From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pj1-x102e.google.com (mail-pj1-x102e.google.com [IPv6:2607:f8b0:4864:20::102e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id D283D3B29D for ; Sat, 22 Oct 2022 10:32:39 -0400 (EDT) Received: by mail-pj1-x102e.google.com with SMTP id f9-20020a17090a654900b00210928389f8so9261933pjs.2 for ; Sat, 22 Oct 2022 07:32:39 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=cc:subject:message-id:date:from:in-reply-to:references:mime-version :from:to:cc:subject:date:message-id:reply-to; bh=Xeldz3G7Y9IhI527ph64MVO4B1mGGJkF2DZiK4Unmtg=; b=Afm60/F5LUVcuiIgKunwtT5ktCG/lu7goR3vnVpWqmObCHlRwcxg0c6ruTGrbC7XcB 7T7Xo4iLKBUxQlUmo1ohjgtwfW8CYWw3ZinVKsCTuknAr4d9Ng3tP36u44TqtYIiukfz soK/SKlMwBqdpT7ywAiEF8flV2nyUB9qKN8BzRsCqIWfBPZo+wpgqZb30lZPIWK4XGC9 InzRzwHaMV3BnzFjx4Fj73B8bod3+kotXAdb9PkIrjIDqIKU1avZ3pssXAM8WwzoJelR E2h+vYHkrfBYm1T6a+xYY1eYKbzuOtdOH2kVQJQ7fSKctqYbBAqd32g05ZTsobOnqk5a 6Lhw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:subject:message-id:date:from:in-reply-to:references:mime-version :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=Xeldz3G7Y9IhI527ph64MVO4B1mGGJkF2DZiK4Unmtg=; b=a4JHbr1unuANANYDFUeSbyMbOA+aTOev14EwaDxiQ45ySGRYe6CFIlMNVhHUuoEalm anMdDxIZ6JcURebe/JXGP6TqOgx2jnHSmCMP5CgMB4ON+irRsktPqyrxDcvAHz9hgbXz vsuH49Hz5kQZW9P1moB4TwcS9DayqjXEALyVd9HavYp10tp2nri/39wFbKSSiS2KXw9m LFnL+pVZq/zrGAdJECHRjj/Tatzk65o9JGPHoo7mJ4Hng2nQdJd9j7pFhg6tA/IDTXeT xu0Nnfr1fQVZguh5VW1xVozuxGwCmuy04zGIFmopNxY3GybQjBml6HnyrCSGRA4coWs5 ZurA== X-Gm-Message-State: ACrzQf1/TDDKI2HSKap2SrZ9haytrHNLLGNDnzYE/oa6vL00kjYc8aPC 6eF5k5eGAj/k6RaDKyGxXKitWBBbWyWhWZLFC+7fuNwY8UI= X-Google-Smtp-Source: AMsMyM5hddk5iWIjVw5aRyrSn07oF4fi4G1j1ETjzuWSrcjzudNxXxloipglfqCtSVMNQLcraW/y+64IAqgYgN4W/08= X-Received: by 2002:a17:902:dad2:b0:17f:7439:20f2 with SMTP id q18-20020a170902dad200b0017f743920f2mr24383695plx.29.1666449158279; Sat, 22 Oct 2022 07:32:38 -0700 (PDT) MIME-Version: 1.0 References: <87bkqatu61.fsf@toke.dk> <759c25c6fd54dceccc00eada5ccf5358d2d1c20c.camel@kau.se> In-Reply-To: From: Herbert Wolverson Date: Sat, 22 Oct 2022 09:32:27 -0500 Message-ID: Cc: "libreqos@lists.bufferbloat.net" Content-Type: multipart/alternative; boundary="0000000000001d981d05eba06e83" Subject: Re: [LibreQoS] In BPF pping - so far X-BeenThere: libreqos@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Many ISPs need the kinds of quality shaping cake can do List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 22 Oct 2022 14:32:40 -0000 --0000000000001d981d05eba06e83 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable This morning I tested cpu-pping with live customers! A little over 1,200 mapped IP addresses, about 600 mbps of real traffic flowing through a big hierarchy of 52 sites. (600 is our "quiet time" traffic) It started very well: the updated xdp-cpumap system dropped in place and the system worked as before. xdp_pping started to show data with correct mappings. CPU load from the mapping system is within 1% of where it was before. After about 20 minutes of continuous execution, it started to run into some scaling issues. The shaping system continued to run wonderfully, and CPU load was still fine. However, it stopped reporting latency data! A bit of debugging showed that once you exceed 16,384 in-flight TCP streams it isn't handling the "map full" situation gracefully - and clearing the map from userspace isn't working correctly. So I hacked away and hacked away. Anyway, it turns out that it does in fact work fine at that scale. There's just a one-line bug in the xdp_pping.c file. I forgot to actually *call* one line of packet cleanup code. Adding that, and everything was awesome. The entire patch that fixed it consists of adding one line: cleanup_packet_ts(packet_ts); Oops. Anyway, with that in place it's running superbly. I did identify a couple of places in which it's being overly verbose with debug information, so I've patched that also. After reducing the overly eager warning about not being able to read a TCP header, CPU performance improved by another 2% on average. Longer-term (i.e. not on a Saturday morning, when I'd rather be playing with my daughter!), I think I'll look at raising some of the buffer sizes. Thanks, Herbert On Wed, Oct 19, 2022 at 11:13 AM Dave Taht wrote: > PS - today's (free) p99 conference is *REALLY AWESOME*. > https://www.p99conf.io/ > > On Wed, Oct 19, 2022 at 9:13 AM Dave Taht wrote: > > > > flent outputs a flent.gz file that I can parse and plot 20 differnt > > ways. Also the graphing tools work on osx > > > > On Wed, Oct 19, 2022 at 9:11 AM Herbert Wolverson via LibreQoS > > wrote: > > > > > > That's true. The 12th gen does seem to have some "special" features..= . > makes for a nice writing platform > > > (this box is primarily my "write books and articles" machine). I'll b= e > doing a wider test on a more normal > > > platform, probably at the weekend (with real traffic, hence the delay > - have to find a time in which I > > > minimize disruption) > > > > > > On Wed, Oct 19, 2022 at 10:49 AM dan wrote: > > >> > > >> Those 'efficiency' threads in Intel 12th gen should probably be > addressed as well. You can't turn them off in BIOS. > > >> > > >> On Wed, Oct 19, 2022 at 8:48 AM Robert Chac=C3=B3n via LibreQoS < > libreqos@lists.bufferbloat.net> wrote: > > >>> > > >>> Awesome work on this! > > >>> I suspect there should be a slight performance bump once > Hyperthreading is disabled and efficient power management is off. > > >>> Hyperthreading/SMT always messes with HTB performance when I leave > it on. Thank you for mentioning that - I now went ahead and added > instructions on disabling hyperthreading on the Wiki for new users. > > >>> Super promising results! > > >>> Interested to see what throughput is with xdp-cpumap-tc vs > cpumap-pping. So far in your VM setup it seems to be doing very well. > > >>> > > >>> On Wed, Oct 19, 2022 at 8:06 AM Herbert Wolverson via LibreQoS < > libreqos@lists.bufferbloat.net> wrote: > > >>>> > > >>>> Also, I forgot to mention that I *think* the current version has > removed the requirement that the inbound > > >>>> and outbound classifiers be placed on the same CPU. I know interdu= o > was particularly keen on packing > > >>>> upload into fewer cores. I'll add that to my list of things to tes= t. > > >>>> > > >>>> On Wed, Oct 19, 2022 at 9:01 AM Herbert Wolverson < > herberticus@gmail.com> wrote: > > >>>>> > > >>>>> I'll definitely take a look - that does look interesting. I don't > have X11 on any of my test VMs, but > > >>>>> it looks like it can work without the GUI. > > >>>>> > > >>>>> Thanks! > > >>>>> > > >>>>> On Wed, Oct 19, 2022 at 8:58 AM Dave Taht > wrote: > > >>>>>> > > >>>>>> could I coax you to adopt flent? > > >>>>>> > > >>>>>> apt-get install flent netperf irtt fping > > >>>>>> > > >>>>>> You sometimes have to compile netperf yourself with --enable-dem= o > on > > >>>>>> some systems. > > >>>>>> There are a bunch of python libs neede for the gui, but only on > the client. > > >>>>>> > > >>>>>> Then you can run a really gnarly test series and plot the result= s > over time. > > >>>>>> > > >>>>>> flent --socket-stats --step-size=3D.05 -t 'the-test-conditions' = -H > > >>>>>> the_server_name rrul # 110 other tests > > >>>>>> > > >>>>>> > > >>>>>> On Wed, Oct 19, 2022 at 6:44 AM Herbert Wolverson via LibreQoS > > >>>>>> wrote: > > >>>>>> > > > >>>>>> > Hey, > > >>>>>> > > > >>>>>> > Testing the current version ( > https://github.com/thebracket/cpumap-pping-hackjob ), it's doing better > than I hoped. This build has shared (not per-cpu) maps, and a userspace > daemon (xdp_pping) to extract and reset stats. > > >>>>>> > > > >>>>>> > My testing environment has grown a bit: > > >>>>>> > * ShaperVM - running Ubuntu Server and LibreQoS, with the new > cpumap-pping-hackjob version of xdp-cpumap. > > >>>>>> > * ExtTest - running Ubuntu Server, set as 10.64.1.1. Hosts an > iperf server. > > >>>>>> > * ClientInt1 - running Ubuntu Server (minimal), set as > 10.64.1.2. Hosts iperf client. > > >>>>>> > * ClientInt2 - running Ubuntu Server (minimal), set as > 10.64.1.3. Hosts iperf client. > > >>>>>> > > > >>>>>> > ClientInt1, ClientInt2 and one interface (LAN facing) of > ShaperVM are on a virtual switch. > > >>>>>> > ExtTest and the other interface (WAN facing) of ShaperVM are o= n > a different virtual switch. > > >>>>>> > > > >>>>>> > These are all on a host machine running Windows 11, a core i7 > 12th gen, 32 Gb RAM and fast SSD setup. > > >>>>>> > > > >>>>>> > TEST 1: DUAL STREAMS, LOW THROUGHPUT > > >>>>>> > > > >>>>>> > For this test, LibreQoS is configured: > > >>>>>> > * Two APs, each with 5gbit/s max. > > >>>>>> > * 100.64.1.2 and 100.64.1.3 setup as CPEs, each limited to > about 100mbit/s. They map to 1:5 and 2:5 respectively (separate CPUs). > > >>>>>> > * Set to use Cake > > >>>>>> > > > >>>>>> > On each client, roughly simultaneously run: iperf -c 100.64.1.