[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <201104191810.18170.arnd@arndb.de>
Date: Tue, 19 Apr 2011 18:10:17 +0200
From: Arnd Bergmann <arnd@...db.de>
To: Mathieu Poirier <mathieu.poirier@...aro.org>
Cc: linux-arm-kernel@...ts.infradead.org, ohad@...ery.com,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] ux500: Adding support for u8500 Hsem functionality V2
On Tuesday 19 April 2011, Mathieu Poirier wrote:
> > One more thing I just noticed: the hwspinlock_internal.h file defines
> > the hwspinlock->id field as "a global, unique, system-wide, index of
> > the lock", but the u8500 hsem just sets it to an integer starting
> > at zero. If there are multiple devices providing hwspinlocks in the
> > same system, that cannot work.
> >
> I have to admit I'm not sure of what your asking here. Hwspinlocks
> should be administered by only one entity and this is what this driver
> is doing.
>
> Please get back to me with a clarification.
You could have a system with multiple instances of the hardware block
that holds your spinlock. I've seen SMP systems that basically take
multiple SOCs and put them into a common address space.
More importantly, there could be a device on an external bus that also
provides a hwspinlock. The API can deal with this, but your driver will
conflict with any other driver trying to do the same.
Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists