[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Pine.LNX.4.44L0.1909161122390.1489-100000@iolanthe.rowland.org>
Date: Mon, 16 Sep 2019 11:25:07 -0400 (EDT)
From: Alan Stern <stern@...land.harvard.edu>
To: Boqun Feng <boqun.feng@...il.com>
cc: LKMM Maintainers -- Akira Yokosawa <akiyks@...il.com>,
Andrea Parri <parri.andrea@...il.com>,
Daniel Lustig <dlustig@...dia.com>,
David Howells <dhowells@...hat.com>,
Jade Alglave <j.alglave@....ac.uk>,
Luc Maranget <luc.maranget@...ia.fr>,
Nicholas Piggin <npiggin@...il.com>,
"Paul E. McKenney" <paulmck@...ux.ibm.com>,
Peter Zijlstra <peterz@...radead.org>,
Will Deacon <will@...nel.org>,
Kernel development list <linux-kernel@...r.kernel.org>
Subject: Re: Documentation for plain accesses and data races
On Mon, 16 Sep 2019, Boqun Feng wrote:
> > In other words, we can define ->vis as:
> >
> > let vis = prop ; ((strong-fence ; [Marked] ; xbstar) | (xbstar & int))
> >
>
> Hmm.. so the problem with this approach is that the (xbstar & int) part
> doesn't satisfy the requirement of visibility... i.e.
>
> X ->prop Z ->(xbstar & int) Y
>
> may not guarantee when Y executes, X is already propagated to Y's CPU.
Yes, it doesn't guarantee this. But the reason it doesn't guarantee
this is because of the prop. The (xbstar & int) part is okay. In
other words, if
Z ->(xbstar & int) Y
then it is certainly true that any store propagating to Z's CPU before
Z executes also propagates to Y's CPU (which is the same one) before Y
executes.
Alan
Powered by blists - more mailing lists