Message ID | 20210829183608.2297877-1-me@ubique.spb.ru |
---|---|
Headers | show |
Series | bpfilter | expand |
On Sun, Aug 29, 2021 at 10:36:07PM +0400, Dmitrii Banshchikov wrote: > /* > - * # Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > + * Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > * *filter ... > - * -A LOCAL -s 10.32.0.0/11 -j FROMDC > - * -A LOCAL -s 10.144.0.0/12 -j FROMDC > - * -A LOCAL -s 10.160.0.0/12 -j FROMDC > - * -A LOCAL -s 10.0.0.0/12 -j FROMDC > - * -A LOCAL -s 10.248.0.0/24 -j FROMDC > - * -A LOCAL -s 10.232.0.0/16 -j FROMDC > - * -A LOCAL -s 10.1.146.131/32 -p udp -m udp --dport 161 -j ACCEPT > - * -A LOCAL -s 10.149.118.14/32 -p udp -m udp --dport 161 -j ACCEPT > - * -A LOCAL -p icmp -j ACCEPT > + * :INPUT ACCEPT [0:0] > + * :FORWARD ACCEPT [0:0] > + * :OUTPUT ACCEPT [0:0] > + * -A INPUT -s 1.1.1.1/32 -d 2.2.2.2/32 -j DROP > + * -A INPUT -s 2.2.0.0/16 -d 3.0.0.0/8 -j DROP > + * -A INPUT -p udp -m udp --sport 100 --dport 500 -j DROP > * COMMIT > */ Patch 10 adds this test, but then patch 12 removes most of it? Keep both? Also hit this on my system with older glibc: ../net/bpfilter/codegen.c: In function ‘codegen_push_subprog’: ../net/bpfilter/codegen.c:67:4: warning: implicit declaration of function ‘reallocarray’ [-Wimplicit-function-declaration] 67 | reallocarray(codegen->subprogs, subprogs_max, sizeof(codegen->subprogs[0])); | ^~~~~~~~~~~~ ../net/bpfilter/codegen.c:66:12: warning: assignment to ‘struct codegen_subprog_desc **’ from ‘int’ makes pointer from integer without a cast [-Wint-conversion] 66 | subprogs = | ^ In libbpf we have libbpf_reallocarray() for this reason. Could you provide an example of generated bpf program? And maybe add Documentation/bpf/bpfilter_design.rst ? The tests don't build for me: $ cd selftests/bpf/bpfilter; make make: *** No rule to make target '-lelf', needed by '.../selftests/bpf/bpfilter/test_match'. Stop. The unit tests are great, btw. test_codegen is not end-to-end, right? Could you add a full test with iptable command line? or netns support is a prerequisite for it?
On Mon, Aug 30, 2021 at 12:45:45PM -0700, Alexei Starovoitov wrote: > On Sun, Aug 29, 2021 at 10:36:07PM +0400, Dmitrii Banshchikov wrote: > > /* > > - * # Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > > + * Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > > * *filter > ... > > - * -A LOCAL -s 10.32.0.0/11 -j FROMDC > > - * -A LOCAL -s 10.144.0.0/12 -j FROMDC > > - * -A LOCAL -s 10.160.0.0/12 -j FROMDC > > - * -A LOCAL -s 10.0.0.0/12 -j FROMDC > > - * -A LOCAL -s 10.248.0.0/24 -j FROMDC > > - * -A LOCAL -s 10.232.0.0/16 -j FROMDC > > - * -A LOCAL -s 10.1.146.131/32 -p udp -m udp --dport 161 -j ACCEPT > > - * -A LOCAL -s 10.149.118.14/32 -p udp -m udp --dport 161 -j ACCEPT > > - * -A LOCAL -p icmp -j ACCEPT > > + * :INPUT ACCEPT [0:0] > > + * :FORWARD ACCEPT [0:0] > > + * :OUTPUT ACCEPT [0:0] > > + * -A INPUT -s 1.1.1.1/32 -d 2.2.2.2/32 -j DROP > > + * -A INPUT -s 2.2.0.0/16 -d 3.0.0.0/8 -j DROP > > + * -A INPUT -p udp -m udp --sport 100 --dport 500 -j DROP > > * COMMIT > > */ > > Patch 10 adds this test, but then patch 12 removes most of it? > Keep both? Sorry, I missed it. I decided that the large blob looks really ugly and switched to the smaller one and forgot to cleanup the patchset. > > Also hit this on my system with older glibc: > > ../net/bpfilter/codegen.c: In function ‘codegen_push_subprog’: > ../net/bpfilter/codegen.c:67:4: warning: implicit declaration of function ‘reallocarray’ [-Wimplicit-function-declaration] > 67 | reallocarray(codegen->subprogs, subprogs_max, sizeof(codegen->subprogs[0])); > | ^~~~~~~~~~~~ > ../net/bpfilter/codegen.c:66:12: warning: assignment to ‘struct codegen_subprog_desc **’ from ‘int’ makes pointer from integer without a cast [-Wint-conversion] > 66 | subprogs = > | ^ > > In libbpf we have libbpf_reallocarray() for this reason. > > Could you provide an example of generated bpf program? > And maybe add Documentation/bpf/bpfilter_design.rst ? I will add documentation in the next iteration when bpf_map_for_each() subprog will be introduced. > > The tests don't build for me: > $ cd selftests/bpf/bpfilter; make > make: *** No rule to make target '-lelf', needed by '.../selftests/bpf/bpfilter/test_match'. Stop. libelf was added because libbpf depends on it. Are you able to build libbpf? > > The unit tests are great, btw. test_codegen is not end-to-end, right? > Could you add a full test with iptable command line? > or netns support is a prerequisite for it? Yeah, as net namespaces aren't supported using iptables binary will modify the root namespace. That is the reason why codegen tests aren't implemented in the end-to-end fashion and rules are represented by blobs. -- Dmitrii Banshchikov
On Tue, Aug 31, 2021 at 12:54:43AM +0400, Dmitrii Banshchikov wrote: > On Mon, Aug 30, 2021 at 12:45:45PM -0700, Alexei Starovoitov wrote: > > On Sun, Aug 29, 2021 at 10:36:07PM +0400, Dmitrii Banshchikov wrote: > > > /* > > > - * # Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > > > + * Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > > > * *filter > > ... > > > - * -A LOCAL -s 10.32.0.0/11 -j FROMDC > > > - * -A LOCAL -s 10.144.0.0/12 -j FROMDC > > > - * -A LOCAL -s 10.160.0.0/12 -j FROMDC > > > - * -A LOCAL -s 10.0.0.0/12 -j FROMDC > > > - * -A LOCAL -s 10.248.0.0/24 -j FROMDC > > > - * -A LOCAL -s 10.232.0.0/16 -j FROMDC > > > - * -A LOCAL -s 10.1.146.131/32 -p udp -m udp --dport 161 -j ACCEPT > > > - * -A LOCAL -s 10.149.118.14/32 -p udp -m udp --dport 161 -j ACCEPT > > > - * -A LOCAL -p icmp -j ACCEPT > > > + * :INPUT ACCEPT [0:0] > > > + * :FORWARD ACCEPT [0:0] > > > + * :OUTPUT ACCEPT [0:0] > > > + * -A INPUT -s 1.1.1.1/32 -d 2.2.2.2/32 -j DROP > > > + * -A INPUT -s 2.2.0.0/16 -d 3.0.0.0/8 -j DROP > > > + * -A INPUT -p udp -m udp --sport 100 --dport 500 -j DROP > > > * COMMIT > > > */ > > > > Patch 10 adds this test, but then patch 12 removes most of it? > > Keep both? > > Sorry, I missed it. > I decided that the large blob looks really ugly and switched to > the smaller one and forgot to cleanup the patchset. > > > > > Also hit this on my system with older glibc: > > > > ../net/bpfilter/codegen.c: In function ‘codegen_push_subprog’: > > ../net/bpfilter/codegen.c:67:4: warning: implicit declaration of function ‘reallocarray’ [-Wimplicit-function-declaration] > > 67 | reallocarray(codegen->subprogs, subprogs_max, sizeof(codegen->subprogs[0])); > > | ^~~~~~~~~~~~ > > ../net/bpfilter/codegen.c:66:12: warning: assignment to ‘struct codegen_subprog_desc **’ from ‘int’ makes pointer from integer without a cast [-Wint-conversion] > > 66 | subprogs = > > | ^ > > > > In libbpf we have libbpf_reallocarray() for this reason. > > > > Could you provide an example of generated bpf program? > > And maybe add Documentation/bpf/bpfilter_design.rst ? > > I will add documentation in the next iteration when > bpf_map_for_each() subprog will be introduced. > > > > > The tests don't build for me: > > $ cd selftests/bpf/bpfilter; make > > make: *** No rule to make target '-lelf', needed by '.../selftests/bpf/bpfilter/test_match'. Stop. > > libelf was added because libbpf depends on it. > Are you able to build