| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Sat, 12 Aug 2017 19:27:02 +0200
From: Paweł Staszewski <pstaszewski@...are.pl>
To: Jesper Dangaard Brouer <brouer@...hat.com>
Cc: Linux Kernel Network Developers <netdev@...r.kernel.org>
Subject: Re: Kernel 4.13.0-rc4-next-20170811 - IP Routing / Forwarding
performance vs Core/RSS number / HT on
Hi and thanks for reply
W dniu 2017-08-12 o 14:23, Jesper Dangaard Brouer pisze:
> On Fri, 11 Aug 2017 19:51:10 +0200 Paweł Staszewski <pstaszewski@...are.pl> wrote:
>
>> Hi
>>
>> I made some tests for performance comparison.
> Thanks for doing this. Feel free to Cc me, if you do more of these
> tests (so I don't miss them on the mailing list).
>
> I don't understand stand if you are reporting a potential problem?
>
> It would be good if you can provide a short summary section (of the
> issue) in the _start_ of the email, and then provide all this nice data
> afterwards, to back your case.
>
> My understanding is, you report:
>
> 1. VLANs on ixgbe show a 30-40% slowdown
> 2. System stopped scaling after 7+ CPUs
This is not only problem/bug report - but some kind of comparision plus
some toughts about possible problems :)
And can help somebody when searching the net for possible expectations :)
Also - dono better list where are the smartest people that know what is
going in kernel with networking :)
Next time i will place summary on top - sorry :)
>
>> Tested HW (FORWARDING HOST):
>>
>> Intel(R) Xeon(R) Gold 6132 CPU @ 2.60GHz
> Interesting, I've not heard about a Intel CPU called "Gold" before now,
> but it does exist:
> https://ark.intel.com/products/123541/Intel-Xeon-Gold-6132-Processor-19_25M-Cache-2_60-GHz
>
>
>> Intel Corporation 82599ES 10-Gigabit SFI/SFP+ Network Connection (rev 01)
> This is one of my all time favorite NICs!
Yes this is a good NIC - will have connectx-4 2x100G by monday so will
also do some tests
>
>> Test diagram:
>>
>>
>> TRAFFIC GENERATOR (ethX) -> (enp216s0f0 - RX Traffic) FORWARDING HOST
>> (enp216s0f1(vlan1000) - TX Traffic) -> (ethY) SINK
>>
>> Forwarder traffic: UDP random ports from 9 to 19 with random hosts from
>> 172.16.0.1 to 172.16.0.255
>>
>> TRAFFIC GENERATOR TX is stable 9.9Mpps (in kernel pktgen)
> What kind of traffic flow? E.g. distribution, many/few source IPs...
Traffic generator is pktgen so udp flows - better paste parameters from
pktgen:
UDP_MIN=9
UDP_MAX=19
pg_set $dev "dst_min 172.16.0.1"
pg_set $dev "dst_max 172.16.0.100"
# Setup random UDP port src range
#pg_set $dev "flag UDPSRC_RND"
pg_set $dev "flag UDPSRC_RND"
pg_set $dev "udp_src_min $UDP_MIN"
pg_set $dev "udp_src_max $UDP_MAX"
>
>
>> Settings used for FORWARDING HOST (changed param. was only number of RSS
>> combined queues + set affinity assignment for them to fit with first
>> numa node where 2x10G port card is installed)
>>
>> ixgbe driver used from kernel (in-kernel build - not a module)
>>
> Nice with a script showing you setup, thanks. I would be good if it had
> comments, telling why you think this is a needed setup adjustment.
>
>> #!/bin/sh
>> ifc='enp216s0f0 enp216s0f1'
>> for i in $ifc
>> do
>> ip link set up dev $i
>> ethtool -A $i autoneg off rx off tx off
> Good:
> Turning off Ethernet flow control, to avoid receiver being the
> bottleneck via pause-frames.
Yes - enabled flow controll is really bad :)
>> ethtool -G $i rx 4096 tx 1024
> You adjust the RX and TX ring queue sizes, this have effects that you
> don't realize. Especially for the ixgbe driver, which have a page
> recycle trick tied to the RX ring queue size.
rx ring 4096 and tx ring 1024
- this is because have best performance then with average packet size
from 64 to 1500 bytes
Can be a little better performance for smaller frames like 64 - with rx
ring set to 1024
below 1 core/1 RSS queue with rx ring set to 1024
0;1;64;1530112;91772160;1529919;88724208
1;1;64;1531584;91872000;1531520;88813196
2;1;64;1531392;91895040;1531262;88831930
3;1;64;1530880;91875840;1531201;88783558
4;1;64;1530688;91829760;1530688;88768826
5;1;64;1530432;91810560;1530624;88764940
6;1;64;1530880;91868160;1530878;88787328
7;1;64;1530496;91845120;1530560;88765114
8;1;64;1530496;91837440;1530687;88772538
9;1;64;1530176;91795200;1530496;88735360
so from 1.47Mpps to 1.53Mpps
But with bigger packets > 200 performance is better when rx is set to 4096
>
>> ip link set $i txqueuelen 1000
> Setting tx queue len to the default 1000 seems redundant.
