Catalin BOIE wrote:
>
> Hi, guys!
>
> I have a Intel440LX mobo, and 2 Pentium II 333 processors.
> On a 2.0.36 both processors are reported correctly:
[SNIP]
>
> But on 2.1.131 'cat
> /proc/cpuinfo' tell me:
>
> processor : 0
[SNIP]
> flags : fpu vme de pse tsc msr 6 mce cx8 9 sep 12 pge 14 cmov 16
> 17 mmx 24
> bogomips : 333.41
[SNIP]...
Before you go looking for exotic hardware trouble, are you sure you
compiled 2.1.131 properly? I have CPU's that give identical reports to
your first CPU on 2.1.131, except for three things: 1) speed is higer on
mine, 2) Stepping is known on mine, and most importantly: 3) The flags
are all recognised and decoded on mine.
Without looking at the code, (lazy) I'd say that the fact that you're
getting numeric CPU flag output in the above (like 6 ... 9 ... 12 etc)
means that your kernel source tree is buggered. I find it unlikely that
an unrecognised stepping would cause failure to recognise CPU flags by
name... ??
Neil
PS: I decided not to be lazy, and read the source code after all: as
long as it find "GenuineIntel" then those flags should be decoded OK.
It does find "GenuineIntel" but the flags aren't being printed. This
implies that the kernel source tree is hosed. Oh and btw, the reason
it's saying "Celeron (Covington)" is due to "detecting" a cache size of
0 on the 2nd CPU. These facts don't rule out CPU remarking, but they do
rule kernel brain-damage firmly in.
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