libbpf? make proceeds to build libbpf just fine, but then it stops with above message. I manually removed -lelf from Makefile. Then run make to see it fail linking and then manually copy pasted gcc command to build it with additional -lelf command line. fwiw make -v GNU Make 4.2.1 > > > > The unit tests are great, btw. test_codegen is not end-to-end, right? > > Could you add a full test with iptable command line? > > or netns support is a prerequisite for it? > > Yeah, as net namespaces aren't supported using iptables binary > will modify the root namespace. That is the reason why codegen > tests aren't implemented in the end-to-end fashion and rules are > represented by blobs. I think when ifindex is no longer hardcoded the netns support doesn't have to be gating. The generic xdp attached to veth in netns should work to do end-to-end test. bpftiler would need to do a bit of magic to figure out the right ifindex. Or we can extend kernel with ifindex-less generic XDP.
On 2021-08-29 2:35 p.m., Dmitrii Banshchikov wrote: [..] > And here are some performance tests. > > The environment consists of two machines(sender and receiver) > connected with 10Gbps link via switch. The sender uses DPDK to > simulate QUIC packets(89 bytes long) from random IP. The switch > measures the generated traffic to be about 7066377568 bits/sec, > 9706553 packets/sec. > > The receiver is a 2 socket 2680v3 + HT and uses either iptables, > nft or bpfilter to filter out UDP traffic. > > Two tests were made. Two rulesets(default policy was to ACCEPT) > were used in each test: > > ``` > iptables -A INPUT -p udp -m udp --dport 1500 -j DROP > ``` > and > ``` > iptables -A INPUT -s 1.1.1.1/32 -p udp -m udp --dport 1000 -j DROP > iptables -A INPUT -s 2.2.2.2/32 -p udp -m udp --dport 2000 -j DROP > ... > iptables -A INPUT -s 31.31.31.31/32 -p udp -m udp --dport 31000 -j DROP > iptables -A INPUT -p udp -m udp --dport 1500 -j DROP > ``` > > The first test measures performance of the receiver via stress-ng > [3] in bogo-ops. The upper-bound(there are no firewall and no > traffic) value for bogo-ops is 8148-8210. The lower bound value > (there is traffic but no firewall) is 6567-6643. > The stress-ng command used: stress-ng -t60 -c48 --metrics-brief. > > The second test measures the number the of dropped packets. The > receiver keeps only 1 CPU online and disables all > others(maxcpus=1 and set number of cores per socket to 1 in > BIOS). The number of the dropped packets is collected via > iptables-legacy -nvL, iptables -nvL and bpftool map dump id. > > Test 1: bogo-ops(the more the better) > iptables nft bpfilter > 1 rule: 6474-6554 6483-6515 7996-8008 > 32 rules: 6374-6433 5761-5804 7997-8042 > > > Test 2: number of dropped packets(the more the better) > iptables nft bpfilter > 1 rule: 234M-241M 220M 900M+ > 32 rules: 186M-196M 97M-98M 900M+ > > > Please let me know if you see a gap in the testing environment. General perf testing will depend on the nature of the use case you are trying to target. What is the nature of the app? Is it just receiving packets and counting? Does it exemplify something something real in your network or is just purely benchmarking? Both are valid. What else can it do (eg are you interested in latency accounting etc)? What i have seen in practise for iptables deployments is a default drop and in general an accept list. Per ruleset IP address aggregation is typically achieved via ipset. So your mileage may vary... Having said that: Our testing[1] approach is typically for a worst case scenario. i.e we make sure you structure the rulesets such that all of the