Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Simon Marchi <simon.marchi@polymtl.ca>
To: Tom Tromey <tom@tromey.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH] python: Fix memleak in do_start_initialization
Date: Mon, 27 Nov 2017 00:36:00 -0000	[thread overview]
Message-ID: <cfb8dba2-1df6-e07e-8be4-e8d6eb1bc61f@polymtl.ca> (raw)
In-Reply-To: <87po84vh77.fsf@tromey.com>

On 2017-11-26 04:54 PM, Tom Tromey wrote:
>>>>>> "Simon" == Simon Marchi <simon.marchi@polymtl.ca> writes:
> 
> Simon> There's already a comment in that regard:
> 
> Hah, I totally missed this.
> 
> Simon> When I read it, I thought it referred only to progname_copy, but it
> Simon> also applies to progname for Python 2.  I integrated your suggestion,
> Simon> wdyt?
> 
> Simon> -  char *progname;
> Simon> +  gdb::unique_xmalloc_ptr<char> progname;
> 
> I think the patch looks good, but would be even clearer if this
> declaration were moved into the #if, so that..
> 
> Simon> -  progname = concat (ldirname (python_libdir).c_str (), SLASH_STRING, "bin",
> Simon> -		     SLASH_STRING, "python", (char *) NULL);
> Simon> +  progname.reset (concat (ldirname (python_libdir).c_str (), SLASH_STRING,
> Simon> +			  "bin", SLASH_STRING, "python", (char *) NULL));
> 
> ...this reset could be replaced with plain initialization.
> 
> thanks,
> Tom
> 

Good idea, here's what I pushed.  Thanks for the comments!


From e8e7d10c39955e7ff99ff42f6f16d6befe2fa12e Mon Sep 17 00:00:00 2001
From: Simon Marchi <simon.marchi@polymtl.ca>
Date: Sun, 26 Nov 2017 19:32:47 -0500
Subject: [PATCH] python: Fix memleak in do_start_initialization

While playing with valgrind, I noticed that with Python 3, the progname
variable in do_start_initialization is not being freed (concat returns a
malloc'ed string).  This patch uses unique_xmalloc_ptr to manage it.
With Python 2, we pass progname it directly to Py_SetProgramName, so it
should not be freed.  We therefore release it before passing it.

gdb/ChangeLog:

	* python/python.c (do_start_initialization): Change progname
	type to gdb::unique_xmalloc_ptr.  Release the pointer when using
	Python 2.
---
 gdb/ChangeLog       |  6 ++++++
 gdb/python/python.c | 12 ++++++------
 2 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index a901409228..b3032ed7d8 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,9 @@
+2017-11-26  Simon Marchi  <simon.marchi@polymtl.ca>
+
+	* python/python.c (do_start_initialization): Change progname
+	type to gdb::unique_xmalloc_ptr.  Release the pointer when using
+	Python 2.
+
 2017-11-26  Tom Tromey  <tom@tromey.com>

 	* common/format.h: Add include guards.
diff --git a/gdb/python/python.c b/gdb/python/python.c
index 5f152611e8..9ed9b6b1ca 100644
--- a/gdb/python/python.c
+++ b/gdb/python/python.c
@@ -1658,7 +1658,6 @@ finalize_python (void *ignore)
 static bool
 do_start_initialization ()
 {
-  char *progname;
 #ifdef IS_PY3K
   int i;
   size_t progsize, count;
@@ -1672,19 +1671,20 @@ do_start_initialization ()
      /foo/bin/python
      /foo/lib/pythonX.Y/...
      This must be done before calling Py_Initialize.  */
-  progname = concat (ldirname (python_libdir).c_str (), SLASH_STRING, "bin",
-		     SLASH_STRING, "python", (char *) NULL);
+  gdb::unique_xmalloc_ptr<char> progname
+    (concat (ldirname (python_libdir).c_str (), SLASH_STRING, "bin",
+	      SLASH_STRING, "python", (char *) NULL));
 #ifdef IS_PY3K
   std::string oldloc = setlocale (LC_ALL, NULL);
   setlocale (LC_ALL, "");
-  progsize = strlen (progname);
+  progsize = strlen (progname.get ());
   progname_copy = (wchar_t *) PyMem_Malloc ((progsize + 1) * sizeof (wchar_t));
   if (!progname_copy)
     {
       fprintf (stderr, "out of memory\n");
       return false;
     }
-  count = mbstowcs (progname_copy, progname, progsize + 1);
+  count = mbstowcs (progname_copy, progname.get (), progsize + 1);
   if (count == (size_t) -1)
     {
       fprintf (stderr, "Could not convert python path to string\n");
@@ -1697,7 +1697,7 @@ do_start_initialization ()
      it is not freed after this call.  */
   Py_SetProgramName (progname_copy);
 #else
-  Py_SetProgramName (progname);
+  Py_SetProgramName (progname.release ());
 #endif
 #endif

-- 
2.15.0


      reply	other threads:[~2017-11-27  0:36 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-26  3:53 Simon Marchi
2017-11-26 19:00 ` Tom Tromey
2017-11-26 20:56   ` Simon Marchi
2017-11-26 21:55     ` Tom Tromey
2017-11-27  0:36       ` Simon Marchi [this message]

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=cfb8dba2-1df6-e07e-8be4-e8d6eb1bc61f@polymtl.ca \
    --to=simon.marchi@polymtl.ca \
    --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