I still end up with the same error: ifconfig :SIOCSLIFNAME for ip: bge0: Invalid argument
On Sun, Jan 18, 2009 at 7:47 PM, Yong Tan <[email protected]> wrote: > Hi Joe, > > That sounds very interesting. Could you try the following steps with the > B100+ driver as root? > rem_drv bge > add_drv -i '"pci14e4,1648"' bge > reboot -- -r > > Hopefully it will work for you. Any problem, please let me know. > > Thanks, > Carson > > > Joe Little wrote: >> >> I have a supermicro-based motherboard, Supermicro H8DA8/H8DAR, in >> three separate systems with is likely an AMD-8131 chipset. It includes >> broadcom gig-e interfaces that worked fine with my B85 and earlier >> based opensolaris versioned systems. I moved up to B102, B103, and >> B104-based solutions recently and in all cases, interface bge0 is >> lost, but bge1 is retained. >> >> A plumb of bge0 results in this dmesg output: >> >> Jan 17 11:47:49 perth ip: [ID 205306 kern.error] bge0: DL_ATTACH_REQ >> failed: DL_SYSERR (errno 22) >> Jan 17 11:47:49 perth ip: [ID 670008 kern.error] bge0: DL_BIND_REQ >> failed: DL_OUTSTATE >> Jan 17 11:47:49 perth ip: [ID 328654 kern.error] bge0: >> DL_PHYS_ADDR_REQ failed: DL_OUTSTATE >> Jan 17 11:47:49 perth ip: [ID 954675 kern.error] bge0: DL_UNBIND_REQ >> failed: DL_OUTSTATE >> >> and this command line output: >> >> ifconfig: SIOCSLIFNAME for ip: bge0: Invalid argument >> >> prtconf shows bge at 1648: >> >> dev_path=/p...@0,0/pci1022,7...@a/pci15d9,1...@5,1:bge1 >> dev_link=/dev/bge1 >> >> and lspci -n shows that the system finds both interfaces, but the >> driver obviously does not configure the first: >> >> 02:05.0 0200: 14e4:1648 (rev 10) >> 02:05.1 0200: 14e4:1648 (rev 10) >> >> Putting the B85 driver in /kerner/drv/bge and rebooting instantly >> returns bge0 back into action. Erast from Nexenta gave me different in >> versions of the bge driver between B85 to B100+, and so far we haven't >> isolated where it broke, but this is obviously a regression. >> >> Is this a known issue? I'm surprised this is no where to be found one >> the list, but seeing as this was a common opteron-based supermicro >> platform I presume others are affected by it. >> > > > -- > Thanks and Regards, > Carson (Yong Tan) > Sun Microsystems China (ERI) > Email: [email protected] > Tel : (86-10)6267-3681 (x51681) > _______________________________________________ opensolaris-discuss mailing list [email protected]
