From: "David Schwartz" <davids@webmaster.com>
To: "Linux-Kernel@Vger. Kernel. Org" <linux-kernel@vger.kernel.org>
Subject: RE: [PATCH 11/23] make atomic_read() and atomic_set() behavior consistent on m32r
Date: Thu, 23 Aug 2007 13:39:54 -0700
Message-ID: <MDEHLPKNGKAHNMBLJOLKGEACGGAC.davids@webmaster.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
In-Reply-To: <alpine.LFD.0.999.0708231310430.30176@woody.linux-foundation.org>
Importance: Normal
Reply-To: davids@webmaster.com
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1123
Lines: 34


> On Thu, 23 Aug 2007, Segher Boessenkool wrote:

> The combining, which I've mentioned *multiple*times* is
>
> 	if (atomic_read(&x) <= 1)
>
> and dammit, if that doesn't result in a *single* instruction, the code
> generation is pure and utter crap.
>
> It should result in
>
> 	cmpl $1,offset(reg)
>
> and nothing else. And there is no way in hell you are doing that with
> "atomic_read()" being inline asm.

Sorry, Linus, I don't agree. The whole point of 'volatile' is to say to the
compiler, "DO NOT OPTIMIZE THIS. What you think is harmless may break things
you do not and should not understand." The combination of the read and the
compare into a single operation is a compiler optimization.

While it would not be unreasonable to still allow this optimization (since I
cannot think of any situations in which it could be harmful), it is just as
reasonable not to.

DS


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/