Christian Biere wrote:
 
> pmc reports wrongly(?) that counters are not supported by this CPU:
> 
> cpu0 at mainbus0: (uniprocessor)
> cpu0: AMD Duron (686-class), 896.22 MHz
> cpu0: features 183f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR>
> cpu0: features 183f9ff<PGE,MCA,CMOV,FGPAT,PSE36,MMX>
> cpu0: features 183f9ff<FXSR>
> cpu0: I-cache 64 KB 64b/line 2-way, D-cache 64 KB 64b/line 2-way
> cpu0: L2 cache 64 KB 64b/line 16-way
> cpu0: ITLB 16 4 KB entries fully associative, 8 4 MB entries fully
> associative cpu0: DTLB 24 4 KB entries fully associative, 8 4 MB entries
> 4-way cpu0: 8 page colors
> 
> pmc_init() sys/arch/i386/i386/pmc.c has cpu_vendor set to "Intel". For
> testing I commented out the 'strcmp(cpu_vendor, "AuthenticAMD")'. pmc
> worked then although I don't know whether the results are correct. Though
> this is just a hack and there must be a better way to explain pmc_init()
> the vendor isn't Intel. With -DCPUDEBUG and -DDIAGNOSTIC identifycpu()
> reports "cpu0: family 6 model 3 step 1".

You're right, pmc_init() is broken.  I think it's safe to say that
pmc is not supported at the moment.

	-- Gregory McGarry <g.mcgarry@ieee.org>