Skip to content

build: use Go 1.26.8 and preserve glibc 2.31 compatibility - #4312

Merged
piotr-roslaniec merged 4 commits into
devfrom
codex/go-toolchain-3991
Sep 30, 2026
Merged

piotr-roslaniec merged 4 commits into
devfrom
codex/go-toolchain-3991

Conversation

@mswilkison

@mswilkison mswilkison commented Sep 8, 2026 •

Copy link
Copy Markdown
Contributor

The Go 1.24 Docker builders reject dependency updates that require Go 1.25.7, including the IPLD update discussed in #3991. Raise the module minimum to 1.25.7 and use Go 1.26.8 for the suggested toolchain, both Docker builders, and CI. Go 1.26 is the older supported release line; Go 1.25 left support when Go 1.27 shipped.

The source builder and runtime move to Alpine 3.23. The release builder retains Bullseye's C compiler and glibc 2.31, copying only Go from the official 1.26.8 image. This preserves cgo for native Linux/amd64 releases and compatibility with Debian 11 and Ubuntu 20.04. Both the client and tag-based release workflows now extract the actual Linux release archive and run the binary in both baseline containers before archiving or publishing it.

All setup-go jobs use v6 and read go.mod, including the vendored btcec and manual benchmark jobs; v6 honors the toolchain directive. The benchmark job now checks out the repository to read that file. Staticcheck moves to 2026.1 for Go 1.26 support.

This targets dev and implements the standalone toolchain step of #3991. All module dependency versions and go.sum remain unchanged; the issue's separate x/crypto, telemetry, and IPLD/libp2p reviews remain follow-ups.

Validation:

  • The original PR's CI Linux release artifact fails at startup on both Debian 11 and Ubuntu 20.04 with missing GLIBC_2.32 and GLIBC_2.34 symbols, confirming that the added check detects the regression.
  • The updated Linux/amd64 builder reports Go 1.26.8, glibc 2.31, and CGO_ENABLED=1.
  • CI on the final commit passes the full Go unit suite, coverage gate, Docker runtime build, Linux/macOS release packaging, and startup of the rebuilt Linux archive on both Debian 11 and Ubuntu 20.04. Formatting, vet, Staticcheck, gosec, and vendored btcec verification/tests also pass.
  • Workflow YAML, the compatibility step's Bash syntax, Go 1.26.8 formatting, and git diff --check pass.
  • The separate client integration job also passes. All applicable CI checks are green on 97b4e1ce5a26a6db467f322e619de059d844b61d.

References: Go release policy and versions, Staticcheck 2026.1.


Prepared rebase onto dev (not pushed yet)

This branch conflicts with dev. A rebase onto dev at bc4a10e was prepared and verified locally: range-diff and net-diff were checked against the current head 97b4e1c, all 3 commits are kept with identical line changes, and nothing from this PR was lost; both workflow YAMLs parse, go build ./... and go mod tidy -diff passed (actionlint and a Docker build were not run). It has not been pushed; the branch is unchanged. When pushed, the original head will be kept as backup/codex/go-toolchain-3991-20260926.

Judgment calls for review:

  • Added commit ci: align Go setup in electrum jobs added on dev: electrum-vendor-byte-identity (setup-go@v5 pinned to 1.24.1) and electrum-lifecycle (setup-go@v5) landed on dev after this branch was cut. Both now use setup-go@v6 with go-version-file: go.mod, like every other job in this PR. They would likely still work without it (GOTOOLCHAIN=auto), so drop the commit if you prefer.
  • client-bench permissions: kept both sides, so the job has actions: read, contents: read (this PR, for its checkout step) and pull-requests: write (from dev's ci(benchmark): automate regression baselines #4340), followed by dev's env block.

