LKML: Jan Kasprzak: PCI/IRQ regressions in 2.6.13.2

Hello,

I've tried to upgrade my Linux boxes to 2.6.13.2, and on some configurations
I have problems that IRQ stopped working or devices are not visible on
the PCI bus. These problems may be completely unrelated, though:

---------------------------------------------------------------------------
Case 1: HP DL-585 quad Opteron box - 2.6.11.10 to 2.6.13.2: The QLA 2312
fibre channel HBA is not visible on the PCI bus - here is a diff -u
between lspci on 2.6.13.2 and 2.6.11.10

--- /tmp/lspci-2.6.13.2 2005-09-23 18:16:24.000000000 +0200
+++ /tmp/lspci-2.6.11.10 2005-09-23 18:20:28.000000000 +0200
@@ -30,3 +30,12 @@
02:04.0 RAID bus controller: Compaq Computer Corporation Smart Array 5i/532 (rev 01)
02:06.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5704 Gigabit Ethernet (rev 10)
02:06.1 Ethernet controller: Broadcom Corporation NetXtreme BCM5704 Gigabit Ethernet (rev 10)
+04:09.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)+04:09.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
+04:0a.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)+04:0a.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
+04:0b.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)+04:0b.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
+04:0c.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)+04:0c.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
+06:0e.0 Fibre Channel: QLogic Corp. QLA2312 Fibre Channel Adapter (rev 02)

--------------------------------------------------------------------------
Case 2: ASUS A7V600 Athlon box (I have six of those boxes, and the problem
is only on one of them): During boot the IDE controller complains
about IRQ probe failing:
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
VP_IDE: IDE controller at PCI slot 0000:00:0f.1
PCI: IRQ 14 for device 0000:00:0f.1 doesn't match PIRQ mask - try pci=usepirqmask
PCI: setting IRQ 14 as level-triggered
PCI: Assigned IRQ 14 for device 0000:00:0f.1
IRQ routing conflict for 0000:00:10.0, have irq 12, want irq 14
IRQ routing conflict for 0000:00:10.1, have irq 12, want irq 14
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
VP_IDE: VIA vt8237 (rev 00) IDE UDMA133 controller on pci0000:00:0f.1
ide0: BM-DMA at 0xa400-0xa407, BIOS settings: hda:DMA, hdb:pio
ide1: BM-DMA at 0xa408-0xa40f, BIOS settings: hdc:DMA, hdd:pio
hda: ST380011A, ATA DISK drive
hda: IRQ probe failed (0xffffbbfa)
hdb: IRQ probe failed (0xfffffbfa)
hdb: IRQ probe failed (0xfffffbfa)
irq 14: nobody cared (try booting with the "irqpoll" option)
[<c01306c4>] __report_bad_irq+0x24/0x80
[<c01307c2>] note_interrupt+0x72/0xc0
[<c01301b0>] __do_IRQ+0xe0/0xf0
[<c0104e59>] do_IRQ+0x19/0x30
[<c0103492>] common_interrupt+0x1a/0x20
[<c011aa50>] __do_softirq+0x30/0x90
[<c011aad6>] do_softirq+0x26/0x30
[<c0104e5e>] do_IRQ+0x1e/0x30
[<c0103492>] common_interrupt+0x1a/0x20
[<c01303a7>] setup_irq+0x77/0xf0
[<c0214da0>] ide_intr+0x0/0x170
[<c013057d>] request_irq+0x9d/0xb0
[<c0218aca>] init_irq+0x19a/0x400
[<c02191b2>] hwif_init+0xd2/0x290
[<c021873d>] probe_hwif_init_with_fixup+0x1d/0x80
[<c021ba93>] ide_setup_pci_device+0x43/0x80
[<c03623f1>] ide_scan_pcidev+0x31/0x60
[<c036243d>] ide_scan_pcibus+0x1d/0xa0
[<c03623a8>] ide_init+0x48/0x60
[<c0350897>] do_initcalls+0x57/0xc0
[<c0154205>] kern_mount+0x15/0x19
[<c0100280>] init+0x0/0x100
[<c01002aa>] init+0x2a/0x100
[<c0100f00>] kernel_thread_helper+0x0/0x10
[<c0100f05>] kernel_thread_helper+0x5/0x10
handlers:
[<c0214da0>] (ide_intr+0x0/0x170)
Disabling IRQ #14
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hdc: WDC WD2500JB-00EVA0, ATA DISK drive
ide1 at 0x170-0x177,0x376 on irq 15
hda: max request size: 1024KiB
hda: lost interrupt
irq 14: nobody cared (try booting with the "irqpoll" option)
[...]
The pci=usepirqmask boot option seems to fix the problem.

The only difference in "lspci -vv" between 2.6.12.3 and 2.6.13.2
is the following line in the IDE controller section:

- Interrupt: pin A routed to IRQ 14
+ Interrupt: pin A routed to IRQ 11

(altough according to /proc/interrupts, ide uses IRQ 14/15 as usual both
on 2.6.12.3 and 2.6.12.2).

-------------------------------------------------------------------------
Case 3: ASUS A7V8X Athlon box (again, I have six of those boxes, only
one of them does not work): During boot the tg3 NIC is detected,
but does not work (probably no interrupts comming through).
The relevant part of dmesg(8) is
PCI: Using IRQ router VIA [1106/3177] at 0000:00:11.0
PCI: IRQ 0 for device 0000:00:07.0 doesn't match PIRQ mask - try pci=usepirqmaskPCI: Found IRQ 10 for device 0000:00:07.0
PCI: Sharing IRQ 10 with 0000:00:08.0
PCI: IRQ 0 for device 0000:00:09.0 doesn't match PIRQ mask - try pci=usepirqmaskPCI: Bridge: 0000:00:01.0
IO window: disabled.
MEM window: d6000000-d7dfffff
PREFETCH window: d7f00000-dfffffff
PCI: Setting latency timer of device 0000:00:01.0 to 64
highmem bounce pool size: 64 pages
Generic RTC Driver v1.07
serio: i8042 AUX port at 0x60,0x64 irq 12
serio: i8042 KBD port at 0x60,0x64 irq 1
Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
tg3.c:v3.37 (August 25, 2005)
PCI: Enabling device 0000:00:09.0 (0014 -> 0016)
PCI: IRQ 0 for device 0000:00:09.0 doesn't match PIRQ mask - try pci=usepirqmaskPCI: setting IRQ 9 as level-triggered
PCI: Assigned IRQ 9 for device 0000:00:09.0
eth0: Tigon3 [partno(BCM95702A20) rev 1002 PHY(5703)] (PCI:33MHz:32-bit) 10/100/1000BaseT Ethernet 00:e0:18:b6:6b:c7
eth0: RXcsums[1] LinkChgREG[0] MIirq[0] ASF[0] Split[0] WireSpeed[1] TSOcap[1]
eth0: dma_rwctrl[763f0000]
Here the pci=usepirqmask does not fix the problem.
Thanks for any solution!

-Yenya

--
| Jan "Yenya" Kasprzak <kas at {fi.muni.cz - work | yenya.net - private}> |
| GPG: ID 1024/D3498839 Fingerprint 0D99A7FB206605D7 8B35FCDE05B18A5E |
| http://www.fi.muni.cz/~kas/ Journal: http://www.fi.muni.cz/~kas/blog/ |
>>> $ cd my-kernel-tree-2.6 <<<
>>> $ dotest /path/to/mbox # yes, Linus has no taste in naming scripts <<<
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/