Yes cause i'm changing this parameter also to see if any impact on
performance we have
>
>> ethtool -C $i rx-usecs 10
> Adjusting this also have effects you might not realize. This actually
> also affect the page recycle scheme of ixgbe. And can sometimes be
> used to solve stalling on DMA TX completions, which could be you issue
> here.
same here - rx-usecs - setting to 10 was kind of compromise to have good
performance with big ans small packet sizes
Same test as above with rx ring 1024 tx ring 1024 and rxusecs set to 256
(1Core/1RSS queue):
0;1;64;1506304;90424320;1506626;87402868
1;1;64;1505536;90343680;1504830;87321088
2;1;64;1506880;90416640;1507522;87388120
3;1;64;1511040;90700800;1511682;87684864
4;1;64;1511040;90681600;1511102;87662476
5;1;64;1511488;90712320;1511614;87673728
6;1;64;1511296;90700800;1511038;87669900
7;1;64;1513344;90773760;1513280;87751680
8;1;64;1513536;90850560;1513470;87807360
9;1;64;1512128;90696960;1512000;87696000
And rx-usecs set to 1
0;1;64;1533632;92037120;1533504;88954368
1;1;64;1533632;92006400;1533570;88943348
2;1;64;1533504;91994880;1533504;88931980
3;1;64;1532864;91979520;1532674;88902516
4;1;64;1533952;92044800;1534080;88961792
5;1;64;1533888;92048640;1534270;88969100
6;1;64;1533952;92037120;1534082;88969216
7;1;64;1533952;92021760;1534208;88969332
8;1;64;1533056;91983360;1532930;88883724
9;1;64;1533760;92021760;1533886;88946828
rx-useck set to 2
0;1;64;1522432;91334400;1522304;88301056
1;1;64;1521920;91330560;1522496;88286208
2;1;64;1522496;91322880;1522432;88304768
3;1;64;1523456;91422720;1523649;88382762
4;1;64;1527680;91676160;1527424;88601728
5;1;64;1527104;91626240;1526912;88572032
6;1;64;1527424;91641600;1527424;88590592
7;1;64;1526336;91572480;1526912;88523776
8;1;64;1527040;91637760;1526912;88579456
9;1;64;1527040;91595520;1526784;88553472
rx-usecs set to 3
0;1;64;1526272;91549440;1526592;88527488
1;1;64;1526528;91560960;1526272;88516352
2;1;64;1525952;91580160;1525888;88527488
3;1;64;1525504;91511040;1524864;88456960
4;1;64;1526272;91568640;1526208;88494080
5;1;64;1525568;91545600;1525312;88494080
6;1;64;1526144;91584000;1526080;88512640
7;1;64;1525376;91530240;1525376;88482944
8;1;64;1526784;91607040;1526592;88549760
9;1;64;1526208;91560960;1526528;88512640
>
>> ethtool -L $i combined 16
>> ethtool -K $i gro on tso on gso off sg on l2-fwd-offload off
>> tx-nocache-copy on ntuple on
> Here are many setting above.
Yes mostly NIC defaults besides the ntuple that is on (for testing some
nfc drop filters - and trying to test also tc-offload )
> GRO/GSO/TSO for _forwarding_ is actually bad... in my tests, enabling
> this result in approx 10% slowdown.
Ok lets give a try :)
gro off tso off gso off sg on l2-fwd-offload off tx-nocache-copy on
ntuple on
rx-usecs 10
1 CPU / 1 RSS QUEUE
0;1;64;1609344;96537600;1609279;93327104
1;1;64;1608320;96514560;1608256;93293812
2;1;64;1608000;96487680;1608125;93267770
3;1;64;1608320;96522240;1608576;93297524
4;1;64;1605888;96387840;1606211;93148986
5;1;64;1601472;96072960;1601600;92870644
6;1;64;1602624;96180480;1602243;92959674
7;1;64;1601728;96107520;1602113;92907764
8;1;64;1602176;96122880;1602176;92933806
9;1;64;1603904;96253440;1603777;93045208
A little better performance 1.6Mpps
But wondering if disabling tso will have no performance impact for tcp
traffic ...
Will try to get some pktgen like pktgen-dpdk that can generate also tcp
traffic - to compare this.
>
> AFAIK "tx-nocache-copy on" was also determined to be a bad option.
I set this to on cause i have better performance (a little 10kpps for
this test)
below same test as above with tx-nocache-copy off
0;1;64;1591552;95496960;1591230;92313654
1;1;64;1596224;95738880;1595842;92555066
2;1;64;1595456;95700480;1595201;92521774
3;1;64;1595456;95723520;1595072;92528966
4;1;64;1595136;95692800;1595457;92503040
5;1;64;1594624;95631360;1594496;92473402
6;1;64;1596224;95761920;1595778;92551180
7;1;64;1595200;95700480;1595331;92521542
8;1;64;1595584;95692800;1595457;92521426
9;1;64;1594624;95662080;1594048;92469574
>
> The "ntuple on" AFAIK disables the flow-director in the NIC. I though
> this would actually help VLAN traffic, but I guess not.
