Message-Id: <201412252111.06004@pali>
Date:	Thu, 25 Dec 2014 21:11:05 +0100
From:	Pali Rohár <pali.rohar@...il.com>
To:	Alex Hung <alex.hung@...onical.com>
Cc:	Gabriele Mazzotta <gabriele.mzt@...il.com>,
	Darren Hart <dvhart@...radead.org>,
	Matthew Garrett <mjg59@...f.ucam.org>,
	"platform-driver-x86@...r.kernel.org" 
	<platform-driver-x86@...r.kernel.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/3] Dell Airplane Mode Switch driver

I will try to recap all information which we have...

*) We should not send wireless key press to userspace when BIOS 
already handles wireless state (and enable/disable wifi): 
http://www.spinics.net/lists/platform-driver-x86/msg05922.html

*) some tested dell machines does not implement GRBT method (or 
report constant value) which could return state of wireless 
(enabled/disabled) -- e.g. Inspiron 7447

*) dell-wireless driver is doing nothing on devices which have 
wireless slider switch (except calling CRBT/ARBT methods)

*) all tested machines emit key with keycode 240 (scancode is 
probably 136 = 0x88) to userspace via i8042 bus/AT keyboard when 
wireless button/slider is pressed/switched

*) both drivers dell-wireless and dell-rbtn do not implement 
setting soft rfkill state (or change wifi state)

So in my opinion: if we decide to use driver for acpi DELLABCE 
device we should use dell-rbnt for devices with hw slider switch 
and dell-wireless for devices with fn button. I think it does not 
make sense to use dell-wireless for devices with hw slider 
because it do nothing and dell-rbtn for devices with fn key 
button as GRBT does not working properly.

And second note: Do we need some driver for acpi DELLABCE device? 
Which problem is trying acpi DELLABCE device to solve? Is not 
everything working fine without driver for DELLABCE device?

My dell-rbtn approach is trying to export rfkill interface from 
DELLABCE device and eliminate using i8042 hook function in smbios 
dell-laptop driver.

Alex, can you check if scancode of wireless change generated by 
BIOS is on all machines same: 136 (0x88)? And is send by keyboard 
controller (not acpi/wmi)?

