Message-ID: <20161108150942.GB10327@mail.corp.redhat.com>
Date:   Tue, 8 Nov 2016 16:09:42 +0100
From:   Benjamin Tissoires <benjamin.tissoires@...il.com>
To:     Nick Dyer <nick@...anahar.org>
Cc:     Dmitry Torokhov <dmitry.torokhov@...il.com>,
        Andrew Duggan <aduggan@...aptics.com>,
        Lyude Paul <thatslyude@...il.com>,
        Christopher Heiny <cheiny@...aptics.com>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        Dennis Wassenberg <dennis.wassenberg@...unet.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        linux-input <linux-input@...r.kernel.org>,
        Jiri Kosina <jikos@...os.cz>, Chris Healy <cphealy@...il.com>
Subject: Re: [PATCH v3 00/18] Synaptics RMI4 and SMBus implementation

Hi Nick,

On Nov 07 2016 or thereabouts, Nick Dyer wrote:
> On Fri, Nov 04, 2016 at 09:23:48AM +0100, Benjamin Tissoires wrote:
> > On Thu, Oct 13, 2016 at 5:50 PM, Benjamin Tissoires
> > <benjamin.tissoires@...hat.com> wrote:
> > > Hi guys,
> > >
> > > This is the third submission of this series, with some addition of RMI4 patches
> > > currently waiting on the list.
> > >
> > > I integrated Bjorn's patch because in the end, it seems to be the right thing
> > > to do. We can create an irqchip in hid-rmi and also provide an IRQ there.
> > > Worse case, if this doesn't work, we can always add a special case to
> > > disable the handling of the IRQs in core.
> > >
> > > I took Nick's patch as it's a nice cleanup and it solves an issue where
> > > the mutex irq_mutex might not be initialized (in the case we are not handling
> > > IRQ directly in core).
> > >
> > > I also took some patches from Andrew that were submitted back in July, and that
> > > were left over. They are required to allow hid-rmi to use rmi4-core without
> > > losing functionality. The DT binding seemed to be problematic, so I just lefted
> > > it out of the series. We can add it later when required (if required).
> > > Andrew, regarding hid-rmi, I'll try to give you a draft of the irqchip
> > > implementation based on what I did for SMBus Host Notify.
> > >
> > > The rest is an update of the SMBus work. The main differences is that now
> > > rmi4-smbus doesn't handle anymore the IRQ and I hope that the SMBus Host Notify
> > > patches I posted on linux-i2c will be approved by Wolfram.
> > 
> > I understand you are busy with many other projects, but it would be
> > great if we could have this series (or a subset of this series)
> > landing ASAP.
> > 
> > We already have at least 2 machines (the Lenovo X1 Tablet and the HP
> > Pavilion x2) that need to be using the function 12 in Synaptics RMI4,
> > and so this series is a requirement to enable those touchpads.
> > 
> > If you really want to postpone the review of the new files in this
> > series, I think merging only the following patches would be
> > acceptable:
> > [v3,01/18] Input: synaptics-rmi4 - Move IRQ handling to rmi_driver
> > [v3,02/18] Input: synaptics-rmi4 - factor out functions from probe
> > [v3,03/18] Input: synaptics-rmi4 - Handle incomplete input data
> > [v3,04/18] Input: synaptics-rmi4 - Add parameters for dribble packets
> > and palm detect gesture
> > [v3,05/18] Input: synaptics-rmi4 - Add support for controlling dribble
> > packets in F12
> > [v3,06/18] Input: synaptics-rmi4 - Set the ABS_MT_TOOL_TYPE bit to
> > report tool type
> > [v3,09/18] Input: synaptics-rmi4 - have only one struct platform data
> > [v3,18/18] Input: synaptics-rmi4 - fix documentation of
> > rmi_2d_sensor_platform_data
> > 
> > This is the bare minimum I think to enable the F12 in hid-rmi. It
> > should apply gracefully (crossing finger for 9/18), and would allow us
> > to switch hid-rmi to rmi4-core.
> > 
> > I guess Jiri will probably want a stable branch with those commits so
> > we can also merge the switch from hid-rmi to use rmi4-core in v4.10.
> 
> Hi Benjamin-
> 
> Just to say, there's a few other patches that we would be keen to also
> get into v4.10 if possible:
> 
> These three enable the F54 support to work properly on more devices:
> 
> Input: synaptics-rmi4 - stop scanning PDT after two empty pages
> https://patchwork.kernel.org/patch/9398323/
> 
> Input: synaptics-rmi4 - add support for F55 sensor tuning
> https://patchwork.kernel.org/patch/9359055/
> 
> Input: synaptics-rmi4 - Propagate correct number of rx and tx electrodes
> to F54
> https://patchwork.kernel.org/patch/9359059/
> 
> 
> I would also be keen to get at least some of the F34 reflash support in:
> http://www.spinics.net/lists/linux-input/msg47479.html
> 
> Would you be able to merge any of these into your series? I am happy to
> do some of the work if you are busy.

I would have answered, yes, sure, but given that Dmitry started to
review/apply part of the F34 series, I am not so sure anymore.

I am a little bit busy this week, but I still kept an eye on the review
I need to do on F34.

I think it would be good to have a common tree with the changes we want
to push in rmi4. An other solution could be to set a group of patches in
patchwork to keep track of them and not lose one.
If you want to lead that, feel free to setup a branch.

Cheers,
Benjamin

> 
> cheers
> 
> Nick
> 
> > > Andrew Duggan (4):
> > >   Input: synaptics-rmi4 - Handle incomplete input data
> > >   Input: synaptics-rmi4 - Add parameters for dribble packets and palm
> > >     detect gesture
> > >   Input: synaptics-rmi4 - Add support for controlling dribble packets in
> > >     F12
> > >   Input: synaptics-rmi4 - Set the ABS_MT_TOOL_TYPE bit to report tool
> > >     type
> > >
> > > Benjamin Tissoires (10):
> > >   Input: synaptics-rmi4 - add SMBus support
> > >   Input: serio - store the pt_buttons in the struct serio directly
> > >   Input: synaptics-rmi4 - have only one struct platform data
> > >   Input: synaptics-rmi4 - Add rmi_find_function()
> > >   Input: synaptics-rmi4 - f30/f03: Forward mechanical buttons on
> > >     buttonpads to PS/2 guest
> > >   Input: synaptics - allocate a Synaptics Intertouch device
> > >   Input: synaptics-rmi4 - add rmi_platform
> > >   Input: synaptics-rmi4 - smbus: call psmouse_deactivate before
> > >     binding/resume
> > >   Input: synaptics-rmi4 - smbus: on resume, try 3 times if init fails
> > >   Input: synaptics-rmi4 - fix documentation of
> > >     rmi_2d_sensor_platform_data
> > >
> > > Bjorn Andersson (1):
> > >   Input: synaptics-rmi4 - Move IRQ handling to rmi_driver
> > >
> > > Dennis Wassenberg (1):
> > >   Input: synaptics-rmi4 - f03: grab data passed by transport device
> > >
> > > Lyude Paul (1):
> > >   Input: synaptics-rmi4 - add support for F03
> > >
> > > Nick Dyer (1):
> > >   Input: synaptics-rmi4 - factor out functions from probe

Powered by blists - more mailing lists