Re: [PATCH 1/2] ASoC: mediatek: mt8188: fix use-after-free in driver remove path

["Trevor Wu (吳文良)" <Trevor.Wu@mediatek.com>]

On Wed, 2023-05-31 at 16:47 -0700, Doug Anderson wrote:
> 
> you have verified the sender or the content.
>  Hi,
> 
> On Tue, May 30, 2023 at 12:25 AM Trevor Wu <trevor.wu@mediatek.com>
> wrote:
> >
> > diff --git a/sound/soc/mediatek/mt8188/mt8188-audsys-clk.c
> b/sound/soc/mediatek/mt8188/mt8188-audsys-clk.c
> > index be1c53bf4729..05d6f9d78e10 100644
> > --- a/sound/soc/mediatek/mt8188/mt8188-audsys-clk.c
> > +++ b/sound/soc/mediatek/mt8188/mt8188-audsys-clk.c
> > @@ -138,6 +138,29 @@ static const struct afe_gate
> aud_clks[CLK_AUD_NR_CLK] = {
> >         GATE_AUD6(CLK_AUD_GASRC11, "aud_gasrc11", "top_asm_h", 11),
> >  };
> >
> > +static void mt8188_audsys_clk_unregister(void *data)
> > +{
> > +       struct mtk_base_afe *afe = (struct mtk_base_afe *)data;
> 
> The above cast is unnecessary since the compiler lets you assign from
> a "void *" to another pointer without a cast. Unnecessary casts are
> considered harmful because they suspend the compiler's ability to do
> type checking. Other than that, this looks good. Sorry for not
> noticing that the same problem affected more than just the driver I
> fixed previously.
> 
> Reviewed-by: Douglas Anderson <dianders@chromium.org>


Got it. I will remove the unnecessary cast in V2. Most importantly,
thank you for bringing this issue to our attention.


Thanks,
Trevor