syndaemon sometimes fails to disable the touchpad

[bitsent]

Binary package hint: xserver-xorg-input-synaptics

syndaemon does not always disable the touchpad when a key is pressed.

Steps to reproduce:

  * Turn on SHMConfig in /etc/X11/xorg.conf and restart X.
  * $ killall syndaemon
  * $ syndaemon -i 3 -k -t
  * Run another application such as gedit and press a single key (not a modifier key) while watching the syndaemon window. It should print 'Disable', wait 3 seconds, and then 'Enable'. Most of the time this works as expected, but a small percentage of the time nothing new is displayed in the syndaemon window (the gedit window displays the key as expected). The timing seems important; if the key is pressed between 1 and 3 seconds after syndaemon displays 'Enable' then the key is more likely to be missed by syndaemon.

This problem sounds minor but has been a major source of aggravation while typing on my laptop because a few times a day I find myself typing in another part of the document or in a different application altogether. Depending on the application which is incorrectly receiving the key presses, the consequences could be serious.

This is on Ubuntu 8.04 and xserver-xorg-input-synaptics 0.14.7~git20070706-mactel1, but I believe this problem has existed for many months. The hardware is an Apple MacBook 2nd generation.

[Scott Thatcher (thatcher)]

This problem is even more apparent to me when running syndaemon -i 0.5 -k

About every third keystroke doesn't trigger the Disable/Enable. While typing, for example, "fffffff...", quickly, it is still very possible to use the trackpad, although it is jumpy. I know syndaemon used to work correctly for me--I remember testing it when I first discovered the command. That was back in Gutsy. I've only recently started noticing this annoying behavior, although I switched to Hardy back in May. Could something else have changed recently?

My hardware is an HP tx1000 laptop, and I get this:
 [ 38.532819] Synaptics Touchpad, model: 1, fw: 6.3, id: 0x180b1, caps: 0xa04713/0x200000
 [ 38.593153] input: SynPS/2 Synaptics TouchPad as /devices/platform/i8042/serio1/input/input9
in my dmesg output.

[William Grant (wgrant)]

This will probably be fixed with the XRecord changes upstream.

[bitinerant (bitinerant)]

This problem has not been fixed in Intrepid.

[Mike (bild85)]

duplicates of this bug are more than 5 years old, and at least for me it causes significant annoyances while typing. In my opinion it should not be classified as "Low" importance. Can it get more attention please?

Also, how can it be both "Triaged" and "Unassigned"? Is this a bug within Launchpad?

[Mike (bild85)]

I wrote to Peter Osterlund, still listed on the Man pages as the author (but he is no longer the maintainer of it.) This is his response:

"I was not aware of this problem (I don't normally use syndaemon), but I was able to reproduce the problem on my fedora machine. The problem is that the default keyboard polling interval is now 200ms, so keypresses shorter than that risk being missed.

There is however an undocumented switch -m to change the polling interval, so try running "syndaemon -i 0.5 -k -m 20", which sets the polling interval to 20ms.

I guess this change was made to conserve battery power on laptops. It would be better if syndaemon didn't have to poll at all, but I don't know how to implement that, or if it is even possible.

Btw, I am no longer maintaining this driver, but feel free to forward my response to whoever is maintaining the driver these days.

Best regards,
Peter Osterlund"

I did test it and while the "-m 20" didn't seem to make much difference, changing it to "-m 2" did resolve the problem. However, as Peter mentioned, it may be too much on the system to poll every 2ms.

[Claudio Pighin (cpighin)]

Thank you itimike for your activity on trying to solve this issue.

I support your request to experts to find a good solution since it's very frustrating to work in such a way and it may be a reason to continue to use Windows instead Linux!

Claudio :)

[Mike (bild85)]

Also affects Ubuntu Natty Narwhal Alpha 3. This is a problem because 11.04 has "Enable mouse clicks with touchpad" pre-selected on a standard installation.

