Re: 2.6.12-rc2-mm2

From: Adrian Bunk
Date: Tue Apr 12 2005 - 19:42:24 EST


On Mon, Apr 11, 2005 at 08:18:32AM -0700, Paul E. McKenney wrote:
> On Mon, Apr 11, 2005 at 12:48:34AM +0200, Adrian Bunk wrote:
> > kernel-rcupdatec-make-the-exports-export_symbol_gpl.patch
> > add-deprecated_for_modules.patch
> > add-deprecated_for_modules-fix.patch
> > deprecate-synchronize_kernel-gpl-replacement.patch
> > deprecate-synchronize_kernel-gpl-replacement-fix.patch
> > change-synchronize_kernel-to-_rcu-and-_sched.patch
> >
> >
> > Please drop these patches.
>
> Please keep them!
>
> > Using these symbols in non-GPL modules is a legal problem at least in
> > the USA except for IBM,
>
> Again, based on what line of reasoning? Again, the obvious lines
> of reasoning do not apply.

Shouldn't the IBM patents be enough reason to prevent everyone except
IBM from using RCU in non-GPL modules?

> > and all we've heard from IBM is that they are
> > not 100% sure that there is really no binary-only module by IBM that
> > might use these symbols.
>
> >From my earlier message (http://lkml.org/lkml/2005/4/4/244):
>
> Agreed, in that I know of no binary module that uses RCU. However,
> I cannot -prove- that there is no such module.
>
> IOW, I am also not 100% sure that there is really no binary-only module
> using these symbols by -anyone-, including someone -other- than IBM.
> In addition, I know of no way that -anyone- could possibly be 100% sure
> that there is really no binary-only module using symbols. Hence the
> approach of providing the year "grace period" before transitioning to
> EXPORT_SYMBOL_GPL().
>...

If kernel development was based on the assumption that every change that
might break binary-only modules would need a one year "grace period", it
was much different from how it's today...

> Thanx, Paul

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

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