Re: [PATCH] net: phy: fix may not suspend when phy has WoL
From: Wentao Guan
Date: Mon Nov 11 2024 - 06:00:15 EST
Well question, but I do not know any knowledge about phydrv->suspend may or not return EBUSY?
Around the WoL state handling is becoming complex for checking the same logic in phy_suspend
and mdio_bus_phy_may_suspend.
I still do not know why that we need to use -EBUSY in 11 years ago
commit("481b5d9" net: phy: provide phy_resume/phy_suspend helpers)
If it is not need, maybe a simple solution is change EBUSY to 0 and add a warning print?
And discussing the patch, it is same as commit 4f534b7f0 in v6.12-rc1, and it context should before
"if (!drv || !phydrv->suspend)" check[It goes wrong when I moving the patch from our dist branch to upstream branch]
, so I send a NAK for the patch.
BRs
Wentao Guan