[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMuHMdVTKRJJDqtpgExmEmwnYcpF8h2xTtbXwKLnVmVPz0bZqw@mail.gmail.com>
Date: Tue, 12 Mar 2019 14:47:01 +0100
From: Geert Uytterhoeven <geert@...ux-m68k.org>
To: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc: Miguel Ojeda Sandonis <miguel.ojeda.sandonis@...il.com>,
Willy Tarreau <willy@...roxy.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v1 1/4] auxdisplay: hd44780: Fix memory leak on ->remove()
Hi Andy,
On Tue, Mar 12, 2019 at 2:18 PM Andy Shevchenko
<andriy.shevchenko@...ux.intel.com> wrote:
> We have to free on ->remove() the allocated resources on ->probe().
>
> Fixes: d47d88361fee ("auxdisplay: Add HD44780 Character LCD support")
> Cc: Geert Uytterhoeven <geert@...ux-m68k.org>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Thanks, nice catch!
> --- a/drivers/auxdisplay/hd44780.c
> +++ b/drivers/auxdisplay/hd44780.c
> @@ -280,6 +280,8 @@ static int hd44780_remove(struct platform_device *pdev)
> struct charlcd *lcd = platform_get_drvdata(pdev);
>
> charlcd_unregister(lcd);
> +
> + kfree(lcd);
(wearing a newer (and hopefully wiser ;-) hat than when I wrote the code)
While this is correct for the current implementation of struct charlcd_priv,
this may be a bit fragile.
What about adding a charlcd_free() wrapper, which can do kfree(to_priv(lcd)),
and be used in the probe failure path, too?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@...ux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Powered by blists - more mailing lists