cvs commit: src/sys/amd64/amd64 identcpu.c src/sys/i386/i386 identcpu.c

[David O'Brien]

On Tue, Oct 18, 2005 at 12:43:52PM -0400, John Baldwin wrote:
> On Tuesday 18 October 2005 11:22 am, David O'Brien wrote:
> > On Tue, Oct 18, 2005 at 12:50:30AM -0400, Daniel Eischen wrote:
> > > On Mon, 17 Oct 2005, Colin Percival wrote:
> > > > David O'Brien wrote:
> > > > > CPU: Dual Core AMD Opteron(tm) Processor 280 (2392.92-MHz K8-class
> > >          ^^^^
> > ...
> > > > This may be clear to *you*, but five years from now I'm sure that *I*
> > > > won't be able to remember what the AMD Opteron 280 is.
> > >
> > > Am I missing something?  Would it still say "Dual" if it were not a
> > > dual core?
> >
> > No it would not.  I'm not sure what this added:
> >     CPU: Dual Core AMD Opteron(tm) Processor 280 (2394.81-MHz K8-class
> >     ..
> >       Cores per package: 2
> 
> Just because your employer formats brand info strings that way doesn't mean 
> other chip manufacturers will.  FreeBSD as a project does not just run on 
> your employer's products, so we can't really make assumptions about the 
> layout of description strings that are free-form and vary from vendor to 
> vendor.  By obtaining the actual values from registers and outputting them in 
> a vendor-neutral way, FreeBSD as a project can provide this info to our users 
> across architectures (i386 and amd64) and across vendors (AMD and Intel) in a 
> uniform way.  This is much more user friendly.

*sigh*  This has become useless.  Please put it under verboseboot or turn
it into something truly useful.  Right now it is simply a waste of dmesg
space.  As written it does not tell you the kernel's POV.  As written it
doesn't accurately document the physical HW.

-- 
-- David  (obrien at FreeBSD.org)