linear rulesets will be iterated and we eventually hit the default ruleset. We also try to reduce variability in the results. A lot of small things could affect your reproducibility, so we try to avoid them. For example, from what you described: You are sending from a random IP - that means each packet will hit a random ruleset (for the case of 32 rulesets). And some rules will likely be hit more often than others. The likelihood of reproducing the same results for multiple runs gets lower as you increase the number of rulesets. From a collection perspective: Looking at the nature of the CPU utilization is important Softirq vs system calls vs user app. Your test workload seems to be very specific to ingress host. So in reality you are more constrained by kernel->user syscalls (which will be hidden if you are mostly dropping in the kernel as opposed to letting packets go to user space). Something is not clear from your email: You seem to indicate that no traffic was running in test 1. If so, why would 32 rulesets give different results than 1? cheers, jamal [1] https://netdevconf.info/0x15/session.html?Linux-ACL-Performance-Analysis
On Mon, Aug 30, 2021 at 09:56:18PM -0400, Jamal Hadi Salim wrote: > On 2021-08-29 2:35 p.m., Dmitrii Banshchikov wrote: > > [..] > > > And here are some performance tests. > > > > The environment consists of two machines(sender and receiver) > > connected with 10Gbps link via switch. The sender uses DPDK to > > simulate QUIC packets(89 bytes long) from random IP. The switch > > measures the generated traffic to be about 7066377568 bits/sec, > > 9706553 packets/sec. > > > > The receiver is a 2 socket 2680v3 + HT and uses either iptables, > > nft or bpfilter to filter out UDP traffic. > > > > Two tests were made. Two rulesets(default policy was to ACCEPT) > > were used in each test: > > > > ``` > > iptables -A INPUT -p udp -m udp --dport 1500 -j DROP > > ``` > > and > > ``` > > iptables -A INPUT -s 1.1.1.1/32 -p udp -m udp --dport 1000 -j DROP > > iptables -A INPUT -s 2.2.2.2/32 -p udp -m udp --dport 2000 -j DROP > > ... > > iptables -A INPUT -s 31.31.31.31/32 -p udp -m udp --dport 31000 -j DROP > > iptables -A INPUT -p udp -m udp --dport 1500 -j DROP > > ``` > > > > The first test measures performance of the receiver via stress-ng > > [3] in bogo-ops. The upper-bound(there are no firewall and no > > traffic) value for bogo-ops is 8148-8210. The lower bound value > > (there is traffic but no firewall) is 6567-6643. > > The stress-ng command used: stress-ng -t60 -c48 --metrics-brief. > > > > The second test measures the number the of dropped packets. The > > receiver keeps only 1 CPU online and disables all > > others(maxcpus=1 and set number of cores per socket to 1 in > > BIOS). The number of the dropped packets is collected via > > iptables-legacy -nvL, iptables -nvL and bpftool map dump id. > > > > Test 1: bogo-ops(the more the better) > > iptables nft bpfilter > > 1 rule: 6474-6554 6483-6515 7996-8008 > > 32 rules: 6374-6433 5761-5804 7997-8042 > > > > > > Test 2: number of dropped packets(the more the better) > > iptables nft bpfilter > > 1 rule: 234M-241M 220M 900M+ > > 32 rules: 186M-196M 97M-98M 900M+ > > > > > > Please let me know if you see a gap in the testing environment. > > General perf testing will depend on the nature of the use case > you are trying to target. > What is the nature of the app? Is it just receiving packets and > counting? Does it exemplify something something real in your > network or is just purely benchmarking? Both are valid. > What else can it do (eg are you interested in latency