From mboxrd@z Thu Jan  1 00:00:00 1970
From: Paul Mackerras <paulus@samba.org>
Subject: Re: MMIO and gcc re-ordering issue
Date: Thu, 12 Jun 2008 22:14:53 +1000
Message-ID: <18513.4925.711881.794221@cargo.ozlabs.ibm.com>
References: <1211852026.3286.36.camel@pasglop>
	<200806111500.30789.nickpiggin@yahoo.com.au>
	<18511.24317.31038.926337@cargo.ozlabs.ibm.com>
	<200806111535.24523.nickpiggin@yahoo.com.au>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Return-path: <linux-arch-owner@vger.kernel.org>
Received: from ozlabs.org ([203.10.76.45]:57581 "EHLO ozlabs.org"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1753862AbYFLMO7 (ORCPT <rfc822;linux-arch@vger.kernel.org>);
	Thu, 12 Jun 2008 08:14:59 -0400
In-Reply-To: <200806111535.24523.nickpiggin@yahoo.com.au>
Sender: linux-arch-owner@vger.kernel.org
List-ID: <linux-arch.vger.kernel.org>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Linus Torvalds <torvalds@linux-foundation.org>, Matthew Wilcox <matthew@wil.cx>, Trent Piepho <tpiepho@freescale.com>, Russell King <rmk+lkml@arm.linux.org.uk>, Benjamin Herrenschmidt <benh@kernel.crashing.org>, David Miller <davem@davemloft.net>, linux-arch@vger.kernel.org, scottwood@freescale.com, linuxppc-dev@ozlabs.org, alan@lxorguk.ukuu.org.uk, linux-kernel@vger.kernel.org

Nick Piggin writes:

>                 /* turn off LED */
>                 val64 = readq(&bar0->adapter_control);
>                 val64 = val64 &(~ADAPTER_LED_ON);
>                 writeq(val64, &bar0->adapter_control);
>                 s2io_link(nic, LINK_DOWN);
>         }
>         clear_bit(__S2IO_STATE_LINK_TASK, &(nic->state));
> 
> Now I can't say definitively that this is going to be wrong on
> powerpc, because I don't know the code well enough. But I'd be
> 90% sure that the unlock is not supposed to be visible to
> other CPUs before the writeqs are queued to the card. On x86 it
> wouldn't be.

Interestingly, there is also a store to cacheable memory
(nic->device_enabled_once), but no smp_wmb or equivalent before the
clear_bit.  So there are other potential problems here besides the I/O
related ones.

Anyway, I have done some tests on a dual G5 here with putting a sync
on both sides of the store in writel etc. (i.e. making readl/writel
strongly ordered w.r.t. everything else), and as you predicted, there
wasn't a noticeable performance degradation, at least not on the
couple of things I tried.  So I am now inclined to accept your
suggestion that we should do that.  I should probably do some similar
checks on POWER6 and a few other machines first, though.

Paul.