From mboxrd@z Thu Jan  1 00:00:00 1970
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Subject: Re: [PATCHv2 1/5] mm: add coherence API for DMA to vmalloc/vmap
 areas
Date: Sun, 03 Jan 2010 08:27:58 +1100
Message-ID: <1262467678.2173.244.camel@pasglop>
References: <1261603345-2494-1-git-send-email-James.Bottomley@suse.de>
	 <1261603345-2494-2-git-send-email-James.Bottomley@suse.de>
	 <20091224100853.GA3165@console-pimps.org>
	 <20091224123913.GD5335@parisc-linux.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Return-path: <linux-arch-owner@vger.kernel.org>
Received: from gate.crashing.org ([63.228.1.57]:59521 "EHLO gate.crashing.org"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751191Ab0ABV3X (ORCPT <rfc822;linux-arch@vger.kernel.org>);
	Sat, 2 Jan 2010 16:29:23 -0500
In-Reply-To: <20091224123913.GD5335@parisc-linux.org>
Sender: linux-arch-owner@vger.kernel.org
List-ID: <linux-arch.vger.kernel.org>
To: Matthew Wilcox <matthew@wil.cx>
Cc: Matt Fleming <matt@console-pimps.org>, James Bottomley <James.Bottomley@suse.de>, linux-arch@vger.kernel.org, linux-parisc@vger.kernel.org, Christoph Hellwig <hch@lst.de>, Russell King <rmk@arm.linux.org.uk>, Paul Mundt <lethal@linux-sh.org>

On Thu, 2009-12-24 at 05:39 -0700, Matthew Wilcox wrote:
> 
> invalidates the processor cache for a given virtual address range
> in the vmap area.  This API addresses the problem that the processor
> may have performed speculative reads into its cache of the vmapped
> area while I/O was occurring to the underlying physical pages.
> 
> Signed-off-by: Matthew Wilcox <willy@linux.intel.com>

Interestingly, our DMA APIs in this regard are sub-optimal as they
should provide 3 hooks, not 2.

Flush is good for ensuring dirty lines have been pushed out before an
outgoing DMA.

But for incoming DMA it would be nice to properly split the 2 calls that
may be needed on some archs, one before, one after the transfer. Sure,
invalidate twice will "work" but the f will also be sub-optimal on some
platforms depending on whether the platform is known to be able or not
to speculatively load cache lines etc...

Maybe just a "before" vs. "after" argument ?

Also, the proposal goes contrary to most of our DMA APIs which don't
actually expose the details of invalidate vs. flush but instead expose
the direction of the transfer (both lack the above subtlety though).

Thoughts ?

Cheers,
Ben.