Pioneer DVR-106D issues
Well, it seemed like a good idea to sort of follow-up on my previous message,
so here it is.
Unlikely what Andy seemed to think, I really wasn't asking this list for
kenrel support. I was actually trying to find out if anyone else was having
the same problem. That problem is under ide-cd burning a dvd-r at 0.40x
speed instead of the stated 4.00x accompanied with 1 to 2 CPU spikes per
second. Cpu usage between spikes is near nil. Also, the mouse cursor
movement will stutter from time to time. Music doesn't skip. This problem
occurs both in growisofs and with cdrecord (the latter I only tested with a
completed iso/udf image so the spikes were only 50% cpu instead of the usual
100%. I'd test with cdrecord and mkisofs to see if it behaves exactly
growisofs if I knew how).
I'm using the linux kernel v2.4.21 + a group of patches (lm_sensors, bttv,
bootsplash, supermount, etc). The drive is *not* mounted under supermount.
Dma is enabled. My chipset information is:
00:00.0 Host bridge: Silicon Integrated Systems [SiS] 735 Host (rev 01)
00:02.5 IDE interface: Silicon Integrated Systems [SiS] 5513 [IDE] (rev d0)
So, if you have a Pioneer DVD-106D drive (revision 107 or otherwise) and your
drive works full speed on ide-cd or runs horrible slow, I'd like to know.
Please tell me what kernel you're using, what revision of firmware, and
obviously the speed of burning a dvd-r. Oh, and knowing the chipset would
probably be useful too. As it stands, I've yet to hear any other users of
this type of drive in Linux. :(
PS: Andy, I guess I didn't make it clear in my original post but the system
was kernel panicing. Since I never was at console all those times, I never
had seen the actuall OOPs, but caps and scroll lock were flashing which told
me it was a kernel panic. After your comment, though, I scratched down the
OOPs dumped on console from burning a dvd-r. The same results occur when
mounting a badly burned dvd-r, for which I posted onto the lkml, if
interested. In either case, the cause was a failure in the ide-scsi reset
function. A hack fix was to just return SCSI_RESET_SNOOZE. The only problem
is that every 3rd or 4th burn's end the drive will stay stuck with its access
light on requiring a reboot to fix.. So, I'm sticking with ide-cd even if
it's horribly slow. And no, I don't expect a fix to be posted here. I'm
stating this information purely for completeness. Besides, Linus says
ide-scsi is a bad hack.
lkml post: http://www.ussg.iu.edu/hypermail/linux/kernel/0311.2/0073.html