Skip to content

chore(deps): bump arbitrary from 1.4.2 to 1.5.0 - #1933

Merged
jsturtevant merged 1 commit into
mainfrom
dependabot/cargo/arbitrary-1.5.0
Oct 29, 2026
Merged

jsturtevant merged 1 commit into
mainfrom
dependabot/cargo/arbitrary-1.5.0

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Oct 14, 2026 •

Copy link
Copy Markdown
Contributor

Bumps arbitrary from 1.4.2 to 1.5.0.

Changelog

Sourced from arbitrary's changelog.

1.5.0

Released 2026-10-05.

Added

  • Added support for deriving Arbitrary for types with raw identifiers.

Changed

  • Updated derive_arbitrary to syn 3. This raises the MSRV from 1.63.0 to 1.71.0, which is the MSRV of syn 3.

Fixed

  • Fixed soundness when generating arrays.
  • Fixed nonzero integer generation to always produce a nonzero value.
  • Updated derive_arbitrary to include fixes for fully qualified paths.

Commits
  • 9807b59 Merge pull request #243 from fitzgen/bump-to-1.5.0
  • 7a49e61 Bump to version 1.5.0
  • aa2802b Merge pull request #242 from drakeo338/claude/241-fix
  • a9f7ca5 Update derive_arbitrary to syn 3 and raise MSRV to 1.71
  • cd35bd2 Merge pull request #239 from Manishearth/fix-soundness
  • 53a651d agent-driven: Fix soundness in try_create_array
  • df8afe9 Merge pull request #238 from baloo/baloo/push-lopsmyosqnpv
  • 15f5c08 Bring fix for fully qualified path
  • 341a8fd Merge pull request #237 from ngildenhuys/downgrading_msrv_deps
  • 9b68ef4 ci: adding update commands to downgrade dependencies to fix MSRV test
  • Additional commits viewable in compare view

@dependabot dependabot Bot added the kind/dependencies For PRs that update dependencies or related components label Oct 14, 2026
@dependabot
dependabot Bot force-pushed the dependabot/cargo/arbitrary-1.5.0 branch from b9dd8f0 to 3db3090 Compare October 14, 2026 16:14
@jsturtevant

Copy link
Copy Markdown
Contributor

@dependabot rebase

@dependabot
dependabot Bot force-pushed the dependabot/cargo/arbitrary-1.5.0 branch from 3db3090 to 3c5fb2c Compare October 21, 2026 20:46
@hyperlight-gh-bot

This comment has been minimized.

@jsturtevant

Copy link
Copy Markdown
Contributor

@dependabot rebase

1 similar comment
@jsturtevant

Copy link
Copy Markdown
Contributor

@dependabot rebase

Bumps [arbitrary](https://github.com/rust-fuzz/arbitrary) from 1.4.2 to 1.5.0.
- [Changelog](https://github.com/rust-fuzz/arbitrary/blob/main/CHANGELOG.md)
- [Commits](rust-fuzz/arbitrary@v1.4.2...v1.5.0)

---
updated-dependencies:
- dependency-name: arbitrary
  dependency-version: 1.5.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot
dependabot Bot force-pushed the dependabot/cargo/arbitrary-1.5.0 branch from 3c5fb2c to c71ab90 Compare October 29, 2026 17:00
@hyperlight-gh-bot

Copy link
Copy Markdown

Benchmark Results

Measured commit: c71ab90411d1
Baseline commit: 36c71f0515bb

kvm / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
382.14 µs (➖ 1.02x slower)
byte_chunks 788.83 ns (➖ 1.02x slower)
vec_bytes 580.31 ns (➖ 1.01x slower)

payload_allocation

slot_pool_segmented
65536 207.64 ns (➖ 1.47x slower)
262144 532.59 ns (➖ 1.02x slower)

sandboxes

create_initialized_and_drop
medium 79.07 ms (➖ 1.03x faster)

slot_pool

alloc_dealloc_4096 alloc_dealloc_128 alloc_dealloc_1500
7.69 ns (➖ 1.00x faster) 7.69 ns (➖ 1.00x slower) 7.70 ns (➖ 1.00x faster)

snapshot_files

load_snapshot_unverified
small 93.39 µs (➖ 1.01x slower)

