netdev - Re: [PATCH net-next 2/3] net: phy: marvell10g: add energy detect power down tunable

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 3 Mar 2020 15:30:13 +0000
From:   Russell King - ARM Linux admin <linux@...linux.org.uk>
To:     Antoine Tenart <antoine.tenart@...tlin.com>
Cc:     Andrew Lunn <andrew@...n.ch>,
        Florian Fainelli <f.fainelli@...il.com>,
        Heiner Kallweit <hkallweit1@...il.com>,
        "David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org
Subject: Re: [PATCH net-next 2/3] net: phy: marvell10g: add energy detect
 power down tunable

On Tue, Mar 03, 2020 at 04:19:58PM +0100, Antoine Tenart wrote:
> On Tue, Mar 03, 2020 at 03:12:32PM +0000, Russell King - ARM Linux admin wrote:
> > On Tue, Mar 03, 2020 at 04:07:41PM +0100, Antoine Tenart wrote:
> > > On Tue, Mar 03, 2020 at 02:44:02PM +0000, Russell King wrote:
> > > >  drivers/net/phy/marvell10g.c | 111 ++++++++++++++++++++++++++++++++++-
> > > >  
> > > > +static int mv3310_maybe_reset(struct phy_device *phydev, u32 unit, bool reset)
> > > > +{
> > > > +	int retries, val, err;
> > > > +
> > > > +	if (!reset)
> > > > +		return 0;
> > > 
> > > You could also call mv3310_maybe_reset after testing the 'reset'
> > > condition, that would make it easier to read the code.
> > 
> > I'm not too convinced:
> > 
> > diff --git a/drivers/net/phy/marvell10g.c b/drivers/net/phy/marvell10g.c
> > index ef1ed9415d9f..3daf73e61dff 100644
> > --- a/drivers/net/phy/marvell10g.c
> > +++ b/drivers/net/phy/marvell10g.c
> > @@ -279,13 +279,10 @@ static int mv3310_power_up(struct phy_device *phydev)
> >  				  MV_V2_PORT_CTRL_PWRDOWN);
> >  }
> >  
> > -static int mv3310_maybe_reset(struct phy_device *phydev, u32 unit, bool reset)
> > +static int mv3310_reset(struct phy_device *phydev, u32 unit)
> >  {
> >  	int retries, val, err;
> >  
> > -	if (!reset)
> > -		return 0;
> > -
> >  	err = phy_modify_mmd(phydev, MDIO_MMD_PCS, unit + MDIO_CTRL1,
> >  			     MDIO_CTRL1_RESET, MDIO_CTRL1_RESET);
> >  	if (err < 0)
> > @@ -684,10 +681,10 @@ static int mv3310_config_mdix(struct phy_device *phydev)
> >  
> >  	err = phy_modify_mmd_changed(phydev, MDIO_MMD_PCS, MV_PCS_CSCR1,
> >  				     MV_PCS_CSCR1_MDIX_MASK, val);
> > -	if (err < 0)
> > +	if (err <= 0)
> >  		return err;
> >  
> > -	return mv3310_maybe_reset(phydev, MV_PCS_BASE_T, err > 0);
> > +	return mv3310_reset(phydev, MV_PCS_BASE_T);
> >  }
> >  
> >  static int mv3310_config_aneg(struct phy_device *phydev)
> > 
> > The change from:
> > 
> > 	if (err < 0)
> > 
> > to:
> > 
> > 	if (err <= 0)
> > 
> > could easily be mistaken as a bug, and someone may decide to try to
> > "fix" that back to being the former instead.  The way I have the code
> > makes the intention explicit.
> 
> Using a single line to test both the error and the 'return 0'
> conditions, yes, I agree. Another solution would be to do something of
> the like:
> 
> 	phy_modify_mmd_changed()
> 	if (err < 0)
> 		return err;
> 
> 	if (err)
> 		mv3310_reset();
> 
> 	return 0;
> 
> I find it more readable, but this kind of thing is also a matter of
> personal taste.

Well, it either becomes:

        err = phy_modify_mmd_changed(phydev, MDIO_MMD_PCS, MV_PCS_CSCR1,
                                     MV_PCS_CSCR1_MDIX_MASK, val);
        if (err < 0)
                return err;

        if (err > 0)
                return mv3310_reset(phydev, MV_PCS_BASE_T);

        return 0;

or:

        err = phy_modify_mmd_changed(phydev, MDIO_MMD_PCS, MV_PCS_CSCR1,
                                     MV_PCS_CSCR1_MDIX_MASK, val);
        if (err > 0)
                err = mv3310_reset(phydev, MV_PCS_BASE_T);

        return err;

In the former case, we have two success-exit paths - one via a successful
mv3310_reset() and one by dropping through to the final return statement.

The latter case looks a bit better, at least to me.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
According to speedtest.net: 11.9Mbps down 500kbps up