[PATCH] net: stmmac: dwmac-rk: Fix ethernet on rk3399 based devices

Commit 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
while getting rid of a runtime PM warning ended up breaking ethernet
on rk3399 based devices. By dropping an extra reference to the device,
the commit ends up enabling suspend / resume of the ethernet device -
which appears to be broken.

While the issue with runtime pm is being investigated, partially
revert commit 2d26f6e39afb to restore the network on rk3399.

Fixes: 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
Suggested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Punit Agrawal <redacted>
Cc: Michael Riesch <redacted>
---
Hi,

There's been a few reports of broken ethernet on rk3399 based
boards. The issue got introduced due to a late commit in the 5.14
cycle.

It would be great if this commit can be taken as a fix for the next rc
as well as applied to the 5.14 stable releases.

Thanks,
Punit

 drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
index ed817011a94a..6924a6aacbd5 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c

@@ -21,6 +21,7 @@
 #include <linux/delay.h>
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
+#include <linux/pm_runtime.h>
 
 #include "stmmac_platform.h"
 

@@ -1528,6 +1529,8 @@ static int rk_gmac_powerup(struct rk_priv_data *bsp_priv)
 		return ret;
 	}
 
+	pm_runtime_get_sync(dev);
+
 	if (bsp_priv->integrated_phy)
 		rk_gmac_integrated_phy_powerup(bsp_priv);
 

@@ -1539,6 +1542,8 @@ static void rk_gmac_powerdown(struct rk_priv_data *gmac)
 	if (gmac->integrated_phy)
 		rk_gmac_integrated_phy_powerdown(gmac);
 
+	pm_runtime_put_sync(&gmac->pdev->dev);
+
 	phy_power_on(gmac, false);
 	gmac_clk_enable(gmac, false);
 }

-- 
2.32.0

Am Mittwoch, 29. September 2021, 15:50:49 CEST schrieb Punit Agrawal:

Commit 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
while getting rid of a runtime PM warning ended up breaking ethernet
on rk3399 based devices. By dropping an extra reference to the device,
the commit ends up enabling suspend / resume of the ethernet device -
which appears to be broken.

While the issue with runtime pm is being investigated, partially
revert commit 2d26f6e39afb to restore the network on rk3399.

Fixes: 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
Suggested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Punit Agrawal <redacted>
Cc: Michael Riesch <redacted>

On a rk3399-puma which has the described issue,
Tested-by: Heiko Stuebner <heiko@sntech.de>

On Wed, 29 Sep 2021 23:02:35 +0200 Heiko Stübner wrote:

Am Mittwoch, 29. September 2021, 15:50:49 CEST schrieb Punit Agrawal:

quoted

Commit 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
while getting rid of a runtime PM warning ended up breaking ethernet
on rk3399 based devices. By dropping an extra reference to the device,
the commit ends up enabling suspend / resume of the ethernet device -
which appears to be broken.

While the issue with runtime pm is being investigated, partially
revert commit 2d26f6e39afb to restore the network on rk3399.

Fixes: 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
Suggested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Punit Agrawal <redacted>
Cc: Michael Riesch <redacted>  

On a rk3399-puma which has the described issue,
Tested-by: Heiko Stuebner <heiko@sntech.de>

Applied, thanks!

On 16:02 01.10.21, Jakub Kicinski wrote:

On Wed, 29 Sep 2021 23:02:35 +0200 Heiko Stübner wrote:

quoted

Am Mittwoch, 29. September 2021, 15:50:49 CEST schrieb Punit Agrawal:

quoted

Commit 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
while getting rid of a runtime PM warning ended up breaking ethernet
on rk3399 based devices. By dropping an extra reference to the device,
the commit ends up enabling suspend / resume of the ethernet device -
which appears to be broken.

While the issue with runtime pm is being investigated, partially
revert commit 2d26f6e39afb to restore the network on rk3399.

Fixes: 2d26f6e39afb ("net: stmmac: dwmac-rk: fix unbalanced pm_runtime_enable warnings")
Suggested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Punit Agrawal <redacted>
Cc: Michael Riesch <redacted>  

On a rk3399-puma which has the described issue,
Tested-by: Heiko Stuebner <heiko@sntech.de>

Applied, thanks!

This also fixed the issue on a RockPi4.

Will any of you submit this to the stable kernels (as this broke within
3.13 for me) or shall I do that?

On Sat, 2 Oct 2021 23:33:03 +0200 Andreas Rammhold wrote:

On 16:02 01.10.21, Jakub Kicinski wrote:

quoted

On Wed, 29 Sep 2021 23:02:35 +0200 Heiko Stübner wrote:  

quoted

On a rk3399-puma which has the described issue,
Tested-by: Heiko Stuebner <heiko@sntech.de>  

Applied, thanks!  

This also fixed the issue on a RockPi4.

Will any of you submit this to the stable kernels (as this broke within
3.13 for me) or shall I do that?

I won't. The patch should be in Linus's tree in around 1 week - at which
point anyone can request the backport.

That said, as you probably know, 4.4 is the oldest active stable branch,
the ship has sailed for anything 3.x.

On 17:20 02.10.21, Jakub Kicinski wrote:

On Sat, 2 Oct 2021 23:33:03 +0200 Andreas Rammhold wrote:

quoted

On 16:02 01.10.21, Jakub Kicinski wrote:

quoted

On Wed, 29 Sep 2021 23:02:35 +0200 Heiko Stübner wrote:  

quoted

On a rk3399-puma which has the described issue,
Tested-by: Heiko Stuebner <heiko@sntech.de>  

Applied, thanks!  

This also fixed the issue on a RockPi4.

Will any of you submit this to the stable kernels (as this broke within
3.13 for me) or shall I do that?

I won't. The patch should be in Linus's tree in around 1 week - at which
point anyone can request the backport.

That said, as you probably know, 4.4 is the oldest active stable branch,
the ship has sailed for anything 3.x.

I am sorry. I meant 5.13.

Am Sonntag, 3. Oktober 2021, 02:41:03 CEST schrieb Andreas Rammhold:

On 17:20 02.10.21, Jakub Kicinski wrote:

quoted

On Sat, 2 Oct 2021 23:33:03 +0200 Andreas Rammhold wrote:

quoted

On 16:02 01.10.21, Jakub Kicinski wrote:

quoted

On Wed, 29 Sep 2021 23:02:35 +0200 Heiko Stübner wrote:  

quoted

On a rk3399-puma which has the described issue,
Tested-by: Heiko Stuebner <heiko@sntech.de>  

Applied, thanks!  

This also fixed the issue on a RockPi4.

Will any of you submit this to the stable kernels (as this broke within
3.13 for me) or shall I do that?

I won't. The patch should be in Linus's tree in around 1 week - at which
point anyone can request the backport.

That said, as you probably know, 4.4 is the oldest active stable branch,
the ship has sailed for anything 3.x.

I am sorry. I meant 5.13.

As the commit has "fixes" tag, it should be picked up automatically
for stable kernels that include the original commit.


Heiko