Purify errors

[Alex Kiernan]

Russ Allbery <rra at stanford.edu> writes:

> Katsuhiro Kondou <kondou at nec.co.jp> writes:
> > Alex Kiernan <alexk at demon.net> wrote;
> 
> > } 	setvbuf        [libc.so.1]
> > } 	ReopenLog      [innd.c:260]
> 
> > I'm not sure how this should be fixed.
> 
> Yeah, that one sure looks like a libc bug to me.
> 

I think they're bogus (they're marked as PLK - potential leaks);
looking at what Solaris is doing, it doesn't store the pointer
returned from malloc, but an offset from it; I suspect thats enough to
confuse purify.

BTW no more UMR/ABR following this mornings change, still trying to
figure out where the memory thats being leaked is from though :-(

-- 
Alex Kiernan, Principal Engineer, Development, Thus PLC