Adds additional CPU details fields.

[Darryl L. Pierce]

+++ Martin S. [04/09/08 13:08 +0200]:
>I had approached your idea back in April 2007 already as I think this is
>valuable information to have in HAL.
>
>See http://lists.freedesktop.org/archives/hal/2007-April/007978.html
>and http://media.sukimashita.com/temp/00-hal-processor-cpuinfo-1.png
>
>It was basically rejected (on IRC) because the whole /proc/cpuinfo was
>better suited in sysfs (which apparently is the correct "upstream"). The
>labels differ from processor/architecture and there is no common spec
>for the contents even among similar archs.
>
>Long time ago I checked LKML and there was some movement to deprecate
>the old cpuinfo and implement a standardized sysfs CPU info interface.
>
>There had been working "corpse patches" on the LKML, not sure where this
>is now...
>
>Also, did you think about:
>- FreeBSD? Solaris?
>- That different CPUs expose different cpuinfo labels?
>- PPC? ARM? IA64? Sparc? ...
>
>Hope we get somewhere this time either upstream or in HAL. :)

As do I. I spoke with a few people weeks ago in #hal on freenode about this
idea and got a generally positive response. For my case, my project at work
needs to identify information on the machine in which it's running and report
it to a central server which then makes determinations based on that. In the
current code we're cobbling things together from disparate sources (libvirt,
HAL and parsing files like /proc/cpuinfo). 

It would make our code more consistent and maintainable if we went to a 
single source for that information. 

-- 
Darryl L. Pierce, Sr. Software Engineer
Red Hat, Inc. - http://www.redhat.com/
oVirt - Virtual Machine Management - http://www.ovirt.org/
"What do you care what other people think, Mr. Feynman?"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/hal/attachments/20080904/2894b3e4/attachment.pgp