Skip to content

bpf: add cpu time counter kfuncs

Puranjay Mohan bpf 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v13 Latest6 patches 4 repliesIn review

Changes in v13 from the cover letter

- Added support for arm64
- Rebased on latest bpf-next/master
- Changed the word helper to kfunc in the commit message to not confuse with bpf helpers.

Patches in v13 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/6bpf: adjust BPF JIT dependency to BPF_SYSCALLkernel/bpf/Kconfig+1 -10 replies
2/6bpf: add bpf_get_cpu_time_counter kfuncarch/x86/net/bpf_jit_comp.c arch/x86/net/bpf_jit_comp32.c include/linux/bpf.h +4 more+75 -10 replies
3/6bpf: add bpf_cpu_time_counter_to_ns kfuncarch/x86/net/bpf_jit_comp.c arch/x86/net/bpf_jit_comp32.c include/linux/bpf.h +2 more+43 -40 replies
4/6selftests/bpf: add selftest to check bpf_get_cpu_time_counter jittools/testing/selftests/bpf/prog_tests/verifier.c tools/testing/selftests/bpf/progs/verifier_cpu_cycles.c+122 -00 replies
5/6selftests/bpf: add usage example for cpu time counter kfuncstools/testing/selftests/bpf/prog_tests/test_cpu_cycles.c tools/testing/selftests/bpf/progs/test_cpu_cycles.c+60 -00 replies
6/6bpf, arm64: Add JIT support for cpu time counter kfuncsarch/arm64/include/asm/insn.h arch/arm64/net/bpf_jit.h arch/arm64/net/bpf_jit_comp.c +1 more+109 -14 replies

Reviews trailers per patch

3 of 6 patches have a Reviewed-by or Acked-by.

PatchEduard ZingermanAndrii NakryikoYonghong Song
1/6
2/6A-b A-b A-b
3/6R-b A-b
4/6A-b
5/6
6/6

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 3 reviewers.

Read the v13 thread

Cover letter [PATCH bpf-next v13 0/6] bpf: add cpu time counter kfuncs

This patchset adds 2 kfuncs to provide a way to precisely measure the time spent running some code. The first patch provides a way to get cpu cycles counter which is used to feed CLOCK_MONOTONIC_RAW. On x86 architecture it is effectively rdtsc_ordered() function. On arm64, it is JITed to a read of CNTVCT_EL0 with ordering or CNTVCTSS_EL0, depending on cpu support.

The second patch adds a kfunc to convert cpu cycles to nanoseconds using shift/mult constants discovered by kernel. The main use-case for this kfunc is to convert deltas of timestamp counter values into nanoseconds. It is not supposed to get CLOCK_MONOTONIC_RAW values as offset part is skipped. JIT version is done for x86 and arm64 for now, on other architectures it falls back to get CLOCK_MONOTONIC_RAW values.

The reason to have these functions is to avoid overhead added by a bpf_ktime_get_ns() call in case of benchmarking, when two timestamps are taken to get delta value. With both functions being JITed, the overhead is minimal and the result has better precision. New functions can be used to benchmark BPF code directly in the program, or can be used in kprobe/uprobe to store timestamp counter in the session cookie and then in kretprobe/uretprobe the delta can be calculated and converted into nanoseconds.

These will be used in under-development BPF benchmarks, the first of which is available at [1].

Pre-requriement for this patch is to adjust CONFIG_BPF_JIT to depend on CONFIG_BPF_SYSCALL as it simplify things and it's actually a long overdue patch.

Selftests are also added to check whether the JIT implementation is correct and to show the simplest usage example.

[1] https://github.com/puranjaymohan/bpf/commits/bpf_bench/

Change log:
v12: https://lore.kernel.org/bpf/[email protected]/
Changes in v13:
- Added support for arm64
- Rebased on latest bpf-next/master
- Changed the word helper to kfunc in the commit message to not confuse with bpf helpers.

