| Excerpts from mail: 5-Nov-101 Re: 420R fails to boot by eeh@netbsd.org 
| > | %TL:3 %TT:32 %TPC:1008590 %TnPC:10085a0
| > | %TSTATE:4482041402  %CWP:2
| > |    %PSTATE:414 AG:0 IE:0 PRIV:1 AM:0 PEF:1 RED:0 MM:0 TLE:0 CLE:0
| > |    MG:1 IG:0
| > |    %ASI:82  %CCR:44  XCC:nZvc   ICC:nZvc
| >  
| > Which then generated a bus fault, which shouldn't happen here...
| I'm assuming you want the disasembly from around this trap. If not, then
| I'm not sure what I'm supposed to be doing.
|
| 1008500           or          %g0, 30, %g3     
| 1008504  +4       sethi       1c00, %g6        
| 1008508  +8       ldxa        [%g3]58, %g3     
| 100850c  +c       sethi       18a2400, %g4     
| 1008510  +10      or          %g6, 3ff, %g6    
| 1008514  +14      ldx         [%g4 + 348], %g4 
| 1008518  +18      srax        %g3, 2b, %g5     
| 100851c  +1c      and         %g3, %g6, %g6    
| 1008520  +20      add         %g5, 1, %g5      
| 1008524  +24      sllx        %g6, 3, %g6      
| 1008528  +28      ldx         [%g4 + %g6], %g4 
| 100852c  +2c      bprnz,pt    %g6, 1008560  +60
| 1008530  +30      sethi       1000000, %g7     
| 1008534  +34      sethi       800000, %g6      
| 1008538  +38      sub         %g3, %g7, %g7    
| 100853c  +3c      subcc       %g7, %g6, %g0    
| 1008540  +40      sethi       1800000, %g7     
| 1008544  +44      or          %g0, 6, %g6      
| 1008548  +48      stb         %g6, [%g7 + 20]  
| 100854c  +4c      tcs         %g0, fffffffffffff001 
| 1008550  +50      sethi       0, %g0         
| 1008554  +54      bpcs,pn     %xcc, 1008640  
| 1008558  +58      or          %g0, 7, %g6    
| 100855c  +5c      stb         %g6, [%g7 + 20]
| 1008560  +60      srlx        %g3, 21, %g6  
| 1008564  +64      subcc       %g5, 1, %g0   
| 1008568  +68      bpgu,pn     %xcc, 1008640 
| 100856c  +6c      srlx        %g3, 17, %g5  
| 1008570  +70      and         %g6, 3ff, %g6 
| 1008574  +74      sll         %g6, 3, %g6   
| 1008578  +78      and         %g5, 3ff, %g5 
| 100857c  +7c      sll         %g5, 3, %g5   
| 1008580  +80      add         %g6, %g4, %g4 
| 1008584  +84      ldxa        [%g4]14, %g4  
| 1008588  +88      srlx        %g3, d, %g6   
| 100858c  +8c      and         %g6, 3ff, %g6 
| 1008590  +90      add         %g5, %g4, %g5 

This is not a memory operation, so it could not have generated
the fault.  Let's assume it was caused by the load at 1008584.
That's an attempt to load an entry from the first-level page table,
which is in the pmap.  The code here uses physical addresses for
the page tables, so it's either accessing a location that doesn't
exist or some device register instead of RAM.  This means that 
either the pointer to the current pmap is wrong, or some of the 
entries in the current pmap are corrupt.

| 1008594  +94      bprz,pn     %g4, 1008608
| 1008598  +98      sethi       0, %g0
| 100859c  +9c      ldxa        [%g5]14, %g4
| 10085a0  +a0      sll         %g6, 3, %g6   

What's the RAM configuration on this machine?  And specifically
which kernel is this?

Eduardo