Fix PR48794 some more
Michael Matz
matz@suse.de
Wed Jan 25 16:04:00 GMT 2012
Hi,
On Wed, 25 Jan 2012, Michael Matz wrote:
> so, the below adjusted testcase from PR48794 still fails for the same
> reasons (regions still referenced from RESX being removed). I was split
> minds about if that's a new bug or just an extension of the old bug, so
> I hijacked the old PR. In any case, remove_unreachable_handlers_no_lp
> needs similar handling like remove_unreachable_handlers.
>
> Patch fixes the segfault and is currently in regstrapping on x86_64-linux.
> Okay for trunk?
Actually, resx/eh_dispatch always are the last BB statements, so the loop
doesn't need to look at all statements in a BB, making it quite somewhat
faster. Consider the tree-eh.c to be looking like so:
Index: tree-eh.c
===================================================================
--- tree-eh.c (revision 183524)
+++ tree-eh.c (working copy)
@@ -3617,14 +3617,40 @@ remove_unreachable_handlers_no_lp (void)
{
eh_region r;
int i;
+ sbitmap r_reachable;
+ basic_block bb;
+
+ r_reachable = sbitmap_alloc (VEC_length (eh_region, cfun->eh->region_array));
+ sbitmap_zero (r_reachable);
+
+ FOR_EACH_BB (bb)
+ {
+ gimple stmt = last_stmt (bb);
+ if (stmt)
+ /* Avoid removing regions referenced from RESX/EH_DISPATCH. */
+ switch (gimple_code (stmt))
+ {
+ case GIMPLE_RESX:
+ SET_BIT (r_reachable, gimple_resx_region (stmt));
+ break;
+ case GIMPLE_EH_DISPATCH:
+ SET_BIT (r_reachable, gimple_eh_dispatch_region (stmt));
+ break;
+ default:
+ break;
+ }
+ }
for (i = 1; VEC_iterate (eh_region, cfun->eh->region_array, i, r); ++i)
- if (r && r->landing_pads == NULL && r->type != ERT_MUST_NOT_THROW)
+ if (r && r->landing_pads == NULL && r->type != ERT_MUST_NOT_THROW
+ && !TEST_BIT (r_reachable, i))
{
if (dump_file)
fprintf (dump_file, "Removing unreachable region %d\n", i);
remove_eh_handler (r);
}
+
+ sbitmap_free (r_reachable);
}
/* Undo critical edge splitting on an EH landing pad. Earlier, we
Ciao,
Michael.
More information about the Gcc-patches
mailing list