CFP-46760: eBPF Runtime Benchmarking - #101
Open
amchelapurath2023 wants to merge 1 commit into
Open
Conversation
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
4 times, most recently
from
June 26, 2026 18:34
be7ec98 to
f248585
Compare
Author
|
Requesting review @jrife |
jrife
requested changes
Jun 26, 2026
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
June 29, 2026 20:28
f248585 to
390408a
Compare
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
July 7, 2026 20:37
390408a to
a33f5c7
Compare
jrife
reviewed
Jul 7, 2026
jrife
approved these changes
Jul 7, 2026
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
July 7, 2026 22:10
a33f5c7 to
9b9aab0
Compare
amchelapurath2023
marked this pull request as ready for review
July 8, 2026 14:46
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
July 8, 2026 14:53
9b9aab0 to
17bfbd9
Compare
Member
|
cc: @cilium/sig-datapath |
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
July 16, 2026 00:11
17bfbd9 to
552b2a8
Compare
dylandreimerink
approved these changes
Jul 16, 2026
pchaigno
approved these changes
Jul 30, 2026
pchaigno
left a comment
pchaigno
left a comment
Member
There was a problem hiding this comment.
This looks good to me. Thanks for the write up!
I'd be interested to know how much overhead enabling this adds. If it is insignificant (which I guess it isn't), we may consider enabling it by default. Knowing the overhead will also be key to document it properly for users.
pchaigno
reviewed
Jul 30, 2026
| **Begin Design Discussion:** 2026-06-26 | ||
| **Cilium Release:** 1.21 | ||
| **Authors:** Anshul Chelapurath <achelapurath@google.com> | ||
| **Status:** Draft |
Member
There was a problem hiding this comment.
@xmulligan This CFP looks ready to go. Is all we need to change this line to Implementable and merge?
Member
There was a problem hiding this comment.
Yes I would just update that line and since it has approval from at least one committer it is ready to merge 🙂
Author
There was a problem hiding this comment.
Made some nit changes based on feedback from cilium/cilium#47186. Thanks for the reviews everyone!
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
July 31, 2026 16:59
552b2a8 to
cc5f2a5
Compare
Signed-off-by: Anshul Chelapurath <achelapurath@google.com>
amchelapurath2023
force-pushed
the
Cilium-BPF-Benchmarking
branch
from
July 31, 2026 17:02
cc5f2a5 to
6bee2e8
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Introduce BPF program execution statistics tracking and telemetry based
on the Linux BPF_ENABLE_STATS syscall. This adds on-demand benchmarking
capabilities via a new 'cilium bpf benchmark' CLI subcommand and routes
cumulative execution counts and runtimes to Prometheus metrics.
Motivation
Cilium's BPF datapath currently lacks runtime visibility, making it difficult to analyze the performance impact of configuration changes or detect latency regressions in large-scale environments without SSHing into individual nodes to run
bpftoolor similar queries manually.[continued in CFP]
Signed-off-by: Anshul Chelapurath achelapurath@google.com