[PATCH bugfix 3.9] PCI: Don't try to disable Bus Master on disconnected PCI devices

[PATCH bugfix 3.9] PCI: Don't try to disable Bus Master on disconnected PCI devices

[Konstantin Khlebnikov]

This is fix for commit 7897e6022761ace7377f0f784fca059da55f5d71 from v3.9-rc1
("PCI: Disable Bus Master unconditionally in pci_device_shutdown()")
in turn that was fix for b566a22c23327f18ce941ffad0ca907e50a53d41 from v3.5-rc1
("PCI: disable Bus Master on PCI device shutdown")

Unfortunately fixing one bug uncovers another:
->shutdown() callback might switch device to deep sleep state.
PCI config space no longer available after that.

Link: https://lkml.org/lkml/2013/3/12/529
Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
Reported-and-Tested-by: Vivek Goyal <vgoyal@redhat.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
 drivers/pci/pci-driver.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 1fa1e48..79277fb 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -390,9 +390,10 @@ static void pci_device_shutdown(struct device *dev)
 
 	/*
 	 * Turn off Bus Master bit on the device to tell it to not
-	 * continue to do DMA
+	 * continue to do DMA. Don't touch devices in D3cold or unknown states.
 	 */
-	pci_clear_master(pci_dev);
+	if (pci_dev->current_state <= PCI_D3hot)
+		pci_clear_master(pci_dev);
 }
 
 #ifdef CONFIG_PM

Re: [PATCH bugfix 3.9] PCI: Don't try to disable Bus Master on disconnected PCI devices

[Konstantin Khlebnikov]

BUMP. This is degradation from 3.8, so this patch must be in 3.9.

I still don't like this forced clearing bus-master bit. But this hack
definitely fixes problems in kexec, so there is reason to keep it here.

Konstantin Khlebnikov wrote:
> This is fix for commit 7897e6022761ace7377f0f784fca059da55f5d71 from v3.9-rc1
> ("PCI: Disable Bus Master unconditionally in pci_device_shutdown()")
> in turn that was fix for b566a22c23327f18ce941ffad0ca907e50a53d41 from v3.5-rc1
> ("PCI: disable Bus Master on PCI device shutdown")
>
> Unfortunately fixing one bug uncovers another:
> ->shutdown() callback might switch device to deep sleep state.
> PCI config space no longer available after that.
>
> Link: https://lkml.org/lkml/2013/3/12/529
> Signed-off-by: Konstantin Khlebnikov<khlebnikov@openvz.org>
> Reported-and-Tested-by: Vivek Goyal<vgoyal@redhat.com>
> Cc: Bjorn Helgaas<bhelgaas@google.com>
> Cc: Rafael J. Wysocki<rafael.j.wysocki@intel.com>
> ---
>   drivers/pci/pci-driver.c |    5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
> index 1fa1e48..79277fb 100644
> --- a/drivers/pci/pci-driver.c
> +++ b/drivers/pci/pci-driver.c
> @@ -390,9 +390,10 @@ static void pci_device_shutdown(struct device *dev)
>
>   	/*
>   	 * Turn off Bus Master bit on the device to tell it to not
> -	 * continue to do DMA
> +	 * continue to do DMA. Don't touch devices in D3cold or unknown states.
>   	 */
> -	pci_clear_master(pci_dev);
> +	if (pci_dev->current_state <= PCI_D3hot)
> +		pci_clear_master(pci_dev);
>   }
>
>   #ifdef CONFIG_PM
>

Re: [PATCH bugfix 3.9] PCI: Don't try to disable Bus Master on disconnected PCI devices

[Bjorn Helgaas]

[+cc linux-pci]

On Mon, Apr 1, 2013 at 2:00 AM, Konstantin Khlebnikov
<khlebnikov@openvz.org> wrote:
> BUMP. This is degradation from 3.8, so this patch must be in 3.9.
>
> I still don't like this forced clearing bus-master bit. But this hack
> definitely fixes problems in kexec, so there is reason to keep it here.

Applied to for-linus for v3.9, thanks!

> Konstantin Khlebnikov wrote:
>>
>> This is fix for commit 7897e6022761ace7377f0f784fca059da55f5d71 from
>> v3.9-rc1
>> ("PCI: Disable Bus Master unconditionally in pci_device_shutdown()")
>> in turn that was fix for b566a22c23327f18ce941ffad0ca907e50a53d41 from
>> v3.5-rc1
>> ("PCI: disable Bus Master on PCI device shutdown")
>>
>> Unfortunately fixing one bug uncovers another:
>> ->shutdown() callback might switch device to deep sleep state.
>> PCI config space no longer available after that.
>>
>> Link: https://lkml.org/lkml/2013/3/12/529
>> Signed-off-by: Konstantin Khlebnikov<khlebnikov@openvz.org>
>> Reported-and-Tested-by: Vivek Goyal<vgoyal@redhat.com>
>> Cc: Bjorn Helgaas<bhelgaas@google.com>
>> Cc: Rafael J. Wysocki<rafael.j.wysocki@intel.com>
>> ---
>>   drivers/pci/pci-driver.c |    5 +++--
>>   1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
>> index 1fa1e48..79277fb 100644
>> --- a/drivers/pci/pci-driver.c
>> +++ b/drivers/pci/pci-driver.c
>> @@ -390,9 +390,10 @@ static void pci_device_shutdown(struct device *dev)
>>
>>         /*
>>          * Turn off Bus Master bit on the device to tell it to not
>> -        * continue to do DMA
>> +        * continue to do DMA. Don't touch devices in D3cold or unknown
>> states.
>>          */
>> -       pci_clear_master(pci_dev);
>> +       if (pci_dev->current_state <= PCI_D3hot)
>> +               pci_clear_master(pci_dev);
>>   }
>>
>>   #ifdef CONFIG_PM
>>
>