Re: kern/50985: uvm_mmap.c::range_test should use runtime limits of the vmspace?

[Martin Husemann <martin%duskware.de@localhost>]

The following reply was made to PR kern/50985; it has been noted by GNATS.

From: Martin Husemann <martin%duskware.de@localhost>
To: matthew green <mrg%eterna.com.au@localhost>
Cc: gnats-bugs%NetBSD.org@localhost
Subject: Re: kern/50985: uvm_mmap.c::range_test should use runtime limits of the vmspace?
Date: Mon, 21 Mar 2016 22:13:34 +0100

 On Tue, Mar 22, 2016 at 07:48:56AM +1100, matthew green wrote:
 > >  -	vaddr_t vm_min_address = VM_MIN_ADDRESS;
 > >  -	vaddr_t vm_max_address = VM_MAXUSER_ADDRESS;
 > >  +	vaddr_t vm_min_address = vm_map_min(map);
 > >  +	vaddr_t vm_max_address = vm_map_max(map);
 > 
 > the max address now is in the kernel space on some platforms, i think.
 
 It comes from exec paccks ep_vm_maxaddr, which usually is VM_MAXUSER_ADDRESS,
 I think.
 
 We can min()/max() it with VM_MAXUSER_ADDRESS/VM_MIN_ADDRESS though, or
 KASSERT that the limit is within the static limit.
 
 Martin