| Message ID | E1mxq4r-00GWrp-Ay@rmk-PC.armlinux.org.uk (mailing list archive) |
|---|---|
| State | RFC |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [CFT,net-next] net: enetc: use .mac_select_pcs() interface | expand |
On Thu, Dec 16, 2021 at 12:41:41PM +0000, Russell King (Oracle) wrote: > Convert the PCS selection to use mac_select_pcs, which allows the PCS > to perform any validation it needs. > > Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Hi, Can anyone test this please? Claudiu? Russell. > --- > drivers/net/ethernet/freescale/enetc/enetc_pf.c | 14 +++++++++----- > 1 file changed, 9 insertions(+), 5 deletions(-) > > diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c > index fe6a544f37f0..1f5bc8fe0a3c 100644 > --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c > +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c > @@ -930,18 +930,21 @@ static void enetc_mdiobus_destroy(struct enetc_pf *pf) > enetc_imdio_remove(pf); > } > > +static struct phylink_pcs * > +enetc_pl_mac_select_pcs(struct phylink_config *config, phy_interface_t iface) > +{ > + struct enetc_pf *pf = phylink_to_enetc_pf(config); > + > + return pf->pcs ? &pf->pcs->pcs : NULL; > +} > + > static void enetc_pl_mac_config(struct phylink_config *config, > unsigned int mode, > const struct phylink_link_state *state) > { > struct enetc_pf *pf = phylink_to_enetc_pf(config); > - struct enetc_ndev_priv *priv; > > enetc_mac_config(&pf->si->hw, state->interface); > - > - priv = netdev_priv(pf->si->ndev); > - if (pf->pcs) > - phylink_set_pcs(priv->phylink, &pf->pcs->pcs); > } > > static void enetc_force_rgmii_mac(struct enetc_hw *hw, int speed, int duplex) > @@ -1058,6 +1061,7 @@ static void enetc_pl_mac_link_down(struct phylink_config *config, > > static const struct phylink_mac_ops enetc_mac_phylink_ops = { > .validate = phylink_generic_validate, > + .mac_select_pcs = enetc_pl_mac_select_pcs, > .mac_config = enetc_pl_mac_config, > .mac_link_up = enetc_pl_mac_link_up, > .mac_link_down = enetc_pl_mac_link_down, > -- > 2.30.2 > >
> -----Original Message----- > From: Russell King <linux@armlinux.org.uk> > Sent: Friday, January 7, 2022 3:41 PM > To: David S. Miller <davem@davemloft.net>; Jakub Kicinski > <kuba@kernel.org> > Cc: Claudiu Manoil <claudiu.manoil@nxp.com>; netdev@vger.kernel.org > Subject: Re: [PATCH CFT net-next] net: enetc: use .mac_select_pcs() > interface > > On Thu, Dec 16, 2021 at 12:41:41PM +0000, Russell King (Oracle) wrote: > > Convert the PCS selection to use mac_select_pcs, which allows the PCS > > to perform any validation it needs. > > > > Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> > > Hi, > > Can anyone test this please? Claudiu? > > Russell. > Hi Russell, drivers/net/ethernet/freescale/enetc/enetc_pf.c: In function 'enetc_pl_mac_select_pcs': drivers/net/ethernet/freescale/enetc/enetc_pf.c:942:27: error: 'struct phylink_pcs' has no member named 'pcs' 942 | return pf->pcs ? &pf->pcs->pcs : NULL; I suppose you meant: - return pf->pcs ? &pf->pcs->pcs : NULL; + return pf->pcs; With this correction I could bring up the SGMII link on my ls1028rdb. The patch needs also rebase. Thanks. Claudiu
