re: bin/50744: uvm/uvm_extern.h: present but cannot be compiled

[matthew green <mrg%eterna.com.au@localhost>]

The following reply was made to PR bin/50744; it has been noted by GNATS.

From: matthew green <mrg%eterna.com.au@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost, nonakap%gmail.com@localhost
Subject: re: bin/50744: uvm/uvm_extern.h: present but cannot be compiled
Date: Fri, 05 Feb 2016 09:08:11 +1100

 Martin Husemann writes:
 > The following reply was made to PR bin/50744; it has been noted by GNATS=
 .
 > =
 
 > From: Martin Husemann <martin%duskware.de@localhost>
 > To: gnats-bugs%NetBSD.org@localhost
 > Cc: =
 
 > Subject: Re: bin/50744: uvm/uvm_extern.h: present but cannot be compiled
 > Date: Tue, 2 Feb 2016 12:02:18 +0100
 > =
 
 >  On Tue, Feb 02, 2016 at 07:45:00AM +0000, nonakap%gmail.com@localhost wrote:
 >  > When net-snmp build, the following message was displayed.
 >  > It was caused by the recent "Hide {p,v}{addr,size}_t and register_t" =
 changes.
 >  =
 
 >  Add =
 
 >  =
 
 >  #define _KMEMUSER      /* NetBSD hides many types for regular userland =
 */
 >  =
 
 >  before the first include in that file.
 
 this doesn't seem like the right fix to me.
 
 <uvm/uvm_extern.h> is a userland API we've expsoed for a long time,
 and it should not require defining special things to get it to
 compile.  if the normal usage is including kernel-specific stuff
 that seems like the right place to fix it to me.
 
 
 .mrg.