At 01:46 PM 7/15/2002, Kamal Prasad wrote:
>I have a program that does an mmap() of 800mb (out of
>1GB total physical memory) (using netbsd-curent). when
>I kill the app (that has done the mmap()), I get a
>panic.
>the trace is given below:-
>--------------------------------------------------------
>0xe9b9ad00: at panic+1b0
>0xe9b9adb0: at pmap_pvo_to_pte+1c4
>0xe9b9ade0: at pmap_pvo_remove+5c
>0xe9b9ae00: at pmap_page_protect+a4
>0xe9b9ae30: at uvm_anfree+1f0
>0xe9b9ae50: at amap_wipeout+120
>0xe9b9ae80: at amap_unref+4c
>0xe9b9aea0: at uvm_map_unreference_amap+54
>0xe9b9aec0: at uvm_unmap_detach+78
>0xe9b9aef0: at uvmspace_free+a0
>0xe9b9af10: at uvm_exit+2c
>0xe9b9af30: at reaper+cc
>0xe9b9af50: at fork_trampoline+10
>-------------------------------------------------
>if anyone knows the problem/fix to this, pl. let me
>know.
>thanks
>-kamal

Since the pmap is my code, I'll try to figure out the problem
now that it is reproducible.  BTW, how much swap does your system
have?


-- 
Matt Thomas               Internet:   matt@3am-software.com
3am Software Foundry      WWW URL:    http://www.3am-software.com/bio/matt/
Cupertino, CA             Disclaimer: I avow all knowledge of this message