yes I enabled this cause was thinking that can help with traffic on vlans
below same test with ntuple off
so all settings for ixgbe:
gro off tso off gso off sg on l2-fwd-offload off tx-nocache-copy off
ntuple off
rx-usecs 10
rx-flow-hash udp4 sdfn
0;1;64;1611840;96691200;1611905;93460794
1;1;64;1610688;96645120;1610818;93427328
2;1;64;1610752;96668160;1610497;93442176
3;1;64;1610624;96664320;1610817;93427212
4;1;64;1610752;96652800;1610623;93412480
5;1;64;1610048;96614400;1610112;93404940
6;1;64;1611264;96641280;1611390;93427212
7;1;64;1611008;96691200;1610942;93468160
8;1;64;1610048;96652800;1609984;93408652
9;1;64;1611136;96641280;1610690;93434636
Performance is a little better
and now with tx-nocache-copy on
0;1;64;1597248;95834880;1597311;92644096
1;1;64;1597888;95865600;1597824;92677446
2;1;64;1597952;95834880;1597822;92644038
3;1;64;1597568;95877120;1597375;92685044
4;1;64;1597184;95827200;1597314;92629190
5;1;64;1597696;95842560;1597565;92625652
6;1;64;1597312;95834880;1597376;92644038
7;1;64;1597568;95873280;1597634;92647924
8;1;64;1598400;95919360;1598849;92699602
9;1;64;1597824;95873280;1598208;92684928
That is weird - so enabling tx-nocache-copy with disabled ntuple have
bad performance impact - but with enabled ntuple there is no performance
impact
>
>
>> ethtool -N $i rx-flow-hash udp4 sdfn
> Why do you change the NICs flow-hash?
whan used 16 cores / 16 rss queues - there was better load distribution
over all cores when sdfn rx-flow-hash enabled
>
>> done
>>
>> ip link set up dev enp216s0f0
>> ip link set up dev enp216s0f1
>>
>> ip a a 10.0.0.1/30 dev enp216s0f0
>>
>> ip link add link enp216s0f1 name vlan1000 type vlan id 1000
>> ip link set up dev vlan1000
>> ip a a 10.0.0.5/30 dev vlan1000
>>
>>
>> ip route add 172.16.0.0/12 via 10.0.0.6
>>
>> ./set_irq_affinity.sh -x 14-27,42-43 enp216s0f0
>> ./set_irq_affinity.sh -x 14-27,42-43 enp216s0f1
>> #cat /sys/devices/system/node/node1/cpulist
>> #14-27,42-55
>> #cat /sys/devices/system/node/node0/cpulist
>> #0-13,28-41
> Is this a NUMA system?
This is 2x CPU 6132 - so have two separate pcie access to the nic - need
to check what cpu is assigned to pcie where network card is connected to
have network card on local cpu where all irq's are binded
>
>
>> #################################################
>>
>>
>> Looks like forwarding performance when using vlans on ixgbe is less that
>> without vlans for about 30-40% (wondering if this is some vlan
>> offloading problem and ixgbe)
> I would see this as a problem/bug that enabling VLANs cost this much.
Yes - was thinking that with tx/rx vlan offloading there will be not
much performance impact when vlans used.
>
>> settings below:
>>
>> ethtool -k enp216s0f0
>> Features for enp216s0f0:
>> Cannot get device udp-fragmentation-offload settings: Operation not
>> supported
>> rx-checksumming: on
>> tx-checksumming: on
>> tx-checksum-ipv4: off [fixed]
>> tx-checksum-ip-generic: on
>> tx-checksum-ipv6: off [fixed]
>> tx-checksum-fcoe-crc: off [fixed]
>> tx-checksum-sctp: on
>> scatter-gather: on
>> tx-scatter-gather: on
>> tx-scatter-gather-fraglist: off [fixed]
>> tcp-segmentation-offload: on
>> tx-tcp-segmentation: on
>> tx-tcp-ecn-segmentation: off [fixed]
>> tx-tcp-mangleid-segmentation: on
>> tx-tcp6-segmentation: on
>> udp-fragmentation-offload: off
>> generic-segmentation-offload: off
>> generic-receive-offload: on
>> large-receive-offload: off
>> rx-vlan-offload: on
>> tx-vlan-offload: on
>> ntuple-filters: on
>> receive-hashing: on
>> highdma: on [fixed]
>> rx-vlan-filter: on
>> vlan-challenged: off [fixed]
>> tx-lockless: off [fixed]
>> netns-local: off [fixed]
>> tx-gso-robust: off [fixed]
>> tx-fcoe-segmentation: off [fixed]
>> tx-gre-segmentation: on
>> tx-gre-csum-segmentation: on
>> tx-ipxip4-segmentation: on
>> tx-ipxip6-segmentation: on
>> tx-udp_tnl-segmentation: on
>> tx-udp_tnl-csum-segmentation: on
>> tx-gso-partial: on
>> tx-sctp-segmentation: off [fixed]
>> tx-esp-segmentation: off [fixed]
>> fcoe-mtu: off [fixed]
>> tx-nocache-copy: on
>> loopback: off [fixed]
>> rx-fcs: off [fixed]
>> rx-all: off
>> tx-vlan-stag-hw-insert: off [fixed]
>> rx-vlan-stag-hw-parse: off [fixed]
>> rx-vlan-stag-filter: off [fixed]
>> l2-fwd-offload: off
>> hw-tc-offload: off
>> esp-hw-offload: off [fixed]
>> esp-tx-csum-hw-offload: off [fixed]
>> rx-udp_tunnel-port-offload: on
>>
>>
>> Another thing is that forwarding performance does not scale with number
>> of cores when 7+ cores are reached
> I've seen problems with using Hyper-Threading CPUs. Could it be that
> above 7 CPUs you are starting to use sibling-cores ?
