PCI/pwrctrl: Skip creating platform device unless CONFIG_PCI_PWRCTL enabled
PATCH 1/1 PCI/pwrctrl: Skip creating platform device unless CONFIG_PCI_PWRCTL enabled
From: Bjorn Helgaas <[email protected]>
If devicetree describes power supplies related to a PCI device, pci_pwrctrl_create_device() previously created a pwrctrl platform device even if CONFIG_PCI_PWRCTL was not enabled.
When pci_pwrctrl_create_device() creates a pwrctrl device and returns it, pci_scan_device() doesn't enumerating the PCI device on the assumption that the pwrctrl core will rescan the bus after turning on the power. If CONFIG_PCI_PWRCTL is not enabled, the rescan never happens.
This breaks PCI enumeration on any system that describes power supplies in devicetree but does not use pwrctrl. Jim reported that some brcmstb platforms break this way.
If CONFIG_PCI_PWRCTL is not enabled, skip creating the pwrctrl platform device and scan the device normally.
---
This an alternate to
https://lore.kernel.org/r/[email protected]
It should accomplish the same thing but I think using #ifdef makes it a little more visible and easier to see that pci_pwrctrl_create_device() is only relevant when CONFIG_PCI_PWRCTL is enabled.
drivers/pci/probe.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/pci/probe.c +4 -0
@@ -2510,6 +2510,7 @@ EXPORT_SYMBOL(pci_bus_read_dev_vendor_id);static struct platform_device *pci_pwrctrl_create_device (struct pci_bus *bus,int devfn) { +#if defined(CONFIG_PCI_PWRCTL ) || defined(CONFIG_PCI_PWRCTL_MODULE )struct pci_host_bridge *host =pci_find_host_bridge (bus);struct platform_device *pdev;struct device_node *np;
@@ -2536,6 +2537,9 @@ static struct platform_device *pci_pwrctrl_create_device(struct pci_bus *bus, in }return pdev; +#else +return NULL ; +#endif }/*
Signature · 1 line
2.43.0
Quoted 18 lines · Bjorn Helgaas
drivers/pci/probe.c +4
-0@@ -2510,6 +2510,7 @@ EXPORT_SYMBOL(pci_bus_read_dev_vendor_id);static struct platform_device *pci_pwrctrl_create_device (struct pci_bus *bus,int devfn) { +#if defined(CONFIG_PCI_PWRCTL ) || defined(CONFIG_PCI_PWRCTL_MODULE )struct pci_host_bridge *host =pci_find_host_bridge (bus);struct platform_device *pdev;struct device_node *np;@@ -2536,6 +2537,9 @@ static struct platform_device *pci_pwrctrl_create_device(struct pci_bus *bus, in }return pdev; +#else +return NULL ; +#endif }
[...]
Quoted 6 lines
This an alternate to
https://lore.kernel.org/r/[email protected]It should accomplish the same thing but I think using #ifdef makes it a little more visible and easier to see that
pci_pwrctrl_create_device()is only relevant whenCONFIG_PCI_PWRCTLis enabled.
Just noting though that section 21 of Documentation/process/coding-style.rst discourages use of #ifdef and recommends IS_ENABLED() and inline stubs instead.
Thanks,
Lukas
Quoted 23 lines · Lukas Wunner
Quoted 18 lines · Bjorn Helgaas
drivers/pci/probe.c +4
-0@@ -2510,6 +2510,7 @@ EXPORT_SYMBOL(pci_bus_read_dev_vendor_id);static struct platform_device *pci_pwrctrl_create_device (struct pci_bus *bus,int devfn) { +#if defined(CONFIG_PCI_PWRCTL ) || defined(CONFIG_PCI_PWRCTL_MODULE )struct pci_host_bridge *host =pci_find_host_bridge (bus);struct platform_device *pdev;struct device_node *np;@@ -2536,6 +2537,9 @@ static struct platform_device *pci_pwrctrl_create_device(struct pci_bus *bus, in }return pdev; +#else +return NULL ; +#endif }[...]
Quoted 3 lines
This an alternate to
https://lore.kernel.org/r/[email protected]
Quoted 7 lines
Quoted 3 lines
It should accomplish the same thing but I think using #ifdef makes it a little more visible and easier to see that
pci_pwrctrl_create_device()is only relevant whenCONFIG_PCI_PWRCTLis enabled.Just noting though that section 21 of
Documentation/process/coding-style.rstdiscourages use of #ifdef and recommendsIS_ENABLED()and inline stubs instead.
True, although I kind of object to IS_ENABLED() in cases like this where what we really want is a no-op function, because IS_ENABLED() forces me to mentally execute the function to see what's going on.
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
But now since we're considering this for v6.15 in a couple days, we need a minimal fix. Mani's original patch is probably the best for that, and I'm fine with that, although I don't think we should throw brcmstb under the bus. The pci_pwrctrl_create_device() assumption that pwrctrl is enabled is just broken, and I don't think the fix is to improve pwrctrl and convert drivers.
Maybe we could use Mani's patch with something like my commit log (it's quite possible I don't understand the situation completely):
If devicetree describes power supplies related to a PCI device, we previously created a pwrctrl device even if CONFIG_PCI_PWRCTL was not enabled.
When pci_pwrctrl_create_device() creates and returns a pwrctrl device, pci_scan_device() doesn't enumerate the PCI device; it assumes the pwrctrl core will rescan the bus after turning on the power. If CONFIG_PCI_PWRCTL is not enabled, the rescan never happens.
This may break PCI enumeration on any system that describes power supplies in devicetree but does not use pwrctrl. Jim reported that some brcmstb platforms break this way.
