Re: [EXT] [PATCH] net: fec: add a check for CONFIG_PM to avoid clock count mis-match

[Simon Horman]

On Thu, Nov 07, 2019 at 01:44:11AM +0000, Andy Duan wrote:
> From: Chuhong Yuan <hslester96@xxxxxxxxx> Sent: Thursday, November 7, 2019 9:19 AM
> > On Wed, Nov 6, 2019 at 6:17 PM Andy Duan <fugang.duan@xxxxxxx> wrote:
> > >
> > > From: Chuhong Yuan <hslester96@xxxxxxxxx> Sent: Wednesday, November
> > 6,
> > > 2019 4:29 PM
> > > > On Wed, Nov 6, 2019 at 4:13 PM Andy Duan <fugang.duan@xxxxxxx>
> > wrote:
> > > > >
> > > > > From: Chuhong Yuan <hslester96@xxxxxxxxx> Sent: Wednesday,
> > > > > November
> > > > 6,
> > > > > 2019 4:01 PM
> > > > > > If CONFIG_PM is enabled, runtime pm will work and call
> > > > > > runtime_suspend automatically to disable clks.
> > > > > > Therefore, remove only needs to disable clks when CONFIG_PM is
> > > > disabled.
> > > > > > Add this check to avoid clock count mis-match caused by
> > double-disable.
> > > > > >
> > > > > > This patch depends on patch
> > > > > > ("net: fec: add missed clk_disable_unprepare in remove").
> > > > > >
> > > > > Please add Fixes tag here.
> > > > >
> > > >
> > > > The previous patch has not been merged to linux, so I do not know
> > > > which commit ID should be used.
> > >
> > > It should be merged into net-next tree.
> > >
> > 
> > I have searched in net-next but did not find it.

Commit ids are stable, so if there is an id in Linus's tree
it will be same in net-next (when the patch appears there).

So you want:

Fixes: c43eab3eddb4 ("net: fec: add missed clk_disable_unprepare in remove")

Also, it is unclear from the patch subject if this patch is targeted at
'net' or 'net-next'. But as c43eab3eddb4 is in Linus's tree I think
it should be for 'net'. So the correct patch subject would be:

[PATCH net] net: fec: add a check for CONFIG_PM to avoid clock

> David, please give the comment. Thanks.
> 
> Regards,
> Andy
> > 
> > > Andy
> > > >
> > > > > Andy
> > > > > > Signed-off-by: Chuhong Yuan <hslester96@xxxxxxxxx>
> > > > > > ---
> > > > > >  drivers/net/ethernet/freescale/fec_main.c | 2 ++
> > > > > >  1 file changed, 2 insertions(+)
> > > > > >
> > > > > > diff --git a/drivers/net/ethernet/freescale/fec_main.c
> > > > > > b/drivers/net/ethernet/freescale/fec_main.c
> > > > > > index a9c386b63581..696550f4972f 100644
> > > > > > --- a/drivers/net/ethernet/freescale/fec_main.c
> > > > > > +++ b/drivers/net/ethernet/freescale/fec_main.c
> > > > > > @@ -3645,8 +3645,10 @@ fec_drv_remove(struct platform_device
> > > > *pdev)
> > > > > >                 regulator_disable(fep->reg_phy);
> > > > > >         pm_runtime_put(&pdev->dev);
> > > > > >         pm_runtime_disable(&pdev->dev);
> > > > > > +#ifndef CONFIG_PM
> > > > > >         clk_disable_unprepare(fep->clk_ahb);
> > > > > >         clk_disable_unprepare(fep->clk_ipg);
> > > > > > +#endif

FWIIW, I am surprised this is the cleanest way to resolve this problem,
though I confess that I have no specific alternative in mind.

> > > > > >         if (of_phy_is_fixed_link(np))
> > > > > >                 of_phy_deregister_fixed_link(np);
> > > > > >         of_node_put(fep->phy_node);
> > > > > > --
> > > > > > 2.23.0
> > > > >