[PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[Hao Ge]

Use devm_platform_ioremap_resource to take the place of of_iomap for
avoid that we don't called iounmap when return some error or remove
device.

Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock drivers and allow module build")
Signed-off-by: Hao Ge <gehao@kylinos.cn>
---
 drivers/clk/mediatek/clk-mt8173-apmixedsys.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/clk/mediatek/clk-mt8173-apmixedsys.c b/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
index 8c2aa8b0f39e..8a2a88c63d15 100644
--- a/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
+++ b/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
@@ -146,8 +146,8 @@ static int clk_mt8173_apmixed_probe(struct platform_device *pdev)
 	struct clk_hw *hw;
 	int r;
 
-	base = of_iomap(node, 0);
-	if (!base)
+	base = devm_of_iomap(&pdev->dev, node, 0, NULL);
+	if (IS_ERR(base))
 		return PTR_ERR(base);
 
 	clk_data = mtk_alloc_clk_data(CLK_APMIXED_NR_CLK);
-- 
2.25.1


No virus found
		Checked by Hillstone Network AntiVirus

Re: [PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[Stephen Boyd]

Quoting Hao Ge (2023-05-08 19:12:14)
> Use devm_platform_ioremap_resource to take the place of of_iomap for
> avoid that we don't called iounmap when return some error or remove
> device.
> 
> Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock drivers and allow module build")
> Signed-off-by: Hao Ge <gehao@kylinos.cn>
> ---

Is this v3?

Re: [PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[Hao Ge]

On Wed, 10 May 2023 14:18:08 -0700
Stephen Boyd <sboyd@kernel.org> wrote:

> Quoting Hao Ge (2023-05-08 19:12:14)
> > Use devm_platform_ioremap_resource to take the place of of_iomap for
> > avoid that we don't called iounmap when return some error or remove
> > device.
> > 
> > Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock
> > drivers and allow module build") Signed-off-by: Hao Ge
> > <gehao@kylinos.cn> ---  
> 
> Is this v3?
Hi Stephen

yes,Perhaps you are too busy,I have not received a response
in the future, so you are unsure about the status of V2.
So I adjusted the code and title make changes smaller and
sent V3 for it.It is precisely because of these that I did 
not continue the original mail thread.

Best regards
Hao

Re: [PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[Stephen Boyd]

Quoting Hao Ge (2023-05-10 18:26:54)
> On Wed, 10 May 2023 14:18:08 -0700
> Stephen Boyd <sboyd@kernel.org> wrote:
> 
> > Quoting Hao Ge (2023-05-08 19:12:14)
> > > Use devm_platform_ioremap_resource to take the place of of_iomap for
> > > avoid that we don't called iounmap when return some error or remove
> > > device.
> > > 
> > > Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock
> > > drivers and allow module build") Signed-off-by: Hao Ge
> > > <gehao@kylinos.cn> ---  
> > 
> > Is this v3?
> Hi Stephen
> 
> yes,Perhaps you are too busy,I have not received a response
> in the future, so you are unsure about the status of V2.
> So I adjusted the code and title make changes smaller and
> sent V3 for it.It is precisely because of these that I did 
> not continue the original mail thread.
> 

I'm waiting for the usual mediatek clk driver folks to review the patch.

Re: [PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[Chen-Yu Tsai]

On Tue, May 9, 2023 at 11:12 AM Hao Ge <gehao@kylinos.cn> wrote:
>
> Use devm_platform_ioremap_resource to take the place of of_iomap for
> avoid that we don't called iounmap when return some error or remove
> device.
>
> Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock drivers and allow module build")
> Signed-off-by: Hao Ge <gehao@kylinos.cn>
> ---
>  drivers/clk/mediatek/clk-mt8173-apmixedsys.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/clk/mediatek/clk-mt8173-apmixedsys.c b/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
> index 8c2aa8b0f39e..8a2a88c63d15 100644
> --- a/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
> +++ b/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
> @@ -146,8 +146,8 @@ static int clk_mt8173_apmixed_probe(struct platform_device *pdev)
>         struct clk_hw *hw;
>         int r;
>
> -       base = of_iomap(node, 0);
> -       if (!base)
> +       base = devm_of_iomap(&pdev->dev, node, 0, NULL);
> +       if (IS_ERR(base))

From lib/devres.c:

 * Please Note: This is not a one-to-one replacement for of_iomap() because the
 * of_iomap() function does not track whether the region is already mapped.  If
 * two drivers try to map the same memory, the of_iomap() function will succeed
 * but the devm_of_iomap() function will return -EBUSY.

