selftests: seccomp: fix format-zero-length warnings

fix the following errors by using string format specifier and an empty
parameter:

seccomp_benchmark.c:197:24: warning: zero-length gnu_printf for

selftests: seccomp: fix format-zero-length warnings

fix the following errors by using string format specifier and an empty
parameter:

seccomp_benchmark.c:197:24: warning: zero-length gnu_printf format
string [-Wformat-zero-length]
197 | ksft_print_msg("");
| ^~
seccomp_benchmark.c:202:24: warning: zero-length gnu_printf format
string [-Wformat-zero-length]
202 | ksft_print_msg("");
| ^~
seccomp_benchmark.c:204:24: warning: zero-length gnu_printf format
string [-Wformat-zero-length]
204 | ksft_print_msg("");
| ^~

Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Suggested-by: Kees Cook <[email protected]>
Signed-off-by: Amer Al Shanawany <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

selftests/seccomp: Pin benchmark to single CPU

The seccomp benchmark test (for validating the benefit of bitmaps) can
be sensitive to scheduling speed, so pin the process to a single CPU,
which appe

selftests/seccomp: Pin benchmark to single CPU

The seccomp benchmark test (for validating the benefit of bitmaps) can
be sensitive to scheduling speed, so pin the process to a single CPU,
which appears to significantly improve reliability, and loosen the
"close enough" checking to allow up to 10% variance instead of 1%.

Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-lkp/[email protected]
Cc: Andy Lutomirski <[email protected]>
Cc: Will Drewry <[email protected]>
Reviewed-by: Mark Brown <[email protected]>
Signed-off-by: Kees Cook <[email protected]>

show more ...

kselftest/seccomp: Report each expectation we assert as a KTAP test

The seccomp benchmark test makes a number of checks on the performance it
measures and logs them to the output but does so in a cu

kselftest/seccomp: Report each expectation we assert as a KTAP test

The seccomp benchmark test makes a number of checks on the performance it
measures and logs them to the output but does so in a custom format which
none of the automated test runners understand meaning that the chances that
anyone is paying attention are slim. Let's additionally log each result in
KTAP format so that automated systems parsing the test output will see each
comparison as a test case. The original logs are left in place since they
provide the actual numbers for analysis.

As part of this rework the flow for the main program so that when we skip
tests we still log all the tests we skip, this is because the standard KTAP
headers and footers include counts of the number of expected and run tests.

Tested-by: Anders Roxell <[email protected]>
Acked-by: Kees Cook <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

kselftest/seccomp: Use kselftest output functions for benchmark

In preparation for trying to output the test results themselves in TAP
format rework all the prints in the benchmark to use the kselft

kselftest/seccomp: Use kselftest output functions for benchmark

In preparation for trying to output the test results themselves in TAP
format rework all the prints in the benchmark to use the kselftest output
functions. The uses of system() all produce single line output so we can
avoid having to deal with fully managing the child process and continue to
use system() by simply printing an empty message before we invoke system().
We also leave one printf() used to complete a line of output in place.

Tested-by: Anders Roxell <[email protected]>
Acked-by: Kees Cook <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

selftests/seccomp: remove ARRAY_SIZE define from seccomp_benchmark

ARRAY_SIZE is defined in several selftests. Remove definitions from
individual test files and include header file for the define in

selftests/seccomp: remove ARRAY_SIZE define from seccomp_benchmark

ARRAY_SIZE is defined in several selftests. Remove definitions from
individual test files and include header file for the define instead.
ARRAY_SIZE define is added in a separate patch to prepare for this
change.

Remove ARRAY_SIZE from seccomp_benchmark and pickup the one defined in
kselftest.h.

Signed-off-by: Shuah Khan <[email protected]>

show more ...

selftests/seccomp: Avoid using "sysctl" for report

Instead of depending on "sysctl" being installed, just use "grep -H" for
sysctl status reporting. Additionally report kernel version for easier
com

selftests/seccomp: Avoid using "sysctl" for report

Instead of depending on "sysctl" being installed, just use "grep -H" for
sysctl status reporting. Additionally report kernel version for easier
comparisons.

Signed-off-by: Kees Cook <[email protected]>

show more ...

selftests/seccomp: Flush benchmark output

When running the seccomp benchmark under a test runner, it wouldn't
provide any feedback on progress. Set stdout unbuffered.