= 1 > -t 500 (for a long run). Running xdp_pping yields correct results: > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:5", "avg" : 4, "min" : 3, "max" : 5, "samples" : 11}, > > >>>>>> > {"tc":"2:5", "avg" : 4, "min" : 3, "max" : 5, "samples" : 11}, > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > Or when I waited a while to gather/reset: > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:5", "avg" : 4, "min" : 3, "max" : 6, "samples" : 60}, > > >>>>>> > {"tc":"2:5", "avg" : 4, "min" : 3, "max" : 5, "samples" : 60}, > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > The ShaperVM shows no errors, just periodic logging that it is > recording data. CPU is about 2-3% on two CPUs, zero on the others (as > expected). > > >>>>>> > > > >>>>>> > After 500 seconds of continual iperfing, each client reported = a > throughput of 104 Mbit/sec and 6.06 GBytes of data transmitted. > > >>>>>> > > > >>>>>> > So for smaller streams, I'd call this a success. > > >>>>>> > > > >>>>>> > TEST 2: DUAL STREAMS, HIGH THROUGHPUT > > >>>>>> > > > >>>>>> > For this test, LibreQoS is configured: > > >>>>>> > * Two APs, each with 5gb/s max. > > >>>>>> > * 100.64.1.2 and 100.64.1.3 setup as CPEs, each limited to > 5Gbit/s! Mapped to 1:5 and 2:5 respectively (separate CPUs). > > >>>>>> > > > >>>>>> > Run iperfc -c 100.64.1.1 -t 500 on each client at the same tim= e. > > >>>>>> > > > >>>>>> > xdp_pping shows results, too: > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:5", "avg" : 4, "min" : 1, "max" : 7, "samples" : 58}, > > >>>>>> > {"tc":"2:5", "avg" : 7, "min" : 3, "max" : 11, "samples" : 58}= , > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:5", "avg" : 5, "min" : 4, "max" : 8, "samples" : 13}, > > >>>>>> > {"tc":"2:5", "avg" : 8, "min" : 7, "max" : 10, "samples" : 13}= , > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > The ShaperVM shows two CPUs pegging between 70 and 90 percent. > > >>>>>> > > > >>>>>> > After 500 seconds of continual iperfing, each client reported = a > throughput of 2.72 Gbits/sec (158 GBytes) and 3.89 Gbits/sec and 226 GByt= es. > > >>>>>> > > > >>>>>> > Maxing out HyperV like this is inducing a bit of latency (whic= h > is to be expected), but it's not bad. I also forgot to disable > hyperthreading, and looking at the host performance it is sometimes runni= ng > the second virtual CPU on an underpowered "fake" CPU. > > >>>>>> > > > >>>>>> > So for two large streams, I think we're doing pretty well also= ! > > >>>>>> > > > >>>>>> > TEST 3: DUAL STREAMS, SINGLE CPU > > >>>>>> > > > >>>>>> > This test is designed to try and blow things up. It's the same > as test 2, but both CPEs are set to the same CPU (1), using TC handles 1:= 5 > and 1:6. > > >>>>>> > > > >>>>>> > ShaperVM CPU1 maxed out in the high 90s, the other CPUs were > idle. The pping stats start to show a bit of degradation in performance f= or > pounding it so hard: > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:6", "avg" : 10, "min" : 9, "max" : 19, "samples" : 24= }, > > >>>>>> > {"tc":"1:5", "avg" : 10, "min" : 8, "max" : 18, "samples" : 24= }, > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > For whatever reason, it smoothed out over time: > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:6", "avg" : 10, "min" : 9, "max" : 12, "samples" : 50= }, > > >>>>>> > {"tc":"1:5", "avg" : 10, "min" : 8, "max" : 13, "samples" : 50= }, > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > Surprisingly (to me), I didn't encounter errors. Each client > received 2.22 Gbit/s performance, over 129 Gbytes of data. > > >>>>>> > > > >>>>>> > TEST 4: DUAL STREAMS, 50 SUB-STREAMS > > >>>>>> > > > >>>>>> > This test is also designed to break things. Same as test 3, bu= t > using iperf -c 100.64.1.1 -P 50 -t 120 - 50 substreams, to try and really > tax the flow tracking. (Shorter time window because I really wanted to go > and find coffee) > > >>>>>> > > > >>>>>> > ShaperVM CPU sat at around 80-97%, tending towards 97%. pping > results show that this torture test is worsening performance, and there's > always lots of samples in the buffer: > > >>>>>> > > > >>>>>> > [ > > >>>>>> > {"tc":"1:6", "avg" : 23, "min" : 19, "max" : 27, "samples" : > 49}, > > >>>>>> > {"tc":"1:5", "avg" : 24, "min" : 19, "max" : 27, "samples" : > 49}, > > >>>>>> > {}] > > >>>>>> > > > >>>>>> > This test also ran better than I expected. You can definitely > see some latency creeping in as I make the system work hard. Each VM show= ed > around 2.4 Gbit/s in total performance at the end of the iperf session. > There's definitely some latency creeping in, which is expected - but I'm > not sure I expected quite that much. > > >>>>>> > > > >>>>>> > WHAT'S NEXT & CONCLUSION > > >>>>>> > > > >>>>>> > I noticed that I forgot to turn off efficient power management > on my VMs and host, and left Hyperthreading on by mistake. So that hurts > overall performance. > > >>>>>> > > > >>>>>> > The base system seems to be working pretty solidly, at least > for small tests.Next up, I'll be removing extraneous debug reporting code= , > removing some code paths that don't do anything but report, and looking f= or > any small optimization opportunities. I'll then re-run these tests. Once > that's done, I hope to find a maintenance window on my WISP and try it wi= th > actual traffic. > > >>>>>> > > > >>>>>> > I also need to re-run these tests without the pping system to > provide some before/after analysis. > > >>>>>> > > > >>>>>> > On Tue, Oct 18, 2022 at 1:01 PM Herbert Wolverson < > herberticus@gmail.com> wrote: > > >>>>>> >> > > >>>>>> >> It's probably not entirely thread-safe right now (ran into > some issues reading per_cpu maps back from userspace; hopefully, I'll get > that figured out) - but the commits I just pushed have it basically worki= ng > on single-stream testing. :-) > > >>>>>> >> > > >>>>>> >> Setup cpumap as usual, and periodically run xdp-pping. This > gives you per-connection RTT information in JSON: > > >>>>>> >> > > >>>>>> >> [ > > >>>>>> >> {"tc":"1:5", "avg" : 5, "min" : 5, "max" : 5, "samples" : 1}, > > >>>>>> >> {}] > > >>>>>> >> > > >>>>>> >> (With the extra {} because I'm not tracking the tail and > haven't done comma removal). The tool also empties the various maps used = to > gather data, acting as a "reset" point. There's a max of 60 samples per > queue, in a ringbuffer setup (so newest will start to overwrite the oldes= t). > > >>>>>> >> > > >>>>>> >> I'll start trying to test on a larger scale now. > > >>>>>> >> > > >>>>>> >> On Mon, Oct 17, 2022 at 3:34 PM Robert Chac=C3=B3n < > robert.chacon@jackrabbitwireless.com> wrote: > > >>>>>> >>> > > >>>>>> >>> Hey Herbert, > > >>>>>> >>> > > >>>>>> >>> Fantastic work! Super exciting to see this coming together, > especially so quickly. > > >>>>>> >>> I'll test it soon. > > >>>>>> >>> I understand and agree with your decision to omit certain > features (ICMP tracking,DNS tracking, etc) to optimize performance for ou= r > use case. Like you said, in order to merge the functionality without a > performance hit, merging them is sort of the only way right now. Otherwis= e > there would be a lot of redundancy and lost throughput for an ISP's use. > Though hopefully long term there will be a way to keep all projects worki= ng > independently but interoperably with a plugin system of some kind. > > >>>>>> >>> > > >>>>>> >>> By the way, I'm making some headway on LibreQoS v1.3. > Focusing on optimizations for high sub counts (8000+ subs) as well as > stateful changes to the queue