PCI: Set up runtime PM on devices that don't support PCI PM
PATCH v2 1/1 PCI: Set up runtime PM on devices that don't support PCI PM
From: Mario Limonciello <[email protected]>
commit 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when initializing") intended to put PCI devices into D0, but in doing so unintentionally changed runtime PM initialization not to occur on devices that don't support PCI PM. This caused a regression in vfio-pci due to an imbalance with it's use.
Adjust the logic in pci_pm_init() so that even if PCI PM isn't supported runtime PM is still initialized.
---
v2:
* remove pointless return
--- drivers/pci/pci.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/pci/pci.c +3 -2
@@ -3222,14 +3222,14 @@ void pci_pm_init(struct pci_dev *dev)/* find PCI PM capability in list */ pm =pci_find_capability (dev,PCI_CAP_ID_PM );if (!pm) -return ; + goto poweron;/* Check device's ability to generate PME# */ pci_read_config_word (dev, pm +PCI_PM_PMC , &pmc);if ((pmc &PCI_PM_CAP_VER_MASK ) >3 ) {pci_err (dev,"unsupported PM cap regs version (%u)\n" , pmc &PCI_PM_CAP_VER_MASK ); -return ; + goto poweron; } dev->pm_cap = pm;
@@ -3274,6 +3274,7 @@ void pci_pm_init(struct pci_dev *dev)pci_read_config_word (dev,PCI_STATUS , &status);if (status &PCI_STATUS_IMM_READY ) dev->imm_ready =1 ; +poweron:pci_pm_power_up_and_verify_state (dev);pm_runtime_forbid (&dev->dev);pm_runtime_set_active (&dev->dev);
Signature · 1 line
2.43.0
Quoted 19 lines · Mario Limonciello
From: Mario Limonciello <[email protected]>
commit
4d4c10f763d7("PCI: Explicitly put devices into D0 when initializing") intended to put PCI devices into D0, but in doing so unintentionally changed runtime PM initialization not to occur on devices that don't support PCI PM. This caused a regression in vfio-pci due to an imbalance with it's use.Adjust the logic in
pci_pm_init()so that even if PCI PM isn't supported runtime PM is still initialized.Cc Alex Williamson <[email protected]>Reported-by Giovanni Cabiddu <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m7e8929d6421690dc8bd6dc639d86c2b4db27cbc4 open hereReported-by Nicolas Dichtel <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m40d277dcdb9be64a1609a82412d1aa906263e201 open hereTested-by Giovanni Cabiddu <[email protected]>Fixes 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when initializing")Signed-off-by Mario Limonciello <[email protected]>
Quoted 1 line
---
Looks fine to me.
Quoted 35 lines
v2:
* remove pointless return--- drivers/pci/pci.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/pci/pci.c +3
-2@@ -3222,14 +3222,14 @@ void pci_pm_init(struct pci_dev *dev)/* find PCI PM capability in list */ pm =pci_find_capability (dev,PCI_CAP_ID_PM );if (!pm) -return ; + goto poweron;/* Check device's ability to generate PME# */ pci_read_config_word (dev, pm +PCI_PM_PMC , &pmc);if ((pmc &PCI_PM_CAP_VER_MASK ) >3 ) {pci_err (dev,"unsupported PM cap regs version (%u)\n" , pmc &PCI_PM_CAP_VER_MASK ); -return ; + goto poweron; } dev->pm_cap = pm;@@ -3274,6 +3274,7 @@ void pci_pm_init(struct pci_dev *dev)pci_read_config_word (dev,PCI_STATUS , &status);if (status &PCI_STATUS_IMM_READY ) dev->imm_ready =1 ; +poweron:pci_pm_power_up_and_verify_state (dev);pm_runtime_forbid (&dev->dev);pm_runtime_set_active (&dev->dev);
Signature · 2 lines
Sathyanarayanan Kuppuswamy Linux Kernel Developer
Le 12/06/2025 à 01:31, Mario Limonciello a écrit :
Quoted 11 lines
From: Mario Limonciello <[email protected]>
commit
4d4c10f763d7("PCI: Explicitly put devices into D0 when initializing") intended to put PCI devices into D0, but in doing so unintentionally changed runtime PM initialization not to occur on devices that don't support PCI PM. This caused a regression in vfio-pci due to an imbalance with it's use.Adjust the logic in
pci_pm_init()so that even if PCI PM isn't supported runtime PM is still initialized.
Quoted 8 lines
Cc Alex Williamson <[email protected]>Reported-by Giovanni Cabiddu <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m7e8929d6421690dc8bd6dc639d86c2b4db27cbc4 open hereReported-by Nicolas Dichtel <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m40d277dcdb9be64a1609a82412d1aa906263e201 open hereTested-by Giovanni Cabiddu <[email protected]>Fixes 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when initializing")Signed-off-by Mario Limonciello <[email protected]>
Thanks for the quick fix!
