Alan J. Flavell wrote:
> I installed 303 on an IBM X-series laptop.
>
> I found to my surprise that it fails to recognise the compact flash
> socket, claiming that the kernel module was not found.  On inserting
> the flash card in the socket, the messages log shows that it knows
> what it wants to load (ide-cs.o), but the module isn't available, so
> it reports that the device is "temporarily" (!) not accessible, or
> words to that effect.
>
> (This worked on RH9, by the way, and I don't recall having to do
> anything to help it along).
>
> Sure enough, the kernel module is missing: the ide-cs.c source is
> right there in the kernel source,
> /usr/src/linux-2.4.21-20.EL/drivers/ide/legacy/ide-cs.c , and the
> kernel has evidently been built with options that would result in this
> driver being built:
>
> CONFIG_BLK_DEV_IDECS=m
>
> but the .o file isn't installed.
>
> Whereas on our RH9 systems:
>
>  -rw-r--r--    1 root     root         6776 Apr 13 23:12
>  /lib/modules/2.4.20-31.9/kernel/drivers/ide/legacy/ide-cs.o
>
>
> After hunting around for it on SL: apparently the .o file is in the
> kernel-unsupported RPM, along with a whole swath of other kernel
> modules.
>
> I hadn't recognised anything on the installation-time packages
> selection menu which indicated that this might be a useful thing to
> install.  Could of course just be my lack of attention...
>
>
> And it seems that on a different laptop, the IBM R40e, the same reason
> explains why the standard sound installation didn't work: it couldn't
> find the trident.o driver for it, until I installed the
> kernel-unsupported RPM.
>
>
> I'm just posting this because it might help someone who's missing one
> or other of the drivers that are in that kernel-unsupported package,
> and are as puzzled as I was, initially, that this or that item wasn't
> working on a default install.
>
> best regards
Hi,
The driver ide-cs.o is found in kernel-unsupported.  So all you will have to
do to get it working is
   yum install kernel-unsupported

This is one of those "don't blame us, that's how redhat does it" things.  But
at least you don't have to reboot or anything to get those kernel modules working.

Troy
--
__________________________________________________
Troy Dawson  [log in to unmask]  (630)840-6468
Fermilab  ComputingDivision/CSS  CSI Group
__________________________________________________