Date: Fri, 28 Apr 2006 14:28:18 -0400 From: Kris Kennaway <kris@obsecurity.org> To: Dmitry Pryanishnikov <dmitry@atlantis.dp.ua> Cc: freebsd-stable@freebsd.org, Kris Kennaway <kris@obsecurity.org> Subject: Re: RELENG_4 -> 5 -> 6: significant performance regression Message-ID: <20060428182818.GA10410@xor.obsecurity.org> In-Reply-To: <20060428122448.K57436@atlantis.atlantis.dp.ua> References: <20060427160536.M96305@atlantis.atlantis.dp.ua> <20060427181226.GA66431@xor.obsecurity.org> <20060428122448.K57436@atlantis.atlantis.dp.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
--jI8keyz6grp/JLjh
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Fri, Apr 28, 2006 at 03:31:17PM +0300, Dmitry Pryanishnikov wrote:
>=20
> Hello!
>=20
> On Thu, 27 Apr 2006, Kris Kennaway wrote:
>=20
> Thanks for your suggestions, they've made a difference (though not as big
> as one could hope).
>=20
> >On Thu, Apr 27, 2006 at 05:08:11PM +0300, Dmitry Pryanishnikov wrote:
> >
> >>makeoptions CONF_CFLAGS=3D-fno-builtin
> >
> >Non-default option; this may conceivably affect performance.
>=20
> The reason why I've initially included this option is the following
> comment (NOTES from RELENG_6):
>=20
> #
> # CONF_CFLAGS gives some extra compiler flags that are added to ${CFLAGS}
> # after most other flags. Here we use it to inhibit use of non-optimal
> # gcc builtin functions (e.g., memcmp).
> #
>=20
> I've read this: using "CONF_CFLAGS=3D-fno-builtin" inhibits use of non-op=
timal
> gcc builtin functions, so this option may be useful for getting max.=20
> performance. Are this comment and my interpretation still correct now?
I don't know, it needs to be tested in your particular case.
> >> %Sys %Intr %Idl
> >>RELENG_4 + rl0 14 14 72
> >>RELENG_4 + fxp0 14 10 76
> >>RELENG_5 + rl0 40 30 30
> >>RELENG_5 + fxp0 35 25 40
> >>RELENG_6 + rl0 45 40 15
> >>RELENG_6 + fxp0 45 35 20
> %Sys %Intr %Idl "time md5 -t" wall clock time
> RELENG_5 + rl0 33 23 44 1:41
> RELENG_5 + fxp0 30 20 50 1:36
>=20
> RELENG_6 + rl0 34 24 42 1:43
> RELENG_6 + fxp0 30 20 50 1:40
>=20
> So performance now is much better then before removal of
>=20
> makeoptions CONF_CFLAGS=3D-fno-builtin
> options INVARIANTS
> options INVARIANT_SUPPORT
> options QUOTA
>=20
> (I'll try to find out which one of these takes which % of overhead when I
> get free time), but still much worse then under RELENG_4, where this
> particular (I'd say "quote common") usage pattern takes 24-28% of CPU tim=
e,=20
> while under RELENG_5 / 6 it takes >=3D 50% ;(
Thanks. Silly question: the data transfer rate is the same on both
4.x and 6.x, right? i.e. the data transfer itself takes the same
time?
The next step is for you to run some profiling tests to see
where the kernel is spending time, e.g. with hwpmc.
Also, when you are trying to quantify performance differences, you
need to run many copies of the test (at least 10) under identical
conditions to account for possible variations. The ministat tool
(/usr/src/tools/tools/ministat) is good for performing statistically
meaningful comparisons of data sets when you have them.
Kris
--jI8keyz6grp/JLjh
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (FreeBSD)
iD8DBQFEUl7BWry0BWjoQKURAmYYAJ9IT2reS/gdAJKV47809KQ8xeAZKACeOlvW
lrD80dnhpdehM6cN76tBQ6I=
=8KTE
-----END PGP SIGNATURE-----
--jI8keyz6grp/JLjh--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060428182818.GA10410>