hi,

I've checked in a fix for this assertion failure to -current.
I'll have it pulled up to 3.x in a few days.

the cc and ld SEGVs you were seeing were most likely due to the ld.elf_so
thing I fixed yesterday.

I fixed the assertion after cpu_reboot() quite a while ago
(but maybe you already knew that).

-Chuck


On Thu, May 05, 2005 at 01:30:53PM +0000, Rui Paulo wrote:
> 
> Hi,
> I'm getting the following panic under NetBSD 3.0_BETA while building the
> 3.0_BETA userland under a 2.0 userland.
> 
> panic: kernel diagnostic assertion "emsr & PSL_EE" failed: file "/amd/quanta/roo
> t/export/netbsd/netbsd-3/src/sys/arch/macppc/macppc/extintr.c", line 811
> 
> I'm using NFS exported sources, as you can see.
> 
> db> bt
> 0xd51bbc30: at panic+0x19c
> 0xd51bbcc0: at __assert+0x28
> 0xd51bbcd0: at do_pending_int+0x280
> 0xd51bbd10: at splx+0x40
> 0xd51bbd20: at pool_do_put+0x1f4
> 0xd51bbd70: at pmap_pvo_remove+0x240
> 0xd51bbd90: at pmap_remove+0x90
> 0xd51bbdb0: at uvm_unmap_remove+0x350
> 0xd51bbdf0: at uvmspace_free+0xe4
> 0xd51bbe10: at uvm_proc_exit+0x54
> 0xd51bbe30: at exit1+0x190
> 0xd51bbeb0: at sys_exit+0x38
> 0xd51bbed0: at syscall_plain+0xe0
> 0xd51bbf40: user SC trap #1 by 0xeffb1c3c: srr1=0xf032
>             r1=0xffffb8a0 cr=0x24000022 xer=0 ctr=0xeffb1c34
> 
> 
> db> ps/a   
>  PID          COMMAND      STRUCT PROC *            UAREA *     VMSPACE/VM_MAP
> >10389            cc1          0x9049c98         0xd51b8000           0x5f5d90
> 
> db> ps     
>  PID           PPID     PGRP        UID S   FLAGS LWPS          COMMAND    WAIT
> >10389         6801     4620       1000 2  0x6002    1              cc1
> 
> db> show map 0x9049c98
> MAP 0x9049c98: [0x1->0x0]
>         #ent=0, sz=105, ref=0, version=0, flags=0x9e76d28
>         pmap=0x9049b08(resident=0, wired=41810)
> 
> db> show uvmexp
> Current UVM status:
>   pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
>   38048 VM pages: 27602 active, 2796 inactive, 120 wired, 2791 free
>   min  10% (25) anon, 10% (25) file, 5% (12) exec
>   max  80% (204) anon, 50% (128) file, 30% (76) exec
>   pages  9560 anon, 20818 file, 2031 exec
>   freemin=64, free-target=85, inactive-target=11076, wired-max=12682
>   faults=6345201, traps=6938954, intrs=3180835, ctxswitch=816409
>   softint=670906, syscalls=0, swapins=92, swapouts=103
>   fault counts:
>     noram=5, noanon=0, pgwait=0, pgrele=0
>     ok relocks(total)=6069(6069), anget(retrys)=469965(13), amapcopy=592359
>     neighbor anon/obj pg=309899/8417368, gets(lock/unlock)=2408044/6056
>     cases: anon=367420, anoncow=101721, obj=1830778, prcopy=577266, przero=33735
> 26
>   daemon and swap counts:
>     woke=146, revs=144, scans=151344, obscans=11788, anscans=1336
>     busy=198, freed=37705, reactivate=122909, deactivate=226588
>     pageouts=90, pending=0, nswget=13
>     nswapdev=1, nanon=68489, nanonneeded=68489 nfreeanon=59633
>     swpages=32841, swpginuse=1367, swpgonly=1354 paging=0
> 
> 
> Also, I'm getting some gcc (cc1 and ld) segmentation faults.
> 
> May these two problems be somewhat related to memory problems or CPU problems
> or has anyone already saw it ?
> 
> P.S.: This has nothing to do with CPU reboot as in 
> http://mail-index.netbsd.org/port-macppc/2005/01/18/0001.html
> 
> Thanks.
> 
> -- 
> Rui Paulo <rpaulo@netbsd-pt.org>