>>>>> "Shmulik" == Hen, Shmulik <shmulik.hen@intel.com> writes: Shmulik> However, Shmulik> I'm pretty sure that the acenic IS broken in that aspect since Shmulik> the value of the register that holds the link status and link Shmulik> speeds bits, and which is used to determine what values to fill Shmulik> in the EthTool fields, does not reflect the correct status when Shmulik> connecting/disconnecting the link and moving between different Shmulik> port types (the debug prints I added were in the driver's code, Shmulik> not in the EthTool application, and they showed an almost fixed Shmulik> value). The changes I made seemed to improve the situation, but Shmulik> only a little. What I was looking for is someone who witnessed Shmulik> the described behavior and may have found a solution (that's Shmulik> why I posted to the acenic list first). I wont rule out that the acenic driver is not fully uptodate on ethtool support. I think Alteon changed the reporting API at some point. I added the ethtool support to the acenic driver long ago, back in the days when ethtool only supported the Sparc drivers hence there might be a couple of things that have changed in that API since. I might get around to looking at it, but it's not on the very top of my priority list right now. If you come up with a patch I shall gladly look at it though. Jes - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org
