Date: Fri, 10 Sep 2004 07:37:54 +0100 (BST)
From: Hugh Dickins <hugh@veritas.com>
To: Kaigai Kohei <kaigai@ak.jp.nec.com>
cc: Andrew Morton <akpm@osdl.org>,
       "Linux Kernel ML(Eng)" <linux-kernel@vger.kernel.org>
Subject: Re: PATCH] atomic_inc_return() [0/5] (Re: atomic_inc_return)
In-Reply-To: <01be01c496eb$6a9dbd60$f97d220a@linux.bs1.fc.nec.co.jp>
Message-ID: <Pine.LNX.4.44.0409100721470.15219-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1367
Lines: 33

On Fri, 10 Sep 2004, Kaigai Kohei wrote:
> > 
> > I believe you have a patch adding those to i386 (including CONFIG_M386
> > runtime check lest it's an actual i386 which cannot do "xadd") and x86_64.
> > I'd be glad to see that go into the tree, would you be ready to submit it
> > to Andrew or Linus based on the current 2.6 BK tree?
> 
> Indeed, I hope to do this.
> atomic_inc_return() is necessary for the 'SELinux performance improvement
> by RCU' patch.

Thanks a lot, the patches 1-5 you've posted look good to me (aside from
some spaces instead of tab in the asm-arm 3/5), and should save me from
my silly race.

If we'd had these earlier, we could have avoiding changing page->count
and page->mapcount over to start from -1 (though I dare say they may
be slightly more efficient as they now are, than if we converted them
back).  Should save making such conversions in future, anyway.

> > I think arm was missing the inc and dec, but has recently gained them.

I was wrong when I said that: asm-arm/atomic.h has recently been reworked,
but does still needs your additions: sorry for the confusion.

Thanks,
Hugh

-
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/