Re: memset bugs.
On Wed, 2009-11-25 at 23:39 -0500, Tom Lane wrote:
John Reiser <jreiser(a)bitwagon.com> writes:
> On 11/25/2009 02:03 PM, Dave Jones wrote:
>> A zero sized memset is always a bug.
> No, memset(,,0) is not always a bug.
I think it's reasonably safe to assume that a *literal constant* zero in
the third argument is a bug. Whether the header macros can distinguish
that from compile-time-constant expressions is an interesting question,
but if they can, +1 for throwing an error.
I logged some time ago a trivial patch of
https://bugzilla.redhat.com/show_bug.cgi?id=532492 which is apparently
in rawhide now to generally avoid a warning on memset(foo, 0, 0)
C.