[regression] 2.6.30 leaking keys on console switch

[regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

One new issue I see in 2.6.30 (I think I saw it in some rc*s) too is
that console switch often leaks the Fkey of the console switch into the 
target console.

The symptome is that e.g. firefox if it was active and you're switching
into X displays its help screen (F1 = help). It happens on non X 
consoles too, but of course the symptoms are different here.

That's a regression against 2.6.29; I don't remember it in .29

Haven't investigated code.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

[added Dmitry to CC]

On Tue, 7 Jul 2009, Andi Kleen wrote:

> One new issue I see in 2.6.30 (I think I saw it in some rc*s) too is 
> that console switch often leaks the Fkey of the console switch into the 
> target console.
> 
> The symptome is that e.g. firefox if it was active and you're switching
> into X displays its help screen (F1 = help). It happens on non X 
> consoles too, but of course the symptoms are different here.
> 
> That's a regression against 2.6.29; I don't remember it in .29
> 
> Haven't investigated code.

Are you seeing with a specific keyboard type(s), or is this a general 
problem? Could you confirm whether it is specific either 
to PS2 or USB keyboards only, or it does happen for both?

Thanks,

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

> Are you seeing with a specific keyboard type(s), or is this a general 
> problem? 
> Could you confirm whether it is specific either 
> to PS2 or USB keyboards only, or it does happen for both?

I'm only seeing it with a USB keyboard, haven't seen it with PS/2
(but I rarely do console switches with PS/2, so might have missed it)

The USB keyboard is a rather ordinary logitech keyboard.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Tue, 7 Jul 2009, Andi Kleen wrote:

> > Are you seeing with a specific keyboard type(s), or is this a general 
> > problem? 
> > Could you confirm whether it is specific either 
> > to PS2 or USB keyboards only, or it does happen for both?
> I'm only seeing it with a USB keyboard, haven't seen it with PS/2
> (but I rarely do console switches with PS/2, so might have missed it)
> The USB keyboard is a rather ordinary logitech keyboard.

Could you please gather 'evtest' output from the /dev/input/eventX device 
corresponding to your keyboard from the situation this problem triggers?

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

On Tue, Jul 07, 2009 at 10:50:01AM +0200, Jiri Kosina wrote:
> 
> [added Dmitry to CC]
> 
> On Tue, 7 Jul 2009, Andi Kleen wrote:
> 
> > One new issue I see in 2.6.30 (I think I saw it in some rc*s) too is 
> > that console switch often leaks the Fkey of the console switch into the 
> > target console.
> > 
> > The symptome is that e.g. firefox if it was active and you're switching
> > into X displays its help screen (F1 = help). It happens on non X 
> > consoles too, but of course the symptoms are different here.
> > 
> > That's a regression against 2.6.29; I don't remember it in .29
> > 
> > Haven't investigated code.
> 
> Are you seeing with a specific keyboard type(s), or is this a general 
> problem? Could you confirm whether it is specific either 
> to PS2 or USB keyboards only, or it does happen for both?
> 

Can't think of any input chnages in this timeftrame that would cause
such effect. Maybe TTY changes?

-- 
Dmitry

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Tue, 7 Jul 2009, Dmitry Torokhov wrote:

> > > One new issue I see in 2.6.30 (I think I saw it in some rc*s) too is 
> > > that console switch often leaks the Fkey of the console switch into the 
> > > target console.
> > > 
> > > The symptome is that e.g. firefox if it was active and you're switching
> > > into X displays its help screen (F1 = help). It happens on non X 
> > > consoles too, but of course the symptoms are different here.
> > > 
> > > That's a regression against 2.6.29; I don't remember it in .29
> > > 
> > > Haven't investigated code.
> > 
> > Are you seeing with a specific keyboard type(s), or is this a general 
> > problem? Could you confirm whether it is specific either 
> > to PS2 or USB keyboards only, or it does happen for both?
> 
> Can't think of any input chnages in this timeftrame that would cause
> such effect. Maybe TTY changes?

For completness, Andi sent me the following evtest output off-list 
yesterday. The LeftAlt being autorepeated at time 1246964430.448470 is a 
little bit suspicious, as it is preceeded by release ... I don't see how 
this could be causing the behavior Andi is observing though.

I am not able to reproduce the problem myself.

