Skip to content

Bump ci-setup-workflow to v2.3.1 - #1984

Merged
ludfjig merged 1 commit into
hyperlight-dev:mainfrom
ludfjig:ci-setup-ubuntu-llvm
Oct 26, 2026
Merged

ludfjig merged 1 commit into
hyperlight-dev:mainfrom
ludfjig:ci-setup-ubuntu-llvm

Conversation

@ludfjig

@ludfjig ludfjig commented Oct 26, 2026 •

Copy link
Copy Markdown
Contributor

Bumps ci-setup-workflow to v2.3.1 in every workflow. ubuntu-latest is Ubuntu 26.04, where apt.llvm.org has no LLVM 18 and v2.3.0 fails. This breaks check-golden-label and post-benchmark-comment. v2.3.1 (hyperlight-dev/ci-setup-workflow#25) installs LLVM 18 from the Ubuntu archive.

@ludfjig ludfjig added the kind/dependencies For PRs that update dependencies or related components label Oct 26, 2026
@ludfjig
ludfjig force-pushed the ci-setup-ubuntu-llvm branch from 91ba730 to 12ec055 Compare October 26, 2026 20:27
@hyperlight-gh-bot

Copy link
Copy Markdown

Benchmark Results

Measured commit: 12ec0553c161
Baseline commit: 559b19185766

kvm / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
372.53 µs (➖ 1.01x faster)
vec_bytes 575.23 ns (➖ 1.01x faster)
byte_chunks 770.69 ns (➖ 1.01x faster)

payload_allocation

slot_pool_segmented
65536 141.92 ns (➖ 1.00x faster)
262144 520.56 ns (➖ 1.01x faster)

sandboxes

create_initialized_and_drop
medium 84.04 ms (➖ 1.02x slower)

slot_pool

alloc_dealloc_128 alloc_dealloc_4096 alloc_dealloc_1500
7.69 ns (➖ 1.00x slower) 7.69 ns (➖ 1.00x faster) 7.68 ns (➖ 1.01x faster)

snapshot_files

load_snapshot_unverified
small 91.78 µs (➖ 1.02x faster)

virtq_readonly

slot_pool_segmented_fragmented slot_pool_segmented
65536 2.02 µs (➖ 1.04x slower) 1.90 µs (➖ 1.10x faster)
262144 8.24 µs (➖ 1.15x slower) 8.13 µs (➖ 1.12x slower)

virtq_readwrite

slot_pool_segmented_fragmented slot_pool_segmented
8192 1.03 µs (➖ 1.02x slower) 1.03 µs (➖ 1.07x faster)
65536 6.14 µs (➖ 1.04x faster) 6.99 µs (➖ 1.11x slower)
262144 26.90 µs (➖ 1.00x faster)
kvm / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
688.48 µs (➖ 1.04x faster)
vec_bytes 557.57 ns (➖ 1.05x faster)
byte_chunks 729.03 ns (➖ 1.06x faster)

payload_allocation

slot_pool_segmented
65536 176.96 ns (➖ 1.02x slower)
262144 564.49 ns (➖ 1.01x slower)

sandboxes

create_initialized_and_drop
medium 79.46 ms (➖ 1.01x slower)

slot_pool

alloc_dealloc_128 alloc_dealloc_4096 alloc_dealloc_1500
7.92 ns (➖ 1.00x slower) 7.80 ns (➖ 1.01x faster) 7.80 ns (➖ 1.01x faster)

snapshot_files

load_snapshot_unverified
small 48.05 µs (➖ 1.03x faster)

virtq_readonly

slot_pool_segmented_fragmented slot_pool_segmented
65536 2.40 µs (➖ 1.04x faster) 2.37 µs (➖ 1.07x faster)
262144 8.51 µs (➖ 1.02x faster) 8.47 µs (➖ 1.03x faster)

virtq_readwrite

slot_pool_segmented_fragmented slot_pool_segmented
8192 806.02 ns (➖ 1.12x faster) 788.62 ns (➖ 1.04x faster)
65536 7.85 µs (➖ 1.03x faster) 7.87 µs (➖ 1.02x faster)
262144 33.98 µs (➖ 1.02x slower)
mshv3 / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
267.41 µs (➖ 1.08x faster)
vec_bytes 702.20 ns (➖ 1.02x faster)
byte_chunks 953.99 ns (➖ 1.03x slower)

