* Linus Torvalds <torvalds@linux-foundation.org> wrote:

quoted text
> On Wed, Nov 17, 2010 at 2:05 PM, Randy Dunlap <randy.dunlap@oracle.com> wrote:
> >
> > smp_lock.h was removed from hardirq.h.  smp_lock.h provided the function prototype
> > for kernel_locked().  Should source files now #include <linux/smp_lock.h> ?
> > even when not being built for SMP?
> 
> Hmm. I think that part was a mistake, but I suspect the simplest fix
> for it is to simply get rid of "kernel_locked()". It has no other
> users than the hardirq.h one, so let's just move it there.
> 
> Something like the attached?
> 
> NOTE! The reason I _only_ take the CONFIG_LOCK_KERNEL version from
> smp_lock.h is because:
> 
>  - LOCK_KERNEL is defined by init/Kconfig as "(SMP || PREEMPT) && BKL"
> 
>  - inside hardirq.h we only use "kernel_locked()" inside "PREEMPT && BKL"
> 
>  - so "PREEMPT && BKL" implies "LOCK_KERNEL"
> 
>  - so the !LOCK_KERNEL kernel_locked() case is irrelevant.
> 
> unless I did a thinko somewhere.
> 
> Does this work in all configurations? TOTALLY UNTESTED! Caveat emptor.

Latest -git fails to build in some circumstances:

 drivers/net/irda/sir_dev.c: In function ‘sirdev_schedule_request’:
 drivers/net/irda/sir_dev.c:292:244: error: dereferencing pointer to incomplete type

Caused by these patches:

 0a5b871ea4c6: hardirq.h: remove now-empty #ifdef/#endif pair
 7957f0a85775: Fix build failure due to hwirq.h needing smp_lock.h
 460781b54253: BKL: remove references to lock_kernel from comments
 451a3c24b013: BKL: remove extraneous #include <smp_lock.h>

I've attached the .config.

It's a problem with one of these:

                if (in_interrupt()  ||  in_atomic()  ||  irqs_disabled()) {

(Havent checked it yet what the problem is.)

Thanks,

	Ingo