Regards,
Nicolas
Quoted 12 lines · Mario Limonciello
From: Mario Limonciello <[email protected]>
commit
4d4c10f763d7("PCI: Explicitly put devices into D0 when initializing") intended to put PCI devices into D0, but in doing so unintentionally changed runtime PM initialization not to occur on devices that don't support PCI PM. This caused a regression in vfio-pci due to an imbalance with it's use.Adjust the logic in
pci_pm_init()so that even if PCI PM isn't supported runtime PM is still initialized.
Quoted 8 lines
Cc Alex Williamson <[email protected]>Reported-by Giovanni Cabiddu <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m7e8929d6421690dc8bd6dc639d86c2b4db27cbc4 open hereReported-by Nicolas Dichtel <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m40d277dcdb9be64a1609a82412d1aa906263e201 open hereTested-by Giovanni Cabiddu <[email protected]>Fixes 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when initializing")Signed-off-by Mario Limonciello <[email protected]>
Quoted 39 lines
---
v2:
* remove pointless return--- drivers/pci/pci.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/pci/pci.c +3
-3@@ -3222,14 +3222,14 @@ void pci_pm_init(struct pci_dev *dev)/* find PCI PM capability in list */ pm =pci_find_capability (dev,PCI_CAP_ID_PM );if (!pm) -return ; + goto poweron;/* Check device's ability to generate PME# */ pci_read_config_word (dev, pm +PCI_PM_PMC , &pmc);if ((pmc &PCI_PM_CAP_VER_MASK ) >3 ) {pci_err (dev,"unsupported PM cap regs version (%u)\n" , pmc &PCI_PM_CAP_VER_MASK ); -return ; + goto poweron; } dev->pm_cap = pm;@@ -3274,6 +3274,7 @@ void pci_pm_init(struct pci_dev *dev)pci_read_config_word (dev,PCI_STATUS , &status);if (status &PCI_STATUS_IMM_READY ) dev->imm_ready =1 ; +poweron:pci_pm_power_up_and_verify_state (dev);pm_runtime_forbid (&dev->dev);pm_runtime_set_active (&dev->dev); --2.43.0
Quoted 11 lines · Mario Limonciello
From: Mario Limonciello <[email protected]>
commit
4d4c10f763d7("PCI: Explicitly put devices into D0 when initializing") intended to put PCI devices into D0, but in doing so unintentionally changed runtime PM initialization not to occur on devices that don't support PCI PM. This caused a regression in vfio-pci due to an imbalance with it's use.Adjust the logic in
pci_pm_init()so that even if PCI PM isn't supported runtime PM is still initialized.
Quoted 8 lines
Cc Alex Williamson <[email protected]>Reported-by Giovanni Cabiddu <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m7e8929d6421690dc8bd6dc639d86c2b4db27cbc4 open hereReported-by Nicolas Dichtel <[email protected]>Closes https://lore.kernel.org/linux-pci/[email protected]/T/#m40d277dcdb9be64a1609a82412d1aa906263e201 open hereTested-by Giovanni Cabiddu <[email protected]>Fixes 4d4c10f763d7 ("PCI: Explicitly put devices into D0 when initializing")Signed-off-by Mario Limonciello <[email protected]>
4d4c10f763d7 appeared in v6.16-rc1, so I applied this to pci/for-linus for v6.16, thanks for the quick resolution!
Quoted 39 lines
---
v2:
* remove pointless return--- drivers/pci/pci.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/pci/pci.c +3
-2@@ -3222,14 +3222,14 @@ void pci_pm_init(struct pci_dev *dev)/* find PCI PM capability in list */ pm =pci_find_capability (dev,PCI_CAP_ID_PM );if (!pm) -return ; + goto poweron;/* Check device's ability to generate PME# */ pci_read_config_word (dev, pm +PCI_PM_PMC , &pmc);if ((pmc &PCI_PM_CAP_VER_MASK ) >3 ) {pci_err (dev,"unsupported PM cap regs version (%u)\n" , pmc &PCI_PM_CAP_VER_MASK ); -return ; + goto poweron; } dev->pm_cap = pm;@@ -3274,6 +3274,7 @@ void pci_pm_init(struct pci_dev *dev)pci_read_config_word (dev,PCI_STATUS , &status);if (status &PCI_STATUS_IMM_READY ) dev->imm_ready =1 ; +poweron:pci_pm_power_up_and_verify_state (dev);pm_runtime_forbid (&dev->dev);pm_runtime_set_active (&dev->dev);Signature · 1 line
2.43.0