I have a Philips/LMS CM 206 cd-rom drive. The CD-Rom howto says that
the Philips/LMS cd-rom drives work with a kernel patch. However I cant
find that patch. Anyone know where I can find it?
The patch is on sunsite in the /pub/Linux/kernel/patches directory if I
recall. However, the patch only works for the CM 205 drive. The 206
is apparently a completely different beastie.
I've just picked up a Philips LMS 206 double speed driver (Fry's was
selling them for $89). After an afternoon of fun setting jumpers,
seating boards and savoring the joys of DOS once again, I managed to
get DOS to recognize the CD-ROM as the D: drive.
On the linux side, I upgraded to the 1.0 kernel (where do you find out
which patch level?) and folded in the patches for the lmscd driver.
When I boot this hybridized kernel, it hangs when trying to recognize
the CD-ROM. Turning on debugging showed that lms_interrupt() was
getting called repeatedly (> 1000 times/sec) with the cd state == 0.
The 1.0 kernel without the CD-ROM driver works just fine.
While I may have made a mistake in applying the patches by hand, I
doubt that this is the problem --- mis-applying the patches would more
likely have caused build problems due to undefined device numbers and
Could anyone either point me at a 206 driver or enough technical info
on the differences between the 205 and 206 to figure out how to patch
Unfortunately, I don't have a 205 handy, so I can't tell you whether
my mutant kernel will, in fact, drive a 205. I suspect it would,
since the 205 driver is compatible with both pre and post 1.0. In
other words, the only problems should be in getting it to build in the
first place. I would be glad to beta (or alpha) test someone else's
attempt --- the 205 driver is nice and clean, it just doesn't seem to
recognize the 206.
Other cards in the system: IDE controller/IO, extra serial port,
SoundBlaster Pro, SVGA, internal modem. I haven't tried rebooting
linux with all these pulled out, but I did have all but the IDE and
SVGA out while mucking with DOS. Didn't seem to make a difference.
Also, I doubt it's an IRQ or IO address conflict, since DOS works (if
you call that working) and the driver is able to get the IRQ when it
requests it from the kernel.
| Easily amused
| Perpetually confused