Hi,
I've put up a modified cd patch set against 2.4.0-test12-pre7. The
changes are roughly as follows:
o Fix SCSI CD-ROM on fs with < 2KB block size. I've briefly tested
with ext2 (1KB) and msdos (512b) and it appears to work. Would
the HFS folks try this too?
o Per command timeout for CD-ROM generic packet
o Per command quiet bit, as not to print a lot of sense stuff when
we know that something may fail
o Major CDROM_SEND_PACKET cleanup
o ide-cd shut up stuff (- sense logging in some cases)
o Make sure that cdrom_sleep actually does that, sleeps
o Let sr retry a PLAYTRKIND command with PLAYMSF for ide-scsi
and ATAPI.
o Minor sr cleanups
*.kernel.org/pub/linux/kernel/people/axboe/patches/2.4.0-test12-pre7/cd-2.bz2
In addition I made another small fix after uploading this one. Audio
ripping can fail with the vm failing to allocate enough contigous
pages, if ripping programs specify large bites of frames to be ripped.
Always default to allocating just a single page.
*.kernel.org/pub/linux/kernel/people/axboe/patches/2.4.0-test12-pre7/cd-2-cdda_alloc.bz2
--
* Jens Axboe <[email protected]>
* SuSE Labs
Hi Jens,
I have tested your latest stuff (cd-2) with the CD that was
causing problems before. The problems still occur.
The CD plays fine, but I get the following errors in
my /var/log/messages:
Dec 8 06:05:29 agate kernel: hdc: packet command error: status=0x51 {
DriveReady SeekComplete Error }
Dec 8 06:05:29 agate kernel: hdc: packet command error: error=0x50
Dec 8 06:05:29 agate kernel: ATAPI device hdc:
Dec 8 06:05:29 agate kernel: Error: Illegal request -- (Sense key=0x05)
Dec 8 06:05:29 agate kernel: Invalid field in command packet --
(asc=0x24, ascq=0x00)
Dec 8 06:05:29 agate kernel: The failed "Play Audio MSF" packet
command was:
Dec 8 06:05:29 agate kernel: "47 00 00 00 02 00 3f 24 ff 00 00 00 "
Dec 8 06:05:29 agate kernel: Error in command packet byte 8 bit 0
Dec 8 06:05:29 agate kernel: Play from track 1 to 9
Dec 8 06:05:29 agate kernel: lba 0 to lba 286050
Here's a pointer to the CD that is causing trouble. Perhaps
you could buy a copy and sort this out. It's a great CD!
http://www.amazon.com/exec/obidos/ASIN/B00000DDPZ
Thanks,
Miles
On Fri, Dec 08 2000, Miles Lane wrote:
> Hi Jens,
>
> I have tested your latest stuff (cd-2) with the CD that was
> causing problems before. The problems still occur.
> The CD plays fine, but I get the following errors in
> my /var/log/messages:
>
> Dec 8 06:05:29 agate kernel: hdc: packet command error: status=0x51 {
> DriveReady SeekComplete Error }
> Dec 8 06:05:29 agate kernel: hdc: packet command error: error=0x50
> Dec 8 06:05:29 agate kernel: ATAPI device hdc:
> Dec 8 06:05:29 agate kernel: Error: Illegal request -- (Sense key=0x05)
> Dec 8 06:05:29 agate kernel: Invalid field in command packet --
> (asc=0x24, ascq=0x00)
> Dec 8 06:05:29 agate kernel: The failed "Play Audio MSF" packet
> command was:
> Dec 8 06:05:29 agate kernel: "47 00 00 00 02 00 3f 24 ff 00 00 00 "
^^
This should not happen, in fact it can't happen. That byte is the ending
frame, which is always modulo 75 (number of frames in a second). So how
this ends up 0xff is a mystery. I'll investigate...
--
* Jens Axboe <[email protected]>
* SuSE Labs
> Dec 8 06:05:29 agate kernel: hdc: packet command error: status=0x51 {
> DriveReady SeekComplete Error }
> Dec 8 06:05:29 agate kernel: hdc: packet command error: error=0x50
> Dec 8 06:05:29 agate kernel: ATAPI device hdc:
> Dec 8 06:05:29 agate kernel: Error: Illegal request -- (Sense key=0x05)
> Dec 8 06:05:29 agate kernel: Invalid field in command packet --
> (asc=0x24, ascq=0x00)
> Dec 8 06:05:29 agate kernel: The failed "Play Audio MSF" packet
> command was:
> Dec 8 06:05:29 agate kernel: "47 00 00 00 02 00 3f 24 ff 00 00 00 "
I think I've seen that as well, though I'm using a test10-ish kernel. I
assumed it was something to do with the CD being duff though.
David