touchpad overwhelms i8042 with int 12

[Rod Davison]

This bug usually manifests itself as a "sticky keyboard", but the keyboard is fine.

1) if I disable the "Internal Pointer Device" in the bios of my laptop, the bug goes away completely.
2) by doing "cat /proc/interrupts | grep i8042" I can watch the interrupt count on int 12 climbing. Even a very light touch on the trackpad (too light to even cause the cursor to move), will cause about 600 interrupts. A 1/2 or 1/4 second touch will generate ~600 interrupts and take about 2 to 3 seconds to process. Any cursor movement using the trackpad prevents any typing for at least 2 seconds.
3) I can reproduce the problem 100% by touching the trackpad while typing. Any typing is delayed or lost while the touchpad is being touched.
4) The trackpad and the keyboard both share the i8042, so it seems pretty clear that problem is around here somewhere.

PS. My laptop is a Toshiba Portege M300. I'm running jaunty, but I've had the problem in feist and hardy too.
> lsb_release -rd
Description: Ubuntu 9.04
Release: 9.04

For those of us afflicted by this bug it makes the system pretty difficult to use.

http://ubuntuforums.org/showthread.php?t=772773

[Victor Vargas (kamus)]

Have you tried to reproduce this issue in latest version of Ubuntu Karmic?

[Rod Davison (roddavison)]

No, I haven't. I'm downloading alpha-6 as I type this. I'll give it a try. Has any work been done on the touchpad driver between Jaunty and Karmic?

[Dmitry Torokhov (dtor)]

Hmm, Protege's keyboard controller indeed can't handle full-rate data stream from synaptics touchpad and we even have DMI workaround in the driver. Please verify that you ave Synaptics touchpad (vs ALPS, look in /proc/bus/input/devices) and if it is synaptics then psmouse.rate=40 module option should help. If it does help then please send me your dmidecode data.

[Rod Davison (roddavison)]

/proc/bus/input/devices has this section:
I: Bus=0011 Vendor=0002 Product=0007 Version=88b1
N: Name="SynPS/2 Synaptics TouchPad"
P: Phys=isa0060/serio1/input0
S: Sysfs=/devices/platform/i8042/serio1/input/input8
U: Uniq=
H: Handlers=mouse1 event8
B: EV=b
B: KEY=6420 0 70000 0 0 0 0 0 0 0 0
B: ABS=11000003

(no reference to ALPS)

Once I found this, I found this incantation:
"sudo modprobe -r psmouse"
and
"sudo modprobe psmouse rate=40"

Having done this, I can now type and mouse at the same time. No lost keyboard events! Whoo, hoo!

Thanks Dmitry. That seems to have done it.

I've pulled the following section out of the dmidecode output. Let me know if you need more of it.

Handle 0x00B0, DMI type 21, 7 bytes
Built-in Pointing Device
 Type: Touch Pad
 Interface: PS/2
 Buttons: 2

[Rod Davison (roddavison)]

Just for the record, I tried playing around with the psmouse.rate=?? option and it seems that 60 is the sweet spot. At 40 there is a noticeable, but slight lag on the trackpad movement. At 60 it seems a bit better, and the mouse and keyboard still seem to be simultaneously operable.

[Dmitry Torokhov (dtor)]

;) The Synaptics hardware supports only 2 modes of reporting when in absolute mode - 40 and 80 packets per second. The driver will set the rate to 80 if you request >= 80 and 40 otherwise.

BTW, I need the entire dmidecode oputput (sans serial numbers/asset tags). You may mail it directly to <email address hidden>

Thanks!

[Dmitry Torokhov (dtor)]

The following patch is scheduled for inclusion in mainline. I would recommend Ubuntu team to pick it up for your kernel update as well, OTOH boxes that would benefit from this fix are pretty old.

[Bryce Harrington (bryce)]

Thanks for attaching a patch for this issue.

Since the patch is against the linux kernel, I'm reassigning the bug report to linux so the kernel team can evaluate.

[Leann Ogasawara (leannogasawara)]

Considering the kernel is now frozen for Karmic, it's likely this will have to wait and be fixed via a Karmic Stable Release Update. Will approve the nomination and get it on our todo list. Thanks!

