Open MPI logo

Hardware Locality Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Hardware Locality Development mailing list

Subject: Re: [hwloc-devel] Strange difference
From: Brice Goglin (Brice.Goglin_at_[hidden])
Date: 2010-03-26 13:25:14


Jeff Squyres wrote:
> On Mar 26, 2010, at 11:33 AM, Samuel Thibault wrote:
>
>
>> Well, yes, it is supposed to display less information :)
>> Which precise difference are you referring to?
>>
>
> The first section of both -- obviously, not the 2nd section of the -v output. :-)
>
> The first section of both displays roughly the same information -- the only addition seems to be the phys number of the socket and the machine memory sizes (which don't agree between -v and non-v -- why?). Indeed, the "-v" output seems to remove a lot of whitespace that was included in the non-v output.
>
> I just thought it was weird that the -v output is almost exactly the same information (in the first part), but expressed in more lines with less whitespace (L1 and core moved to their own lines). More specifically, if adding the phys number of the socket and the machine sizes is the only information difference, why not add to the same output format as the non-v output?
>

The reason was that the "phys" attribute is mostly useless for anything
but Proc and NUMANode.

Current other differences when adding -v are:
* "Proc" if -v instead of "P"
* print local memory size
* print machine DMI attributes
* print memory sizes in verbose mode (not rounded to GB or so)

Brice