virtq_readonly

slot_pool_segmented slot_pool_segmented_fragmented
65536 1.96 µs (➖ 1.07x faster) 2.13 µs (➖ 1.13x slower)
262144 7.40 µs (➖ 1.06x slower) 7.25 µs (➖ 1.02x slower)

virtq_readwrite

slot_pool_segmented slot_pool_segmented_fragmented
8192 1.06 µs (➖ 1.05x slower) 1.02 µs (➖ 1.03x slower)
65536 6.16 µs (➖ 1.07x faster) 6.16 µs (➖ 1.03x faster)
262144 28.04 µs (➖ 1.04x slower)
kvm / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
672.28 µs (➖ 1.08x faster)
byte_chunks 680.32 ns (➖ 1.05x faster)
vec_bytes 520.13 ns (➖ 1.05x faster)

payload_allocation

slot_pool_segmented
65536 181.18 ns (➖ 1.03x slower)
262144 560.48 ns (➖ 1.10x slower)

sandboxes

create_initialized_and_drop
medium 72.78 ms (➖ 1.05x faster)

slot_pool

alloc_dealloc_4096 alloc_dealloc_128 alloc_dealloc_1500
7.28 ns (➖ 1.07x slower) 7.22 ns (➖ 1.06x slower) 7.24 ns (➖ 1.06x slower)

snapshot_files

load_snapshot_unverified
small 46.47 µs (➖ 1.08x faster)

virtq_readonly

slot_pool_segmented slot_pool_segmented_fragmented
65536 2.18 µs (➖ 1.01x slower) 2.19 µs (➖ 1.02x slower)
262144 7.91 µs (➖ 1.04x slower) 7.95 µs (➖ 1.06x slower)

virtq_readwrite

slot_pool_segmented slot_pool_segmented_fragmented
8192 730.62 ns (➖ 1.01x slower) 731.93 ns (➖ 1.02x faster)
65536 7.22 µs (➖ 1.01x slower) 7.19 µs (➖ 1.01x faster)
262144 29.50 µs (➖ 1.07x faster)
mshv3 / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
376.01 µs (➖ 1.17x faster)
byte_chunks 963.30 ns (➖ 1.04x slower)
vec_bytes 711.00 ns (➖ 1.00x faster)

payload_allocation

slot_pool_segmented
65536 194.10 ns (➖ 1.00x slower)
262144 757.81 ns (➖ 1.01x faster)

sandboxes

create_initialized_and_drop
medium 61.78 ms (➖ 1.05x slower)

slot_pool

alloc_dealloc_4096 alloc_dealloc_128 alloc_dealloc_1500
9.58 ns (➖ 1.00x faster) 9.62 ns (➖ 1.14x faster) 9.93 ns (➖ 1.01x faster)

snapshot_files

load_snapshot_unverified
small 87.89 µs (➖ 1.00x faster)

virtq_readonly

slot_pool_segmented slot_pool_segmented_fragmented
65536 2.26 µs (➖ 1.01x faster) 2.40 µs (➖ 1.03x faster)
262144 9.71 µs (➖ 1.00x slower) 10.73 µs (➖ 1.07x slower)

virtq_readwrite

slot_pool_segmented slot_pool_segmented_fragmented
8192 1.24 µs (➖ 1.03x slower) 1.23 µs (➖ 1.03x faster)
65536 8.74 µs (➖ 1.20x faster) 8.49 µs (➖ 1.14x faster)
262144 35.97 µs (➖ 1.08x faster)
mshv3 / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
725.95 µs (➖ 1.10x faster)
byte_chunks 914.33 ns (➖ 1.01x faster)
vec_bytes 657.48 ns (➖ 1.00x slower)

payload_allocation

slot_pool_segmented
65536 270.79 ns (➖ 1.04x slower)
262144 674.39 ns (➖ 1.01x slower)

sandboxes

create_initialized_and_drop
medium 80.40 ms (➖ 1.10x faster)

slot_pool

alloc_dealloc_4096 alloc_dealloc_128 alloc_dealloc_1500
8.71 ns (➖ 1.09x faster) 8.80 ns (➖ 1.00x faster) 8.71 ns (➖ 1.00x faster)

