linux-kernel - Re: [PATCH net v2 1/1] net: pcs: xpcs: fix incorrect CL37 AN sequence

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] [day] [month] [year] [list]

Message-ID: <20210929131338.6zppmzraznym5rsv@skbuf>
Date:   Wed, 29 Sep 2021 13:13:39 +0000
From:   Vladimir Oltean <vladimir.oltean@....com>
To:     Wong Vee Khee <vee.khee.wong@...ux.intel.com>
CC:     Jose Abreu <Jose.Abreu@...opsys.com>, Andrew Lunn <andrew@...n.ch>,
        Heiner Kallweit <hkallweit1@...il.com>,
        Russell King <linux@...linux.org.uk>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Wong Vee Khee <veekhee@...il.com>,
        Wong Vee Khee <vee.khee.wong@...el.com>
Subject: Re: [PATCH net v2 1/1] net: pcs: xpcs: fix incorrect CL37 AN sequence

On Wed, Sep 29, 2021 at 08:05:34PM +0800, Wong Vee Khee wrote:
> According to Synopsys DesignWare Cores Ethernet PCS databook, it is
> required to disable Clause 37 auto-negotiation by programming bit-12
> (AN_ENABLE) to 0 if it is already enabled, before programming various
> fields of VR_MII_AN_CTRL registers.
> 
> After all these programming are done, it is then required to enable
> Clause 37 auto-negotiation by programming bit-12 (AN_ENABLE) to 1.
> 
> Cc: Vladimir Oltean <vladimir.oltean@....com>
> Signed-off-by: Wong Vee Khee <vee.khee.wong@...ux.intel.com>
> ---

netdev/fixes_present 	fail 	Series targets non-next tree, but doesn't contain any Fixes tags
https://patchwork.kernel.org/project/netdevbpf/patch/20210929120534.411157-1-vee.khee.wong@linux.intel.com/

> v1 -> v2:
>  - Removed use of xpcs_modify() helper function.
>  - Add conditional check on inband auto-negotiation.
> 
>  drivers/net/pcs/pcs-xpcs.c | 30 ++++++++++++++++++++++++++----
>  1 file changed, 26 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
> index fb0a83dc09ac..f34d5caeaba1 100644
> --- a/drivers/net/pcs/pcs-xpcs.c
> +++ b/drivers/net/pcs/pcs-xpcs.c
> @@ -700,11 +700,14 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs, unsigned int mode)
>  	int ret;
>  
>  	/* For AN for C37 SGMII mode, the settings are :-
> -	 * 1) VR_MII_AN_CTRL Bit(2:1)[PCS_MODE] = 10b (SGMII AN)
> -	 * 2) VR_MII_AN_CTRL Bit(3) [TX_CONFIG] = 0b (MAC side SGMII)
> +	 * 1) VR_MII_MMD_CTRL Bit(12) [AN_ENABLE] = 0b (Disable SGMII AN in case
> +	      it is already enabled)
> +	 * 2) VR_MII_AN_CTRL Bit(2:1)[PCS_MODE] = 10b (SGMII AN)
> +	 * 3) VR_MII_AN_CTRL Bit(3) [TX_CONFIG] = 0b (MAC side SGMII)
>  	 *    DW xPCS used with DW EQoS MAC is always MAC side SGMII.
> -	 * 3) VR_MII_DIG_CTRL1 Bit(9) [MAC_AUTO_SW] = 1b (Automatic
> +	 * 4) VR_MII_DIG_CTRL1 Bit(9) [MAC_AUTO_SW] = 1b (Automatic
>  	 *    speed/duplex mode change by HW after SGMII AN complete)
> +	 * 5) VR_MII_MMD_CTRL Bit(12) [AN_ENABLE] = 1b (Enable SGMII AN)
>  	 *
>  	 * Note: Since it is MAC side SGMII, there is no need to set
>  	 *	 SR_MII_AN_ADV. MAC side SGMII receives AN Tx Config from
> @@ -712,6 +715,12 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs, unsigned int mode)
>  	 *	 between PHY and Link Partner. There is also no need to
>  	 *	 trigger AN restart for MAC-side SGMII.
>  	 */
> +	ret = xpcs_read(xpcs, MDIO_MMD_VEND2, DW_VR_MII_MMD_CTRL);
> +	if (ret < 0)
> +		return ret;
> +	ret &= ~AN_CL37_EN;
> +	xpcs_write(xpcs, MDIO_MMD_VEND2, DW_VR_MII_MMD_CTRL, ret);

For consistency can you please check the return code from this xpcs_write too?

> +
>  	ret = xpcs_read(xpcs, MDIO_MMD_VEND2, DW_VR_MII_AN_CTRL);
>  	if (ret < 0)
>  		return ret;
> @@ -736,7 +745,20 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs, unsigned int mode)
>  	else
>  		ret &= ~DW_VR_MII_DIG_CTRL1_MAC_AUTO_SW;
>  
> -	return xpcs_write(xpcs, MDIO_MMD_VEND2, DW_VR_MII_DIG_CTRL1, ret);
> +	ret = xpcs_write(xpcs, MDIO_MMD_VEND2, DW_VR_MII_DIG_CTRL1, ret);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = xpcs_read(xpcs, MDIO_MMD_VEND2, DW_VR_MII_MMD_CTRL);
> +	if (ret < 0)
> +		return ret;
> +
> +	if (phylink_autoneg_inband(mode))
> +		ret |= AN_CL37_EN;
> +	else
> +		ret &= ~AN_CL37_EN;
> +
> +	return xpcs_write(xpcs, MDIO_MMD_VEND2, DW_VR_MII_MMD_CTRL, ret);

Not sure whether this is too much for "net" or not, but I suppose it is
pretty much relevant.

Clearing AN_CL37_EN during the first step is only necessary if it was
set in the first place. Otherwise, you could do only the xpcs_read but
not the xpcs_write.

At the end, you don't really need to read DW_VR_MII_MMD_CTRL again, you
just read it above, didn't you, can't you save it in a local variable?

And you only need to perform the write if the state of the AN_CL37_EN
bit changed, i.e. it was 0 before, and is now 1.

What I'm trying to say is that for the sja1105 driver, the XPCS is
accessed over SPI commands, and I would like for the number of those to
be minimized if possible. Thanks.

I've tested the change and it still works, by the way. I will add the
appropriate tags when you resend the v3.

>  }
>  
>  static int xpcs_config_2500basex(struct dw_xpcs *xpcs)
> -- 
> 2.25.1
>