Remove the print_debug and build-metadata cargo features by ludfjig · Pull Request #1980 · hyperlight-dev/hyperlight

ludfjig · 2026-10-23T18:00:15Z

This PR removes two host cargo features and tidies up feature lists.

Remove the print_debug feature and the exported debug! macro. The macro called tracing::debug! and also printed to stdout in debug builds. Its cfg was evaluated in the caller's crate, so the print never worked for external users.
Remove the build-metadata feature and its built and git2 build dependencies. The feature logged build details on the first sandbox creation. Its git fields were empty for crates.io users.
Clean up feature lists:
- The std feature of hyperlight-common no longer pulls in tracing.
- Guest crates only list features of their direct dependencies. The resolved features are unchanged.
- Remove the empty hyperlight-libc features table and stale feature names in hack/clippy-package-features.sh.

Dropping git2, which builds libgit2 from C, speeds up a clean debug build of hyperlight-host with default features on 8 cores:

	Before	After	Change
Wall time	65.6 s	42.5 s	−35%
CPU time	319 s	209 s	−34%
Crates compiled	131	120	−11
Target directory	847 MB	788 MB	−7%

Breaking: Crates that enable print_debug or build-metadata must remove them. Replace hyperlight_host::debug! with tracing::debug!.

Closes #1979

print_debug made the exported debug! macro also print to stdout in debug builds. tracing::debug! already emits the same message. The macro's cfg is evaluated in the caller's crate, so it never printed for external users. Signed-off-by: Ludvig Liljenberg <[email protected]>

build-metadata used built with git2 to log build details on the first sandbox creation. Its git fields were empty for crates.io users. Dropping git2, which builds libgit2 from C, speeds up a clean debug build of hyperlight-host with default features on 8 cores. Wall time drops 35%, from 65.6 s to 42.5 s. CPU time drops 34%, from 319 s to 209 s. The build compiles 11 fewer crates, and the target directory shrinks 7%, from 847 MB to 788 MB. Signed-off-by: Ludvig Liljenberg <[email protected]>

* Enabling std on hyperlight-common no longer pulls in tracing. * Feature lists name only the crates each package uses directly. * Drop the empty hyperlight-libc features table and stale feature names from the clippy script. Signed-off-by: Ludvig Liljenberg <[email protected]>

Copilot

🟢 Approval recommended

Static review found no blocking issues, and the removals are consistent across callers, feature forwarding, build recipes, and documentation.

No open issues 🎉

What changed in this PR

This PR reduces Hyperlight host build overhead by removing two Cargo features and their logging machinery. It also simplifies feature forwarding and documents the breaking changes.

Changes:

Remove print_debug, hyperlight_host::debug!, and build-metadata logging with its build dependencies.
Simplify guest feature lists and keep common-crate tracing optional.
Update build recipes, coverage documentation, and the changelog.

File	Description
`src/tests/rust_guests/witguest/Cargo.toml`	Simplifies guest feature forwarding.
`src/tests/rust_guests/simpleguest/Cargo.toml`	Simplifies guest feature forwarding.
`src/tests/rust_guests/dummyguest/Cargo.toml`	Simplifies memory-profile forwarding.
`src/hyperlight_libc/Cargo.toml`	Removes the empty features table.
`src/hyperlight_host/src/sandbox/uninitialized.rs`	Removes build logging, uses tracing directly, and ungates tracing tests.
`src/hyperlight_host/src/sandbox/trace/context.rs`	Uses `tracing::debug!` directly.
`src/hyperlight_host/src/lib.rs`	Removes the debug macro and build-metadata machinery.
`src/hyperlight_host/Cargo.toml`	Removes both features and the `built` dependency.
`src/hyperlight_host/build.rs`	Removes metadata generation and the debug-print configuration alias.
`src/hyperlight_guest_bin/Cargo.toml`	Simplifies tracing feature forwarding.
`src/hyperlight_common/Cargo.toml`	Keeps tracing optional when enabling `std`.
`Justfile`	Removes obsolete features from checks and test commands.
`hack/clippy-package-features.sh`	Removes stale feature exclusions.
`docs/how-to-run-coverage.md`	Aligns documented feature combinations with coverage commands.
`CHANGELOG.md`	Documents feature removals and migration guidance.
`Cargo.lock`	Removes unused build-metadata dependencies.

