From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:39165)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <rusty@ozlabs.org>) id 1V8g3v-0005U9-EY
	for qemu-devel@nongnu.org; Sun, 11 Aug 2013 20:32:47 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <rusty@ozlabs.org>) id 1V8g3q-0005PD-86
	for qemu-devel@nongnu.org; Sun, 11 Aug 2013 20:32:43 -0400
Received: from ozlabs.org ([203.10.76.45]:50469)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <rusty@ozlabs.org>) id 1V8g3p-0005Oa-U1
	for qemu-devel@nongnu.org; Sun, 11 Aug 2013 20:32:38 -0400
From: Rusty Russell <rusty@rustcorp.com.au>
In-Reply-To: <8761vflzu5.fsf@codemonkey.ws>
References: <1375938949-22622-1-git-send-email-rusty@rustcorp.com.au>
	<1375938949-22622-2-git-send-email-rusty@rustcorp.com.au>
	<87li4cgvh1.fsf@codemonkey.ws> <87ob97nz7x.fsf@rustcorp.com.au>
	<8761vflzu5.fsf@codemonkey.ws>
Date: Mon, 12 Aug 2013 09:16:55 +0930
Message-ID: <87a9knoknk.fsf@rustcorp.com.au>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: Re: [Qemu-devel] [PATCH 1/7] virtio: allow byte swapping for vring
	and config access
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Anthony Liguori <anthony@codemonkey.ws>, qemu-devel@nongnu.org

Anthony Liguori <anthony@codemonkey.ws> writes:
> Rusty Russell <rusty@rustcorp.com.au> writes:
>> (Qemu run under eatmydata to eliminate syncs)
>
> FYI, cache=unsafe is equivalent to using eatmydata.

Ah, thanks!

> I can reproduce this although I also see a larger standard deviation.
>
> BEFORE:
> 	MIN: 496
> 	MAX: 1055
>         AVG: 873.22
>         STDEV: 136.88
>
> AFTER:
>         MIN: 494
>         MAX: 1456
>         AVG: 947.77
>         STDEV: 150.89

BTW, how did you generate these stats?  Consider this my plug for my
little stats filter:
        https://github.com/rustyrussell/stats

>
> In my datasets, the stdev is higher in the after case implying that
> there is more variation.  Indeed, the MIN is pretty much the same.
>
> GCC is inlining the functions, I'm still surprised that it's measurable
> at all.

GCC won't inline across compilation units without -flto though, so the
stub call won't be inlined, right?

Cheers,
Rusty.