payload_allocation

slot_pool_segmented
65536 193.23 ns (➖ 1.02x faster)
262144 730.78 ns (➖ 1.00x faster)

sandboxes

create_initialized_and_drop
medium 62.88 ms (➖ 1.10x slower)

slot_pool

alloc_dealloc_128 alloc_dealloc_4096 alloc_dealloc_1500
9.59 ns (➖ 1.00x faster) 10.22 ns (➖ 1.13x faster) 9.65 ns (➖ 1.01x faster)

snapshot_files

load_snapshot_unverified
small 85.81 µs (➖ 1.01x faster)

virtq_readonly

slot_pool_segmented_fragmented slot_pool_segmented
65536 2.41 µs (➖ 1.02x faster) 2.25 µs (➖ 1.01x slower)
262144 10.71 µs (➖ 1.06x slower) 9.67 µs (➖ 1.03x faster)

virtq_readwrite

slot_pool_segmented_fragmented slot_pool_segmented
8192 1.29 µs (➖ 1.04x slower) 1.22 µs (➖ 1.02x slower)
65536 8.22 µs (➖ 1.21x faster) 9.72 µs (➖ 1.01x slower)
262144 36.26 µs (➖ 1.08x faster)
mshv3 / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
802.85 µs (➖ 1.01x faster)
vec_bytes 693.04 ns (➖ 1.05x slower)
byte_chunks 914.86 ns (➖ 1.02x faster)

payload_allocation

slot_pool_segmented
65536 261.54 ns (➖ 1.01x faster)
262144 717.15 ns (➖ 1.08x slower)

sandboxes

create_initialized_and_drop
medium 69.64 ms (➖ 1.05x faster)

slot_pool

alloc_dealloc_128 alloc_dealloc_4096 alloc_dealloc_1500
8.68 ns (➖ 1.00x faster) 8.79 ns (➖ 1.01x slower) 8.70 ns (➖ 1.30x faster)

snapshot_files

load_snapshot_unverified
small 44.59 µs (➖ 1.01x slower)

virtq_readonly

slot_pool_segmented_fragmented slot_pool_segmented
65536 2.28 µs (➖ 1.01x slower) 2.27 µs (➖ 1.01x faster)
262144 7.95 µs (➖ 1.01x faster) 7.94 µs (➖ 1.00x slower)

virtq_readwrite

slot_pool_segmented_fragmented slot_pool_segmented
8192 908.31 ns (➖ 1.01x faster) 901.37 ns (➖ 1.03x faster)
65536 7.51 µs (➖ 1.01x faster) 7.45 µs (➖ 1.01x faster)
262144 38.73 µs (➖ 1.03x faster)
hyperv-ws2025 / amd (Windows) (❌ 1.68x)

Top regressions

  • payload_allocation/slot_pool_segmented/262144 — ❌ 1.68x slower

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
2.12 ms (➖ 1.03x faster)
vec_bytes 788.85 ns (➖ 1.04x faster)
byte_chunks 1.14 µs (➖ 1.00x slower)

payload_allocation

slot_pool_segmented
65536 258.55 ns (➖ 1.12x slower)
262144 1.33 µs (❌ 1.68x slower)

sandboxes

create_initialized_and_drop
medium 100.27 ms (➖ 1.06x faster)

slot_pool

alloc_dealloc_128 alloc_dealloc_4096 alloc_dealloc_1500
10.79 ns (➖ 1.01x slower) 10.67 ns (➖ 1.01x slower) 11.19 ns (➖ 1.06x slower)

snapshot_files

load_snapshot_unverified
small 684.84 µs (➖ 1.05x slower)

virtq_readonly

slot_pool_segmented_fragmented slot_pool_segmented
65536 2.52 µs (➖ 1.10x faster) 2.49 µs (➖ 1.11x faster)
262144 8.56 µs (➖ 1.01x faster) 8.55 µs (➖ 1.11x faster)

virtq_readwrite

slot_pool_segmented_fragmented slot_pool_segmented
8192 1.38 µs (➖ 1.01x faster) 1.37 µs (➖ 1.03x faster)
65536 8.77 µs (➖ 1.02x faster) 8.54 µs (➖ 1.12x faster)
262144 37.85 µs (➖ 1.23x faster)
hyperv-ws2025 / intel (Windows) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
3.09 ms (➖ 1.05x slower)
vec_bytes 770.65 ns (➖ 1.01x faster)
byte_chunks 1.23 µs (➖ 1.02x slower)

