[PATCH 0/4 v2] Serdes: s32g: Add support for serdes subsystem

[Jan Petrous]

On Tue, Feb 03, 2026 at 05:19:13PM +0100, Vincent Guittot wrote:
> s32g SoC family includes 2 serdes subsystems which are made of one PCIe
> controller, 2 XPCS and a shared Phy. The Phy got 2 lanes that can be
> configured to output PCIe lanes and/or SGMII.
>     
> Implement PCIe phy and XPCS support.
>     
> Change since v1:
> - Fix compile_test
> - Use devm_reset_control_get_exclusive()
> - Fix s32g_serdes_phy_set_mode_ext()
> - Manage devm_clk_bulk_get_all() returns 0
> - Fix s32g_serdes_parse_lanes() error management
> - Move xpcs filein drivers/net/pcs/
> - Add pcs_inband_caps()
> - Fix functions in phylink_pcs_ops
> - Fix MAINTAINERS
> 
> 
> Vincent Guittot (4):
>   dt-bindings: serdes: s32g: Add NXP serdes subsystem
>   phy: s32g: Add serdes subsystem phy
>   phy: s32g: Add serdes xpcs subsystem
>   MAINTAINERS: Add MAINTAINER for NXP S32G Serdes driver
> 
>  .../bindings/phy/nxp,s32g-serdes.yaml         |  154 +++
>  MAINTAINERS                                   |   10 +
>  drivers/net/pcs/Makefile                      |    1 +
>  drivers/net/pcs/pcs-nxp-s32g-xpcs.c           | 1006 +++++++++++++++++
>  drivers/phy/freescale/Kconfig                 |   10 +
>  drivers/phy/freescale/Makefile                |    1 +
>  drivers/phy/freescale/phy-nxp-s32g-serdes.c   |  953 ++++++++++++++++
>  include/linux/pcs/pcs-nxp-s32g-xpcs.h         |   50 +
>  8 files changed, 2185 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/phy/nxp,s32g-serdes.yaml
>  create mode 100644 drivers/net/pcs/pcs-nxp-s32g-xpcs.c
>  create mode 100644 drivers/phy/freescale/phy-nxp-s32g-serdes.c
>  create mode 100644 include/linux/pcs/pcs-nxp-s32g-xpcs.h
> 
> -- 
> 2.43.0
> 

Hi Vincent, all,
I'm taking over the S32G SerDes/XPCS upstreaming. The effort has moved in-house
at NXP and I'll be carrying it forward, continuing from this v2 rather than
restarting from zero.

Vincent, thanks for the v1->v2 groundwork. I'll keep your authorship on the
patches that originate from your series (Co-developed-by plus your
Signed-off-by) and build on top; I'll send you v3 off-list first, as you
offered.

A v3 is in preparation and will come as an RFC, with the v2 review comments
addressed.

Vincent, if you're OK with the handoff, a short ack here would help make the
transition visible to the reviewers.

Thanks.
/Jan