2013-09-14 01:27:17

by Rafael J. Wysocki

Subject: [PATCH] PCI / ACPI / PM: Clear pme_poll for devices in D3cold on wakeup

From: Rafael J. Wysocki <[email protected]>

Commit 448bd85 (PCI/PM: add PCIe runtime D3cold support) added a
piece of code to pci_acpi_wake_dev() causing that function to behave
in a special way for devices in D3cold (so that their configuration
registeres are not accessed before those devices are resumed).
However, it didn't take the clearing of the pme_poll flag into
account. That has to be done for all devices, even if they are in
D3cold, or pci_pme_list_scan() will not know that wakeup has been
signaled for the device and will poll its PME Status bit
unnecessarily.

Fix the problem by moving the clearing of the pme_poll flag in
pci_acpi_wake_dev() before the code introduced by commit 448bd85.

Reported-and-tested-by: David E. Box <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
Cc: 3.6+ <[email protected]> # 3.6+
---
drivers/pci/pci-acpi.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)

Index: linux-pm/drivers/pci/pci-acpi.c
===================================================================
--- linux-pm.orig/drivers/pci/pci-acpi.c
+++ linux-pm/drivers/pci/pci-acpi.c
@@ -47,6 +47,9 @@ static void pci_acpi_wake_dev(acpi_handl
if (event != ACPI_NOTIFY_DEVICE_WAKE || !pci_dev)
return;

+ if (pci_dev->pme_poll)
+ pci_dev->pme_poll = false;
+
if (pci_dev->current_state == PCI_D3cold) {
pci_wakeup_event(pci_dev);
pm_runtime_resume(&pci_dev->dev);
@@ -57,9 +60,6 @@ static void pci_acpi_wake_dev(acpi_handl
if (pci_dev->pme_support)
pci_check_pme_status(pci_dev);

- if (pci_dev->pme_poll)
- pci_dev->pme_poll = false;
-
pci_wakeup_event(pci_dev);
pm_runtime_resume(&pci_dev->dev);

2013-09-19 20:51:38

by Bjorn Helgaas

Subject: Re: [PATCH] PCI / ACPI / PM: Clear pme_poll for devices in D3cold on wakeup

On Fri, Sep 13, 2013 at 7:38 PM, Rafael J. Wysocki <[email protected]> wrote:
> From: Rafael J. Wysocki <[email protected]>
>
> Commit 448bd85 (PCI/PM: add PCIe runtime D3cold support) added a
> piece of code to pci_acpi_wake_dev() causing that function to behave
> in a special way for devices in D3cold (so that their configuration
> registeres are not accessed before those devices are resumed).

s/registeres/registers/

> However, it didn't take the clearing of the pme_poll flag into
> account. That has to be done for all devices, even if they are in
> D3cold, or pci_pme_list_scan() will not know that wakeup has been
> signaled for the device and will poll its PME Status bit
> unnecessarily.
>
> Fix the problem by moving the clearing of the pme_poll flag in
> pci_acpi_wake_dev() before the code introduced by commit 448bd85.
>
> Reported-and-tested-by: David E. Box <[email protected]>
> Signed-off-by: Rafael J. Wysocki <[email protected]>

Acked-by: Bjorn Helgaas <[email protected]>

> Cc: 3.6+ <[email protected]> # 3.6+
> ---
> drivers/pci/pci-acpi.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> Index: linux-pm/drivers/pci/pci-acpi.c
> ===================================================================
> --- linux-pm.orig/drivers/pci/pci-acpi.c
> +++ linux-pm/drivers/pci/pci-acpi.c
> @@ -47,6 +47,9 @@ static void pci_acpi_wake_dev(acpi_handl
> if (event != ACPI_NOTIFY_DEVICE_WAKE || !pci_dev)
> return;
>
> + if (pci_dev->pme_poll)
> + pci_dev->pme_poll = false;
> +
> if (pci_dev->current_state == PCI_D3cold) {
> pci_wakeup_event(pci_dev);
> pm_runtime_resume(&pci_dev->dev);
> @@ -57,9 +60,6 @@ static void pci_acpi_wake_dev(acpi_handl
> if (pci_dev->pme_support)
> pci_check_pme_status(pci_dev);
>
> - if (pci_dev->pme_poll)
> - pci_dev->pme_poll = false;
> -
> pci_wakeup_event(pci_dev);
> pm_runtime_resume(&pci_dev->dev);
>
>