From mboxrd@z Thu Jan  1 00:00:00 1970
From: linux@arm.linux.org.uk (Russell King - ARM Linux)
Date: Wed, 14 Nov 2012 10:21:27 +0000
Subject: [PATCH 1/2] arm/mm: L2CC shared mutex with ARM TZ
In-Reply-To: <0154077FE026E54BB093CA7EB3FD1AE32B57AF1E29@SAFEX1MAIL3.st.com>
References: <0154077FE026E54BB093CA7EB3FD1AE32B57AF1B59@SAFEX1MAIL3.st.com>
 <20121113170711.GI28327@n2100.arm.linux.org.uk>
 <0154077FE026E54BB093CA7EB3FD1AE32B57AF1E29@SAFEX1MAIL3.st.com>
Message-ID: <20121114102127.GH3290@n2100.arm.linux.org.uk>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Wed, Nov 14, 2012 at 10:18:31AM +0100, Etienne CARRIERE ST wrote:
> > From: Russell King - ARM Linux [mailto:linux at arm.linux.org.uk] 
> > Sent: Tuesday, November 13, 2012 6:07 PM
> > 
> > Right, so, what this tells me is that the implementation of the spinlock 
> > in the secure software is potentially the same as the kernel's spinlock.
> > The kernel's spinlock implementation is GPL'd.  If the secure side spinlock 
> > implementation is a copy of the kernel's spinlock implementation, then 
> > that implementation is also GPL'd (...).
> > 
> > I think someone has some explaining to do.
> 
> It does.
> 
> The TZ implementation of the shared mutex is the ARM native mutex
> instructions set LDREX/STREX, protected by a native ARM IRQ disabling.
> I believe this is native ARMv7 natural handling and not subject to
> GPLization.

Maybe, maybe not.  There's more to locking than just the instructions
used - there's the values which are used to represent "locked" and
"unlocked", and then there's the ticket locking implementation found
in recent kernels which defines the value to have more meaning than
just "locked" and "unlocked".

With ticket locking, you need to know the details of the implementation
to be able to lock successfully, and that basically means copying the
implementation.