>
Turbostats can help here:
Package Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI
C1 C2 C1% C2% CPU%c1 CPU%c6 CoreTmp PkgTmp PkgWatt
RAMWatt PKG_% RAM_%
- - - 72 2.27 3188 2600 194844 0
64 69282 0.07 97.83 18.38 79.36 -4 54 123.49
16.08 0.00 0.00
0 0 0 8 0.74 1028 2600 1513 0
32 1462 1.50 97.99 10.92 88.34 47 51 58.34
5.34 0.00 0.00
0 0 28 7 0.67 1015 2600 1255 0
12 1249 0.96 98.61 10.99
0 1 1 7 0.68 1019 2600 1260 0
0 1260 0.00 99.54 8.44 90.88 49
0 1 29 9 0.71 1208 2600 1252 0
0 1253 0.00 99.48 8.41
0 2 2 7 0.67 1019 2600 1261 0
0 1260 0.00 99.54 8.44 90.89 48
0 2 30 7 0.67 1017 2600 1255 0
0 1255 0.00 99.55 8.44
0 3 3 7 0.68 1019 2600 1260 0
0 1259 0.00 99.53 8.46 90.86 -4
0 3 31 7 0.67 1017 2600 1256 0
0 1256 0.00 99.55 8.46
0 4 4 7 0.67 1027 2600 1260 0
0 1260 0.00 99.54 8.43 90.90 -4
0 4 32 7 0.66 1018 2600 1255 0
0 1255 0.00 99.55 8.44
0 5 5 7 0.68 1020 2600 1260 0
0 1257 0.00 99.54 8.44 90.89 50
0 5 33 7 0.68 1019 2600 1255 0
0 1255 0.00 99.55 8.43
0 6 6 7 0.70 1019 2600 1260 0
0 1259 0.00 99.53 8.43 90.87 -4
0 6 34 7 0.70 1019 2600 1255 0
0 1255 0.00 99.54 8.43
0 8 7 7 0.68 1019 2600 1262 0
0 1261 0.00 99.52 8.42 90.90 50
0 8 35 7 0.67 1019 2600 1255 0
0 1255 0.00 99.55 8.43
0 9 8 7 0.68 1019 2600 1260 0
0 1257 0.00 99.54 8.40 90.92 49
0 9 36 7 0.66 1017 2600 1255 0
0 1255 0.00 99.55 8.41
0 10 9 7 0.66 1018 2600 1257 0
0 1257 0.00 99.54 8.40 90.94 -4
0 10 37 7 0.66 1018 2600 1255 0
0 1255 0.00 99.55 8.41
0 11 10 7 0.66 1019 2600 1257 0
0 1259 0.00 99.54 8.56 90.77 -4
0 11 38 7 0.66 1018 2600 1255 0
3 1252 0.19 99.36 8.57
0 12 11 7 0.67 1019 2600 1260 0
0 1260 0.00 99.54 8.44 90.88 -4
0 12 39 7 0.67 1019 2600 1255 0
0 1256 0.00 99.55 8.44
0 13 12 7 0.68 1019 2600 1257 0
4 1254 0.32 99.22 8.67 90.65 -4
0 13 40 7 0.69 1019 2600 1256 0
4 1253 0.24 99.31 8.66
0 14 13 7 0.71 1020 2600 1260 0
0 1259 0.00 99.53 8.41 90.88 -4
0 14 41 7 0.72 1020 2600 1255 0
0 1255 0.00 99.54 8.40
1 0 14 3564 99.19 3594 2600 125472 0
0 0 0.00 0.00 0.81 0.00 54 54 65.15
10.74 0.00 0.00
1 0 42 3 0.07 3701 2600 1255 0
0 1255 0.00 99.95 99.93
1 1 15 11 0.32 3301 2600 1257 0
0 1257 0.00 99.81 26.37 73.31 42
1 1 43 10 0.31 3301 2600 1255 0
0 1255 0.00 99.82 26.38
1 2 16 10 0.31 3301 2600 1257 0
0 1257 0.00 99.81 26.37 73.32 39
1 2 44 10 0.32 3301 2600 1255 0
0 1255 0.00 99.82 26.36
1 3 17 10 0.32 3301 2600 1257 0
0 1257 0.00 99.81 26.40 73.28 39
1 3 45 11 0.32 3301 2600 1255 0
0 1255 0.00 99.81 26.40
1 4 18 10 0.32 3301 2600 1257 0
0 1257 0.00 99.82 26.40 73.28 40
1 4 46 11 0.32 3301 2600 1255 0
0 1255 0.00 99.82 26.40
1 5 19 11 0.33 3301 2600 1257 0
0 1257 0.00 99.81 26.40 73.27 39
1 5 47 11 0.33 3300 2600 1255 0
0 1255 0.00 99.82 26.40
1 6 20 12 0.35 3301 2600 1257 0
0 1257 0.00 99.81 26.38 73.27 42
1 6 48 12 0.36 3301 2600 1255 0
0 1255 0.00 99.81 26.37
1 8 21 11 0.33 3301 2600 1257 0
0 1257 0.00 99.82 26.37 73.29 42
1 8 49 11 0.33 3301 2600 1255 0
0 1255 0.00 99.82 26.38
1 9 22 10 0.32 3300 2600 1257 0
0 1257 0.00 99.82 26.35 73.34 41
1 9 50 10 0.30 3301 2600 1255 0
0 1255 0.00 99.82 26.36
1 10 23 10 0.31 3301 2600 1257 0
0 1257 0.00 99.82 26.37 73.33 41
1 10 51 10 0.31 3301 2600 1255 0
0 1255 0.00 99.82 26.36
1 11 24 10 0.32 3301 2600 1257 0
0 1257 0.00 99.81 26.62 73.06 41
1 11 52 10 0.32 3301 2600 1255 0
4 1251 0.32 99.50 26.62
1 12 25 11 0.33 3301 2600 1257 0
0 1257 0.00 99.81 26.39 73.28 41
1 12 53 11 0.33 3301 2600 1258 0
0 1254 0.00 99.82 26.38
1 13 26 12 0.36 3317 2600 1259 0
0 1258 0.00 99.79 26.41 73.23 39
1 13 54 11 0.34 3301 2600 1255 0
0 1254 0.00 99.82 26.42
1 14 27 12 0.36 3301 2600 1257 0
5 1251 0.24 99.58 26.54 73.10 41
