Messages in this thread | | | Date | Tue, 24 May 2016 18:57:41 +0200 | From | Peter Zijlstra <> | Subject | Re: [RFC][PATCH 2/3] locking: Annotate spin_unlock_wait() users |
| |
On Tue, May 24, 2016 at 09:17:13AM -0700, Linus Torvalds wrote:
> This needs to be either hidden inside the basic spinlock functions, > _or_ it needs to be a clear and unambiguous interface. Anything that > starts talking about control dependencies is not it. > > Note that this really is about naming and use, not about > implementation. So something like "spin_sync_after_unlock_wait()" is > acceptable, even if the actual _implementation_ were to be exactly the > same as the "after_ctrl_dep()" crap.
OK; so I would prefer to keep the smp_acquire__after_ctrl_dep() crap for common use in smp_cond_acquire() and such, but I'd be more than happy to just stuff it unconditionally into spin_unlock_wait().
Most users really need it, and its restores intuitive semantics to the primitive.
I'm assuming the explicit use then left in ipc/sem.c (as paired with the spin_is_locked) is fine with you; that's certainly not driver code.
Todays series was really more about auditing all the spin_unlock_wait() usage sites.
| |