public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [C++/66270]  another may_alias crash
@ 2015-05-25 21:09 Nathan Sidwell
  2015-05-26  7:52 ` Jason Merrill
  0 siblings, 1 reply; 6+ messages in thread
From: Nathan Sidwell @ 2015-05-25 21:09 UTC (permalink / raw)
  To: Jason Merrill, GCC Patches

[-- Attachment #1: Type: text/plain, Size: 1798 bytes --]

This patch addresses 66270, another case where may_alias disrupted the canonical 
type system.  We ICE as TYPE_CANONICALs differ, but comptypes think they are the 
same.

There seems to be a bit of confusion as to whether pointers that differ only in 
TYPE_REF_CAN_ALIAS_ALL are the same canonical type or not.

Firstly, in tree.c build_pointer_type_for_mode, when the pointed-to type is not 
its own canonical type, that means the newly constructed pointer type is 
(possibly) not canonical either.  So we explicitly build a canonical type with:

   else if (TYPE_CANONICAL (to_type) != to_type)
     TYPE_CANONICAL (t)
       = build_reference_type_for_mode (TYPE_CANONICAL (to_type),
				       mode, false);

But we're passing 'false' in as 'can_alias_all', rather than pass the value 
passed into us.  That'll make a difference if the caller passed in true and 
to_type doesn't have may_alias set.  This is inconsistent at least, because we 
could sometimes end up with canonical types with T_R_C_A_A set (to-type is 
canonical) and sometimes with it not set.  It seems the right solution is to 
consider T_R_C_A_A as a distinguisher, thus we should pass can_alias_all to the 
canonical type builder.  Note that it is ok to pass the possibly modified 
can_alias_all in, and not the incoming value, because we only ever modify it to 
make it true -- and in that case the same behavior would happen in the canonical 
type builder because to_type and TYPE_CANONICAL (to_type) should have the same 
may_alias attribute.

Anyway, that's a bit of collateral confusion I fell over investigating.  With 
that out of the way, we have  to teach comptypes that T_R_C_A_A affects pointer 
type equality.  Hence add such a check to POINTER_TYPE case there.

bootstrapped on x86-linux & tested, ok?

nathan

[-- Attachment #2: 66270.patch --]
[-- Type: text/x-patch, Size: 2077 bytes --]

2015-05-25  Nathan Sidwell  <nathan@acm.org>


	PR c++/66270
	* tree.c (build_pointer_type_for_mode): Pass can_alias_all to
	canonical type builder.
	(build_reference_type_for_mode): Likewise.

	cp/
	* typeck.c (structural_comptypes) [POINTER_TYPE]: Compare
	TYPE_REF_CAN_ALIAS_ALL.

	testsuite/
	* g++.dg/ext/alias-canon3.C: New.

Index: cp/typeck.c
===================================================================
--- cp/typeck.c	(revision 223636)
+++ cp/typeck.c	(working copy)
@@ -1307,6 +1307,8 @@ structural_comptypes (tree t1, tree t2,
       if (TYPE_MODE (t1) != TYPE_MODE (t2)
 	  || !same_type_p (TREE_TYPE (t1), TREE_TYPE (t2)))
 	return false;
+      if (TYPE_REF_CAN_ALIAS_ALL (t1) != TYPE_REF_CAN_ALIAS_ALL (t2))
+	return false;
       break;
 
     case METHOD_TYPE:
Index: tree.c
===================================================================
--- tree.c	(revision 223636)
+++ tree.c	(working copy)
@@ -7759,7 +7759,7 @@ build_pointer_type_for_mode (tree to_typ
   else if (TYPE_CANONICAL (to_type) != to_type)
     TYPE_CANONICAL (t)
       = build_pointer_type_for_mode (TYPE_CANONICAL (to_type),
-				     mode, false);
+				     mode, can_alias_all);
 
   /* Lay out the type.  This function has many callers that are concerned
      with expression-construction, and this simplifies them all.  */
@@ -7826,7 +7826,7 @@ build_reference_type_for_mode (tree to_t
   else if (TYPE_CANONICAL (to_type) != to_type)
     TYPE_CANONICAL (t)
       = build_reference_type_for_mode (TYPE_CANONICAL (to_type),
-				       mode, false);
+				       mode, can_alias_all);
 
   layout_type (t);
 
Index: testsuite/g++.dg/ext/alias-canon3.C
===================================================================
--- testsuite/g++.dg/ext/alias-canon3.C	(revision 0)
+++ testsuite/g++.dg/ext/alias-canon3.C	(working copy)
@@ -0,0 +1,12 @@
+// { dg-do compile }
+// PR c++/66270
+
+typedef float __m256 __attribute__ (( __vector_size__(32), __may_alias__ ));
+struct A {
+  __m256 ymm;
+  const float &f() const;
+};
+
+const float &A::f() const {
+  return ymm[1];
+}

^ permalink raw reply	[flat|nested] 6+ messages in thread
end of thread, other threads:[~2015-05-27 21:04 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-05-25 21:09 [C++/66270] another may_alias crash Nathan Sidwell
2015-05-26  7:52 ` Jason Merrill
2015-05-26 19:18   ` Nathan Sidwell
2015-05-27 12:42     ` Nathan Sidwell
2015-05-27 16:36     ` Jason Merrill
2015-05-27 21:26       ` Nathan Sidwell

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).