Hi Kyle, > From: Kyle Larose [mailto:eomereadig at gmail.com] > On Wed, Jan 20, 2016 at 9:28 AM, Harry van Haaren > <harry.van.haaren at intel.com> wrote: > > This patch exposes link duplex, speed, and status via the > > existing xstats API. > > I'm slightly confused by this. Why are we exposing operational > properties of the chip through an API which I thought was primarily > targeting statistics?
In a fault-detection situation, link state is a good item to monitor - just like the rest of the statistics on the NIC. > When I think of statistics and a NIC, I think of > values which are monotonically increasing. I think of values that are > derived primary from the packets flowing through the system. I do not > think of link state, link speed and duplex, which have nothing to do > with packets, and are not monotonic. Link state, and speed seem a good fit to me. I'll admit I'm not sure about duplex, and would be happy to respin the patch without duplex if the community would prefer that. > Should we not have a separate API to get this type information? I > mean, just because we have a generic "string to uint64_t" map doesn't > mean we should toss in anything that can fit into a uin64_t. In theory we could create a new API for this, but I think the current xstats API is a good fit for exposing this info, so why create extra APIs? As a client of the DPDK API, I would prefer more statistics in a single API than have to research and implement two or more APIs to retrieve the information to monitor. I'm working on exposing keep-alive statistics using an xstats style API, I'll the patches later today so we can discuss them too. Regards, -Harry
