public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Sergio Durigan Junior <sergiodj@redhat.com>
To: GDB Patches <gdb-patches@sourceware.org>
Cc: Tom Tromey <tom@tromey.com>,	Sergio Durigan Junior <sergiodj@redhat.com>
Subject: [PATCH v2] Guard against 'current_directory == NULL' on gdb_abspath (PR gdb/23613)
Date: Thu, 21 Nov 2019 19:37:00 -0000	[thread overview]
Message-ID: <20191121193729.26315-1-sergiodj@redhat.com> (raw)
In-Reply-To: <20190710234615.14800-1-sergiodj@redhat.com>

Changes from v1:

- Adjust all spots where 'current_directory' was being used with
  'concat' to determine the path of a file.

- Expand 'gdb_abspath' comment and mention what happens when
  'current_directory' is NULL.


Ref.: https://bugzilla.redhat.com/show_bug.cgi?id=1728147
Ref.: https://sourceware.org/bugzilla/show_bug.cgi?id=23613

Hi,

This bug has been reported against Fedora GDB, but there's also an
upstream bug.  The problem reported is that GDB segfaults when the
working directory is deleted.  It's pretty use to reproduce it:

  mkdir bla
  cd bla
  rmdir ../bla
  gdb echo

Debugging the problem is a bit tricky, because, since the current
directory doesn't exist anymore, a corefile cannot be saved there.
After a few attempts, I came up with the following:

  gdb -ex 'shell mkdir bla' -ex 'cd bla' -ex 'shell rmdir ../bla' -ex 'r echo' ./gdb/gdb

This assumes that you're inside a build directory which contains
./gdb/gdb, of course.

After investigating it, I found that the problem happens at
gdb_abspath, where we're dereferencing 'current_directory' without
checking if it's NULL:

    ...
    (concat (current_directory,
	     IS_DIR_SEPARATOR (current_directory[strlen (current_directory) - 1])
	     ? "" : SLASH_STRING,
    ...

So I fixed the problem with the patch below.  The idea is that, if
'current_directory' is NULL, then the final string returned should be
just the "path".

After fixing the bug, I found a similar one reported against our
bugzilla: PR gdb/23613.  The problem is the same, but the reproducer
is a bit different.

I really tried writing a testcase for this, but unfortunately it's
apparently not possible to start GDB inside a non-existent directory
with DejaGNU.

I regression tested this patch on the BuildBot, and no regressions
were found.

gdb/ChangeLog:
2019-11-21  Sergio Durigan Junior  <sergiodj@redhat.com>

	https://bugzilla.redhat.com/show_bug.cgi?id=1728147
	PR gdb/23613
	* bsd-kvm.c (bsd_kvm_target_open): Use 'gdb_abspath'.
	* corelow.c: Include 'gdbsupport/pathstuff.h'.
	(core_target_open): Use 'gdb_abspath'.
	* gdbsupport/pathstuff.c (gdb_abspath): Guard against
	'current_directory == NULL' case.
	* gdbsupport/pathstuff.h (gdb_abspath): Expand comment and
	explain what happens when 'current_directory' is NULL.
	* go32-nat.c (go32_nat_target::wait): Check if
	'current_directory' is NULL before call to 'chdir'.
	* source.c (add_path): Use 'gdb_abspath'.
	* top.c: Include 'gdbsupport/pathstuff.h'.
	(init_history): Use 'gdb_abspath'.
	(set_history_filename): Likewise.
	* tracefile-tfile.c: Include 'gdbsupport/pathstuff.h'.
	(tfile_target_open): Use 'gdb_abspath'.

