Re: frlock and barrier discussion

[Andrea Arcangeli <andrea@suse.de>]

On Wed, Jan 29, 2003 at 05:15:55PM -0800, Stephen Hemminger wrote:
> On Tue, 2003-01-28 at 23:06, Richard Henderson wrote:
> > On Tue, Jan 28, 2003 at 03:42:21PM -0800, Stephen Hemminger wrote:
> > > +static inline void fr_write_begin(frlock_t *rw)
> > > +{
> > > +	preempt_disable();
> > > +	rw->pre_sequence++;
> > > +	wmb();
> > > +}
> > > +
> > > +static inline void fr_write_end(frlock_t *rw)
> > > +{
> > > +	wmb();
> > > +	rw->post_sequence++;
> > 
> > These need to be mb(), not wmb(), if you want the bits in between
> > to actually happen in between, as with your xtime example.  At
> > present there's nothing stoping xtime from being *read* before
> > your read from pre_sequence happens.
> 
> 
> First, write_begin/end can only be safely used when there is separate
> writer synchronization such as a spin_lock or semaphore.  
> As far as I know, semaphore or spin_lock guarantees a barrier.
> So xtime or anything else can not be read before the spin_lock.
> 
> Using mb() is more paranoid than necessary. 

yes, it should only generate a superflous lock on x86.

it shouldn't even be necessary in fr_write_trylock.

Andrea