[Discuss] new ide interrupt problem

[Jeff East]

Things just started stopping.  I'd get up to a certain point in the kernel
boot, but then it would complain that hda had interrupt timeouts - dma, I
think.  Based on a hint in the bootmessages, I added "irqpoll", and am
currently running.  But I have the following:


Probing IDE interface ide0...
hda: MAXTOR 6L060J3, ATA DISK drive
hdb: LITE-ON DVDRW SHM-165H6S, ATAPI CD/DVD-ROM drive
ide0 at 0xb000-0xb007,0xa802 on irq 18
Probing IDE interface ide1...
hda: max request size: 128KiB
hda: 117266688 sectors (60040 MB) w/1819KiB Cache, CHS=65535/16/63, UDMA
(33) hda: cache flushes supported
 hda: hda1 hda2 hda3 < hda5 hda6 hda7 >
hdb: ATAPI 48X DVD-ROM DVD-R-RAM CD-R/RW drive, 2048kB Cache, UDMA(33)
Uniform CD-ROM driver Revision: 3.20
hdb: cdrom_pc_intr: The drive appears confused (ireason = 0x01). Trying to
recover by ending request.
hdb: cdrom_pc_intr: The drive appears confused (ireason = 0x01). Trying to
recover by ending request.
hdb: cdrom_pc_intr: The drive appears confused (ireason = 0x01). Trying to
recover by ending request.
...
irq 18: nobody cared (try booting with the "irqpoll" option)
 [<c0134466>] __report_bad_irq+0x36/0x7f
 [<c0134534>] note_interrupt+0x69/0x9c
 [<c0134b57>] handle_fasteoi_irq+0x81/0xa0
 [<c0104b89>] do_IRQ+0x5d/0x74
 [<c021c23d>] ide_intr+0x77/0x11f
 [<c0217510>] skge_intr+0x19/0x1c3
 [<c01032eb>] common_interrupt+0x23/0x28
 [<c0133cbd>] handle_IRQ_event+0x16/0x3f
 [<c0134c3e>] handle_edge_irq+0xc8/0x106
 [<c0104b89>] do_IRQ+0x5d/0x74
 [<c01032eb>] common_interrupt+0x23/0x28
 [<c0100dec>] mwait_idle_with_hints+0x34/0x38
 [<c0100cb4>] cpu_idle+0x5e/0x73
 [<c03ba7eb>] start_kernel+0x19e/0x1a2
 [<c03ba293>] unknown_bootoption+0x0/0x181
 =======================
handlers:
[<c021c1c6>] (ide_intr+0x0/0x11f)
Disabling IRQ #18


I had reset BIOS settings, but have changed nothing else, or taken the
cover off the machine.  From what I can tell, I might have a bad IRQ
signal coming from the hardware.  "/proc/interrupts" says:

 18:     100000          0   IO-APIC-fasteoi   ide0

and that CPU0 number is not changing.  Right now everything seems to be
working, including seti/boinc and hdparm.  Ideas desired.

---
One more time:	Plain-text emails good; HTML emails bad!
		http://expita.com/nomime.html