On 1/23/25 5:45 AM, Jakub Jelinek wrote:
Hi!
After committing the append_ctor_to_tree_vector patch, I've realized
that for the larger constructors make_tree_vector_from_ctor unnecessarily
wastes one GC vector; make_tree_vector () / release_tree_vector () only
caches GC vectors from 4 to 16 allocated tree elements, so in the likely
case of a rather small ctor using make_tree_vector () can be beneficial,
we can pick something from the cache and if we don't need it later,
pt.cc calls release_tree_vector on it to return it back to the cache.
But for the larger ctors, we just eat one vector from the cache, never
use it (because the vec_safe_reserve will immediately allocate a different
vector) and never return it back to the cache.
So, the following patch passes NULL for the larger vectors, which
append_ctor_to_tree_vector handles just fine now (vec_safe_reserve will
just allocate appropriately sized vector).
Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
OK.
2025-01-23 Jakub Jelinek <ja...@redhat.com>
* c-common.cc (make_tree_vector_from_ctor): Only use make_tree_vector
for ctors with <= 16 elements.
--- gcc/c-family/c-common.cc.jj 2025-01-22 09:22:53.037015592 +0100
+++ gcc/c-family/c-common.cc 2025-01-22 19:04:28.113360324 +0100
@@ -9044,7 +9044,8 @@ append_ctor_to_tree_vector (vec<tree, va
vec<tree, va_gc> *
make_tree_vector_from_ctor (tree ctor)
{
- vec<tree,va_gc> *ret = make_tree_vector ();
+ vec<tree,va_gc> *ret
+ = CONSTRUCTOR_NELTS (ctor) <= 16 ? make_tree_vector () : NULL;
return append_ctor_to_tree_vector (ret, ctor);
}
Jakub
