Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 6 Mar 2019 17:58:31 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Akira Yokosawa <akiyks@gmail.com>
Cc:     "Paul E. McKenney" <paulmck@linux.ibm.com>,
        Borislav Petkov <bp@alien8.de>,
        Andrea Parri <andrea.parri@amarulasolutions.com>,
        linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
        Alan Stern <stern@rowland.harvard.edu>,
        Will Deacon <will.deacon@arm.com>,
        Boqun Feng <boqun.feng@gmail.com>,
        Nicholas Piggin <npiggin@gmail.com>,
        David Howells <dhowells@redhat.com>,
        Jade Alglave <j.alglave@ucl.ac.uk>,
        Luc Maranget <luc.maranget@inria.fr>,
        Daniel Lustig <dlustig@nvidia.com>
Subject: Re: [RFC PATCH] tools/memory-model: Remove (dep ; rfi) from ppo
Message-ID: <20190306165831.GZ32477@hirez.programming.kicks-ass.net>
References: <20190226093009.GS32477@hirez.programming.kicks-ass.net>
 <20190226104551.GF32534@hirez.programming.kicks-ass.net>
 <20190226112133.GG32534@hirez.programming.kicks-ass.net>
 <20190226112521.GH32534@hirez.programming.kicks-ass.net>
 <20190226113008.GI32534@hirez.programming.kicks-ass.net>
 <20190226113813.GA14753@zn.tnic>
 <20190226134906.GG32494@hirez.programming.kicks-ass.net>
 <20190226142845.GK4072@linux.ibm.com>
 <20190226150450.GW32477@hirez.programming.kicks-ass.net>
 <ab631777-7cbd-4658-fb99-05d60ec5a1ea@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ab631777-7cbd-4658-fb99-05d60ec5a1ea@gmail.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Mar 07, 2019 at 12:46:05AM +0900, Akira Yokosawa wrote:
> So, I'm looking at the macro RELOC_HIDE() defined in include/linux/compiler-gcc.h.

> Am I the only one who was not aware of this gcc-specific macro?

It's one I regularly see, but had forgotten about in this context.

However; you can also fix things by adding asm volatile ("":::"memory");
in places.

But that's not really the point; I would really rather have a cmdline
knob to fix things. That way we can compile the kernel with and without
and look for differences. -fno-unicorns or something :-)

While I understand some compiler people revel in UB and love to make
unicorns happen, I think in this case the produces result is utterly
insane.