Re: [PATCH net-next v2] tcp: reduce cpu usage under tcp memory pressure when SO_SNDBUF is set

[Jason Baron <jbaron@akamai.com>]

On 08/11/2015 10:49 AM, Eric Dumazet wrote:
> On Tue, 2015-08-11 at 14:38 +0000, Jason Baron wrote:
>> From: Jason Baron <jbaron@akamai.com>
> 
>> In my testing, this brought a single threaad's cpu usage down from 100% to ~1%
>> while maintaining the same level of throughput.
>>
> 
> Hi Jason. Could you give more details on this test ?
> 
> How many flows are competing ?
> 
>

Yes, so the test case I'm using to test against is somewhat contrived.
In that I am simply allocating around 40,000 sockets that are idle to
create a 'permanent' memory pressure in the background. Then, I have
just 1 flow that sets SO_SNDBUF, which results in the: poll(), write() loop.

That said, we encountered this issue initially where we had 10,000+
flows and whenever the system would get into memory pressure, we would
see all the cpus spin at 100%.

So the testcase I wrote, was just a simplistic version for testing. But
I am going to try and test against the more realistic workload where
this issue was initially observed.

Thanks,

-Jason