Peter Seebach writes:
> Hi!  This is NOT a NetBSD bug, but I'm really curious as to whether anyone's
> seen it.
> 
> This program reliably hangs three different AMD64 machines, all of them
> Athlons, and all with nForce4 chipsets.  I would love to know more cases
> where it works or doesn't.  Hangs have been observed under 64-bit Linux
> and 32-bit Windows.
> 
>         #include <string.h>
>         unsigned char a[4*1024*1024];
> 
>         int
>         main(void) {
>                 while (1)
>                         memset(a, 0, 4*1024*1024);
>                 return 0;
>         }
> 
> This program 100% reliably hangs my Athlon64 box.  It takes under 10 seconds
> under Linux, several minutes on Windows.  And no, that doesn't make any sense
> .

For the record, the above program runs without any problems on a
Sun V40z with 4 of these little guys:

cpu0 at mainbus0: apid 0 (boot processor)
cpu0: AMD Opteron(tm) Processor 850, 2393.34 MHz
cpu0: features: e7dbfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR>
cpu0: features: e7dbfbff<PGE,MCA,CMOV,PAT,PSE36,MPC,NOX,MMXX,MMX>
cpu0: features: e7dbfbff<FXSR,SSE,SSE2,LONG,3DNOW2,3DNOW>
cpu0: I-cache 64 KB 64B/line 2-way, D-cache 64 KB 64B/line 2-way
cpu0: L2 cache 1 MB 64B/line 16-way
cpu0: ITLB 32 4 KB entries fully associative, 8 4 MB entries fully associative
cpu0: DTLB 32 4 KB entries fully associative, 8 4 MB entries fully associative
cpu0: calibrating local timer
cpu0: apic clock running at 199 MHz
cpu0: 16 page colors

Even ran 4 concurrent copies of the program for a while, with zero 
problems/hangs/whatever...

Later...

Greg Oster