I recently learnt that barebox Ethernet drivers need to call ->halt
themselves, either directly or indirectly.
The first patch does this. Sascha, do you do fixups into master? If so,
you could fix the first commit up into my previous patch.
After applying the patch and doing a round of boot bnet, the first MDIO
access during probe times out:
ERROR: <NULL>: MDIO not idle at entry
This is fixed by the other two commits.
The driver isn't in a release yet, so I hope they could join the driver
in master.
Cheers
Ahmad
Ahmad Fatoum (3):
fixup! net: designware: eqos: stop DMA on halt
net: designware: eqos: enable clocks before mdio_register
net: designware: eqos: fix NULL pointer use in dev_printf
drivers/net/designware_eqos.c | 18 ++++++------
drivers/net/designware_eqos.h | 4 ---
drivers/net/designware_stm32.c | 33 +---------------------
drivers/net/designware_tegra186.c | 47 ++++++-------------------------
4 files changed, 20 insertions(+), 82 deletions(-)
--
2.20.1
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
