PCI: qcom: Fix compile error

  Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
to acquire bus lock") has added an argument to acquire bus lock
in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
without this argument, resulting in below build error.

drivers/pci/controller/dwc/pcie-qcom.c:973:9: error: too few arguments to function 'pci_enable_link_state'

This commit fixes the compilation error by passing the sem argument
to pci_enable_link_state in the qcom_pcie_enable_aspm function.

Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
---
 drivers/pci/controller/dwc/pcie-qcom.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
  

On Tue, Nov 28, 2023 at 09:50:26AM +0530, Vignesh Raman wrote:
> Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
> to acquire bus lock") has added an argument to acquire bus lock
> in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
> without this argument, resulting in below build error.
> 

Where do you see this error? That patch is not even merged. Looks like you are
sending the patch against some downstream tree.

- Mani

> drivers/pci/controller/dwc/pcie-qcom.c:973:9: error: too few arguments to function 'pci_enable_link_state'
> 
> This commit fixes the compilation error by passing the sem argument
> to pci_enable_link_state in the qcom_pcie_enable_aspm function.
> 
> Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
> ---
>  drivers/pci/controller/dwc/pcie-qcom.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c
> index 6902e97719d1..e846e3531d8e 100644
> --- a/drivers/pci/controller/dwc/pcie-qcom.c
> +++ b/drivers/pci/controller/dwc/pcie-qcom.c
> @@ -970,7 +970,7 @@ static int qcom_pcie_enable_aspm(struct pci_dev *pdev, void *userdata)
>  {
>  	/* Downstream devices need to be in D0 state before enabling PCI PM substates */
>  	pci_set_power_state(pdev, PCI_D0);
> -	pci_enable_link_state(pdev, PCIE_LINK_STATE_ALL);
> +	pci_enable_link_state(pdev, PCIE_LINK_STATE_ALL, false);
>  
>  	return 0;
>  }
> -- 
> 2.40.1
> 
>
  

Hi Mani,

On 28/11/23 10:44, Manivannan Sadhasivam wrote:
> On Tue, Nov 28, 2023 at 09:50:26AM +0530, Vignesh Raman wrote:
>> Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
>> to acquire bus lock") has added an argument to acquire bus lock
>> in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
>> without this argument, resulting in below build error.
>>
> 
> Where do you see this error? That patch is not even merged. Looks like you are
> sending the patch against some downstream tree.

I got this error with drm-tip - git://anongit.freedesktop.org/drm-tip

This commit is merged in drm-intel/topic/core-for-CI - 
https://cgit.freedesktop.org/drm-intel/log/?h=topic/core-for-CI

Regards,
Vignesh
  

On Tue, Nov 28, 2023 at 11:44:26AM +0530, Vignesh Raman wrote:
> Hi Mani,
> 
> On 28/11/23 10:44, Manivannan Sadhasivam wrote:
> > On Tue, Nov 28, 2023 at 09:50:26AM +0530, Vignesh Raman wrote:
> > > Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
> > > to acquire bus lock") has added an argument to acquire bus lock
> > > in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
> > > without this argument, resulting in below build error.
> > > 
> > 
> > Where do you see this error? That patch is not even merged. Looks like you are
> > sending the patch against some downstream tree.
> 
> I got this error with drm-tip - git://anongit.freedesktop.org/drm-tip
> 
> This commit is merged in drm-intel/topic/core-for-CI -
> https://cgit.freedesktop.org/drm-intel/log/?h=topic/core-for-CI
> 

Okay. Since this patch is just for CI, please do not CC linux-pci as it causes
confusion.

- Mani

> Regards,
> Vignesh
  

On 28/11/23 12:21, Manivannan Sadhasivam wrote:
> On Tue, Nov 28, 2023 at 11:44:26AM +0530, Vignesh Raman wrote:
>> Hi Mani,
>>
>> On 28/11/23 10:44, Manivannan Sadhasivam wrote:
>>> On Tue, Nov 28, 2023 at 09:50:26AM +0530, Vignesh Raman wrote:
>>>> Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
>>>> to acquire bus lock") has added an argument to acquire bus lock
>>>> in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
>>>> without this argument, resulting in below build error.
>>>>
>>>
>>> Where do you see this error? That patch is not even merged. Looks like you are
>>> sending the patch against some downstream tree.
>>
>> I got this error with drm-tip - git://anongit.freedesktop.org/drm-tip
>>
>> This commit is merged in drm-intel/topic/core-for-CI -
>> https://cgit.freedesktop.org/drm-intel/log/?h=topic/core-for-CI
>>
> 
> Okay. Since this patch is just for CI, please do not CC linux-pci as it causes
> confusion.