payload_allocation

slot_pool_segmented
65536 267.42 ns (➖ 1.28x slower)
262144 920.45 ns (➖ 1.22x slower)

sandboxes

create_initialized_and_drop
medium 111.54 ms (➖ 1.00x slower)

slot_pool

alloc_dealloc_128 alloc_dealloc_4096 alloc_dealloc_1500
12.02 ns (➖ 1.24x slower) 12.70 ns (➖ 1.27x slower) 11.47 ns (➖ 1.19x slower)

snapshot_files

load_snapshot_unverified
small 527.61 µs (➖ 1.27x faster)

virtq_readonly

slot_pool_segmented_fragmented slot_pool_segmented
65536 2.27 µs (➖ 1.00x slower) 2.45 µs (➖ 1.07x slower)
262144 7.66 µs (➖ 1.02x slower) 7.50 µs (➖ 1.02x faster)

virtq_readwrite

slot_pool_segmented_fragmented slot_pool_segmented
8192 1.17 µs (➖ 1.12x slower) 1.07 µs (➖ 1.03x slower)
65536 8.25 µs (➖ 1.00x faster) 8.38 µs (➖ 1.01x faster)
262144 40.69 µs (➖ 1.19x faster)

Reported by cargo ci bench-report --candidate run:40395854628 --config-file bench_report.toml.

ubuntu-latest is Ubuntu 26.04, where apt.llvm.org has no LLVM 18 and ci-setup-workflow v2.3.0 fails. v2.3.1 installs LLVM 18 from the Ubuntu archive.

Signed-off-by: Ludvig Liljenberg <[email protected]>
@ludfjig
ludfjig force-pushed the ci-setup-ubuntu-llvm branch from 12ec055 to 11a83b8 Compare October 26, 2026 21:29
@ludfjig ludfjig changed the title Pin ci-setup-workflow to the Ubuntu archive LLVM fix Bump ci-setup-workflow to v2.3.1 Oct 26, 2026
@ludfjig
ludfjig marked this pull request as ready for review October 26, 2026 21:29
Copilot AI balanced review requested due to automatic review settings October 26, 2026 21:29

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

The consistent release pins incorporate a targeted upstream fix without interface changes or identified blocking issues.

No open issues 🎉

What changed in this PR

Updates Hyperlight’s CI setup action to v2.3.1 to address LLVM 18 installation failures on Ubuntu 26.04, including PR validation and benchmark comment jobs.

Changes:

  • Pins all setup-action references across 15 workflows to the v2.3.1 commit.
  • Incorporates the Ubuntu archive installation fix from ci-setup-workflow#25, preserving existing inputs and environment settings.
File Description
.github/​workflows/​ValidatePullRequest.yml Updates setup for validation and benchmark comments.
.github/​workflows/​RustNightly.yml Updates nightly toolchain setup.
.github/​workflows/​RegenSnapshotGoldens.yml Updates snapshot regeneration setup.
.github/​workflows/​PrimeCaches.yml Updates cache priming setup.
.github/​workflows/​dep_update_guest_locks.yml Updates guest lockfile maintenance setup.
.github/​workflows/​dep_run_examples.yml Updates example execution setup.
.github/​workflows/​dep_fuzzing.yml Updates fuzzing setup.
.github/​workflows/​dep_code_checks.yml Updates both code-check setup references.
.github/​workflows/​dep_build_test.yml Updates build and test setup.
.github/​workflows/​dep_build_guests.yml Updates guest build setup.
.github/​workflows/​dep_benchmarks.yml Updates benchmark setup.
.github/​workflows/​CreateRelease.yml Updates all three release setup references.
.github/​workflows/​Coverage.yml Updates coverage setup.
.github/​workflows/​copilot-setup-steps.yml Updates Copilot environment setup.
.github/​workflows/​CargoPublish.yml Updates crate publishing setup.

Review effort: Balanced

@ludfjig
ludfjig enabled auto-merge (squash) October 26, 2026 21:37
@ludfjig
ludfjig merged commit f3ade9e into hyperlight-dev:main Oct 26, 2026
63 of 64 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

kind/dependencies For PRs that update dependencies or related components

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants