[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- To: "Paul E. McKenney" <paulmck@xxxxxxxxxxxxxxxxxx>
- From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
- Date: Wed, 27 Jan 2016 09:39:05 +0100
- Cc: linux-mips@xxxxxxxxxxxxxx, linux-ia64@xxxxxxxxxxxxxxx, "Michael S. Tsirkin" <mst@xxxxxxxxxx>, Will Deacon <will.deacon@xxxxxxx>, virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx, "H. Peter Anvin" <hpa@xxxxxxxxx>, sparclinux@xxxxxxxxxxxxxxx, Ingo Molnar <mingo@xxxxxxxxxx>, linux-arch@xxxxxxxxxxxxxxx, linux-s390@xxxxxxxxxxxxxxx, Russell King - ARM Linux <linux@xxxxxxxxxxxxxxxx>, user-mode-linux-devel@xxxxxxxxxxxxxxxxxxxxx, linux-sh@xxxxxxxxxxxxxxx, Michael Ellerman <mpe@xxxxxxxxxxxxxx>, x86@xxxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxxx, Ingo Molnar <mingo@xxxxxxx>, linux-xtensa@xxxxxxxxxxxxxxxx, james.hogan@xxxxxxxxxx, Arnd Bergmann <arnd@xxxxxxxx>, Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>, adi-buildroot-devel@xxxxxxxxxxxxxxxxxxxxx, Leonid Yegoshin <Leonid.Yegoshin@xxxxxxxxxx>, ddaney.cavm@xxxxxxxxx, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, linux-metag@xxxxxxxxxxxxxxx, linux-arm-kernel@xxxxxxxxxxxxxxxxxxx, Andrew Cooper <andrew.cooper3@xxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, Ralf Baechle <ralf@xxxxxxxxxxxxxx>, Joe Perches <joe@xxxxxxxxxxx>, linuxppc-dev@xxxxxxxxxxxxxxxx, David Miller <davem@xxxxxxxxxxxxx>
- Delivery-date: Wed, 27 Jan 2016 08:39:35 +0000
- List-id: Xen developer discussion <xen-devel.lists.xen.org>
On Tue, Jan 26, 2016 at 12:13:39PM -0800, Paul E. McKenney wrote:
> On Tue, Jan 26, 2016 at 11:19:27AM +0100, Peter Zijlstra wrote:
> > So isn't smp_mb__after_unlock_lock() exactly such a scenario? And would
> > not someone trying to implement RCsc locks using locally transitive
> > RELEASE/ACQUIRE operations need exactly this stuff?
> >
> > That is, I am afraid we need to cover the mix of local and global
> > transitive operations at least in overview.
>
> True, but we haven't gotten to locking yet.
The mythical smp_mb__after_release_acquire() then ;-)
(and yes, I know you're going to say we don't have that)
> That said, I would argue
> that smp_mb__after_unlock_lock() upgrades locks to transitive, and
> thus would not be an exception to the "no combining transitive and
> non-transitive steps in cycles" rule.
But But But ;-) It does that exactly by combining. I suspect this is
(partly) the source of your SC chains with one PC link example.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel
- References:
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
- Re: [Xen-devel] [v3,11/41] mips: reuse asm-generic/barrier.h
|