Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Tom Tromey <tom@tromey.com>
To: Simon Marchi <simon.marchi@polymtl.ca>
Cc: Tom Tromey <tom@tromey.com>,
	gdb-patches@sourceware.org,
	Joel Brobecker <brobecker@adacore.com>
Subject: Re: [PATCH 3/3] Cache the result of find_file_and_directory
Date: Sun, 05 Dec 2021 13:17:38 -0700	[thread overview]
Message-ID: <874k7mu7q5.fsf@tromey.com> (raw)
In-Reply-To: <adfcb528-41f9-3da0-2d19-9e4acfb0d4a4@polymtl.ca> (Simon Marchi's message of "Sat, 4 Dec 2021 22:47:24 -0500")

>> Sorry about that.  I'll fix it soon.  Or you can back it out if you
>> prefer.

Simon> I'm not personally affected so I do not mind, I can wait for the fix.

Here is what I am checking in.
I was able to reproduce the failure with valgrind, and confirmed that
this patch fixes it.

Tom

commit 33af066d07d495c81c7c102125aec8dbac62c27b
Author: Tom Tromey <tom@tromey.com>
Date:   Sun Dec 5 13:13:33 2021 -0700

    Preserve artificial CU name in process_psymtab_comp_unit_reader
    
    This fixes a use-after-free that Simon pointed out.
    process_psymtab_comp_unit_reader was allocating an artificial name for
    a CU, and then discarding it.  However, this name was preserved in the
    cached file_and_directory.  This patch arranges for the allocated name
    to be preserved there.

diff --git a/gdb/dwarf2/file-and-dir.h b/gdb/dwarf2/file-and-dir.h
index 1a9ccf35829..c56922ff90d 100644
--- a/gdb/dwarf2/file-and-dir.h
+++ b/gdb/dwarf2/file-and-dir.h
@@ -84,9 +84,10 @@ struct file_and_directory
   }
 
   /* Set the filename.  */
-  void set_name (const char *name)
+  void set_name (gdb::unique_xmalloc_ptr<char> name)
   {
-    m_name = name;
+    m_name_storage = std::move (name);
+    m_name = m_name_storage.get ();
   }
 
 private:
@@ -94,6 +95,9 @@ struct file_and_directory
   /* The filename.  */
   const char *m_name;
 
+  /* Storage for the filename, if needed.  */
+  gdb::unique_xmalloc_ptr<char> m_name_storage;
+
   /* The compilation directory.  NULL if not known.  If we needed to
      compute a new string, it will be stored in the comp_dir_storage
      member, and this will be NULL.  Otherwise, points directly to the
diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
index ff5758eb0a4..f2d7da7de52 100644
--- a/gdb/dwarf2/read.c
+++ b/gdb/dwarf2/read.c
@@ -6986,15 +6986,15 @@ process_psymtab_comp_unit_reader (const struct die_reader_specs *reader,
   prepare_one_comp_unit (cu, comp_unit_die, pretend_language);
 
   /* Allocate a new partial symbol table structure.  */
-  gdb::unique_xmalloc_ptr<char> debug_filename;
   static const char artificial[] = "<artificial>";
   file_and_directory &fnd = find_file_and_directory (comp_unit_die, cu);
   if (strcmp (fnd.get_name (), artificial) == 0)
     {
-      debug_filename.reset (concat (artificial, "@",
-				    sect_offset_str (per_cu->sect_off),
-				    (char *) NULL));
-      fnd.set_name (debug_filename.get ());
+      gdb::unique_xmalloc_ptr<char> debug_filename
+	(concat (artificial, "@",
+		 sect_offset_str (per_cu->sect_off),
+		 (char *) NULL));
+      fnd.set_name (std::move (debug_filename));
     }
 
   pst = create_partial_symtab (per_cu, per_objfile, fnd.get_name ());

  reply	other threads:[~2021-12-05 20:17 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-30  1:33 [PATCH 0/3] Refactor find_file_and_directory Tom Tromey
2021-11-30  1:33 ` [PATCH 1/3] Remove Irix case from find_file_and_directory Tom Tromey
2021-12-04 10:43   ` Joel Brobecker via Gdb-patches
2021-11-30  1:33 ` [PATCH 2/3] Move file_and_directory to new file and C++-ize Tom Tromey
2021-11-30 16:18   ` Lancelot SIX via Gdb-patches
2021-11-30 17:44     ` Tom Tromey
2021-12-04 10:38   ` Joel Brobecker via Gdb-patches
2021-12-04 18:22     ` Tom Tromey
2021-11-30  1:33 ` [PATCH 3/3] Cache the result of find_file_and_directory Tom Tromey
2021-12-04 10:42   ` Joel Brobecker via Gdb-patches
2021-12-04 18:22     ` Tom Tromey
2021-12-05  2:31       ` Simon Marchi via Gdb-patches
2021-12-05  3:46         ` Tom Tromey
2021-12-05  3:47           ` Simon Marchi via Gdb-patches
2021-12-05 20:17             ` Tom Tromey [this message]
2021-12-06  1:54               ` Simon Marchi via Gdb-patches
2021-12-07  4:30                 ` Tom Tromey

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=874k7mu7q5.fsf@tromey.com \
    --to=tom@tromey.com \
    --cc=brobecker@adacore.com \
    --cc=gdb-patches@sourceware.org \
    --cc=simon.marchi@polymtl.ca \
    /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