[Barak (barak-naveh)]

I've installed Ubuntu 10.10 on a MacBook Pro and i'm getting hit by this bug.

It's terrible to work with it, so I must disable mouse clicks with touchpad, or use and external keyboard.

[Mike (bild85)]

"it's very frustrating to work in such a way and it may be a reason to continue to use Windows instead Linux!"

Is it possible to add this to the 100 papercuts project, or at least be raised from low importance?

[disappearedng (disappearedng)]

Ok this is a really big problem for a lot of users. Please mark the importance to a level higher.

[Goofy (gfiala)]

Thx, i have an Acer NB and i'am not sure if/why my bugs has been marked as a duplicate of this one, since the problem is even worse than this - it does not only fail to disable, it also is not able to set the sensitivity of the touchpad at all, can't find the reference to my bug anymore, but it has been rated as a synaptic-driver problem, which seems also not be fixed until now.

[Mike (bild85)]

Hi Goofy. I marked your bug (#57764) a dupe of this primarily because of the following statements:
"When i change the settings in kcm-touchpad module and check with "synclient -l" the settings are indeed changed, however, they have no effect..."

and:
"Under Windows the pad works as expected, but under kubuntu its driving my wife crazy because she always hits the pad while typing text with quite unexpected results ;-)"

You mentioned other symptoms such as the inability to define FingerHigh and FingerLow boundaries, which may be a separate bug if you're still experiencing it.

[Jonas Petersson (catellie)]

@itismike - thanks for the hint in #743352

However, I'm a but stumped as I've failed to figure out the correct way to enable SHMConfig in Natty. Things appear to have changed both inside the bcm5974 and with the .fdi files recently so none of the tricks I know allow synclient to connect.

In the old version of bcm5974 I simply adjusted a compile time flag, but it seems to be gone now. Note: What I'm after is really to disable all kinds of tapping that turn into clicks - I just want to the "real" click and the apply 1, 2 or 3 fingers to select actual mouse button. I don't mind using syndaemon if that is now considered a better route, though.

Any and all hints appreciated.

(OT again: I can confirm that the bcmwl 11+ channels being disabled in the driver is the problem - adjugstin my old channel 13 down to 6 made it perfectly visible as in the previous driver release.)

[Mike (bild85)]

That's odd. I had no problem disabling the "touch" in the touchpad settings after I noticed it was enabled by default from a clean Natty install. Sounds like a separate bug if you can't disable it through the GUI.

[Jonas Petersson (catellie)]

@itismike
Well, it might be sort of related. I'm pretty sure it worked when I booted from the live CD (as did the screen backlight keys, but I'm not sure it was using nvidia at the time), so something is different from that state and it may well be that the difference is the SHMConfig flag. If so, it makes fair sense that without SHMConfig the syndaemon will be unable to adjust the touchpad.

[Goofy (gfiala)]

In the meantime i also spent some time on analysing, and found out,
that
Option "SHMConfig" "On" (not "true")
solved the issue of not beeing able to adjust fingerHigh/fingerLow with
synclient for me.
The sensitivity settings one can make in the kde-gui-touchpad are still not
useful and when the notebook heats up, one might have to change sensitivity
again. The SHMConfig flag and some related settings have been deactivated by the
update process, according to the comments inserted in xorg.conf due to HAL things.

...so it's now just snd-hda-intel (mic) that forces me to use Windows.

