}
> >> +
> >> +static int ov7251_write_seq_regs(struct ov7251 *ov7251, u16 reg, u8 *val,
> >> +				 u8 num)
> >> +{
> >> +	const u8 maxregbuf = 5;
> >> +	u8 regbuf[maxregbuf];
> >> +	u8 nregbuf = sizeof(reg) + num * sizeof(*val);
> >> +	int ret = 0;
> >> +
> >> +	if (nregbuf > maxregbuf)
> >> +		return -EINVAL;
> >> +
> >> +	regbuf[0] = reg >> 8;
> >> +	regbuf[1] = reg & 0xff;
> >> +
> >> +	memcpy(regbuf + 2, val, num);
> >> +
> >> +	ret = i2c_master_send(ov7251->i2c_client, regbuf, nregbuf);
> >> +	if (ret < 0) {
> >> +		dev_err(ov7251->dev, "%s: write seq regs error %d: first reg=%x\n",
> > 
> > This line is over 80... 
> 
> Yes indeed. Somehow checkpatch does not report this line, I don't know why.
> 
> > 
> > If you're happy with these, I can make the changes, too; they're trivial.
> 
> Only the second one? Thanks :)

Works for me. I'd still think the overhead of managing the buffer is
irrelevant where to having an extra function to do essentially the same
thing is a source of maintenance and review work. Note that we're even now
spending time to discuss it. ;-)

-- 
Kind regards,

Sakari Ailus
e-mail: sakari.ailus@....fi