Reshape the MDIO settle fix before proposing it upstream #19

New issue

Open

opened 2026-08-27 23:22:26 +00:00 by tiagoagueda · 0 comments

tiagoagueda commented

2026-08-27 23:22:26 +00:00

Owner

The working code pokes dwmac1000 MII_ADDR/MII_DATA directly from stmmac_dvr_probe().
It is correct and it fixed a real bug — 40 ms of settle measured on a cold boot, 0 ms warm —
but it is a measurement harness, not a patch.

Proper form is a retry in the sunxi glue, or a settle property. See the end of
34-ethernet-cold-boot.md.

Done when

reworked into a form that could be sent to netdev without embarrassment

The working code pokes dwmac1000 `MII_ADDR`/`MII_DATA` directly from `stmmac_dvr_probe()`. It is correct and it fixed a real bug — 40 ms of settle measured on a cold boot, 0 ms warm — but it is a measurement harness, not a patch. Proper form is a retry in the sunxi glue, or a settle property. See the end of `34-ethernet-cold-boot.md`. **Done when** - [ ] reworked into a form that could be sent to netdev without embarrassment

tiagoagueda added the

upstream

P3-normal

labels

2026-08-27 23:22:26 +00:00

tiagoagueda referenced this issue

2026-08-28 05:59:30 +00:00

U-Boot: no MAC driver is built — enable the GMAC to unlock TFTP boot #39

tiagoagueda referenced this issue

2026-08-30 12:57:16 +00:00

eth0 does not come up after a cold boot: networkd gives up permanently #7

tiagoagueda referenced this issue