Generated by Claude Code

@coderabbitai

coderabbitai Bot commented Sep 8, 2026 •

Copy link
Copy Markdown

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: 39ab54d6-812d-4001-a807-bd605b865f91

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

Keep Bullseye's C compiler and libc while copying the Go 1.26.8 toolchain from the official image. Run packaged Linux binaries on Debian 11 and Ubuntu 20.04 before publication. Align the benchmark job inherited from dev with the module-selected Go toolchain.
@mswilkison mswilkison changed the title build: align Go builders and CI on 1.26.8 build: use Go 1.26.8 and preserve glibc 2.31 compatibility Sep 8, 2026
@mswilkison
mswilkison changed the base branch from main to dev September 8, 2026 15:26
@mswilkison
mswilkison force-pushed the codex/go-toolchain-3991 branch from f583006 to 10b0812 Compare September 8, 2026 15:26
# Conflicts:
#	.github/workflows/client.yml
@github-actions

Copy link
Copy Markdown

client-bench results

Baseline: latest successful push on the target branch. Regressions >12% are advisory on pull requests (hard-fail on push to dev and on the nightly schedule).

benchstat vs baseline
goos: linux
goarch: amd64
pkg: github.com/keep-network/keep-core/pkg/altbn128
cpu: AMD EPYC 7763 64-Core Processor                
                                │  bench.txt   │
                                │    sec/op    │
CompressG1-4                      561.1n ± 54%
DecompressG1-4                    42.59µ ±  3%
CompressDecompressRoundTripG1-4   42.55µ ±  3%
CompressDecompressRoundTripG2-4   1.884m ±  1%
geomean                           37.20µ

                                │  bench.txt   │
                                │     B/op     │
CompressG1-4                        224.0 ± 0%
DecompressG1-4                    4.625Ki ± 3%
CompressDecompressRoundTripG1-4   4.746Ki ± 4%
CompressDecompressRoundTripG2-4   1.156Mi ± 2%
geomean                           8.683Ki

                                │  bench.txt  │
                                │  allocs/op  │
CompressG1-4                       5.000 ± 0%
DecompressG1-4                     147.0 ± 5%
CompressDecompressRoundTripG1-4    149.5 ± 8%
CompressDecompressRoundTripG2-4   16.12k ± 2%
geomean                            205.1

cpu: Intel(R) Xeon(R) 6973P-C
                                │ bench-prev/bench.txt │
                                │        sec/op        │
CompressG1-4                             208.5n ± 116%
DecompressG1-4                           25.51µ ±   3%
CompressDecompressRoundTripG1-4          25.70µ ±   4%
CompressDecompressRoundTripG2-4          1.269m ±   1%
geomean                                  20.41µ

                                │ bench-prev/bench.txt │
                                │         B/op         │
CompressG1-4                                224.0 ± 0%
DecompressG1-4                            4.626Ki ± 2%
CompressDecompressRoundTripG1-4           4.880Ki ± 2%
CompressDecompressRoundTripG2-4           1.145Mi ± 1%
geomean                                   8.724Ki

                                │ bench-prev/bench.txt │
                                │      allocs/op       │
CompressG1-4                                5.000 ± 0%
DecompressG1-4                              145.5 ± 7%
CompressDecompressRoundTripG1-4             153.5 ± 4%
CompressDecompressRoundTripG2-4            15.95k ± 1%
geomean                                     205.4

pkg: github.com/keep-network/keep-core/pkg/beacon/gjkr
cpu: AMD EPYC 7763 64-Core Processor                
                                            │  bench.txt  │
                                            │   sec/op    │
MarshalEphemeralPublicKeyMessage-4            1.314µ ± 1%
UnmarshalEphemeralPublicKeyMessage-4          51.96µ ± 0%
MarshalEphemeralPublicKeyMessage_64Keys-4     24.32µ ± 1%
UnmarshalEphemeralPublicKeyMessage_64Keys-4   1.620m ± 0%
geomean                                       40.49µ

                                            │  bench.txt   │
                                            │     B/op     │
MarshalEphemeralPublicKeyMessage-4              784.0 ± 0%
UnmarshalEphemeralPublicKeyMessage-4          1.255Ki ± 0%
MarshalEphemeralPublicKeyMessage_64Keys-4     15.91Ki ± 0%
UnmarshalEphemeralPublicKeyMessage_64Keys-4   32.00Ki ± 0%
geomean                                       4.703Ki