structure. > > >>>>>> >>> I'm working to set up a physical lab to test high throughput > and high client count scenarios. > > >>>>>> >>> When testing beyond ~32,000 filters we get "no space left on > device" from xdp-cpumap-tc, which I think relates to the bpf map size > limitation you mentioned. Maybe in the coming months we can take a look a= t > that. > > >>>>>> >>> > > >>>>>> >>> Anyway great work on the cpumap-pping program! Excited to se= e > more on this. > > >>>>>> >>> > > >>>>>> >>> Thanks, > > >>>>>> >>> Robert > > >>>>>> >>> > > >>>>>> >>> On Mon, Oct 17, 2022 at 12:45 PM Herbert Wolverson via > LibreQoS wrote: > > >>>>>> >>>> > > >>>>>> >>>> Hey, > > >>>>>> >>>> > > >>>>>> >>>> My current (unfinished) progress on this is now available > here: https://github.com/thebracket/cpumap-pping-hackjob > > >>>>>> >>>> > > >>>>>> >>>> I mean it about the warnings, this isn't at all stable, > debugged - and can't promise that it won't unleash the nasal demons > > >>>>>> >>>> (to use a popular C++ phrase). The name is descriptive! ;-) > > >>>>>> >>>> > > >>>>>> >>>> With that said, I'm pretty happy so far: > > >>>>>> >>>> > > >>>>>> >>>> * It runs only on the classifier - which xdp-cpumap-tc has > nicely shunted onto a dedicated CPU. It has to run on both > > >>>>>> >>>> the inbound and outbound classifiers, since otherwise it > would only see half the conversation. > > >>>>>> >>>> * It does assume that your ingress and egress CPUs are > mapped to the same interface; I do that anyway in BracketQoS. Not doing > > >>>>>> >>>> that opens up a potential world of pain, since writes to > the shared maps would require a locking scheme. Too much locking, and you > lose all of the benefit of using multiple CPUs to begin with. > > >>>>>> >>>> * It is pretty wasteful of RAM, but most of the shaper > systems I've worked with have lots of it. > > >>>>>> >>>> * I've been gradually removing features that I don't want > for BracketQoS. A hypothetical future "useful to everyone" version wouldn= 't > do that. > > >>>>>> >>>> * Rate limiting is working, but I removed the requirement > for a shared configuration provided from userland - so right now it's > always set to report at 1 second intervals per stream. > > >>>>>> >>>> > > >>>>>> >>>> My testbed is currently 3 Hyper-V VMs - a simple "client" > and "world", and a "shaper" VM in between running a slightly hacked-up > LibreQoS. > > >>>>>> >>>> iperf from "client" to "world" (with Libre set to allow > 10gbit/s max, via a cake/HTB queue setup) is around 5 gbit/s at present, = on > my > > >>>>>> >>>> test PC (the host is a core i7, 12th gen, 12 cores - 64gb > RAM and fast SSDs) > > >>>>>> >>>> > > >>>>>> >>>> Output currently consists of debug messages reading: > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 515.399222: > bpf_trace_printk: (tc) Flow open event > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 515.399239: > bpf_trace_printk: (tc) Send performance event (5,1), 374696 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 515.399466: > bpf_trace_printk: (tc) Flow open event > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 515.399475: > bpf_trace_printk: (tc) Send performance event (5,1), 247069 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 516.405151: > bpf_trace_printk: (tc) Send performance event (5,1), 5217155 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 517.405248: > bpf_trace_printk: (tc) Send performance event (5,1), 4515394 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 518.406117: > bpf_trace_printk: (tc) Send performance event (5,1), 4481289 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 519.406255: > bpf_trace_printk: (tc) Send performance event (5,1), 4255268 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 520.407864: > bpf_trace_printk: (tc) Send performance event (5,1), 5249493 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 521.406664: > bpf_trace_printk: (tc) Send performance event (5,1), 3795993 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 522.407469: > bpf_trace_printk: (tc) Send performance event (5,1), 3949519 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 523.408126: > bpf_trace_printk: (tc) Send performance event (5,1), 4365335 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 524.408929: > bpf_trace_printk: (tc) Send performance event (5,1), 4154910 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 525.410048: > bpf_trace_printk: (tc) Send performance event (5,1), 4405582 > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 525.434080: > bpf_trace_printk: (tc) Send flow event > > >>>>>> >>>> cpumap/0/map:4-1371 [000] D..2. 525.482714: > bpf_trace_printk: (tc) Send flow event > > >>>>>> >>>> > > >>>>>> >>>> The times haven't been tweaked yet. The (5,1) is tc handle > major/minor, allocated by the xdp-cpumap parent. > > >>>>>> >>>> I get pretty low latency between VMs; I'll set up a test > with some real-world data very soon. > > >>>>>> >>>> > > >>>>>> >>>> I plan to keep hacking away, but feel free to take a peek. > > >>>>>> >>>> > > >>>>>> >>>> Thanks, > > >>>>>> >>>> Herbert > > >>>>>> >>>> > > >>>>>> >>>> On Mon, Oct 17, 2022 at 10:14 AM Simon Sundberg < > Simon.Sundberg@kau.se> wrote: > > >>>>>> >>>>> > > >>>>>> >>>>> Hi, thanks for adding me to the conversation. Just a coupl= e > of quick > > >>>>>> >>>>> notes. > > >>>>>> >>>>> > > >>>>>> >>>>> On Mon, 2022-10-17 at 16:13 +0200, Toke H=C3=B8iland-J=C3= =B8rgensen > wrote: > > >>>>>> >>>>> > [ Adding Simon to Cc ] > > >>>>>> >>>>> > > > >>>>>> >>>>> > Herbert Wolverson via LibreQoS < > libreqos@lists.bufferbloat.net> writes: > > >>>>>> >>>>> > > > >>>>>> >>>>> > > Hey, > > >>>>>> >>>>> > > > > >>>>>> >>>>> > > I've had some pretty good success with merging > xdp-pping ( > > >>>>>> >>>>> > > > https://github.com/xdp-project/bpf-examples/blob/master/pping/pping.h ) > > >>>>>> >>>>> > > into xdp-cpumap-tc ( > https://github.com/xdp-project/xdp-cpumap-tc ). > > >>>>>> >>>>> > > > > >>>>>> >>>>> > > I ported over most of the xdp-pping code, and then > changed the entry point > > >>>>>> >>>>> > > and packet parsing code to make use of the work alread= y > done in > > >>>>>> >>>>> > > xdp-cpumap-tc (it's already parsed a big chunk of the > packet, no need to do > > >>>>>> >>>>> > > it twice). Then I switched the maps to per-cpu maps, > and had to pin them - > > >>>>>> >>>>> > > otherwise the two tc instances don't properly share > data. > > >>>>>> >>>>> > > > > >>>>>> >>>>> > > >>>>>> >>>>> I guess the xdp-cpumap-tc ensures that the same flow is > processed on > > >>>>>> >>>>> the same CPU core at both ingress or egress. Otherwise, if > a flow may > > >>>>>> >>>>> be processed by different cores on ingress and egress the > per-CPU maps > > >>>>>> >>>>> will not really work reliably as each core will have a > different view > > >>>>>> >>>>> on the state of the flow, if there's been a previous packe= t > with a > > >>>>>> >>>>> certain TSval from that flow etc. > > >>>>>> >>>>> > > >>>>>> >>>>> Furthermore, if a flow is always processed on the same cor= e > (on both > > >>>>>> >>>>> ingress and egress) I think per-CPU maps may be a bit > wasteful on > > >>>>>> >>>>> memory. From my understanding the keys for per-CPU maps ar= e > still > > >>>>>> >>>>> shared across all CPUs, it's just that each CPU gets its > own value. So > > >>>>>> >>>>> all CPUs will then have their own data for each flow, but > it's only the > > >>>>>> >>>>> CPU processing the flow that will have any relevant data > for the flow > > >>>>>> >>>>> while the remaining CPUs will just have an empty state for > that flow. > > >>>>>> >>>>> Under the same assumption that packets within the same flo= w > are always > > >>>>>> >>>>> processed on the same core there should generally not be a= ny > > >>>>>> >>>>> concurrency issues with having a global (non-per-CPU) > either as packets > > >>>>>> >>>>> from the same flow cannot be processed concurrently then > (and thus no > > >>>>>> >>>>> concurrent access to the same value in the map). I am > however still > > >>>>>> >>>>> very unclear on if there's any considerable performance > impact between > > >>>>>> >>>>> global and per-CPU map versions if the same key is not > accessed > > >>>>>> >>>>> concurrently. > > >>>>>> >>>>> > > >>>>>> >>>>> > > Right now, output > > >>>>>> >>>>> > > is just stubbed - I've still got to port the perfmap > output code. Instead, > > >>>>>> >>>>> > > I'm dumping a bunch of extra data to the kernel debug > pipe, so I can see > > >>>>>> >>>>> > > roughly what the output would look like. > > >>>>>> >>>>> > > > > >>>>>> >>>>> > > With debug enabled and just logging I'm now getting > about 4.9 Gbits/sec on > > >>>>>> >>>>> > > single-stream iperf between two VMs (with a shaper VM > in the middle). :-) > > >>>>>> >>>>> > > > >>>>>> >>>>> > Just FYI, that "just logging" is probably the biggest > source of > > >>>>>> >>>>> > overhead, then. What Simon found was that sending the > data from kernel > > >>>>>> >>>>> > to userspace is one of the most expensive bits of epping= , > at least when > > >>>>>> >>>>> > the number of data points goes up (which is does as > additional flows are > > >>>>>> >>>>> > added). > > >>>>>> >>>>> > > >>>>>> >>>>> Yhea, reporting individual RTTs when there's lots of them > (you may get > > >>>>>> >>>>> upwards of 1000 RTTs/s per flow) is not only problematic i= n > terms of > > >>>>>> >>>>> direct overhead from the tool itself, but also becomes > demanding for > > >>>>>> >>>>> whatever you use all those RTT samples for (i.e. need to > log, parse, > > >>>>>> >>>>> analyze etc. a very large amount of RTTs). One way to deal > with that is > > >>>>>> >>>>> of course to just apply some sort of sampling (the > -r/--rate-limit and > > >>>>>> >>>>> -R/--rtt-rate > > >>>>>> >>>>> > > > >>>>>> >>>>> > > So my question: how would you prefer to receive this > data? I'll have to > > >>>>>> >>>>> > > write a daemon that provides userspace control > (periodic cleanup as well as > > >>>>>> >>>>> > > reading the performance stream), so the world's kinda > our oyster. I can > > >>>>>> >>>>> > > stick to Kathie's original format (and dump it to a > named pipe, perhaps?), > > >>>>>> >>>>> > > a condensed format that only shows what you want to > use, an efficient > > >>>>>> >>>>> > > binary format if you feel like parsing that... > > >>>>>> >>>>> > > > >>>>>> >>>>> > It would be great if we could combine efforts a bit here > so we don't > > >>>>>> >>>>> > fork the codebase more than we have to. I.e., if > "upstream" epping and > > >>>>>> >>>>> > whatever daemon you end up writing can agree on data > format etc that > > >>>>>> >>>>> > would be fantastic! Added Simon to Cc to facilitate this > :) > > >>>>>> >>>>> > > > >>>>>> >>>>> > Briefly what I've discussed before with Simon was to hav= e > the ability to > > >>>>>> >>>>> > aggregate the metrics in the kernel (WiP PR [0]) and hav= e > a userspace > > >>>>>> >>>>> > utility periodically pull them out. What we discussed wa= s > doing this > > >>>>>> >>>>> > using an LPM map (which is not in that PR yet). The idea > would be that > > >>>>>> >>>>> > userspace would populate the LPM map with the keys > (prefixes) they > > >>>>>> >>>>> > wanted statistics for (in LibreQOS context that could be > one key per > > >>>>>> >>>>> > customer, for instance). Epping would then do a map > lookup into the LPM, > > >>>>>> >>>>> > and if it gets a match it would update the statistics in > that map entry > > >>>>>> >>>>> > (keeping a histogram of latency values seen, basically). > Simon's PR > > >>>>>> >>>>> > below uses this technique where userspace will "reset" > the histogram > > >>>>>> >>>>> > every time it loads it by swapping out two different map > entries when it > > >>>>>> >>>>> > does a read; this allows you to control the sampling rat= e > from > > >>>>>> >>>>> > userspace, and you'll just get the data since the last > time you polled. > > >>>>>> >>>>> > > >>>>>> >>>>> Thank's Toke for summarzing both the current state and the > plan going > > >>>>>> >>>>> forward. I will just note that this PR (and all my other > work with > > >>>>>> >>>>> ePPing/BPF-PPing/XDP-PPing/I-suck-at-names-PPing) will be > more or less > > >>>>>> >>>>> on hold for a couple of weeks right now as I'm trying to > finish up a > > >>>>>> >>>>> paper. > > >>>>>> >>>>> > > >>>>>> >>>>> > I was thinking that if we all can agree on the map > format, then your > > >>>>>> >>>>> > polling daemon could be one userspace "client" for that, > and the epping > > >>>>>> >>>>> > binary itself could be another; but we could keep > compatibility between > > >>>>>> >>>>> > the two, so we don't duplicate effort. > > >>>>>> >>>>> > > > >>>>>> >>>>> > Similarly, refactoring of the epping code itself so it > can be plugged > > >>>>>> >>>>> > into the cpumap-tc code would be a good goal... > > >>>>>> >>>>> > > >>>>>> >>>>> Should probably do that...at some point. In general I thin= k > it's a bit > > >>>>>> >>>>> of an interesting problem to think about how to chain > multiple XDP/tc > > >>>>>> >>>>> programs together in an efficent way. Most XDP and tc > programs will do > > >>>>>> >>>>> some amount of packet parsing and when you have many > chained programs > > >>>>>> >>>>> parsing the same packets this obviously becomes a bit > wasteful. In the > > >>>>>> >>>>> same time it would be nice if one didn't need to manually > merge > > >>>>>> >>>>> multiple programs together into a single one like this to > get rid of > > >>>>>> >>>>> this duplicated parsing, or at least make that process of > merging those > > >>>>>> >>>>> programs as simple as possible. > > >>>>>> >>>>> > > >>>>>> >>>>> > > >>>>>> >>>>> > -Toke > > >>>>>> >>>>> > > > >>>>>> >>>>> > [0] https://github.com/xdp-project/bpf-examples/pull/59 > > >>>>>> >>>>> > > >>>>>> >>>>> N=C3=A4r du skickar e-post till Karlstads universitet beha= ndlar > vi dina personuppgifter. > > >>>>>> >>>>> When you send an e-mail to Karlstad University, we will > process your personal data. > > >>>>>> >>>> > > >>>>>> >>>> _______________________________________________ > > >>>>>> >>>> LibreQoS mailing list > > >>>>>> >>>> LibreQoS@lists.bufferbloat.net > > >>>>>> >>>> https://lists.bufferbloat.net/listinfo/libreqos > > >>>>>> >>> > > >>>>>> >>> > > >>>>>> >>> > > >>>>>> >>> -- > > >>>>>> >>> Robert Chac=C3=B3n > > >>>>>> >>> CEO | JackRabbit Wireless LLC > > >>>>>> > > > >>>>>> > _______________________________________________ > > >>>>>> > LibreQoS mailing list > > >>>>>> > LibreQoS@lists.bufferbloat.net > > >>>>>> > https://lists.bufferbloat.net/listinfo/libreqos > > >>>>>> > > >>>>>> > > >>>>>> > > >>>>>> -- > > >>>>>> This song goes out to all the folk that thought Stadia would wor= k: > > >>>>>> > https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-698136666= 5607352320-FXtz > > >>>>>> Dave T=C3=A4ht CEO, TekLibre, LLC > > >>>> > > >>>> _______________________________________________ > > >>>> LibreQoS mailing list > > >>>> LibreQoS@lists.bufferbloat.net > > >>>> https://lists.bufferbloat.net/listinfo/libreqos > > >>> > > >>> > > >>> > > >>> -- > > >>> Robert Chac=C3=B3n > > >>> CEO | JackRabbit Wireless LLC > > >>> _______________________________________________ > > >>> LibreQoS mailing list > > >>> LibreQoS@lists.bufferbloat.net > > >>> https://lists.bufferbloat.net/listinfo/libreqos > > > > > > _______________________________________________ > > > LibreQoS mailing list > > > LibreQoS@lists.bufferbloat.net > > > https://lists.bufferbloat.net/listinfo/libreqos > > > > > > > > -- > > This song goes out to all the folk that thought Stadia would work: > > > https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-698136666= 5607352320-FXtz > > Dave T=C3=A4ht CEO, TekLibre, LLC > > > > -- > This song goes out to all the folk that thought Stadia would work: > > https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-698136666= 5607352320-FXtz > Dave T=C3=A4ht CEO, TekLibre, LLC > --0000000000001d981d05eba06e83 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
This morning I tested cpu-pping with live customers! =
A little over 1,200 mapped IP addresses, about 600 mbps of r= eal traffic flowing through a big
hierarchy of 52 sites. (600 is = our "quiet time" traffic)

It started= very well: the updated xdp-cpumap system dropped in place and the system w= orked as
before. xdp_pping started to show data with correct mapp= ings. CPU load from the mapping
system is within 1% of where it w= as before.

After about 20 minutes of continuous ex= ecution, it started to run into some scaling issues.
The shaping = system continued to run wonderfully, and CPU load was still fine. However,<= /div>
it stopped reporting latency data! A bit of debugging showed that= once you exceed
16,384 in-flight TCP streams it isn't handli= ng the "map full" situation gracefully - and
clearing t= he map from userspace isn't working correctly. So I hacked away and hac= ked
away.

Anyway, it turns out that it d= oes in fact work fine at that scale. There's just a one-line
= bug in the xdp_pping.c file. I forgot to actually *call* one line of packet= cleanup code.
Adding that, and everything was awesome.

The entire patch that fixed it consists of adding one = line:
cleanup_packet_ts(pac= ket_ts);

Oops.

Any= way, with that in place it's running superbly. I did identify a couple = of places in
which it's being overly verbose with debug infor= mation, so I've patched that also.

After reduc= ing the overly eager warning about not being able to read a TCP header,
CPU performance improved by another 2% on average.
Longer-term (i.e. not on a Saturday morning, when I'd rathe= r be playing with my
daughter!), I think I'll look at raising= some of the buffer sizes.

Thanks,
Herbe= rt

On Wed, Oct 19, 2022 at 11:13 AM Dave Taht <dave.taht@gmail.com> wrote:
PS - today's (free) p99 confe= rence is *REALLY AWESOME*. https://www.p99conf.io/

On Wed, Oct 19, 2022 at 9:13 AM Dave Taht <dave.taht@gmail.com> wrote:
>
> flent outputs a flent.gz file that I can parse and plot 20 differnt > ways. Also the graphing tools work on osx
>
> On Wed, Oct 19, 2022 at 9:11 AM Herbert Wolverson via LibreQoS
> <libreqos@lists.bufferbloat.net> wrote:
> >
> > That's true. The 12th gen does seem to have some "specia= l" features... makes for a nice writing platform
> > (this box is primarily my "write books and articles" ma= chine). I'll be doing a wider test on a more normal
> > platform, probably at the weekend (with real traffic, hence the d= elay - have to find a time in which I
> > minimize disruption)
> >
> > On Wed, Oct 19, 2022 at 10:49 AM dan <dandenson@gmail.com> wrote:
> >>
> >> Those 'efficiency' threads in Intel 12th gen should p= robably be addressed as well.=C2=A0 You can't turn them off in BIOS. > >>
> >> On Wed, Oct 19, 2022 at 8:48 AM Robert Chac=C3=B3n via LibreQ= oS <= libreqos@lists.bufferbloat.net> wrote:
> >>>
> >>> Awesome work on this!
> >>> I suspect there should be a slight performance bump once = Hyperthreading is disabled and efficient power management is off.
> >>> Hyperthreading/SMT always messes with HTB performance whe= n I leave it on. Thank you for mentioning that - I now went ahead and added= instructions on disabling hyperthreading on the Wiki for new users.
> >>> Super promising results!
> >>> Interested to see what throughput is with xdp-cpumap-tc v= s cpumap-pping. So far in your VM setup it seems to be doing very well.
> >>>
> >>> On Wed, Oct 19, 2022 at 8:06 AM Herbert Wolverson via Lib= reQoS <libreqos@lists.bufferbloat.net> wrote:
> >>>>
> >>>> Also, I forgot to mention that I *think* the current = version has removed the requirement that the inbound
> >>>> and outbound classifiers be placed on the same CPU. I= know interduo was particularly keen on packing
> >>>> upload into fewer cores. I'll add that to my list= of things to test.
> >>>>
> >>>> On Wed, Oct 19, 2022 at 9:01 AM Herbert Wolverson <= ;herberticus@gma= il.com> wrote:
> >>>>>
> >>>>> I'll definitely take a look - that does look = interesting. I don't have X11 on any of my test VMs, but
> >>>>> it looks like it can work without the GUI.
> >>>>>
> >>>>> Thanks!
> >>>>>
> >>>>> On Wed, Oct 19, 2022 at 8:58 AM Dave Taht <dave.taht@gmail.com> wrote:
> >>>>>>
> >>>>>> could I coax you to adopt flent?
> >>>>>>
> >>>>>> apt-get install flent netperf irtt fping
> >>>>>>
> >>>>>> You sometimes have to compile netperf yoursel= f with --enable-demo on
> >>>>>> some systems.
> >>>>>> There are a bunch of python libs neede for th= e gui, but only on the client.
> >>>>>>
> >>>>>> Then you can run a really gnarly test series = and plot the results over time.
> >>>>>>
> >>>>>> flent --socket-stats --step-size=3D.05 -t = 9;the-test-conditions' -H
> >>>>>> the_server_name rrul # 110 other tests
> >>>>>>
> >>>>>>
> >>>>>> On Wed, Oct 19, 2022 at 6:44 AM Herbert Wolve= rson via LibreQoS
> >>>>>> <libreqos@lists.bufferbloat.net> wrote: > >>>>>> >
> >>>>>> > Hey,
> >>>>>> >
> >>>>>> > Testing the current version ( https://github.com/thebracket/cpumap-pping-hackjob ), it= 's doing better than I hoped. This build has shared (not per-cpu) maps,= and a userspace daemon (xdp_pping) to extract and reset stats.
> >>>>>> >
> >>>>>> > My testing environment has grown a bit:<= br> > >>>>>> > * ShaperVM - running Ubuntu Server and L= ibreQoS, with the new cpumap-pping-hackjob version of xdp-cpumap.
> >>>>>> > * ExtTest - running Ubuntu Server, set a= s 10.64.1.1. Hosts an iperf server.
> >>>>>> > * ClientInt1 - running Ubuntu Server (mi= nimal), set as 10.64.1.2. Hosts iperf client.
> >>>>>> > * ClientInt2 - running Ubuntu Server (mi= nimal), set as 10.64.1.3. Hosts iperf client.
> >>>>>> >
> >>>>>> > ClientInt1, ClientInt2 and one interface= (LAN facing) of ShaperVM are on a virtual switch.
> >>>>>> > ExtTest and the other interface (WAN fac= ing) of ShaperVM are on a different virtual switch.
> >>>>>> >
> >>>>>> > These are all on a host machine running = Windows 11, a core i7 12th gen, 32 Gb RAM and fast SSD setup.
> >>>>>> >
> >>>>>> > TEST 1: DUAL STREAMS, LOW THROUGHPUT
> >>>>>> >
> >>>>>> > For this test, LibreQoS is configured: > >>>>>> > * Two APs, each with 5gbit/s max.
> >>>>>> > * 100.64.1.2 and 100.64.1.3 setup as CPE= s, each limited to about 100mbit/s. They map to 1:5 and 2:5 respectively (s= eparate CPUs).
