Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site mtxinu.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!umcp-cs!gymble!lll-crg!dual!unisoft!mtxinu!ed
From: ed@mtxinu.UUCP (Ed Gould)
Newsgroups: net.unix-wizards
Subject: Re: UPAGES kernel constant
Message-ID: <411@mtxinu.UUCP>
Date: Mon, 17-Jun-85 20:02:34 EDT
Article-I.D.: mtxinu.411
Posted: Mon Jun 17 20:02:34 1985
Date-Received: Sun, 23-Jun-85 10:25:52 EDT
References: <11029@brl-tgr.ARPA> <1462@utah-gr.UUCP>
Reply-To: ed@mtxinu.UUCP (Ed Gould)
Organization: mt Xinu, Berkeley, CA
Lines: 22

>In article <11029@brl-tgr.ARPA> droms@PURDUE.ARPA (Ralph E Droms) writes:
>>... compiled with KERNEL not defined.  This forces sys/h/param.h to include
>><machine/param.h>, rather than ../machine/param.h.  So, UPAGES must be
>>changed in both .h files for the kernel to be remade correctly.

In article <1462@utah-gr.UUCP> thomas@utah-gr.UUCP (Spencer W. Thomas) writes:
>
>If you had /usr/include/machine set up as a symbolic link to
>/sys/machine, this would not be a problem.

True, but it doesn't help if you're compiling a test kernel with
its own headers.  Adding a declaration of the appropriate
variable (uba_hd if I remember correctly) solves the "undefined"
problem:

	#define	KERNEL
	...
	struct	uba_hd	uba_hd[1];

-- 
Ed Gould		    mt Xinu, 2910 Seventh St., Berkeley, CA  94710  USA
{ucbvax,decvax}!mtxinu!ed   +1 415 644 0146