From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH v5] KVM: nVMX: Fully support of nested VMX preemption timer Date: Thu, 03 Oct 2013 10:09:26 +0200 Message-ID: <524D2636.8070906@redhat.com> References: <1379319104-10266-1-git-send-email-yzt356@gmail.com> <52444CF6.1020102@redhat.com> <52493F8C.6040009@web.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Arthur Chunqi Li , kvm@vger.kernel.org, gleb@redhat.com, "Zhang, Yang Z" To: Jan Kiszka Return-path: Received: from mx1.redhat.com ([209.132.183.28]:27326 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751448Ab3JCIJK (ORCPT ); Thu, 3 Oct 2013 04:09:10 -0400 In-Reply-To: <52493F8C.6040009@web.de> Sender: kvm-owner@vger.kernel.org List-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Il 30/09/2013 11:08, Jan Kiszka ha scritto: > On 2013-09-26 17:04, Paolo Bonzini wrote: >> Hi all, >>=20 >> the test fails for me if the preemption timer value is set to a=20 >> value that is above ~2000 (which means ~65000 TSC cycles on this=20 >> machine). The preemption timer seems to count faster than what >> is expected, for example only up to 4 million cycles if you set >> it to one million. So, I am leaving the patch out of kvm/queue >> for now, until I can test it on more processors. >=20 > I've done some measurements with the help of ftrace on the time it=20 > takes to let the preemption timer trigger (no adjustments via=20 > Arthur's patch were involved): On my Core i7-620M, the preemption=20 > timer seems to tick almost 10 times faster than spec and scale=20 > value (5) suggests. I've loaded a value of 100000, and it took=20 > about 130 =B5s until I got a vmexit with reason PREEMPTION_TIMER (no=20 > other exists in between). 10x is similar to what I was observing. Paolo -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.21 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJSTSY2AAoJEBvWZb6bTYbyZmMQAJ7i2irCo/CDAK/zS//BHI/E OE2jsqKnUVkcSee5XJ+bdyclqwLmp7J/vEySKlXpdrzr2lcZ/FNP2muXQ1HXeK8L I3eoSbuYTY3DPHKTjR09GVNfzIZIC6H8TwKb8RdbtZgwci1r9kjmwcpAJt3Qh0UR xM/6a5gnOubdCGx6SdFBVPL+OfZ3zu1Si6Aw+3mnNYO9KvLpbtA3lO4u4HteTchM KOszK0rmR3Y1LoWQdUhuTgrP1DMFZyZKhW4nHOIq/DWK+/al+0knJ5z5QaeZgP3z GkvwzDbzJKGTGRJpAiAixQ40uKIycHkv6IKjvJn2iT6/XhI78W2+X5OVjhU4kt8p aJ3h+KacopaYQpxaHBZo8jSvT0U+vX5mxvoBbn6okaKMw/iZ7eXVR6n0Pi3zHF35 g8jUalHKbunY8q0bFjZcvVfmAIVf+oBEQ67nkxZSUw3vW4zvx+Vmtx7j0MwzmLc7 2IWSAWehSaYCsKzHh7Qf+j8/8qCDaPZDmaAIrzE9ODglMHd/fZk/zA3rhJ7+5MnZ =46KVNV/ABcnvm/TP3s0bIL7hWHUawZ/PlxJ0IxKvD25TUm70kpG9FIbrbRZ25ltsh mmI+aL5HK5ztJR4lSj0jk51YBYlLWoef7yWzDSnVgnebceY5KP7o1WVzSPt+AJlD Z6dK0g1rJk39kQV5nrXh =3DBPPg -----END PGP SIGNATURE-----