_{Review effort: Balanced}

ludfjig · 2026-10-23T18:12:25Z

@simongdavies I know you have said in the past that build-metadata can be useful, but I don't agree. Applications can check their cargo.lock to see what version they use, and if they really want to include it in their tracing at runtime, they can do that too for any dependency they care about. I don't think hyperlight needs special feature for this

hyperlight-gh-bot · 2026-10-23T18:45:58Z

Benchmark Results

Measured commit: 7a74694c5e9d
Baseline commit: 8a7e6b8794fe

kvm / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`744.61 ns` (➖ 1.03x faster)
`vec_bytes`	`592.03 ns` (➖ 1.02x slower)
		`372.30 µs` (➖ 1.01x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`524.75 ns` (➖ 1.23x faster)
`65536`	`145.21 ns` (➖ 1.01x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`81.36 ms` (➖ 1.01x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`7.76 ns` (➖ 1.01x slower)	`7.69 ns` (➖ 1.00x faster)	`7.72 ns` (➖ 1.00x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`93.13 µs` (➖ 1.00x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.16 µs` (➖ 1.00x faster)	`7.06 µs` (➖ 1.02x faster)
`65536`	`2.05 µs` (➖ 1.07x slower)	`1.92 µs` (➖ 1.10x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`6.31 µs` (➖ 1.06x faster)	`6.26 µs` (➖ 1.03x faster)
`8192`	`1.00 µs` (➖ 1.01x faster)	`1.03 µs` (➖ 1.04x faster)
`262144`		`28.32 µs` (➖ 1.05x slower)

kvm / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`748.29 ns` (➖ 1.06x faster)
`vec_bytes`	`572.53 ns` (➖ 1.02x faster)
		`730.89 µs` (➖ 1.04x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`594.24 ns` (➖ 1.01x faster)
`65536`	`205.56 ns` (➖ 1.01x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`86.12 ms` (➖ 1.01x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`8.06 ns` (➖ 1.01x faster)	`7.94 ns` (➖ 1.03x faster)	`8.05 ns` (➖ 1.02x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`48.49 µs` (➖ 1.10x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`8.90 µs` (➖ 1.02x faster)	`8.90 µs` (➖ 1.01x faster)
`65536`	`2.55 µs` (➖ 1.02x faster)	`2.54 µs` (➖ 1.02x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.31 µs` (➖ 1.03x faster)	`8.34 µs` (➖ 1.02x faster)
`8192`	`849.27 ns` (➖ 1.03x faster)	`840.54 ns` (➖ 1.01x faster)
`262144`		`33.95 µs` (➖ 1.03x faster)

mshv3 / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`915.15 ns` (➖ 1.00x slower)
`vec_bytes`	`704.53 ns` (➖ 1.01x slower)
		`322.50 µs` (➖ 1.01x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`673.84 ns` (➖ 1.03x faster)
`65536`	`185.30 ns` (➖ 1.02x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`60.68 ms` (➖ 1.06x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`9.89 ns` (➖ 1.00x faster)	`9.88 ns` (➖ 1.17x faster)	`9.91 ns` (➖ 1.00x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`86.69 µs` (➖ 1.01x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`9.25 µs` (➖ 1.01x slower)	`9.57 µs` (➖ 1.04x slower)
`65536`	`2.61 µs` (➖ 1.03x faster)	`2.47 µs` (➖ 1.00x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.12 µs` (➖ 1.01x faster)	`8.10 µs` (➖ 1.05x faster)
`8192`	`1.27 µs` (➖ 1.02x faster)	`1.25 µs` (➖ 1.03x faster)
`262144`		`34.73 µs` (➖ 1.06x faster)