> -----Original Message----- > From: Claudiu Manoil > Sent: Monday, January 10, 2022 7:38 PM > To: Russell King <linux@armlinux.org.uk>; David S. Miller > <davem@davemloft.net>; Jakub Kicinski <kuba@kernel.org> > Cc: netdev@vger.kernel.org > Subject: RE: [PATCH CFT net-next] net: enetc: use .mac_select_pcs() > interface > > > -----Original Message----- > > From: Russell King <linux@armlinux.org.uk> > > Sent: Friday, January 7, 2022 3:41 PM > > To: David S. Miller <davem@davemloft.net>; Jakub Kicinski > > <kuba@kernel.org> > > Cc: Claudiu Manoil <claudiu.manoil@nxp.com>; netdev@vger.kernel.org > > Subject: Re: [PATCH CFT net-next] net: enetc: use .mac_select_pcs() > > interface > > > > On Thu, Dec 16, 2021 at 12:41:41PM +0000, Russell King (Oracle) wrote: > > > Convert the PCS selection to use mac_select_pcs, which allows the PCS > > > to perform any validation it needs. > > > > > > Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> > > > > Hi, > > > > Can anyone test this please? Claudiu? > > > > Russell. > > > > Hi Russell, > > drivers/net/ethernet/freescale/enetc/enetc_pf.c: In function > 'enetc_pl_mac_select_pcs': > drivers/net/ethernet/freescale/enetc/enetc_pf.c:942:27: error: 'struct > phylink_pcs' has no member named 'pcs' > 942 | return pf->pcs ? &pf->pcs->pcs : NULL; > > I suppose you meant: > - return pf->pcs ? &pf->pcs->pcs : NULL; > + return pf->pcs; > > With this correction I could bring up the SGMII link on my ls1028rdb. > The patch needs also rebase. > Actually it's just a matter of rebasing the patch on latest net-next, since commit e7026f15564f "net: phy: lynx: refactor Lynx PCS module to use generic phylink_pcs" replaced 'struct lynx_pcs' with 'struct phylink_pcs'.
On Mon, Jan 10, 2022 at 05:37:55PM +0000, Claudiu Manoil wrote: > Hi Russell, > > drivers/net/ethernet/freescale/enetc/enetc_pf.c: In function 'enetc_pl_mac_select_pcs': > drivers/net/ethernet/freescale/enetc/enetc_pf.c:942:27: error: 'struct phylink_pcs' has no member named 'pcs' > 942 | return pf->pcs ? &pf->pcs->pcs : NULL; > > I suppose you meant: > - return pf->pcs ? &pf->pcs->pcs : NULL; > + return pf->pcs; > > With this correction I could bring up the SGMII link on my ls1028rdb. Thanks. > The patch needs also rebase. Yes, due to other patches have been merged into net-next since the patch was sent for testing. However, now is not the time to be sending it for net-next as the merge window is open, so there's little point rebasing it until after the merge window has closed.
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c index fe6a544f37f0..1f5bc8fe0a3c 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -930,18 +930,21 @@ static void enetc_mdiobus_destroy(struct enetc_pf *pf) enetc_imdio_remove(pf); } +static struct phylink_pcs * +enetc_pl_mac_select_pcs(struct phylink_config *config, phy_interface_t iface) +{ + struct enetc_pf *pf = phylink_to_enetc_pf(config); + + return pf->pcs ? &pf->pcs->pcs : NULL; +} + static void enetc_pl_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { struct enetc_pf *pf = phylink_to_enetc_pf(config); - struct enetc_ndev_priv *priv; enetc_mac_config(&pf->si->hw, state->interface); - - priv = netdev_priv(pf->si->ndev); - if (pf->pcs) - phylink_set_pcs(priv->phylink, &pf->pcs->pcs); } static void enetc_force_rgmii_mac(struct enetc_hw *hw, int speed, int duplex) @@ -1058,6 +1061,7 @@ static void enetc_pl_mac_link_down(struct phylink_config *config, static const struct phylink_mac_ops enetc_mac_phylink_ops = { .validate = phylink_generic_validate, + .mac_select_pcs = enetc_pl_mac_select_pcs, .mac_config = enetc_pl_mac_config, .mac_link_up = enetc_pl_mac_link_up, .mac_link_down = enetc_pl_mac_link_down,
Convert the PCS selection to use mac_select_pcs, which allows the PCS to perform any validation it needs. Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> --- drivers/net/ethernet/freescale/enetc/enetc_pf.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-)