accounting etc)? > What i have seen in practise for iptables deployments is a default > drop and in general an accept list. Per ruleset IP address aggregation > is typically achieved via ipset. So your mileage may vary... This was a pure benchmarking with the single goal - show that there might exist scenarios when using bpfilter may provide some performance benefits. > > Having said that: > Our testing[1] approach is typically for a worst case scenario. > i.e we make sure you structure the rulesets such that all of the > linear rulesets will be iterated and we eventually hit the default > ruleset. > We also try to reduce variability in the results. A lot of small > things could affect your reproducibility, so we try to avoid them. > For example, from what you described: > You are sending from a random IP - that means each packet will hit > a random ruleset (for the case of 32 rulesets). And some rules will > likely be hit more often than others. The likelihood of reproducing the > same results for multiple runs gets lower as you increase the number > of rulesets. > From a collection perspective: > Looking at the nature of the CPU utilization is important > Softirq vs system calls vs user app. > Your test workload seems to be very specific to ingress host. > So in reality you are more constrained by kernel->user syscalls > (which will be hidden if you are mostly dropping in the kernel > as opposed to letting packets go to user space). > > Something is not clear from your email: > You seem to indicate that no traffic was running in test 1. > If so, why would 32 rulesets give different results than 1? I mentioned the lower and upper bound values for bogo-ops on the machine. The lower bound is when there is traffic and no firewall at all. The upper bound is when there is no firewall and no traffic. Then the first test measures bogo-ops for two rule sets when there is traffic for either iptables, nft or bpfilter. > > cheers, > jamal > > [1] https://netdevconf.info/0x15/session.html?Linux-ACL-Performance-Analysis -- Dmitrii Banshchikov
On Mon, Aug 30, 2021 at 04:45:15PM -0700, Alexei Starovoitov wrote: > On Tue, Aug 31, 2021 at 12:54:43AM +0400, Dmitrii Banshchikov wrote: > > On Mon, Aug 30, 2021 at 12:45:45PM -0700, Alexei Starovoitov wrote: > > > On Sun, Aug 29, 2021 at 10:36:07PM +0400, Dmitrii Banshchikov wrote: > > > > /* > > > > - * # Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > > > > + * Generated by iptables-save v1.8.2 on Sat May 8 05:22:41 2021 > > > > * *filter > > > ... > > > > - * -A LOCAL -s 10.32.0.0/11 -j FROMDC > > > > - * -A LOCAL -s 10.144.0.0/12 -j FROMDC > > > > - * -A LOCAL -s 10.160.0.0/12 -j FROMDC > > > > - * -A LOCAL -s 10.0.0.0/12 -j FROMDC > > > > - * -A LOCAL -s 10.248.0.0/24 -j FROMDC > > > > - * -A LOCAL -s 10.232.0.0/16 -j FROMDC > > > > - * -A LOCAL -s 10.1.146.131/32 -p udp -m udp --dport 161 -j ACCEPT > > > > - * -A LOCAL -s 10.149.118.14/32 -p udp -m udp --dport 161 -j ACCEPT > > > > - * -A LOCAL -p icmp -j ACCEPT > > > > + * :INPUT ACCEPT [0:0] > > > > + * :FORWARD ACCEPT [0:0] > > > > + * :OUTPUT ACCEPT [0:0] > > > > + * -A INPUT -s 1.1.1.1/32 -d 2.2.2.2/32 -j DROP > > > > + * -A INPUT -s 2.2.0.0/16 -d 3.0.0.0/8 -j DROP > > > > + * -A INPUT -p udp -m udp --sport 100 --dport 500 -j DROP > > > > * COMMIT > > > > */ > > > > > > Patch 10 adds this test, but then patch 12 removes most of it? > > > Keep both? > > > > Sorry, I missed it. > > I decided that the large blob looks really ugly and switched to > > the smaller one and forgot to cleanup the patchset. > > > > > > > > Also hit this on my system with older