                                            │ bench.txt  │
                                            │ allocs/op  │
MarshalEphemeralPublicKeyMessage-4            16.00 ± 0%
UnmarshalEphemeralPublicKeyMessage-4          24.00 ± 0%
MarshalEphemeralPublicKeyMessage_64Keys-4     384.0 ± 0%
UnmarshalEphemeralPublicKeyMessage_64Keys-4   584.0 ± 0%
geomean                                       96.33

cpu: Intel(R) Xeon(R) 6973P-C
                                            │ bench-prev/bench.txt │
                                            │        sec/op        │
MarshalEphemeralPublicKeyMessage-4                     881.6n ± 2%
UnmarshalEphemeralPublicKeyMessage-4                   40.92µ ± 0%
MarshalEphemeralPublicKeyMessage_64Keys-4              16.88µ ± 3%
UnmarshalEphemeralPublicKeyMessage_64Keys-4            1.281m ± 3%
geomean                                                29.71µ

                                            │ bench-prev/bench.txt │
                                            │         B/op         │
MarshalEphemeralPublicKeyMessage-4                      784.0 ± 0%
UnmarshalEphemeralPublicKeyMessage-4                  1.255Ki ± 0%
MarshalEphemeralPublicKeyMessage_64Keys-4             15.91Ki ± 0%
UnmarshalEphemeralPublicKeyMessage_64Keys-4           32.00Ki ± 0%
geomean                                               4.703Ki

                                            │ bench-prev/bench.txt │
                                            │      allocs/op       │
MarshalEphemeralPublicKeyMessage-4                      16.00 ± 0%
UnmarshalEphemeralPublicKeyMessage-4                    24.00 ± 0%
MarshalEphemeralPublicKeyMessage_64Keys-4               384.0 ± 0%
UnmarshalEphemeralPublicKeyMessage_64Keys-4             584.0 ± 0%
geomean                                                 96.33

pkg: github.com/keep-network/keep-core/pkg/bitcoin
cpu: AMD EPYC 7763 64-Core Processor                
                                  │  bench.txt  │
                                  │   sec/op    │
ComputeSignatureHashes_1Input-4     1.413µ ± 1%
ComputeSignatureHashes_5Inputs-4    4.323µ ± 0%
ComputeSignatureHashes_20Inputs-4   15.11µ ± 1%
geomean                             4.519µ

                                  │  bench.txt   │
                                  │     B/op     │
ComputeSignatureHashes_1Input-4       752.0 ± 0%
ComputeSignatureHashes_5Inputs-4    2.367Ki ± 0%
ComputeSignatureHashes_20Inputs-4   8.672Ki ± 0%
geomean                             2.470Ki

                                  │ bench.txt  │
                                  │ allocs/op  │
ComputeSignatureHashes_1Input-4     18.00 ± 0%
ComputeSignatureHashes_5Inputs-4    72.00 ± 0%
ComputeSignatureHashes_20Inputs-4   270.0 ± 0%
geomean                             70.47

cpu: Intel(R) Xeon(R) 6973P-C
                                  │ bench-prev/bench.txt │
                                  │        sec/op        │
ComputeSignatureHashes_1Input-4              1.010µ ± 2%
ComputeSignatureHashes_5Inputs-4             2.987µ ± 3%
ComputeSignatureHashes_20Inputs-4            10.32µ ± 3%
geomean                                      3.145µ

                                  │ bench-prev/bench.txt │
                                  │         B/op         │
ComputeSignatureHashes_1Input-4               752.0 ± 0%
ComputeSignatureHashes_5Inputs-4            2.367Ki ± 0%
ComputeSignatureHashes_20Inputs-4           8.672Ki ± 0%
geomean                                     2.470Ki

