[PATCH] Fix memory leak in printf_positional

[Joseph Myers]

On Thu, 27 Aug 2015, Carlos O'Donell wrote:

> I agree, but I don't think anyone should spend more than an hour trying
> to find such a test case. The static analysis tools can show you a failure,

I really don't think this case should take that long to find how to 
trigger the leak.

There will of course be cases where the reason for no test case is 
something like "this bug only appears if you allocate memory occupying at 
least 3/4 of the address space, which is impossible on most current 64-bit 
systems and unreasonable for the testsuite to do on 32-bit systems even if 
possible" - or cases where it's unclear whether any combination of 
circumstances can cause the code in question to be reached in practice.  
But the starting point for most user-visible bug fixes should be that a 
test is added if it's straightforward to figure out how to test for the 
bug reliably within the context of the existing test infrastructure.

-- 
Joseph S. Myers
joseph@codesourcery.com