Building the Linux kernel with Clang and LLVM
 help / color / mirror / Atom feed
From: Christian Marangi <ansuelsmth@gmail.com>
To: Andrew Lunn <andrew@lunn.ch>
Cc: Maxime Chevallier <maxime.chevallier@bootlin.com>,
	Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Simon Horman <horms@kernel.org>, Jonathan Corbet <corbet@lwn.net>,
	Shuah Khan <skhan@linuxfoundation.org>,
	Lorenzo Bianconi <lorenzo@kernel.org>,
	Heiner Kallweit <hkallweit1@gmail.com>,
	Russell King <linux@armlinux.org.uk>,
	Philipp Zabel <p.zabel@pengutronix.de>,
	Nathan Chancellor <nathan@kernel.org>,
	Nick Desaulniers <ndesaulniers@google.com>,
	Bill Wendling <morbo@google.com>,
	Justin Stitt <justinstitt@google.com>,
	netdev@vger.kernel.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-mediatek@lists.infradead.org, llvm@lists.linux.dev
Subject: Re: [PATCH net-next v11 00/12] net: pcs: Introduce support for fwnode PCS
Date: Sun, 9 Aug 2026 19:49:54 +0200	[thread overview]
Message-ID: <6a78bdc6.89544747.3fcae.b8e0@mx.google.com> (raw)
In-Reply-To: <4c595c15-17d9-4c23-bb44-344c48d36e9a@lunn.ch>

On Sun, Aug 09, 2026 at 07:44:55PM +0200, Andrew Lunn wrote:
> On Fri, Aug 07, 2026 at 03:12:42PM +0200, Christian Marangi wrote:
> > This series introduce a most awaited feature that is correctly
> > provide PCS with fwnode without having to use specific export symbol
> > and additional handling of PCS in phylink.
> 
> One of Russells complaints was the phylink has too many bits of code
> for ideas which did not work out and were replaced by new ways of
> doing things.
> 
> We want .mac_select_pcs to go away, this framework should replace it.
> Please could you convert one of the exist users of .mac_select_pcs, so
> we get an idea of what that involved. If that conversion is pretty
> mechanical, i think we should have a second patchset which converts
> all the current users, and then remove support for .mac_select_pcs.
> 
> Overall i think the direction of this patchset is good.
>

Thanks for the initial check.

Maxime converted one device with this and the patch delta was small... let
me copy those patch here. (it was referenced in v9)

I think the most annoying one will be stmmac with the 3 different PCS
implementation but we already handled that in a upcoming ipq50xx driver
currently present in OpenWrt mainline.

Maybe if Maxime is O.K. I can include his patch in this series.

From 38ad94b1e62bf3523097983dfb515f59b1634477 Mon Sep 17 00:00:00 2001
From: Maxime Chevallier <maxime.chevallier@bootlin.com>
Date: Tue, 21 Jul 2026 13:52:57 +0200
Subject: [PATCH] net: marvell: mvpp2: Convert to the new PCS API

Following the introduction of the PCS framework, port mvpp2 to the new
PCS API.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
---
 .../net/ethernet/marvell/mvpp2/mvpp2_main.c   | 161 +++++++++++-------
 1 file changed, 95 insertions(+), 66 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index ccc24a1301f2..8d9663c7a023 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -6499,21 +6499,6 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode,
                writel(ctrl4, port->base + MVPP22_GMAC_CTRL_4_REG);
 }

-static struct phylink_pcs *mvpp2_select_pcs(struct phylink_config *config,
-                                           phy_interface_t interface)
-{
-       struct mvpp2_port *port = mvpp2_phylink_to_port(config);
-
-       /* Select the appropriate PCS operations depending on the
-        * configured interface mode. We will only switch to a mode
-        * that the validate() checks have already passed.
-        */
-       if (mvpp2_is_xlg(interface))
-               return &port->pcs_xlg;
-       else
-               return &port->pcs_gmac;
-}
-
 static int mvpp2_mac_prepare(struct phylink_config *config, unsigned int mode,
                             phy_interface_t interface)
 {
@@ -6786,7 +6771,6 @@ static int mvpp2_mac_enable_tx_lpi(struct phylink_config *config, u32 timer,
 }

 static const struct phylink_mac_ops mvpp2_phylink_ops = {
-       .mac_select_pcs = mvpp2_select_pcs,
        .mac_prepare = mvpp2_mac_prepare,
        .mac_config = mvpp2_mac_config,
        .mac_finish = mvpp2_mac_finish,
@@ -6808,7 +6792,10 @@ static void mvpp2_acpi_start(struct mvpp2_port *port)
        };
        struct phylink_pcs *pcs;

-       pcs = mvpp2_select_pcs(&port->phylink_config, port->phy_interface);
+       if (mvpp2_is_xlg(port->phy_interface))
+               pcs = &port->pcs_xlg;
+       else
+               pcs = &port->pcs_gmac;

        mvpp2_mac_prepare(&port->phylink_config, MLO_AN_INBAND,
                          port->phy_interface);
@@ -6823,6 +6810,78 @@ static void mvpp2_acpi_start(struct mvpp2_port *port)
                          SPEED_UNKNOWN, DUPLEX_UNKNOWN, false, false);
 }

+static int mvpp2_port_fill_pcs(struct phylink_config *config,
+                              struct phylink_pcs **available_pcs,
+                              unsigned int num_possible_pcs)
+{
+       struct mvpp2_port *port = mvpp2_phylink_to_port(config);
+
+       available_pcs[0] = &port->pcs_gmac;
+
+       if (mvpp2_port_supports_xlg(port)) {
+               if (num_possible_pcs < 2)
+                       return -EINVAL;
+
+               available_pcs[1] = &port->pcs_xlg;
+       }
+
+       return 0;
+}
+
+static void mvpp2_port_init_pcs_xlg(struct phylink_pcs *pcs, bool has_comphy,
+                                   phy_interface_t phy_mode)
+{
+       if (has_comphy) {
+               __set_bit(PHY_INTERFACE_MODE_5GBASER,
+                         pcs->supported_interfaces);
+               __set_bit(PHY_INTERFACE_MODE_10GBASER,
+                         pcs->supported_interfaces);
+               __set_bit(PHY_INTERFACE_MODE_XAUI,
+                         pcs->supported_interfaces);
+       } else if (phy_mode == PHY_INTERFACE_MODE_5GBASER) {
+               __set_bit(PHY_INTERFACE_MODE_5GBASER,
+                         pcs->supported_interfaces);
+       } else if (phy_mode == PHY_INTERFACE_MODE_10GBASER) {
+               __set_bit(PHY_INTERFACE_MODE_10GBASER,
+                         pcs->supported_interfaces);
+       } else if (phy_mode == PHY_INTERFACE_MODE_XAUI) {
+               __set_bit(PHY_INTERFACE_MODE_XAUI,
+                         pcs->supported_interfaces);
+       }
+}
+
+static void mvpp2_port_init_pcs_gmac(struct phylink_pcs *pcs, bool has_comphy,
+                                    phy_interface_t phy_mode)
+{
+       if (has_comphy) {
+               /* If a COMPHY is present, we can support any of the
+                * serdes modes and switch between them.
+                */
+               __set_bit(PHY_INTERFACE_MODE_SGMII,
+                         pcs->supported_interfaces);
+               __set_bit(PHY_INTERFACE_MODE_1000BASEX,
+                         pcs->supported_interfaces);
+               __set_bit(PHY_INTERFACE_MODE_2500BASEX,
+                         pcs->supported_interfaces);
+       } else if (phy_mode == PHY_INTERFACE_MODE_2500BASEX) {
+               /* No COMPHY, with only 2500BASE-X mode supported */
+               __set_bit(PHY_INTERFACE_MODE_2500BASEX,
+                         pcs->supported_interfaces);
+       } else if (phy_mode == PHY_INTERFACE_MODE_1000BASEX ||
+                  phy_mode == PHY_INTERFACE_MODE_SGMII) {
+               /* No COMPHY, we can switch between 1000BASE-X and SGMII
+                */
+               __set_bit(PHY_INTERFACE_MODE_1000BASEX,
+                         pcs->supported_interfaces);
+               __set_bit(PHY_INTERFACE_MODE_SGMII,
+                         pcs->supported_interfaces);
+       }
+
+       /* RGMII and MII are still routed through the gmac PCS */
+       phy_interface_set_rgmii(pcs->supported_interfaces);
+       __set_bit(PHY_INTERFACE_MODE_MII, pcs->supported_interfaces);
+}
+
 /* In order to ensure backward compatibility for ACPI, check if the port
  * firmware node comprises the necessary description allowing to use phylink.
  */
@@ -7082,28 +7141,17 @@ static int mvpp2_port_probe(struct platform_device *pdev,
                        port->phylink_config.mac_capabilities |=
                                MAC_SYM_PAUSE | MAC_ASYM_PAUSE;

