is 2.6.6-1.435.2.3. With 2.6.7-1.494.2.2 and 2.6.8-1.521 it would
fail at bring-up with the dreaded HW autonegotiation failed.
2.6.6-1.435.2.3 was rock solid BTW. I also took the latest tg3 from Linus' BK tree
and compiled it for 2.6.8-1.521 but that didn't help. (This system doesn't have >4GB
RAM - seem to remember that as being a problem on bugzilla for RHEL users though supposedly
solved in U4.)
Is bcm5700 still an execrable abomination ;-))? (There is a very recent driver July 28 and even the IBM HS20 blade site points to this one) - anyway out of desperation I tried the shiny new 7.1.35 bcm5700 and it brought the NICs up nicely on 2.6.87-1.521.
Bugzilla on tg3 has such a variety of problems and some seem similar to mine -
basically can't auto-neg or bring up the interface. But some of the bug reporters
systems are RHEL 2.4.xx kernels.
Should we try to incorporate bcm5700 in parallel with tg3 - at least it gives
us an alternative?
-richard-
- : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