                                  │ bench-prev/bench.txt │
                                  │      allocs/op       │
ComputeSignatureHashes_1Input-4               18.00 ± 0%
ComputeSignatureHashes_5Inputs-4              72.00 ± 0%
ComputeSignatureHashes_20Inputs-4             270.0 ± 0%
geomean                                       70.47

pkg: github.com/keep-network/keep-core/pkg/bls
cpu: AMD EPYC 7763 64-Core Processor                
                     │  bench.txt  │
                     │   sec/op    │
Sign-4                 196.6µ ± 0%
Verify-4               1.989m ± 1%
AggregateBLS/N=10-4    1.902m ± 1%
AggregateBLS/N=50-4    2.023m ± 1%
AggregateBLS/N=100-4   2.146m ± 0%
ThresholdVerify-4      3.299m ± 2%
geomean                1.483m

                     │  bench.txt   │
                     │     B/op     │
Sign-4                 14.16Ki ± 0%
Verify-4               95.27Ki ± 0%
AggregateBLS/N=10-4    82.36Ki ± 0%
AggregateBLS/N=50-4    82.36Ki ± 0%
AggregateBLS/N=100-4   82.36Ki ± 0%
ThresholdVerify-4      359.4Ki ± 0%
geomean                80.44Ki

                     │  bench.txt  │
                     │  allocs/op  │
Sign-4                  710.0 ± 0%
Verify-4               1.452k ± 0%
AggregateBLS/N=10-4     764.0 ± 0%
AggregateBLS/N=50-4     764.0 ± 0%
AggregateBLS/N=100-4    764.0 ± 0%
ThresholdVerify-4      9.118k ± 0%
geomean                1.270k

                  │  bench.txt  │
                  │ p50-sec/op  │
ThresholdVerify-4   3.256m ± 2%

                  │  bench.txt  │
                  │ p95-sec/op  │
ThresholdVerify-4   3.663m ± 3%

                  │  bench.txt  │
                  │ p99-sec/op  │
ThresholdVerify-4   3.830m ± 3%

cpu: Intel(R) Xeon(R) 6973P-C
                     │ bench-prev/bench.txt │
                     │        sec/op        │
Sign-4                          130.4µ ± 3%
Verify-4                        1.251m ± 4%
AggregateBLS/N=10-4             1.192m ± 1%
AggregateBLS/N=50-4             1.254m ± 0%
AggregateBLS/N=100-4            1.331m ± 0%
ThresholdVerify-4               3.624m ± 0%
geomean                         1.027m

                     │ bench-prev/bench.txt │
                     │         B/op         │
Sign-4                         14.21Ki ± 0%
Verify-4                       114.3Ki ± 0%
AggregateBLS/N=10-4            101.4Ki ± 0%
AggregateBLS/N=50-4            101.4Ki ± 0%
AggregateBLS/N=100-4           101.4Ki ± 0%
ThresholdVerify-4              897.7Ki ± 0%
geomean                        107.2Ki

                     │ bench-prev/bench.txt │
                     │      allocs/op       │
Sign-4                           712.0 ± 0%
Verify-4                        1.660k ± 0%
AggregateBLS/N=10-4              971.0 ± 0%
AggregateBLS/N=50-4              971.0 ± 0%
AggregateBLS/N=100-4             971.0 ± 0%
ThresholdVerify-4               21.12k ± 0%
geomean                         1.685k

pkg: github.com/keep-network/keep-core/pkg/net/libp2p
cpu: AMD EPYC 7763 64-Core Processor                
                               │  bench.txt  │
                               │   sec/op    │
ChannelDeliver_SingleHandler-4   17.34n ± 1%
ChannelDeliver_10Handlers-4      148.5n ± 0%
ProcessPubsubMessage-4           297.1n ± 0%
geomean                          91.44n

                               │  bench.txt   │
                               │     B/op     │
ChannelDeliver_SingleHandler-4   0.000 ± 0%
ChannelDeliver_10Handlers-4      80.00 ± 0%
ProcessPubsubMessage-4           192.0 ± 0%
geomean                                     ¹
¹ summaries must be >0 to compute geomean

