public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r13-4081] ragen-op-float: Fix up float_binary_op_range_finish [PR107668]
@ 2022-11-16  6:36 Jakub Jelinek
  0 siblings, 0 replies; only message in thread
From: Jakub Jelinek @ 2022-11-16  6:36 UTC (permalink / raw)
  To: gcc-cvs

https://gcc.gnu.org/g:7c6cd9c05efca29a1a9635b81c86cbad25bbdbbe

commit r13-4081-g7c6cd9c05efca29a1a9635b81c86cbad25bbdbbe
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed Nov 16 07:30:07 2022 +0100

    ragen-op-float: Fix up float_binary_op_range_finish [PR107668]
    
    The following testcase ICEs, because when !HONOR_NANS but
    HONOR_SIGNED_ZEROS, if we see
    lhs = op1 * op2;
    and know that lhs is [-0.0, 0.0] and op2 is [0.0, 0.0], the
    division of these two yields UNDEFINED and clear_nan () on it
    fails an assert.  With HONOR_NANS it would actually result in
    a known NAN, but when NANs aren't honored, we clear the NAN bits.
    Now, for the above case we actually don't know anything about
    the op1 range (except that it isn't a NAN/INF because of
    !HONOR_NANS !HONOR_INFINITIES), so I think the best is just
    to return VARYING for the case we get UNDEFINED as well.
    
    If we want, the op[12]_range methods perhaps can handle the
    corner cases earlier separately, say for
    lhs [0.0, 0.0] and op2 [0.0, 0.0] when HONOR_SIGNED_ZEROS this
    would be just [0.0, MAX].
    
    2022-11-16  Jakub Jelinek  <jakub@redhat.com>
    
            PR tree-optimization/107668
            * range-op-float.cc (float_binary_op_range_finish): Set VARYING
            also when r is UNDEFINED.
    
            * gcc.dg/ubsan/pr107668.c: New test.

Diff:
---
 gcc/range-op-float.cc                 |  3 ++-
 gcc/testsuite/gcc.dg/ubsan/pr107668.c | 12 ++++++++++++
 2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/gcc/range-op-float.cc b/gcc/range-op-float.cc
index 53a0928c6aa..0c4ec8dd178 100644
--- a/gcc/range-op-float.cc
+++ b/gcc/range-op-float.cc
@@ -1891,8 +1891,9 @@ float_binary_op_range_finish (bool ret, frange &r, tree type,
   // or the reverse operation introduced a known NAN.
   // Say for lhs = op1 * op2 if lhs is [-0, +0] and op2 is too,
   // 0 / 0 is known NAN.  Just punt in that case.
+  // If NANs aren't honored, we get for 0 / 0 UNDEFINED, so punt as well.
   // Or if lhs is a known NAN, we also don't know anything.
-  if (r.known_isnan () || lhs.known_isnan ())
+  if (r.known_isnan () || lhs.known_isnan () || r.undefined_p ())
     {
       r.set_varying (type);
       return true;
diff --git a/gcc/testsuite/gcc.dg/ubsan/pr107668.c b/gcc/testsuite/gcc.dg/ubsan/pr107668.c
new file mode 100644
index 00000000000..e94d6cdb530
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/ubsan/pr107668.c
@@ -0,0 +1,12 @@
+/* PR tree-optimization/107668 */
+/* { dg-do compile } */
+/* { dg-options "-ffast-math -fno-associative-math -fsanitize=float-cast-overflow -fno-guess-branch-probability -fsigned-zeros" } */
+
+_Complex int c;
+int i;
+
+void
+foo (void)
+{
+  c /= (_Complex) i;
+}

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2022-11-16  6:36 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-11-16  6:36 [gcc r13-4081] ragen-op-float: Fix up float_binary_op_range_finish [PR107668] Jakub Jelinek

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).