Date: Thu, 23 Sep 2010 06:05:59 -0700 (PDT) From: Bryce <bryce@bryce.net> To: nickolasbug@gmail.com Cc: freebsd-stable@freebsd.org Subject: Re: SuperMicro i7 (UP) - very slow performance Message-ID: <bb30311e-ef0c-4d64-aa6b-8ffd8610a277@t7g2000vbj.googlegroups.com> In-Reply-To: <mailpost.1285167713.3595345.64974.mailing.freebsd.stable@FreeBSD.cs.nctu.edu.tw> References: <AANLkTik3bFuh9FR8vvwf=FfO2ojQYzHxWD8MODj22x1a@mail.gmail.com> <mailpost.1285167713.3595345.64974.mailing.freebsd.stable@FreeBSD.cs.nctu.edu.tw>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sep 22, 9:35=A0am, nickolas...@gmail.com wrote: > > md5 -t is quite a small benchmark, even with his misfunctioning CPU it > > took <6 seconds to complete. > > > If his problem is a misapplied heatsink/fan, then his CPU could be > > throttling when it gets hot, the hotter it gets the more it throttles, > > which could explain his massive buildworld walltime. Perhaps running > > something like: > > > =A0apply -0 "md5 -t" `jot 10` > > > would display a notable difference. > I don't think is heat & throttling related. Here's the temps while doing a make -j16 buildworld: dev.cpu.0.temperature: 57.0C dev.cpu.1.temperature: 57.0C dev.cpu.2.temperature: 54.0C dev.cpu.3.temperature: 54.0C dev.cpu.4.temperature: 57.0C dev.cpu.5.temperature: 57.0C dev.cpu.6.temperature: 55.0C dev.cpu.7.temperature: 55.0C IMO, the problem is preventing the system from even loading up the CPU's enough to even get hot - I have never seen it over 61 C. If sit & watch top -P during a buildworld, it rarely get's all 8 CPU's loaded up with idle numbers near 0%. A vast majority of the time one CPU is maxed and the others are mostly idle. > You may also run openssl speed or "linpack" test > (/usr/ports/math/linpack) or hpl (/usr/ports/benchmarks/hpl) to load > your CPU > _______________________________________________ > 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"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bb30311e-ef0c-4d64-aa6b-8ffd8610a277>