On Thursday 25 December 2014 04:13:02 Alex Hung wrote:
> I usually prefer to stick with formal document as everything
> else can be changed without reasons.
> 
> I am not certain whether the keypress is defined in Dell's
> document, and I will confirm with Dell. If this keypress is
> well defined, it is a good idea to use it.
> 
> 
> On Wed, Dec 24, 2014 at 7:40 PM, Gabriele Mazzotta
> 
> <gabriele.mzt@...il.com> wrote:
> > On Wednesday 24 December 2014 17:13:45 Alex Hung wrote:
> >> I uploaded acpidump files [1] (except for XPS 13 which is
> >> not available), and this should help clarify what has been
> >> tested.
> >> 
> >> Does Inspirion 5721 does not have either DELLABCE or
> >> DELRBTN. It is used for comparison. My apologies that I
> >> did not point this out in previous email.
> >> 
> >> When calling ARBT(1), BIOS will no longer issue scancode
> >> and will not pull low hardware pin "W_DISABLE#" on mini
> >> card. This essentially gives all wireless control to OS.
> >> This is likely the answer to Microsoft's Windows
> >> Certification Program
> >> System.Client.RadioManagement.HardwareButton [2] as below:
> >> 
> >> =============================================
> >> If a PC has a physical (hardware) button switch on a PC
> >> that turns wireless radios on and off, it must be software
> >> controllable and interact appropriately with the Radio
> >> Management UI
> >> =============================================
> >> 
> >> Dell's BIOS does issue a notify(RBTN, 0x80). This is done
> >> purposely to re-enumerate the state of radio switch which
> >> may be changed when system is in S3 or S4. I think this
> >> should not occur when CRBT returns 0 or 1 (for hotkey that
> >> cannot be changed during S3 or S4), but that's how it is
> >> done currently.
> > 
> > The notification is sent on my XPS13 (CRBT returns 0),
> > toggling the WiFi state on resume.
> > 
> >> dell-wireless does not handle this notification in S3 or S4
> >> for following reasons:
> >> 
> >> 1. dell-wireless does not handle slider (i.e. CRBT = 2 or
> >> 3). Device drivers should read the hardware pin,
> >> "W_DISABLE#" on mini spec and change hard block
> >> accordingly. This pin is commonly used by OEM today.
> >> 
> >> 2. it is not possible to distinguish the notification
> >> (0x80) from hotkey press or S3/S4. I also concerned this
> >> may mis-trigger state change when resuming from S3 or S4,
> >> but it does not. Does any know how to ignore this
> >> notification during resume only?
> >> 
> >> dell-rbtn can use this notification + method (GRBT) [2] to
> >> solve the problem that slider state.
> > 
> > Unfortunately this won't solve the problem for me. After
> > ARBT is called with 1 as parameter, it seems that GRBT
> > always returns 1.
> > 
> > I don't know how to ignore the notification on resume, if
> > not through a flag set by a PM callback.
> > 
> > Given that all the tested laptops reported a keypress on the
> > i8042 bus, isn't it better to rely on that instead?
> > 
> >> [1] http://people.canonical.com/~alexhung/dell-acpidump/
> >> [2]
> >> http://msdn.microsoft.com/en-us/library/windows/hardware/j
> >> j128256.aspx [3] It seems GRBT may not always be
> >> implemented...
> >> 
> >> I'd love to do more tests and share the results on any
> >> particular systems, but I may need some more detailed
> >> instructions.
> >> 
> >> 
> >> 
> >> On Tue, Dec 23, 2014 at 3:16 AM, Gabriele Mazzotta
> >> 
> >> <gabriele.mzt@...il.com> wrote:
> >> > On Monday 22 December 2014 15:27:57 Alex Hung wrote:
> >> >> = Testing =
> >> >> 
> >> >> I tested six Dell systems for two sets of patches for
> >> >> dell radio button - two system with radio slider and
> >> >> four with radio hotkey. There are also two systems with
> >> >> working ARBT method.
> >> >> 
> >> >> == Basic Information ==
> >> >> Based OS: Ubuntu 14.10 (kernel 3.16 [1]) and kernel 3.18
> >> >> [2]
> >> >> 
> >> >> Patches:
> >> >> 1. dell-wireless v3 = original v2 + Gabriele's
> >> >> suggestion [3] 2. dell-rbtn [4]
> >> >> 
> >> >> Method:
> >> >> 1. run "rfkill list" and press hotkey / toggle slider
> >> >> during runtime 2. run "rfkill list" and toggle slider
> >> >> during S3
> >> >> 
> >> >> == Results ==
> >> >> 
> >> >> I summarized the tests in Google sheet as below. Please
> >> >> advise if anyone has problem reading it.
> >> >> 
> >> >> https://docs.google.com/spreadsheets/d/1voffS6dNglwAExSG
> >> >> h3UmG__UAO2qfZ829CkJLPo06aI/edit?usp=sharing
> >> >> 
> >> >> PS. The document will stay as long as possible for
> >> >> future references.
> >> >> 
> >> >> == Summary ==
> >> >> 
> >> >> 1. I did not observed a duplicated event. However,
> >> >> keycode 240 (unknown) is generated on many UUT. It is
> >> >> not issued by dell-laptop or del-wmi. I am suspecting
> >> >> it is the other event Pali observes but it can be the
> >> >> result of different distro.
> >> >> 
> >> >> 2. Some system issues scancode "0xe0 0x73 0xe0 0xf3". It
> >> >> can also be used toggle wireless state but this can
> >> >> also be distro-dependent. This scancode does nothing on
> >> >> Ubuntu 14.10.
> >> >> 
> >> >> 2. There are two systems with working ARBT (XPS 13 9333
> >> >> and Inspiron 7447). Calling ARBT(1) changes BIOS
> >> >> behaviours, and this matches to Dell's document. We
> >> >> should include it in the patch for maximum capability.
> >> >> 
> >> >> 
> >> >> [1] dell-wireless is only tested 3.16.
> >> >> [2] dell-rbtn is tested on 3.16 and 3.18, but no
> >> >> differences are observed. [3]
> >> >> http://people.canonical.com/~alexhung/dell-wireless/
> >> >> [4] http://people.canonical.com/~alexhung/dell-rbtn/
> >> > 
> >> > I've just tried the last revision of dell-wireless and
> >> > noticed that a notification (0x80) is sent to DELLABCE
> >> > after a transition from S3 to S0, causing dell-wireless
> >> > to send KEY_RFKILL. This shouldn't happen. Same thing
> >> > for transitions from S4 to S0.
> >> > 
> >> > Gabriele

-- 
Pali Rohár
pali.rohar@...il.com

Download attachment "signature.asc " of type "application/pgp-signature" (199 bytes)

Powered by blists - more mailing lists