The register range for apmixedsys given in mt8173.dtsi overlaps with
the HDMI phy. I'm concerned that would cause issues.

ChenYu

>                 return PTR_ERR(base);
>
>         clk_data = mtk_alloc_clk_data(CLK_APMIXED_NR_CLK);
> --
> 2.25.1
>
>
> No virus found
>                 Checked by Hillstone Network AntiVirus

Re: [PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[AngeloGioacchino Del Regno]

Il 12/05/23 02:39, Chen-Yu Tsai ha scritto:
> On Tue, May 9, 2023 at 11:12 AM Hao Ge <gehao@kylinos.cn> wrote:
>>
>> Use devm_platform_ioremap_resource to take the place of of_iomap for
>> avoid that we don't called iounmap when return some error or remove
>> device.
>>
>> Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock drivers and allow module build")
>> Signed-off-by: Hao Ge <gehao@kylinos.cn>
>> ---
>>   drivers/clk/mediatek/clk-mt8173-apmixedsys.c | 4 ++--
>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/clk/mediatek/clk-mt8173-apmixedsys.c b/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
>> index 8c2aa8b0f39e..8a2a88c63d15 100644
>> --- a/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
>> +++ b/drivers/clk/mediatek/clk-mt8173-apmixedsys.c
>> @@ -146,8 +146,8 @@ static int clk_mt8173_apmixed_probe(struct platform_device *pdev)
>>          struct clk_hw *hw;
>>          int r;
>>
>> -       base = of_iomap(node, 0);
>> -       if (!base)
>> +       base = devm_of_iomap(&pdev->dev, node, 0, NULL);
>> +       if (IS_ERR(base))
> 
>  From lib/devres.c:
> 
>   * Please Note: This is not a one-to-one replacement for of_iomap() because the
>   * of_iomap() function does not track whether the region is already mapped.  If
>   * two drivers try to map the same memory, the of_iomap() function will succeed
>   * but the devm_of_iomap() function will return -EBUSY.
> 
> The register range for apmixedsys given in mt8173.dtsi overlaps with
> the HDMI phy. I'm concerned that would cause issues.

I agree, that will cause issues.

Regards,
Angelo

Re: [PATCH] mediatek/clk-mt8173-apmixedsys: use devm_of_iomap to avoid resource leak in clk_mt8173_apmixed_probe

[Hao Ge]



> On May 11, 2023, at 09:27, Hao Ge <gehao618@163.com> wrote:
> 
> On Wed, 10 May 2023 14:18:08 -0700
> Stephen Boyd <sboyd@kernel.org> wrote:
> 
>> Quoting Hao Ge (2023-05-08 19:12:14)
>>> Use devm_platform_ioremap_resource to take the place of of_iomap for
>>> avoid that we don't called iounmap when return some error or remove
>>> device.
>>> Fixes: 4c02c9af3cb9 ("clk: mediatek: mt8173: Break down clock
>>> drivers and allow module build") Signed-off-by: Hao Ge
>>> <gehao@kylinos.cn> ---  
>> Is this v3?
> Hi Stephen
> 
> yes,Perhaps you are too busy,I have not received a response
> in the future, so you are unsure about the status of V2.
> So I adjusted the code and title make changes smaller and
> sent V3 for it.It is precisely because of these that I did 
> not continue the original mail thread.
> 
> Best regards
> Hao
Hi Stephen

I made a few typos in my previous email. To avoid misunderstandings, I corrected them and sent a new one.

Yes,Perhaps you are too busy,I have not received a response in the future, so I am unsure about the status of V2.So I adjusted the code and title make changes smaller and
sent V3 for it.It is precisely because of these that I did not continue the original mail thread.

Best regards
Hao