Re: [PATCH net] ixgbe: check return value of napi_complete_done()

On 9/21/18 6:33 AM, Eric Dumazet wrote:

On 09/21/2018 12:17 AM, Song Liu wrote:

quoted

quoted

On Sep 20, 2018, at 4:49 PM, Eric Dumazet [off-list ref] wrote:



On 09/20/2018 04:43 PM, Song Liu wrote:

quoted

quoted

I tried to totally skip ndo_poll_controller() here. It did avoid hitting
the issue. However, netpoll will drop (fail to send) more packets.

Why is it failing ?

If you are under high memory pressure, then maybe if you absolutely want memory to send
netpoll packets, you want to grab all NAPI contexts as a way to prevent other cpus
from feeding incoming packets to the host and add more memory pressure ;)

I did the test with Eric's latest patch (and disable ndo_poll_controller
in driver). The result didn't show significant increase in drop packets.
I guess packet drops in my earlier test was caused by some other changes
I mixed there.

So I think this patch does fix the issue. Thanks Eric!

Great, this is awesome.

I will prepare a patch series for net tree.

The core infrastructure is just better at being able to drain TX completions
without risking stealing the NAPI context forever.

should we remove ndo_poll_controller then?
My understanding that the patch helps by not letting
drivers do napi_schedule() for all queues into this_cpu, right?
But most of the drivers do exactly that in their ndo_poll_controller
implementations. Means most of the drivers will experience
this nasty behavior.