All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vitaliy Ivanov <vitalivanov@gmail.com>
To: Richard Weinberger <richard@nod.at>
Cc: Jeff Dike <jdike@addtoit.com>,
	user-mode-linux-devel@lists.sourceforge.net,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] uml: drivers/net_user.c memory leak fix
Date: Fri, 08 Jul 2011 16:00:44 +0300	[thread overview]
Message-ID: <1310130044.26209.3.camel@vivanov> (raw)
In-Reply-To: <201107081244.53433.richard@nod.at>


> > > This control logic is a bit strange.
> > > When change_tramp() fails we should not printk() the output variable.
> > > 
> > > if (pid < 0){
> > >  free(output);
> > >  return;
> > > }
> > > 
> > > Would be much cleaner.
> > 
> > I just didn't want to clone this free-return stuff. So, what you
> > proposing is like this:
> > ------------
> > ...
> >         output = uml_kmalloc(output_len, UM_GFP_KERNEL);
> >         if (output == NULL)
> >                 printk(UM_KERN_ERR "change : failed to allocate output "
> >                        "buffer\n");
> > 
> >         pid = change_tramp(argv, output, output_len);
> >         if (pid < 0) {
> >                 free(output);               <---------- I'm not sure but 'output' can be NULL here.
> >                 return;
> >         }
> > 
> >         if (output != NULL) {
> >                 printk("%s", output);
> >                 kfree(output);
> >         }
> > }
> > ------------
> > 
> > I was trying to print 'output' only in case change_tramp is
> > successful. That's the old logic. And at the same time perform free
> > only in case output is not NULL.
> 
> Why?
> Freeing a NULL pointer is perfectly fine.

I should agree that something that you propose has better readability.

So, here is updated patch.

Thanks
---
>>From b0c5ca0336cc94b2fda251e0da7918394e59c5cd Mon Sep 17 00:00:00 2001
From: Vitaliy Ivanov <vitalivanov@gmail.com>
Date: Fri, 8 Jul 2011 14:54:29 +0300
Subject: [PATCH] uml: drivers/net_user.c memory leak fix

Perform memory cleanup on exit.
On receiving invalid 'pid' we still should clean 'output' variable.

Signed-off-by: Vitaliy Ivanov <vitalivanov@gmail.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 arch/um/drivers/net_user.c |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/arch/um/drivers/net_user.c b/arch/um/drivers/net_user.c
index 9415dd9..5201188 100644
--- a/arch/um/drivers/net_user.c
+++ b/arch/um/drivers/net_user.c
@@ -228,7 +228,10 @@ static void change(char *dev, char *what, unsigned char *addr,
 		       "buffer\n");
 
 	pid = change_tramp(argv, output, output_len);
-	if (pid < 0) return;
+	if (pid < 0) {
+		kfree(output);
+		return;
+	}
 
 	if (output != NULL) {
 		printk("%s", output);
-- 
1.7.0.4




WARNING: multiple messages have this Message-ID (diff)
From: Vitaliy Ivanov <vitalivanov@gmail.com>
To: Richard Weinberger <richard@nod.at>
Cc: Jeff Dike <jdike@addtoit.com>,
	user-mode-linux-devel@lists.sourceforge.net,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] uml: drivers/net_user.c memory leak fix
Date: Fri, 08 Jul 2011 16:00:44 +0300	[thread overview]
Message-ID: <1310130044.26209.3.camel@vivanov> (raw)
In-Reply-To: <201107081244.53433.richard@nod.at>


> > > This control logic is a bit strange.
> > > When change_tramp() fails we should not printk() the output variable.
> > > 
> > > if (pid < 0){
> > >  free(output);
> > >  return;
> > > }
> > > 
> > > Would be much cleaner.
> > 
> > I just didn't want to clone this free-return stuff. So, what you
> > proposing is like this:
> > ------------
> > ...
> >         output = uml_kmalloc(output_len, UM_GFP_KERNEL);
> >         if (output == NULL)
> >                 printk(UM_KERN_ERR "change : failed to allocate output "
> >                        "buffer\n");
> > 
> >         pid = change_tramp(argv, output, output_len);
> >         if (pid < 0) {
> >                 free(output);               <---------- I'm not sure but 'output' can be NULL here.
> >                 return;
> >         }
> > 
> >         if (output != NULL) {
> >                 printk("%s", output);
> >                 kfree(output);
> >         }
> > }
> > ------------
> > 
> > I was trying to print 'output' only in case change_tramp is
> > successful. That's the old logic. And at the same time perform free
> > only in case output is not NULL.
> 
> Why?
> Freeing a NULL pointer is perfectly fine.

I should agree that something that you propose has better readability.

So, here is updated patch.

Thanks
---
>From b0c5ca0336cc94b2fda251e0da7918394e59c5cd Mon Sep 17 00:00:00 2001
From: Vitaliy Ivanov <vitalivanov@gmail.com>
Date: Fri, 8 Jul 2011 14:54:29 +0300
Subject: [PATCH] uml: drivers/net_user.c memory leak fix

Perform memory cleanup on exit.
On receiving invalid 'pid' we still should clean 'output' variable.

Signed-off-by: Vitaliy Ivanov <vitalivanov@gmail.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 arch/um/drivers/net_user.c |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/arch/um/drivers/net_user.c b/arch/um/drivers/net_user.c
index 9415dd9..5201188 100644
--- a/arch/um/drivers/net_user.c
+++ b/arch/um/drivers/net_user.c
@@ -228,7 +228,10 @@ static void change(char *dev, char *what, unsigned char *addr,
 		       "buffer\n");
 
 	pid = change_tramp(argv, output, output_len);
-	if (pid < 0) return;
+	if (pid < 0) {
+		kfree(output);
+		return;
+	}
 
 	if (output != NULL) {
 		printk("%s", output);
-- 
1.7.0.4




  reply	other threads:[~2011-07-08 13:00 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-07-07 16:36 [uml-devel] [PATCH 1/3] uml: drivers/net_user.c memory leak fix Vitaliy Ivanov
2011-07-07 16:36 ` Vitaliy Ivanov
2011-07-07 17:27 ` Richard Weinberger
2011-07-08 10:30   ` Vitaliy Ivanov
2011-07-08 10:30     ` Vitaliy Ivanov
2011-07-08 10:44     ` [uml-devel] " Richard Weinberger
2011-07-08 10:44       ` Richard Weinberger
2011-07-08 13:00       ` Vitaliy Ivanov [this message]
2011-07-08 13:00         ` Vitaliy Ivanov
2011-07-08 15:42         ` Richard Weinberger

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=1310130044.26209.3.camel@vivanov \
    --to=vitalivanov@gmail.com \
    --cc=jdike@addtoit.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=richard@nod.at \
    --cc=user-mode-linux-devel@lists.sourceforge.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.