[PATCH 0/4] Followup to "Make clk_hw::init NULL after clk registration"

[PATCH 0/4] Followup to "Make clk_hw::init NULL after clk registration"

[Stephen Boyd]

I found some more cases where the init structure is referenced from
within the clk_hw struct after clk_registration is called. I suspect the
rtc driver fix is useful to avoid crashes on Allwinner devices, reported
by kernel-ci.

Cc: <bot@kernelci.org>                                                                                                                                      
Cc: <tomeu.vizoso@collabora.com>
Cc: <guillaume.tucker@collabora.com>
Cc: <mgalka@collabora.com>
Cc: <broonie@kernel.org>
Cc: <matthew.hart@linaro.org>
Cc: <khilman@baylibre.com>
Cc: <enric.balletbo@collabora.com>
Cc: Sylwester Nawrocki <s.nawrocki@samsung.com>      
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Chen-Yu Tsai <wens@csie.org>
Cc: Jun Nie <jun.nie@linaro.org>
Cc: Maxime Ripard <maxime.ripard@bootlin.com>
Cc: Shawn Guo <shawnguo@kernel.org>
Cc: Sugaya Taichi <sugaya.taichi@socionext.com>
Cc: Taniya Das <tdas@codeaurora.org>

Stephen Boyd (4):
  clk: milbeaut:  Don't reference clk_init_data after registration
  clk: zx296718: Don't reference clk_init_data after registration
  rtc: sun6i: Don't reference clk_init_data after registration
  clk: qcom: Remove error prints from DFS registration

 drivers/clk/clk-milbeaut.c     |   2 +-
 drivers/clk/qcom/clk-rcg2.c    |   8 +--
 drivers/clk/zte/clk-zx296718.c | 109 +++++++++++++++------------------
 drivers/rtc/rtc-sun6i.c        |   2 +-
 4 files changed, 52 insertions(+), 69 deletions(-)

-- 
Sent by a computer through tubes

[PATCH 3/4] rtc: sun6i: Don't reference clk_init_data after registration

[Stephen Boyd]

A future patch is going to change semantics of clk_register() so that
clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid
referencing this member here so that we don't run into NULL pointer
exceptions.

Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Maxime Ripard <maxime.ripard@bootlin.com>
Cc: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
---

Please ack so I can take this through clk tree.

 drivers/rtc/rtc-sun6i.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/rtc/rtc-sun6i.c b/drivers/rtc/rtc-sun6i.c
index c0e75c373605..d50ee023b559 100644
--- a/drivers/rtc/rtc-sun6i.c
+++ b/drivers/rtc/rtc-sun6i.c
@@ -279,7 +279,7 @@ static void __init sun6i_rtc_clk_init(struct device_node *node,
 
 	of_property_read_string_index(node, "clock-output-names", 1,
 				      &clkout_name);
-	rtc->ext_losc = clk_register_gate(NULL, clkout_name, rtc->hw.init->name,
+	rtc->ext_losc = clk_register_gate(NULL, clkout_name, init.name,
 					  0, rtc->base + SUN6I_LOSC_OUT_GATING,
 					  SUN6I_LOSC_OUT_GATING_EN_OFFSET, 0,
 					  &rtc->lock);
-- 
Sent by a computer through tubes

Re: [PATCH 3/4] rtc: sun6i: Don't reference clk_init_data after registration

[Chen-Yu Tsai]

On Fri, Aug 16, 2019 at 12:00 AM Stephen Boyd <sboyd@kernel.org> wrote:
>
> A future patch is going to change semantics of clk_register() so that
> clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid
> referencing this member here so that we don't run into NULL pointer
> exceptions.
>
> Cc: Alessandro Zummo <a.zummo@towertech.it>
> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
> Cc: Maxime Ripard <maxime.ripard@bootlin.com>
> Cc: Chen-Yu Tsai <wens@csie.org>
> Signed-off-by: Stephen Boyd <sboyd@kernel.org>

Acked-by: Chen-Yu Tsai <wens@csie.org>

Re: [PATCH 3/4] rtc: sun6i: Don't reference clk_init_data after registration

[Stephen Boyd]

Quoting Stephen Boyd (2019-08-15 09:00:19)
> A future patch is going to change semantics of clk_register() so that
> clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid
> referencing this member here so that we don't run into NULL pointer
> exceptions.
> 
> Cc: Alessandro Zummo <a.zummo@towertech.it>
> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
> Cc: Maxime Ripard <maxime.ripard@bootlin.com>
> Cc: Chen-Yu Tsai <wens@csie.org>
> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
> ---
> 

Looks like this fixed the sunxi boot crashes in -next. I'll add a tag
for kernelci credit.

Re: [PATCH 3/4] rtc: sun6i: Don't reference clk_init_data after registration

[Alexandre Belloni]

On 15/08/2019 09:00:19-0700, Stephen Boyd wrote:
> A future patch is going to change semantics of clk_register() so that
> clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid
> referencing this member here so that we don't run into NULL pointer
> exceptions.
> 
> Cc: Alessandro Zummo <a.zummo@towertech.it>
> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
> Cc: Maxime Ripard <maxime.ripard@bootlin.com>
> Cc: Chen-Yu Tsai <wens@csie.org>
> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

> ---
> 
> Please ack so I can take this through clk tree.
> 
>  drivers/rtc/rtc-sun6i.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/rtc/rtc-sun6i.c b/drivers/rtc/rtc-sun6i.c
> index c0e75c373605..d50ee023b559 100644
> --- a/drivers/rtc/rtc-sun6i.c
> +++ b/drivers/rtc/rtc-sun6i.c
> @@ -279,7 +279,7 @@ static void __init sun6i_rtc_clk_init(struct device_node *node,
>  
>  	of_property_read_string_index(node, "clock-output-names", 1,
>  				      &clkout_name);
> -	rtc->ext_losc = clk_register_gate(NULL, clkout_name, rtc->hw.init->name,
> +	rtc->ext_losc = clk_register_gate(NULL, clkout_name, init.name,
>  					  0, rtc->base + SUN6I_LOSC_OUT_GATING,
>  					  SUN6I_LOSC_OUT_GATING_EN_OFFSET, 0,
>  					  &rtc->lock);
> -- 
> Sent by a computer through tubes
> 

-- 
Alexandre Belloni, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

Re: [PATCH 3/4] rtc: sun6i: Don't reference clk_init_data after registration

[Stephen Boyd]

Quoting Stephen Boyd (2019-08-15 09:00:19)
> A future patch is going to change semantics of clk_register() so that
> clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid
> referencing this member here so that we don't run into NULL pointer
> exceptions.
> 
> Cc: Alessandro Zummo <a.zummo@towertech.it>
> Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
> Cc: Maxime Ripard <maxime.ripard@bootlin.com>
> Cc: Chen-Yu Tsai <wens@csie.org>
> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
> ---

Applied to clk-next

Re: [PATCH 0/4] Followup to "Make clk_hw::init NULL after clk registration"

[Guillaume Tucker]

On 15/08/2019 17:00, Stephen Boyd wrote:
> I found some more cases where the init structure is referenced from
> within the clk_hw struct after clk_registration is called. I suspect the
> rtc driver fix is useful to avoid crashes on Allwinner devices, reported
> by kernel-ci.

Please feel free to add this trailer where appropriate:

  Reported-by: "kernelci.org bot" <bot@kernelci.org>


Thanks,
Guillaume