Re: [RFC][PATCH] Fix a race between rwsem and the scheduler

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Tue, 2016-08-30 at 20:34 +0200, Peter Zijlstra wrote:
> 
> I'm not actually sure it does. There is the comment from 8643cda549ca4
> which explain the program order guarantees.
> 
> But I'm not sure who or what would simply a full smp_mb() when you call
> schedule() -- I mean, its true on x86, but that's 'trivial'.

It's always been a requirement that if you actually context switch a
full mb() is implied (though that isn't the case if you don't actually
switch, ie, you are back to RUNNING before you even hit schedule).

On powerpc we have a sync deep in _switch to achieve that.

This is necessary so that a process who wakes up on a different CPU sees
all of its own load/stores.

> > I mean, I thought that the LOAD/STORE's done by some task can't
> > be re-ordered with LOAD/STORE's done by another task which was
> > running on the same CPU. Wrong?
> 
> If so, I'm not sure how :/
> 
> So smp_mb__before_spinlock() stops stores from @prev, and the ACQUIRE
> from spin_lock(&rq->lock) stops both loads/stores from @next, but afaict
> nothing stops the loads from @prev seeing stores from @next.
> 
> Also not sure this matters though, if they're threads in the same
> process its a data race already and nobody cares. If they're not threads
> in the same process, they're separated by address space and can't 'see'
> each other anyway.

The architecture switch_to() has to do the right thing.

Cheers,
Ben.