glibc: > > > > > > ../net/bpfilter/codegen.c: In function ‘codegen_push_subprog’: > > > ../net/bpfilter/codegen.c:67:4: warning: implicit declaration of function ‘reallocarray’ [-Wimplicit-function-declaration] > > > 67 | reallocarray(codegen->subprogs, subprogs_max, sizeof(codegen->subprogs[0])); > > > | ^~~~~~~~~~~~ > > > ../net/bpfilter/codegen.c:66:12: warning: assignment to ‘struct codegen_subprog_desc **’ from ‘int’ makes pointer from integer without a cast [-Wint-conversion] > > > 66 | subprogs = > > > | ^ > > > > > > In libbpf we have libbpf_reallocarray() for this reason. > > > > > > Could you provide an example of generated bpf program? > > > And maybe add Documentation/bpf/bpfilter_design.rst ? > > > > I will add documentation in the next iteration when > > bpf_map_for_each() subprog will be introduced. > > > > > > > > The tests don't build for me: > > > $ cd selftests/bpf/bpfilter; make > > > make: *** No rule to make target '-lelf', needed by '.../selftests/bpf/bpfilter/test_match'. Stop. > > > > libelf was added because libbpf depends on it. > > Are you able to build libbpf? > > make proceeds to build libbpf just fine, but then it stops with above message. > I manually removed -lelf from Makefile. Then run make to see it fail linking > and then manually copy pasted gcc command to build it with additional -lelf > command line. > fwiw make -v > GNU Make 4.2.1 Will take a look on it. Thanks. > > > > > > > The unit tests are great, btw. test_codegen is not end-to-end, right? > > > Could you add a full test with iptable command line? > > > or netns support is a prerequisite for it? > > > > Yeah, as net namespaces aren't supported using iptables binary > > will modify the root namespace. That is the reason why codegen > > tests aren't implemented in the end-to-end fashion and rules are > > represented by blobs. > > I think when ifindex is no longer hardcoded the netns support > doesn't have to be gating. The generic xdp attached to veth in netns > should work to do end-to-end test. bpftiler would need to do a bit of magic > to figure out the right ifindex. Or we can extend kernel with ifindex-less > generic XDP. Is it ok to add an external dependency to tests? The unit test will need to execute iptables binary. -- Dmitrii Banshchikov
On 2021-08-31 8:48 a.m., Dmitrii Banshchikov wrote: > On Mon, Aug 30, 2021 at 09:56:18PM -0400, Jamal Hadi Salim wrote: >> On 2021-08-29 2:35 p.m., Dmitrii Banshchikov wrote: >> >> Something is not clear from your email: >> You seem to indicate that no traffic was running in test 1. >> If so, why would 32 rulesets give different results than 1? > > I mentioned the lower and upper bound values for bogo-ops on the > machine. The lower bound is when there is traffic and no firewall > at all. The upper bound is when there is no firewall and no > traffic. Then the first test measures bogo-ops for two rule sets > when there is traffic for either iptables, nft or bpfilter. > Thanks, I totally misread that. I did look at stress-ng initially to use it to stress the system and emulate some real world setup (polluting cache etc) while testing but the variability of the results was bad - so dropped it earlier. Maybe we should look at it again. cheers, jamal
On Tue, Aug 31, 2021 at 5:52 AM Dmitrii Banshchikov <me@ubique.spb.ru> wrote: > > Is it ok to add an external dependency to tests? The unit test > will need to execute iptables binary. Ideally not, but sometimes it's unavoidable. iptables cmd is generally available. selftests/bpf already have few tests that shell out to it. They're not part of test_progs though.