Event: time 1246964411.828591, type 4 (Misc), code 4 (ScanCode), value 70028
Event: time 1246964411.828596, type 1 (Key), code 28 (Enter), value 0
Event: time 1246964411.828598, -------------- Report Sync ------------
Event: time 1246964412.260590, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964412.260594, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964412.260597, -------------- Report Sync ------------
Event: time 1246964412.512460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.512462, -------------- Report Sync ------------
Event: time 1246964412.548460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.548462, -------------- Report Sync ------------
Event: time 1246964412.584458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.584460, -------------- Report Sync ------------
Event: time 1246964412.620458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.620460, -------------- Report Sync ------------
Event: time 1246964412.656460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.656461, -------------- Report Sync ------------
Event: time 1246964412.692458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.692460, -------------- Report Sync ------------
Event: time 1246964412.728459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.728461, -------------- Report Sync ------------
Event: time 1246964412.764458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.764459, -------------- Report Sync ------------
Event: time 1246964412.800461, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.800463, -------------- Report Sync ------------
Event: time 1246964412.836457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.836459, -------------- Report Sync ------------
Event: time 1246964412.872458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.872460, -------------- Report Sync ------------
Event: time 1246964412.908459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.908460, -------------- Report Sync ------------
Event: time 1246964412.944457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.944459, -------------- Report Sync ------------
Event: time 1246964412.980458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964412.980460, -------------- Report Sync ------------
Event: time 1246964412.996593, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964412.996597, type 1 (Key), code 65 (F7), value 1
Event: time 1246964412.996599, -------------- Report Sync ------------
Event: time 1246964413.000467, type 17 (LED), code 0 (NumLock), value 0
Event: time 1246964413.000469, -------------- Report Sync ------------
Event: time 1246964413.156598, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964413.156604, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964413.156608, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964413.156610, type 1 (Key), code 65 (F7), value 0
Event: time 1246964413.156611, -------------- Report Sync ------------
Event: time 1246964417.492588, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964417.492592, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964417.492595, -------------- Report Sync ------------
Event: time 1246964417.508611, type 17 (LED), code 0 (NumLock), value 1
Event: time 1246964417.508613, -------------- Report Sync ------------
Event: time 1246964417.668469, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964417.668475, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964417.668478, -------------- Report Sync ------------
Event: time 1246964417.988595, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964417.988602, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964417.988605, -------------- Report Sync ------------
Event: time 1246964418.240460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.240463, -------------- Report Sync ------------
Event: time 1246964418.276459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.276461, -------------- Report Sync ------------
Event: time 1246964418.312459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.312461, -------------- Report Sync ------------
Event: time 1246964418.348457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.348459, -------------- Report Sync ------------
Event: time 1246964418.384458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.384460, -------------- Report Sync ------------
Event: time 1246964418.420459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.420461, -------------- Report Sync ------------
Event: time 1246964418.456457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.456459, -------------- Report Sync ------------
Event: time 1246964418.492457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.492459, -------------- Report Sync ------------
Event: time 1246964418.528459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.528461, -------------- Report Sync ------------
Event: time 1246964418.564459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964418.564461, -------------- Report Sync ------------
Event: time 1246964418.564597, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964418.564600, type 1 (Key), code 65 (F7), value 1
Event: time 1246964418.564602, -------------- Report Sync ------------
Event: time 1246964418.568467, type 17 (LED), code 0 (NumLock), value 0
Event: time 1246964418.568470, -------------- Report Sync ------------
Event: time 1246964418.724597, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964418.724602, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964418.724605, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964418.724607, type 1 (Key), code 65 (F7), value 0
Event: time 1246964418.724609, -------------- Report Sync ------------
Event: time 1246964424.148592, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964424.148597, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964424.148600, -------------- Report Sync ------------
Event: time 1246964424.224609, type 17 (LED), code 0 (NumLock), value 1
Event: time 1246964424.224611, -------------- Report Sync ------------
Event: time 1246964424.388595, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964424.388601, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964424.388604, -------------- Report Sync ------------
Event: time 1246964424.644596, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964424.644603, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964424.644606, -------------- Report Sync ------------
Event: time 1246964424.896459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964424.896461, -------------- Report Sync ------------
Event: time 1246964424.932457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964424.932459, -------------- Report Sync ------------
Event: time 1246964424.968460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964424.968461, -------------- Report Sync ------------
Event: time 1246964425.004459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964425.004460, -------------- Report Sync ------------
Event: time 1246964425.040457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964425.040459, -------------- Report Sync ------------
Event: time 1246964425.076459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964425.076461, -------------- Report Sync ------------
Event: time 1246964425.112459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964425.112461, -------------- Report Sync ------------
Event: time 1246964425.148458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964425.148460, -------------- Report Sync ------------
