Re: radeon in dom0/ivtv in domU: irq 16 nobody cared

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Wed, Apr 07, 2010 at 08:19:16PM -0400, Konrad Rzeszutek Wilk wrote:
> On Thu, Apr 08, 2010 at 12:45:42AM +0200, Mark Hurenkamp wrote:
> > Hi,
> >
> >
> > With a fairly recent (~march 29th) xen/next kernel and a recent  
> > xen-4.0.rc9-pre build, i still get
> > occasional 'nobody cared' messages on a IRQ which is shared between dom0 
> > radeon
> > driver, and domU ivtv driver. Usually this happens while i'm working on 
> > the desktop,
> > and a recording is active in domU.
> >
> > Is there something i can do about this?
> 
> Yes,
> 
> Please e-mail your full serial log output, your cat /proc/interrupts,
> and 'lspci -vvv' output. This is to say, for both Dom0 and DomU.

I think I am able to reproduce this with one device (in DomU) that shares the IRQ
(17) with another device that is in Dom0. In Dom0 I get:

-sh-3.1# 
-sh-3.1# [ 2349.534294] irq 17: nobody cared (try booting with the
"irqpoll" option)
[ 2349.534477] Pid: 0, comm: swapper Not tainted 2Trace:
[ 2349.534728]  <IRQ>  [<ffffffff810ea3c7>] __report_bad_irq+0x54/0xe2
[ 2349.534887]  [<ffffffff810ea6a2>] note_interrupt+0x24d/0x2b8
[ 2349.535019]  [<ffffffff810eb95d>] handle_level_irq+0xef/0x17b
[ 2349.535151]  [<ffffffff81370257>] xen_evtchn_do_upcall+0x156/0x254
[ 2349.535282]  [<ffffffff81017e7e>]
xen_do_hypervisor_callback+0x1e/0x30
[ 2349.535282]  <EOI>  [<ffffffff810093aa>] ?
hypercall_page+0x3aa/0x1000
[ 2349.535282]  [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1000
[ 2349.535282]  [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1000
[ 2349.535282]  [<ffffffff81010f3b>] ? xen_safe_halt+0x1e/0x3d
[ 2349.535282]  [<ffffffff8100cb87>] ? xen_idle+0x10b/0x130
[ 2349.535282]  [<ffffffff81015887>] ? cpu_idle+0x167/0x1d5
[ 2349.535282]  [<ffffffff816641e1>] ? rest_init+0xb5/0xbe
[ 2349.535282]  [<ffffffff81a447a5>] ? start_kernel+0x777/0x78a
[ 2349.535282]  [<ffffffff81a43326>] ?
x86_64_start_reservations+0x111/0x11c
[ 2349.535282]  [<ffffffff81a48e25>] ? xen_start_kernel+0x678/0x686
[ 2349.535282] handlers:
[ 2349.535282] [<ffffffffa00c2420>] (lpfc_sli_intr_handler+0x0/0x22a
[lpfc])
[ 2349.535282] [<ffffffffa00240a5>] (tg3_interrupt_tagged+0x0/0xe6
[tg3])
[ 2349.535282] Disabling IRQ #17
[ 2382.845061] lpfc 0000:05:04.0: 0:0459 Adapter heartbeat failure,
taking this port offline.
[ 2397.052375] device-mapper: multipath: Failing path 8:0.
[ 2397.053041] ata3: lost interrupt (Status 0x50)
[ 2397.053275] [ 2398.054372] device-mapper: multipath: Failing path
8:16.
[ 2398.055179] ata4: lost interrupt (Status 0x50)
[ 2398.055413][ 2447.701115] ata3: lost interrupt (Status 0x50)
[ 2447.701389] sd 2:0:0:0: [sda] Unhandled error code
[ 2447.701515] sd 2:0


.. and it also kills the ata_piix controller which is not on the same
IRQ (??)

In DomU I get (not surprising since the IRQ just got turned off):

[  811.703067] ------------[ cut here ]------------
[  811.704050] WARNING: at
/home/konrad/git/neb.64/linux-up/net/sched/sch_generic.c:255
dev_watchdog+0xdf/0x139()
[  811.704050] NETDEV WATCHDOG: eth1 (skge): transmit queue 0 timed out
[  811.704050] Modules linked in: iscsi_ibft skge [last unloaded:
dump_dma]
[  811.704050] Pid: 0, comm: swapper Not tainted
2.6.34-rc3NEB-00089-g11969d1 #252
[  811.704050] Call Trace:
[  811.704050]  <IRQ>  [<ffffffff81042560>]
warn_slowpath_common+0x77/0x8f
[  811.704050]  [<ffffffff810425c5>] warn_slowpath_fmt+0x3c/0x3e
[  811.704050]  [<ffffffff81062989>] ? tick_dev_program_event+0x37/0xf7
[  811.704050]  [<ffffffff813ba070>] dev_watchdog+0xdf/0x139
[  811.704050]  [<ffffffff81062a9e>] ? tick_program_event+0x25/0x27
[  811.704050]  [<ffffffff8100740c>] ? xen_spin_lock+0xb/0xd
[  811.704050]  [<ffffffff8104c5c2>] run_timer_softirq+0x199/0x20b
[  811.704050]  [<ffffffff81047c4f>] __do_softirq+0xa5/0x142
[  811.704050]  [<ffffffff81249f62>] ? unmask_evtchn+0x2d/0xbc
[  811.704050]  [<ffffffff8100aa5c>] call_softirq+0x1c/0x30
[  811.704050]  [<ffffffff8100c411>] do_softirq+0x61/0xbf
[  811.704050]  [<ffffffff810479ad>] irq_exit+0x36/0x78
[  811.704050]  [<ffffffff8124a623>] xen_evtchn_do_upcall+0x18c/0x1a8
[  811.704050]  [<ffffffff8100aaae>]
xen_do_hypervisor_callback+0x1e/0x30
[  811.704050]  <EOI>  [<ffffffff810013aa>] ?
hypercall_page+0x3aa/0x1006
[  811.704050]  [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1006
[  811.704050]  [<ffffffff810062ef>] ? xen_safe_halt+0x10/0x1a
[  811.704050]  [<ffffffff81003ebb>] ? xen_idle+0x63/0x6c
[  811.704050]  [<ffffffff81008dcc>] ? cpu_idle+0x66/0xa4
[  811.704050]  [<ffffffff814a9155>] ? rest_init+0x79/0x7b
[  811.704050]  [<ffffffff81abedeb>] ? start_kernel+0x3e9/0x3f4
[  811.704050]  [<ffffffff81abe2c8>] ?
x86_64_start_reservations+0xb3/0xb7
[  811.704050]  [<ffffffff81ac145b>] ? xen_start_kernel+0x5a3/0x5a7
[  811.704050] ---[ end trace 742d7bf32c76ea72 ]---