2.5.64-mm6: kernel BUG at kernel/timer.c:155!
----- Original Message -----
From: Jens Axboe <[email protected]>
Date: Thu, 13 Mar 2003 18:54:54 +0100
To: Jeremy Fitzhardinge <[email protected]>
Subject: Re: 2.5.64-mm6: kernel BUG at kernel/timer.c:155!
Quoted 6 lines · Jeremy Fitzhardinge
Quoted 2 lines
I was reading back a freshly burned CD from my shiny new Plexwriter 48/24/48A. I'm using ide-scsi, so this is an iso9660 filesystem mounted
out of curiousity, why? ide-cd should work much better than ide-scsi in 2.5, if it doesn't I'd like to know.
There are still userspace CD burning programs that do not yet support ATAPI burning interface. "cdrecord" does support it, but K3B (a KDE burning frontend to cdrecord and company) only works with SCSI or IDE-SCSI burners (well, or at least I have been unable to convince it to use the ATAPI interface to my Sony burner).
Best regards,
Felipe
Signature · 5 lines
______________________________________________ http://www.linuxmail.org/ Now with e-mail forwarding for only US$5.95/yr Powered by Outblaze
Quoted 13 lines · Felipe Alfaro Solana
----- Original Message -----
From: Jens Axboe <[email protected]>
Date: Thu, 13 Mar 2003 18:54:54 +0100
To: Jeremy Fitzhardinge <[email protected]>
Subject: Re: 2.5.64-mm6: kernel BUG atkernel/timer.c:155!Quoted 6 lines · Jeremy Fitzhardinge
Quoted 2 lines
I was reading back a freshly burned CD from my shiny new Plexwriter 48/24/48A. I'm using ide-scsi, so this is an iso9660 filesystem mounted
out of curiousity, why? ide-cd should work much better than ide-scsi in 2.5, if it doesn't I'd like to know.
Quoted 6 lines
There are still userspace CD burning programs that do not yet support ATAPI burning interface. "cdrecord" does support it, but K3B (a KDE burning frontend to cdrecord and company) only works with SCSI or IDE-SCSI burners (well, or at least I have been unable to convince it to use the ATAPI interface to my Sony burner).
That is quite possible. Maybe someone could compile a list of programs that still use read/write to /dev/sg*?
I had at some point planned to allow arbitrary block device -> /dev/sg* mapping, maybe it would be a good idea to finish that thought. Then you wouldn't have to rely on programs using SG_IO ioctl, which does seem like a bad idea.
Signature · 1 line
Jens Axboe