> >>>>>> > * Set to use Cake
> >>>>>> >
> >>>>>> > On each client, roughly simultaneously r= un: iperf -c 100.64.1.1 -t 500 (for a long run). Running xdp_pping yields c= orrect results:
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:5", "a= vg" : 4, "min" : 3, "max" : 5, "samples"= : 11},
> >>>>>> > {"tc":"2:5", "a= vg" : 4, "min" : 3, "max" : 5, "samples"= : 11},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > Or when I waited a while to gather/reset= :
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:5", "a= vg" : 4, "min" : 3, "max" : 6, "samples"= : 60},
> >>>>>> > {"tc":"2:5", "a= vg" : 4, "min" : 3, "max" : 5, "samples"= : 60},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > The ShaperVM shows no errors, just perio= dic logging that it is recording data.=C2=A0 CPU is about 2-3% on two CPUs,= zero on the others (as expected).
> >>>>>> >
> >>>>>> > After 500 seconds of continual iperfing,= each client reported a throughput of 104 Mbit/sec and 6.06 GBytes of data = transmitted.
> >>>>>> >
> >>>>>> > So for smaller streams, I'd call thi= s a success.
> >>>>>> >
> >>>>>> > TEST 2: DUAL STREAMS, HIGH THROUGHPUT > >>>>>> >
> >>>>>> > For this test, LibreQoS is configured: > >>>>>> > * Two APs, each with 5gb/s max.
> >>>>>> > * 100.64.1.2 and 100.64.1.3 setup as CPE= s, each limited to 5Gbit/s! Mapped to 1:5 and 2:5 respectively (separate CP= Us).
> >>>>>> >
> >>>>>> > Run iperfc -c 100.64.1.1 -t 500 on each = client at the same time.
> >>>>>> >
> >>>>>> > xdp_pping shows results, too:
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:5", "a= vg" : 4, "min" : 1, "max" : 7, "samples"= : 58},
> >>>>>> > {"tc":"2:5", "a= vg" : 7, "min" : 3, "max" : 11, "samples"= ; : 58},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:5", "a= vg" : 5, "min" : 4, "max" : 8, "samples"= : 13},
> >>>>>> > {"tc":"2:5", "a= vg" : 8, "min" : 7, "max" : 10, "samples"= ; : 13},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > The ShaperVM shows two CPUs pegging betw= een 70 and 90 percent.
> >>>>>> >
> >>>>>> > After 500 seconds of continual iperfing,= each client reported a throughput of 2.72 Gbits/sec (158 GBytes) and 3.89 = Gbits/sec and 226 GBytes.
> >>>>>> >
> >>>>>> > Maxing out HyperV like this is inducing = a bit of latency (which is to be expected), but it's not bad. I also fo= rgot to disable hyperthreading, and looking at the host performance it is s= ometimes running the second virtual CPU on an underpowered "fake"= CPU.
> >>>>>> >
> >>>>>> > So for two large streams, I think we'= ;re doing pretty well also!
> >>>>>> >
> >>>>>> > TEST 3: DUAL STREAMS, SINGLE CPU
> >>>>>> >
> >>>>>> > This test is designed to try and blow th= ings up. It's the same as test 2, but both CPEs are set to the same CPU= (1), using TC handles 1:5 and 1:6.
> >>>>>> >
> >>>>>> > ShaperVM CPU1 maxed out in the high 90s,= the other CPUs were idle. The pping stats start to show a bit of degradati= on in performance for pounding it so hard:
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:6", "a= vg" : 10, "min" : 9, "max" : 19, "samples&quo= t; : 24},
> >>>>>> > {"tc":"1:5", "a= vg" : 10, "min" : 8, "max" : 18, "samples&quo= t; : 24},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > For whatever reason, it smoothed out ove= r time:
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:6", "a= vg" : 10, "min" : 9, "max" : 12, "samples&quo= t; : 50},
> >>>>>> > {"tc":"1:5", "a= vg" : 10, "min" : 8, "max" : 13, "samples&quo= t; : 50},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > Surprisingly (to me), I didn't encou= nter errors. Each client received 2.22 Gbit/s performance, over 129 Gbytes = of data.
> >>>>>> >
> >>>>>> > TEST 4: DUAL STREAMS, 50 SUB-STREAMS
> >>>>>> >
> >>>>>> > This test is also designed to break thin= gs. Same as test 3, but using iperf -c 100.64.1.1 -P 50 -t 120 - 50 substre= ams, to try and really tax the flow tracking. (Shorter time window because = I really wanted to go and find coffee)
> >>>>>> >
> >>>>>> > ShaperVM CPU sat at around 80-97%, tendi= ng towards 97%. pping results show that this torture test is worsening perf= ormance, and there's always lots of samples in the buffer:
> >>>>>> >
> >>>>>> > [
> >>>>>> > {"tc":"1:6", "a= vg" : 23, "min" : 19, "max" : 27, "samples&qu= ot; : 49},
> >>>>>> > {"tc":"1:5", "a= vg" : 24, "min" : 19, "max" : 27, "samples&qu= ot; : 49},
> >>>>>> > {}]
> >>>>>> >
> >>>>>> > This test also ran better than I expecte= d. You can definitely see some latency creeping in as I make the system wor= k hard. Each VM showed around 2.4 Gbit/s in total performance at the end of= the iperf session. There's definitely some latency creeping in, which = is expected - but I'm not sure I expected quite that much.
> >>>>>> >
> >>>>>> > WHAT'S NEXT & CONCLUSION
> >>>>>> >
> >>>>>> > I noticed that I forgot to turn off effi= cient power management on my VMs and host, and left Hyperthreading on by mi= stake. So that hurts overall performance.
> >>>>>> >
> >>>>>> > The base system seems to be working pret= ty solidly, at least for small tests.Next up, I'll be removing extraneo= us debug reporting code, removing some code paths that don't do anythin= g but report, and looking for any small optimization opportunities. I'l= l then re-run these tests. Once that's done, I hope to find a maintenan= ce window on my WISP and try it with actual traffic.
> >>>>>> >
> >>>>>> > I also need to re-run these tests withou= t the pping system to provide some before/after analysis.
> >>>>>> >
> >>>>>> > On Tue, Oct 18, 2022 at 1:01 PM Herbert = Wolverson <he= rberticus@gmail.com> wrote:
> >>>>>> >>
> >>>>>> >> It's probably not entirely threa= d-safe right now (ran into some issues reading per_cpu maps back from users= pace; hopefully, I'll get that figured out) - but the commits I just pu= shed have it basically working on single-stream testing. :-)
> >>>>>> >>
> >>>>>> >> Setup cpumap as usual, and periodica= lly run xdp-pping. This gives you per-connection RTT information in JSON: > >>>>>> >>
> >>>>>> >> [
> >>>>>> >> {"tc":"1:5", &qu= ot;avg" : 5, "min" : 5, "max" : 5, "samples&q= uot; : 1},
> >>>>>> >> {}]
> >>>>>> >>
> >>>>>> >> (With the extra {} because I'm n= ot tracking the tail and haven't done comma removal). The tool also emp= ties the various maps used to gather data, acting as a "reset" po= int. There's a max of 60 samples per queue, in a ringbuffer setup (so n= ewest will start to overwrite the oldest).
> >>>>>> >>
> >>>>>> >> I'll start trying to test on a l= arger scale now.
> >>>>>> >>
> >>>>>> >> On Mon, Oct 17, 2022 at 3:34 PM Robe= rt Chac=C3=B3n <robert.chacon@jackrabbitwireless.com> wrote:
> >>>>>> >>>
> >>>>>> >>> Hey Herbert,
> >>>>>> >>>
> >>>>>> >>> Fantastic work! Super exciting t= o see this coming together, especially so quickly.
> >>>>>> >>> I'll test it soon.
> >>>>>> >>> I understand and agree with your= decision to omit certain features (ICMP tracking,DNS tracking, etc) to opt= imize performance for our use case. Like you said, in order to merge the fu= nctionality without a performance hit, merging them is sort of the only way= right now. Otherwise there would be a lot of redundancy and lost throughpu= t for an ISP's use. Though hopefully long term there will be a way to k= eep all projects working independently but interoperably with a plugin syst= em of some kind.
> >>>>>> >>>
> >>>>>> >>> By the way, I'm making some = headway on LibreQoS v1.3. Focusing on optimizations for high sub counts (80= 00+ subs) as well as stateful changes to the queue structure.
> >>>>>> >>> I'm working to set up a phys= ical lab to test high throughput and high client count scenarios.
> >>>>>> >>> When testing beyond ~32,000 filt= ers we get "no space left on device" from xdp-cpumap-tc, which I = think relates to the bpf map size limitation you mentioned. Maybe in the co= ming months we can take a look at that.
> >>>>>> >>>
> >>>>>> >>> Anyway great work on the cpumap-= pping program! Excited to see more on this.
> >>>>>> >>>
> >>>>>> >>> Thanks,
> >>>>>> >>> Robert
> >>>>>> >>>
> >>>>>> >>> On Mon, Oct 17, 2022 at 12:45 PM= Herbert Wolverson via LibreQoS <libreqos@lists.bufferbloat.net> wrote:<= br> > >>>>>> >>>>
> >>>>>> >>>> Hey,
> >>>>>> >>>>
> >>>>>> >>>> My current (unfinished) prog= ress on this is now available here: https://github= .com/thebracket/cpumap-pping-hackjob
> >>>>>> >>>>
> >>>>>> >>>> I mean it about the warnings= , this isn't at all stable, debugged - and can't promise that it wo= n't unleash the nasal demons
> >>>>>> >>>> (to use a popular C++ phrase= ). The name is descriptive! ;-)
> >>>>>> >>>>
> >>>>>> >>>> With that said, I'm pret= ty happy so far:
> >>>>>> >>>>
> >>>>>> >>>> * It runs only on the classi= fier - which xdp-cpumap-tc has nicely shunted onto a dedicated CPU. It has = to run on both
> >>>>>> >>>>=C2=A0 =C2=A0the inbound and = outbound classifiers, since otherwise it would only see half the conversati= on.
> >>>>>> >>>> * It does assume that your i= ngress and egress CPUs are mapped to the same interface; I do that anyway i= n BracketQoS. Not doing
> >>>>>> >>>>=C2=A0 =C2=A0that opens up a = potential world of pain, since writes to the shared maps would require a lo= cking scheme. Too much locking, and you lose all of the benefit of using mu= ltiple CPUs to begin with.
> >>>>>> >>>> * It is pretty wasteful of R= AM, but most of the shaper systems I've worked with have lots of it. > >>>>>> >>>> * I've been gradually re= moving features that I don't want for BracketQoS. A hypothetical future= "useful to everyone" version wouldn't do that.
> >>>>>> >>>> * Rate limiting is working, = but I removed the requirement for a shared configuration provided from user= land - so right now it's always set to report at 1 second intervals per= stream.
> >>>>>> >>>>
> >>>>>> >>>> My testbed is currently 3 Hy= per-V VMs - a simple "client" and "world", and a "= shaper" VM in between running a slightly hacked-up LibreQoS.
> >>>>>> >>>> iperf from "client"= ; to "world" (with Libre set to allow 10gbit/s max, via a cake/HT= B queue setup) is around 5 gbit/s at present, on my
> >>>>>> >>>> test PC (the host is a core = i7, 12th gen, 12 cores - 64gb RAM and fast SSDs)
> >>>>>> >>>>
> >>>>>> >>>> Output currently consists of= debug messages reading:
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0515.399222: bpf_trace_printk: (tc)= Flow open event
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0515.399239: bpf_trace_printk: (tc)= Send performance event (5,1), 374696
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0515.399466: bpf_trace_printk: (tc)= Flow open event
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0515.399475: bpf_trace_printk: (tc)= Send performance event (5,1), 247069
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0516.405151: bpf_trace_printk: (tc)= Send performance event (5,1), 5217155
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0517.405248: bpf_trace_printk: (tc)= Send performance event (5,1), 4515394
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0518.406117: bpf_trace_printk: (tc)= Send performance event (5,1), 4481289
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0519.406255: bpf_trace_printk: (tc)= Send performance event (5,1), 4255268
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0520.407864: bpf_trace_printk: (tc)= Send performance event (5,1), 5249493
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0521.406664: bpf_trace_printk: (tc)= Send performance event (5,1), 3795993
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0522.407469: bpf_trace_printk: (tc)= Send performance event (5,1), 3949519
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0523.408126: bpf_trace_printk: (tc)= Send performance event (5,1), 4365335
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0524.408929: bpf_trace_printk: (tc)= Send performance event (5,1), 4154910
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0525.410048: bpf_trace_printk: (tc)= Send performance event (5,1), 4405582
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0525.434080: bpf_trace_printk: (tc)= Send flow event
> >>>>>> >>>>=C2=A0 =C2=A0cpumap/0/map:4-1= 371=C2=A0 =C2=A0 [000] D..2.=C2=A0 =C2=A0525.482714: bpf_trace_printk: (tc)= Send flow event
> >>>>>> >>>>
> >>>>>> >>>> The times haven't been t= weaked yet. The (5,1) is tc handle major/minor, allocated by the xdp-cpumap= parent.
> >>>>>> >>>> I get pretty low latency bet= ween VMs; I'll set up a test with some real-world data very soon.
> >>>>>> >>>>
> >>>>>> >>>> I plan to keep hacking away,= but feel free to take a peek.
> >>>>>> >>>>
> >>>>>> >>>> Thanks,
> >>>>>> >>>> Herbert
> >>>>>> >>>>
> >>>>>> >>>> On Mon, Oct 17, 2022 at 10:1= 4 AM Simon Sundberg <Simon.Sundberg@kau.se> wrote:
> >>>>>> >>>>>
> >>>>>> >>>>> Hi, thanks for adding me= to the conversation. Just a couple of quick
> >>>>>> >>>>> notes.
> >>>>>> >>>>>
> >>>>>> >>>>> On Mon, 2022-10-17 at 16= :13 +0200, Toke H=C3=B8iland-J=C3=B8rgensen wrote:
> >>>>>> >>>>> > [ Adding Simon to C= c ]
> >>>>>> >>>>> >
> >>>>>> >>>>> > Herbert Wolverson v= ia LibreQoS <libreqos@lists.bufferbloat.net> writes:
> >>>>>> >>>>> >
> >>>>>> >>>>> > > Hey,
> >>>>>> >>>>> > >
> >>>>>> >>>>> > > I've had s= ome pretty good success with merging xdp-pping (
> >>>>>> >>>>> > > https://github.com/xdp-project/bpf-examples/b= lob/master/pping/pping.h )
> >>>>>> >>>>> > > into xdp-cpuma= p-tc ( https://github.com/xdp-project/xdp-cpumap-tc = ).
> >>>>>> >>>>> > >
> >>>>>> >>>>> > > I ported over = most of the xdp-pping code, and then changed the entry point
> >>>>>> >>>>> > > and packet par= sing code to make use of the work already done in
> >>>>>> >>>>> > > xdp-cpumap-tc = (it's already parsed a big chunk of the packet, no need to do
> >>>>>> >>>>> > > it twice). The= n I switched the maps to per-cpu maps, and had to pin them -
> >>>>>> >>>>> > > otherwise the = two tc instances don't properly share data.
> >>>>>> >>>>> > >
> >>>>>> >>>>>
> >>>>>> >>>>> I guess the xdp-cpumap-t= c ensures that the same flow is processed on
> >>>>>> >>>>> the same CPU core at bot= h ingress or egress. Otherwise, if a flow may
> >>>>>> >>>>> be processed by differen= t cores on ingress and egress the per-CPU maps
> >>>>>> >>>>> will not really work rel= iably as each core will have a different view
> >>>>>> >>>>> on the state of the flow= , if there's been a previous packet with a
> >>>>>> >>>>> certain TSval from that = flow etc.
> >>>>>> >>>>>
> >>>>>> >>>>> Furthermore, if a flow i= s always processed on the same core (on both
> >>>>>> >>>>> ingress and egress) I th= ink per-CPU maps may be a bit wasteful on
> >>>>>> >>>>> memory. From my understa= nding the keys for per-CPU maps are still
> >>>>>> >>>>> shared across all CPUs, = it's just that each CPU gets its own value. So
> >>>>>> >>>>> all CPUs will then have = their own data for each flow, but it's only the
> >>>>>> >>>>> CPU processing the flow = that will have any relevant data for the flow
> >>>>>> >>>>> while the remaining CPUs= will just have an empty state for that flow.
> >>>>>> >>>>> Under the same assumptio= n that packets within the same flow are always
> >>>>>> >>>>> processed on the same co= re there should generally not be any
> >>>>>> >>>>> concurrency issues with = having a global (non-per-CPU) either as packets
> >>>>>> >>>>> from the same flow canno= t be processed concurrently then (and thus no
> >>>>>> >>>>> concurrent access to the= same value in the map). I am however still
> >>>>>> >>>>> very unclear on if there= 's any considerable performance impact between
> >>>>>> >>>>> global and per-CPU map v= ersions if the same key is not accessed
> >>>>>> >>>>> concurrently.
