[git pull] Input updates for 2.6.33-rc2

[git pull] Input updates for 2.6.33-rc2

[Dmitry Torokhov]

Hi Linus,

Please pull from:

	git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git for-linus
or
	master.kernel.org:/pub/scm/linux/kernel/git/dtor/input.git for-linus

to receive updates for the input subsystem.

Changelog:
---------

Dmitry Torokhov (5):
      Input: speed up suspend/shutdown for PS/2 mice and keyboards
      Input: serio - do not mark kseriod freezable anymore
      Input: ff-memless - another fix for signed to unsigned overflow
      Input: iforce - fix oops on device disconnect
      Input: matrix-keypad - handle cases when GPIOs can't be wakeup sources

Jari Vanhala (2):
      Input: ff-memless - start playing FF effects immediately
      Input: ff-memless - add notion of direction to for rumble effects


Diffstat:
--------

 drivers/input/ff-memless.c                  |   48 +++++++++++++++++++++++----
 drivers/input/joystick/iforce/iforce-main.c |   26 +++------------
 drivers/input/joystick/iforce/iforce-usb.c  |   28 ++++------------
 drivers/input/joystick/iforce/iforce.h      |    2 -
 drivers/input/keyboard/atkbd.c              |    5 ++-
 drivers/input/keyboard/matrix_keypad.c      |   29 ++++++++++++----
 drivers/input/mouse/psmouse-base.c          |    5 ++-
 drivers/input/serio/serio.c                 |   11 +-----
 8 files changed, 84 insertions(+), 70 deletions(-)

-- 
Dmitry

Re: [git pull] Input updates for 2.6.33-rc2

[Linus Torvalds]

On Sun, 27 Dec 2009, Dmitry Torokhov wrote:
>
>  8 files changed, 84 insertions(+), 70 deletions(-)

I get

 14 files changed, 94 insertions(+), 96 deletions(-)

so unpulled.

If you updated your tree on purpose, you need to send me an email saying 
so (preferably as a reply to your previous "please pull", so that when I 
see the old one and it doesn't match what I get, I can then also see the 
new one without having to search for it)

		Linus

Re: [git pull] Input updates for 2.6.33-rc2

[Dmitry Torokhov]

On Dec 30, 2009, at 12:58 PM, Linus Torvalds <torvalds@linux-foundation.org 
 > wrote:

>
>
> On Sun, 27 Dec 2009, Dmitry Torokhov wrote:
>>
>> 8 files changed, 84 insertions(+), 70 deletions(-)
>
> I get
>
> 14 files changed, 94 insertions(+), 96 deletions(-)
>
> so unpulled.
>
> If you updated your tree on purpose, you need to send me an email  
> saying
> so (preferably as a reply to your previous "please pull", so that  
> when I
> see the old one and it doesn't match what I get, I can then also see  
> the
> new one without having to search for it)

Yes, I just updated that branch with more fixes, not seeing any pulls  
after rc2 i thought you were gone till after New Year. I will resend  
updated pull request later tonight.

-- 
Dmitry

Re: [git pull] Input updates for 2.6.33-rc2

[Dmitry Torokhov]

On Wed, Dec 30, 2009 at 01:16:31PM -0800, Dmitry Torokhov wrote:
> 
> 
> 
> 
> On Dec 30, 2009, at 12:58 PM, Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
> 
> >
> >
> >On Sun, 27 Dec 2009, Dmitry Torokhov wrote:
> >>
> >>8 files changed, 84 insertions(+), 70 deletions(-)
> >
> >I get
> >
> >14 files changed, 94 insertions(+), 96 deletions(-)
> >
> >so unpulled.
> >
> >If you updated your tree on purpose, you need to send me an email
> >saying
> >so (preferably as a reply to your previous "please pull", so that
> >when I
> >see the old one and it doesn't match what I get, I can then also
> >see the
> >new one without having to search for it)
> 
> Yes, I just updated that branch with more fixes, not seeing any
> pulls after rc2 i thought you were gone till after New Year. I will
> resend updated pull request later tonight.
> 

Hi Linus,

Please pull from:

	git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git for-linus
or
	master.kernel.org:/pub/scm/linux/kernel/git/dtor/input.git for-linus

to receive updates for the input subsystem.

