Re: [PATCH RFC 3/9] RCU: Preemptible RCU

["Paul E. McKenney" <paulmck@linux.vnet.ibm.com>]

On Mon, Oct 01, 2007 at 11:44:25AM -0700, Davide Libenzi wrote:
> On Sun, 30 Sep 2007, Paul E. McKenney wrote:
> 
> > On Sun, Sep 30, 2007 at 04:02:09PM -0700, Davide Libenzi wrote:
> > > On Sun, 30 Sep 2007, Oleg Nesterov wrote:
> > > 
> > > > Ah, but I asked the different question. We must see CPU 1's stores by
> > > > definition, but what about CPU 0's stores (which could be seen by CPU 1)?
> > > > 
> > > > Let's take a "real life" example,
> > > > 
> > > >                 A = B = X = 0;
> > > >                 P = Q = &A;
> > > > 
> > > > CPU_0           CPU_1           CPU_2
> > > > 
> > > > P = &B;         *P = 1;         if (X) {
> > > >                 wmb();                  rmb();
> > > >                 X = 1;                  BUG_ON(*P != 1 && *Q != 1);
> > > >                                 }
> > > > 
> > > > So, is it possible that CPU_1 sees P == &B, but CPU_2 sees P == &A ?
> > > 
> > > That can't be. CPU_2 sees X=1, that happened after (or same time at most - 
> > > from a cache inv. POV) to *P=1, that must have happened after P=&B (in 
> > > order for *P to assign B). So P=&B happened, from a pure time POV, before 
> > > the rmb(), and the rmb() should guarantee that CPU_2 sees P=&B too.
> > 
> > Actually, CPU designers have to go quite a ways out of their way to
> > prevent this BUG_ON from happening.  One way that it would happen
> > naturally would be if the cache line containing P were owned by CPU 2,
> > and if CPUs 0 and 1 shared a store buffer that they both snooped.  So,
> > here is what could happen given careless or sadistic CPU designers:
> 
> Ohh, I misinterpreted that rmb(), sorry. Somehow I gave it for granted
> that it was a cross-CPU sync point (ala read_barrier_depends). If that's a
> local CPU load ordering only, things are different, clearly. But ...
> 
> > o	CPU 0 stores &B to P, but misses the cache, so puts the
> > 	result in the store buffer.  This means that only CPUs 0 and 1
> > 	can see it.
> > 
> > o	CPU 1 fetches P, and sees &B, so stores a 1 to B.  Again,
> > 	this value for P is visible only to CPUs 0 and 1.
> > 
> > o	CPU 1 executes a wmb(), which forces CPU 1's stores to happen
> > 	in order.  But it does nothing about CPU 0's stores, nor about CPU
> > 	1's loads, for that matter (and the only reason that POWER ends
> > 	up working the way you would like is because wmb() turns into
> > 	"sync" rather than the "eieio" instruction that would have been
> > 	used for smp_wmb() -- which is maybe what Oleg was thinking of,
> > 	but happened to abbreviate.  If my analysis is buggy, Anton and
> > 	Paulus will no doubt correct me...)
> 
> If a store buffer is shared between CPU_0 and CPU_1, it is very likely 
> that a sync done on CPU_1 is going to sync even CPU_0 stores that are 
> held in the buffer at the time of CPU_1's sync.

That would indeed be one approach that CPU designers could take to
avoid being careless or sadistic.  ;-)

Another approach would be to make CPU 1 refrain from snooping CPU 0's
entries in the shared store queue.

							Thanx, Paul