linux-kernel - Re: [PATCH 2/3] Input: of_touchscreen - fix setting max values on X/Y axis

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: <20150708150841.GA40116@dtor-ws>
Date:	Wed, 8 Jul 2015 08:08:41 -0700
From:	Dmitry Torokhov <dmitry.torokhov@...il.com>
To:	Roger Quadros <rogerq@...com>
Cc:	linux-input@...r.kernel.org,
	Maxime Ripard <maxime.ripard@...e-electrons.com>,
	Sebastian Reichel <sre@...nel.org>,
	Pavel Machek <pavel@....cz>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3] Input: of_touchscreen - fix setting max values on
 X/Y axis

On Wed, Jul 08, 2015 at 10:59:04AM +0300, Roger Quadros wrote:
> Dmitry,
> 
> On 07/07/15 19:25, Dmitry Torokhov wrote:
> >Hi Roger,
> >
> >On Tue, Jul 07, 2015 at 12:37:31PM +0300, Roger Quadros wrote:
> >>Hi Dmitry,
> >>
> >>On 07/07/15 03:27, Dmitry Torokhov wrote:
> >>>The binding specification says that "touchscreen-size-x" and "-y" specify
> >>>horizontal and vertical resolution of the touchscreen and therefore maximum
> >>>absolute coordinates should be reduced by 1 since we are starting with 0.
> >>>
> >>>Signed-off-by: Dmitry Torokhov <dmitry.torokhov@...il.com>
> >>>---
> >>>  drivers/input/touchscreen/of_touchscreen.c | 10 ++++++----
> >>>  1 file changed, 6 insertions(+), 4 deletions(-)
> >>>
> >>>diff --git a/drivers/input/touchscreen/of_touchscreen.c b/drivers/input/touchscreen/of_touchscreen.c
> >>>index 759cf4b..50bc0f2 100644
> >>>--- a/drivers/input/touchscreen/of_touchscreen.c
> >>>+++ b/drivers/input/touchscreen/of_touchscreen.c
> >>>@@ -71,23 +71,25 @@ void touchscreen_parse_of_params(struct input_dev *dev, bool multitouch)
> >>>
> >>>  	axis = multitouch ? ABS_MT_POSITION_X : ABS_X;
> >>>  	data_present = touchscreen_get_prop_u32(np, "touchscreen-size-x",
> >>>-						input_abs_get_max(dev, axis),
> >>>+						input_abs_get_max(dev,
> >>>+								  axis) + 1,
> >>
> >>Why do we need to pass default_value to touchscreen_get_prop_u32()?
> >>If the property doesn't exist we are not updating the parameter
> >>anyway right?
> >
> >The binding can specify max, fuzz, both, or neither. If only one is
> >specified we do not want to "undo" whatever the driver did (for example
> 
> why not? At least that is not what touchscreen_parse_properties() does.
> It will update the touchscreen params if any one of the DT property
> is present.

Yes, but while keeping the old properties (max, fuzz) intact if they
have not been specified in dts. That is why we are passing default value
to touchscreen_get_prop_u32.

> 
> >tsc2005 sets up the default maximums before trying to parse OF), so we
> >fetch the current value and pass it on as default one.
> 
> This is what would happen in tsc2005 case
> 
> -driver first sets ABS_X_max to 0xfff and ABS_Y_max to 0xfff.
> -calls touchscreen_parse_properties()
> -which calls touchscreen_get_prop_u32(dev, "touchscreen-size-x",
> 0xfff+1, &maximum)
> -if DT properties are missing, touchscreen_get_prop_u32() changes
> ABS_X_max to 0x1000 and returns false.
> - as data_present is false we don't do a -1 and so we have ABS_X_max
> changed to 0x1000.

If data_present is false we do not call input_set_abs_params() at all
ans so nothing changes in input device settings.  Now, if fuzz would be
set up (but size was not) we would be calling input_set_abs_params(),
but given that we subtract 1 from maximum returned by
touchscreen_get_prop_u32() we'd be setting input_set_abs_params(input,
ABS_X, 0, 0x1000 - 1, fuzz) thus preserving 0xfff value set by the
driver.

> 
> This doesn't look right.
> 
> So IMO default_value parameter can be removed from
> touchscreen_get_prop_u32() to make things simpler.
> 
> If the property doesn't exist it will not change *value and return false.

But it will change if one of the properties set. For example, if you set
fuzz but not size.

Thanks.

-- 
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/