Event: time 1246964425.172593, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964425.172597, type 1 (Key), code 65 (F7), value 1
Event: time 1246964425.172599, -------------- Report Sync ------------
Event: time 1246964425.176466, type 17 (LED), code 0 (NumLock), value 0
Event: time 1246964425.176468, -------------- Report Sync ------------
Event: time 1246964425.332599, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964425.332605, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964425.332608, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964425.332610, type 1 (Key), code 65 (F7), value 0
Event: time 1246964425.332611, -------------- Report Sync ------------
Event: time 1246964430.368609, type 17 (LED), code 0 (NumLock), value 1
Event: time 1246964430.368612, -------------- Report Sync ------------
Event: time 1246964430.448470, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964430.448474, -------------- Report Sync ------------
Event: time 1246964430.484465, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964430.484468, -------------- Report Sync ------------
Event: time 1246964430.520461, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964430.520464, -------------- Report Sync ------------
Event: time 1246964430.532595, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964430.532599, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964430.532602, -------------- Report Sync ------------
Event: time 1246964430.980597, type 4 (Misc), code 4 (ScanCode), value 700e0
Event: time 1246964430.980603, type 1 (Key), code 29 (LeftControl), value 1
Event: time 1246964430.980608, -------------- Report Sync ------------
Event: time 1246964431.012596, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964431.012602, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964431.012605, -------------- Report Sync ------------
Event: time 1246964431.264471, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.264476, -------------- Report Sync ------------
Event: time 1246964431.300471, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.300476, -------------- Report Sync ------------
Event: time 1246964431.336466, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.336470, -------------- Report Sync ------------
Event: time 1246964431.372485, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.372488, -------------- Report Sync ------------
Event: time 1246964431.408472, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.408476, -------------- Report Sync ------------
Event: time 1246964431.444467, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.444470, -------------- Report Sync ------------
Event: time 1246964431.480501, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.480505, -------------- Report Sync ------------
Event: time 1246964431.516461, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.516464, -------------- Report Sync ------------
Event: time 1246964431.552468, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.552473, -------------- Report Sync ------------
Event: time 1246964431.588480, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.588484, -------------- Report Sync ------------
Event: time 1246964431.624460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.624463, -------------- Report Sync ------------
Event: time 1246964431.660458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.660460, -------------- Report Sync ------------
Event: time 1246964431.696464, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.696467, -------------- Report Sync ------------
Event: time 1246964431.732458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.732460, -------------- Report Sync ------------
Event: time 1246964431.768459, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.768461, -------------- Report Sync ------------
Event: time 1246964431.804458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964431.804460, -------------- Report Sync ------------
Event: time 1246964431.828598, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964431.828602, type 1 (Key), code 65 (F7), value 1
Event: time 1246964431.828604, -------------- Report Sync ------------
Event: time 1246964431.832468, type 17 (LED), code 0 (NumLock), value 0
Event: time 1246964431.832470, -------------- Report Sync ------------
Event: time 1246964431.988595, type 4 (Misc), code 4 (ScanCode), value 700e0
Event: time 1246964431.988601, type 1 (Key), code 29 (LeftControl), value 0
Event: time 1246964431.988603, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964431.988605, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964431.988608, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964431.988610, type 1 (Key), code 65 (F7), value 0
Event: time 1246964431.988611, -------------- Report Sync ------------
Event: time 1246964440.800611, type 17 (LED), code 0 (NumLock), value 1
Event: time 1246964440.800614, -------------- Report Sync ------------
Event: time 1246964440.916463, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964440.916468, -------------- Report Sync ------------
Event: time 1246964440.952463, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964440.952465, -------------- Report Sync ------------
Event: time 1246964440.964469, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964440.964472, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964440.964475, -------------- Report Sync ------------
Event: time 1246964441.172596, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964441.172601, type 1 (Key), code 56 (LeftAlt), value 1
Event: time 1246964441.172605, -------------- Report Sync ------------
Event: time 1246964441.424461, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.424464, -------------- Report Sync ------------
Event: time 1246964441.460458, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.460460, -------------- Report Sync ------------
Event: time 1246964441.496457, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.496459, -------------- Report Sync ------------
Event: time 1246964441.532460, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.532462, -------------- Report Sync ------------
Event: time 1246964441.568467, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.568470, -------------- Report Sync ------------
Event: time 1246964441.604475, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.604480, -------------- Report Sync ------------
Event: time 1246964441.640465, type 1 (Key), code 56 (LeftAlt), value 2
Event: time 1246964441.640469, -------------- Report Sync ------------
Event: time 1246964441.652601, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964441.652607, type 1 (Key), code 65 (F7), value 1
Event: time 1246964441.652609, -------------- Report Sync ------------
Event: time 1246964441.656468, type 17 (LED), code 0 (NumLock), value 0
Event: time 1246964441.656471, -------------- Report Sync ------------
Event: time 1246964441.812597, type 4 (Misc), code 4 (ScanCode), value 700e2
Event: time 1246964441.812603, type 1 (Key), code 56 (LeftAlt), value 0
Event: time 1246964441.812607, type 4 (Misc), code 4 (ScanCode), value 70040
Event: time 1246964441.812609, type 1 (Key), code 65 (F7), value 0
Event: time 1246964441.812610, -------------- Report Sync ------------
Event: time 1246964445.504609, type 17 (LED), code 0 (NumLock), value 1
Event: time 1246964445.504612, -------------- Report Sync ------------
Event: time 1246964446.452594, type 4 (Misc), code 4 (ScanCode), value 700e0
Event: time 1246964446.452600, type 1 (Key), code 29 (LeftControl), value 1
Event: time 1246964446.452603, -------------- Report Sync ------------

