New: touchpad become crazy: disable irq 21

Bugme-new Andrew Morton 4 participants 4 messages Sep 8, 2010 – Nov 26, 2010 1 message not in the archive ACPI

lore

Share & tools

lore

Andrew Morton

author

2010-09-08 22:32 UTC #1 lore raw

In reply to a message not in the archive

(switched to email. Please respond via emailed reply-to-all, not via the bugzilla web interface).

Quoted 14 lines · On Sun, 5 Sep 2010 11:47:41 GMT [email protected] wrote:

https://bugzilla.kernel.org/show_bug.cgi?id=17802
Summary: touchpad become crazy: disable irq 21
Product: Other
Version: 2.5
Platform: All
OS/Version: Linux
Tree: Mainline
Status: NEW
Severity: high
Priority: P1
Component: Other
AssignedTo: [email protected]
ReportedBy: [email protected]

heh.

Quoted 49 lines

Regression: No

Hi
on macbook 4.1 (kernel 2.6.31.12) often the touchpad become crazy and irq 21 is disable, in syslog I see:

Sep  5 13:32:05 pollon kernel: [ 2215.099257] ide-atapi: hda:
ide_check_ireason: wrong transfer direction!
Sep  5 13:32:05 pollon kernel: [ 2215.150195] ide-atapi: hda:
ide_check_ireason: wrong transfer direction!
Sep  5 13:32:23 pollon kernel: [ 2233.081273] hda: request sense failure:
status=0xd0 { Busy }
Sep  5 13:32:23 pollon kernel: [ 2233.081285] hda: possibly failed opcode: 0xa0
Sep  5 13:32:37 pollon kernel: [ 2247.317239] irq 21: nobody cared (try booting
with the "irqpoll" option)
Sep  5 13:32:37 pollon kernel: [ 2247.317244] Pid: 0, comm: swapper Tainted: P 
         2.6.31.12-pollon #2
Sep  5 13:32:37 pollon kernel: [ 2247.317246] Call Trace:
Sep  5 13:32:37 pollon kernel: [ 2247.317253]  [<c0169b7f>] ?
__report_bad_irq+0x24/0x69
Sep  5 13:32:37 pollon kernel: [ 2247.317256]  [<c0169b86>] ?
__report_bad_irq+0x2b/0x69
Sep  5 13:32:37 pollon kernel: [ 2247.317259]  [<c0169cb1>] ?
note_interrupt+0xed/0x13c
Sep  5 13:32:37 pollon kernel: [ 2247.317262]  [<c016a196>] ?
handle_fasteoi_irq+0x7a/0x97
Sep  5 13:32:37 pollon kernel: [ 2247.317265]  [<c01098d4>] ?
handle_irq+0x17/0x1b
Sep  5 13:32:37 pollon kernel: [ 2247.317268]  [<c0109110>] ? do_IRQ+0x38/0x89
Sep  5 13:32:37 pollon kernel: [ 2247.317270]  [<c0107c69>] ?
common_interrupt+0x29/0x30
Sep  5 13:32:37 pollon kernel: [ 2247.317289]  [<f845623a>] ?
acpi_idle_enter_simple+0xf7/0x129 [processor]
Sep  5 13:32:37 pollon kernel: [ 2247.317293]  [<c02f79d6>] ?
cpuidle_idle_call+0x64/0x96
Sep  5 13:32:37 pollon kernel: [ 2247.317297]  [<c01068a3>] ?
cpu_idle+0x46/0x5f
Sep  5 13:32:37 pollon kernel: [ 2247.317298] handlers:
Sep  5 13:32:37 pollon kernel: [ 2247.317300] [<f809c147>]
(usb_hcd_irq+0x0/0x66 [usbcore])
Sep  5 13:32:37 pollon kernel: [ 2247.317323] [<f80dbbbe>] (ide_intr+0x0/0x1b9
[ide_core])
Sep  5 13:32:37 pollon kernel: [ 2247.317343] [<f80dbbbe>] (ide_intr+0x0/0x1b9
[ide_core])
Sep  5 13:32:37 pollon kernel: [ 2247.317361] [<f809c147>]
(usb_hcd_irq+0x0/0x66 [usbcore])
Sep  5 13:32:37 pollon kernel: [ 2247.317382] Disabling IRQ #21

Quoted 4 lines

also suspend stops and I can only reboot the system.
If I add irqpoll there is not this problem but suspend doesn't run.

Not heh. Ugly.

Could be ACPI, could be x86, could be something else.

Guys, could you please take a look, suggest how we should tackle this?

Thanks.

Dmitry Torokhov

2010-09-08 22:58 UTC #2 lore raw

Quoted 22 lines · Andrew Morton

(switched to email. Please respond via emailed reply-to-all, not via the bugzilla web interface).
Quoted 14 lines · On Sun, 5 Sep 2010 11:47:41 GMT [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=17802
Summary: touchpad become crazy: disable irq 21
Product: Other
Version: 2.5
Platform: All
OS/Version: Linux
Tree: Mainline
Status: NEW
Severity: high
Priority: P1
Component: Other
AssignedTo: [email protected]
ReportedBy: [email protected]

Quoted 8 lines

heh.
Quoted 6 lines
Regression: No
Hi
on macbook 4.1 (kernel 2.6.31.12) often the touchpad become crazy and irq 21 is disable, in syslog I see:

Hmm, by touchpad being crazy you mean the pointer jumps around in X? Does the touchpad driver use IRQ 21?

