Re: 4.0.0-rc mouse/keyboard events get stuck

From: Hans de Bruin
Date: Wed Apr 08 2015 - 15:06:32 EST

On 04/08/2015 05:29 PM, Hans de Goede wrote:

On 08-04-15 12:59, Hans de Bruin wrote:
On 04/06/2015 11:23 PM, Dmitry Torokhov wrote:
Hi Hans,

On Mon, Apr 6, 2015 at 11:39 AM, Hans de Bruin <jmdebruin@xxxxxxxxx>
Since 4.0.0-rc something mu laptop has some mouse /keyboard problems.
When I play KDE kpatience, a card game, card stick to my
mousepointer: left
mouse button down on a card, move mouse pointer and card to another
left mouse button up to drop the card. This should drop the card,
but often
the mouse button up events gets stuck somewhere until I move he mouse
pointer a little bit.
In web based mail clients something similar happens when I drag mail to
another folder.
Sometime after I start a xterm and start typing, the letters don't
until move the mouse pointer.

Bisecting is out of the question. This kernel realease was so
dramatic I was
glad to have functioning kde screen when rc4 arrived.

My laptop is a Dell Latitude D430, it has a touchpad and a pointer
an two
sets of buttons.

[ 0.636492] input: AT Translated Set 2 keyboard as
[ 0.633108] mousedev: PS/2 mouse device common for all mice
[ 80.543044] input: PS/2 ALPS Mouse as

Linus just merged 2 fixes to ALPS driver that might help you. Could
you please try compiling tip of the tree and telling me if it fixes
your issue?

I am at rc.7 and the problem is still there. But maybe you can get
away with "user error". Here is a picture of the layout of my laptop:

Lets call the blue thing ts. The buttons directly beneath the keys
ts-l and ts-r. For the touchpad tp, tp-l and tp-r. And because one can
never have to many input devices um, um-l and um-r for a usb mouse.

When I play kpatience and stick to one device ts or ts-l or mp and
mp-l no problems arise. I can also mix devices for example um and
tp-l. There are several combinations. Only one of these ts and tp-l
show the problematic behavior. This is the way i use my laptop (user

Ok, this one is a regression which is caused by:

I've attached a fix which I will formally send / submit upstream in a
few minutes too.

This patch fixes the regression on my laptop. Thanks for the quick response.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at