xfs-masters
[Top] [All Lists]

[xfs-masters] Re: [interesting] smattering of possible memory ordering b

To: Nick Piggin <nickpiggin@xxxxxxxxxxxx>
Subject: [xfs-masters] Re: [interesting] smattering of possible memory ordering bugs
From: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
Date: Fri, 26 Oct 2007 14:32:27 +1000
Cc: Linux Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>, paulus@xxxxxxxxx, shaggy@xxxxxxxxxxxxxx, adaplas@xxxxxxxxx, "Morton, Andrew" <akpm@xxxxxxxxxxxxxxxxxxxx>, xfs-masters@xxxxxxxxxxx
In-reply-to: <200710261347.35545.nickpiggin@yahoo.com.au>
References: <200710261209.58519.nickpiggin@yahoo.com.au> <1193369717.7018.56.camel@pasglop> <200710261347.35545.nickpiggin@yahoo.com.au>
Reply-to: xfs-masters@xxxxxxxxxxx
Sender: xfs-masters-bounce@xxxxxxxxxxx
On Fri, 2007-10-26 at 13:47 +1000, Nick Piggin wrote:

> I don't think the previous code was wrong... it's not a locked section
> > and we don't care about ordering previous stores. It's an
> allocation, it
> > should be fine. In general, bitmap allocators should be allright.
> 
> Well if it is just allocating an arbitrary _number_ out of a bitmap
> and nothing else (eg. like the pid allocator), then you don't need
> barriers.

Yup, that's what it does.

Cheers,
Ben.



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