I've noticed when compiling 2.5.2 that certain modules (such as
via82cxxx_audio) do not want to compile. After a few hours of digging, I
noticed that it seems as if the method for dealing with i_rdev has
changed. I eventually got my code to compile (and, as a matter of fact,
ALSA and NVIDIA's driver) by amending all instances of i_rdev to
i_rdev.value, but I'll be the first to admit that's my first "kernel
patch" (is it worthy of such a name?) and though the modifications seem
to produce a flawlessly working kernel I'm still tinkering with cogs in
a giant machine.
That said, if my modification is not an example of human stupidity in
action, I'd like to notify those reading that if you have any similar
places that must be cleaned, I've crafted a simple perl script to patch
sources accordingly.
http://tucb.com/chase/fix-i_rdev.html
Commentary is appreciated. But more importantly - why did this change?
(Please CC all replies to this message to [email protected] as I
am not yet a member of this mailing list [hides in corner])
Thanks,
Chase Venters
On Sat, 2 Mar 2002, Chase Venters wrote:
> I've noticed when compiling 2.5.2 that certain modules (such as
Please try again with 2.5.6-pre2. Unless you can say "it still happens
with the latest 2.5 release (or better: prerelease)" people here will not
pay too much attention to it. For example: ALSA made it into the
mainstream 2.5 kernel in 2.5.5-pre1. This means that any problem related
to ALSA in kernels before 2.5.5-pre1 will be ignored here, unless the
problems are still there.
Developping the linux kernel sure is life in the fast lane...
Jos