Signature · 1 line

Dmitry

Zhang Rui

2010-09-09 00:14 UTC #3 lore raw

In reply to Andrew Morton

Quoted 21 lines · Andrew Morton

(switched to email. Please respond via emailed reply-to-all, not via the bugzilla web interface).
Quoted 14 lines · On Sun, 5 Sep 2010 11:47:41 GMT [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=17802
Summary: touchpad become crazy: disable irq 21
Product: Other
Version: 2.5
Platform: All
OS/Version: Linux
Tree: Mainline
Status: NEW
Severity: high
Priority: P1
Component: Other
AssignedTo: [email protected]
ReportedBy: [email protected]

Quoted 55 lines

heh.

Quoted 53 lines

Regression: No

Hi
on macbook 4.1 (kernel 2.6.31.12) often the touchpad become crazy and irq 21 is disable, in syslog I see:

Sep  5 13:32:05 pollon kernel: [ 2215.099257] ide-atapi: hda:
ide_check_ireason: wrong transfer direction!
Sep  5 13:32:05 pollon kernel: [ 2215.150195] ide-atapi: hda:
ide_check_ireason: wrong transfer direction!
Sep  5 13:32:23 pollon kernel: [ 2233.081273] hda: request sense failure:
status=0xd0 { Busy }
Sep  5 13:32:23 pollon kernel: [ 2233.081285] hda: possibly failed opcode: 0xa0
Sep  5 13:32:37 pollon kernel: [ 2247.317239] irq 21: nobody cared (try booting
with the "irqpoll" option)
Sep  5 13:32:37 pollon kernel: [ 2247.317244] Pid: 0, comm: swapper Tainted: P 
         2.6.31.12-pollon #2
Sep  5 13:32:37 pollon kernel: [ 2247.317246] Call Trace:
Sep  5 13:32:37 pollon kernel: [ 2247.317253]  [<c0169b7f>] ?
__report_bad_irq+0x24/0x69
Sep  5 13:32:37 pollon kernel: [ 2247.317256]  [<c0169b86>] ?
__report_bad_irq+0x2b/0x69
Sep  5 13:32:37 pollon kernel: [ 2247.317259]  [<c0169cb1>] ?
note_interrupt+0xed/0x13c
Sep  5 13:32:37 pollon kernel: [ 2247.317262]  [<c016a196>] ?
handle_fasteoi_irq+0x7a/0x97
Sep  5 13:32:37 pollon kernel: [ 2247.317265]  [<c01098d4>] ?
handle_irq+0x17/0x1b
Sep  5 13:32:37 pollon kernel: [ 2247.317268]  [<c0109110>] ? do_IRQ+0x38/0x89
Sep  5 13:32:37 pollon kernel: [ 2247.317270]  [<c0107c69>] ?
common_interrupt+0x29/0x30
Sep  5 13:32:37 pollon kernel: [ 2247.317289]  [<f845623a>] ?
acpi_idle_enter_simple+0xf7/0x129 [processor]
Sep  5 13:32:37 pollon kernel: [ 2247.317293]  [<c02f79d6>] ?
cpuidle_idle_call+0x64/0x96
Sep  5 13:32:37 pollon kernel: [ 2247.317297]  [<c01068a3>] ?
cpu_idle+0x46/0x5f
Sep  5 13:32:37 pollon kernel: [ 2247.317298] handlers:
Sep  5 13:32:37 pollon kernel: [ 2247.317300] [<f809c147>]
(usb_hcd_irq+0x0/0x66 [usbcore])
Sep  5 13:32:37 pollon kernel: [ 2247.317323] [<f80dbbbe>] (ide_intr+0x0/0x1b9
[ide_core])
Sep  5 13:32:37 pollon kernel: [ 2247.317343] [<f80dbbbe>] (ide_intr+0x0/0x1b9
[ide_core])
Sep  5 13:32:37 pollon kernel: [ 2247.317361] [<f809c147>]
(usb_hcd_irq+0x0/0x66 [usbcore])
Sep  5 13:32:37 pollon kernel: [ 2247.317382] Disabling IRQ #21

also suspend stops and I can only reboot the system.

If I add irqpoll there is not this problem but suspend doesn't run.

please attach the content of /proc/interrupts.
please attach the acpidump as well.

thanks,
rui

Len Brown

reviewer

2010-11-26 20:15 UTC #4 lore raw

In reply to Andrew Morton

CPU0 CPU1

   0:     287671     109637   IO-APIC-edge      timer
   8:          0          1   IO-APIC-edge      rtc0
   9:        483        479   IO-APIC-fasteoi   acpi
  16:      36491      15935   IO-APIC-fasteoi   uhci_hcd:usb4, 
uhci_hcd:usb5, ndiswrapper
  18:      23811      10147   IO-APIC-fasteoi   uhci_hcd:usb6, ata_piix
  19:          1          3   IO-APIC-fasteoi   ohci1394
  20:        869        298   IO-APIC-fasteoi   ehci_hcd:usb2, 
uhci_hcd:usb3, HDA Intel
  21:      97807      68959   IO-APIC-fasteoi   ide0, ide1, ehci_hcd:usb1, 
uhci_hcd:usb7
  27:          1          0   PCI-MSI-edge      eth0
  28:      18118      22763   PCI-MSI-edge      i915@pci:0000:00:02.0

The DSDT shows that there are no programmable IRQs in IOAPIC mode on this machine, so it is unlikely that there is an ACPI interrupt routing issue.

It is more likely that the problem is in one of the devices (above) using IRQ21. I assume that the touchpad is uhci_hcd:usb7?

What does "lsusb" show on this machine?

Of course I'd recommend trying to reproduce the issue after excluding "ndiswrapper" from the configuration.

cheers,
Len Brown, Intel Open Source Technology Center