public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r13-4501] range-op-float: Fix up ICE in lower_bound [PR107975]
@ 2022-12-06  9:24 Jakub Jelinek
  0 siblings, 0 replies; only message in thread
From: Jakub Jelinek @ 2022-12-06  9:24 UTC (permalink / raw)
  To: gcc-cvs

https://gcc.gnu.org/g:0525a7fad2a5b1d933a9662c11aa074b38cfa3d5

commit r13-4501-g0525a7fad2a5b1d933a9662c11aa074b38cfa3d5
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Tue Dec 6 10:23:55 2022 +0100

    range-op-float: Fix up ICE in lower_bound [PR107975]
    
    According to https://gcc.gnu.org/pipermail/gcc-regression/2022-December/077258.html
    my patch caused some ICEs, e.g. the following testcase ICEs.
    The problem is that lower_bound and upper_bound methods on a france assert
    that the range isn't VR_NAN or VR_UNDEFINED.
    All the op1_range/op2_range methods already return early if lhs.undefined_p,
    but the other cases (when lhs is VR_NAN or the other opN is VR_NAN or
    VR_UNDEFINED) aren't.  float_binary_op_range_finish will DTRT for those
    cases already.
    
    2022-12-06  Jakub Jelinek  <jakub@redhat.com>
    
            PR tree-optimization/107975
            * range-op-float.cc (foperator_mult::op1_range,
            foperator_div::op1_range, foperator_div::op2_range): Just
            return float_binary_op_range_finish result if lhs is known
            NAN, or the other operand is known NAN or UNDEFINED.
    
            * gcc.dg/pr107975.c: New test.

Diff:
---
 gcc/range-op-float.cc           |  6 ++++++
 gcc/testsuite/gcc.dg/pr107975.c | 15 +++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/gcc/range-op-float.cc b/gcc/range-op-float.cc
index e9455a929ae..ac57c8a61d0 100644
--- a/gcc/range-op-float.cc
+++ b/gcc/range-op-float.cc
@@ -2146,6 +2146,8 @@ public:
     bool ret = rdiv.fold_range (r, type, lhs, op2);
     if (ret == false)
       return false;
+    if (lhs.known_isnan () || op2.known_isnan () || op2.undefined_p ())
+      return float_binary_op_range_finish (ret, r, type, lhs);
     const REAL_VALUE_TYPE &lhs_lb = lhs.lower_bound ();
     const REAL_VALUE_TYPE &lhs_ub = lhs.upper_bound ();
     const REAL_VALUE_TYPE &op2_lb = op2.lower_bound ();
@@ -2296,6 +2298,8 @@ public:
     bool ret = fop_mult.fold_range (r, type, lhs, op2);
     if (!ret)
       return ret;
+    if (lhs.known_isnan () || op2.known_isnan () || op2.undefined_p ())
+      return float_binary_op_range_finish (ret, r, type, lhs);
     const REAL_VALUE_TYPE &lhs_lb = lhs.lower_bound ();
     const REAL_VALUE_TYPE &lhs_ub = lhs.upper_bound ();
     const REAL_VALUE_TYPE &op2_lb = op2.lower_bound ();
@@ -2325,6 +2329,8 @@ public:
     bool ret = fold_range (r, type, op1, lhs);
     if (!ret)
       return ret;
+    if (lhs.known_isnan () || op1.known_isnan () || op1.undefined_p ())
+      return float_binary_op_range_finish (ret, r, type, lhs);
     const REAL_VALUE_TYPE &lhs_lb = lhs.lower_bound ();
     const REAL_VALUE_TYPE &lhs_ub = lhs.upper_bound ();
     const REAL_VALUE_TYPE &op1_lb = op1.lower_bound ();
diff --git a/gcc/testsuite/gcc.dg/pr107975.c b/gcc/testsuite/gcc.dg/pr107975.c
new file mode 100644
index 00000000000..7710f6f4a5d
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/pr107975.c
@@ -0,0 +1,15 @@
+/* PR tree-optimization/107975 */
+/* { dg-do compile } */
+/* { dg-options "-O2" } */
+/* { dg-add-options ieee } */
+
+double
+foo (double x, double y)
+{
+  if (x == 42.0)
+    return 1.0;
+  double r = x * y;
+  if (!__builtin_isnan (r))
+    __builtin_unreachable ();
+  return r;
+}

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

only message in thread, other threads:[~2022-12-06  9:24 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-12-06  9:24 [gcc r13-4501] range-op-float: Fix up ICE in lower_bound [PR107975] 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).