Messages in this thread | | | Date | Thu, 15 Apr 1999 15:30:21 -0400 | From | Dan Reish <> | Subject | Re: Another 2.2.5 crash, and it's not X, dammit! |
| |
On Wed, Apr 14, 1999 at 05:35:37PM +0200, Markus Fischer wrote (in a private message - I hope he doesn't mind my quoting him in public):
> Maybe you have an dma problem with one of you disks/cdrom ? > We know linux does not jet support all types of chipsets 100%. > I have some own my own which hangs the machine when i try to > mount a cdrom (dma enabled).
This machine had two drives, one on each IDE controller. hda had DMA on after booting; on hdc, DMA was off, although I was able to turn it on once with no apparent ill effects. (It was off at the time of the crash.)
A couple of days ago, I got these kernel messages:
Apr 13 14:29:23 dreish kernel: hda: timeout waiting for DMA Apr 13 14:29:23 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:29:23 dreish kernel: hda: DMA disabled Apr 13 14:29:23 dreish kernel: ide0: reset: success Apr 13 14:29:33 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:29:33 dreish kernel: ide0: reset: success Apr 13 14:29:43 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:29:43 dreish kernel: end_request: I/O error, dev 03:02 (hda), sector 424106 Apr 13 14:29:43 dreish kernel: hda: status timeout: status=0xd0 { Busy } Apr 13 14:29:43 dreish kernel: hda: drive not ready for command Apr 13 14:29:44 dreish kernel: ide0: reset: success Apr 13 14:29:54 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:29:54 dreish kernel: ide0: reset: success Apr 13 14:30:04 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:30:04 dreish kernel: end_request: I/O error, dev 03:02 (hda), sector 424108 Apr 13 14:30:04 dreish kernel: hda: status timeout: status=0xd0 { Busy } Apr 13 14:30:04 dreish kernel: hda: drive not ready for command Apr 13 14:30:04 dreish kernel: ide0: reset: success Apr 13 14:30:14 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:30:14 dreish kernel: ide0: reset: success Apr 13 14:30:24 dreish kernel: hda: irq timeout: status=0xd0 { Busy } Apr 13 14:30:24 dreish kernel: end_request: I/O error, dev 03:02 (hda), sector 424110 Apr 13 14:30:24 dreish kernel: hda: status timeout: status=0xd0 { Busy } Apr 13 14:30:24 dreish kernel: hda: drive not ready for command Apr 13 14:30:24 dreish kernel: ide0: reset: success
At which point, I immediately shut down and replaced the drive, which was getting too small anyway. It was two years old, and had been running 24/7, so I guess this supports both the hardware failure theory and this DMA bug theory.
Some information on the dearly departed disk:
Apr 13 13:48:13 dreish kernel: hda: WDC AC22500L, ATA DISK drive Apr 13 13:48:13 dreish kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 Apr 13 13:48:13 dreish kernel: hda: WDC AC22500L, 2441MB w/256kB Cache, CHS=620/128/63, DMA
And I don't know how useful this is, but here's what /proc provides about the controller, ide0:
# cat /proc/ide/ide0/config pci bus 00 device 39 vid 8086 did 7010 channel 0 86 80 10 70 05 00 80 02 00 80 01 01 00 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 01 f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 77 e3 07 a3 09 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 10 0f 00 00 00 00 00 00 # cat /proc/ide/ide0/channel 0 # cat /proc/ide/ide0/mate ide1 # cat /proc/ide/ide0/model pci
-- Dan
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |