Re: [patch] 2.4.16 for pointers to __devexit functions

Jakub Jelinek (jakub@redhat.com)
Thu, 6 Dec 2001 04:50:41 -0500


On Thu, Dec 06, 2001 at 09:40:19AM +0000, David Woodhouse wrote:
>
> davem@redhat.com said:
> > > Why not __attribute__((weak)) ?
> > This doesn't work on all platforms unfortunately :(
>
> Doesn't work at all, or just doesn't work with the (current) minimum
> recommended compiler? We have to increase those minima at some point.

Actually, I think all GCCs will error on it, or at least should.
Most of the __devexit routines are static, you get something like:
test.c:2: weak declaration of `foo' must be public

It is very weird thing to have a non-public weak, and assemblers will do
weird things if you tweak it in assembly output.

Jakub
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/