>> > > * 2) check for sudden death of the NIC: >> > > * It seems that a reference set for this chip went out with >> incorrect info, >> > > * and there exist boards that aren't quite right. An >> unexpected voltage >> > > * drop can cause the PHY to get itself in a weird state >> (basically reset). >> > > * NOTE: this only seems to affect revC chips. >> >> > Code commented out and NIC is working OK. Strange. >> > eth0: DSPCFG accepted after 0 usec. >> > eth0: link up. >> > eth0: Setting full-duplex based on negotiated link capability. >> > dspcfg = 0x00000000 np->dspcfg = 0x00005060 >> >> Oh, that's entertaining. I have to confess that I've never seen an that >> triggered the workaround before - adding the maintainer, Tim Hockin, who >> may be able to shed some light on the expected behaviour here? > > It's been quite a while since I dealt with this issue, so I am going > on faulty memory. A particular reference design for this chip had bad > resistor values, or something similar. That caused the chip to get > very very confused and need a reset. Can You send me documentation? I can't find anything in datasheet. I will replace bad resitors with correct ones. > So the driver is finding your chip to be hosed over and over again. > dspcfg = 0x000000 is bad. I'd be very surprised if you don't get > other wierdness - bad performance or noise or who knows what. No. It is much better. Much less packets need to be retransmitted. I was blaming w3cache.tkdami.net earlier. > You could take out the error message and just let the driver do it's > thing, or you can try to run with that logic removed. But I'd measure > both and see what they do. Specifically - look for packet errors. With code commented out I have 1 error / 30000 transmitted packets from DP83815C. I have 1 error / 100000 transmitted packets to DP83815C. Maybe it works at all because I have short cable, only 10m long. I don't remember any errors with plain 2.6.21.1. > Tim Rafał
---------------------------------------------------------------------- NIE KUPUJ!!! ...zanim nie porownasz cen >> http://link.interia.pl/f1a5e - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html