[vortex] The mysteries of on-chip MII.

Bill Cattey wdc@MIT.EDU
Thu Mar 21 13:00:01 2002


The values for the EEPROM that I set were ones I hard-coded into
a modified vortex-diag to set.  They are the values that the systems
arrived with.  Setting other values did not cause the problem.

Stated another way:

I'll set the values you suggest and see if the vanilla 1.1.16 driver
will fail to find the MII's.

But the test case I needed was: "Will the 3c59x driver initialize these
boxes with the EEPROM values they arrived with from the factory
that were pathological."

I will run the other tests you propose.

One thing I can tell you is that issuing:

	mii-diag -A 100baseTx-FD

three times, the speed got better.  But this is the same behavior I get
from the unmodified 1.1.16 that complained of no MII's in the original
symptoms.

That one-minute timer thing is probably the reason why it took me 3 tries.
It could have been that the autonegotiate didn't really take effect
until one minute after I issued mii-diag.

-wdc