[Tim Gardner (timg-tpi)]

http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-karmic.git;a=commit;h=9d1049b97d302c0970ffc6a5470cc584d741cf16

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux - 2.6.32-2.2

---------------
linux (2.6.32-2.2) lucid; urgency=low

  [ Andy Whitcroft ]

  * install the full changelog with the binary package
  * changelog -- explicitly note rebases and clean history
  * reinstate armel.mk with no flavours
    - LP: #449637
  * [Upstream] block: silently error unsupported empty barriers too
    - LP: #420423
  * [Config] udate configs following karmic resync
  * [Config] update ports configs following karmic resync
  * [Upstream] lirc -- follow removal of .id element

  [ Colin Watson ]

  * Use section 'admin' rather than 'base'
  * Add more e100 firmware to nic-modules
    - LP: #451872
  * Add qla1280 firmware to scsi-modules
    - LP: #381037

  [ John Johansen ]

  * SAUCE: AppArmor: Set error code after structure initialization.
    - LP: #427948
  * SAUCE: AppArmor: Fix off by 2 error in getprocattr mem allocation
    - LP: #446595
  * SAUCE: AppArmor: Fix mediation of "deleted" paths

  [ Kees Cook ]

  * SAUCE: [x86] fix report of cs-limit nx-emulation
    - LP: #454285

  [ Leann Ogasawara ]

  * SAUCE: (drop after 2.6.31) input: Add support for filtering input
    events
    - LP: #430809
  * SAUCE: (drop after 2.6.31) dell-laptop: Trigger rfkill updates on wifi
    toggle switch press
    - LP: #430809

  [ Luke Yelavich ]

  * SAUCE: Add sr_mod to the scsi-modules udeb for powerpc
  * [Config] Add sd_mod to scsi-modules udeb for powerpc

  [ Mario Limonciello ]

  * SAUCE: Update to LIRC 0.8.6
    - LP: #432678
  * SAUCE: dell-laptop: Store the HW switch status internally rather than
    requerying every time
    - LP: #430809
  * SAUCE: dell-laptop: Blacklist machines not supporting dell-laptop
    - LP: #430809

  [ Stefan Bader ]

  * [Upstream] acerhdf: Limit modalias matching to supported boards
    - LP: #435958

  [ Tim Gardner ]

  * [Upstream] i915: Fix i2c init message
    - LP: #409361
  * [Config] Add sym53c8xx.ko to virtual sub-flavour
    - LP: #439415
  * [Config] Add d101m_ucode.bin to d-i/firmware/nic-modules
    - LP: #439456
  * [Config] Set default I/O scheduler back to CFQ for desktop flavours
    - LP: #381300
  * SAUCE: Created MODULE_EXPORT/MODULE_IMPORT macros
    - LP: #430694
  * SAUCE: Use MODULE_IMPORT macro to tie intel_agp to i915
    - LP: #430694
  * [Config] CONFIG_GFS2_FS_LOCKING_DLM=y
    - LP: #416325
  * SAUCE: Fix MODULE_IMPORT/MODULE_EXPORT
    - LP: #430694
  * SAUCE: Raise the default console 'quiet' level to 2
  * [Config] CONFIG_X86_PAT=y
  * [Config] Add armel arch to linux-libc-dev arches.
    - LP: #449637
  * [Config] CONFIG_X86_MCE
  * [Upstream] (drop after 2.6.31) Input: synaptics - add another Protege
    M300 to rate blacklist
    - LP: #433801

  [ Upstream Kernel Changes ]

  * sgi-gru: Fix kernel stack buffer overrun, CVE-2009-2584
  * drm/i915: Fix FDI M/N setting according with correct color depth
    - LP: #416792

linux (2.6.32-1.1) lucid; urgency=low

  [ Andy Whitcroft ]

  * rebase to v2.6.32-rc3
  * [Config] update configs following rebase to 2.6.32-rc3
  * [Config] update ports configs following rebase to 2.6.32-rc3
  * AppArmor -- fix pstrace_may_access rename
  * staging/android -- disa...

[Martin Pitt (pitti)]

Accepted linux into karmic-proposed, the package will build now and be available in a few hours. Please test and give feedback here. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

[Alberto Milone (albertomilone)]

@Rod
Can you test the package in -proposed, please?

