Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Andreas Dilger <adilger@dilger.ca>
Message-Id: <028BF9CD-2E43-47DF-9373-A2D6EA7A3CF5@dilger.ca>
Content-Type: multipart/signed;
 boundary="Apple-Mail=_7D235CBA-60A1-49A6-B812-08865E719A4F";
 protocol="application/pgp-signature"; micalg=pgp-sha256
Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\))
Subject: Re: [PATCH] mbcache: Speed up cache entry creation
Date:   Thu, 25 Jul 2019 15:16:04 -0600
In-Reply-To: <20190724040118.GA31214@sultan-box.localdomain>
Cc:     Alexander Viro <viro@zeniv.linux.org.uk>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux List Kernel Mailing <linux-kernel@vger.kernel.org>
To:     Sultan Alsawaf <sultan@kerneltoast.com>
References: <20190723053549.14465-1-sultan@kerneltoast.com>
 <5EDDA127-031C-4F16-9B9B-8DBC94C7E471@dilger.ca>
 <20190724040118.GA31214@sultan-box.localdomain>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


--Apple-Mail=_7D235CBA-60A1-49A6-B812-08865E719A4F
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

On Jul 23, 2019, at 10:01 PM, Sultan Alsawaf <sultan@kerneltoast.com> =
wrote:
>=20
> On Tue, Jul 23, 2019 at 10:56:05AM -0600, Andreas Dilger wrote:
>> Do you have any kind of performance metrics that show this is an =
actual
>> improvement in performance?  This would be either macro-level =
benchmarks
>> (e.g. fio, but this seems unlikely to show any benefit), or =
micro-level
>> measurements (e.g. flame graph) that show a net reduction in CPU =
cycles,
>> lock contention, etc. in this part of the code.
>=20
> Hi Andreas,
>=20
> Here are some basic micro-benchmark results:
>=20
> Before:
> [    3.162896] mb_cache_entry_create: AVG cycles: 75
> [    3.054701] mb_cache_entry_create: AVG cycles: 78
> [    3.152321] mb_cache_entry_create: AVG cycles: 77
>=20
> After:
> [    3.043380] mb_cache_entry_create: AVG cycles: 68
> [    3.194321] mb_cache_entry_create: AVG cycles: 71
> [    3.038100] mb_cache_entry_create: AVG cycles: 69

This information should be included in the patch description, since that
allows making a decision on whether the patch is worthwhile to land or =
not.

> The performance difference is probably more drastic when free memory =
is low,
> since an unnecessary call to kmem_cache_alloc() can result in a long =
wait for
> pages to be freed.

Cheers, Andreas


--Apple-Mail=_7D235CBA-60A1-49A6-B812-08865E719A4F
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
	filename=signature.asc
Content-Type: application/pgp-signature;
	name=signature.asc
Content-Description: Message signed with OpenPGP

-----BEGIN PGP SIGNATURE-----
Comment: GPGTools - http://gpgtools.org
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=3yFi
-----END PGP SIGNATURE-----

--Apple-Mail=_7D235CBA-60A1-49A6-B812-08865E719A4F--