Re: [gentoo-dev] CPUs reported by /proc in SMP machines and different archs

Re: [gentoo-dev] CPUs reported by /proc in SMP machines and different archs

[Meir Kriheli]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wednesday 24 December 2003 20:40, Kevyn Shortell wrote:
> On Wed, 2003-12-24 at 08:04, Meir Kriheli wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> >
> > Hi,
> >
> > While going over the output of /proc/cpuinfo (among the rest) you've sent
> > for - -stats (10x for that), it seems that the kernel reports multiple
> > cpus in a different way for each arch, so some revision of the code is
> > needed.
> >
> > Question: is it granted that on multiple cpu machines (and different
> > archs) each cpu will have the same properties (e.g: Clock Speed) ?
> >
> > - From the S in SMP one would guess it's true, but I still need to be
> > sure.
> >
> > For x86 archs it reports each processor, while on other archs (alpha,
> > sparc) it jusr reports total num cpus (andone know the status on Macs ?)
>
> processor       : 0
> cpu             : PPC970, altivec supported
> clock           : 2000MHz
> revision        : 2.2
>
> processor       : 1
> cpu             : PPC970, altivec supported
> clock           : 2000MHz
> revision        : 2.2
> machine         : PowerMac7,2
> motherboard     : PowerMac7,2 MacRISC4 Power Macintosh
> detected as     : 336 (PowerMac G5)
> pmac flags      : 00000000
> memory          : 1024MB
> pmac-generation : NewWorld
>
>
> this is a much newer kernel so more closer to what you'd find out in the
> world.

10x for /proc/cpuinfo. Maybe I should rephrase the question:

Has anyone seen a multiple cpu machine with different properties for the cpus 
inside (clock speed etc) ?

- -- 
Meir Kriheli
MKsoft systems
http://www.mksoft.co.il
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/7kAFRkS5DWK1mZkRAqXAAKCcrYF+YCLYWNz14Zrbm8isu9KyRACfdTdb
9E7ow3Jgvlvq91N/RNwL15k=
=oDUf
-----END PGP SIGNATURE-----

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] CPUs reported by /proc in SMP machines and different archs

[Bob Miller]

Meir Kriheli wrote:

> Has anyone seen a multiple cpu machine with different properties for
> the cpus inside (clock speed etc) ?

Yes, SGI's old multi-CPU MIPS boxes (either Challenge/Onyx or
Origin/Onyx2, I forget which) could be configured with different speed
R4000-family CPUs.  I used one once to debug a problem in IRIX where
some middleware assumed the CPUs were all the same sped.  It was a
supported configuration from SGI.

Oh, did you mean Linux machines? (-:

-- 
Bob Miller                              K<bob>
kbobsoft software consulting
http://kbobsoft.com                     kbob@jogger-egg.com

--
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] CPUs reported by /proc in SMP machines and different archs

[Philippe Lafoucrière]

> Has anyone seen a multiple cpu machine with different properties for the
> cpus inside (clock speed etc) ?


I'm not sure this is what you're looking for, so excuse me if it's not :

# cat /proc/cpuinfo

processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 6
model           : 6
model name      : AMD Athlon(tm) MP 1800+
stepping        : 2
cpu MHz         : 1533.707
cache size      : 256 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca 
cmov pat pse36 mmx fxsr sse syscall mp mmxext 3dnowext 3dnow
bogomips        : 3014.65

processor       : 1
vendor_id       : AuthenticAMD
cpu family      : 6
model           : 6
model name      : AMD Athlon(tm) Processor
stepping        : 2
cpu MHz         : 1533.707
cache size      : 256 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes


See the Model name of the CPUs : 1 is a MP, the other is shown as 'classic' 
athlon (which is not, my Tyan MB wouldn't boot with a non-mp athlon).


--
gentoo-dev@gentoo.org mailing list