Re: [RFC net-next v3 4/6] net: phy: nxp-c45-tja11xx: add MACsec support

[Simon Horman]

On Wed, Sep 06, 2023 at 07:01:32PM +0300, Radu Pirea (NXP OSS) wrote:
> Add MACsec support.
> The MACsec block has four TX SCs and four RX SCs. The driver supports up
> to four SecY. Each SecY with one TX SC and one RX SC.
> The RX SCs can have two keys, key A and key B, written in hardware and
> enabled at the same time.
> The TX SCs can have two keys written in hardware, but only one can be
> active at a given time.
> On TX, the SC is selected using the MAC source address. Due of this
> selection mechanism, each offloaded netdev must have a unique MAC
> address.
> On RX, the SC is selected by SCI(found in SecTAG or calculated using MAC
> SA), or using RX SC 0 as implicit.
> 
> Signed-off-by: Radu Pirea (NXP OSS) <radu-nicolae.pirea@xxxxxxxxxxx>

...

> +static struct nxp_c45_sa *nxp_c45_find_sa(struct list_head *sa_list,
> +					  enum nxp_c45_sa_type sa_type, u8 an)
> +{
> +	struct nxp_c45_sa *pos, *tmp;
> +
> +	list_for_each_entry_safe(pos, tmp, sa_list, list)
> +		if (pos->an == an && pos->type == sa_type)
> +			return pos;
> +
> +	return ERR_PTR(-EINVAL);
> +}

...

> +void nxp_c45_handle_macsec_interrupt(struct phy_device *phydev,
> +				     irqreturn_t *ret)
> +{
> +	struct nxp_c45_phy *priv = phydev->priv;
> +	struct nxp_c45_secy *pos, *tmp;
> +	struct nxp_c45_sa *sa;
> +	u8 encoding_sa;
> +	int secy_id;
> +	u32 reg = 0;
> +
> +	if (!phydev->macsec_ops)
> +		return;
> +
> +	do {
> +		nxp_c45_macsec_read(phydev, MACSEC_EVR, &reg);
> +		if (!reg)
> +			return;
> +
> +		secy_id = MACSEC_REG_SIZE - ffs(reg);
> +		list_for_each_entry_safe(pos, tmp, &priv->macsec->secy_list,
> +					 list)
> +			if (pos->secy_id == secy_id)
> +				break;
> +
> +		encoding_sa = pos->secy->tx_sc.encoding_sa;
> +		phydev_dbg(phydev, "pn_wrapped: TX SC %d, encoding_sa %u\n",
> +			   pos->secy_id, encoding_sa);
> +
> +		sa = nxp_c45_find_sa(&pos->sa_list, TX_SA, encoding_sa);
> +		if (!IS_ERR(sa))
> +			macsec_pn_wrapped(pos->secy, sa->sa);
> +		else
> +			WARN_ON(!sa);

Hi Radu,

Smatch doesn't seem to think that sa can be NULL: it is either a valid
pointer or an error pointer.

> +
> +		nxp_c45_macsec_write(phydev, MACSEC_EVR,
> +				     TX_SC_BIT(pos->secy_id));
> +		*ret = IRQ_HANDLED;
> +	} while (reg);
> +}

...