From owner-freebsd-stable@FreeBSD.ORG Thu Sep 23 00:45:24 2010 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 22B911065670 for ; Thu, 23 Sep 2010 00:45:24 +0000 (UTC) (envelope-from bryce@bryce.net) Received: from mail-gx0-f186.google.com (mail-gx0-f186.google.com [209.85.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id CE5EA8FC14 for ; Thu, 23 Sep 2010 00:45:23 +0000 (UTC) Received: by gxk1 with SMTP id 1so1006439gxk.13 for ; Wed, 22 Sep 2010 17:45:23 -0700 (PDT) MIME-Version: 1.0 Received: by 10.100.235.38 with SMTP id i38mr42103anh.7.1285202723105; Wed, 22 Sep 2010 17:45:23 -0700 (PDT) Received: by c16g2000vbp.googlegroups.com with HTTP; Wed, 22 Sep 2010 17:45:22 -0700 (PDT) Date: Wed, 22 Sep 2010 17:45:22 -0700 (PDT) In-Reply-To: X-IP: 70.140.155.124 References: User-Agent: G2/1.0 X-HTTP-UserAgent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10,gzip(gfe) Message-ID: <12466724-cf4e-449b-8a6a-464e08c1c17f@c16g2000vbp.googlegroups.com> From: Bryce To: Ian Smith Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: freebsd-stable@freebsd.org Subject: Re: SuperMicro i7 (UP) - very slow performance X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 23 Sep 2010 00:45:24 -0000 On Sep 22, 11:43=A0am, smi...@nimnet.asn.au (Ian Smith) wrote: > On Wed, 22 Sep 2010, Tom Evans wrote: > > =A0> On Wed, Sep 22, 2010 at 2:50 PM, Ian Smith wr= ote: > =A0> > It seems far more than just CPU performance is awry. =A0Adam's dat= a from > =A0> > his i7 shows 2.7 times Bryce's speed for the md5 -t, maybe a lower= EST > =A0> > rate? - but that could no way account for buildworld taking 22.5 h= ours. > =A0> > > =A0> > Recent buildworld (albeit i386) on my Thinkpad T23 ran just shy of= 3.5 > =A0> > hours, without -j on an 1133MHz P3-M, 768MB of 133MHz RAM, 5400rpm= UFS > =A0> > disk - with X/KDE running meanwhile (~5-7% CPU penalty). > =A0> > =A0> md5 -t is quite a small benchmark, even with his misfunctioning CPU = it > =A0> took <6 seconds to complete. > =A0> > =A0> If his problem is a misapplied heatsink/fan, then his CPU could be > =A0> throttling when it gets hot, the hotter it gets the more it throttle= s, > =A0> which could explain his massive buildworld walltime. Perhaps running > =A0> something like: > =A0> > =A0> =A0 apply -0 "md5 -t" `jot 10` > =A0> > =A0> would display a notable difference. > =A0> > =A0> Intel chips are quite good at running without much cooling and not > =A0> dieing, using thermal throttling to preserve the CPU. > > I guess you mean on-package, without p4tcc or ACPI throttle support? > > >From Bryce's original message: > > # Disable throttle control (and rely on EIST) > hint.p4tcc.0.disabled=3D"1" > hint.acpi_throttle.0.disabled=3D"1" > > which is sensible, and seems to have been applied to all CPUs, but fromht= tp://www.bryce.net/files/dmesg.bootwe see for each of cpu[0-7]: > > est0: on cpu0 > est0: Invalid id16 (set, cur) =3D (20, 21) > est0: Can't check freq 2667, it may be invalid > est0: Invalid id16 (set, cur) =3D (19, 21) > est0: Can't check freq 2533, it may be invalid > est0: Invalid id16 (set, cur) =3D (18, 21) > est0: Can't check freq 2400, it may be invalid > est0: Invalid id16 (set, cur) =3D (17, 21) > est0: Can't check freq 2267, it may be invalid > est0: Invalid id16 (set, cur) =3D (16, 21) > est0: Can't check freq 2133, it may be invalid > est0: Invalid id16 (set, cur) =3D (15, 21) > est0: Can't check freq 2000, it may be invalid > est0: Invalid id16 (set, cur) =3D (14, 21) > est0: Can't check freq 1867, it may be invalid > est0: Invalid id16 (set, cur) =3D (13, 21) > est0: Can't check freq 1733, it may be invalid > est0: Invalid id16 (set, cur) =3D (12, 21) > est0: Can't check freq 1600, it may be invalid > > which looks a bit ominous? =A0What does 'sysctl hw.acpi dev.cpu' say? hw.acpi.supported_sleep_state: S1 S4 S5 hw.acpi.power_button_state: S5 hw.acpi.sleep_button_state: S1 hw.acpi.lid_switch_state: NONE hw.acpi.standby_state: S1 hw.acpi.suspend_state: NONE hw.acpi.sleep_delay: 1 hw.acpi.s4bios: 0 hw.acpi.verbose: 1 hw.acpi.disable_on_reboot: 0 hw.acpi.handle_reboot: 0 hw.acpi.reset_video: 0 hw.acpi.cpu.cx_lowest: C1 dev.cpu.0.