[PATCH] net: stmmac: Do not enable/disable runtime PM for PCI devices

Mikhail Malyshev posted 1 patch 1 year, 10 months ago
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
[PATCH] net: stmmac: Do not enable/disable runtime PM for PCI devices
Posted by Mikhail Malyshev 1 year, 10 months ago
Common function stmmac_dvr_probe is called for both PCI and non-PCI
device. For PCI devices pm_runtime_enable/disable are called by framework
and should not be called by the driver.

For PCI devices plat->pdev != NULL. Use this fact to detect PCI devices

Signed-off-by: Mikhail Malyshev <mike.malyshev@gmail.com>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 24cd80490d19..db45d8dbc1eb 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -7743,7 +7743,9 @@ int stmmac_dvr_probe(struct device *device,
 
 	pm_runtime_get_noresume(device);
 	pm_runtime_set_active(device);
-	if (!pm_runtime_enabled(device))
+
+	/* For PCI devices PM is disabled/enabled by the framework */
+	if (!priv->plat->pdev)
 		pm_runtime_enable(device);
 
 	if (priv->hw->pcs != STMMAC_PCS_TBI &&
@@ -7846,7 +7848,10 @@ void stmmac_dvr_remove(struct device *dev)
 	mutex_destroy(&priv->lock);
 	bitmap_free(priv->af_xdp_zc_qps);
 
-	pm_runtime_disable(dev);
+	/* For PCI devices PM is disabled/enabled by the framework */
+	if (!priv->plat->pdev)
+		pm_runtime_disable(dev);
+
 	pm_runtime_put_noidle(dev);
 }
 EXPORT_SYMBOL_GPL(stmmac_dvr_remove);

---
base-commit: 23956900041d968f9ad0f30db6dede4daccd7aa9
change-id: 20240321-stmmac-fix-f506d52882d2

Best regards,
-- 
Mikhail Malyshev <mike.malyshev@gmail.com>
Re: [PATCH] net: stmmac: Do not enable/disable runtime PM for PCI devices
Posted by Jiri Pirko 1 year, 10 months ago
Thu, Mar 21, 2024 at 02:54:15PM CET, mike.malyshev@gmail.com wrote:
>Common function stmmac_dvr_probe is called for both PCI and non-PCI
>device. For PCI devices pm_runtime_enable/disable are called by framework
>and should not be called by the driver.

I don't follow. The rest of the pm_runtime* functions are okay to call,
but enable() is not. Why? You need to provide more reasoning.


>
>For PCI devices plat->pdev != NULL. Use this fact to detect PCI devices

Sentence ends with "."

I assume this is a bug fix. Do you have a trace or some other symptoms?
Please add it to the patch description. You also need to add "Fixes"
tag.

Make sure you read:
https://www.kernel.org/doc/html/next/process/maintainer-netdev.html?highlight=network#tl-dr

Thanks

pw-bot: cr


>
>Signed-off-by: Mikhail Malyshev <mike.malyshev@gmail.com>
>---
> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++++++--
> 1 file changed, 7 insertions(+), 2 deletions(-)
>
>diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>index 24cd80490d19..db45d8dbc1eb 100644
>--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>@@ -7743,7 +7743,9 @@ int stmmac_dvr_probe(struct device *device,
> 
> 	pm_runtime_get_noresume(device);
> 	pm_runtime_set_active(device);
>-	if (!pm_runtime_enabled(device))
>+
>+	/* For PCI devices PM is disabled/enabled by the framework */
>+	if (!priv->plat->pdev)
> 		pm_runtime_enable(device);
> 
> 	if (priv->hw->pcs != STMMAC_PCS_TBI &&
>@@ -7846,7 +7848,10 @@ void stmmac_dvr_remove(struct device *dev)
> 	mutex_destroy(&priv->lock);
> 	bitmap_free(priv->af_xdp_zc_qps);
> 
>-	pm_runtime_disable(dev);
>+	/* For PCI devices PM is disabled/enabled by the framework */
>+	if (!priv->plat->pdev)
>+		pm_runtime_disable(dev);
>+
> 	pm_runtime_put_noidle(dev);
> }
> EXPORT_SYMBOL_GPL(stmmac_dvr_remove);
>
>---
>base-commit: 23956900041d968f9ad0f30db6dede4daccd7aa9
>change-id: 20240321-stmmac-fix-f506d52882d2
>
>Best regards,
>-- 
>Mikhail Malyshev <mike.malyshev@gmail.com>
>
>