svn commit: r302099 - head/sys/netinet

[Bjoern A. Zeeb]

On 23 Jun 2016, at 8:17, Marko Zec wrote:

> On Thu, 23 Jun 2016 00:34:03 +0000
> "Bjoern A. Zeeb" <bz at freebsd.org> wrote:
>
>> Author: bz
>> Date: Thu Jun 23 00:34:03 2016
>> New Revision: 302099
>> URL: https://svnweb.freebsd.org/changeset/base/302099
>>
>> Log:
>>   Check the V_tcbinfo.ipi_count to hit 0 before doing the full TCP
>> cleanup. That way timers can finish cleanly and we do not gamble with
>> a DELAY().
>>   Reviewed by:		gnn, jtl
>>   Approved by:		re (gjb)
>>   Obtained from:		projects/vnet
>>   MFC after:		2 weeks
>>   Sponsored by:		The FreeBSD Foundation
>>   Differential Revision:	https://reviews.freebsd.org/D6923
>
> As much as this change is welcome, it unnecesarily introduces a
> mandatory 100 ms delay on each vnet teardown, which I already pointed
> out in a comment to r301601 two weeks ago, which remained unanswered,

sorry about that.   While VNET teardown is a non-criticial slow path and
the pause doesn’t really matter a lot (unless you need the resources to be
freed)  you actually made me go an “benchmark” this in terms
of how often my test cases do run into the pause after your change. It was
indeed very rare.


> along with the question why a delay of 100 ms was introduced here, when
> before r302099 the delay was only a single clock tick?  And furthermore
> the delay computation expresion here is not style(9) compliant...
>
> Hence, please rectify the above objections, perhaps by something like:

Done.   Thanks a lot!

/bz