Change-Id: Ibb0932fa25bc5c2d3ae4a7f64bd7f32885ca403b
---
 gdb/bsd-kvm.c              |  7 +++----
 gdb/corelow.c              |  4 ++--
 gdb/gdbsupport/pathstuff.c |  2 +-
 gdb/gdbsupport/pathstuff.h |  5 ++++-
 gdb/go32-nat.c             |  3 ++-
 gdb/source.c               |  3 +--
 gdb/top.c                  | 18 ++++++++++++------
 gdb/tracefile-tfile.c      |  4 ++--
 8 files changed, 27 insertions(+), 19 deletions(-)

diff --git a/gdb/bsd-kvm.c b/gdb/bsd-kvm.c
index 21f978728d..895686ef62 100644
--- a/gdb/bsd-kvm.c
+++ b/gdb/bsd-kvm.c
@@ -114,14 +114,13 @@ bsd_kvm_target_open (const char *arg, int from_tty)
 
   if (arg)
     {
-      char *temp;
-
       filename = tilde_expand (arg);
       if (filename[0] != '/')
 	{
-	  temp = concat (current_directory, "/", filename, (char *)NULL);
+	  gdb::unique_xmalloc_ptr<char> temp (gdb_abspath (filename));
+
 	  xfree (filename);
-	  filename = temp;
+	  filename = temp.release ();
 	}
     }
 
diff --git a/gdb/corelow.c b/gdb/corelow.c
index fa1661ee61..c656306fe5 100644
--- a/gdb/corelow.c
+++ b/gdb/corelow.c
@@ -43,6 +43,7 @@
 #include "gdb_bfd.h"
 #include "completer.h"
 #include "gdbsupport/filestuff.h"
+#include "gdbsupport/pathstuff.h"
 
 #ifndef O_LARGEFILE
 #define O_LARGEFILE 0
@@ -373,8 +374,7 @@ core_target_open (const char *arg, int from_tty)
 
   gdb::unique_xmalloc_ptr<char> filename (tilde_expand (arg));
   if (!IS_ABSOLUTE_PATH (filename.get ()))
-    filename.reset (concat (current_directory, "/",
-			    filename.get (), (char *) NULL));
+    filename = gdb_abspath (filename.get ());
 
   flags = O_BINARY | O_LARGEFILE;
   if (write_files)
diff --git a/gdb/gdbsupport/pathstuff.c b/gdb/gdbsupport/pathstuff.c
index fafecd543d..aa51e8f36e 100644
--- a/gdb/gdbsupport/pathstuff.c
+++ b/gdb/gdbsupport/pathstuff.c
@@ -134,7 +134,7 @@ gdb_abspath (const char *path)
   if (path[0] == '~')
     return gdb_tilde_expand_up (path);
 
-  if (IS_ABSOLUTE_PATH (path))
+  if (IS_ABSOLUTE_PATH (path) || current_directory == NULL)
     return make_unique_xstrdup (path);
 
   /* Beware the // my son, the Emacs barfs, the botch that catch...  */
diff --git a/gdb/gdbsupport/pathstuff.h b/gdb/gdbsupport/pathstuff.h
index 7d7c8cd028..e84685b654 100644
--- a/gdb/gdbsupport/pathstuff.h
+++ b/gdb/gdbsupport/pathstuff.h
@@ -44,7 +44,10 @@ extern gdb::unique_xmalloc_ptr<char>
 
    Contrary to "gdb_realpath", this function uses CURRENT_DIRECTORY
    for the path expansion.  This may lead to scenarios the current
-   working directory (CWD) is different than CURRENT_DIRECTORY.  */
+   working directory (CWD) is different than CURRENT_DIRECTORY.
+
+   If CURRENT_DIRECTORY is NULL, this function returns a copy of
+   PATH.  */
 
 extern gdb::unique_xmalloc_ptr<char> gdb_abspath (const char *path);
 
diff --git a/gdb/go32-nat.c b/gdb/go32-nat.c
index 8c71bd7650..94972a2f5c 100644
--- a/gdb/go32-nat.c
+++ b/gdb/go32-nat.c
@@ -507,7 +507,8 @@ go32_nat_target::wait (ptid_t ptid, struct target_waitstatus *status,
     }
 
   getcwd (child_cwd, sizeof (child_cwd)); /* in case it has changed */
-  chdir (current_directory);
+  if (current_directory != NULL)
+    chdir (current_directory);
 
