[Top] [All Lists]

Re: [PATCH 26/30] loongson: flush irq write operation

To: yanh <>
Subject: Re: [PATCH 26/30] loongson: flush irq write operation
From: Ralf Baechle <>
Date: Wed, 20 May 2009 08:07:19 +0100
Cc: Wu Zhangjin <>,, Arnaud Patard <>,,, Philippe Vachon <>, Zhang Le <>, Erwan Lerale <>
In-reply-to: <1242786494.4382.58.camel@localhost.localdomain>
Original-recipient: rfc822;
References: <1242426527.10164.174.camel@falcon> <> <1242700637.4382.21.camel@localhost.localdomain> <> <1242786494.4382.58.camel@localhost.localdomain>
User-agent: Mutt/1.5.18 (2008-05-17)
On Wed, May 20, 2009 at 10:28:14AM +0800, yanh wrote:

> I have test this patch just now. It works well on yeeloong. 

Thanks for testing.

> I have one question what's the difference between the two patch? 

Uncached writes can't be re-ordered.  By adding a read after the last
write my patch forces not only completion of the preceding write but due
to this ordering constraint also completion of all preceding writes is

I/O space writes are slow.  I mean slower than slugs in space.  So my
patch is an optimization but that was not the point; I really only meant
to verify that we understood what's going on and we seem to.  Now let's
fix the real issue and make outX() non-posted.


<Prev in Thread] Current Thread [Next in Thread>