v11 -> v12:
* drop x86_32 JIT implementation because previous implementation of mul_u64_u32_shr() was not fully correct and the amount of work to properly implement it in asm is not worth the result.
* add comment explaining stablility of shift and mult values
* add comment explaining the reasoning behind CLOCK_MONOTONIC_RAW usage in fallback mode
* drop KF_FASTCALL tag from bpf_get_cpu_time_counter() as it cannot be implemented as fastcall in the verifier
* re-implement verifier checks for possible JIT of kfuncs
* adjust selftests to use JIT code for x86_64 and verifier inlining for arm64
v10 -> v11:
* add missing IS_ENABLED(CONFIG_BPF_SYSCALL)
* reword "cycles" -> "counter"
v9 -> v10:
* rework fallback implementation to avoid using vDSO data from kernel space.
* add comment about using "LFENCE; RDTSC" instead of "RDTSCP"
* guard x86 JIT implementation to be sure that TSC is enabled and stable
* v9 link:
https://lore.kernel.org/bpf/[email protected]/
v8 -> v9:
* rewording of commit messages, no code changes
* move change log from each patch into cover letter
v7 -> v8:
* rename kfuncs again to bpf_get_cpu_time_counter() and bpf_cpu_time_counter_to_ns()
* use cyc2ns_read_begin()/cyc2ns_read_end() to get mult and shift constants in bpf_cpu_time_counter_to_ns()
v6 -> v7:
* change boot_cpu_has() to cpu_feature_enabled() (Borislav)
* return constant clock_mode in __arch_get_hw_counter() call v5 -> v6:
* added cover letter
* add comment about dropping S64_MAX manipulation in jitted implementation of rdtsc_oredered (Alexey)
* add comment about using 'lfence;rdtsc' variant (Alexey)
* change the check in fixup_kfunc_call() (Eduard)
* make __arch_get_hw_counter() call more aligned with vDSO implementation (Yonghong)
v4 -> v5:
* use #if instead of #ifdef with IS_ENABLED
v3 -> v4:
* change name of the helper to bpf_get_cpu_cycles (Andrii)
* Hide the helper behind CONFIG_GENERIC_GETTIMEOFDAY to avoid exposing it on architectures which do not have vDSO functions and data
* reduce the scope of check of inlined functions in verifier to only 2, which are actually inlined.
* change helper name to bpf_cpu_cycles_to_ns.
* hide it behind CONFIG_GENERIC_GETTIMEOFDAY to avoid exposing on unsupported architectures.
v2 -> v3:
* change name of the helper to bpf_get_cpu_cycles_counter to explicitly mention what counter it provides (Andrii)
* move kfunc definition to bpf.h to use it in JIT.
* introduce another kfunc to convert cycles into nanoseconds as more meaningful time units for generic tracing use case (Andrii)
v1 -> v2:
* Fix incorrect function return value type to u64
* Introduce bpf_jit_inlines_kfunc_call() and use it in mark_fastcall_pattern_for_call() to avoid clobbering in case of running programs with no JIT (Eduard)
* Avoid rewriting instruction and check function pointer directly in JIT (Alexei)
* Change includes to fix compile issues on non x86 architectures

Puranjay Mohan (1):
bpf, arm64: Add JIT support for cpu time counter kfuncs

Vadim Fedorenko (5):
bpf: adjust BPF JIT dependency to BPF_SYSCALL
bpf: add bpf_get_cpu_time_counter kfunc
bpf: add bpf_cpu_time_counter_to_ns kfunc selftests/bpf: add selftest to check bpf_get_cpu_time_counter jit selftests/bpf: add usage example for cpu time counter kfuncs

 arch/arm64/include/asm/insn.h                 |   2 +
 arch/arm64/net/bpf_jit.h                      |   4 +
 arch/arm64/net/bpf_jit_comp.c                 |  54 ++++++
 arch/x86/net/bpf_jit_comp.c                   |  72 ++++++++
 arch/x86/net/bpf_jit_comp32.c                 |   2 +
 include/linux/bpf.h                           |   4 +
 include/linux/filter.h                        |   1 +
 kernel/bpf/Kconfig                            |   2 +-
 kernel/bpf/core.c                             |  11 ++
 kernel/bpf/helpers.c                          |  17 ++
 kernel/bpf/verifier.c                         |  14 +-
 .../bpf/prog_tests/test_cpu_cycles.c          |  35 ++++
 .../selftests/bpf/prog_tests/verifier.c       |   2 +
 .../selftests/bpf/progs/test_cpu_cycles.c     |  25 +++
 .../selftests/bpf/progs/verifier_cpu_cycles.c | 168 ++++++++++++++++++
 15 files changed, 408 insertions(+), 5 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/test_cpu_cycles.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_cpu_cycles.c
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_cpu_cycles.c

base-commit: eb0d6d97c27c29cd7392c8fd74f46edf7dff7ec2

Signature · 1 line
2.52.0