Re: [PATCH] introduce sys_membarrier(): process-wide memory barrier (v10)

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

On Mon, Apr 05, 2010 at 01:23:55PM -0700, Randy Dunlap wrote:
> Paul E. McKenney wrote:
> > On Mon, Apr 05, 2010 at 03:10:57PM -0400, Mathieu Desnoyers wrote:
> >> * Randy Dunlap (randy.dunlap@oracle.com) wrote:
> >>> On Mon, 5 Apr 2010 13:57:37 -0400 Mathieu Desnoyers wrote:
> > 
> > [ . . . ]
> > 
> >>>> +#else /* #ifdef CONFIG_SMP */
> >>> I don't know that we have a known convention for that, but I would use:
> >>>
> >>> #else /* not CONFIG_SMP */
> >>>
> >>> or
> >>>
> >>> #else /* !CONFIG_SMP */
> >>>
> >>>> +
> >>>> +SYSCALL_DEFINE1(membarrier, unsigned int, flags)
> >>>> +{
> >>>> +	return 0;
> >>>> +}
> >>>> +
> >>>> +#endif /* #else #ifdef CONFIG_SMP */
> 
> or just:
> 
> #endif /* #else #ifdef CONFIG_SMP : tell the reader that the #else part of the #ifdef CONFIG_SMP just ended */
> 
> ad nauseum.

You lost me on this one.

> >>> and:
> >>>
> >>> #endif /* CONFIG_SMP */
> >>>
> >>> The "#else #ifdef" is both ugly and too wordy IMO.
> > 
> > The extra words make it very clear that we are in at the end of the #else
> > clause of a #ifdef with the given condition.  With "#endif /* CONFIG_SMP
> > */", is the immediately preceding code compiled under CONFIG_SMP or
> > !CONFIG_SMP?  You have to dig back and see whether or not there is a
> > #else clause.
> > 
> > But there is no accounting for taste.  ;-)
> 
> IYHO.

Indeed, in both directions.

							Thanx, Paul