Zitat von Jonas Petersson <email address hidden>:

> @itismike
> Well, it might be sort of related. I'm pretty sure it worked when I booted
> from the live CD (as did the screen backlight keys, but I'm not sure it was
> using nvidia at the time), so something is different from that state and it
> may well be that the difference is the SHMConfig flag. If so, it makes fair
> sense that without SHMConfig the syndaemon will be unable to adjust the
> touchpad.
>
> --
> You received this bug notification because you are a direct subscriber
> of a duplicate bug (577764).
> https://bugs.launchpad.net/bugs/240738
>
> Title:
> syndaemon sometimes fails to disable the touchpad
>
> Status in “xserver-xorg-input-synaptics” package in Ubuntu:
> Confirmed
>
> Bug description:
> Binary package hint: xserver-xorg-input-synaptics
>
> syndaemon does not always disable the touchpad when a key is pressed.
>
> Steps to reproduce:
>
> * Turn on SHMConfig in /etc/X11/xorg.conf and restart X.
> * $ killall syndaemon
> * $ syndaemon -i 3 -k -t
> * Run another application such as gedit and press a single key (not a
> modifier key) while watching the syndaemon window. It should print
> 'Disable', wait 3 seconds, and then 'Enable'. Most of the time this works as
> expected, but a small percentage of the time nothing new is displayed in the
> syndaemon window (the gedit window displays the key as expected). The timing
> seems important; if the key is pressed between 1 and 3 seconds after
> syndaemon displays 'Enable' then the key is more likely to be missed by
> syndaemon.
>
> This problem sounds minor but has been a major source of aggravation
> while typing on my laptop because a few times a day I find myself
> typing in another part of the document or in a different application
> altogether. Depending on the application which is incorrectly
> receiving the key presses, the consequences could be serious.
>
> This is on Ubuntu 8.04 and xserver-xorg-input-synaptics
> 0.14.7~git20070706-mactel1, but I believe this problem has existed for
> many months. The hardware is an Apple MacBook 2nd generation.
>
> To unsubscribe from this bug, go to:
>
https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-input-synaptics/+bug/240738/+subscribe
>
>

[zeimm (zeimm)]

Hello,

I am running ubuntu 10.10 on a Dell Inspiron laptop n 5010. I get exactly the same touchpad problem as described by everyone else even though I've ticked the Disable touchpad while typing option.
I also think this is an extremely frustrating issue and the importance of resolving it must be raised.

Best regards,
Zea

[Mike (bild85)]

I am puzzled why there are only ten people affected by this bug. It drives me absolutely crazy and I am forced to disable touchpad clicks. How can so few people be complaining? Is it hardware-specific? Do other laptop touchpads disable themselves in the hardware rather than relying on the synaptics driver?

[Barak (barak-naveh)]

I'm sure it affects many more people.

[bitinerant (bitinerant)]

I think the problem is that most people never actually figure out what is happening well enough to be able to locate this bug report. I agree--it must be affecting many more people. As well as the MacBook mentioned at the top, it also affects both of our two other laptops. That is 100% of our laptops which are from 3 different manufacturers. On one laptop, we've had to disable tap-to-click, making the touchpad less useful.

[bitinerant (bitinerant)]

There are also 18 people affected by the various duplicates of this bug, but it's hard to know how many of those are unique.

[Mike (bild85)]

that's a very good point bitinerant. I think I'll check to see if a feature-request exists in Launchpad to add those totals together :)

[Mike (bild85)]

Here is the bug for summing the number of people affected by a bug and its duplicates. I realize this doesn't belong here but figured others might want to see it or mark themselves as affected :)
https://bugs.launchpad.net/launchpad/+bug/678090

[spbrereton (simon-brereton)]

Well, I decided to add myself as this is seriously annoying. I have a MacBook Pro with a HUGE trackpad and anytime I go near a the y, u, 6 or 7 keys the cursor moves and my text ends up being inserted elsewhere. Othertimes it will randomly copy a block of text and then the next key press deletes/replaces said block.

I've tried all the hacks. I have syndaemon set to run at login without much effect.

[HS Lin (lin-ifa)]

I believe there are a lot more people affected. This is the only reason that I am still using OS X since I got the MCP about 9 months ago.

I also think this is a problem affecting all the laptops with large touchpad. Before I bought the MBP, I tried the HP Envy 15, with exactly the same problem. I also see that the new Dell XPS 15z Laptop also has this problem.

Hope the Linux developer community can get this fixed soon.

