Re: All kinds of irq 16: nobody cared with Sandy Bridge Asus P8H67-MMB and multiple drivers

From: Bill Davidsen
Date: Wed Jun 29 2011 - 10:12:23 EST


Marc MERLIN wrote:
[please Cc me on replies so that I can see them faster]

I've found various reports on the net, but usually they are one driver
or one card that's not doing the right thing.

In my case, it happens with multiple cards and drivers and I'm wondering if
it can be a motherboard bug somehow, and if there are linux kernel options
other than irqpoll which does not help, that can shed some light on this.

I just bought a new Sandy Bridge board:
Manufacturer: ASUSTeK Computer INC.
Product Name: P8H67-M PRO
Vendor: American Megatrends Inc.
Version: 1003
Release Date: 05/10/2011

I suppose the onboard pata could be problematic, but if so why would it only
fail when some special combination of cards share its irq?
03:00.0 IDE interface: VIA Technologies, Inc. Unknown device 0415
pata_via 0000:03:00.0: version 0.3.4
pata_via 0000:03:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
pata_via 0000:03:00.0: setting latency timer to 64

I started with 2.6.36 and now have the same problems with 2.6.39.1.
The MB has 2 PCIe slots, and 2 PCI slots.
3 slots get force routed on irq 16, and the last PCI slot is irq 19.
The onboard IDE is also irq 16.

I have tried 6 different PCI and PCIe cards:
- tg3 (gige)
- rtl8169 (gige)
- e100
- CMI8738-MC6
- sata_sil24 (sil 3132)
- sata_mv (marvell)

Pretty much all combinations where I have more than one PCI card plugged
in a slot routed to irq16 causes

irq 16: nobody cared (try booting with the "irqpoll" option)
(...)
handlers:
[<c037171b>] (ata_bmdma_interrupt+0x0/0x170)
[<f8450f1f>] (sil24_interrupt+0x0/0x443 [sata_sil24])
[<f8449d7a>] (e100_intr+0x0/0xa2 [e100])
Disabling IRQ #16

irqpoll makes no difference.

Moving the cards around helps if I only use 2 cards, one in the slot
with irq19 (currently sound card) and one in the slot with irq16.

Putting more than one card on irq 16 (not counting onboard pata) pretty
much always causes the dreaded message.

I however got one very puzzling error with an e100 in the separate irq19
slot once too:
kernel: irq 19: nobody cared (try booting with the "irqpoll" option)
kernel: Pid: 0, comm: swapper Tainted: G W 2.6.36.0-core2smp-volpreempt-noide-hm64-20100724 #1

Does this happen with an untainted kernel?

--
Bill Davidsen <davidsen@xxxxxxx>
"We have more to fear from the bungling of the incompetent than from
the machinations of the wicked." - from Slashdot
--
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/