                               │  bench.txt   │
                               │  allocs/op   │
ChannelDeliver_SingleHandler-4   0.000 ± 0%
ChannelDeliver_10Handlers-4      1.000 ± 0%
ProcessPubsubMessage-4           3.000 ± 0%
geomean                                     ¹
¹ summaries must be >0 to compute geomean

cpu: Intel(R) Xeon(R) 6973P-C
                               │ bench-prev/bench.txt │
                               │        sec/op        │
ChannelDeliver_SingleHandler-4            23.82n ± 2%
ChannelDeliver_10Handlers-4               90.00n ± 0%
ProcessPubsubMessage-4                    201.0n ± 1%
geomean                                   75.53n

                               │ bench-prev/bench.txt │
                               │         B/op         │
ChannelDeliver_SingleHandler-4             8.000 ± 0%
ChannelDeliver_10Handlers-4                80.00 ± 0%
ProcessPubsubMessage-4                     192.0 ± 0%
geomean                                    49.72

                               │ bench-prev/bench.txt │
                               │      allocs/op       │
ChannelDeliver_SingleHandler-4             1.000 ± 0%
ChannelDeliver_10Handlers-4                1.000 ± 0%
ProcessPubsubMessage-4                     3.000 ± 0%
geomean                                    1.442

pkg: github.com/keep-network/keep-core/pkg/net/retransmission
cpu: AMD EPYC 7763 64-Core Processor                
                       │  bench.txt   │
                       │    sec/op    │
BackoffStrategyTick-4     5.301n ± 0%
StandardStrategyTick-4   0.6245n ± 0%
geomean                   1.819n

                       │  bench.txt   │
                       │     B/op     │
BackoffStrategyTick-4    0.000 ± 0%
StandardStrategyTick-4   0.000 ± 0%
geomean                             ¹
¹ summaries must be >0 to compute geomean

                       │  bench.txt   │
                       │  allocs/op   │
BackoffStrategyTick-4    0.000 ± 0%
StandardStrategyTick-4   0.000 ± 0%
geomean                             ¹
¹ summaries must be >0 to compute geomean

cpu: Intel(R) Xeon(R) 6973P-C
                       │ bench-prev/bench.txt │
                       │        sec/op        │
BackoffStrategyTick-4             13.68n ± 4%
StandardStrategyTick-4           0.3599n ± 7%
geomean                           2.219n

                       │ bench-prev/bench.txt │
                       │         B/op         │
BackoffStrategyTick-4            0.000 ± 0%
StandardStrategyTick-4           0.000 ± 0%
geomean                                     ¹
¹ summaries must be >0 to compute geomean

                       │ bench-prev/bench.txt │
                       │      allocs/op       │
BackoffStrategyTick-4            0.000 ± 0%
StandardStrategyTick-4           0.000 ± 0%
geomean                                     ¹
¹ summaries must be >0 to compute geomean

pkg: github.com/keep-network/keep-core/pkg/tbtc
cpu: AMD EPYC 7763 64-Core Processor                
                                │  bench.txt  │
                                │   sec/op    │
GetRecentWindows_100Windows-4     69.88µ ± 2%
GetRecentWindows_1000Windows-4    808.6µ ± 1%
GetSummary_100Windows-4           62.27µ ± 2%
GetSummary_1000Windows-4          734.8µ ± 1%
CleanupOldWindows_1000Windows-4   119.7µ ± 1%
geomean                           198.7µ

                                │  bench.txt   │
                                │     B/op     │
GetRecentWindows_100Windows-4     90.87Ki ± 0%
GetRecentWindows_1000Windows-4    906.7Ki ± 0%
GetSummary_100Windows-4           89.94Ki ± 0%
GetSummary_1000Windows-4          898.6Ki ± 0%
CleanupOldWindows_1000Windows-4   8.056Ki ± 0%
geomean                           139.9Ki

