Bug 97131 - Unhandled IRQ 16 from ehci_hcd makes mouse unusable
Summary: Unhandled IRQ 16 from ehci_hcd makes mouse unusable
Status: NEW
Alias: None
Product: Drivers
Classification: Unclassified
Component: USB (show other bugs)
Hardware: Intel Linux
: P1 normal
Assignee: Greg Kroah-Hartman
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-04-23 01:47 UTC by kynnjo
Modified: 2015-04-23 07:24 UTC (History)
0 users

See Also:
Kernel Version: 3.2.0.-4-amd64
Subsystem:
Regression: No
Bisected commit-id:


Attachments
Output of diagnostic commands (dmesg, etc.) (68.23 KB, text/plain)
2015-04-23 01:47 UTC, kynnjo
Details

Description kynnjo 2015-04-23 01:47:19 UTC
Created attachment 174821 [details]
Output of diagnostic commands (dmesg, etc.)

NB: The attachment to this report includes the output of the following commands:

    # dmesg
    # cat /etc/rc.local
    # lspci -nn
    # lspci -v
    # dpkg -l | grep "$(uname -r)"
    # cat /proc/interrupts

================================================================================

My new computer (Dell Precision T1700) works fine until the system
goes to sleep for the first time after booting.  After waking, the
cursor on the screen lags so much behind the physical motion of the
mouse that the latter is basically unusable.  This mouse malfunction
persists until the system is rebooted.

The only new log messages I see when I bring the system back from its
first post-boot sleep is this (both in /var/log/kern.log and
/var/log/syslog, and part of it also in /var/log/messages):

    Apr 22 09:21:00 capitan kernel: [   67.832836] irq 16: nobody cared (try booting with the "irqpoll" option)
    Apr 22 09:21:00 capitan kernel: [   67.832848] Pid: 0, comm: swapper/0 Not tainted 3.2.0-4-amd64 #1 Debian 3.2.65-1+deb7u2
    Apr 22 09:21:00 capitan kernel: [   67.832853] Call Trace:
    Apr 22 09:21:00 capitan kernel: [   67.832856]  <IRQ>  [<ffffffff81092ddd>] ? __report_bad_irq+0x2c/0xb5
    Apr 22 09:21:00 capitan kernel: [   67.832878]  [<ffffffff810931e2>] ? note_interrupt+0x1b8/0x23a
    Apr 22 09:21:00 capitan kernel: [   67.832886]  [<ffffffff81091554>] ? handle_irq_event_percpu+0x15f/0x17d
    Apr 22 09:21:00 capitan kernel: [   67.832893]  [<ffffffff810915a6>] ? handle_irq_event+0x34/0x52
    Apr 22 09:21:00 capitan kernel: [   67.832903]  [<ffffffff8106c305>] ? arch_local_irq_save+0x11/0x17
    Apr 22 09:21:00 capitan kernel: [   67.832911]  [<ffffffff81093959>] ? handle_fasteoi_irq+0x7c/0xaf
    Apr 22 09:21:00 capitan kernel: [   67.832921]  [<ffffffff8100fa51>] ? handle_irq+0x1d/0x21
    Apr 22 09:21:00 capitan kernel: [   67.832929]  [<ffffffff8100f62a>] ? do_IRQ+0x42/0x98
    Apr 22 09:21:00 capitan kernel: [   67.832939]  [<ffffffff813511ae>] ? common_interrupt+0x6e/0x6e
    Apr 22 09:21:00 capitan kernel: [   67.832943]  <EOI>  [<ffffffff81024404>] ? lapic_next_event+0xe/0x13
    Apr 22 09:21:00 capitan kernel: [   67.832977]  [<ffffffffa01dc35b>] ? arch_local_irq_enable+0x7/0x8 [processor]
    Apr 22 09:21:00 capitan kernel: [   67.832991]  [<ffffffffa01dd0b3>] ? acpi_idle_enter_c1+0x8d/0xb3 [processor]
    Apr 22 09:21:00 capitan kernel: [   67.833002]  [<ffffffff8127180d>] ? cpuidle_idle_call+0xec/0x179
    Apr 22 09:21:00 capitan kernel: [   67.833010]  [<ffffffff8100d242>] ? cpu_idle+0xa5/0xf2
    Apr 22 09:21:00 capitan kernel: [   67.833020]  [<ffffffff816aab3b>] ? start_kernel+0x3bd/0x3c8
    Apr 22 09:21:00 capitan kernel: [   67.833029]  [<ffffffff816aa140>] ? early_idt_handlers+0x140/0x140
    Apr 22 09:21:00 capitan kernel: [   67.833037]  [<ffffffff816aa3c4>] ? x86_64_start_kernel+0x104/0x111
    Apr 22 09:21:00 capitan kernel: [   67.833041] handlers:
    Apr 22 09:21:00 capitan kernel: [   67.833059] [<ffffffffa00a9216>] usb_hcd_irq
    Apr 22 09:21:00 capitan kernel: [   67.833067] [<ffffffffa0295cbd>] azx_interrupt
    Apr 22 09:21:00 capitan kernel: [   67.833072] Disabling IRQ #16


I figure that it is the disabling of IRQ #16 (last line) that is
responsible for the subsequent malfunctioning of the mouse.

The driver associated with this interrupt is

    # grep '^ 16:' /proc/interrupts
     16:       1211          0          0          0  IR-IO-APIC-fasteoi   ehci_hcd:usb1

(When I first looked into this problem, the IRQ 16 line was actually
like this:

    # grep '^ 16:' /proc/interrupts
     16:       1210          0          0          0  IR-IO-APIC-fasteoi   ehci_hcd:usb1, snd_hda_intel

Since snd_hda_intel was being loaded twice at boot time (e.g. it
showed up also in the line for IRQ 45), I tried to fix the problem
with the mouse by disabling this instance of snd_hda_intel.  For
details, see the content of my /etc/rc.local file further below.  As
it turned out, even though snd_hda_intel no longer appears in the IRQ
16 line, the unhandled IRQ 16 error keeps happening, and the problem
with the mouse persists.)


================================================================================
A few more observations:

1. I don't see this problem at all if I reboot the machine into
   Windows 7 (mouse works the same before and after sleep).

2. the mouse does not seem to be the source of the problem, because the
   both the unhandled interrupt and the subsequent problems with the
   mouse occur even if I:

     1. disconnect the mouse;
     2. reboot the machine (into Debian);
     3. put the system to sleep (with `sleep 0.1; xset dpms force off` );
     4. wake the system;
     5. connect the mouse

3. the problem also occurs if, instead of putting the system to sleep
   (as described in step (2.3) above), I simply turn off the monitor.
Comment 1 Greg Kroah-Hartman 2015-04-23 07:24:49 UTC
On Thu, Apr 23, 2015 at 01:47:19AM +0000, bugzilla-daemon@bugzilla.kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=97131
> 
>             Bug ID: 97131
>            Summary: Unhandled IRQ 16 from ehci_hcd makes mouse unusable

Please send to the linux-usb@vger.kernel.org mailing list

Note You need to log in before you can comment on or make changes to this bug.