-               if (mvpp2_port_supports_xlg(port)) {
-                       /* If a COMPHY is present, we can support any of
-                        * the serdes modes and switch between them.
-                        */
-                       if (comphy) {
-                               __set_bit(PHY_INTERFACE_MODE_5GBASER,
-                                         port->phylink_config.supported_interfaces);
-                               __set_bit(PHY_INTERFACE_MODE_10GBASER,
-                                         port->phylink_config.supported_interfaces);
-                               __set_bit(PHY_INTERFACE_MODE_XAUI,
-                                         port->phylink_config.supported_interfaces);
-                       } else if (phy_mode == PHY_INTERFACE_MODE_5GBASER) {
-                               __set_bit(PHY_INTERFACE_MODE_5GBASER,
-                                         port->phylink_config.supported_interfaces);
-                       } else if (phy_mode == PHY_INTERFACE_MODE_10GBASER) {
-                               __set_bit(PHY_INTERFACE_MODE_10GBASER,
-                                         port->phylink_config.supported_interfaces);
-                       } else if (phy_mode == PHY_INTERFACE_MODE_XAUI) {
-                               __set_bit(PHY_INTERFACE_MODE_XAUI,
-                                         port->phylink_config.supported_interfaces);
-                       }
+               if (!mvpp2_port_supports_xlg(port))
+                       port->phylink_config.num_possible_pcs = 1;
+               else
+                       port->phylink_config.num_possible_pcs = 2;
+
+               port->phylink_config.fill_available_pcs = mvpp2_port_fill_pcs;

+               mvpp2_port_init_pcs_xlg(&port->pcs_xlg, comphy, phy_mode);
+               mvpp2_port_init_pcs_gmac(&port->pcs_gmac, comphy, phy_mode);
+
+               if (mvpp2_port_supports_xlg(port)) {
                        if (comphy)
                                port->phylink_config.mac_capabilities |=
                                        MAC_10000FD | MAC_5000FD;
@@ -7115,35 +7163,16 @@ static int mvpp2_port_probe(struct platform_device *pdev,
                                        MAC_10000FD;
                }

-               if (mvpp2_port_supports_rgmii(port)) {
-                       phy_interface_set_rgmii(port->phylink_config.supported_interfaces);
-                       __set_bit(PHY_INTERFACE_MODE_MII,
-                                 port->phylink_config.supported_interfaces);
-               }
+               phy_interface_copy(port->phylink_config.pcs_interfaces,
+                                  port->pcs_gmac.supported_interfaces);

