I only noticed after doing a bootstrap/test and pushing the
patch that I messed up the conditional. There was no failure
there so I suspect it was not being tested and most likely
does not run into the issue of why no_delete was added either.
Pushed the fix after a build.
gcc/ChangeLog:
* tree-ssa-dce.cc (simple_dce_from_worklist): Fix conditional
for no_delete with a call.
Signed-off-by: Andrea Pinski <[email protected]>
---
gcc/tree-ssa-dce.cc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/tree-ssa-dce.cc b/gcc/tree-ssa-dce.cc
index 9fdd102442a..18bc851b2cd 100644
--- a/gcc/tree-ssa-dce.cc
+++ b/gcc/tree-ssa-dce.cc
@@ -2224,8 +2224,8 @@ simple_dce_from_worklist (bitmap worklist, bitmap
need_eh_cleanup,
if (gimple_has_side_effects (t))
{
gcall *call = dyn_cast <gcall *> (t);
- // For no delete don't remove the lhs.
- if (call && no_delete)
+ // Remove the lhs for a call if not no_delete.
+ if (call && !no_delete)
{
gimple_call_set_lhs (call, NULL_TREE);
update_stmt (call);
--
2.43.0
