Re: NetBSD 7 and current issues

[Anders Magnusson <ragge%ludd.ltu.se@localhost>]

John Klos skrev den 2015-08-16 21:18:
> Has anyone besides me noticed that NetBSD 7 and current panic if you 
> try to run an aout binary?
>
> # ./dry2-1.5
> panic: usrptmap space leakage
> Stopped in pid 13.1 (dry2-1.5) at       netbsd:vpanic+0xce: pushl   $n
> db> bt
> panic: usrptmap space leakage
> Stack traceback :
> 0x84058c80: vpanic+0xce(0x8026e08a,0x84058d20)
> 0x84058ca0: snprintf+0x0(0x8026e08a)
> 0x84058cd4: pmap_getusrptes+0x195(0x816ed2d0,0x800000)
> 0x84058d38: pmap_enter+0x241(0x816ed2d0,0x40005000,0x1370000,0x5,0x20)
> 0x84058d88: uvm_fault_internal+0xb9b(0x81663d40,0x40004000,0x1,0)
> 0x84058ef8: trap+0x35d(0x84058fb4)
Interesting panic.
This panic occurs if no space in the user page table map can be found.
It tries to unmap process after process to find space, using the same 
way as swapping processes do.

However, we can see that pmap_enter tries tries to map in space at 
0x40005000, which means that it needs to map virtually the whole P1 pte 
space, which is the problem.
So the interesting thing is why this specific address is mapped. mmap'ed 
space should go just below the stack, and not in the bottom of P1.

-- Ragge