If CONFIG_PCI_PWRCTL is not enabled, skip creating the pwrctrl platform device and scan the device normally.
Looking at pci_pwrctrl_create_device() raised more questions for me:
- If pwrctrl is enabled, devicetree describes a regulator, and the power is already on, do we really want to defer enumeration?
- If pwrctrl *not* enabled, devicetree describes a regulator, and the power is off, is there anything in dmesg that gives us a clue about why we don't enumerate the device? Could/should we emit a message as a hint?
- Creating a pwrctrl device, returning a pointer to it, and then throwing that pointer away seems ... weird? Makes me wonder about the lifetime management of that device.
Bjorn
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
pci_pwrctrl_create_device() is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.
Thanks,
Lukas
Quoted 10 lines · Lukas Wunner
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
pci_pwrctrl_create_device()is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.
+1
- Mani
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Quoted 6 lines · Manivannan Sadhasivam
Quoted 5 lines · Lukas Wunner
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
Quoted 7 lines
Quoted 5 lines
pci_pwrctrl_create_device()is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.+1
I dropped the ball here and didn't get any fix for this in v6.15.
Why do we need pci_pwrctrl_create_device() in drivers/pci/probe.c? The obvious thing would have been to put the implementation in drivers/pci/pwrctrl with a stub in drivers/pci/pci.h, so I assume there's some reason we can't do that?
Quoted 14 lines · Bjorn Helgaas
Quoted 12 lines · Manivannan Sadhasivam
Quoted 10 lines · Lukas Wunner
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
pci_pwrctrl_create_device()is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.
Quoted 3 lines
Quoted 1 line
+1
I dropped the ball here and didn't get any fix for this in v6.15.
:-(
Quoted 6 lines
Why do we need
pci_pwrctrl_create_device()indrivers/pci/probe.c? The obvious thing would have been to put the implementation indrivers/pci/pwrctrlwith a stub indrivers/pci/pci.h, so I assume there's some reason we can't do that?
I was wondering if we could confine PWRCTL/_SLOT to work on a per PCIe controller basis. For example, if we allow the port DT node to have boolean "pwrctrl;" property, it would direct the PWRCTL code to operate on the regulators within that node. This would allow CONFIG_PWRCTL/_SLOT and the pcie-brcmstb.c way of controlling regulators to happily coexist.
One could argue that "pwrctrl;" does not describe the HW as a DT property should, but I think it should be considered nevertheless.
Regards,
Jim Quinlan
Broadcom STB
Quoted 35 lines · Jim Quinlan
Quoted 14 lines · Bjorn Helgaas
Quoted 12 lines · Manivannan Sadhasivam
Quoted 10 lines · Lukas Wunner
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
pci_pwrctrl_create_device()is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.Quoted 3 lines
Quoted 1 line
+1
I dropped the ball here and didn't get any fix for this in v6.15.
:-(
Quoted 6 lines
Why do we need
pci_pwrctrl_create_device()indrivers/pci/probe.c? The obvious thing would have been to put the implementation indrivers/pci/pwrctrlwith a stub indrivers/pci/pci.h, so I assume there's some reason we can't do that?I was wondering if we could confine PWRCTL/_SLOT to work on a per PCIe controller basis. For example, if we allow the port DT node to have boolean "pwrctrl;" property, it would direct the PWRCTL code to operate on the regulators within that node. This would allow CONFIG_PWRCTL/_SLOT and the
pcie-brcmstb.cway of controlling regulators to happily coexist.
Quoted 3 lines
One could argue that "pwrctrl;" does not describe the HW as a DT property should, but I think it should be considered nevertheless.
No. There were a lot of attempts previously before the advent of pwrctrl framework to add a DT property for powering the devices, but DT maintainer clearly said that there should be no DT property for this purpose. So we came up with this approach.
- Mani
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Quoted 13 lines · Bjorn Helgaas
Quoted 12 lines · Manivannan Sadhasivam
Quoted 10 lines · Lukas Wunner
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
pci_pwrctrl_create_device()is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.
Quoted 4 lines
Quoted 1 line
+1
I dropped the ball here and didn't get any fix for this in v6.15.
That's unfortunate. Let's try to get it backported through stable list.
Quoted 4 lines
Why do we need
pci_pwrctrl_create_device()indrivers/pci/probe.c? The obvious thing would have been to put the implementation indrivers/pci/pwrctrlwith a stub indrivers/pci/pci.h, so I assume there's some reason we can't do that?
I assumed initially that pwrctrl dependency would always be there with PCI core. Since now it is clear that we need to build the function conditionally, we could move it to drivers/pci/pwrctrl/core.c with stub in drivers/pci/pci.h.
I will submit a patch for that.
- Mani
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Quoted 13 lines · Bjorn Helgaas' via BCM-KERNEL-FEEDBACK-LIST,PDL
Quoted 12 lines · Manivannan Sadhasivam
Quoted 10 lines · Lukas Wunner
Quoted 3 lines · Bjorn Helgaas
What I would prefer is something like the first paragraph in that
section: the #ifdef in a header file that declares the function and defines a no-op stub, with the implementation in some pwrctrl file.
pci_pwrctrl_create_device()is static, but it is possible to #ifdef the whole function in the .c file and provide the stub in an #else branch. That's easier to follow than #ifdef'ing portions of the function.
Quoted 3 lines
Quoted 1 line
+1
I dropped the ball here and didn't get any fix for this in v6.15.
Any chance we could target a fix to be back ported into 6.15 stable? Do you need people to help you test it?
Signature · 1 line
Florian