1 14 55 12 0.36 3300 2600 1255 0
0 1254 0.00 99.82 26.54
So it looks like in all tests i'm using core+sibling
But side effect of this is that :
33 * 100.0 = 3300.0 MHz max turbo 28 active cores
33 * 100.0 = 3300.0 MHz max turbo 24 active cores
33 * 100.0 = 3300.0 MHz max turbo 20 active cores
33 * 100.0 = 3300.0 MHz max turbo 14 active cores
34 * 100.0 = 3400.0 MHz max turbo 12 active cores
34 * 100.0 = 3400.0 MHz max turbo 8 active cores
35 * 100.0 = 3500.0 MHz max turbo 4 active cores
37 * 100.0 = 3700.0 MHz max turbo 2 active cores
So more cores = less MHz per core/sibling
>
>> perf top:
>>
>> PerfTop: 77835 irqs/sec kernel:99.7% exact: 0.0% [4000Hz
>> cycles], (all, 56 CPUs)
>> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>>
>> 16.32% [kernel] [k] skb_dst_force
>> 16.30% [kernel] [k] dst_release
>> 15.11% [kernel] [k] rt_cache_valid
>> 12.62% [kernel] [k] ipv4_mtu
> It seems a little strange that these 4 functions are on the top
Yes dono why there is ipv4_mtu called and taking soo much cycles
>
>> 5.60% [kernel] [k] do_raw_spin_lock
> Why is calling/taking this lock? (Use perf call-graph recording).
can be hard to paste it here:)
attached file
>
>> 3.03% [kernel] [k] fib_table_lookup
>> 2.70% [kernel] [k] ip_finish_output2
>> 2.10% [kernel] [k] dev_gro_receive
>> 1.89% [kernel] [k] eth_type_trans
>> 1.81% [kernel] [k] ixgbe_poll
>> 1.15% [kernel] [k] ixgbe_xmit_frame_ring
>> 1.06% [kernel] [k] __build_skb
>> 1.04% [kernel] [k] __dev_queue_xmit
>> 0.97% [kernel] [k] ip_rcv
>> 0.78% [kernel] [k] netif_skb_features
>> 0.74% [kernel] [k] ipt_do_table
> Unloading netfilter modules, will give more performance, but it
> semifake to do so.
Compiled in kernel - only in filter mode - with ipv4+ipv6 - no other
modules conntrack or other .
>> 0.70% [kernel] [k] acpi_processor_ffh_cstate_enter
>> 0.64% [kernel] [k] ip_forward
>> 0.59% [kernel] [k] __netif_receive_skb_core
>> 0.55% [kernel] [k] dev_hard_start_xmit
>> 0.53% [kernel] [k] ip_route_input_rcu
>> 0.53% [kernel] [k] ip_rcv_finish
>> 0.51% [kernel] [k] page_frag_free
>> 0.50% [kernel] [k] kmem_cache_alloc
>> 0.50% [kernel] [k] udp_v4_early_demux
>> 0.44% [kernel] [k] skb_release_data
>> 0.42% [kernel] [k] inet_gro_receive
>> 0.40% [kernel] [k] sch_direct_xmit
>> 0.39% [kernel] [k] __local_bh_enable_ip
>> 0.33% [kernel] [k] netdev_pick_tx
>> 0.33% [kernel] [k] validate_xmit_skb
>> 0.28% [kernel] [k] fib_validate_source
>> 0.27% [kernel] [k] deliver_ptype_list_skb
>> 0.25% [kernel] [k] eth_header
>> 0.23% [kernel] [k] get_dma_ops
>> 0.22% [kernel] [k] skb_network_protocol
>> 0.21% [kernel] [k] ip_output
>> 0.21% [kernel] [k] vlan_dev_hard_start_xmit
>> 0.20% [kernel] [k] ixgbe_alloc_rx_buffers
>> 0.18% [kernel] [k] nf_hook_slow
>> 0.18% [kernel] [k] apic_timer_interrupt
>> 0.18% [kernel] [k] virt_to_head_page
>> 0.18% [kernel] [k] build_skb
>> 0.16% [kernel] [k] swiotlb_map_page
>> 0.16% [kernel] [k] ip_finish_output
>> 0.16% [kernel] [k] udp4_gro_receive
>>
>>
>> RESULTS:
>>
>> CSV format - delimeter ";"
>>
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;1;64;1470912;88247040;1470720;85305530
>> 1;1;64;1470912;88285440;1470977;85335110
>> 2;1;64;1470464;88247040;1470402;85290508
>> 3;1;64;1471424;88262400;1471230;85353728
>> 4;1;64;1468736;88166400;1468672;85201652
>> 5;1;64;1470016;88181760;1469949;85234944
>> 6;1;64;1470720;88247040;1470466;85290624
>> 7;1;64;1471232;88277760;1471167;85346246
>> 8;1;64;1469184;88170240;1469249;85216326
>> 9;1;64;1470592;88227840;1470847;85294394
> Single core 1.47Mpps seems a little low, I would expect 2Mpps.
