drivers: mdio_nxp_enet_qos: Add Clause 45 support. #93442
Conversation
zephyrbot
requested review from
dbaluta,
decsny,
dleach02,
maass-hamburg,
manuargue and
mmahadevan108
drivers/mdio/mdio_nxp_enet_qos.c
Outdated
| @@ -111,8 +131,7 @@ static int do_transaction(struct mdio_transaction *mdio) | |||
| return ret; | |||
| } | |||
|
|
|||
| static int nxp_enet_qos_mdio_read(const struct device *dev, | |||
| uint8_t portaddr, uint8_t regaddr, | |||
| static int nxp_enet_qos_mdio_read(const struct device *dev, uint8_t portaddr, uint8_t regaddr, | |||
| uint16_t *read_data) | |||
There was a problem hiding this comment.
should probably avoid doing these kinds of whitespace changes in the commit with the functional changes. Challenges the reviewer to stare at this more to confirm there is not functional change.
There was a problem hiding this comment.
Sorry, I’m an avid clang-format user, which automatically formats code. Apparently, the original code was checked in unformatted.
PetervdPerk-NXP
force-pushed
the
pr-enet-qos-cl45
branch
from
2b2be8d to
c4af6ca
Compare
Adds handler for C45 read/write phy transactions. Signed-off-by: Peter van der Perk <peter.vanderperk@nxp.com>
PetervdPerk-NXP
force-pushed
the
pr-enet-qos-cl45
branch
from
c4af6ca to
3254c5b
Compare
|
|
Alright, upstream break on main is fixed/reverted, CI is green again @dleach02 |
There was a problem hiding this comment.
LGTM.
Just a nit about naming in the struct mdio_transaction, you could have uint16_t regaddr that store both the u8 regaddr for c22 and the u16 regaddr for c45, and a uint8_t devaddr that stores the device addr for c45 and is not used in c22. No functional difference, just to not have two regaddr/regaddr_c45.
Adds support for CL45 enabled PHY's for the nxp_enet_qos mac.
By adding handlers for C45 read/write.