On Tue, 9 Oct 2001, Greg MATTHEWS wrote:

> hi...
>
> just built xlockmore from source on sparc/NetBSD 1.5.1
>
> seemed to work fine until, lots of this:
>
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: chunk is already free.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: chunk is already free.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: chunk is already free.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
> xlock in free(): warning: junk pointer, too high to make sense.
>
> and then:
>
> Access control list restored.
> xlock: caught signal 11 while running pacman mode (uid 1000).

	The malloc() library in NetBSD 1.5 and later (taken from FreeBSD
	IIRC) is a lot better at warning about invalid free() calls.
	Usually they are harmless (if signs of sloppy programming :)

	Can you reproduce the problem with 'xlock -mode pacman -nolock'.
	Both my 1.5.2 i386 and 1.5Y sparc64 boxes seem fine with that
	(I don't have a 1.5.x sparc box with X installed :)

-- 
		David/absolute		-- www.netbsd.org: No hype required --