>
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;2;64;2413120;144802560;2413245;139975924
>> 1;2;64;2415296;144913920;2415356;140098188
>> 2;2;64;2416768;144898560;2416573;140105670
>> 3;2;64;2418176;145056000;2418110;140261806
>> 4;2;64;2416512;144990720;2416509;140172950
>> 5;2;64;2415168;144860160;2414466;140064780
>> 6;2;64;2416960;144983040;2416833;140190930
>> 7;2;64;2413632;144768000;2413568;140001734
>> 8;2;64;2415296;144898560;2414589;140087168
>> 9;2;64;2416576;144963840;2416892;140190930
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;3;64;3419008;205155840;3418882;198239244
>> 1;3;64;3428032;205585920;3427971;198744234
>> 2;3;64;3425472;205536000;3425344;198677260
>> 3;3;64;3425088;205470720;3425156;198603136
>> 4;3;64;3427648;205693440;3426883;198773888
>> 5;3;64;3426880;205670400;3427392;198796044
>> 6;3;64;3429120;205678080;3430140;198848186
>> 7;3;64;3422976;205355520;3423490;198458136
>> 8;3;64;3423168;205336320;3423486;198495372
>> 9;3;64;3424384;205493760;3425538;198617868
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;4;64;4406464;264364800;4405244;255560296
>> 1;4;64;4404672;264349440;4405122;255541504
>> 2;4;64;4402368;264049920;4403326;255188864
>> 3;4;64;4401344;264076800;4400702;255207134
>> 4;4;64;4385536;263074560;4386620;254312716
>> 5;4;64;4386560;263189760;4385404;254379532
>> 6;4;64;4398784;263857920;4399031;255025288
>> 7;4;64;4407232;264445440;4407998;255637900
>> 8;4;64;4413184;264698880;4413758;255875816
>> 9;4;64;4411328;264526080;4411906;255712372
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;5;64;5094464;305871360;5094464;295657262
>> 1;5;64;5090816;305514240;5091201;295274810
>> 2;5;64;5088384;305387520;5089792;295175108
>> 3;5;64;5079296;304869120;5079484;294680368
>> 4;5;64;5092992;305544960;5094207;295349166
>> 5;5;64;5092416;305502720;5093372;295334260
>> 6;5;64;5080896;304896000;5081090;294677004
>> 7;5;64;5085376;305114880;5086401;294933058
>> 8;5;64;5092544;305575680;5092036;295356938
>> 9;5;64;5093056;305652480;5093832;295449506
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;6;64;5705088;342351360;5705784;330965110
>> 1;6;64;5710272;342743040;5707591;331373952
>> 2;6;64;5703424;342182400;5701826;330776552
>> 3;6;64;5708736;342604800;5707963;331147462
>> 4;6;64;5710144;342654720;5712067;331202910
>> 5;6;64;5712064;342777600;5711361;331292288
>> 6;6;64;5710144;342585600;5708607;331144272
>> 7;6;64;5699840;342021120;5697853;330609222
>> 8;6;64;5701184;342124800;5702909;330653592
>> 9;6;64;5711360;342735360;5713283;331247686
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;7;64;6244416;374603520;6243591;362180072
>> 1;7;64;6230912;374016000;6231490;361534126
>> 2;7;64;6244800;374776320;6244866;362224326
>> 3;7;64;6238720;374376960;6238261;361838510
>> 4;7;64;6218816;373079040;6220413;360683962
>> 5;7;64;6224320;373566720;6225086;361017404
>> 6;7;64;6224000;373570560;6221370;360936088
>> 7;7;64;6210048;372741120;6210627;360212654
>> 8;7;64;6231616;374035200;6231537;361445502
>> 9;7;64;6227840;373724160;6228802;361162752
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;8;64;6251840;375144960;6251849;362609678
>> 1;8;64;6250816;375014400;6250881;362547038
>> 2;8;64;6257728;375432960;6257160;362911104
>> 3;8;64;6255552;375325440;6255622;362822074
>> 4;8;64;6243776;374576640;6243270;362120622
>> 5;8;64;6237184;374296320;6237690;361790080
>> 6;8;64;6240960;374415360;6240714;361927366
>> 7;8;64;6222784;373317120;6223746;360854424
>> 8;8;64;6225920;373593600;6227014;361154980
>> 9;8;64;6238528;374304000;6237701;361845238
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;14;64;6486144;389184000;6486135;376236488
>> 1;14;64;6454912;387390720;6454222;374466734
>> 2;14;64;6441152;386480640;6440431;373572780
>> 3;14;64;6450240;386972160;6450870;374070014
>> 4;14;64;6465600;387997440;6467221;375089654
>> 5;14;64;6448384;386860800;6448000;373980230
>> 6;14;64;6452352;387095040;6452148;374168904
>> 7;14;64;6441984;386507520;6443203;373665058
>> 8;14;64;6456704;387340800;6455744;374429092
>> 9;14;64;6464640;387901440;6465218;374949004
>> ID;CPU_CORES / RSS QUEUES;PKT_SIZE;PPS_RX;BPS_RX;PPS_TX;BPS_TX
>> 0;16;64;6939008;416325120;6938696;402411192
>> 1;16;64;6941952;416444160;6941745;402558918
>> 2;16;64;6960576;417584640;6960707;403698718
>> 3;16;64;6940736;416486400;6941820;402503876
>> 4;16;64;6927680;415741440;6927420;401853870
>> 5;16;64;6929792;415687680;6929917;401839196
>> 6;16;64;6950400;416989440;6950661;403026166
>> 7;16;64;6953664;417216000;6953454;403260544
>> 8;16;64;6948480;416851200;6948800;403023266
>> 9;16;64;6924160;415422720;6924092;401542468
> I've seen Linux scale beyond 6.9Mpps, thus I also see this as an
> issue/bug. You could be stalling on DMA TX completion being too slow,
> but you already increased the interval and increased the TX ring queue
> size. You could play with those setting and see if it changes this?
