[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d8478697-e53b-456c-a285-7e1a1c297ed9@163.com>
Date: Tue, 8 Apr 2025 10:02:30 +0800
From: Hans Zhang <18255117159@....com>
To: Christophe JAILLET <christophe.jaillet@...adoo.fr>, lpieralisi@...nel.org
Cc: manivannan.sadhasivam@...aro.org, thierry.reding@...il.com, kw@...ux.com,
robh@...nel.org, bhelgaas@...gle.com, jonathanh@...dia.com,
linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-tegra@...r.kernel.org
Subject: Re: [v3] PCI: tegra194: Fix debugfs directory creation when
CONFIG_PCIEASPM is disabled
On 2025/4/8 03:43, Christophe JAILLET wrote:
> Hi,
>
> Nitpick: PATCH is missing within the [] in the subject.
>
Hi Christophe,
I will pay attention in the future.
>
> Le 07/04/2025 à 14:43, Hans Zhang a écrit :
>> Previously, the debugfs directory was unconditionally created in
>> tegra_pcie_config_rp() regardless of the CONFIG_PCIEASPM setting.
>> This led to unnecessary directory creation when ASPM support was
>> disabled.
>>
>> Move the debugfs directory creation into init_debugfs() which is
>> conditionally compiled based on CONFIG_PCIEASPM. This ensures:
>> - The directory is only created when ASPM-related debugfs entries are
>> needed.
>> - Proper error handling for directory creation failures.
>> - Avoids cluttering debugfs with empty directories when ASPM is disabled.
>>
>> Signed-off-by: Hans Zhang <18255117159@....com>
>> ---
>> Changes since v2:
>> https://lore.kernel.org/linux-pci/20250406134355.49036-1-18255117159@163.com/
>>
>> - Maintainer recommends ignoring the devm_kasprintf return value. The
>> module should still work correctly. So just a return;
>
> Note, that I'm not a maintainer ;-)
>
Maybe I made a mistake. ;-)
No.72 Christophe JAILLET <christophe ! jaillet () wanadoo ! fr>
2104(0.17%) @Hobbyists @French
Best regards,
Hans
> For what it worth:
>
> Reviewed-by: Christophe JAILLET <christophe.jaillet@...adoo.fr>
>
>
> CJ
>>
>> Changes since v1:
>> https://lore.kernel.org/linux-pci/20250405145459.26800-1-18255117159@163.com/
>>
>> - The first version was committed incorrectly because the judgment
>> parameter in "debugfs_remove_recursive" was not noticed.
>> ---
>> drivers/pci/controller/dwc/pcie-tegra194.c | 19 ++++++++++---------
>> 1 file changed, 10 insertions(+), 9 deletions(-)
>>
>> diff --git a/drivers/pci/controller/dwc/pcie-tegra194.c
>> b/drivers/pci/controller/dwc/pcie-tegra194.c
>> index 5103995cd6c7..bc419688527a 100644
>> --- a/drivers/pci/controller/dwc/pcie-tegra194.c
>> +++ b/drivers/pci/controller/dwc/pcie-tegra194.c
>> @@ -713,7 +713,16 @@ static void init_host_aspm(struct tegra_pcie_dw
>> *pcie)
>> static void init_debugfs(struct tegra_pcie_dw *pcie)
>> {
>> - debugfs_create_devm_seqfile(pcie->dev, "aspm_state_cnt",
>> pcie->debugfs,
>> + struct device *dev = pcie->dev;
>> + char *name;
>> +
>> + name = devm_kasprintf(dev, GFP_KERNEL, "%pOFP", dev->of_node);
>> + if (!name)
>> + return;
>> +
>> + pcie->debugfs = debugfs_create_dir(name, NULL);
>> +
>> + debugfs_create_devm_seqfile(dev, "aspm_state_cnt", pcie->debugfs,
>> aspm_state_cnt);
>> }
>> #else
>> @@ -1634,7 +1643,6 @@ static void tegra_pcie_deinit_controller(struct
>> tegra_pcie_dw *pcie)
>> static int tegra_pcie_config_rp(struct tegra_pcie_dw *pcie)
>> {
>> struct device *dev = pcie->dev;
>> - char *name;
>> int ret;
>> pm_runtime_enable(dev);
>> @@ -1664,13 +1672,6 @@ static int tegra_pcie_config_rp(struct
>> tegra_pcie_dw *pcie)
>> goto fail_host_init;
>> }
>> - name = devm_kasprintf(dev, GFP_KERNEL, "%pOFP", dev->of_node);
>> - if (!name) {
>> - ret = -ENOMEM;
>> - goto fail_host_init;
>> - }
>> -
>> - pcie->debugfs = debugfs_create_dir(name, NULL);
>> init_debugfs(pcie);
>> return ret;
>>
>> base-commit: a8662bcd2ff152bfbc751cab20f33053d74d0963
Powered by blists - more mailing lists