* Re: [Bloat] fq_codel on macosx
2018-09-08 15:24 ` Dave Taht
@ 2018-09-08 15:35 ` Sebastian Moeller
2018-09-08 15:54 ` Dave Taht
2018-09-08 21:17 ` Pete Heist
2 siblings, 0 replies; 9+ messages in thread
From: Sebastian Moeller @ 2018-09-08 15:35 UTC (permalink / raw)
To: Dave Täht; +Cc: bloat
I ran a quick rrul_cs8 test and I see some counters are actually increasing, though most stay at zero (which might be okay, I have no shaper configured on the macbook):
macbook:~ user$ sudo netstat -I en0 -qq
en0:
[ sched: FQ_CODEL qlength: 0/128 ]
[ pkts: 0 bytes: 0 dropped pkts: 480 bytes: 77809 ]
=====================================================
[ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
[ queued pkts: 0 bytes: 0 ]
[ dequeued pkts: 223138 bytes: 77920371 ]
[ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
[ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
[ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
[ flows total: 0 new: 0 old: 0 ]
[ throttle on: 0 off: 0 drop: 0 ]
=====================================================
[ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
[ queued pkts: 0 bytes: 0 ]
[ dequeued pkts: 1290573 bytes: 156658208 ]
[ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
[ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
[ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
[ flows total: 0 new: 0 old: 0 ]
[ throttle on: 0 off: 0 drop: 0 ]
=====================================================
[ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
[ queued pkts: 0 bytes: 0 ]
[ dequeued pkts: 6910907 bytes: 1684837312 ]
[ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
[ flow control: 1 feedback: 1 stalls: 0 failed: 0 ]
[ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
[ flows total: 0 new: 0 old: 0 ]
[ throttle on: 0 off: 0 drop: 0 ]
=====================================================
[ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
[ queued pkts: 0 bytes: 0 ]
[ dequeued pkts: 1535662 bytes: 438122731 ]
[ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
[ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
[ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
[ flows total: 0 new: 0 old: 0 ]
[ throttle on: 0 off: 0 drop: 0 ]
macbook:~ user $
I also have the impression, that VI seems to be optimized for ~2 MTUs probably to elicit one ACK per txop?
Best Regards
Sebastian
> On Sep 8, 2018, at 17:24, Dave Taht <dave.taht@gmail.com> wrote:
>
> On Sat, Sep 8, 2018 at 8:12 AM Sebastian Moeller <moeller0@gmx.de> wrote:
>>
>> Hi Dave,
>>
>> Jason Woringen just told me that macosx seems to include fq_codel now as well as some involved priority queues that I can not make much sense of. It might simply have inherited this from its freebsd underpinnings...
>
> Wow. I had no idea. I know that they enabled tcp ecn universally last
> year... but not fq_codel!! Holy cow!
>
> Same here on my air....
>
> daves-Air-3:wiki d$ netstat -I en0 -qq
>
> en0:
>
> [ sched: FQ_CODEL qlength: 0/128 ]
>
> [ pkts: 0 bytes: 0 dropped pkts: 13262
> bytes: 8469105 ]
>
> =====================================================
>
> [ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 380295 bytes: 47356766 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>
> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
> [ throttle on: 0 off: 0 drop: 0 ]
>
> =====================================================
>
> [ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 4339 bytes: 321840 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>
> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
> [ throttle on: 0 off: 0 drop: 0 ]
>
> =====================================================
>
> [ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 40497473 bytes: 11269032336 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 86 feedback: 86 stalls: 4 failed: 0 ]
>
> [ drop overflow: 3116 early: 854 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
> [ throttle on: 0 off: 0 drop: 0 ]
>
> =====================================================
>
> [ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 1553687 bytes: 358969659 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>
> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
>
>
>> en4 is 1Gbps ethernet
>> macbook:~ user$ netstat -I en4 -qq
>> en4:
>> [ sched: FQ_CODEL qlength: 0/128 ]
>> [ pkts: 100 bytes: 28920 dropped pkts: 1 bytes: 342 ]
>> =====================================================
>> [ pri: CTL (0) srv_cl: 0x480190 quantum: 600 drr_max: 8 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 91 bytes: 27706 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: RV (3) srv_cl: 0x300110 quantum: 3000 drr_max: 6 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: AV (4) srv_cl: 0x280120 quantum: 3000 drr_max: 6 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: OAM (5) srv_cl: 0x200020 quantum: 1500 drr_max: 4 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: RD (6) srv_cl: 0x180010 quantum: 1500 drr_max: 4 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 9 bytes: 1214 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: BK_SYS (9) srv_cl: 0x80090 quantum: 1500 drr_max: 2 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 0 bytes: 0 ]
>> [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> macbook:~ user$
>>
>>
>> and on wifi (en0)
>> macbook:~ user $ netstat -I en0 -qq
>> en0:
>> [ sched: FQ_CODEL qlength: 0/128 ]
>> [ pkts: 0 bytes: 0 dropped pkts: 480 bytes: 77809 ]
>> =====================================================
>> [ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 72537 bytes: 7015758 ]
>> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 1139670 bytes: 80872250 ]
>> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 6737285 bytes: 1608200668 ]
>> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>> [ flow control: 1 feedback: 1 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> =====================================================
>> [ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
>> [ queued pkts: 0 bytes: 0 ]
>> [ dequeued pkts: 1396402 bytes: 382861004 ]
>> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>> [ flows total: 0 new: 0 old: 0 ]
>> [ throttle on: 0 off: 0 drop: 0 ]
>> macbook:~ user $
>>
>>
>>
>>
>>
>>
>
>
> --
>
> Dave Täht
> CEO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-669-226-2619
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [Bloat] fq_codel on macosx
2018-09-08 15:24 ` Dave Taht
2018-09-08 15:35 ` Sebastian Moeller
@ 2018-09-08 15:54 ` Dave Taht
2018-09-08 21:17 ` Pete Heist
2 siblings, 0 replies; 9+ messages in thread
From: Dave Taht @ 2018-09-08 15:54 UTC (permalink / raw)
To: Sebastian Moeller; +Cc: bloat
[ flow control: 86 feedback: 86 stalls: 4 failed: 0 ]
[ drop overflow: 3116 early: 854 memfail: 0 duprexmt:0 ]
Not sure what this means. I think they are using fairly short queues
(which is fine given this is a wifi box),
so they are taking advantage of the fq, but tail dropping a lot.
"Early" would be my guess for codel itself
engaging. As for flow control? ecn? Dunno. I've not looked at direct
packet captures much on osx. Now I shall.
I note I recently added something to increase the signal strength in
fq_codel's bulk dropper, based on my observation that with openwrt's
shorter queues (1000 packets) it still tail dropped a lot, and you
could clearly see a bifurcation in flow behaviors when that started
happening.
https://github.com/dtaht/fq_codel_fast
On Sat, Sep 8, 2018 at 8:24 AM Dave Taht <dave.taht@gmail.com> wrote:
>
> On Sat, Sep 8, 2018 at 8:12 AM Sebastian Moeller <moeller0@gmx.de> wrote:
> >
> > Hi Dave,
> >
> > Jason Woringen just told me that macosx seems to include fq_codel now as well as some involved priority queues that I can not make much sense of. It might simply have inherited this from its freebsd underpinnings...
>
> Wow. I had no idea. I know that they enabled tcp ecn universally last
> year... but not fq_codel!! Holy cow!
>
> Same here on my air....
>
> daves-Air-3:wiki d$ netstat -I en0 -qq
>
> en0:
>
> [ sched: FQ_CODEL qlength: 0/128 ]
>
> [ pkts: 0 bytes: 0 dropped pkts: 13262
> bytes: 8469105 ]
>
> =====================================================
>
> [ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 380295 bytes: 47356766 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>
> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
> [ throttle on: 0 off: 0 drop: 0 ]
>
> =====================================================
>
> [ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 4339 bytes: 321840 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>
> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
> [ throttle on: 0 off: 0 drop: 0 ]
>
> =====================================================
>
> [ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 40497473 bytes: 11269032336 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 86 feedback: 86 stalls: 4 failed: 0 ]
>
> [ drop overflow: 3116 early: 854 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
> [ throttle on: 0 off: 0 drop: 0 ]
>
> =====================================================
>
> [ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
>
> [ queued pkts: 0 bytes: 0 ]
>
> [ dequeued pkts: 1553687 bytes: 358969659 ]
>
> [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
>
> [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
>
> [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
>
> [ flows total: 0 new: 0 old: 0 ]
>
>
>
> > en4 is 1Gbps ethernet
> > macbook:~ user$ netstat -I en4 -qq
> > en4:
> > [ sched: FQ_CODEL qlength: 0/128 ]
> > [ pkts: 100 bytes: 28920 dropped pkts: 1 bytes: 342 ]
> > =====================================================
> > [ pri: CTL (0) srv_cl: 0x480190 quantum: 600 drr_max: 8 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 91 bytes: 27706 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: RV (3) srv_cl: 0x300110 quantum: 3000 drr_max: 6 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: AV (4) srv_cl: 0x280120 quantum: 3000 drr_max: 6 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: OAM (5) srv_cl: 0x200020 quantum: 1500 drr_max: 4 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: RD (6) srv_cl: 0x180010 quantum: 1500 drr_max: 4 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 9 bytes: 1214 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: BK_SYS (9) srv_cl: 0x80090 quantum: 1500 drr_max: 2 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 0 bytes: 0 ]
> > [ budget: 0 target qdelay: 5.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > macbook:~ user$
> >
> >
> > and on wifi (en0)
> > macbook:~ user $ netstat -I en0 -qq
> > en0:
> > [ sched: FQ_CODEL qlength: 0/128 ]
> > [ pkts: 0 bytes: 0 dropped pkts: 480 bytes: 77809 ]
> > =====================================================
> > [ pri: VO (1) srv_cl: 0x400180 quantum: 600 drr_max: 8 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 72537 bytes: 7015758 ]
> > [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: VI (2) srv_cl: 0x380100 quantum: 3000 drr_max: 6 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 1139670 bytes: 80872250 ]
> > [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: BE (7) srv_cl: 0x0 quantum: 1500 drr_max: 4 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 6737285 bytes: 1608200668 ]
> > [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
> > [ flow control: 1 feedback: 1 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > =====================================================
> > [ pri: BK (8) srv_cl: 0x100080 quantum: 1500 drr_max: 2 ]
> > [ queued pkts: 0 bytes: 0 ]
> > [ dequeued pkts: 1396402 bytes: 382861004 ]
> > [ budget: 0 target qdelay: 10.00 msec update interval:100.00 msec ]
> > [ flow control: 0 feedback: 0 stalls: 0 failed: 0 ]
> > [ drop overflow: 0 early: 0 memfail: 0 duprexmt:0 ]
> > [ flows total: 0 new: 0 old: 0 ]
> > [ throttle on: 0 off: 0 drop: 0 ]
> > macbook:~ user $
> >
> >
> >
> >
> >
> >
>
>
> --
>
> Dave Täht
> CEO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-669-226-2619
--
Dave Täht
CEO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-669-226-2619
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [Bloat] fq_codel on macosx
2018-09-08 15:24 ` Dave Taht
2018-09-08 15:35 ` Sebastian Moeller
2018-09-08 15:54 ` Dave Taht
@ 2018-09-08 21:17 ` Pete Heist
2018-09-08 21:58 ` Dave Taht
2 siblings, 1 reply; 9+ messages in thread
From: Pete Heist @ 2018-09-08 21:17 UTC (permalink / raw)
To: Dave Taht; +Cc: Sebastian Moeller, bloat
[-- Attachment #1: Type: text/plain, Size: 718 bytes --]
> On Sep 8, 2018, at 5:24 PM, Dave Taht <dave.taht@gmail.com> wrote:
>
> On Sat, Sep 8, 2018 at 8:12 AM Sebastian Moeller <moeller0@gmx.de <mailto:moeller0@gmx.de>> wrote:
>>
>> Hi Dave,
>>
>> Jason Woringen just told me that macosx seems to include fq_codel now as well as some involved priority queues that I can not make much sense of. It might simply have inherited this from its freebsd underpinnings...
>
> Wow. I had no idea. I know that they enabled tcp ecn universally last
> year... but not fq_codel!! Holy cow!
Woah is right. It’s not in OS/X 10.11, but I also see it in 10.13, so it was either added in 10.12 or 10.13.
fq_codel: good enough for Apple, but not systemd. :)
Pete
[-- Attachment #2: Type: text/html, Size: 4931 bytes --]
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Bloat] fq_codel on macosx
2018-09-08 21:17 ` Pete Heist
@ 2018-09-08 21:58 ` Dave Taht
2018-09-09 18:49 ` Toke Høiland-Jørgensen
0 siblings, 1 reply; 9+ messages in thread
From: Dave Taht @ 2018-09-08 21:58 UTC (permalink / raw)
To: Pete Heist; +Cc: Sebastian Moeller, bloat
On Sat, Sep 8, 2018 at 2:17 PM Pete Heist <pete@heistp.net> wrote:
>
>
> On Sep 8, 2018, at 5:24 PM, Dave Taht <dave.taht@gmail.com> wrote:
>
> On Sat, Sep 8, 2018 at 8:12 AM Sebastian Moeller <moeller0@gmx.de> wrote:
>
>
> Hi Dave,
>
> Jason Woringen just told me that macosx seems to include fq_codel now as well as some involved priority queues that I can not make much sense of. It might simply have inherited this from its freebsd underpinnings...
>
>
> Wow. I had no idea. I know that they enabled tcp ecn universally last
> year... but not fq_codel!! Holy cow!
>
>
> Woah is right. It’s not in OS/X 10.11, but I also see it in 10.13, so it was either added in 10.12 or 10.13.
>
> fq_codel: good enough for Apple, but not systemd. :)
>
Well I should probably write more on that subject than I already have.
> Pete
>
--
Dave Täht
CEO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-669-226-2619
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Bloat] fq_codel on macosx
2018-09-08 21:58 ` Dave Taht
@ 2018-09-09 18:49 ` Toke Høiland-Jørgensen
2018-09-09 18:51 ` Jonathan Morton
0 siblings, 1 reply; 9+ messages in thread
From: Toke Høiland-Jørgensen @ 2018-09-09 18:49 UTC (permalink / raw)
To: Dave Taht, Pete Heist; +Cc: bloat
Dave Taht <dave.taht@gmail.com> writes:
> On Sat, Sep 8, 2018 at 2:17 PM Pete Heist <pete@heistp.net> wrote:
>>
>>
>> On Sep 8, 2018, at 5:24 PM, Dave Taht <dave.taht@gmail.com> wrote:
>>
>> On Sat, Sep 8, 2018 at 8:12 AM Sebastian Moeller <moeller0@gmx.de> wrote:
>>
>>
>> Hi Dave,
>>
>> Jason Woringen just told me that macosx seems to include fq_codel now as well as some involved priority queues that I can not make much sense of. It might simply have inherited this from its freebsd underpinnings...
>>
>>
>> Wow. I had no idea. I know that they enabled tcp ecn universally last
>> year... but not fq_codel!! Holy cow!
>>
>>
>> Woah is right. It’s not in OS/X 10.11, but I also see it in 10.13, so it was either added in 10.12 or 10.13.
>>
>> fq_codel: good enough for Apple, but not systemd. :)
>>
>
> Well I should probably write more on that subject than I already have.
Erm, fq_codel has been the default in systemd for quite a while now?
-Toke
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Bloat] fq_codel on macosx
2018-09-09 18:49 ` Toke Høiland-Jørgensen
@ 2018-09-09 18:51 ` Jonathan Morton
2018-09-09 20:38 ` Dave Taht
0 siblings, 1 reply; 9+ messages in thread
From: Jonathan Morton @ 2018-09-09 18:51 UTC (permalink / raw)
To: Toke Høiland-Jørgensen; +Cc: Dave Taht, Pete Heist, bloat
> On 9 Sep, 2018, at 9:49 pm, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>
> Erm, fq_codel has been the default in systemd for quite a while now?
Yes, but there's an issue open to change it to sch_fq, which is apparently being taken seriously.
- Jonathan Morton
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Bloat] fq_codel on macosx
2018-09-09 18:51 ` Jonathan Morton
@ 2018-09-09 20:38 ` Dave Taht
0 siblings, 0 replies; 9+ messages in thread
From: Dave Taht @ 2018-09-09 20:38 UTC (permalink / raw)
To: Jonathan Morton; +Cc: Toke Høiland-Jørgensen, Pete Heist, bloat
On Sun, Sep 9, 2018 at 11:51 AM Jonathan Morton <chromatix99@gmail.com> wrote:
>
> > On 9 Sep, 2018, at 9:49 pm, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> >
> > Erm, fq_codel has been the default in systemd for quite a while now?
>
> Yes, but there's an issue open to change it to sch_fq, which is apparently being taken seriously.
I'm pretty sure I squashed that: https://github.com/systemd/systemd/issues/9725
That, and this: https://github.com/systemd/systemd/issues/9748
were my impetus to get the ecn-sane project off the ground.
Anyone for fq'd socket (udp) buffers in linux? A udp_sent_lowat setsockopt?
> - Jonathan Morton
>
--
Dave Täht
CEO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-669-226-2619
^ permalink raw reply [flat|nested] 9+ messages in thread