On Wed, Sep 27, 2023 at 11:15:26AM +0000, Richard Biener wrote:
> > tree-vect-patterns.cc:2947  unprom.quick_grow (nops);
> > T = vect_unpromoted_value
> > Go for quick_grow_cleared?  Something else?
> 
> The CTOR zero-initializes everything, so maybe it can go.  In theory
> .set_op could also be changed to .push_op ...

So, I had a look at this and I think using quick_grow_cleared is best choice
here.  The nops is 2 or 1 most of the time, worst case 3, so the price of
extra initialization of 4 pointer-sized-or-less members times 1, 2 or 3
doesn't seem worth bothering, it is similar to the bitmap_head case where
we already pay the price for just one structure anytime we do
  vect_unpromoted_value unprom_diff;
(and later set_op on it) or even
  vect_unpromoted_value unprom0[2];

With this patch and Richard S's poly_int_pod removal the static_assert can
be enabled as well and gcc builds.

Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

The second patch waits for the poly_int_pod removal commit and has been just
build tested but not bootstrapped yet.

2023-09-29  Jakub Jelinek  <jakub@redhat.com>

	* tree-vect-patterns.cc (vect_recog_over_widening_pattern): Use
	quick_grow_cleared method on unprom rather than quick_grow.

--- gcc/tree-vect-patterns.cc.jj	2023-08-24 15:37:29.321410276 +0200
+++ gcc/tree-vect-patterns.cc	2023-09-29 09:45:27.980168865 +0200
@@ -2944,7 +2944,7 @@ vect_recog_over_widening_pattern (vec_in
   /* Check the operands.  */
   unsigned int nops = gimple_num_ops (last_stmt) - first_op;
   auto_vec <vect_unpromoted_value, 3> unprom (nops);
-  unprom.quick_grow (nops);
+  unprom.quick_grow_cleared (nops);
   unsigned int min_precision = 0;
   bool single_use_p = false;
   for (unsigned int i = 0; i < nops; ++i)


	Jakub