%desc: ACPI CPU dev.cpu.0.%driver: cpu dev.cpu.0.%location: handle=3D\_PR_.P001 dev.cpu.0.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.0.%parent: acpi0 dev.cpu.0.freq: 2801 dev.cpu.0.freq_levels: 2801/143000 2667/114000 2533/100000 2400/87000 2267/76000 2133/68000 2000/59000 1867/51000 1733/44000 1600/40000 dev.cpu.0.cx_supported: C1/32 C2/96 C3/128 dev.cpu.0.cx_lowest: C1 dev.cpu.0.cx_usage: 100.00% 0.00% 0.00% last 372us dev.cpu.0.temperature: 51.0C dev.cpu.1.%desc: ACPI CPU dev.cpu.1.%driver: cpu dev.cpu.1.%location: handle=3D\_PR_.P002 dev.cpu.1.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.1.%parent: acpi0 dev.cpu.1.cx_supported: C1/32 C2/96 C3/128 dev.cpu.1.cx_lowest: C1 dev.cpu.1.cx_usage: 100.00% 0.00% 0.00% last 178us dev.cpu.1.temperature: 51.0C dev.cpu.2.%desc: ACPI CPU dev.cpu.2.%driver: cpu dev.cpu.2.%location: handle=3D\_PR_.P003 dev.cpu.2.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.2.%parent: acpi0 dev.cpu.2.cx_supported: C1/32 C2/96 C3/128 dev.cpu.2.cx_lowest: C1 dev.cpu.2.cx_usage: 100.00% 0.00% 0.00% last 386us dev.cpu.2.temperature: 48.0C dev.cpu.3.%desc: ACPI CPU dev.cpu.3.%driver: cpu dev.cpu.3.%location: handle=3D\_PR_.P004 dev.cpu.3.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.3.%parent: acpi0 dev.cpu.3.cx_supported: C1/32 C2/96 C3/128 dev.cpu.3.cx_lowest: C1 dev.cpu.3.cx_usage: 100.00% 0.00% 0.00% last 391us dev.cpu.3.temperature: 49.0C dev.cpu.4.%desc: ACPI CPU dev.cpu.4.%driver: cpu dev.cpu.4.%location: handle=3D\_PR_.P005 dev.cpu.4.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.4.%parent: acpi0 dev.cpu.4.cx_supported: C1/32 C2/96 C3/128 dev.cpu.4.cx_lowest: C1 dev.cpu.4.cx_usage: 100.00% 0.00% 0.00% last 413us dev.cpu.4.temperature: 50.0C dev.cpu.5.%desc: ACPI CPU dev.cpu.5.%driver: cpu dev.cpu.5.%location: handle=3D\_PR_.P006 dev.cpu.5.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.5.%parent: acpi0 dev.cpu.5.cx_supported: C1/32 C2/96 C3/128 dev.cpu.5.cx_lowest: C1 dev.cpu.5.cx_usage: 100.00% 0.00% 0.00% last 381us dev.cpu.5.temperature: 51.0C dev.cpu.6.%desc: ACPI CPU dev.cpu.6.%driver: cpu dev.cpu.6.%location: handle=3D\_PR_.P007 dev.cpu.6.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.6.%parent: acpi0 dev.cpu.6.cx_supported: C1/32 C2/96 C3/128 dev.cpu.6.cx_lowest: C1 dev.cpu.6.cx_usage: 100.00% 0.00% 0.00% last 411us dev.cpu.6.temperature: 48.0C dev.cpu.7.%desc: ACPI CPU dev.cpu.7.%driver: cpu dev.cpu.7.%location: handle=3D\_PR_.P008 dev.cpu.7.%pnpinfo: _HID=3Dnone _UID=3D0 dev.cpu.7.%parent: acpi0 dev.cpu.7.cx_supported: C1/32 C2/96 C3/128 dev.cpu.7.cx_lowest: C1 dev.cpu.7.cx_usage: 100.00% 0.00% 0.00% last 330us dev.cpu.7.temperature: 48.0C > > Running multiple md5s or say 'dd if=3D/dev/random of=3D/dev/null bs=3D1M = &' > in a short sleep loop echoing "`date` `sysctl -n dev.cpu.0.freq` plus > indicative coretemp sysctls might reveal something as it heats up? =A0 > Surprisingly (?) the dmesg shows no ACPI thermal zones (detected). I don't think it is temperature, I have never seen temps above the low 60's C and the speed never goes down from 2.8 Ghz. This is what I see when running your dd for a while: bryce@tahiti[~]>sysctl -a | grep temperature dev.cpu.0.temperature: 55.0C dev.cpu.1.temperature: 55.0C dev.cpu.2.temperature: 51.0C dev.cpu.3.temperature: 52.0C dev.cpu.4.temperature: 59.0C dev.cpu.5.temperature: 61.0C dev.cpu.6.temperature: 51.0C dev.cpu.7.temperature: 52.0C bryce@tahiti[~]>sysctl -n dev.cpu.0.freq 2801 > > cheers, Ian > > _______________________________________________ > freebsd-sta...@freebsd.org mailing listhttp://lists.freebsd.org/mailman/l= istinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"