snapshot_files

load_snapshot_unverified
small 49.25 µs (➖ 1.33x faster)

virtq_readonly

slot_pool_segmented slot_pool_segmented_fragmented
65536 2.31 µs (➖ 1.01x slower) 2.31 µs (➖ 1.00x slower)
262144 8.02 µs (➖ 1.00x slower) 8.06 µs (➖ 1.01x slower)

virtq_readwrite

slot_pool_segmented slot_pool_segmented_fragmented
8192 923.15 ns (➖ 1.00x slower) 901.23 ns (➖ 1.04x faster)
65536 7.46 µs (➖ 1.00x faster) 7.50 µs (➖ 1.01x faster)
262144 39.41 µs (➖ 1.05x slower)
hyperv-ws2025 / amd (Windows) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
1.92 ms (➖ 1.02x slower)
byte_chunks 1.18 µs (➖ 1.01x faster)
vec_bytes 861.48 ns (➖ 1.09x slower)

payload_allocation

slot_pool_segmented
65536 231.82 ns (➖ 1.00x faster)
262144 787.37 ns (➖ 1.01x slower)

sandboxes

create_initialized_and_drop
medium 84.40 ms (➖ 1.05x slower)

slot_pool

alloc_dealloc_4096 alloc_dealloc_128 alloc_dealloc_1500
10.04 ns (➖ 1.01x slower) 10.08 ns (➖ 1.01x slower) 10.11 ns (➖ 1.02x slower)

snapshot_files

load_snapshot_unverified
small 669.11 µs (➖ 1.04x slower)

virtq_readonly

slot_pool_segmented slot_pool_segmented_fragmented
65536 2.48 µs (➖ 1.08x slower) 2.54 µs (➖ 1.08x slower)
262144 9.95 µs (➖ 1.05x slower) 10.43 µs (➖ 1.00x slower)

virtq_readwrite

slot_pool_segmented slot_pool_segmented_fragmented
8192 1.41 µs (➖ 1.14x slower) 1.39 µs (➖ 1.13x slower)
65536 9.39 µs (➖ 1.07x slower) 10.29 µs (➖ 1.13x slower)
262144 41.82 µs (➖ 1.05x faster)
hyperv-ws2025 / intel (Windows) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

decode_vec_bytes_copy encode_control
3.25 ms (➖ 1.06x slower)
byte_chunks 1.19 µs (➖ 1.00x slower)
vec_bytes 766.66 ns (➖ 1.02x faster)

payload_allocation

slot_pool_segmented
65536 212.95 ns (➖ 1.01x slower)
262144 752.35 ns (➖ 1.01x faster)

sandboxes

create_initialized_and_drop
medium 105.21 ms (➖ 1.00x faster)

slot_pool

alloc_dealloc_4096 alloc_dealloc_128 alloc_dealloc_1500
10.20 ns (➖ 1.01x slower) 9.95 ns (➖ 1.02x slower) 9.78 ns (➖ 1.00x faster)

snapshot_files

load_snapshot_unverified
small 568.64 µs (➖ 1.53x faster)

virtq_readonly

slot_pool_segmented slot_pool_segmented_fragmented
65536 2.49 µs (➖ 1.00x faster) 2.28 µs (➖ 1.01x faster)
262144 7.71 µs (➖ 1.03x slower) 7.54 µs (➖ 1.00x slower)

virtq_readwrite

slot_pool_segmented slot_pool_segmented_fragmented
8192 1.09 µs (➖ 1.02x slower) 1.23 µs (➖ 1.16x slower)
65536 8.64 µs (➖ 1.01x slower) 8.77 µs (➖ 1.06x slower)
262144 51.61 µs (➖ 1.00x faster)

Reported by cargo ci bench-report --candidate run:40917156702 --config-file bench_report.toml.

@jsturtevant
jsturtevant enabled auto-merge (squash) October 29, 2026 20:31
@jsturtevant
jsturtevant merged commit 3233086 into main Oct 29, 2026
63 checks passed
@jsturtevant
jsturtevant deleted the dependabot/cargo/arbitrary-1.5.0 branch October 29, 2026 20:32
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.

1 participant