>
> Could you try my napi_monitor tool in:
> https://github.com/netoptimizer/prototype-kernel/tree/master/kernel/samples/bpf
>
> Also provide the output from:
> mpstat -P ALL -u -I SCPU -I SUM 2
with 16 cores / 16 RSS queues
Average: CPU %usr %nice %sys %iowait %irq %soft %steal
%guest %gnice %idle
Average: all 0.00 0.00 0.01 0.00 0.00 28.57 0.00
0.00 0.00 71.42
Average: 0 0.00 0.00 0.04 0.00 0.00 0.08 0.00
0.00 0.00 99.88
Average: 1 0.00 0.00 0.12 0.00 0.00 0.00 0.00
0.00 0.00 99.88
Average: 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 3 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 4 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 5 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 6 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 8 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 9 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 10 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 11 0.08 0.00 0.04 0.00 0.00 0.00 0.00
0.00 0.00 99.88
Average: 12 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 13 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 14 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 15 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 16 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 17 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 18 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 19 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 20 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 21 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 22 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 23 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 24 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 25 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 26 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 27 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 28 0.00 0.00 0.04 0.00 0.00 0.00 0.00
0.00 0.00 99.96
Average: 29 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 30 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 31 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 32 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 33 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 34 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 35 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 36 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 37 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 38 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 39 0.04 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 99.96
Average: 40 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 41 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 42 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 43 0.00 0.00 0.00 0.00 0.00 100.00 0.00
0.00 0.00 0.00
Average: 44 0.00 0.00 0.04 0.17 0.00 0.00 0.00
0.00 0.00 99.79
Average: 45 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 46 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 47 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 48 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 49 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 50 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 51 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 52 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 53 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 54 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: 55 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 100.00
Average: CPU intr/s
Average: all 123596.08
Average: 0 646.38
Average: 1 500.54
Average: 2 511.67
Average: 3 534.25
Average: 4 542.21
Average: 5 531.54
Average: 6 554.58
Average: 7 535.88
Average: 8 544.58
Average: 9 536.42
Average: 10 575.46
Average: 11 601.12
Average: 12 502.08
Average: 13 575.46
Average: 14 5917.92
Average: 15 5949.58
Average: 16 7021.29
Average: 17 7299.71
Average: 18 7391.67
Average: 19 7354.25
Average: 20 7543.42
Average: 21 7354.25
Average: 22 7322.33
Average: 23 7368.71
Average: 24 7429.00
Average: 25 7406.46
Average: 26 7400.67
Average: 27 7447.21
Average: 28 517.00
Average: 29 549.54
Average: 30 529.33
Average: 31 533.83
Average: 32 541.25
Average: 33 541.17
Average: 34 532.50
Average: 35 545.17
Average: 36 528.96
