> > 66.36 BogoMips which is in the expected range for a
> > P166. However, for cpu1 I get 53.04 BogoMips which is
...
> As I told an earlier questioner, the BogoMIPS number should not concern
> you. One CPU is not running significantly slower than the other. The
> word "Bogo" appears in BogoMIPS for a very good reason: the number is
> bogus.
that's a little strong. two of the same model of processors, clocked
the same, should produce the same bogomips. bogomips are merely the
calibration constant for a particular timing loop. that loop has
changed in various kernels, and is effected by the compiler. within
a single model of a processor, it should scale with clock, since that's
its whole purpose.
> that it is the position of the CPU that makes the difference; that is,
> the second CPU will always be the one that has lower BogoMIPS.
nonsense.
> In an effort to further convince you that BogoMIPS is b*llsh*t and
> should not be cared about, take my /proc/cpuinfo as a striking example:
..
> cpu MHz : 561.105
...
> bogomips : 1120.67
I can't imagine why the author things this is bullshit. celerons are
execute the BM loop very fast. that's not to say that BM have any meaning
across different models of CPU, cache, or BM code. obviously.
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/dmentre/smp-howto/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
