Date: Sat, 28 Jan 2012 22:38:57 +0100 From: Marius Strobl <marius@alchemy.franken.de> To: Randy Bush <randy@psg.com> Cc: FreeBSD Stable <freebsd-stable@freebsd.org>, FreeBSD Net <net@freebsd.org> Subject: Re: 9-stable - ifmedia_set: no match for 0x0/0xfffffff Message-ID: <20120128213857.GB39861@alchemy.franken.de> In-Reply-To: <m239aztxcj.wl%randy@psg.com> References: <m2y5svb4ra.wl%randy@psg.com> <m2sjj3b1m3.wl%randy@psg.com> <m2r4ynb0ys.wl%randy@psg.com> <20120128121830.GA38513@alchemy.franken.de> <m24nvg57lv.wl%randy@psg.com> <20120128131830.GZ44286@alchemy.franken.de> <m239aztxcj.wl%randy@psg.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jan 29, 2012 at 04:56:28AM +0900, Randy Bush wrote: > > Hrm, the problem apparently is that while when probing, the PHY > > still knows about the media it supports, it just has forgotten > > about it after the reset during attach. There was a change prior > > to 8.2 which would turn this from silently being ignored (which > > generally might or might not work) into resulting what you see > > now (the upper layers arguably shouldn't trigger a panic in this > > case though). I can't remember a change to either bge(4) or > > brgphy(4) between 8.2 and now which could trigger this though. > > Have you tried to set the loader-tunable hw.bge.allow_asf to 0? > > The default for that option still is different between 8 and 9+. > > it no longer panics when booting, but the interface comes up not seeing > carrier > > bge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500 > options=8009b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,LINKSTATE> > ether 00:30:48:82:11:a2 > inet 198.180.150.1 netmask 0xffffff80 broadcast 198.180.150.127 > inet6 fe80::230:48ff:fe82:11a2%bge0 prefixlen 64 scopeid 0x1 > inet6 2001:418:8006::1 prefixlen 64 > inet 198.180.150.2 netmask 0xffffffff broadcast 198.180.150.2 > nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL> > media: Ethernet 1000baseT (none) > status: no carrier > Are you sure that the other end is also forced to 1000baseT half-duplex? What happens if you set hw.bge.allow_asf to 0 and use auto-negotiation on both sides? Marius
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120128213857.GB39861>