                                │  bench.txt  │
                                │  allocs/op  │
GetRecentWindows_100Windows-4      804.0 ± 0%
GetRecentWindows_1000Windows-4    8.004k ± 0%
GetSummary_100Windows-4            801.0 ± 0%
GetSummary_1000Windows-4          8.001k ± 0%
CleanupOldWindows_1000Windows-4    3.000 ± 0%
geomean                            658.4

cpu: Intel(R) Xeon(R) 6973P-C
                                │ bench-prev/bench.txt │
                                │        sec/op        │
GetRecentWindows_100Windows-4              55.48µ ± 1%
GetRecentWindows_1000Windows-4             618.4µ ± 1%
GetSummary_100Windows-4                    52.83µ ± 4%
GetSummary_1000Windows-4                   616.3µ ± 3%
CleanupOldWindows_1000Windows-4            86.89µ ± 2%
geomean                                    157.5µ

                                │ bench-prev/bench.txt │
                                │         B/op         │
GetRecentWindows_100Windows-4             90.87Ki ± 0%
GetRecentWindows_1000Windows-4            906.7Ki ± 0%
GetSummary_100Windows-4                   89.94Ki ± 0%
GetSummary_1000Windows-4                  898.6Ki ± 0%
CleanupOldWindows_1000Windows-4           8.055Ki ± 0%
geomean                                   139.9Ki

                                │ bench-prev/bench.txt │
                                │      allocs/op       │
GetRecentWindows_100Windows-4               804.0 ± 0%
GetRecentWindows_1000Windows-4             8.004k ± 0%
GetSummary_100Windows-4                     801.0 ± 0%
GetSummary_1000Windows-4                   8.001k ± 0%
CleanupOldWindows_1000Windows-4             3.000 ± 0%
geomean                                     658.4

pkg: github.com/keep-network/keep-core/pkg/tecdsa/dkg
cpu: AMD EPYC 7763 64-Core Processor                
                                             │  bench.txt  │
                                             │   sec/op    │
MarshalEphemeralPublicKeyMessage-4             1.145µ ± 3%
UnmarshalEphemeralPublicKeyMessage-4           1.233µ ± 1%
MarshalEphemeralPublicKeyMessage_100Keys-4     29.00µ ± 1%
UnmarshalEphemeralPublicKeyMessage_100Keys-4   37.90µ ± 1%
RoundTripDKGMessage-4                          2.508µ ± 2%
geomean                                        5.223µ

                                             │  bench.txt   │
                                             │     B/op     │
MarshalEphemeralPublicKeyMessage-4               624.0 ± 0%
UnmarshalEphemeralPublicKeyMessage-4             917.0 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4     14.33Ki ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4   21.42Ki ± 0%
RoundTripDKGMessage-4                          1.422Ki ± 0%
geomean                                        2.988Ki

                                             │ bench.txt  │
                                             │ allocs/op  │
MarshalEphemeralPublicKeyMessage-4             12.00 ± 0%
UnmarshalEphemeralPublicKeyMessage-4           12.00 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4     402.0 ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4   314.0 ± 0%
RoundTripDKGMessage-4                          25.00 ± 0%
geomean                                        53.89

cpu: Intel(R) Xeon(R) 6973P-C
                                             │ bench-prev/bench.txt │
                                             │        sec/op        │
MarshalEphemeralPublicKeyMessage-4                      774.4n ± 4%
UnmarshalEphemeralPublicKeyMessage-4                    794.5n ± 3%
MarshalEphemeralPublicKeyMessage_100Keys-4              18.51µ ± 9%
UnmarshalEphemeralPublicKeyMessage_100Keys-4            24.33µ ± 1%
RoundTripDKGMessage-4                                   1.590µ ± 4%
geomean                                                 3.379µ

                                             │ bench-prev/bench.txt │
                                             │         B/op         │
MarshalEphemeralPublicKeyMessage-4                       624.0 ± 0%
UnmarshalEphemeralPublicKeyMessage-4                     917.0 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4             14.33Ki ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4           21.42Ki ± 0%
RoundTripDKGMessage-4                                  1.422Ki ± 0%
geomean                                                2.988Ki

