From: Marek Polacek <polacek@redhat.com>
To: GCC Patches <gcc-patches@gcc.gnu.org>,
Joseph Myers <joseph@codesourcery.com>,
David Malcolm <dmalcolm@redhat.com>
Subject: C PATCH to fix ICE with -Wformat and zero-length array (PR c/80919)
Date: Fri, 02 Jun 2017 17:19:00 -0000 [thread overview]
Message-ID: <20170602171912.GH3413@redhat.com> (raw)
In the C FE, zero-length arrays require structural equality, so we can't
compare their canonical types, 'cause they're NULL. But matching_type_p didn't
know that so we were crashing. With this patch the ICE is gone and the warning
messages are the same as with e.g. "int a[1]".
Bootstrapped/regtested on x86_64-linux, ok for trunk/7?
2017-06-02 Marek Polacek <polacek@redhat.com>
PR c/80919
* c-format.c (matching_type_p): Return false if any of the types
requires structural equality.
* gcc.dg/format/pr80919.c: New test.
diff --git gcc/c-family/c-format.c gcc/c-family/c-format.c
index faef267..732339b 100644
--- gcc/c-family/c-format.c
+++ gcc/c-family/c-format.c
@@ -3278,6 +3278,12 @@ matching_type_p (tree spec_type, tree arg_type)
gcc_assert (spec_type);
gcc_assert (arg_type);
+ /* If any of the types requires structural equality, we can't compare
+ their canonical types. */
+ if (TYPE_STRUCTURAL_EQUALITY_P (spec_type)
+ || TYPE_STRUCTURAL_EQUALITY_P (arg_type))
+ return false;
+
spec_type = TYPE_CANONICAL (spec_type);
arg_type = TYPE_CANONICAL (arg_type);
diff --git gcc/testsuite/gcc.dg/format/pr80919.c gcc/testsuite/gcc.dg/format/pr80919.c
index e69de29..510c2e4 100644
--- gcc/testsuite/gcc.dg/format/pr80919.c
+++ gcc/testsuite/gcc.dg/format/pr80919.c
@@ -0,0 +1,16 @@
+/* PR c/80919 */
+/* { dg-do compile } */
+/* { dg-options "-Wall" } */
+
+void
+fn (void)
+{
+ int a[0];
+ __builtin_printf("%d\n", &a); /* { dg-warning "expects argument of type" } */
+ __builtin_printf("%i\n", &a); /* { dg-warning "expects argument of type" } */
+
+ __builtin_printf("%o\n", &a); /* { dg-warning "expects argument of type" } */
+ __builtin_printf("%u\n", &a); /* { dg-warning "expects argument of type" } */
+ __builtin_printf("%x\n", &a); /* { dg-warning "expects argument of type" } */
+ __builtin_printf("%X\n", &a); /* { dg-warning "expects argument of type" } */
+}
Marek
next reply other threads:[~2017-06-02 17:19 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-02 17:19 Marek Polacek [this message]
2017-06-02 19:51 ` Joseph Myers
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=20170602171912.GH3413@redhat.com \
--to=polacek@redhat.com \
--cc=dmalcolm@redhat.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=joseph@codesourcery.com \
/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).