   if (a_tss.tss_irqn == 0x21)
     {
diff --git a/gdb/source.c b/gdb/source.c
index d9cd5f32ca..b456a517e3 100644
--- a/gdb/source.c
+++ b/gdb/source.c
@@ -540,8 +540,7 @@ add_path (const char *dirname, char **which_path, int parse_separators)
 	new_name_holder.reset (concat (name, ".", (char *) NULL));
 #endif
       else if (!IS_ABSOLUTE_PATH (name) && name[0] != '$')
-	new_name_holder.reset (concat (current_directory, SLASH_STRING, name,
-				       (char *) NULL));
+	new_name_holder = gdb_abspath (name);
       else
 	new_name_holder.reset (savestring (name, p - name));
       name = new_name_holder.get ();
diff --git a/gdb/top.c b/gdb/top.c
index 2953eac819..0893fe057f 100644
--- a/gdb/top.c
+++ b/gdb/top.c
@@ -54,6 +54,7 @@
 #include "gdb_select.h"
 #include "gdbsupport/scope-exit.h"
 #include "gdbarch.h"
+#include "gdbsupport/pathstuff.h"
 
 /* readline include files.  */
 #include "readline/readline.h"
@@ -1982,12 +1983,13 @@ init_history (void)
          that was read.  */
 #ifdef __MSDOS__
       /* No leading dots in file names are allowed on MSDOS.  */
-      history_filename = concat (current_directory, "/_gdb_history",
-				 (char *)NULL);
+      const char *fname = "_gdb_history";
 #else
-      history_filename = concat (current_directory, "/.gdb_history",
-				 (char *)NULL);
+      const char *fname = ".gdb_history";
 #endif
+
+      gdb::unique_xmalloc_ptr<char> temp (gdb_abspath (fname));
+      history_filename = temp.release ();
     }
   read_history (history_filename);
 }
@@ -2065,8 +2067,12 @@ set_history_filename (const char *args,
      directories the file written will be the same as the one
      that was read.  */
   if (!IS_ABSOLUTE_PATH (history_filename))
-    history_filename = reconcat (history_filename, current_directory, "/", 
-				 history_filename, (char *) NULL);
+    {
+      gdb::unique_xmalloc_ptr<char> temp (gdb_abspath (history_filename));
+
+      xfree (history_filename);
+      history_filename = temp.release ();
+    }
 }
 
 static void
diff --git a/gdb/tracefile-tfile.c b/gdb/tracefile-tfile.c
index 5c3837ec5b..0953b03781 100644
--- a/gdb/tracefile-tfile.c
+++ b/gdb/tracefile-tfile.c
@@ -32,6 +32,7 @@
 #include "xml-tdesc.h"
 #include "target-descriptions.h"
 #include "gdbsupport/buffer.h"
+#include "gdbsupport/pathstuff.h"
 #include <algorithm>
 
 #ifndef O_LARGEFILE
@@ -470,8 +471,7 @@ tfile_target_open (const char *arg, int from_tty)
 
   gdb::unique_xmalloc_ptr<char> filename (tilde_expand (arg));
   if (!IS_ABSOLUTE_PATH (filename.get ()))
-    filename.reset (concat (current_directory, "/", filename.get (),
-			    (char *) NULL));
+    filename = gdb_abspath (filename.get ());
 
   flags = O_BINARY | O_LARGEFILE;
   flags |= O_RDONLY;
-- 
2.21.0

  parent reply	other threads:[~2019-11-21 19:37 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-10 23:46 [PATCH] " Sergio Durigan Junior
2019-07-11 16:05 ` Tom Tromey
2019-07-12  1:21   ` Sergio Durigan Junior
2019-11-21 19:37 ` Sergio Durigan Junior [this message]
2019-12-10 16:21   ` [PATCH v2] " Sergio Durigan Junior
2019-12-13 22:23   ` Tom Tromey
2019-12-14  4:50     ` Sergio Durigan Junior

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=20191121193729.26315-1-sergiodj@redhat.com \
    --to=sergiodj@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=tom@tromey.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).