Average: 37 509.92
Average: 38 520.12
Average: 39 523.29
Average: 40 530.75
Average: 41 542.33
Average: 42 5921.71
Average: 43 5949.42
Average: 44 503.04
Average: 45 542.75
Average: 46 582.50
Average: 47 581.71
Average: 48 495.29
Average: 49 524.38
Average: 50 527.92
Average: 51 528.12
Average: 52 456.38
Average: 53 477.00
Average: 54 440.92
Average: 55 568.83
Average: CPU HI/s TIMER/s NET_TX/s NET_RX/s BLOCK/s
IRQ_POLL/s TASKLET/s SCHED/s HRTIMER/s RCU/s
Average: 0 0.00 250.00 0.17 87.00 0.00
0.00 45.46 250.00 0.00 13.75
Average: 1 0.00 233.42 0.00 0.00 0.00
0.00 0.00 249.92 0.00 17.21
Average: 2 0.00 249.04 0.00 0.00 0.00
0.00 0.00 249.96 0.00 12.67
Average: 3 0.00 249.92 0.00 0.00 0.00
0.00 0.00 249.92 0.00 34.42
Average: 4 0.00 248.67 0.17 0.00 0.00
0.00 0.00 249.96 0.00 43.42
Average: 5 0.00 249.46 0.00 0.00 0.00
0.00 0.00 249.92 0.00 32.17
Average: 6 0.00 249.79 0.00 0.00 0.00
0.00 0.00 249.87 0.00 54.92
Average: 7 0.00 240.12 0.00 0.00 0.00
0.00 0.00 249.96 0.00 45.79
Average: 8 0.00 247.42 0.00 0.00 0.00
0.00 0.00 249.92 0.00 47.25
Average: 9 0.00 249.29 0.00 0.00 0.00
0.00 0.00 249.96 0.00 37.17
Average: 10 0.00 248.75 0.00 0.00 0.00
0.00 0.00 249.92 0.00 76.79
Average: 11 0.00 249.29 0.00 0.00 0.00
0.00 42.79 249.83 0.00 59.21
Average: 12 0.00 249.83 0.00 0.00 0.00
0.00 0.00 249.96 0.00 2.29
Average: 13 0.00 249.92 0.00 0.00 0.00
0.00 0.00 249.92 0.00 75.62
Average: 14 0.00 148.21 0.17 5758.04 0.00
0.00 0.00 8.42 0.00 3.08
Average: 15 0.00 148.42 0.46 5789.25 0.00
0.00 0.00 8.33 0.00 3.12
Average: 16 0.00 142.62 0.79 6866.46 0.00
0.00 0.00 8.29 0.00 3.12
Average: 17 0.00 143.17 0.42 7145.00 0.00
0.00 0.00 8.08 0.00 3.04
Average: 18 0.00 153.62 0.42 7226.42 0.00
0.00 0.00 8.04 0.00 3.17
Average: 19 0.00 150.46 0.46 7192.21 0.00
0.00 0.00 8.04 0.00 3.08
Average: 20 0.00 145.21 0.17 7386.50 0.00
0.00 0.00 8.29 0.00 3.25
Average: 21 0.00 150.96 0.46 7191.37 0.00
0.00 0.00 8.25 0.00 3.21
Average: 22 0.00 146.67 0.54 7163.96 0.00
0.00 0.00 8.04 0.00 3.12
Average: 23 0.00 151.38 0.42 7205.75 0.00
0.00 0.00 8.00 0.00 3.17
Average: 24 0.00 153.33 0.17 7264.12 0.00
0.00 0.00 8.08 0.00 3.29
Average: 25 0.00 153.21 0.17 7241.83 0.00
0.00 0.00 7.96 0.00 3.29
Average: 26 0.00 153.96 0.17 7234.88 0.00
0.00 0.00 8.38 0.00 3.29
Average: 27 0.00 151.71 0.79 7283.25 0.00
0.00 0.00 8.04 0.00 3.42
Average: 28 0.00 245.71 0.00 0.00 0.00
0.00 0.00 249.50 0.00 21.79
Average: 29 0.00 233.21 0.00 0.00 0.00
0.00 0.00 249.87 0.00 66.46
Average: 30 0.00 248.92 0.00 0.00 0.00
0.00 0.00 250.00 0.00 30.42
Average: 31 0.00 249.92 0.00 0.00 0.00
0.00 0.00 249.96 0.00 33.96
Average: 32 0.00 248.67 0.00 0.00 0.00
0.00 0.00 249.96 0.00 42.62
Average: 33 0.00 249.46 0.00 0.00 0.00
0.00 0.00 249.92 0.00 41.79
Average: 34 0.00 249.79 0.00 0.00 0.00
0.00 0.00 249.87 0.00 32.83
Average: 35 0.00 240.12 0.00 0.00 0.00
0.00 0.00 249.96 0.00 55.08
Average: 36 0.00 247.42 0.00 0.00 0.00
0.00 0.00 249.96 0.00 31.58
Average: 37 0.00 249.29 0.00 0.00 0.00
0.00 0.00 249.92 0.00 10.71
Average: 38 0.00 248.75 0.00 0.00 0.00
0.00 0.00 249.87 0.00 21.50
Average: 39 0.00 249.50 0.00 0.00 0.00
0.00 0.00 249.83 0.00 23.96
Average: 40 0.00 249.83 0.00 0.00 0.00
0.00 0.00 249.96 0.00 30.96
Average: 41 0.00 249.92 0.00 0.00 0.00
0.00 0.00 249.92 0.00 42.50
Average: 42 0.00 148.38 0.71 5761.00 0.00
0.00 0.00 8.25 0.00 3.38
Average: 43 0.00 147.21 0.50 5790.33 0.00
0.00 0.00 8.00 0.00 3.38
Average: 44 0.00 248.96 0.00 0.00 0.00
0.00 0.00 248.13 0.00 5.96
Average: 45 0.00 249.04 0.00 0.00 0.00
0.00 0.00 248.88 0.00 44.83
Average: 46 0.00 248.96 0.00 0.00 0.00
0.00 0.00 248.58 0.00 84.96
Average: 47 0.00 249.00 0.00 0.00 0.00
0.00 0.00 248.75 0.00 83.96
Average: 48 0.00 249.12 0.00 0.00 0.00
0.00 0.00 132.83 0.00 113.33
Average: 49 0.00 249.12 0.00 0.00 0.00
0.00 0.00 248.62 0.00 26.62
Average: 50 0.00 248.92 0.00 0.00 0.00
0.00 0.00 248.58 0.00 30.42
Average: 51 0.00 249.08 0.00 0.00 0.00
0.00 0.00 248.42 0.00 30.63
Average: 52 0.00 249.21 0.00 0.00 0.00
0.00 0.00 131.96 0.00 75.21
Average: 53 0.00 249.08 0.00 0.00 0.00
0.00 0.00 136.12 0.00 91.79
Average: 54 0.00 249.00 0.00 0.00 0.00
0.00 0.00 136.79 0.00 55.12
Average: 55 0.00 249.04 0.00 0.00 0.00
0.00 0.00 248.71 0.00 71.08
>
View attachment "perf-report" of type "text/plain" (2361492 bytes)
Powered by blists - more mailing lists