diff mbox

[v2] PCI: enable ASPM configuration in PCIE POWERSAVE mode

Message ID 1405317877-15060-1-git-send-email-vidyas@nvidia.com (mailing list archive)
State New, archived
Delegated to: Bjorn Helgaas
Headers show

Commit Message

vidya sagar July 14, 2014, 6:04 a.m. UTC
From: Vidya Sagar <sagar.tv@gmail.com>

'commit 1a680b7c3258 ("PCI: PCIe links may not get configured for ASPM
 under POWERSAVE mode")' moved pcie_aspm_powersave_config_link() out of
pci_raw_set_power_state() to pci_set_power_state() which would enable
ASPM. But, with 'commit db288c9c5f9d ("PCI / PM: restore the original
 behavior of pci_set_power_state()")', which re-introduced the following check
./drivers/pci/pci.c: pci_set_power_state()
+   /* Check if we're already there */
+   if (dev->current_state == state)
+       return 0;
in pci_set_power_state(), call to pcie_aspm_powersave_config_link() is never
made leaving ASPM broken.
Fix it by not returning from when the above condition is true, rather, jump to
ASPM configuration code and exit from there eventually.

Signed-off-by: Vidya Sagar <sagar.tv@gmail.com>
---
v2:
* Calling ASPM config code is moved from pci_set_power_state() to
* do_pci_enable_device()

 drivers/pci/pci.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

Comments

Bjorn Helgaas July 15, 2014, 8:51 p.m. UTC | #1
[+cc Matthew, Rafael]

On Mon, Jul 14, 2014 at 12:04 AM, Vidya Sagar <sagar.tv@gmail.com> wrote:
> From: Vidya Sagar <sagar.tv@gmail.com>
>
> 'commit 1a680b7c3258 ("PCI: PCIe links may not get configured for ASPM
>  under POWERSAVE mode")' moved pcie_aspm_powersave_config_link() out of
> pci_raw_set_power_state() to pci_set_power_state() which would enable
> ASPM. But, with 'commit db288c9c5f9d ("PCI / PM: restore the original
>  behavior of pci_set_power_state()")', which re-introduced the following check
> ./drivers/pci/pci.c: pci_set_power_state()
> +   /* Check if we're already there */
> +   if (dev->current_state == state)
> +       return 0;
> in pci_set_power_state(), call to pcie_aspm_powersave_config_link() is never
> made leaving ASPM broken.
> Fix it by not returning from when the above condition is true, rather, jump to
> ASPM configuration code and exit from there eventually.
>
> Signed-off-by: Vidya Sagar <sagar.tv@gmail.com>

I'm OK with the patch (obviously, since I wrote it), but you neglected
to copy Matthew and Rafael.  And you forgot to rewrite the changelog.
I'll take care of the changelog, but I would like to see something
from Matthew and Rafael, since this is a significant change in how we
configure ASPM.

Matthew, Rafael, my observation was: "But I wonder if we can untangle
ASPM from pci_set_power_state().  I don't think they're really
related.  There are a zillion .suspend() and .resume() methods that
call pci_set_power_state().  I doubt that we need to do ASPM
configuration during suspend, and I'm dubious about resume, too."

So the new patch moves the pcie_aspm_powersave_config_link() call out
of pci_set_power_state() and into do_pci_enable_device().

Bjorn

> ---
> v2:
> * Calling ASPM config code is moved from pci_set_power_state() to
> * do_pci_enable_device()
>
>  drivers/pci/pci.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index 1c8592b..81d49d3 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -839,12 +839,6 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
>
>         if (!__pci_complete_power_transition(dev, state))
>                 error = 0;
> -       /*
> -        * When aspm_policy is "powersave" this call ensures
> -        * that ASPM is configured.
> -        */
> -       if (!error && dev->bus->self)
> -               pcie_aspm_powersave_config_link(dev->bus->self);
>
>         return error;
>  }
> @@ -1195,12 +1189,18 @@ int __weak pcibios_enable_device(struct pci_dev *dev, int bars)
>  static int do_pci_enable_device(struct pci_dev *dev, int bars)
>  {
>         int err;
> +       struct pci_dev *bridge;
>         u16 cmd;
>         u8 pin;
>
>         err = pci_set_power_state(dev, PCI_D0);
>         if (err < 0 && err != -EIO)
>                 return err;
> +
> +       bridge = pci_upstream_bridge(dev);
> +       if (bridge)
> +               pcie_aspm_powersave_config_link(bridge);
> +
>         err = pcibios_enable_device(dev, bars);
>         if (err < 0)
>                 return err;
> --
> 1.8.1.5
>
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thierry Reding July 16, 2014, 8:15 a.m. UTC | #2
On Mon, Jul 14, 2014 at 11:34:37AM +0530, Vidya Sagar wrote:
> From: Vidya Sagar <sagar.tv@gmail.com>

The commit message looks somewhat cluttered:

> 'commit 1a680b7c3258 ("PCI: PCIe links may not get configured for ASPM
>  under POWERSAVE mode")'

Usually you'd cite a commit without the extra single quotes. More like
this:

Commit 1a680b7c3258 ("PCI: PCIe links may not get configured for ASPM
under POWERSAVE mode") moved...

> moved pcie_aspm_powersave_config_link() out of
> pci_raw_set_power_state() to pci_set_power_state() which would enable
> ASPM. But, with 'commit db288c9c5f9d ("PCI / PM: restore the original
>  behavior of pci_set_power_state()")',

Same here.

>                                        which re-introduced the following check
> ./drivers/pci/pci.c: pci_set_power_state()
> +   /* Check if we're already there */
> +   if (dev->current_state == state)
> +       return 0;

To make this stand out better, it could be preceded and followed by a
single blank line (and perhaps indented using tabs).

> in pci_set_power_state(), call to pcie_aspm_powersave_config_link() is never
> made leaving ASPM broken.
> Fix it by not returning from when the above condition is true, rather, jump to
> ASPM configuration code and exit from there eventually.
> 
> Signed-off-by: Vidya Sagar <sagar.tv@gmail.com>

This patch uses code suggested by Bjorn and it's customary to credit the
original author using something like:

	Suggested-by: Bjorn Helgaas <bhelgaas@google.com>

or

	Originally-by: Bjorn Helgaas <bhelgaas@google.com>

Thierry
diff mbox

Patch

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 1c8592b..81d49d3 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -839,12 +839,6 @@  int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 
 	if (!__pci_complete_power_transition(dev, state))
 		error = 0;
-	/*
-	 * When aspm_policy is "powersave" this call ensures
-	 * that ASPM is configured.
-	 */
-	if (!error && dev->bus->self)
-		pcie_aspm_powersave_config_link(dev->bus->self);
 
 	return error;
 }
@@ -1195,12 +1189,18 @@  int __weak pcibios_enable_device(struct pci_dev *dev, int bars)
 static int do_pci_enable_device(struct pci_dev *dev, int bars)
 {
 	int err;
+	struct pci_dev *bridge;
 	u16 cmd;
 	u8 pin;
 
 	err = pci_set_power_state(dev, PCI_D0);
 	if (err < 0 && err != -EIO)
 		return err;
+
+	bridge = pci_upstream_bridge(dev);
+	if (bridge)
+		pcie_aspm_powersave_config_link(bridge);
+
 	err = pcibios_enable_device(dev, bars);
 	if (err < 0)
 		return err;