Changelog:
---------

Dmitry Torokhov (7):
      Input: speed up suspend/shutdown for PS/2 mice and keyboards
      Input: serio - do not mark kseriod freezable anymore
      Input: ff-memless - another fix for signed to unsigned overflow
      Input: iforce - fix oops on device disconnect
      Input: matrix-keypad - handle cases when GPIOs can't be wakeup sources
      Input: lifebook - add CONFIG_DMI dependency
      Input: iforce - wait for command completion when closing the device

Felipe Balbi (2):
      Input: twl4030_keypad - switch to using threaded IRQ
      Input: twl4030-pwrbutton - switch to using threaded IRQ

Jari Vanhala (2):
      Input: ff-memless - start playing FF effects immediately
      Input: ff-memless - add notion of direction to for rumble effects

Rakib Mullick (1):
      Input: wistron - fix test for CONFIG_PM

René Bolldorf (1):
      Input: psmouse - fix compile warning in hgpk module


Diffstat:
--------

 drivers/input/ff-memless.c                  |   48 +++++++++++++++++++++++----
 drivers/input/joystick/iforce/iforce-main.c |   29 ++++------------
 drivers/input/joystick/iforce/iforce-usb.c  |   29 ++++------------
 drivers/input/joystick/iforce/iforce.h      |    2 -
 drivers/input/keyboard/atkbd.c              |    5 ++-
 drivers/input/keyboard/matrix_keypad.c      |   29 ++++++++++++----
 drivers/input/keyboard/twl4030_keypad.c     |   11 +-----
 drivers/input/misc/twl4030-pwrbutton.c      |   14 +-------
 drivers/input/misc/wistron_btns.c           |    2 +-
 drivers/input/mouse/Kconfig                 |    2 +-
 drivers/input/mouse/hgpk.c                  |    1 -
 drivers/input/mouse/lifebook.c              |    2 -
 drivers/input/mouse/psmouse-base.c          |    5 ++-
 drivers/input/serio/serio.c                 |   11 +-----
 14 files changed, 94 insertions(+), 96 deletions(-)

-- 
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [git pull] Input updates for 2.6.33-rc2

[Linus Torvalds]

On Wed, 30 Dec 2009, Dmitry Torokhov wrote:
>
> Dmitry Torokhov (7):
>       Input: speed up suspend/shutdown for PS/2 mice and keyboards

Btw, the thing I like best about this commit is how it adds some comments 
on the keyboard commands.

That said, the naming and the comments aren't all that wonderful. I don't 
think you should use the word "RESET" for command f5/f6: the command is 
called "Set Default" (and ".. Disable" for f5), and there is no actual 
reset involved. 

So I suspect the command should be renamed to

	#define ATKBD_CMD_SET_DEFAULTS_AND_DISABLE	0x00f5
	#define ATKBD_CMD_SET_DEFAULTS			0x00f6

and then the comments wouldn't even be needed (and anybody reading the 
code would not have to look them up in the header file).

And then "ATKBD_CMD_RESET_BAT" should probably just be "ATKBD_CMD_RESET" 
(I wonder what the "BAT" part is all about?)

Oh, and command F5h is not always a "set defaults and disable". For some 
(all?) PS/2 mice it seems to be _just_ a "disable", and I'm not at all 
sure you should have used it for the psmouse_cleanup() function. You used 
to do a

	psmouse_reset(psmouse);

which sent a full reset to the mouse (and waits for the two-byte ACK). It 
did _not_ disable the mouse - that happened earlier in 
psmouse_deactivate(), when you sent it the PSMOUSE_CMD_DISABLE command.

You changed that 'psmouse_reset()' to

	/*
	 * Reset the mouse to defaults (bare PS/2 protocol).
	 */
	        ps2_command(&psmouse->ps2dev, NULL, PSMOUSE_CMD_RESET_DIS);

which now disables it again, only to then later do
        
	/*      
	 * Some boxes, such as HP nx7400, get terribly confused if mouse
	 * is not fully enabled before suspending/shutting down.
	 */     
	        ps2_command(&psmouse->ps2dev, NULL, PSMOUSE_CMD_ENABLE);

which seems odd. IOW, Why do you do a "set defaults and disable" followed 
by a "enable", when you already had it disabled?

I think the PSMOUSE_CMD_RESET that _used_ to be there would re-enable most 
mice, but now that you don't even do that any more, I get the feeling that 
the "set defaults and disable" should be just a "set defaults".

