On Fri, 22 Jun 2001 11:13:17 -0600,
"D. Stimits" <stimits@xxxxxxxxxx> wrote:
>of device drivers or hardware that might be compiled, if it does
>something like #include <linux/something.h>, and something.h has changed
>in the kernel, it could cause other problems than just compile
>incompatibilities. For example, modules that are compiled separately
>from the kernel (because they are not patched in to the kernel or
>otherwise not part of the kernel source tree, such as cvs releases of
>many modules).
Seperate compilation of modules, in particular for binary only modules,
is known to be riddled with risks. Module symbol versions attempt to
detect mismatches between the kernel and the module by creating a hash
over the API of each exported symbol, if the hash does not match then
the module does not load.
Alas module symbol versions has several fundamental design flaws. The
version data is abused by third party binary modules in such a way as
to defeat the check. Modversions causes so many compile and build
problems that most people disable modversions on their own kernels. I
certainly do and I am the kernel build and modutils maintainer!
Kernel 2.5 and modutils 2.5 will have a completely new method for
checking the ABI between kernels and modules. Or it will once I find
time to do the coding.
|