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Tue, 7 Jul 2009, Dmitry Torokhov wrote:

> > Are you seeing with a specific keyboard type(s), or is this a general 
> > problem? Could you confirm whether it is specific either to PS2 or USB 
> > keyboards only, or it does happen for both?
> Can't think of any input chnages in this timeftrame that would cause
> such effect. Maybe TTY changes?

By the way, if I understand it correctly, Andi is able to see the very 
same effect even inside X, so tty can be possibly ruled out, right?

Still, I don't see the leaking F key in the evtest output (see the mail 
from yesterday), which is quite puzzling.

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Wed, 8 Jul 2009, Jiri Kosina wrote:

> For completness, Andi sent me the following evtest output off-list 
> yesterday. The LeftAlt being autorepeated at time 1246964430.448470 is a 
> little bit suspicious, as it is preceeded by release ... I don't see how 
> this could be causing the behavior Andi is observing though.
> 
> I am not able to reproduce the problem myself.

Umm, actually, Andi, could you verify whether reverting e7b5c1ef4d changes 
the behavior you are observing in any way?

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

On Thu, Jul 09, 2009 at 01:20:56PM +0200, Jiri Kosina wrote:
> On Tue, 7 Jul 2009, Dmitry Torokhov wrote:
> 
> > > Are you seeing with a specific keyboard type(s), or is this a general 
> > > problem? Could you confirm whether it is specific either to PS2 or USB 
> > > keyboards only, or it does happen for both?
> > Can't think of any input chnages in this timeftrame that would cause
> > such effect. Maybe TTY changes?
> 
> By the way, if I understand it correctly, Andi is able to see the very 
> same effect even inside X, so tty can be possibly ruled out, right?

I see it both in X (with firefox) and on the console (e.g. links
starts doing something)

-Andi
-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

On Thu, Jul 09, 2009 at 01:29:55PM +0200, Jiri Kosina wrote:
> On Wed, 8 Jul 2009, Jiri Kosina wrote:
> 
> > For completness, Andi sent me the following evtest output off-list 
> > yesterday. The LeftAlt being autorepeated at time 1246964430.448470 is a 
> > little bit suspicious, as it is preceeded by release ... I don't see how 
> > this could be causing the behavior Andi is observing though.
> > 
> > I am not able to reproduce the problem myself.
> 
> Umm, actually, Andi, could you verify whether reverting e7b5c1ef4d changes 
> the behavior you are observing in any way?