(You may have good reasons for having picked the "and disable" command. I 
just wonder what they are)

		Linus

Re: [git pull] Input updates for 2.6.33-rc2

[Dmitry Torokhov]

On Thu, Dec 31, 2009 at 11:50:59AM -0800, Linus Torvalds wrote:
> 
> 
> On Wed, 30 Dec 2009, Dmitry Torokhov wrote:
> >
> > Dmitry Torokhov (7):
> >       Input: speed up suspend/shutdown for PS/2 mice and keyboards
> 
> Btw, the thing I like best about this commit is how it adds some comments 
> on the keyboard commands.
> 
> That said, the naming and the comments aren't all that wonderful. I don't 
> think you should use the word "RESET" for command f5/f6: the command is 
> called "Set Default" (and ".. Disable" for f5), and there is no actual 
> reset involved. 
> 

I guess it is just a matter of wording. For me "reset" and "revert to
initial state" are pretty much the same.

> So I suspect the command should be renamed to
> 
> 	#define ATKBD_CMD_SET_DEFAULTS_AND_DISABLE	0x00f5

ATKBD_CMD_SET_DEFAULTS_AND_DISABLE_DATA_REPORTING :) I was just trying
to follow the existing naming style. It was there before me and it does
make sense to me still so I did not feel the neet to change it.

> 	#define ATKBD_CMD_SET_DEFAULTS			0x00f6
> 
> and then the comments wouldn't even be needed (and anybody reading the 
> code would not have to look them up in the header file).
> 
> And then "ATKBD_CMD_RESET_BAT" should probably just be "ATKBD_CMD_RESET" 
> (I wonder what the "BAT" part is all about?)

That came from Vojtech but I am pretty sure this stands for Basic
Assurance Test since full reset for touchpads for example involves
recalibration. See this excerpt from Synaptics docs:

"At power-on, the PS/2 TouchPad performs a self-test and calibration,
then transmits the completion code $AA and ID code $00. If the TouchPad
fails its self-test, it transmits error code $FC and ID code $00.  This
processing also occurs when a software Reset ($FF) command is received.
The host should not attempt to send commands to the TouchPad until the
calibration/self-test is complete."

> 
> Oh, and command F5h is not always a "set defaults and disable". For some 
> (all?) PS/2 mice it seems to be _just_ a "disable",

Exactly, that is why it is called PSMOUSE_CMD_DISABLE when used in mouse
context. However snippets of docs I have indicate that for keyboards it
also causes them to reset to defaults.

> and I'm not at all 
> sure you should have used it for the psmouse_cleanup() function.

I am using PSMOUSE_CMD_RESET_DIS (0xf6) in psmouse_cleanup().

> You used 
> to do a
> 
> 	psmouse_reset(psmouse);
> 
> which sent a full reset to the mouse (and waits for the two-byte ACK). It 
> did _not_ disable the mouse - 

It does actually, at least for some devices (incidentally the ones that
may give trouble to the BIOS if left as is):

"The reset state of the TouchPad is as follows:
- Reported sample rate is 100 samples per second (see page 30).
- Reported resolution is 4 counts per mm (see page 32).
- Scaling is 1:1.
- Stream mode is selected.
- Data reporting is disabled. <---------------
- Absolute mode is disabled."

> that happened earlier in 
> psmouse_deactivate(), when you sent it the PSMOUSE_CMD_DISABLE command.

What happened before doing psmouse_reset() (which is PSMOUSE_CMD_RESET_BAT)
should not matter because all previous state should be reset after it.

> 
> You changed that 'psmouse_reset()' to
> 
> 	/*
> 	 * Reset the mouse to defaults (bare PS/2 protocol).
> 	 */
> 	        ps2_command(&psmouse->ps2dev, NULL, PSMOUSE_CMD_RESET_DIS);
> 
> which now disables it again, only to then later do
>         
> 	/*      
> 	 * Some boxes, such as HP nx7400, get terribly confused if mouse
> 	 * is not fully enabled before suspending/shutting down.
> 	 */     
> 	        ps2_command(&psmouse->ps2dev, NULL, PSMOUSE_CMD_ENABLE);
> 
> which seems odd. IOW, Why do you do a "set defaults and disable" followed 
> by a "enable", when you already had it disabled?

We do not have "light reset" option, so we have to use "light reset and
disable" to return to the bare 3-byte PS/2 protocol. Then we have to
re-enable the device because BIOS on some HP notebooks would get
confused and not suspend (or resume, I don't quite remember) if mouse
left disabled.

> 
> I think the PSMOUSE_CMD_RESET that _used_ to be there would re-enable most 
> mice, but now that you don't even do that any more, I get the feeling that 
> the "set defaults and disable" should be just a "set defaults".

There is no such command for mice unfortunately.

> 
> (You may have good reasons for having picked the "and disable" command. I 
> just wonder what they are)
> 
> 		Linus

-- 
Dmitry