linux-kernel - Re: [PATCH v4 04/11] leds: aw200xx: calculate dts property display

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]

Message-ID: <20231124094146.qsgmmbwulemjikpg@CAB-WSD-L081021>
Date:   Fri, 24 Nov 2023 12:41:46 +0300
From:   Dmitry Rokosov <ddrokosov@...utedevices.com>
To:     Lee Jones <lee@...nel.org>
CC:     <pavel@....cz>, <robh+dt@...nel.org>,
        <krzysztof.kozlowski+dt@...aro.org>, <conor+dt@...nel.org>,
        <andy.shevchenko@...il.com>, <kernel@...rdevices.ru>,
        <rockosov@...il.com>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <linux-leds@...r.kernel.org>,
        George Stark <gnstark@...utedevices.com>
Subject: Re: [PATCH v4 04/11] leds: aw200xx: calculate dts property
 display_rows in the driver

On Thu, Nov 23, 2023 at 04:32:52PM +0000, Lee Jones wrote:
> On Tue, 21 Nov 2023, Dmitry Rokosov wrote:
> 
> > From: George Stark <gnstark@...utedevices.com>
> > 
> > Get rid of device tree property "awinic,display-rows". The property
> > value actually means number of current switches and depends on how leds
> 
> Nit: LEDs
> 
> > are connected to the device. It should be calculated manually by max
> > used led number. In the same way it is computed automatically now.
> 
> As above - I won't mention this again.
> 
> > Max used led is taken from led definition subnodes.
> > 
> > Signed-off-by: George Stark <gnstark@...utedevices.com>
> > Signed-off-by: Dmitry Rokosov <ddrokosov@...utedevices.com>
> > ---
> >  drivers/leds/leds-aw200xx.c | 39 +++++++++++++++++++++++++------------
> >  1 file changed, 27 insertions(+), 12 deletions(-)
> > 
> > diff --git a/drivers/leds/leds-aw200xx.c b/drivers/leds/leds-aw200xx.c
> > index 7762b3a132ac..4bce5e7381c0 100644
> > --- a/drivers/leds/leds-aw200xx.c
> > +++ b/drivers/leds/leds-aw200xx.c
> > @@ -379,6 +379,30 @@ static void aw200xx_disable(const struct aw200xx *const chip)
> >  	return gpiod_set_value_cansleep(chip->hwen, 0);
> >  }
> >  
> > +static bool aw200xx_probe_get_display_rows(struct device *dev, struct aw200xx *chip)
> > +{
> > +	struct fwnode_handle *child;
> > +	u32 max_source = 0;
> > +
> > +	device_for_each_child_node(dev, child) {
> > +		u32 source;
> > +		int ret;
> > +
> > +		ret = fwnode_property_read_u32(child, "reg", &source);
> > +		if (ret || source >= chip->cdef->channels)
> 
> Shouldn't the second clause fail instantly?
> 

We already have such logic in the aw200xx_probe_fw() function, which
skips the LED node with the wrong reg value too. Furthermore, we have
strict reg constraints in the dt-bindings parts (in the current patch
series), so we assume that the DT developer will not create an LED with
the wrong reg value.

> > +			continue;
> > +
> > +		max_source = max(max_source, source);
> > +	}
> > +
> > +	if (!max_source)
> 
> Since max_source is an integer, please use an '== 0' comparison.
> 

Okay

> > +		return false;
> > +
> > +	chip->display_rows = max_source / chip->cdef->display_size_columns + 1;
> > +
> > +	return true;
> > +}
> > +
> >  static int aw200xx_probe_fw(struct device *dev, struct aw200xx *chip)
> >  {
> >  	struct fwnode_handle *child;
> > @@ -386,18 +410,9 @@ static int aw200xx_probe_fw(struct device *dev, struct aw200xx *chip)
> >  	int ret;
> >  	int i;
> >  
> > -	ret = device_property_read_u32(dev, "awinic,display-rows",
> > -				       &chip->display_rows);
> > -	if (ret)
> > -		return dev_err_probe(dev, ret,
> > -				     "Failed to read 'display-rows' property\n");
> > -
> > -	if (!chip->display_rows ||
> > -	    chip->display_rows > chip->cdef->display_size_rows_max) {
> > -		return dev_err_probe(dev, ret,
> > -				     "Invalid leds display size %u\n",
> > -				     chip->display_rows);
> > -	}
> > +	if (!aw200xx_probe_get_display_rows(dev, chip))
> 
> Function calls in side if() statements in general is rough.
> 
> Please break it out and use 'ret' as we usually do.
> 
> > +		return dev_err_probe(dev, -EINVAL,
> 
> Make this the return value from aw200xx_probe_get_display_rows() and use
> 'ret' instead.
> 

No problem, I'll prepare a new version.

> > +				     "No valid led definitions found\n");
> >  
> >  	current_max = aw200xx_imax_from_global(chip, AW200XX_IMAX_MAX_uA);
> >  	current_min = aw200xx_imax_from_global(chip, AW200XX_IMAX_MIN_uA);
> > -- 
> > 2.36.0
> > 
> 
> -- 
> Lee Jones [李琼斯]

-- 
Thank you,
Dmitry