Even if I exit all other CD-ROM-accessing processes and also open and
close the tray (for the purpose of possibly resetting some kernel IDE
registers) prior to starting up cdda2wav, the same thing happens.
* cdda2wav "hangs" immediately after startup as soon as the drive
has spun up.
* "hdc: lost interrupt" occurs and repeats with a 10 sec. interval on
active console and in log.
* The HDD-activity-LED stays illuminated.
* The CD-ROM-drive keeps spinning, with its tray locked (no response on
eject-button), but its activity-LED is not illuminated.
* The cdda2wav process sleeps uninterruptibly in "down_failed", as
do other CD-ROM-accessing processes (either started up before or
after the malfunction). None of these processes die on SIGKILL.
* System load stabilizes around 1 when the system is otherwise idle,
or 2 or more if other processes are sleeping uninterruptibly as
well. CPU-utilization remains normal (~98% idle as reported by top).
I have no problem using cdda2wav as long as I have not done audio-CD
playback first. Doing audio-CD playback after successful use of
cdda2wav is neither a problem (but then again cdda2wav does not work
after that).
I also tried 2.2.13 in the hope that the patches in
drivers/block/ide-cd.[ch] and drivers/block/ide.c (some 200 lines away
from where the "lost interrupt" message is produced) would do the
thing, but it did not help.
My machine is a Pentium 200 MMX with 64 MB RAM, a SAMSUNG "24XMax"
CD-ROM-drive, and Creative SB AWE64 soundcard.
cdda2wav is version 1.0-i686-pc-linux-gnu
>From dmesg:
[...]
| PCI: PCI BIOS revision 2.10 entry at 0xfb0b0
| PCI: Using configuration type 1
| PCI: Probing PCI hardware
[...]
| PIIX3: IDE controller on PCI bus 00 dev 39
| PIIX3: not 100% native mode: will probe irqs later
| ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:pio, hdb:pio
| ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
| hda: IBM-DHEA-38451, ATA DISK drive
| hdc: SAMSUNG CD-ROM SCR-2431, ATAPI CDROM drive
| ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
| ide1 at 0x170-0x177,0x376 on irq 15
| hda: IBM-DHEA-38451, 8063MB w/472kB Cache, CHS=1027/255/63, (U)DMA
| hdc: ATAPI 20X CD-ROM drive, 382kB Cache
| Uniform CDROM driver Revision: 2.56
[...]
This is the contents of the files in /proc/ide/ide1/hdc:
/proc/ide/ide1/hdc/capacity: 2147483647
/proc/ide/ide1/hdc/driver: ide-cdrom version 4.54
/proc/ide/ide1/hdc/media: cdrom
/proc/ide/ide1/hdc/model: SAMSUNG CD-ROM SCR-2431
/proc/ide/ide1/hdc/settings:
| name value min max mode
| ---- ----- --- --- ----
| breada_readahead 4 0 127 rw
| dsc_overlap 0 0 1 rw
| file_readahead 124 0 2097151 rw
| io_32bit 0 0 3 rw
| keepsettings 0 0 1 rw
| max_kb_per_request 64 1 127 rw
| nice1 1 0 1 rw
| pio_mode write-only 0 255 w
| slow 0 0 1 rw
| unmaskirq 0 0 1 rw
| using_dma 1 0 1 rw
Reading /proc/ide/ide1/hdc/identify caused another uninterruptible
sleep, and now (after having performed yet another boot) I realize
that it is probably uninteresting anyway.
I guess I could do some more testing with different parameters, but it
is a bit annoying having to boot (and fsck /home which is not
unmounted on shutdown due to (I guess) unkillable processes having
locks on it) between every test.
It seems to me as if this malfunction is partly caused by the specific
behavior of my CD-ROM-drive as this does not happen on my sister's
machine with a pretty modern DVD-capable drive. But unless my drive is
irreconcilably strange or I have somehow misconfigured my kernel (or
BIOS) severely, I guess this is something that should be looked into.
Hope someone has a clue as to how this could be fixed.
Sincerely,
Knut Aksel Røysland
-
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/