linux-kernel - Re: [PATCH 03/10] Input: synaptics-rmi4 - f30/f03: Forward mechanical buttons on buttonpads to PS/2 guest

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 for Android: free password hash cracker in your pocket

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

Date:   Tue,  7 Feb 2017 17:25:37 +0100
From:   Benjamin Tissoires <benjamin.tissoires@...hat.com>
To:     Dmitry Torokhov <dmitry.torokhov@...il.com>,
        KT Liao <kt.liao@....com.tw>,
        Andrew Duggan <aduggan@...aptics.com>
Cc:     Adrian Alves <aalves@...il.com>, linux-kernel@...r.kernel.org,
        linux-input@...r.kernel.org
Subject: Re: [PATCH 03/10] Input: synaptics-rmi4 - f30/f03: Forward mechanical buttons on buttonpads to PS/2 guest

Hi Dmitry,

On Feb 06 2017 or thereabouts, Dmitry Torokhov wrote:
> Hi Benjamin,
> 
> On Tue, Jan 10, 2017 at 05:11:21PM +0100, Benjamin Tissoires wrote:
> > +void rmi_f03_commit_buttons(struct rmi_function *fn)
> > +{
> > +   struct f03_data *f03 = dev_get_drvdata(&fn->dev);
> > +   int i;
> > +
> > +   f03->serio->extra_byte = f03->overwrite_buttons;
> > +
> > +   for (i = 0; i < 3; i++)
> > +           serio_interrupt(f03->serio, 0x00, 0x00);
> 
> How does this work with protocols larger than 3 bytes (i.e. anything but
> bare PS/2)? Or F03 limits the kind of devices we connect to RMI4 device?

Right, there is no such constraint.

> 
> Overall, I am very uncomfortable with the concept of "extra byte" at
> serio level. Maybe we should define a new flag, something like
> SERIO_SYNTHETIC or SERIO_OOB_DATA, and have something like:
> 
> void rmi_f03_commit_buttons(struct rmi_function *fn)
> {
>       struct serio *serio = f03->serio;
> 
>       serio_pause_rx(serio);
>       if (serio->drv) {
>               serio->drv->interrupt(serio, PSMOUSE_EXTRA_BTNS,
>                                     SERIO_OOB_DATA);
>               serio->drv->interrupt(serio, f03->overwrite_buttons.
>                                     SERIO_OOB_DATA);
>       }
>       serio_pause_rx(serio);
> }
> 
> and have psmouse driver store and use it as it sees fit.

Sounds good to me. Would the 2 following patches be OK?

Cheers,
Benjamin


Benjamin Tissoires (2):
  Input: psmouse - add a custom serio protocol to send extra information
  Input: synaptics-rmi4 - f30/f03: Forward mechanical buttons on
    buttonpads to PS/2 guest

 drivers/input/mouse/psmouse-base.c | 39 +++++++++++++++++++--
 drivers/input/mouse/psmouse.h      |  5 +++
 drivers/input/rmi4/rmi_driver.h    | 14 ++++++++
 drivers/input/rmi4/rmi_f03.c       | 39 +++++++++++++++++++++
 drivers/input/rmi4/rmi_f30.c       | 72 ++++++++++++++++++++++++++++++--------
 include/uapi/linux/serio.h         |  7 ++--
 6 files changed, 155 insertions(+), 21 deletions(-)

-- 
2.9.3