Re: : Fwd: Research code to run
David Mason <dmason-bqArmZWzea/GcjXNFnLQ/[email protected]> Fri, 5 Sep 2014 15:51:45 -0400
| Newsgroups | gmane.org.user-groups.linux.tolug |
|---|---|
| Message-ID | <CALFgxqBmXMbnu=fgrxwOaV=AqhG8BtrXH+6ZfA-Jasz3RRT4bg@mail.gmail.com> |
On 5 September 2014 15:22, Lennart Sorensen <lsorense-1wCw9BSqJbv44Nm34jS7GywD8/[email protected]> wrote: > What did you expect sysctl -n machdep.cpu.brand_string to accomplish? > I can't find any linux system where that does anything. I wanted this to run on any *ix system and produce output that I can parse to figure out the machine. I expect Linux boxes to have /proc/cpuinfo, and everything else to have sysctl. That particular string is from Mac OS X, but this has also run on FreeBSD and Solaris (where sysctl produces rather different output. Turns out some Linuxes also have sysctl, with various contents. The only thing I really care about it is as much detail as possible about the CPU on the system. I'm not interested in how numbers from any particular machine compare to another machine, just how the different runs compare. The output "should" have static < bimorph* < dynamic, but it doesn't seem to be that simple... :-) Thanks if you get a chance to run this on something interesting. I have an Alpha, SPARC, PowerPC, AMD, and a plethora of Intel machines. ../Dave -- The Toronto Linux Users Group. Meetings: http://gtalug.org/ TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists