Re: OFFTOPIC: binary modules, bad idea!

Michael Elizabeth Chastain (
Sun, 21 Dec 1997 16:51:30 -0600

Ted Ts'o writes:

> That's not quite true. foo.ver is created when "make depend" is run,
> not when resident.c is compiled.

foo.ver also depends on resident.c. This dependency causes the ugly
massive recompilation whenever any symbol-bearing file changes.

> Yes, but they only get changed when a kernel hacker modifies them, and I
> think it's fair to require a kernel hacker to remember to run "make
> depend".

What if foo.ver depended on foo.h, rather than resident.c? That would
be a good dependency chain, but then the EXPORT_SYMBOL declarations
would need to be in there, too.

> > I do plan to run genksyms on each compile of: a file that exports symbols
> > or a file that is compiled with -DMODULE.
> Unfortunately, unless you actually use a make rule to only run genksyms
> when it is necessary, the resulting speed hit when compiling modules (of
> running genksyms all the time) is going to be quite fierce, I suspect.

I'm not worried about that. genksyms runs very fast compared to cc1.

Michael Chastain
"love without fear"