public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jakub Jelinek <jakub@redhat.com>
To: Jason Merrill <jason@redhat.com>,
	"Joseph S. Myers" <joseph@codesourcery.com>,
	Marek Polacek <polacek@redhat.com>
Cc: gcc-patches@gcc.gnu.org
Subject: [PATCH] c, c++: Allow ignoring -Winit-self through pragmas [PR105593]
Date: Fri, 13 Jan 2023 01:32:38 +0100	[thread overview]
Message-ID: <Y8CmpuU+Js7Tlsqz@tucnak> (raw)

Hi!

As mentioned in the PR, various x86 intrinsics need to return
an uninitialized vector.  Currently they use self initialization
to avoid -Wuninitialized warnings, which works fine in C, but
doesn't work in C++ where -Winit-self is enabled in -Wall.
We don't have an attribute to mark a variable as knowingly
uninitialized (the uninitialized attribute exists but means
something else, only in the -ftrivial-auto-var-init context),
and trying to suppress either -Wuninitialized or -Winit-self
inside of the _mm_undefined_ps etc. intrinsic definitions
doesn't work, one needs to currently disable through pragmas
-Wuninitialized warning at the point where _mm_undefined_ps etc.
result is actually used, but that goes against the intent of
those intrinsics.

The -Winit-self warning option actually doesn't do any warning,
all we do is record a suppression for -Winit-self if !warn_init_self
on the decl definition and later look that up in uninit pass.

The following patch changes those !warn_init_self tests which
are true only based on the command line option setting, not based
on GCC diagnostic pragma overrides to
!warning_enabled_at (DECL_SOURCE_LOCATION (decl), OPT_Winit_self)
such that it takes them into account.

Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

Will post incremental patch for the intrinsic headers.

2023-01-13  Jakub Jelinek  <jakub@redhat.com>

	PR c++/105593
gcc/c/
	* c-parser.cc (c_parser_initializer): Check warning_enabled_at
	at the DECL_SOURCE_LOCATION (decl) for OPT_Winit_self instead
	of warn_init_self.
gcc/cp/
	* decl.cc (cp_finish_decl): Check warning_enabled_at
	at the DECL_SOURCE_LOCATION (decl) for OPT_Winit_self instead
	of warn_init_self.
gcc/testsuite/
	* c-c++-common/Winit-self3.c: New test.
	* c-c++-common/Winit-self4.c: New test.
	* c-c++-common/Winit-self5.c: New test.

--- gcc/c/c-parser.cc.jj	2023-01-11 22:18:25.560492345 +0100
+++ gcc/c/c-parser.cc	2023-01-12 15:30:10.460233783 +0100
@@ -5701,7 +5701,7 @@ c_parser_initializer (c_parser *parser,
 	  && !DECL_EXTERNAL (decl)
 	  && !TREE_STATIC (decl)
 	  && ret.value == decl
-	  && !warn_init_self)
+	  && !warning_enabled_at (DECL_SOURCE_LOCATION (decl), OPT_Winit_self))
 	suppress_warning (decl, OPT_Winit_self);
       if (TREE_CODE (ret.value) != STRING_CST
 	  && (TREE_CODE (ret.value) != COMPOUND_LITERAL_EXPR
--- gcc/cp/decl.cc.jj	2023-01-04 18:42:24.597997547 +0100
+++ gcc/cp/decl.cc	2023-01-12 15:26:01.257817526 +0100
@@ -8407,7 +8407,7 @@ cp_finish_decl (tree decl, tree init, bo
       if (!DECL_EXTERNAL (decl)
 	  && !TREE_STATIC (decl)
 	  && decl == tree_strip_any_location_wrapper (init)
-	  && !warn_init_self)
+	  && !warning_enabled_at (DECL_SOURCE_LOCATION (decl), OPT_Winit_self))
 	suppress_warning (decl, OPT_Winit_self);
     }
 
--- gcc/testsuite/c-c++-common/Winit-self3.c.jj	2023-01-12 15:49:56.759172518 +0100
+++ gcc/testsuite/c-c++-common/Winit-self3.c	2023-01-12 15:50:51.512384963 +0100
@@ -0,0 +1,36 @@
+/* PR c++/105593 */
+/* { dg-do compile } */
+/* { dg-options "-W -Wall" } */
+
+void bar (int);
+
+static inline int
+baz (void)
+{
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Winit-self"
+  int u = u;		/* { dg-bogus "'u' is used uninitialized" } */
+#pragma GCC diagnostic pop
+  return u;
+}
+
+void
+foo (void)
+{
+  int u = baz ();
+  bar (u);
+}
+
+static inline int
+qux (void)
+{
+  int u = u;		/* { dg-warning "'u' is used uninitialized" "" { target c++ } } */
+  return u;		/* { dg-message "'u' was declared here" "" { target c++ } .-1 } */
+}
+
+void
+corge (void)
+{
+  int u = qux ();
+  bar (u);
+}
--- gcc/testsuite/c-c++-common/Winit-self4.c.jj	2023-01-12 15:50:15.233906776 +0100
+++ gcc/testsuite/c-c++-common/Winit-self4.c	2023-01-12 15:50:42.445515372 +0100
@@ -0,0 +1,36 @@
+/* PR c++/105593 */
+/* { dg-do compile } */
+/* { dg-options "-W -Wall -Winit-self" } */
+
+void bar (int);
+
+static inline int
+baz (void)
+{
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Winit-self"
+  int u = u;		/* { dg-bogus "'u' is used uninitialized" } */
+#pragma GCC diagnostic pop
+  return u;
+}
+
+void
+foo (void)
+{
+  int u = baz ();
+  bar (u);
+}
+
+static inline int
+qux (void)
+{
+  int u = u;		/* { dg-warning "'u' is used uninitialized" } */
+  return u;		/* { dg-message "'u' was declared here" "" { target *-*-* } .-1 } */
+}
+
+void
+corge (void)
+{
+  int u = qux ();
+  bar (u);
+}
--- gcc/testsuite/c-c++-common/Winit-self5.c.jj	2023-01-12 15:51:01.045247847 +0100
+++ gcc/testsuite/c-c++-common/Winit-self5.c	2023-01-12 15:51:12.929076923 +0100
@@ -0,0 +1,36 @@
+/* PR c++/105593 */
+/* { dg-do compile } */
+/* { dg-options "-W -Wall -Wno-init-self" } */
+
+void bar (int);
+
+static inline int
+baz (void)
+{
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Winit-self"
+  int u = u;		/* { dg-bogus "'u' is used uninitialized" } */
+#pragma GCC diagnostic pop
+  return u;
+}
+
+void
+foo (void)
+{
+  int u = baz ();
+  bar (u);
+}
+
+static inline int
+qux (void)
+{
+  int u = u;		/* { dg-bogus "'u' is used uninitialized" } */
+  return u;
+}
+
+void
+corge (void)
+{
+  int u = qux ();
+  bar (u);
+}

	Jakub


             reply	other threads:[~2023-01-13  0:32 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-13  0:32 Jakub Jelinek [this message]
2023-01-13  2:49 ` Jason Merrill
2023-01-13 13:48   ` Marek Polacek

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=Y8CmpuU+Js7Tlsqz@tucnak \
    --to=jakub@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jason@redhat.com \
    --cc=joseph@codesourcery.com \
    --cc=polacek@redhat.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).