CD Mounting Errors (ISO9660)
 

Part of the output from dmesg:

attempt to access beyond end of device
sr0: rw=0, want=1341044, limit=1311284
ISOFS: unable to read i-node block
isofs_fill_super: get root inode failed

------------------------------------------

hdparm /dev/sr0

/dev/sr0:
multcount = 0 (off)
IO_support = 0 (default)
readonly = 0 (off)
readahead = 256 (on)
HDIO_GETGEO failed: Inappropriate ioctl for device

----------------------------------------------------

A similar problem has plagued me in the past only to raise its ugly head again. The WWW is filled with similar problems. First, my drive and discs are fine and work perfectly under BW64 13.0/Slackware 13.0. Obviously this problem is related to the software, likely the kernel. Physically shutting down the computer and waiting about 10 seconds and then rebooting is the ONLY way to mount a CD (only the 1st time, however) without this error. I'm using the stock huge kernel so I suspect switching to a generic kernel or rebuilding the stock kernel to allow me access to the disk driver modules might allow me to physically unload and then reload the modules without having to cut the power to the computer.

Any ideas? As it is, Slackware64 13.1 is basically useless unless some of these bugs can be identified and eliminated.

Some indications that ATI's binary driver might be causing this problem. I do have Catalyst 10.5 installed... simply conjecture at this point.

Well, for giggles, I blacklisted the fglrx driver while leaving ATI's OpenGL libraries installed. No difference as I expected.

Whatever is causing this problem: HDIO_GETGEO failed: Inappropriate ioctl for device

Critical bug:

Bump!

Sorry, I do not have any actual help except to say that I have three systems now installed with Slackware 13.1 and I have no CD problems.

Two are 32 bit the other is 64 bit... all CD/DVDs mount fine here.

Count yourself fortunate, because this bug is a major and critical one.

Don't use the huge kernel, compile your own or use generic.

There's few more threads about this issue in slack LQ. It seems like only sata dvd drives are affected.

If that's the case try putting the SATA controller in AHCI mode in the BIOS options. This may help because this will force it to use the ahci driver over whatever SATA or PATA driver it might be currently using, which may have a bug in it.

Hi,

I still think it would be best if the 'OP' would switch to the generic instead of using a installer kernel.
:hattip:

BTW, I AM using the -generic 2.6.33.4 kernel, but the problem remains. I'm hoping an upgrade to a newer kernel in the 2.6.33.x series beyond 2.6.33.5 will fix this problem. This problem might not be related to the kernel, though.

IIRC *, my BIOS is pretty restrictive in changing parameters of the drives. I do remember that this particular DVD recorder doesn't use SMART, however. I'll do more checking. Thanks for any clues.

* I just checked the BIOS. The ONLY options are to either enable or disable the SATA controller which naturally isn't an option. ;) This PC came with a PATA controller and hard drive along with a SATA controller and DVD recorder. On the main menu of the BIOS, selecting the DVD recorder only brings up a screen whose options are NOT configurable, i.e. the options are greyed out. One of the greyed out options states the DVD recorder isn't SMART compatible. No reference to AHCI can be found in the BIOS.

Hi,
By the above statement it did seem you were still using the 'huge' instead of the 'generic'.

No where until now did we know you are using the generic kernel.

What does your 'mkinitrd' line look like?

What is the motherboard model, BIOS version & manufacture?
:hattip:

I am pretty sure this is not a kernel problem. I have tried the huge-smp kernel in Slack 13.1, and I tried my own custom 2.6.34 kernel, no difference. I also am pretty sure this is not a AHCI vs ATA native issue, I've tried both ways, same result.

I still think it's something in hal/udev.

I tend to agree. When I first noticed this horrific bug, I noticed that the CD/DVD seemed to only mount properly after a cold boot and ONLY the first time. * So, it seems logical to look for the bug in software dealing with the mounting/umounting of optical devices (removable) attached to a SATA controller. I also have a hard drive attached to this SATA controller, but I don't use it when I'm booted into SW64 13.1.

* I'm not 100% positive about the accuracy of this statement, though, from my initial tests only.

Ok, now the only option left is compiling a newer kernel, that's the only thing I can think of that might fix it.

What SATA controller do you have? I have an ASUS SATA DVD drive and do not see this problem at all. lspci shows the controller to be...