Because of the possilble failure of devm_kzalloc(), name might be NULL and
will cause null pointer derefrence later.
Therefore, it might be better to check it and directly return -ENOMEM.
Fixes: d39fb172760e ("clk: microchip: add PolarFire SoC fabric clock support")
Signed-off-by: Hui Tang <[email protected]>
---
drivers/clk/microchip/clk-mpfs-ccc.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/clk/microchip/clk-mpfs-ccc.c b/drivers/clk/microchip/clk-mpfs-ccc.c
index 7be028dced63..32aae880a14f 100644
--- a/drivers/clk/microchip/clk-mpfs-ccc.c
+++ b/drivers/clk/microchip/clk-mpfs-ccc.c
@@ -166,6 +166,9 @@ static int mpfs_ccc_register_outputs(struct device *dev, struct mpfs_ccc_out_hw_
struct mpfs_ccc_out_hw_clock *out_hw = &out_hws[i];
char *name = devm_kzalloc(dev, 23, GFP_KERNEL);
+ if (!name)
+ return -ENOMEM;
+
snprintf(name, 23, "%s_out%u", parent->name, i);
out_hw->divider.hw.init = CLK_HW_INIT_HW(name, &parent->hw, &clk_divider_ops, 0);
out_hw->divider.reg = data->pll_base[i / MPFS_CCC_OUTPUTS_PER_PLL] +
@@ -200,6 +203,9 @@ static int mpfs_ccc_register_plls(struct device *dev, struct mpfs_ccc_pll_hw_clo
struct mpfs_ccc_pll_hw_clock *pll_hw = &pll_hws[i];
char *name = devm_kzalloc(dev, 18, GFP_KERNEL);
+ if (!name)
+ return -ENOMEM;
+
pll_hw->base = data->pll_base[i];
snprintf(name, 18, "ccc%s_pll%u", strchrnul(dev->of_node->full_name, '@'), i);
pll_hw->name = (const char *)name;
--
2.17.1
On Sat, Nov 19, 2022 at 01:48:58PM +0800, Hui Tang wrote:
> Because of the possilble failure of devm_kzalloc(), name might be NULL and
> will cause null pointer derefrence later.
In theory, yeah?
(note to self, s/refrence/reference/, s/possilble/possible)
> Therefore, it might be better to check it and directly return -ENOMEM.
I agree with your use of might here. If the allocations do fail, we
likely aren't getting the system off the ground anyway - but there is
no harm in checking.
Reviewed-by: Conor Dooley <[email protected]>
@Claudiu, supposedly I can push to the at91 repo now so I will try to do
that.
Thanks,
Conor.
>
> Fixes: d39fb172760e ("clk: microchip: add PolarFire SoC fabric clock support")
> Signed-off-by: Hui Tang <[email protected]>
> ---
> drivers/clk/microchip/clk-mpfs-ccc.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/drivers/clk/microchip/clk-mpfs-ccc.c b/drivers/clk/microchip/clk-mpfs-ccc.c
> index 7be028dced63..32aae880a14f 100644
> --- a/drivers/clk/microchip/clk-mpfs-ccc.c
> +++ b/drivers/clk/microchip/clk-mpfs-ccc.c
> @@ -166,6 +166,9 @@ static int mpfs_ccc_register_outputs(struct device *dev, struct mpfs_ccc_out_hw_
> struct mpfs_ccc_out_hw_clock *out_hw = &out_hws[i];
> char *name = devm_kzalloc(dev, 23, GFP_KERNEL);
>
> + if (!name)
> + return -ENOMEM;
> +
> snprintf(name, 23, "%s_out%u", parent->name, i);
> out_hw->divider.hw.init = CLK_HW_INIT_HW(name, &parent->hw, &clk_divider_ops, 0);
> out_hw->divider.reg = data->pll_base[i / MPFS_CCC_OUTPUTS_PER_PLL] +
> @@ -200,6 +203,9 @@ static int mpfs_ccc_register_plls(struct device *dev, struct mpfs_ccc_pll_hw_clo
> struct mpfs_ccc_pll_hw_clock *pll_hw = &pll_hws[i];
> char *name = devm_kzalloc(dev, 18, GFP_KERNEL);
>
> + if (!name)
> + return -ENOMEM;
> +
> pll_hw->base = data->pll_base[i];
> snprintf(name, 18, "ccc%s_pll%u", strchrnul(dev->of_node->full_name, '@'), i);
> pll_hw->name = (const char *)name;
> --
> 2.17.1
>
On 2022/11/19 18:43, Conor Dooley wrote:
> On Sat, Nov 19, 2022 at 01:48:58PM +0800, Hui Tang wrote:
>> Because of the possilble failure of devm_kzalloc(), name might be NULL and
>> will cause null pointer derefrence later.