[Rainer Rohde (rainer-rohde)]

This affects me as well on my MacBook Pro.

[Oliver Joos (oliver-joos)]

I just marked another 3 duplicates and agree that many people could be affected.

I tested a Samsung SF510 with an Elantech touchpad where this bug really hurts: its palm detection is very unreliable. Finger clicks are detected if a palm is 3 mm *ABOVE* the touchpad surface! Even worse: the clickable left/right buttons are part of the touchpad! Their surface does move the pointer too and makes them nearly unusable. So users are forced to enable tapping to click, staggering again into this bug while typing.

Many Samsung laptops or Asus EeePCs have Elantech touchpads and might be affected. Fixing this bug could ease such problems with bad touchpad drivers or hardware.

@admin: please consider to raise the importance of this bug.

[Oliver-joos-freenet (oliver-joos-freenet)]

I use Ubuntu 10.04.2 LTS with xorg 1:7.5+5ubuntu1 and xserver-xorg-input-synaptics 1.2.2-1ubuntu4

To reproduce start a terminal and execute
1. killall syndaemon
2. syndaemon -i 4.0 -k
3. then press keys and move pointer with touchpad at once

Two things go wrong:
* syndaemon writes "Disable" and "Enable" to stdout, but
  enables not always 4.0 secs after the last keypress.
* quite often the pointer still moves although syndaemon
  says it is disabled.

Details have been well explained by Stanley Sokolow as comment #20 on http://launchpad.net/bugs/27541 which is one of many duplicates of http://launchpad.net/bugs/240738. The low default timeout of 0.5 secs in Ubuntu makes it even worse and seems a workaround for http://launchpad.net/bugs/801763

Due to recent diversity of touchpad hardware this bug gains publicity. I found that Macbooks, newer Samsung laptops and Asus EeePCs are among the affected systems. See my findings with Elantech touchpads in comment #27 of http://launchpad.net/bugs/240738

[Mike (bild85)]

Not to toot my own horn here, but I contacted Peter Osterlund, the author of the syndaemon man page, and his response suggests that it isn't just a simple change of the default timing as suggested in comment 20 of the aforementioned bug:
https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-input-synaptics/+bug/240738/comments/5

[Peter Hutterer (peter-hutterer)]

Did you try using the record extension instead of polling?

syndaemon -R -i 4.0 -k

[N,N-dimethyl (tryptamine)]

Could you please expand on how palm detection is unreliable? I agree that setting a two second delay sucks as a general cure for wandering hands and fingers, but as far as I know the PalmMinZ/PalmMinWidth (I think those are correct) and general touch sensitivity settings still work--or are they the underlying issue now? The touchpad's detection above the surface is normal behavior, but it is a settings failure that causes it to not filter out these very light changes in capacitance as clicks. Turning syndaemon or just palm checking on or off without specifying these values for the particular device, as a few of you have mentioned, probably isn't going to help.

I don't know how easy or difficult it would be to actually code, but I think a few sliders attached to a small graphical control system to change these sensitivity values would be the most acceptable solution, as touchpads seem to vary entirely too much between software and hardware implementations for a single setting to work for everyone (per ALPS and Synaptics' page, they actually intend for OEM's to customize their devices, although how much of this lies in the hard- and software realms I can't say).

[N,N-dimethyl (tryptamine)]

I'm terribly sorry for the dumb double-post, but I forgot to specify that these settings are located in xorg.conf's input section or in a more flexible xorg.conf.d rule file. This could also probably be accomplished with an init rule, although that seems a bit heavy-handed, especially if the target users affected by this problem are in general not going to be wholly comfortable modifying init rules directly (myself included, for the most oart), or if the system is configured with a kludge of several daemons/programs that can control or even compete over these values (that I perceive most DE's as falling under this header is reason I don't like to change rules directly--I'm not sure how valid it is :).