Tested that, no change unfortunately.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

On Thu, Jul 09, 2009 at 01:20:56PM +0200, Jiri Kosina wrote:
> On Tue, 7 Jul 2009, Dmitry Torokhov wrote:
> 
> > > Are you seeing with a specific keyboard type(s), or is this a general 
> > > problem? Could you confirm whether it is specific either to PS2 or USB 
> > > keyboards only, or it does happen for both?
> > Can't think of any input chnages in this timeftrame that would cause
> > such effect. Maybe TTY changes?
> 
> By the way, if I understand it correctly, Andi is able to see the very 
> same effect even inside X, so tty can be possibly ruled out, right?
> 
> Still, I don't see the leaking F key in the evtest output (see the mail 
> from yesterday), which is quite puzzling.

Everything quite puzzling.

I find it puzzling that people cannot reproduce this. I see this on two
quite different systems (different keyboard, system). Does really nobody
else see it?

I checked now and it happens with both when I switch with Alt-Fx
and Ctrl-Alt-Fx (at least on SUSE the later works from the consoles
too, so i usually use that)

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Thu, 9 Jul 2009, Andi Kleen wrote:

> > Umm, actually, Andi, could you verify whether reverting e7b5c1ef4d changes 
> > the behavior you are observing in any way?
> Tested that, no change unfortunately.

Is there any possibility to connect PS/2 keyboard to those systems, to see 
if it is specific to USB keyboards or whether it is rather generic input 
problem?

On Thu, 9 Jul 2009, Andi Kleen wrote:

> > By the way, if I understand it correctly, Andi is able to see the very 
> > same effect even inside X, so tty can be possibly ruled out, right?
> > 
> > Still, I don't see the leaking F key in the evtest output (see the mail 
> > from yesterday), which is quite puzzling.
> 
> Everything quite puzzling.
> 
> I find it puzzling that people cannot reproduce this. I see this on two
> quite different systems (different keyboard, system). Does really nobody
> else see it?

I really tried, and I don't. Could you please send me your config?

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

On Thu, Jul 09, 2009 at 01:20:56PM +0200, Jiri Kosina wrote:
> On Tue, 7 Jul 2009, Dmitry Torokhov wrote:
> 
> > > Are you seeing with a specific keyboard type(s), or is this a general 
> > > problem? Could you confirm whether it is specific either to PS2 or USB 
> > > keyboards only, or it does happen for both?
> > Can't think of any input chnages in this timeftrame that would cause
> > such effect. Maybe TTY changes?
> 
> By the way, if I understand it correctly, Andi is able to see the very 
> same effect even inside X, so tty can be possibly ruled out, right?
> 

Umm, doesn't SuSE still use legacy keyboard driver instead of evdev in
X? Thomas mentioned recently it still did...

-- 
Dmitry

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Sun, 12 Jul 2009, Dmitry Torokhov wrote:

> > > > Are you seeing with a specific keyboard type(s), or is this a 
> > > > general problem? Could you confirm whether it is specific either 
> > > > to PS2 or USB keyboards only, or it does happen for both?
> > > Can't think of any input chnages in this timeftrame that would cause
> > > such effect. Maybe TTY changes?
> > By the way, if I understand it correctly, Andi is able to see the very 
> > same effect even inside X, so tty can be possibly ruled out, right?
> Umm, doesn't SuSE still use legacy keyboard driver instead of evdev in
> X? Thomas mentioned recently it still did...

It does.

If I understand Andi's problem though, he is able to trigger the problem 
even solely on console, with X not being involved at all.

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

On Mon, Jul 13, 2009 at 09:07:22AM +0200, Jiri Kosina wrote:
> On Sun, 12 Jul 2009, Dmitry Torokhov wrote:
> 
> > > > > Are you seeing with a specific keyboard type(s), or is this a 
> > > > > general problem? Could you confirm whether it is specific either 
> > > > > to PS2 or USB keyboards only, or it does happen for both?
> > > > Can't think of any input chnages in this timeftrame that would cause
> > > > such effect. Maybe TTY changes?
> > > By the way, if I understand it correctly, Andi is able to see the very 
> > > same effect even inside X, so tty can be possibly ruled out, right?
> > Umm, doesn't SuSE still use legacy keyboard driver instead of evdev in
> > X? Thomas mentioned recently it still did...
> 
> It does.
> 
> If I understand Andi's problem though, he is able to trigger the problem 
> even solely on console, with X not being involved at all.