Suggested-by: Will Drewry <wad

selftests/seccomp: Flush benchmark output

When running the seccomp benchmark under a test runner, it wouldn't
provide any feedback on progress. Set stdout unbuffered.

Suggested-by: Will Drewry <[email protected]>
Signed-off-by: Kees Cook <[email protected]>

show more ...

selftests/seccomp: Compare bitmap vs filter overhead

As part of the seccomp benchmarking, include the expectations with
regard to the timing behavior of the constant action bitmaps, and report
incon

selftests/seccomp: Compare bitmap vs filter overhead

As part of the seccomp benchmarking, include the expectations with
regard to the timing behavior of the constant action bitmaps, and report
inconsistencies better.

Example output with constant action bitmaps on x86:

$ sudo ./seccomp_benchmark 100000000
Current BPF sysctl settings:
net.core.bpf_jit_enable = 1
net.core.bpf_jit_harden = 0
Benchmarking 200000000 syscalls...
129.359381409 - 0.008724424 = 129350656985 (129.4s)
getpid native: 646 ns
264.385890006 - 129.360453229 = 135025436777 (135.0s)
getpid RET_ALLOW 1 filter (bitmap): 675 ns
399.400511893 - 264.387045901 = 135013465992 (135.0s)
getpid RET_ALLOW 2 filters (bitmap): 675 ns
545.872866260 - 399.401718327 = 146471147933 (146.5s)
getpid RET_ALLOW 3 filters (full): 732 ns
696.337101319 - 545.874097681 = 150463003638 (150.5s)
getpid RET_ALLOW 4 filters (full): 752 ns
Estimated total seccomp overhead for 1 bitmapped filter: 29 ns
Estimated total seccomp overhead for 2 bitmapped filters: 29 ns
Estimated total seccomp overhead for 3 full filters: 86 ns
Estimated total seccomp overhead for 4 full filters: 106 ns
Estimated seccomp entry overhead: 29 ns
Estimated seccomp per-filter overhead (last 2 diff): 20 ns
Estimated seccomp per-filter overhead (filters / 4): 19 ns
Expectations:
native ≤ 1 bitmap (646 ≤ 675): ✔️
native ≤ 1 filter (646 ≤ 732): ✔️
per-filter (last 2 diff) ≈ per-filter (filters / 4) (20 ≈ 19): ✔️
1 bitmapped ≈ 2 bitmapped (29 ≈ 29): ✔️
entry ≈ 1 bitmapped (29 ≈ 29): ✔️
entry ≈ 2 bitmapped (29 ≈ 29): ✔️
native + entry + (per filter * 4) ≈ 4 filters total (755 ≈ 752): ✔️

[YiFei: Changed commit message to show stats for this patch series]
Signed-off-by: Kees Cook <[email protected]>
Link: https://lore.kernel.org/r/1b61df3db85c5f7f1b9202722c45e7b39df73ef2.1602431034.git.yifeifz2@illinois.edu

show more ...

selftests/seccomp: Improve calibration loop

The seccomp benchmark calibration loop did not need to take so long.
Instead, use a simple 1 second timeout and multiply up to target. It
does not need to

selftests/seccomp: Improve calibration loop

The seccomp benchmark calibration loop did not need to take so long.
Instead, use a simple 1 second timeout and multiply up to target. It
does not need to be accurate.

Signed-off-by: Kees Cook <[email protected]>

show more ...

selftests/seccomp: Expand benchmark to per-filter measurements

It's useful to see how much (at a minimum) each filter adds to the
syscall overhead. Add additional calculations.

Signed-off-by: Kees

selftests/seccomp: Expand benchmark to per-filter measurements

It's useful to see how much (at a minimum) each filter adds to the
syscall overhead. Add additional calculations.

Signed-off-by: Kees Cook <[email protected]>

show more ...

selftests/seccomp: Add simple seccomp overhead benchmark

This attempts to produce a comparison between native getpid() and a
RET_ALLOW-filtered getpid(), to measure the overhead cost of using
seccom

selftests/seccomp: Add simple seccomp overhead benchmark

This attempts to produce a comparison between native getpid() and a
RET_ALLOW-filtered getpid(), to measure the overhead cost of using
seccomp().

Signed-off-by: Kees Cook <[email protected]>

show more ...