-               if (comphy) {
-                       /* If a COMPHY is present, we can support any of the
-                        * serdes modes and switch between them.
-                        */
-                       __set_bit(PHY_INTERFACE_MODE_SGMII,
-                                 port->phylink_config.supported_interfaces);
-                       __set_bit(PHY_INTERFACE_MODE_1000BASEX,
-                                 port->phylink_config.supported_interfaces);
-                       __set_bit(PHY_INTERFACE_MODE_2500BASEX,
-                                 port->phylink_config.supported_interfaces);
-               } else if (phy_mode == PHY_INTERFACE_MODE_2500BASEX) {
-                       /* No COMPHY, with only 2500BASE-X mode supported */
-                       __set_bit(PHY_INTERFACE_MODE_2500BASEX,
-                                 port->phylink_config.supported_interfaces);
-               } else if (phy_mode == PHY_INTERFACE_MODE_1000BASEX ||
-                          phy_mode == PHY_INTERFACE_MODE_SGMII) {
-                       /* No COMPHY, we can switch between 1000BASE-X and SGMII
-                        */
-                       __set_bit(PHY_INTERFACE_MODE_1000BASEX,
-                                 port->phylink_config.supported_interfaces);
-                       __set_bit(PHY_INTERFACE_MODE_SGMII,
-                                 port->phylink_config.supported_interfaces);
-               }
+               if (mvpp2_port_supports_xlg(port))
+                       phy_interface_or(port->phylink_config.pcs_interfaces,
+                                        port->phylink_config.pcs_interfaces,
+                                        port->pcs_xlg.supported_interfaces);
+
+               phy_interface_copy(port->phylink_config.supported_interfaces,
+                                  port->phylink_config.pcs_interfaces);

                phylink = phylink_create(&port->phylink_config, port_fwnode,
                                         phy_mode, &mvpp2_phylink_ops);
--
2.55.0



-- 
	Ansuel

  reply	other threads:[~2026-08-09 17:50 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-07 13:12 [PATCH net-next v11 00/12] net: pcs: Introduce support for fwnode PCS Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 01/12] net: phylink: keep and use MAC supported_interfaces in phylink struct Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 02/12] net: phylink: introduce internal phylink PCS handling Christian Marangi
2026-08-09 17:25   ` Andrew Lunn
2026-08-07 13:12 ` [PATCH net-next v11 03/12] net: pcs: implement Firmware node support for PCS driver Christian Marangi
2026-08-07 20:29   ` Randy Dunlap
2026-08-07 20:49     ` Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 04/12] net: phylink: save phylink instance fwnode on phylink_create Christian Marangi
2026-08-09 17:29   ` Andrew Lunn
2026-08-07 13:12 ` [PATCH net-next v11 05/12] net: phylink: support PCS provider release Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 06/12] net: phylink: support late PCS provider attach Christian Marangi
2026-08-09 17:35   ` Andrew Lunn
2026-08-07 13:12 ` [PATCH net-next v11 07/12] net: Document PCS subsystem Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 08/12] MAINTAINERS: add myself as PCS subsystem maintainer Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 09/12] net: phylink: add .pcs_link_down PCS OP Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 10/12] dt-bindings: net: pcs: Document support for Airoha Ethernet PCS Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 11/12] net: pcs: airoha: add PCS driver for Airoha AN7581 SoC Christian Marangi
2026-08-07 13:12 ` [PATCH net-next v11 12/12] net: airoha: add phylink support Christian Marangi
2026-08-09 17:44 ` [PATCH net-next v11 00/12] net: pcs: Introduce support for fwnode PCS Andrew Lunn
2026-08-09 17:49   ` Christian Marangi [this message]
2026-08-09 20:43     ` Andrew Lunn
2026-08-09 20:59       ` Christian Marangi

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=6a78bdc6.89544747.3fcae.b8e0@mx.google.com \
    --to=ansuelsmth@gmail.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=andrew@lunn.ch \
    --cc=conor+dt@kernel.org \
    --cc=corbet@lwn.net \
    --cc=davem@davemloft.net \
    --cc=devicetree@vger.kernel.org \
    --cc=edumazet@google.com \
    --cc=hkallweit1@gmail.com \
    --cc=horms@kernel.org \
    --cc=justinstitt@google.com \
    --cc=krzk+dt@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=linux@armlinux.org.uk \
    --cc=llvm@lists.linux.dev \
    --cc=lorenzo@kernel.org \
    --cc=maxime.chevallier@bootlin.com \
    --cc=morbo@google.com \
    --cc=nathan@kernel.org \
    --cc=ndesaulniers@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=p.zabel@pengutronix.de \
    --cc=pabeni@redhat.com \
    --cc=robh@kernel.org \
    --cc=skhan@linuxfoundation.org \
    /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