From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michal =?iso-8859-1?Q?Koutn=FD?= Subject: Re: [PATCH 0/7] introduce cpu.headroom knob to cpu controller Date: Tue, 21 May 2019 15:47:30 +0200 Message-ID: <20190521134730.GA12346@blackbody.suse.cz> References: <20190408214539.2705660-1-songliubraving@fb.com> <20190410115907.GE19434@e105550-lin.cambridge.arm.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="r5Pyd7+fXNt84Ff3" Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-ID: To: Song Liu Cc: Morten Rasmussen , Kernel Team , "peterz@infradead.org" , "vincent.guittot@linaro.org" , "tglx@linutronix.de" , "mingo@redhat.com" , "cgroups@vger.kernel.org" , linux-kernel --r5Pyd7+fXNt84Ff3 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hello Song. On Wed, Apr 10, 2019 at 07:43:35PM +0000, Song Liu = wrote: > The load level above is measured as requests-per-second.=20 >=20 > When there is no side workload, the system has about 45% busy CPU with=20 > load level of 1.0; and about 75% busy CPU at load level of 1.5.=20 >=20 > The saturation starts before the system hitting 100% utilization. This is > true for many different resources: ALUs in SMT systems, cache lines,=20 > memory bandwidths, etc.=20 I have read through the thread continuation and it appears to me there is some misunderstanding on the latency metric (scheduler latency <=3D your latency <=3D request wall time?). Could you please describe how is the latency that you report defined and measured? Thanks, Michal --r5Pyd7+fXNt84Ff3 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEE+amhwRV4jZeXdUhoK2l36XSZ9y4FAlzkAW4ACgkQK2l36XSZ 9y6i5Q/+PtvIsPw08gmvSIO/fdv4U5u+tI7WHJcUXGWD+sTVdVSlB+MWDfdlblUH HIyGgIBDGL184mq4kiYzxvSMPyorrtlI2Nbvq9SYqLJWPnZGMTWtOgVgb7roVJQG xk96LUF3nmWjy6wU5wY0GD3J1yaygsqMLOuCP9LQLx8Mqv7yZ5xY+J98Sdm6s1Uo y9ePTJLq8r8vA/oHSaLyRTiQeN+QYnFmqJvMaqdImM9FH9pO6WADa1Hp/UfN6KT5 06NwWzaqPJ1e5JZqZ5DZMKZSCaLHNS7pywwuTgHgxoGacbBYywQ3VOF67SDwb5zK oOsE0YL//9Wd2cH6qh3i5301z+lx+WvFyXu3hnKBUKt2RzlT/CqwickpZxamJ3Pk PtV/7U/gJV404Z2TXE1cMOwfEfgjIE2WanjfMSyIuhkdxGjruQ0Ik+9DDIWMed08 lgH33hEOLGu1qK4TKVXR/MJjdsUNAR60c4SgAWB5oc9hSjrdFDV9wD+UWviizgjM WMasTRwVoOaP2H9iiXlXQXX2Tei04aHR2fbW0XawsFO/bne47S/r4eHL2TvxuRZR W/6Q5p5o4nHvOuRQpZ2zTQ5/SXpcRSMHM7xnvxNzFJxjiR3LJmVb1fsGv7D+xIY4 J74b5569lroFCynIj+8XxVv21JHeVtBgYEH2HKUowuk6/DyftnE= =vB4R -----END PGP SIGNATURE----- --r5Pyd7+fXNt84Ff3--