Re: Fix for kern/20700

[buhrow%lothlorien.nfbcal.org@localhost (Brian Buhrow)]

        Hello.  I think a patch which restricts autonegotiation to 1000-base-T
when  1000-base-T is requested will cover all the cases I'm aware of:

1.  10-base-T and 10-base-T full-duplex (works with media auto and
hard-coded)

2.  100-base-T and 100-base-T full-duplex (works with media auto and
hard-coded)

3.  1000-base-T and 1000-baseT full-duplex (works with auto, and, with
patches for 2070, with hard-coded media)

4.  1000-base-Sx and 1000-Base-SX full-duplex (works with media auto and
hard-coded up as well.)

        The problem we have right now is that 1000-base-T requires
autonegotiation be enabled, even if you're only willing to negotiate
1000-base-T.  I'm working on patches to address this issue, and I believe
all other bases are currently covered and working.
-thanks
-Brian

On Dec 4,  6:51pm, Manuel Bouyer wrote:
} Subject: Re: Fix for kern/20700
} On Thu, Dec 01, 2011 at 09:32:53PM +0000, David Laight wrote:
} > [...]
} > 
} > At a guess it isn't setting the ANAR (autonegotiate advertise register)
} > correctly.
} > 
} > I've long been an advocate of constraining the 'media' by limiting
} > the ANAR values - rather then fixing the mode.
} > IIRC the ANAR register is one of the standard 8 registers (if the
} > original definition has enough bits for Ge and 10Ge).
} 
} There are cases where we need to fix the mode and not use autoneg.
} E.g. early 100Mbs cisco devices didn't negotiate properly, the only
} way to get things working properly was to force media on both ends.
} 
} If we only set the ANAR negotiation may still be failing.
} Maybe we should introduce new media types or options to specify
} if we want fixed media with or without autoneg ?
} 
} For kern/20700, keeping autoneg and limiting ANAR to 1000baseT
} should be enough.
} 
} -- 
} Manuel Bouyer <bouyer%antioche.eu.org@localhost>
}      NetBSD: 26 ans d'experience feront toujours la difference
} --
>-- End of excerpt from Manuel Bouyer