Re: [RFC][PATCH 4/4] Rename lock_cpu_hotplug/unlock_cpu_hotplug

[Nick Piggin <nickpiggin@yahoo.com.au>]

Arjan van de Ven wrote:
> On Thu, 2006-08-24 at 19:33 +0530, Gautham R Shenoy wrote:
>
> > On Thu, Aug 24, 2006 at 01:00:00PM +0200, Arjan van de Ven wrote:
> >
> > > On Thu, 2006-08-24 at 16:04 +0530, Gautham R Shenoy wrote:
> > >
> > > > This patch renames lock_cpu_hotplug to cpu_hotplug_disable and
> > > > unlock_cpu_hotplug to cpu_hotplug_enable throughout the kernel.
> > >
> > > Hi,
> > >
> > > to be honest I dislike the new names too. You turned it into a refcount,
> > > which is good, but the normal linux name for such refcount functions is
> > > _get and _put.....  and in addition the refcount technically isn't
> > > hotplug specific, all you want is to keep the kernel data for the
> > > processor as being "used", so cpu_get() and cpu_put() would sound
> > > reasonable names to me, or cpu_data_get() cpu_data_put().
> >
> > Thus, choice of 'cpu_hotplug_disable' and 'cpu_hotplug_enable'
> > was determined on the basis of its purpose, as in *what* it does 
> > as opposed to *how* it does it. :)
>
>
> well.. it comes down to the difference of locking to protect data versus
> locking to protect against a specific piece of code. Almost always the
> later turns out to be a mistake...

But it is not protecting a cpu from going away, it is protecting ALL
cpus from coming or leaving. In that respect it is much more like a
cpu_online_map lock rather than a data structure refcount.

It really is just like a reentrant rw semaphore... I don't see the
point of the name change, but I guess we don't like reentrant locks so
calling it something else might go down better with Linus ;)

--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/