+2011-02-08 Richard Guenther <rguenther@suse.de>
+
+ PR middle-end/47639
+ * tree-vect-generic.c (expand_vector_operations_1): Update
+ stmts here ...
+ (expand_vector_operations): ... not here. Cleanup EH info
+ and the CFG if required.
+
2011-02-08 Richard Guenther <rguenther@suse.de>
PR tree-optimization/47641
+2011-02-08 Richard Guenther <rguenther@suse.de>
+
+ PR middle-end/47639
+ * g++.dg/opt/pr47639.c: New testcase.
+
2011-02-08 Richard Guenther <rguenther@suse.de>
PR tree-optimization/47632
--- /dev/null
+// { dg-do compile }
+// { dg-options "-fnon-call-exceptions" }
+
+typedef int __attribute__ ((vector_size (8))) vec;
+
+vec foo (vec v1, vec v2)
+{
+ try
+ {
+ return v1 / v2;
+ }
+ catch (...)
+ {
+ throw;
+ }
+}
+
way to do it is change expand_vector_operation and its callees to
return a tree_code, RHS1 and RHS2 instead of a tree. */
gimple_assign_set_rhs_from_tree (gsi, new_rhs);
-
- gimple_set_modified (gsi_stmt (*gsi), true);
+ update_stmt (gsi_stmt (*gsi));
}
\f
/* Use this to lower vector operations introduced by the vectorizer,
{
gimple_stmt_iterator gsi;
basic_block bb;
+ bool cfg_changed = false;
FOR_EACH_BB (bb)
{
for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
{
expand_vector_operations_1 (&gsi);
- update_stmt_if_modified (gsi_stmt (gsi));
+ /* ??? If we do not cleanup EH then we will ICE in
+ verification. But in reality we have created wrong-code
+ as we did not properly transition EH info and edges to
+ the piecewise computations. */
+ if (maybe_clean_eh_stmt (gsi_stmt (gsi))
+ && gimple_purge_dead_eh_edges (bb))
+ cfg_changed = true;
}
}
- return 0;
+
+ return cfg_changed ? TODO_cleanup_cfg : 0;
}
struct gimple_opt_pass pass_lower_vector =