[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201006144302.GY2628@hirez.programming.kicks-ass.net>
Date: Tue, 6 Oct 2020 16:43:02 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: "stern@...land.harvard.edu" <stern@...land.harvard.edu>
Cc: David Laight <David.Laight@...lab.com>,
"linux-toolchains@...r.kernel.org" <linux-toolchains@...r.kernel.org>,
Will Deacon <will@...nel.org>,
Paul McKenney <paulmck@...nel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"parri.andrea@...il.com" <parri.andrea@...il.com>,
"boqun.feng@...il.com" <boqun.feng@...il.com>,
"npiggin@...il.com" <npiggin@...il.com>,
"dhowells@...hat.com" <dhowells@...hat.com>,
"j.alglave@....ac.uk" <j.alglave@....ac.uk>,
"luc.maranget@...ia.fr" <luc.maranget@...ia.fr>,
"akiyks@...il.com" <akiyks@...il.com>,
"dlustig@...dia.com" <dlustig@...dia.com>,
"joel@...lfernandes.org" <joel@...lfernandes.org>,
"torvalds@...ux-foundation.org" <torvalds@...ux-foundation.org>
Subject: Re: Control Dependencies vs C Compilers
On Tue, Oct 06, 2020 at 10:23:24AM -0400, stern@...land.harvard.edu wrote:
> On Tue, Oct 06, 2020 at 03:31:15PM +0200, Peter Zijlstra wrote:
> > Only if we get the compiler people on board and have them provide means
> > are we guaranteed safe from the optimizer. Otherwise we'll just keep
> > playing whack-a-mole with fancy new optimization techniques. And given
> > how horridly painful it is to debug memory ordering problems, I feel it
> > is best to make sure we're not going to have to more than necessary.
>
> Given that you would have to add a compiler annotation, isn't it just as
> easy to use READ_ONCE and WRITE_ONCE?
>
> Or are you worried that even with READ_ONCE and WRITE_ONCE, the compiler
> might still somehow defeat the control dependency?
Yes.
Also, not all instances actually have the WRITE_ONCE() on. The one in
the perf ringbuffer for example uses local_read() for the load (which is
basically READ_ONCE()), but doesn't have WRITE_ONCE() on the inside.
Also, afaiu WRITE_ONCE() also doesn't stop the compiler from lifting it
if it thinks it can get away with it -- memory-barriers.txt has
examples.
And then there's the case where the common branch has a store, I know
ARM64 ARM isn't clear on that, but I'm thinking any actual hardware will
still have to respect it, and it's a matter of 'fixing' the ARM.
Mostly I just want the compiler people to say they'll guarantee the
behaviour if we do 'X'. If 'X' happens to be 'any dynamic branch headed
by a volatile load' that's fine by me.
If they want a new keyword or attribute, that's also fine. But I want to
have the compiler people tell me what they want and guarantee they'll
not come and wreck things.
Powered by blists - more mailing lists