Sorry, I tried again and I can't reproduce the console misbehaviour anymore.
Only see it in X.  I thought I had seen it in links resetting to the first
page, but I don't see that now.

-andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Jiri Kosina]

On Mon, 13 Jul 2009, Andi Kleen wrote:

> > > > > > Are you seeing with a specific keyboard type(s), or is this a 
> > > > > > general problem? Could you confirm whether it is specific either 
> > > > > > to PS2 or USB keyboards only, or it does happen for both?
> > > > > Can't think of any input chnages in this timeftrame that would cause
> > > > > such effect. Maybe TTY changes?
> > > > By the way, if I understand it correctly, Andi is able to see the very 
> > > > same effect even inside X, so tty can be possibly ruled out, right?
> > > Umm, doesn't SuSE still use legacy keyboard driver instead of evdev in
> > > X? Thomas mentioned recently it still did...
> > 
> > It does.
> > 
> > If I understand Andi's problem though, he is able to trigger the problem 
> > even solely on console, with X not being involved at all.
> 
> Sorry, I tried again and I can't reproduce the console misbehaviour anymore.
> Only see it in X.  I thought I had seen it in links resetting to the first
> page, but I don't see that now.

So by total coincidence I reproduced exactly this behavior (*) (while not 
trying to do so) on system running 2.6.24-rc3, with PS/2 keyboard. 

Andi, are you absolutely sure that this appeared between .29 and .30 for 
you?

And if so, is there any chance you could try bisecting it?

(*) have been on virtual desktop where firefox was running, switched to 
console using Ctrl-Alt-F1, and switched back to X using F7. Firefox then 
popped up some dialog about configuring some feature that is triggered by 
pressing F7/AltF7, or whatever. Is that what you are seeing?

-- 
Jiri Kosina
SUSE Labs

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

> Andi, are you absolutely sure that this appeared between .29 and .30 for 
> you?

I only noticed it after the .30 update at least.

> And if so, is there any chance you could try bisecting it?

Ok.

> 
> (*) have been on virtual desktop where firefox was running, switched to 
> console using Ctrl-Alt-F1, and switched back to X using F7. Firefox then 
> popped up some dialog about configuring some feature that is triggered by 
> pressing F7/AltF7, or whatever. Is that what you are seeing?

Yes, I typically see it with the help screen. It also happens with
other applications, not just firefox.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

[ressurecting old thread...]

On Wed, Jul 08, 2009 at 01:20:23PM +0200, Jiri Kosina wrote:
> On Tue, 7 Jul 2009, Dmitry Torokhov wrote:
> 
> > > > One new issue I see in 2.6.30 (I think I saw it in some rc*s) too is 
> > > > that console switch often leaks the Fkey of the console switch into the 
> > > > target console.
> > > > 
> > > > The symptome is that e.g. firefox if it was active and you're switching
> > > > into X displays its help screen (F1 = help). It happens on non X 
> > > > consoles too, but of course the symptoms are different here.
> > > > 
> > > > That's a regression against 2.6.29; I don't remember it in .29
> > > > 
> > > > Haven't investigated code.
> > > 
> > > Are you seeing with a specific keyboard type(s), or is this a general 
> > > problem? Could you confirm whether it is specific either 
> > > to PS2 or USB keyboards only, or it does happen for both?
> > 
> > Can't think of any input chnages in this timeftrame that would cause
> > such effect. Maybe TTY changes?
> 
> For completness, Andi sent me the following evtest output off-list 
> yesterday. The LeftAlt being autorepeated at time 1246964430.448470 is a 
> little bit suspicious, as it is preceeded by release ... I don't see how 
> this could be causing the behavior Andi is observing though.
> 
> I am not able to reproduce the problem myself.
> 

...

> Event: time 1246964431.804460, -------------- Report Sync ------------
> Event: time 1246964431.828598, type 4 (Misc), code 4 (ScanCode), value 70040
> Event: time 1246964431.828602, type 1 (Key), code 65 (F7), value 1
> Event: time 1246964431.828604, -------------- Report Sync ------------
> Event: time 1246964431.832468, type 17 (LED), code 0 (NumLock), value 0
> Event: time 1246964431.832470, -------------- Report Sync ------------
> Event: time 1246964431.988595, type 4 (Misc), code 4 (ScanCode), value 700e0
> Event: time 1246964431.988601, type 1 (Key), code 29 (LeftControl), value 0
> Event: time 1246964431.988603, type 4 (Misc), code 4 (ScanCode), value 700e2
> Event: time 1246964431.988605, type 1 (Key), code 56 (LeftAlt), value 0
> Event: time 1246964431.988608, type 4 (Misc), code 4 (ScanCode), value 70040
> Event: time 1246964431.988610, type 1 (Key), code 65 (F7), value 0
> Event: time 1246964431.988611, -------------- Report Sync ------------

Btw, what kind of keyboard is this? I was playing today but I could not
get my USB keyboards to report release of more than 1 key per report....

