Arch Linux

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Uevent disables USB-ports

Good Evening,

since kernel 2.6.39, an uevent disables my usb ports:
Kernel Log output

Oct 18 09:01:04 hades kernel: [    1.436484] irq 16: nobody cared (try booting with the "irqpoll" option)
Oct 18 09:01:04 hades kernel: [    1.436486] Pid: 0, comm: swapper Not tainted 3.0-ARCH #1
Oct 18 09:01:04 hades kernel: [    1.436487] Call Trace:
Oct 18 09:01:04 hades kernel: [    1.436488]  <IRQ>  [<ffffffff810c147a>] __report_bad_irq+0x3a/0xd0
Oct 18 09:01:04 hades kernel: [    1.436495]  [<ffffffff810c1896>] note_interrupt+0x136/0x1f0
Oct 18 09:01:04 hades kernel: [    1.436497]  [<ffffffff810bf989>] handle_irq_event_percpu+0xc9/0x2a0
Oct 18 09:01:04 hades kernel: [    1.436499]  [<ffffffff810bfba5>] handle_irq_event+0x45/0x70
Oct 18 09:01:04 hades kernel: [    1.436501]  [<ffffffff810c21c7>] handle_fasteoi_irq+0x57/0xd0
Oct 18 09:01:04 hades kernel: [    1.436503]  [<ffffffff8100d9f2>] handle_irq+0x22/0x40
Oct 18 09:01:04 hades kernel: [    1.436506]  [<ffffffff813f66ea>] do_IRQ+0x5a/0xe0
Oct 18 09:01:04 hades kernel: [    1.436508]  [<ffffffff813f43d3>] common_interrupt+0x13/0x13
Oct 18 09:01:04 hades kernel: [    1.436509]  <EOI>  [<ffffffff81273d7b>] ? intel_idle+0xcb/0x120
Oct 18 09:01:04 hades kernel: [    1.436513]  [<ffffffff81273d5d>] ? intel_idle+0xad/0x120
Oct 18 09:01:04 hades kernel: [    1.436516]  [<ffffffff81313e3d>] cpuidle_idle_call+0x9d/0x350
Oct 18 09:01:04 hades kernel: [    1.436519]  [<ffffffff8100a21a>] cpu_idle+0xba/0x100
Oct 18 09:01:04 hades kernel: [    1.436521]  [<ffffffff813d1f52>] rest_init+0x96/0xa4
Oct 18 09:01:04 hades kernel: [    1.436523]  [<ffffffff81748c23>] start_kernel+0x3de/0x3eb
Oct 18 09:01:04 hades kernel: [    1.436525]  [<ffffffff81748347>] x86_64_start_reservations+0x132/0x136
Oct 18 09:01:04 hades kernel: [    1.436527]  [<ffffffff81748140>] ? early_idt_handlers+0x140/0x140
Oct 18 09:01:04 hades kernel: [    1.436529]  [<ffffffff8174844d>] x86_64_start_kernel+0x102/0x111
Oct 18 09:01:04 hades kernel: [    1.436530] handlers:
Oct 18 09:01:04 hades kernel: [    1.436533] [<ffffffffa00064f0>] usb_hcd_irq
Oct 18 09:01:04 hades kernel: [    1.436534] Disabling IRQ #16

Later disabling IRQ #16 again:

