Re: [regression, bisected] adb trackpad disappears after suspend to ram

From: Rafael J. Wysocki
Date: Thu May 28 2009 - 12:58:42 EST


On Thursday 28 May 2009, Jan Scholz wrote:
> Jan Scholz <scholz@xxxxxxxxxxxxxxxxxxxxx> writes:
>
> > Hi,
> >
> > On my PowerBook6,5 (iBook G4) the trackpad does not show up on the
> > adb-bus after resuming from suspend to ram. This happens in roughly 50%
> > of the resume cycles.
> > Affected kernel versions: v2.6.30_rc1 through _rc7
>
> I bisected the regression and found the first bad commit to be:
> 2ed8d2b3a81bdbb0418301628ccdb008ac9f40b7
> "PM: Rework handling of interrupts during suspend-resume"

Thanks for bisecting.

We are too late in the cycle to revert this commit and it really is needed to
fix a more serious issue. Nevertheless knowing that it caused the problem to
appear on your system is also important.

Ben, do you have an idea what may be going on here? Does __disable_irq() mask
the interrupt on this platform?

> To those of you I cc'ed: sorry for the noise, but you were the ones that
> get_maintainers.pl spit out when asking for *adb*-files.
>
> In addition to the diff of the lshal output I posted before, here is the
> diff in the output of lsadb:
>
> --- lsadb-ok.out 2009-05-28 08:53:09.000000000 +0200
> +++ lsadb-notouchpad.out 2009-05-28 08:55:59.000000000 +0200
> @@ -1,35 +1,24 @@
>
> Devices connected to the ADB-Bus: (PMU: 12)
>
> D R raw data decoded data
> -------------------------------------------------------------------------------------
> 2.3: [62 c4] Encode Device
> Apple Macintosh PowerBook G3 Keyboard (ISO layout)
> Exceptional Event enabled
> Service Request enabled
> 2: [ff ff] Qualifier: (not all must be supported)
> Fn-key Delete Caps-Lock Reset Control
> Shift Option Apple Num-Lock Scroll-Lock
> Scroll-Lock-LED Caps-Lock-LED Num-Lock-LED
> 1: [05 01] Special keys need Fn Key
> 0:
>
> - 3.3: [23 04] Relative-Position Device
> - Extended Mouse Protokoll
> - Exceptional Event disabled
> - Service Request enabled
> - 2: [08 48 03 00 00 00 00 00]
> - 1: [74 70 61 64 01 90 03 04] Mouse type: tpad
> - resolution: 400 dpi
> - Device class: Trackpad
> - buttons: 4
> - 0:
> -
> 7.3: [67 1f] Misc. Device
> Powerbook G3 (Pismo) Button Device
> Exceptional Event enabled
> Service Request enabled
> 2:
> 1:
> 0:
>
> >
> > Here is a diff of the output of lshal before and after the trackpad vanished:
> >
> > --- lshal-ok.out 2009-05-25 17:00:28.000000000 +0200
> > +++ lshal-notouchpad.out 2009-05-25 17:04:42.000000000 +0200
> > @@ -1,5 +1,5 @@
> >
> > -Dumping 81 device(s) from the Global Device List:
> > +Dumping 80 device(s) from the Global Device List:
> > -------------------------------------------------
> > udi = '/org/freedesktop/Hal/devices/computer'
> > info.addons = {'hald-addon-cpufreq'} (string list)
> > @@ -95,21 +95,6 @@
> > linux.subsystem = 'input' (string)
> > linux.sysfs_path = '/sys/devices/virtual/input/input4/event4' (string)
> >
> > -udi = '/org/freedesktop/Hal/devices/computer_logicaldev_input_2'
> > - info.capabilities = {'input', 'input.mouse'} (string list)
> > - info.category = 'input' (string)
> > - info.parent = '/org/freedesktop/Hal/devices/computer' (string)
> > - info.product = 'ADB mouse' (string)
> > - info.subsystem = 'input' (string)
> > - info.udi = '/org/freedesktop/Hal/devices/computer_logicaldev_input_2' (string)
> > - input.device = '/dev/input/event3' (string)
> > - input.product = 'ADB mouse' (string)
> > - input.x11_driver = 'evdev' (string)
> > - linux.device_file = '/dev/input/event3' (string)
> > - linux.hotplug_type = 2 (0x2) (int)
> > - linux.subsystem = 'input' (string)
> > - linux.sysfs_path = '/sys/devices/virtual/input/input3/event3' (string)
> > -
> > udi = '/org/freedesktop/Hal/devices/computer_logicaldev_input_1'
> > info.addons.singleton = {'hald-addon-input'} (string list)
> > info.capabilities = {'input', 'input.keys', 'button'} (string list)
> > @@ -1820,6 +1805,6 @@
> > pci.vendor_id = 4203 (0x106b) (int)
> >
> >
> > -Dumped 81 device(s) from the Global Device List.
> > +Dumped 80 device(s) from the Global Device List.
> > ------------------------------------------------
> >
> > Please let me know if you need more info.

I'd like to see a boot log, preferably containing a suspend-resume in which
the problem was reproduced.

Best,
Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/