-- 
Dmitry

Re: [regression] 2.6.30 leaking keys on console switch

[Andi Kleen]

> Btw, what kind of keyboard is this? I was playing today but I could not
> get my USB keyboards to report release of more than 1 key per report....

The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
to a USB port through it.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

On Wed, Oct 14, 2009 at 09:19:50AM +0200, Andi Kleen wrote:
> > Btw, what kind of keyboard is this? I was playing today but I could not
> > get my USB keyboards to report release of more than 1 key per report....
> 
> The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
> to a USB port through it.
> 

Hmm, I don't think I have one of those... It is curious beast - we have
software autorepeat for HID but PS/2 keyboards do autorepeat
themselves... What will combination do?

-- 
Dmitry

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

On Wed, Oct 14, 2009 at 12:39:28AM -0700, Dmitry Torokhov wrote:
> On Wed, Oct 14, 2009 at 09:19:50AM +0200, Andi Kleen wrote:
> > > Btw, what kind of keyboard is this? I was playing today but I could not
> > > get my USB keyboards to report release of more than 1 key per report....
> > 
> > The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
> > to a USB port through it.
> > 
> 
> Hmm, I don't think I have one of those... It is curious beast - we have
> software autorepeat for HID but PS/2 keyboards do autorepeat
> themselves... What will combination do?
> 

Could you please try this incredibly stupid patch?

-- 
Dmitry


Input: autorepeat debug patch

From: Dmitry Torokhov <dmitry.torokhov@gmail.com>

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---

 drivers/input/input.c |   12 ++++++++++--
 1 files changed, 10 insertions(+), 2 deletions(-)


diff --git a/drivers/input/input.c b/drivers/input/input.c
index 6ba676b..77fa3da 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -137,7 +137,7 @@ static void input_repeat_key(unsigned long data)
 	if (test_bit(dev->repeat_key, dev->key) &&
 	    is_event_supported(dev->repeat_key, dev->keybit, KEY_MAX)) {
 
-		input_pass_event(dev, EV_KEY, dev->repeat_key, 2);
+		input_pass_event(dev, EV_KEY, dev->repeat_key, 200);
 
 		if (dev->sync) {
 			/*
@@ -208,7 +208,15 @@ static void input_handle_event(struct input_dev *dev,
 		if (is_event_supported(code, dev->keybit, KEY_MAX) &&
 		    !!test_bit(code, dev->key) != value) {
 
-			if (value != 2) {
+			if (value == 2 || value == 200) {
+				/* autorepeat */
+				if (!test_bit(code, dev->key)) {
+					printk(KERN_ERR "%s autorepeat for released key %d\n",
+						value == 200 ? "software" : "hardware", code);
+					break;
+				}
+				value = 2;
+			} else {
 				__change_bit(code, dev->key);
 				if (value)
 					input_start_autorepeat(dev, code);

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

On Thu, Jul 16, 2009 at 11:24:06AM +0200, Andi Kleen wrote:
> > Andi, are you absolutely sure that this appeared between .29 and .30 for 
> > you?
> 
> I only noticed it after the .30 update at least.
> 
> > And if so, is there any chance you could try bisecting it?
> 
> Ok.
> 
> > 
> > (*) have been on virtual desktop where firefox was running, switched to 
> > console using Ctrl-Alt-F1, and switched back to X using F7. Firefox then 
> > popped up some dialog about configuring some feature that is triggered by 
> > pressing F7/AltF7, or whatever. Is that what you are seeing?
> 
> Yes, I typically see it with the help screen. It also happens with
> other applications, not just firefox.
> 

I don't really see the issue here, except for maybe X being stupid - it
seems that it processes release events as press+release in cases when it
did not see the press event.

