Re: [RFC PATCH] -Wmemset-transposed-args (PR middle-end/61294)

[Jakub Jelinek <jakub@redhat.com>]

On Wed, Jul 09, 2014 at 12:51:32PM +0200, Richard Biener wrote:
> At least it shouldn't (they are not required to be shared and usually are not
> if they've gone a transition from TREE_OVERFLOW to !TREE_OVERFLOW).
> 
> Well, still feels ugly to me - but it's Jasons call in the end.

Another possibility is to keep that info on the side, something e.g.
the C FE already does.  There, c_parser_expr_list fills for the first 3
arguments (if requested by the caller) info about whether the argument used
to be originally a sizeof, and then the caller can look at this info.
For the literal zeros, it can be stored as a bitmask in a single char.

All of these warnings (-Wsizeof-pointer-memaccess, -Wsizeof-array-argument
and -Wmemset-transposed-args) are implemented in a hackish way, because we
fold everything too early.  Perhaps for such analysis we want a FOLDED_EXPR
which would have arguments what it has been folded to and the original tree,
for the purposes of code generation the first argument would be used and 
the second one only for the analysis.  We don't have that many spots where
we need the original trees to be analyzed yet for it to be worth it though
IMHO.

	Jakub