diff options
| author |   Vladimir Oltean <vladimir.oltean@nxp.com> | 2023-04-18 22:01:41 +0300 |
|---|---|---|
| committer |   Jakub Kicinski <kuba@kernel.org> | 2023-04-21 20:04:09 -0700 |
| commit | f3b766d981310989d524b076065a93b8f11bcab2 (patch) | |
| tree | 735fb3bd127b78e1feaa2a165b0eb8ccebaee835 /drivers/net/phy/Makefile | |
| parent | Merge tag 'wireless-next-2023-04-21' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next (diff) | |
| download | linux-f3b766d981310989d524b076065a93b8f11bcab2.tar.xz linux-f3b766d981310989d524b076065a93b8f11bcab2.zip | |
net: phy: add basic driver for NXP CBTX PHY
The CBTX PHY is a Fast Ethernet PHY integrated into the SJA1110 A/B/C
automotive Ethernet switches.
It was hoped it would work with the Generic PHY driver, but alas, it
doesn't. The most important reason why is that the PHY is powered down
by default, and it needs a vendor register to power it on.
It has a linear memory map that is accessed over SPI by the SJA1110
switch driver, which exposes a fake MDIO controller. It has the
following (and only the following) standard clause 22 registers:
0x0: MII_BMCR
0x1: MII_BMSR
0x2: MII_PHYSID1
0x3: MII_PHYSID2
0x4: MII_ADVERTISE
0x5: MII_LPA
0x6: MII_EXPANSION
0x7: the missing MII_NPAGE for Next Page Transmit Register
Every other register is vendor-defined.
The register map expands the standard clause 22 5-bit address space of
0x20 registers, however the driver does not need to access the extra
registers for now (and hopefully never). If it ever needs to do that, it
is possible to implement a fake (software) page switching mechanism
between the PHY driver and the SJA1110 MDIO controller driver.
Also, Auto-MDIX is turned off by default in hardware, the driver turns
it on by default and reports the current status. I've tested this with a
VSC8514 link partner and a crossover cable, by forcing the mode on the
link partner, and seeing that the CBTX PHY always sees the reverse of
the mode forced on the VSC8514 (and that traffic works). The link
doesn't come up (as expected) if MDI modes are forced on both ends in
the same way (with the cross-over cable, that is).
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20230418190141.1040562-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'drivers/net/phy/Makefile')
| -rw-r--r-- | drivers/net/phy/Makefile | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index b5138066ba04..ae11bf20b46e 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -79,6 +79,7 @@ obj-$(CONFIG_MOTORCOMM_PHY) += motorcomm.o obj-$(CONFIG_NATIONAL_PHY) += national.o obj-$(CONFIG_NCN26000_PHY) += ncn26000.o obj-$(CONFIG_NXP_C45_TJA11XX_PHY) += nxp-c45-tja11xx.o +obj-$(CONFIG_NXP_CBTX_PHY) += nxp-cbtx.o obj-$(CONFIG_NXP_TJA11XX_PHY) += nxp-tja11xx.o obj-$(CONFIG_QSEMI_PHY) += qsemi.o obj-$(CONFIG_REALTEK_PHY) += realtek.o |