public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Andrew MacLeod <amacleod@redhat.com>
To: gcc-patches <gcc-patches@gcc.gnu.org>
Subject: [PATCH] Fix a few incorrect accesses.
Date: Fri, 2 Dec 2022 09:12:23 -0500	[thread overview]
Message-ID: <cbc8f41c-fe12-b7af-c906-e19f1ce1224e@redhat.com> (raw)


[-- Attachment #1.1: Type: text/plain, Size: 712 bytes --]

This consists of 3 changes which stronger type checking has indicated 
are non-compliant with the type field.

I doubt they are super important because there has not been a trap 
triggered by them, and they have been in the source base since sometime 
before 2017.  However, we should probably fix them.

I also notice that those are all uses of VOID_TYPE_P, which 
coincidentally does not check if its a type node being checked:

    /* Nonzero if this type is the (possibly qualified) void type.  */
    #define VOID_TYPE_P(NODE) (TREE_CODE (NODE) == VOID_TYPE)

So I guess it wouldn't trap anyway, just silently never trigger.

Bootstraps on x86_64-pc-linux-gnu with no regressions.  OK for trunk?

Andrew

[-- Attachment #2: 0001-Fix-a-few-incorrect-accesses.patch --]
[-- Type: text/x-patch, Size: 2319 bytes --]

From d1003e853d1813105eef6e441578e5bea9de8d03 Mon Sep 17 00:00:00 2001
From: Andrew MacLeod <amacleod@redhat.com>
Date: Tue, 29 Nov 2022 13:07:28 -0500
Subject: [PATCH] Fix a few incorrect accesses.

This consists of 3 changes which stronger type checking has indicated
are incorrect.

	gcc/
	* fold-const.cc (fold_unary_loc): Check TREE_TYPE of node.
	(tree_invalid_nonnegative_warnv_p): Likewise.

	gcc/c-family/
	* c-attribs.cc (handle_deprecated_attribute): Use type when
	using TYPE_NAME.
---
 gcc/c-family/c-attribs.cc | 2 +-
 gcc/fold-const.cc         | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/gcc/c-family/c-attribs.cc b/gcc/c-family/c-attribs.cc
index 07bca68e9b9..b36dd97802b 100644
--- a/gcc/c-family/c-attribs.cc
+++ b/gcc/c-family/c-attribs.cc
@@ -4240,7 +4240,7 @@ handle_deprecated_attribute (tree *node, tree name,
       if (type && TYPE_NAME (type))
 	{
 	  if (TREE_CODE (TYPE_NAME (type)) == IDENTIFIER_NODE)
-	    what = TYPE_NAME (*node);
+	    what = TYPE_NAME (type);
 	  else if (TREE_CODE (TYPE_NAME (type)) == TYPE_DECL
 		   && DECL_NAME (TYPE_NAME (type)))
 	    what = DECL_NAME (TYPE_NAME (type));
diff --git a/gcc/fold-const.cc b/gcc/fold-const.cc
index 114258fa182..e80be8049e1 100644
--- a/gcc/fold-const.cc
+++ b/gcc/fold-const.cc
@@ -9369,8 +9369,8 @@ fold_unary_loc (location_t loc, enum tree_code code, tree type, tree op0)
 	      && TREE_CODE (tem) == COND_EXPR
 	      && TREE_CODE (TREE_OPERAND (tem, 1)) == code
 	      && TREE_CODE (TREE_OPERAND (tem, 2)) == code
-	      && ! VOID_TYPE_P (TREE_OPERAND (tem, 1))
-	      && ! VOID_TYPE_P (TREE_OPERAND (tem, 2))
+	      && ! VOID_TYPE_P (TREE_TYPE (TREE_OPERAND (tem, 1)))
+	      && ! VOID_TYPE_P (TREE_TYPE (TREE_OPERAND (tem, 2)))
 	      && (TREE_TYPE (TREE_OPERAND (TREE_OPERAND (tem, 1), 0))
 		  == TREE_TYPE (TREE_OPERAND (TREE_OPERAND (tem, 2), 0)))
 	      && (! (INTEGRAL_TYPE_P (TREE_TYPE (tem))
@@ -15002,7 +15002,7 @@ tree_invalid_nonnegative_warnv_p (tree t, bool *strict_overflow_p, int depth)
 
 	/* If the initializer is non-void, then it's a normal expression
 	   that will be assigned to the slot.  */
-	if (!VOID_TYPE_P (t))
+	if (!VOID_TYPE_P (TREE_TYPE (t)))
 	  return RECURSE (t);
 
 	/* Otherwise, the initializer sets the slot in some way.  One common
-- 
2.38.1


             reply	other threads:[~2022-12-02 14:12 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-02 14:12 Andrew MacLeod [this message]
2022-12-02 14:52 ` Richard Biener
2022-12-02 17:02   ` Andrew MacLeod
2022-12-07 10:08 ` Thomas Schwinge
2022-12-07 14:58   ` Andrew MacLeod

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=cbc8f41c-fe12-b7af-c906-e19f1ce1224e@redhat.com \
    --to=amacleod@redhat.com \
    --cc=gcc-patches@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).