mshv3 / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`935.14 ns` (➖ 1.01x slower)
`vec_bytes`	`670.71 ns` (➖ 1.00x slower)
		`1.02 ms` (➖ 1.13x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`725.73 ns` (➖ 1.15x slower)
`65536`	`192.12 ns` (➖ 1.30x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`76.48 ms` (➖ 1.05x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`9.06 ns` (➖ 1.00x faster)	`9.10 ns` (➖ 1.06x faster)	`8.97 ns` (➖ 1.02x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`49.55 µs` (➖ 1.12x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.99 µs` (➖ 1.01x faster)	`7.96 µs` (➖ 1.01x faster)
`65536`	`2.30 µs` (➖ 1.00x slower)	`2.30 µs` (➖ 1.01x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`7.48 µs` (➖ 1.01x faster)	`7.45 µs` (➖ 1.01x faster)
`8192`	`905.68 ns` (➖ 1.02x faster)	`918.98 ns` (➖ 1.00x slower)
`262144`		`40.08 µs` (➖ 1.03x slower)

hyperv-ws2025 / amd (Windows) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`1.13 µs` (➖ 1.01x faster)
`vec_bytes`	`824.66 ns` (➖ 1.09x slower)
		`1.95 ms` (➖ 1.02x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`815.30 ns` (➖ 1.04x faster)
`65536`	`233.65 ns` (➖ 1.06x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`82.90 ms` (➖ 1.03x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`10.78 ns` (➖ 1.01x faster)	`10.64 ns` (➖ 1.14x faster)	`10.61 ns` (➖ 1.25x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`700.88 µs` (➖ 1.20x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`9.42 µs` (➖ 1.09x slower)	`8.58 µs` (➖ 1.00x slower)
`65536`	`2.76 µs` (➖ 1.09x slower)	`2.48 µs` (➖ 1.02x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.79 µs` (➖ 1.02x slower)	`9.17 µs` (➖ 1.03x slower)
`8192`	`1.53 µs` (➖ 1.02x faster)	`1.38 µs` (➖ 1.00x slower)
`262144`		`38.03 µs` (➖ 1.01x faster)

hyperv-ws2025 / intel (Windows) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`1.18 µs` (➖ 1.00x slower)
`vec_bytes`	`758.23 ns` (➖ 1.01x faster)
		`3.25 ms` (➖ 1.01x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`740.00 ns` (➖ 1.01x faster)
`65536`	`212.69 ns` (➖ 1.01x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`119.65 ms` (➖ 1.04x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`10.67 ns` (➖ 1.11x slower)	`10.32 ns` (➖ 1.03x slower)	`9.70 ns` (➖ 1.00x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`574.66 µs` (➖ 1.18x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.53 µs` (➖ 1.01x faster)	`7.48 µs` (➖ 1.01x slower)
`65536`	`2.32 µs` (➖ 1.04x slower)	`2.49 µs` (➖ 1.03x slower)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.29 µs` (➖ 1.01x slower)	`8.40 µs` (➖ 1.02x faster)
`8192`	`1.39 µs` (➖ 1.03x faster)	`1.09 µs` (➖ 1.05x slower)
`262144`		`42.89 µs` (➖ 1.03x slower)

_{Reported by cargo ci bench-report --candidate run:39999720120 --config-file bench_report.toml.}

ludfjig added 3 commits October 23, 2026 10:38

Copilot AI balanced review requested due to automatic review settings October 23, 2026 18:00

ludfjig requested a review from danbugs as a code owner October 23, 2026 18:00

ludfjig added the kind/refactor For PRs that restructure or remove code without adding new functionality. label Oct 23, 2026

ludfjig requested review from andreiltd, dblnz, devigned, jprendes, jsturtevant, simongdavies, squillace and syntactically as code owners October 23, 2026 18:00

Copilot started reviewing on behalf of ludfjig October 23, 2026 18:10

View session

Copilot AI reviewed Oct 23, 2026

View reviewed changes

Conversation

ludfjig commented Oct 23, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

🟢 Approval recommended

Uh oh!

ludfjig commented Oct 23, 2026

Uh oh!

hyperlight-gh-bot Bot commented Oct 23, 2026

Benchmark Results

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Uh oh!

This branch has not been deployed

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants