On Tue, Feb 7, 2017 at 8:20 AM, Tariq Toukan [off-list ref] wrote:
On 07/02/2017 5:02 AM, Eric Dumazet wrote:
quoted
Same technique than some Intel drivers, for arches where PAGE_SIZE = 4096
In most cases, pages are reused because they were consumed
before we could loop around the RX ring.
This might not be true when multiple streams are handled in the same ring.
As the process time in stack will be longer.
We will test this scenario as well.
Sure thing. I already did ;)
I had a local patch adding a new "ethtool -S ... | grep
rx_page_allocs" counter.
I could probably add it to the series. (I noticed that
rx_alloc_failed was a dead counter)
Thanks.