Dave Taht writes:
> I am curious as to the cpu and thruput difference between pfifo, htb +
> fqcodel v cake at these insane speeds.
Sure, actual test results at ludicrous speed is on my list :)
-Toke
___
Cake mailing list
Cake@lists.bufferbloat.net
ht
I am curious as to the cpu and thruput difference between pfifo, htb +
fqcodel v cake at these insane speeds.
On Fri, Jul 6, 2018, 3:27 AM Toke Høiland-Jørgensen wrote:
> Jonathan Morton writes:
>
> >> On 6 Jul, 2018, at 4:21 pm, Toke Høiland-Jørgensen
> wrote:
> >>
> >>> I'm handling it wit
Jonathan Morton writes:
>> On 6 Jul, 2018, at 4:21 pm, Toke Høiland-Jørgensen wrote:
>>
>>> I'm handling it without using any extra permanent state, just making
>>> use of what's already there. Take a look at latest commit.
>>
>> Yup, that also fixes the infinite loop issue. I'm fine with doin
> On 6 Jul, 2018, at 4:21 pm, Toke Høiland-Jørgensen wrote:
>
>> I'm handling it without using any extra permanent state, just making
>> use of what's already there. Take a look at latest commit.
>
> Yup, that also fixes the infinite loop issue. I'm fine with doing it
> this way; however, it doe
Jonathan Morton writes:
>> On 6 Jul, 2018, at 3:00 pm, Toke Høiland-Jørgensen wrote:
>>
>> Toke Høiland-Jørgensen writes:
>>
>>> Since sch->q.qlen can be changed from outside cake, we can't rely on it for
>>> break loops etc. So keep an internal counter that mirrors it and use that
>>> for al
> On 6 Jul, 2018, at 3:00 pm, Toke Høiland-Jørgensen wrote:
>
> Toke Høiland-Jørgensen writes:
>
>> Since sch->q.qlen can be changed from outside cake, we can't rely on it for
>> break loops etc. So keep an internal counter that mirrors it and use that
>> for all checks.
>
> Sent this as a pat
Toke Høiland-Jørgensen writes:
> Since sch->q.qlen can be changed from outside cake, we can't rely on it for
> break loops etc. So keep an internal counter that mirrors it and use that
> for all checks.
Sent this as a patch to avoid stepping on your toes if you are working
on the code now :)
-T
Since sch->q.qlen can be changed from outside cake, we can't rely on it for
break loops etc. So keep an internal counter that mirrors it and use that
for all checks.
This the issue with inifinte loops on multi-queue hardware.
Signed-off-by: Toke Høiland-Jørgensen
---
sch_cake.c | 16 +++
