This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: interpretation of %CPU in 'procps' output for multi-cpu & hyperthreading


Tom Rodman wrote:
On Thu 10/11/07 10:05 CDT Matthew Woehlke wrote:
Tom Rodman wrote:
Is there a way to prove that a given process with more than 1 thread,
is still restricted to just one CPU?
Unless you have manually set affinity, why would this be true? More likely, only one thread is actually doing anything.

Thanks Matthew.


I meant to ask:

  Is there a way to prove that a given process with more than 1
  thread, must always have all it's threads on a single CPU at
  any given time ( over the life of the process, I assume the all
  it's threads could shift from CPU to CPU)?

Well... unless you have set affinity, I don't know of any reason why a multi-threaded application would have all of its threads end up on only a single CPU to begin with*. As for whether or not threads can migrate, I would assume they can, although I don't really know for sure. (On Linux, under constant load, they don't seem to. On Windows it seems like even single-threaded applications tend to be distributed across multiple cores.)


(* for the nitpickers: "unless of course you only /have/ one CPU".)

Anyway, this isn't really a Cygwin question any more, you would do better asking on a Windows forum. If you wish to discuss this further, it should probably be taken to cygwin-talk.

--
Matthew
If a signature
is not read by anyone,
does it make a sound?

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]