Oct 18 09:01:04 hades kernel: [   15.963949] irq 16: nobody cared (try booting with the "irqpoll" option)
Oct 18 09:01:04 hades kernel: [   15.963955] Pid: 0, comm: swapper Tainted: G         C  3.0-ARCH #1
Oct 18 09:01:04 hades kernel: [   15.963957] Call Trace:
Oct 18 09:01:04 hades kernel: [   15.963959]  <IRQ>  [<ffffffff810c147a>] __report_bad_irq+0x3a/0xd0
Oct 18 09:01:04 hades kernel: [   15.963975]  [<ffffffff810c1896>] note_interrupt+0x136/0x1f0
Oct 18 09:01:04 hades kernel: [   15.963977]  [<ffffffff810bf989>] handle_irq_event_percpu+0xc9/0x2a0
Oct 18 09:01:04 hades kernel: [   15.963979]  [<ffffffff810bfba5>] handle_irq_event+0x45/0x70
Oct 18 09:01:04 hades kernel: [   15.963981]  [<ffffffff810c21c7>] handle_fasteoi_irq+0x57/0xd0
Oct 18 09:01:04 hades kernel: [   15.963984]  [<ffffffff8100d9f2>] handle_irq+0x22/0x40
Oct 18 09:01:04 hades kernel: [   15.963986]  [<ffffffff813f66ea>] do_IRQ+0x5a/0xe0
Oct 18 09:01:04 hades kernel: [   15.963988]  [<ffffffff813f43d3>] common_interrupt+0x13/0x13
Oct 18 09:01:04 hades kernel: [   15.963989]  <EOI>  [<ffffffff81273d7b>] ? intel_idle+0xcb/0x120
Oct 18 09:01:04 hades kernel: [   15.963994]  [<ffffffff81273d5d>] ? intel_idle+0xad/0x120
Oct 18 09:01:04 hades kernel: [   15.963997]  [<ffffffff81313e3d>] cpuidle_idle_call+0x9d/0x350
Oct 18 09:01:04 hades kernel: [   15.964000]  [<ffffffff8100a21a>] cpu_idle+0xba/0x100
Oct 18 09:01:04 hades kernel: [   15.964002]  [<ffffffff813d1f52>] rest_init+0x96/0xa4
Oct 18 09:01:04 hades kernel: [   15.964004]  [<ffffffff81748c23>] start_kernel+0x3de/0x3eb
Oct 18 09:01:04 hades kernel: [   15.964006]  [<ffffffff81748347>] x86_64_start_reservations+0x132/0x136
Oct 18 09:01:04 hades kernel: [   15.964008]  [<ffffffff81748140>] ? early_idt_handlers+0x140/0x140
Oct 18 09:01:04 hades kernel: [   15.964009]  [<ffffffff8174844d>] x86_64_start_kernel+0x102/0x111
Oct 18 09:01:04 hades kernel: [   15.964010] handlers:
Oct 18 09:01:04 hades kernel: [   15.964016] [<ffffffffa00064f0>] usb_hcd_irq
Oct 18 09:01:04 hades kernel: [   15.964018] [<ffffffffa02a4220>] mei_interrupt_quick_handler threaded [<ffffffffa02a4570>] mei_interrupt_thread_handler
Oct 18 09:01:04 hades kernel: [   15.964021] Disabling IRQ #16

Several output information:

lspci:

00:16.0 Communication controller: Intel Corporation 6 Series/C200 Series Chipset Family MEI Controller #1 (rev 04)

/proc/interrupts:

      CPU0                     CPU1  CPU2  CPU3       
16: 1687314878          0          0          0   IO-APIC-fasteoi   ehci_hcd:usb1, mei

Kernel Log mei info:

Oct 18 09:01:04 hades kernel: [   15.464667] mei: module is from the staging directory, the quality is unknown, you have been warned.
Oct 18 09:01:04 hades kernel: [   15.464922] mei 0000:00:16.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Oct 18 09:01:04 hades kernel: [   15.464928] mei 0000:00:16.0: setting latency timer to 64

Kernel Log ehci info:

Oct 18 09:01:04 hades kernel: [    0.999944] ehci_hcd 0000:00:1d.0: PCI INT A -> GSI 23 (level, low) -> IRQ 23
Oct 18 09:01:04 hades kernel: [    0.999964] ehci_hcd 0000:00:1d.0: setting latency timer to 64
Oct 18 09:01:04 hades kernel: [    0.999966] ehci_hcd 0000:00:1d.0: EHCI Host Controller
Oct 18 09:01:04 hades kernel: [    0.999971] ehci_hcd 0000:00:1d.0: new USB bus registered, assigned bus number 4
Oct 18 09:01:04 hades kernel: [    0.999996] ehci_hcd 0000:00:1d.0: debug port 2
Oct 18 09:01:04 hades kernel: [    1.003885] ehci_hcd 0000:00:1d.0: cache line size of 64 is not supported
Oct 18 09:01:04 hades kernel: [    1.003897] ehci_hcd 0000:00:1d.0: irq 23, io mem 0xf0426800
Oct 18 09:01:04 hades kernel: [    1.016476] ehci_hcd 0000:00:1d.0: USB 2.0 started, EHCI 1.00

Module Info:
mei:

filename:       /lib/modules/3.0-ARCH/kernel/drivers/staging/mei/mei.ko.gz
version:        7.1.20.1
description:    Intel(R) Management Engine Interface
author:         Intel Corporation
srcversion:     0A1192D3177950D60ACC4F4
staging:        Y
vermagic:       3.0-ARCH SMP preempt mod_unload 
parm:           watchdog_timeout:Intel(R) AMT Watchdog timeout value in seconds. (default=120, disable=0) (ushort)

ehci_hcd:

filename:       /lib/modules/3.0-ARCH/kernel/drivers/usb/host/ehci-hcd.ko.gz
author:         David Brownell
description:    USB 2.0 'Enhanced' Host Controller (EHCI) Driver
alias:          pci:v*d*sv*sd*bc0Csc03i20*
depends:        usbcore
vermagic:       3.0-ARCH SMP preempt mod_unload 
parm:           log2_irq_thresh:log2 IRQ latency, 1-64 microframes (int)
parm:           park:park setting; 1-3 back-to-back async packets (uint)
parm:           ignore_oc:ignore bogus hardware overcurrent indications (bool)
parm:           hird:host initiated resume duration, +1 for each 75us

I'm using the DP67BG mainboard from intel.
Please let me know, if you need additional information.

Greetings,
kinky

Edit:
Forgot to mention that I already tried following kernel parameters (acpi=routeirq, noirqdebug, pci=msi, irqpoll, irqfixup) of which only noirqdebug worked (as to say that IRQ #16 was not disables, so my keyboard did work), but which caused a decrease in graphics performance (low fps rates (shells refresh slowly), tearing).

Last edited by kinky (2011-11-07 20:44:48)

tomk

Forum Fellow

From: Ireland

Registered: 2004-07-21

Posts: 9,839

Re: Uevent disables USB-ports

mei: module is from the staging directory, the quality is unknown, you have been warned.

I'd try blacklisting mei, see if that helps.

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

Did blacklist it, no change whatsoever.
However, I would like to point out, that my usb ports are disabled _before_
the modules are loaded (I'm using lvm+luks), precisely after the udev events,
right before entering the luks password.

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

*bump*

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

*bumping* again.

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

another *bump*

Lone_Wolf

Administrator

From: Netherlands, Europe

Registered: 2005-10-04

Posts: 13,389

Re: Uevent disables USB-ports

looks like a conflict between ehci_hd and mei, but to be sure there are no other devices handling irq 16, please post

$ lspci -v

What extension cards do you have in your system ?

---

Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.

clean chroot building not flexible enough ?
Try clean chroot manager by graysky

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

lspci -v:
http://paste.xinu.at/wCwKj/

The only extension card would be my graphics card. (http://paste.xinu.at/wCwKj/#n118)

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

Another day, another *bump*.

Lone_Wolf

Administrator

From: Netherlands, Europe

Registered: 2005-10-04

Posts: 13,389

Re: Uevent disables USB-ports

ok, there are only 2 devices using IRQ 16

00:16.0 Communication controller: Intel Corporation 6 Series/C200 Series Chipset Family MEI Controller #1 (rev 04)

00:1a.0 USB controller: Intel Corporation 6 Series/C200 Series Chipset Family USB Enhanced Host Controller #2 (rev 05) (prog-if 20 [EHCI])

there are many reports about intel chipset systems having the same problem after kernel 2.6.38 / 2.6.39 .
you basically have 2 options :
1. If you want to have all usb ports available:
test with the LTS kernel 2.6.32.48, verify it doesn't have those problems (99% sure)
stick to LTS-kernel atleast until the problem is solved in the mainline kernel

2. Disable / don't use the problem ports.
your system has multiple USB controllers :

00:1a.0 USB controller: Intel Corporation 6 Series/C200 Series Chipset Family USB Enhanced Host Controller #2 (rev 05) (prog-if 20 [EHCI])

00:1d.0 USB controller: Intel Corporation 6 Series/C200 Series Chipset Family USB Enhanced Host Controller #1 (rev 05) (prog-if 20 [EHCI])

03:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host Controller (rev 03) (prog-if 30 [XHCI])

test which / how many ports do work
If there are not enough usb ports working, your system has several pci express slots.
you could consider purchasing a pci express extension card with usb ports.
While it may use the same driver, it will get a different irq.

---

Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.

clean chroot building not flexible enough ?
Try clean chroot manager by graysky

nihues

Member

From: Balneário Camboriú / SC

Registered: 2007-11-09

Posts: 2

Website

Re: Uevent disables USB-ports

The problem is the USB3 controller (NEC uPD720200), I've got a post in ubuntu forum that partialy resolve... use pci=nomsi in grub seems to work, but with me it disables my wifi....

another way to use without nomsi is connecting to the usb port  and do a rmmod xhci_hcd then modprobe xhci_hcd

There is a patch sent to the 3.2 kernel maybe resolves the problem...

---

Criação Desenvolvimento de sites
Ecolibra Consultoria Ambiental
Oceanografia Brasil

kinky

Member

From: Germany

Registered: 2009-09-14

Posts: 17

Re: Uevent disables USB-ports

The problem is the USB3 controller (NEC uPD720200), I've got a post in ubuntu forum that partialy resolve... use pci=nomsi in grub seems to work, but with me it disables my wifi....

another way to use without nomsi is connecting to the usb port  and do a rmmod xhci_hcd then modprobe xhci_hcd

There is a patch sent to the 3.2 kernel maybe resolves the problem...

The funny thing is, that I tried all USB-ports and it did work with my USB3.0 ports...
Keyboard and mouse receive power, as well as the normal USB2 ports after boot. Anyway, I hope that the problem will be fixed in 3.2 because someday I will surely need the USB3 ports for something other than mouse & keyboard.

I won't mark the thread as [solved] because my workaround isn't really a solution, but I will provide feedback, if the kernel 3.2 did solve the problem.