> >>>>>> >>>>>
> >>>>>> >>>>> > > Right now, out= put
> >>>>>> >>>>> > > is just stubbe= d - I've still got to port the perfmap output code. Instead,
> >>>>>> >>>>> > > I'm dumpin= g a bunch of extra data to the kernel debug pipe, so I can see
> >>>>>> >>>>> > > roughly what t= he output would look like.
> >>>>>> >>>>> > >
> >>>>>> >>>>> > > With debug ena= bled and just logging I'm now getting about 4.9 Gbits/sec on
> >>>>>> >>>>> > > single-stream = iperf between two VMs (with a shaper VM in the middle). :-)
> >>>>>> >>>>> >
> >>>>>> >>>>> > Just FYI, that &quo= t;just logging" is probably the biggest source of
> >>>>>> >>>>> > overhead, then. Wha= t Simon found was that sending the data from kernel
> >>>>>> >>>>> > to userspace is one= of the most expensive bits of epping, at least when
> >>>>>> >>>>> > the number of data = points goes up (which is does as additional flows are
> >>>>>> >>>>> > added).
> >>>>>> >>>>>
> >>>>>> >>>>> Yhea, reporting individu= al RTTs when there's lots of them (you may get
> >>>>>> >>>>> upwards of 1000 RTTs/s p= er flow) is not only problematic in terms of
> >>>>>> >>>>> direct overhead from the= tool itself, but also becomes demanding for
> >>>>>> >>>>> whatever you use all tho= se RTT samples for (i.e. need to log, parse,
> >>>>>> >>>>> analyze etc. a very larg= e amount of RTTs). One way to deal with that is
> >>>>>> >>>>> of course to just apply = some sort of sampling (the -r/--rate-limit and
> >>>>>> >>>>> -R/--rtt-rate
> >>>>>> >>>>> >
> >>>>>> >>>>> > > So my question= : how would you prefer to receive this data? I'll have to
> >>>>>> >>>>> > > write a daemon= that provides userspace control (periodic cleanup as well as
> >>>>>> >>>>> > > reading the pe= rformance stream), so the world's kinda our oyster. I can
> >>>>>> >>>>> > > stick to Kathi= e's original format (and dump it to a named pipe, perhaps?),
> >>>>>> >>>>> > > a condensed fo= rmat that only shows what you want to use, an efficient
> >>>>>> >>>>> > > binary format = if you feel like parsing that...
> >>>>>> >>>>> >
> >>>>>> >>>>> > It would be great i= f we could combine efforts a bit here so we don't
> >>>>>> >>>>> > fork the codebase m= ore than we have to. I.e., if "upstream" epping and
> >>>>>> >>>>> > whatever daemon you= end up writing can agree on data format etc that
> >>>>>> >>>>> > would be fantastic!= Added Simon to Cc to facilitate this :)
> >>>>>> >>>>> >
> >>>>>> >>>>> > Briefly what I'= ve discussed before with Simon was to have the ability to
> >>>>>> >>>>> > aggregate the metri= cs in the kernel (WiP PR [0]) and have a userspace
> >>>>>> >>>>> > utility periodicall= y pull them out. What we discussed was doing this
> >>>>>> >>>>> > using an LPM map (w= hich is not in that PR yet). The idea would be that
> >>>>>> >>>>> > userspace would pop= ulate the LPM map with the keys (prefixes) they
> >>>>>> >>>>> > wanted statistics f= or (in LibreQOS context that could be one key per
> >>>>>> >>>>> > customer, for insta= nce). Epping would then do a map lookup into the LPM,
> >>>>>> >>>>> > and if it gets a ma= tch it would update the statistics in that map entry
> >>>>>> >>>>> > (keeping a histogra= m of latency values seen, basically). Simon's PR
> >>>>>> >>>>> > below uses this tec= hnique where userspace will "reset" the histogram
> >>>>>> >>>>> > every time it loads= it by swapping out two different map entries when it
> >>>>>> >>>>> > does a read; this a= llows you to control the sampling rate from
> >>>>>> >>>>> > userspace, and you&= #39;ll just get the data since the last time you polled.
> >>>>>> >>>>>
> >>>>>> >>>>> Thank's Toke for sum= marzing both the current state and the plan going
> >>>>>> >>>>> forward. I will just not= e that this PR (and all my other work with
> >>>>>> >>>>> ePPing/BPF-PPing/XDP-PPi= ng/I-suck-at-names-PPing) will be more or less
> >>>>>> >>>>> on hold for a couple of = weeks right now as I'm trying to finish up a
> >>>>>> >>>>> paper.
> >>>>>> >>>>>
> >>>>>> >>>>> > I was thinking that= if we all can agree on the map format, then your
> >>>>>> >>>>> > polling daemon coul= d be one userspace "client" for that, and the epping
> >>>>>> >>>>> > binary itself could= be another; but we could keep compatibility between
> >>>>>> >>>>> > the two, so we don&= #39;t duplicate effort.
> >>>>>> >>>>> >
> >>>>>> >>>>> > Similarly, refactor= ing of the epping code itself so it can be plugged
> >>>>>> >>>>> > into the cpumap-tc = code would be a good goal...
> >>>>>> >>>>>
> >>>>>> >>>>> Should probably do that.= ..at some point. In general I think it's a bit
> >>>>>> >>>>> of an interesting proble= m to think about how to chain multiple XDP/tc
> >>>>>> >>>>> programs together in an = efficent way. Most XDP and tc programs will do
> >>>>>> >>>>> some amount of packet pa= rsing and when you have many chained programs
> >>>>>> >>>>> parsing the same packets= this obviously becomes a bit wasteful. In the
> >>>>>> >>>>> same time it would be ni= ce if one didn't need to manually merge
> >>>>>> >>>>> multiple programs togeth= er into a single one like this to get rid of
> >>>>>> >>>>> this duplicated parsing,= or at least make that process of merging those
> >>>>>> >>>>> programs as simple as po= ssible.
> >>>>>> >>>>>
> >>>>>> >>>>>
> >>>>>> >>>>> > -Toke
> >>>>>> >>>>> >
> >>>>>> >>>>> > [0] https://github.com/xdp-project/bpf-examples/pull/59
> >>>>>> >>>>>
> >>>>>> >>>>> N=C3=A4r du skickar e-po= st till Karlstads universitet behandlar vi dina personuppgifter<https://w= ww.kau.se/gdpr>.
> >>>>>> >>>>> When you send an e-mail = to Karlstad University, we will process your personal data<https://www.= kau.se/en/gdpr>.
> >>>>>> >>>>
> >>>>>> >>>> ____________________________= ___________________
> >>>>>> >>>> LibreQoS mailing list
> >>>>>> >>>> LibreQoS@lists.bufferbloat.net<= br> > >>>>>> >>>> https:= //lists.bufferbloat.net/listinfo/libreqos
> >>>>>> >>>
> >>>>>> >>>
> >>>>>> >>>
> >>>>>> >>> --
> >>>>>> >>> Robert Chac=C3=B3n
> >>>>>> >>> CEO | JackRabbit Wireless LLC > >>>>>> >
> >>>>>> > ________________________________________= _______
> >>>>>> > LibreQoS mailing list
> >>>>>> > LibreQoS@lists.bufferbloat.net
> >>>>>> > https://lists.buff= erbloat.net/listinfo/libreqos
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>> --
> >>>>>> This song goes out to all the folk that thoug= ht Stadia would work:
> >>>>>> https://www.linkedin.com/posts/dtaht_the-mushroom-song-ac= tivity-6981366665607352320-FXtz
> >>>>>> Dave T=C3=A4ht CEO, TekLibre, LLC
> >>>>
> >>>> _______________________________________________
> >>>> LibreQoS mailing list
> >>>> LibreQoS@lists.bufferbloat.net
> >>>> https://lists.bufferbloat.net/l= istinfo/libreqos
> >>>
> >>>
> >>>
> >>> --
> >>> Robert Chac=C3=B3n
> >>> CEO | JackRabbit Wireless LLC
> >>> _______________________________________________
> >>> LibreQoS mailing list
> >>> LibreQoS@lists.bufferbloat.net
> >>> https://lists.bufferbloat.net/listi= nfo/libreqos
> >
> > _______________________________________________
> > LibreQoS mailing list
> > LibreQoS@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/li= breqos
>
>
>
> --
> This song goes out to all the folk that thought Stadia would work:
> https:/= /www.linkedin.com/posts/dtaht_the-mushroom-song-activity-698136666560735232= 0-FXtz
> Dave T=C3=A4ht CEO, TekLibre, LLC



--
This song goes out to all the folk that thought Stadia would work:
https://www.= linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXt= z
Dave T=C3=A4ht CEO, TekLibre, LLC
--0000000000001d981d05eba06e83--