netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Lunn <andrew@lunn.ch>
To: "Radu Pirea (NXP OSS)" <radu-nicolae.pirea@oss.nxp.com>
Cc: hkallweit1@gmail.com, linux@armlinux.org.uk, davem@davemloft.net,
	kuba@kernel.org, netdev@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] phy: nxp-c45: add driver for tja1103
Date: Mon, 12 Apr 2021 20:04:20 +0200	[thread overview]
Message-ID: <YHSLpGQclt6EshDF@lunn.ch> (raw)
In-Reply-To: <20210409184106.264463-1-radu-nicolae.pirea@oss.nxp.com>

> +static const struct nxp_c45_phy_stats nxp_c45_hw_stats[] = {
> +	{ "phy_symbol_error_cnt", MDIO_MMD_VEND1, SYMBOL_ERROR_COUNTER, 0, GENMASK(15, 0) },
> +	{ "phy_link_status_drop_cnt", MDIO_MMD_VEND1, LINK_DROP_COUNTER, 8, GENMASK(13, 8) },
> +	{ "phy_link_availability_drop_cnt", MDIO_MMD_VEND1, LINK_DROP_COUNTER, 0, GENMASK(5, 0) },

netdev tries to keep with the old 80 character limit. Please wrap the
long lines.

> +static void nxp_c45_set_delays(struct phy_device *phydev)
> +{
> +	struct nxp_c45_phy *priv = phydev->priv;
> +	u64 tx_delay = priv->tx_delay;
> +	u64 rx_delay = priv->rx_delay;
> +	u64 degree;
> +
> +	if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID ||
> +	    phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) {
> +		degree = tx_delay / PS_PER_DEGREE;
> +		phy_write_mmd(phydev, MDIO_MMD_VEND1, VND1_TXID,
> +			      ID_ENABLE | nxp_c45_get_phase_shift(degree));
> +	}

You are missing an else clause. You need to ensure the delay is 0 if
delays are not required. You have no idea what the bootloader has
done.

> +static int nxp_c45_get_delays(struct phy_device *phydev)
> +{
> +	struct nxp_c45_phy *priv = phydev->priv;
> +	int ret;
> +
> +	if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID ||
> +	    phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) {
> +		ret = device_property_read_u32(&phydev->mdio.dev, "tx-internal-delay-ps",
> +					       &priv->tx_delay);
> +		if (ret) {
> +			phydev_err(phydev, "tx-internal-delay-ps property missing\n");

This is not normally mandatory. Default to 2ns if not specified in DT.

> +static int nxp_c45_set_phy_mode(struct phy_device *phydev)
> +{
> +	int ret;
> +
> +	ret = phy_read_mmd(phydev, MDIO_MMD_VEND1, ABILITIES);
> +	phydev_dbg(phydev, "Clause 45 managed PHY abilities 0x%x\n", ret);
> +
> +	switch (phydev->interface) {
> +	case PHY_INTERFACE_MODE_RGMII:
> +		if (!(ret & RGMII_ABILITY)) {
> +			phydev_err(phydev, "rgmii mode not supported\n");
> +			return -EINVAL;
> +		}
> +		phy_write_mmd(phydev, MDIO_MMD_VEND1, MII_BASIC_CONFIG, MII_BASIC_CONFIG_RGMII);
> +		break;
> +	case PHY_INTERFACE_MODE_RGMII_ID:
> +	case PHY_INTERFACE_MODE_RGMII_TXID:
> +	case PHY_INTERFACE_MODE_RGMII_RXID:
> +		if (!(ret & RGMII_ID_ABILITY)) {
> +			phydev_err(phydev, "rgmii-id, rgmii-txid, rgmii-rxid modes are not supported\n");
> +			return -EINVAL;
> +		}
> +		phy_write_mmd(phydev, MDIO_MMD_VEND1, MII_BASIC_CONFIG, MII_BASIC_CONFIG_RGMII);
> +		ret = nxp_c45_get_delays(phydev);
> +		if (ret)
> +			return ret;
> +
> +		nxp_c45_set_delays(phydev);
> +		break;

Again, for PHY_INTERFACE_MODE_RGMII you need to ensure the hardware is
not inserting a delay.

> +	case PHY_INTERFACE_MODE_SGMII:
> +		if (!(ret & SGMII_ABILITY)) {
> +			phydev_err(phydev, "sgmii mode not supported\n");
> +			return -EINVAL;
> +		}
> +		phy_write_mmd(phydev, MDIO_MMD_VEND1, MII_BASIC_CONFIG, MII_BASIC_CONFIG_SGMII);
> +		break;

Interested. What gets reported over the inband signalling?

	    Andrew

      parent reply	other threads:[~2021-04-12 18:04 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-09 18:41 [PATCH] phy: nxp-c45: add driver for tja1103 Radu Pirea (NXP OSS)
2021-04-09 19:18 ` Heiner Kallweit
2021-04-12  9:10   ` Radu Nicolae Pirea (NXP OSS)
2021-04-09 19:31 ` Jakub Kicinski
2021-04-09 19:36 ` Andrew Lunn
2021-04-12 10:02   ` Radu Nicolae Pirea (NXP OSS)
2021-04-12 12:57     ` Andrew Lunn
2021-04-12 14:11       ` Radu Nicolae Pirea (NXP OSS)
2021-04-12 14:23         ` Andrew Lunn
2021-04-12 14:49           ` Radu Nicolae Pirea (NXP OSS)
2021-04-12 16:52             ` Andrew Lunn
2021-04-13  6:56               ` Christian Herber
2021-04-13 13:30                 ` Andrew Lunn
2021-04-13 13:44                   ` Christian Herber
2021-04-13 13:57                     ` Andrew Lunn
2021-04-13 14:02                       ` Christian Herber
2021-04-13 14:04                         ` Andrew Lunn
2021-04-11  2:33 ` kernel test robot
2021-04-12  9:50 ` Russell King - ARM Linux admin
2021-04-13 13:44   ` Radu Nicolae Pirea (NXP OSS)
2021-04-12 18:04 ` Andrew Lunn [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YHSLpGQclt6EshDF@lunn.ch \
    --to=andrew@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=hkallweit1@gmail.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@armlinux.org.uk \
    --cc=netdev@vger.kernel.org \
    --cc=radu-nicolae.pirea@oss.nxp.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).