RE: [PATCH 1/2] net: phylink: add sync flag mac_ready to fix resume issue with WoL enabled

[Clark Wang]

> -----Original Message-----
> From: Russell King <linux@xxxxxxxxxxxxxxx>
> Sent: 2022年11月30日 19:33
> To: Clark Wang <xiaoning.wang@xxxxxxx>
> Cc: peppe.cavallaro@xxxxxx; alexandre.torgue@xxxxxxxxxxx;
> joabreu@xxxxxxxxxxxx; davem@xxxxxxxxxxxxx; edumazet@xxxxxxxxxx;
> kuba@xxxxxxxxxx; pabeni@xxxxxxxxxx; mcoquelin.stm32@xxxxxxxxx;
> andrew@xxxxxxx; hkallweit1@xxxxxxxxx; netdev@xxxxxxxxxxxxxxx;
> linux-stm32@xxxxxxxxxxxxxxxxxxxxxxxxxxxx;
> linux-arm-kernel@xxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH 1/2] net: phylink: add sync flag mac_ready to fix resume
> issue with WoL enabled
> 
> On Wed, Nov 30, 2022 at 11:23:42AM +0000, Russell King (Oracle) wrote:
> > On Wed, Nov 30, 2022 at 07:11:47PM +0800, Clark Wang wrote:
> > > Issue we met:
> > > On some platforms, mac cannot work after resumed from the suspend
> > > with WoL enabled.
> > >
> > > The cause of the issue:
> > > 1. phylink_resolve() is in a workqueue which will not be executed
> immediately.
> > >    This is the call sequence:
> > >        phylink_resolve()->phylink_link_up()->pl->mac_ops->mac_link_up()
> > >    For stmmac driver, mac_link_up() will set the correct speed/duplex...
> > >    values which are from link_state.
> > > 2. In stmmac_resume(), it will call stmmac_hw_setup() after called the
> > >    phylink_resume(). stmmac_core_init() is called in function
> > > stmmac_hw_setup(),
> >
> > ... and that is where the problem is. Don't call phylink_resume()
> > before your hardware is ready to see a link-up event.
> 
> ... and while that is being fixed, maybe the stupid code in
> stmmac_resume() can also be fixed:
> 
>         rtnl_lock();
>         if (device_may_wakeup(priv->device) && priv->plat->pmt) {
>                 phylink_resume(priv->phylink);
>         } else {
>                 phylink_resume(priv->phylink);
>                 if (device_may_wakeup(priv->device))
>                         phylink_speed_up(priv->phylink);
>         }
>         rtnl_unlock();
> 
>         rtnl_lock();
> 
> 1. phylink_resume() is always called after that first rtnl_lock(), so there's no
> point it being stupidly in each side of the if().
> 
> 2. the rtnl_unlock() followed by rtnl_lock() is completely unnecessary.
> 
> Thanks.
> 
> --
> RMK's Patch system:
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ar
> mlinux.org.uk%2Fdeveloper%2Fpatches%2F&amp;data=05%7C01%7Cxiaoning.
> wang%40nxp.com%7Cc0961fa17589464a742308dad2c69502%7C686ea1d3bc2
> b4c6fa92cd99c5c301635%7C0%7C0%7C638054047593066949%7CUnknown%
> 7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwi
> LCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=wURkTxhK2IlqktOKmJzx
> Nr8E4KCzd1gjDgHK3iSy6EA%3D&amp;reserved=0
> FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!