On Wed, 2017-03-15 at 22:39 -0700, Alexei Starovoitov wrote: > when there is no room in the rx fifo the hw will increment the counter. > That's the same as oom causing alloc fails and rx ring not being replenished. > When there is nothing free in rx ring to dma the packet to, the hw will > increment that counter.
After 4096 frames were lost buy linux mlx4 driver, then hw will finally catch up and start increasing its own hw counter. While you pretended that you received millions of frames and wanted that not a single one could be dropped without a report. This is simply not true. I do not believe I will continue the discussion. We can not really agree it seems, so why annoying all netdev readers.