Start up xev, switch to text console and then back to X (don't release
the CTRL+ALT+Fx combo). Let X switch fully and then release the combo.
Observe the 2 events reported by xev for Fx key. You may try running
evtest in other window and observe that there are no down events at the
time of 2 xev events, only autorepeats and release event (and
autorepeats are ignored by X who does it's own software autorepeat).

-- 
Dmitry

Re: [regression] 2.6.30 leaking keys on console switch

[H. Peter Anvin]

On 10/14/2009 12:39 AM, Dmitry Torokhov wrote:
> On Wed, Oct 14, 2009 at 09:19:50AM +0200, Andi Kleen wrote:
>>> Btw, what kind of keyboard is this? I was playing today but I could not
>>> get my USB keyboards to report release of more than 1 key per report....
>>
>> The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
>> to a USB port through it.
>>
> 
> Hmm, I don't think I have one of those... It is curious beast - we have
> software autorepeat for HID but PS/2 keyboards do autorepeat
> themselves... What will combination do?
> 

Autorepeat on PS/2 keyboards is a mode setting.  I would guess the HID
adapter would turn it off if it is halfway sane.

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.

Re: [regression] 2.6.30 leaking keys on console switch

[Pavel Machek]

On Wed 2009-10-14 00:39:28, Dmitry Torokhov wrote:
> On Wed, Oct 14, 2009 at 09:19:50AM +0200, Andi Kleen wrote:
> > > Btw, what kind of keyboard is this? I was playing today but I could not
> > > get my USB keyboards to report release of more than 1 key per report....
> > 
> > The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
> > to a USB port through it.
> > 
> 
> Hmm, I don't think I have one of those... It is curious beast - we have
> software autorepeat for HID but PS/2 keyboards do autorepeat
> themselves... What will combination do?

I always wondered how that's supposed to work... on loaded system. As
linux is not realtime system, it can't, afaict.


-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Re: [regression] 2.6.30 leaking keys on console switch

[Dmitry Torokhov]

On Mon, Oct 19, 2009 at 11:38:36PM +0200, Pavel Machek wrote:
> On Wed 2009-10-14 00:39:28, Dmitry Torokhov wrote:
> > On Wed, Oct 14, 2009 at 09:19:50AM +0200, Andi Kleen wrote:
> > > > Btw, what kind of keyboard is this? I was playing today but I could not
> > > > get my USB keyboards to report release of more than 1 key per report....
> > > 
> > > The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
> > > to a USB port through it.
> > > 
> > 
> > Hmm, I don't think I have one of those... It is curious beast - we have
> > software autorepeat for HID but PS/2 keyboards do autorepeat
> > themselves... What will combination do?
> 
> I always wondered how that's supposed to work... on loaded system. As
> linux is not realtime system, it can't, afaict.
> 

What exactly requires realtime? Surely not autorepeat...

-- 
Dmitry

Re: [regression] 2.6.30 leaking keys on console switch

[Pavel Machek]

On Wed 2009-10-21 01:15:30, Dmitry Torokhov wrote:
> On Mon, Oct 19, 2009 at 11:38:36PM +0200, Pavel Machek wrote:
> > On Wed 2009-10-14 00:39:28, Dmitry Torokhov wrote:
> > > On Wed, Oct 14, 2009 at 09:19:50AM +0200, Andi Kleen wrote:
> > > > > Btw, what kind of keyboard is this? I was playing today but I could not
> > > > > get my USB keyboards to report release of more than 1 key per report....
> > > > 
> > > > The log was with a PS/2<->USB adapter, with a PS/2 keyboard connected
> > > > to a USB port through it.
> > > > 
> > > 
> > > Hmm, I don't think I have one of those... It is curious beast - we have
> > > software autorepeat for HID but PS/2 keyboards do autorepeat
> > > themselves... What will combination do?
> > 
> > I always wondered how that's supposed to work... on loaded system. As
> > linux is not realtime system, it can't, afaict.
> 
> What exactly requires realtime? Surely not autorepeat...

Well, ps/2 does not provide timestamps, neither does USB AFAICT. So if
you hit say interrupt latency, you'll repeat keys one too many times
or something.

[But if X does autorepeat in software, that's even worse. That means
that scheduling latency can cause autorepeat glitches, and that's
probably what I'm seeing here. And I haaaaaaaaaaaaaaate when my
tyyyyyyping goes like this just because I have background load.]

									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html