                                             │ bench-prev/bench.txt │
                                             │      allocs/op       │
MarshalEphemeralPublicKeyMessage-4                       12.00 ± 0%
UnmarshalEphemeralPublicKeyMessage-4                     12.00 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4               402.0 ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4             314.0 ± 0%
RoundTripDKGMessage-4                                    25.00 ± 0%
geomean                                                  53.89

pkg: github.com/keep-network/keep-core/pkg/tecdsa/signing
cpu: AMD EPYC 7763 64-Core Processor                
                                             │  bench.txt  │
                                             │   sec/op    │
MarshalEphemeralPublicKeyMessage-4             1.133µ ± 6%
UnmarshalEphemeralPublicKeyMessage-4           1.232µ ± 1%
MarshalEphemeralPublicKeyMessage_100Keys-4     28.67µ ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4   37.06µ ± 1%
MarshalSigningShareMessage-4                   1.142µ ± 0%
UnmarshalSigningShareMessage-4                 1.227µ ± 0%
RoundTripEphemeralKey-4                        2.495µ ± 1%
geomean                                        3.394µ

                                             │  bench.txt   │
                                             │     B/op     │
MarshalEphemeralPublicKeyMessage-4               624.0 ± 0%
UnmarshalEphemeralPublicKeyMessage-4             917.0 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4     14.33Ki ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4   21.42Ki ± 0%
MarshalSigningShareMessage-4                     592.0 ± 0%
UnmarshalSigningShareMessage-4                   864.0 ± 0%
RoundTripEphemeralKey-4                        1.505Ki ± 0%
geomean                                        1.989Ki

                                             │ bench.txt  │
                                             │ allocs/op  │
MarshalEphemeralPublicKeyMessage-4             12.00 ± 0%
UnmarshalEphemeralPublicKeyMessage-4           12.00 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4     402.0 ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4   314.0 ± 0%
MarshalSigningShareMessage-4                   12.00 ± 0%
UnmarshalSigningShareMessage-4                 13.00 ± 0%
RoundTripEphemeralKey-4                        24.00 ± 0%
geomean                                        35.28

cpu: Intel(R) Xeon(R) 6973P-C
                                             │ bench-prev/bench.txt │
                                             │        sec/op        │
MarshalEphemeralPublicKeyMessage-4                      725.0n ± 2%
UnmarshalEphemeralPublicKeyMessage-4                    777.5n ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4              17.73µ ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4            23.35µ ± 0%
MarshalSigningShareMessage-4                            719.7n ± 0%
UnmarshalSigningShareMessage-4                          767.3n ± 0%
RoundTripEphemeralKey-4                                 1.761µ ± 3%
geomean                                                 2.170µ

                                             │ bench-prev/bench.txt │
                                             │         B/op         │
MarshalEphemeralPublicKeyMessage-4                       624.0 ± 0%
UnmarshalEphemeralPublicKeyMessage-4                     917.0 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4             14.33Ki ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4           21.42Ki ± 0%
MarshalSigningShareMessage-4                             592.0 ± 0%
UnmarshalSigningShareMessage-4                           864.0 ± 0%
RoundTripEphemeralKey-4                                1.505Ki ± 0%
geomean                                                1.989Ki

                                             │ bench-prev/bench.txt │
                                             │      allocs/op       │
MarshalEphemeralPublicKeyMessage-4                       12.00 ± 0%
UnmarshalEphemeralPublicKeyMessage-4                     12.00 ± 0%
MarshalEphemeralPublicKeyMessage_100Keys-4               402.0 ± 0%
UnmarshalEphemeralPublicKeyMessage_100Keys-4             314.0 ± 0%
MarshalSigningShareMessage-4                             12.00 ± 0%
UnmarshalSigningShareMessage-4                           13.00 ± 0%
RoundTripEphemeralKey-4                                  24.00 ± 0%
geomean                                                  35.28

@piotr-roslaniec
piotr-roslaniec merged commit d72bca6 into dev Sep 30, 2026
45 checks passed
@piotr-roslaniec
piotr-roslaniec deleted the codex/go-toolchain-3991 branch September 30, 2026 08:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants