memecc_error() reads, in part,

        printf("\taddress: 0x%x%x\n", efar0 & ECC_AFR_PAH, efar1);

Surely that sbould be 0x%x%08x?  Otherwise (for example) a failure at
PA 0x102030405 will print as 0x12030405.  Confusing at best (though
not, strictly, ambiguous, since ECC_AFR_PAH is only one nibble wide and
thus the first %x will always generate exactly one digit - does anyone
have a NetBSD SPARC with enough RAM for the PAH bits to be nonzero?).

Also, is there any easy way to tell which module is responsible for a
given error?  I just got

NMI: system interrupts: 10000000<VME=0,SBUS=0,M>
memory error:
       EFSR: 7931<CE,DW=3,SYNDROME=79>
       MBus transaction: 8fffcd50<VAH=0,TYPE=5,SIZE=5,C,VA=ff,S,MID=8>
       address: 0x029a9bb8

and I'm wondering which stick of RAM needs reseating/replacing (if,
indeed, it isn't purely random - the error _was_ corrected, I note).

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B