Sure, thank you.

Jani, is this fix required for topic/core-for-CI ?

Regards,
Vignesh
  

On Tue, 28 Nov 2023, Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> wrote:
> On Tue, Nov 28, 2023 at 11:44:26AM +0530, Vignesh Raman wrote:
>> Hi Mani,
>> 
>> On 28/11/23 10:44, Manivannan Sadhasivam wrote:
>> > On Tue, Nov 28, 2023 at 09:50:26AM +0530, Vignesh Raman wrote:
>> > > Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
>> > > to acquire bus lock") has added an argument to acquire bus lock
>> > > in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
>> > > without this argument, resulting in below build error.
>> > > 
>> > 
>> > Where do you see this error? That patch is not even merged. Looks like you are
>> > sending the patch against some downstream tree.
>> 
>> I got this error with drm-tip - git://anongit.freedesktop.org/drm-tip
>> 
>> This commit is merged in drm-intel/topic/core-for-CI -
>> https://cgit.freedesktop.org/drm-intel/log/?h=topic/core-for-CI
>> 
>
> Okay. Since this patch is just for CI, please do not CC linux-pci as it causes
> confusion.

Agreed. More on-topic for linux-pci is what happened with [1].

We've been running CI with that for months now to avoid lockdep splats,
and it's obviously in everyone's best interest to get a fix upstream.

David, Bjorn?


BR,
Jani.


[1] https://lore.kernel.org/all/20230321233849.3408339-1-david.e.box@linux.intel.com/




>
> - Mani
>
>> Regards,
>> Vignesh
  

On Tue, Nov 28, 2023 at 12:39:02PM +0200, Jani Nikula wrote:
> On Tue, 28 Nov 2023, Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> wrote:
> > On Tue, Nov 28, 2023 at 11:44:26AM +0530, Vignesh Raman wrote:
> >> Hi Mani,
> >> 
> >> On 28/11/23 10:44, Manivannan Sadhasivam wrote:
> >> > On Tue, Nov 28, 2023 at 09:50:26AM +0530, Vignesh Raman wrote:
> >> > > Commit a2458d8f618a ("PCI/ASPM: pci_enable_link_state: Add argument
> >> > > to acquire bus lock") has added an argument to acquire bus lock
> >> > > in pci_enable_link_state, but qcom_pcie_enable_aspm calls it
> >> > > without this argument, resulting in below build error.
> >> > > 
> >> > 
> >> > Where do you see this error? That patch is not even merged. Looks like you are
> >> > sending the patch against some downstream tree.
> >> 
> >> I got this error with drm-tip - git://anongit.freedesktop.org/drm-tip
> >> 
> >> This commit is merged in drm-intel/topic/core-for-CI -
> >> https://cgit.freedesktop.org/drm-intel/log/?h=topic/core-for-CI
> >> 
> >
> > Okay. Since this patch is just for CI, please do not CC linux-pci as it causes
> > confusion.
> 
> Agreed. More on-topic for linux-pci is what happened with [1].
> 
> We've been running CI with that for months now to avoid lockdep splats,
> and it's obviously in everyone's best interest to get a fix upstream.
> 

Yes, ofc. Right now we have 2 series/patches to fix the locking issue:

https://lore.kernel.org/all/20230321233849.3408339-1-david.e.box@linux.intel.com/
https://lore.kernel.org/linux-pci/20231128081512.19387-1-johan+linaro@kernel.org/

Bjorn has to choose one among them.

- Mani

> David, Bjorn?
> 
> 
> BR,
> Jani.
> 
> 
> [1] https://lore.kernel.org/all/20230321233849.3408339-1-david.e.box@linux.intel.com/
> 
> 
> 
> 
> >
> > - Mani
> >
> >> Regards,
> >> Vignesh
> 
> -- 
> Jani Nikula, Intel