>
> In theory, yeah?
>
> (note to self, s/refrence/reference/, s/possilble/possible)
Sorry, I make spelling mistakes.
Thanks.
On 19.11.2022 12:43, Conor Dooley wrote:
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
>
> On Sat, Nov 19, 2022 at 01:48:58PM +0800, Hui Tang wrote:
>> Because of the possilble failure of devm_kzalloc(), name might be NULL and
>> will cause null pointer derefrence later.
>
> In theory, yeah?
>
> (note to self, s/refrence/reference/, s/possilble/possible)
Applied to clk-microchip-fixes with these adjustments, thanks!
>
>> Therefore, it might be better to check it and directly return -ENOMEM.
>
> I agree with your use of might here. If the allocations do fail, we
> likely aren't getting the system off the ground anyway - but there is
> no harm in checking.
>
> Reviewed-by: Conor Dooley <[email protected]>
>
> @Claudiu, supposedly I can push to the at91 repo now so I will try to do
> that.
>
> Thanks,
> Conor.
>
>>
>> Fixes: d39fb172760e ("clk: microchip: add PolarFire SoC fabric clock support")
>> Signed-off-by: Hui Tang <[email protected]>
>> ---
>> drivers/clk/microchip/clk-mpfs-ccc.c | 6 ++++++
>> 1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/clk/microchip/clk-mpfs-ccc.c b/drivers/clk/microchip/clk-mpfs-ccc.c
>> index 7be028dced63..32aae880a14f 100644
>> --- a/drivers/clk/microchip/clk-mpfs-ccc.c
>> +++ b/drivers/clk/microchip/clk-mpfs-ccc.c
>> @@ -166,6 +166,9 @@ static int mpfs_ccc_register_outputs(struct device *dev, struct mpfs_ccc_out_hw_
>> struct mpfs_ccc_out_hw_clock *out_hw = &out_hws[i];
>> char *name = devm_kzalloc(dev, 23, GFP_KERNEL);
>>
>> + if (!name)
>> + return -ENOMEM;
>> +
>> snprintf(name, 23, "%s_out%u", parent->name, i);
>> out_hw->divider.hw.init = CLK_HW_INIT_HW(name, &parent->hw, &clk_divider_ops, 0);
>> out_hw->divider.reg = data->pll_base[i / MPFS_CCC_OUTPUTS_PER_PLL] +
>> @@ -200,6 +203,9 @@ static int mpfs_ccc_register_plls(struct device *dev, struct mpfs_ccc_pll_hw_clo
>> struct mpfs_ccc_pll_hw_clock *pll_hw = &pll_hws[i];
>> char *name = devm_kzalloc(dev, 18, GFP_KERNEL);
>>
>> + if (!name)
>> + return -ENOMEM;
>> +
>> pll_hw->base = data->pll_base[i];
>> snprintf(name, 18, "ccc%s_pll%u", strchrnul(dev->of_node->full_name, '@'), i);
>> pll_hw->name = (const char *)name;
>> --
>> 2.17.1
>>