Date: Sat, 13 Jul 2013 01:52:32 -0400 From: Kurt Lidl <lidl@pix.net> To: Peter Grehan <grehan@freebsd.org> Cc: freebsd-virtualization@freebsd.org Subject: Re: calcru runtime went backwards Message-ID: <51E0EB20.3070103@pix.net> In-Reply-To: <51DB8987.3010806@freebsd.org> References: <51DB7845.1050704@pix.net> <51DB8987.3010806@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 7/8/13 11:54 PM, Peter Grehan wrote: > Hi Kurt, > >> CPU: Intel(R) Xeon(R) CPU E5-2630 0 @ 2.30GHz (2300.13-MHz K8-class CPU) > > Looks like this Sandy Bridge has turbo-boost. First step is to disable > that in the BIOS and see if that has any effect. > > How many vCPUs in the guest ? I was finally able to do this. I turned off the Turbo-Boost in the BIOS, and restarted everything. My virtual machine is configured with 4 cpus and 10240 (10GB) of memory. I got the same error today when compiling python in the virtual machine: calcru: runtime went backwards from 48142872748812902 usec to 593237864 usec for pid 73714 (make) calcru: runtime went backwards from 48142872748812902 usec to 593157552 usec for pid 73714 (make) Looking at the console log, I also found the following on the console when it was booting: calcru: runtime went backwards from 8023812124181248 usec to 979712 usec for pid 13 (geom) So, disabling Turbo-Boost doesn't resolve the problem. -Kurt
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?51E0EB20.3070103>