[Rod Davison (roddavison)]

I just grabbed the latest "proposed" updates and the problem is resolved.

Fixed!

Rod

On Tue, Nov 10, 2009 at 8:13 AM, Alberto Milone <
<email address hidden>> wrote:

> @Rod
> Can you test the package in -proposed, please?
>
> --
> touchpad overwhelms i8042 with int 12
> https://bugs.launchpad.net/bugs/433801
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “linux” package in Ubuntu: Fix Released
> Status in “linux” source package in Karmic: Fix Committed
>
> Bug description:
> This bug usually manifests itself as a "sticky keyboard", but the keyboard
> is fine.
>
> 1) if I disable the "Internal Pointer Device" in the bios of my laptop, the
> bug goes away completely.
> 2) by doing "cat /proc/interrupts | grep i8042" I can watch the interrupt
> count on int 12 climbing. Even a very light touch on the trackpad (too light
> to even cause the cursor to move), will cause about 600 interrupts. A 1/2 or
> 1/4 second touch will generate ~600 interrupts and take about 2 to 3 seconds
> to process. Any cursor movement using the trackpad prevents any typing for
> at least 2 seconds.
> 3) I can reproduce the problem 100% by touching the trackpad while typing.
> Any typing is delayed or lost while the touchpad is being touched.
> 4) The trackpad and the keyboard both share the i8042, so it seems pretty
> clear that problem is around here somewhere.
>
> PS. My laptop is a Toshiba Portege M300. I'm running jaunty, but I've had
> the problem in feist and hardy too.
> > lsb_release -rd
> Description: Ubuntu 9.04
> Release: 9.04
>
> For those of us afflicted by this bug it makes the system pretty difficult
> to use.
>
> http://ubuntuforums.org/showthread.php?t=772773
>

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux - 2.6.31-15.50

---------------
linux (2.6.31-15.50) karmic-proposed; urgency=low

  [ Kees Cook ]

  * SAUCE: Fix nx_enable reporting
    - LP: #454285

linux (2.6.31-15.49) karmic-proposed; urgency=low

  [ Benjamin Herrenschmidt ]

  * [Upstream] (drop after 2.6.31) usb-storage: Workaround devices with
    bogus sense size
    - LP: #446146

  [ John Johansen ]

  * SAUCE: AppArmor: AppArmor wrongly reports allow perms as denied
    - LP: #453335
  * SAUCE: AppArmor: Policy load and replacement can fail to alloc mem
    - LP: #458299
  * SAUCE: AppArmor: AppArmor fails to audit change_hat correctly
    - LP: #462824
  * SAUCE: AppArmor: AppArmor disallows truncate of deleted files.
    - LP: #451375

  [ Kees Cook ]

  * SAUCE: [x86] fix report of cs-limit nx-emulation
    - LP: #454285

  [ Scott James Remnant ]

  * Revert "SAUCE: trace: add trace_event for the open() syscall"
  * SAUCE: trace: add trace events for open(), exec() and uselib()
    - LP: #462111

  [ Stefan Bader ]

  * SAUCE: Fix sub-flavour script to not stop on missing directories
    - LP: #453073

  [ Tim Gardner ]

  * [Upstream] (drop after 2.6.31) Input: synaptics - add another Protege
    M300 to rate blacklist
    - LP: #433801

  [ Upstream Kernel Changes ]

  * PM: Make warning in suspend_test_finish() less likely to happen
    - LP: #464552
 -- Stefan Bader <email address hidden> Tue, 10 Nov 2009 14:31:52 +0100

[logari81 (logari81)]

This fix breaks the following device:

I: Bus=0011 Vendor=0002 Product=0008 Version=7325
N: Name="AlpsPS/2 ALPS GlidePoint"
P: Phys=isa0060/serio1/input0
S: Sysfs=/devices/platform/i8042/serio1/input/input9
U: Uniq=
H: Handlers=mouse3 event9
B: EV=f
B: KEY=420 0 670000 0 0 0 0 0 0 0 0
B: REL=3
B: ABS=1000003

Tested on a fresh karmic installation. I attach the dmesg results from 2.6.31-14 and 2.6.31-15.

[logari81 (logari81)]

oops, after a reboot to 2.6.31-15 the touchpad now works with this kernel too! Anyway don't take my last post into account.