public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
From: Richard Biener <rguenth@gcc.gnu.org>
To: gcc-cvs@gcc.gnu.org
Subject: [gcc r14-9135] tree-optimization/114048 - ICE in copy_reference_ops_from_ref
Date: Thu, 22 Feb 2024 14:48:22 +0000 (GMT)	[thread overview]
Message-ID: <20240222144822.2B3DD3858413@sourceware.org> (raw)

https://gcc.gnu.org/g:92c40297991f51e7fa942f29517bc4398fce33f9

commit r14-9135-g92c40297991f51e7fa942f29517bc4398fce33f9
Author: Richard Biener <rguenther@suse.de>
Date:   Thu Feb 22 14:22:29 2024 +0100

    tree-optimization/114048 - ICE in copy_reference_ops_from_ref
    
    The following adds another omission to the assert verifying we're
    not running into spurious off == -1.
    
            PR tree-optimization/114048
            * tree-ssa-sccvn.cc (copy_reference_ops_from_ref): MEM_REF
            can also produce -1 off.
    
            * gcc.dg/torture/pr114048.c: New testcase.

Diff:
---
 gcc/testsuite/gcc.dg/torture/pr114048.c | 25 +++++++++++++++++++++++++
 gcc/tree-ssa-sccvn.cc                   |  2 ++
 2 files changed, 27 insertions(+)

diff --git a/gcc/testsuite/gcc.dg/torture/pr114048.c b/gcc/testsuite/gcc.dg/torture/pr114048.c
new file mode 100644
index 000000000000..338000b30062
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/torture/pr114048.c
@@ -0,0 +1,25 @@
+/* { dg-do compile } */
+
+typedef struct {
+  void *child[2];
+  char otherbits;
+} critbit0_node;
+
+int allprefixed_traverse(char *top)
+{
+  if (top)
+    {
+      critbit0_node *q = (void *)top - 1;
+      int direction = 0;
+      for (;; ++direction)
+	switch (allprefixed_traverse(q->child[direction]))
+	  {
+	  case 1:
+	    break;
+	  case 0:
+	    return 0;
+	  default:
+	    return 1;
+	  }
+    }
+}
diff --git a/gcc/tree-ssa-sccvn.cc b/gcc/tree-ssa-sccvn.cc
index 3e93685e80a6..2587eb1c5052 100644
--- a/gcc/tree-ssa-sccvn.cc
+++ b/gcc/tree-ssa-sccvn.cc
@@ -1185,6 +1185,8 @@ copy_reference_ops_from_ref (tree ref, vec<vn_reference_op_s> *result)
 	      else
 		{
 		  gcc_assert (known_ne (op.off, -1)
+			      /* The constant offset can be -1.  */
+			      || op.opcode == MEM_REF
 			      /* Out-of-bound indices can compute to
 				 a known -1 offset.  */
 			      || ((op.opcode == ARRAY_REF

                 reply	other threads:[~2024-02-22 14:48 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240222144822.2B3DD3858413@sourceware.org \
    --to=rguenth@gcc.gnu.org \
    --cc=gcc-cvs@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).