[Tim Kuijsten (kuijsten)]

I'm also affected by this seemingly small, but extremely annoying bug and it took me a couple of days before finding this report.

My MacBook 5,1 and Asus Eee PC are both affected by it and it already made me upgrade to Oneiric in the hope it was fixed and then back to OS X before giving it a new try. I'm wondering why nothing is mentioned on the Macbook pages (https://help.ubuntu.com/community/MacBook5-1/Natty).

Adding the earlier mentioned undocumented -m option to syndaemon (-m 20) makes the situation a lot better (although I'm not sure how this affects battery power) together with the new sensitivity slider in Oneiric it might get pleasant to use my Macbook touchpad in Ubuntu once Oneiric final is out ;-)

On a side note, it might increase the likelyhood someone finds this bug by changing the subject to contain what is in the english gui, on Natty this is "Disable touchpad while typing".

ps. does someone know how to add this -m option to syndaemon on every login? I grepped on syndaemon but didn't find anything in /etc or ~, also nothing in "Startup Applications Preferences".

[Bugs-freedesktop-org-20-kuijsten (bugs-freedesktop-org-20-kuijsten)]

Using the undocumented option "-m 20" on syndaemon, as mentioned in LP #240738, made the situation a lot better for me. At least most keypresses are catched now.

[Bugs-freedesktop-org-20-kuijsten (bugs-freedesktop-org-20-kuijsten)]

(In reply to comment #2)
> Did you try using the record extension instead of polling?
>
> syndaemon -R -i 4.0 -k

If you press a key real quick it still misses it sometimes (apart from the fact that not being able to click for 4 seconds after the last letter you typed is very annoying by itself).

I experimented with this undocumented switch -m to change the polling interval and running "syndaemon -R -i 0.5 -k -m 20", which changes the default polling interval of 200ms to 20ms, results in no more noticeable key presses being missed.

[Peter Hutterer (peter-hutterer)]

If you use -R, the -m argument is ignored.

[N,N-dimethyl (tryptamine)]

Tim, try this page: https://help.ubuntu.com/community/UbuntuBootupHowto (I don't want to state something incorrect :) ).

I'm still a little confused as to why using syndaemon is necessary at all, though, when synclient doesn't seem to be affected by whether it's running or not and has many more useful options for handling this rather than a timeout. You guys are trying to avoid palm touches registering when you're typing on a relatively small keyboard/on a netbook-sized device, yes? I can post my xorg.conf for my Aspire One D260 with good, palm-touch-less settings if anyone would like.

[Mike (bild85)]

tryptamine, I'm sure you are trying to be helpful but I'm not sure you have a good understanding of this bug. If it isn't affecting you then perhaps your hardware has a built-in system that handles accidental touchpad clicks automatically. Consider yourself fortunate. Maybe reading some of the duplicates will give you a better understanding of how others are experiencing the core problem.

[Peter Hutterer (peter-hutterer)]

(In reply to comment #3)
> Using the undocumented option "-m 20" on syndaemon,

commit 72d5b4886927aee5fbc871b5c3d0300be92d8ecc
Author: Peter Hutterer <email address hidden>
Date: Thu Jul 28 10:43:38 2011 +1000

    man: document syndaemon -m switch

[Peter Hutterer (peter-hutterer)]

Closing as fixed as per comment #4. Since -m 20 is ignored if -R is given, I can assume that -R is enough to fix the issue. GNOME also uses the -R flag nowadays, see https://bugzilla.gnome.org/show_bug.cgi?id=639623

Long story short, any polling suffers from race condition, so this cannot be fixed in normal mode.

[Mike (bild85)]

Thanks Peter! I did't get notification of your updates to this bug but just upgraded to 12.04 LTS and can confirm that the trackpad is successfully disabled while typing! Thank you so much!!!
Plus I noticed I can click-and-hold with one finger while sliding another finger for click-and-drag functionality just like in OS X. We've come a long way :)

[Timo Aaltonen (tjaalton)]

closing as fixed