[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFBinCB-WZmnMtvG1P=sWASQZCg9F9d-Hg0OzVNUsEutUFX0Cg@mail.gmail.com>
Date: Thu, 16 Jun 2022 22:17:19 +0200
From: Martin Blumenstingl <martin.blumenstingl@...glemail.com>
To: Liang He <windhl@....com>
Cc: Neil Armstrong <narmstrong@...libre.com>, khilman@...libre.com,
jbrunet@...libre.com, linux-arm-kernel@...ts.infradead.org,
linux-amlogic@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] soc: amlogic: Fix refcount leak in meson-secure-pwrc.c
Hello,
On Thu, Jun 16, 2022 at 4:50 PM Liang He <windhl@....com> wrote:
>
> In meson_secure_pwrc_probe(), there is a refcount leak in one fail
> path.
>
> Signed-off-by: Liang He <windhl@....com>
Acked-by: Martin Blumenstingl <martin.blumenstingl@...glemail.com>
[...]
> pwrc = devm_kzalloc(&pdev->dev, sizeof(*pwrc), GFP_KERNEL);
> - if (!pwrc)
> + if (!pwrc) {
> + of_node_put(sm_np);
> return -ENOMEM;
> + }
Another approach would be to just move devm_kzalloc() a few lines
further up (above of_find_compatible_node()).
That would catch similar issues in future when someone wants to add
more code right after devm_kzalloc(). That said, I don't think that
this is a likely scenario so the patch is fine for me as-is.
Thanks a lot for submitting this!
Best regards,
Martin
Powered by blists - more mailing lists