Weakly link c_guest_dispatch_function by ludfjig · Pull Request #1848 · hyperlight-dev/hyperlight

ludfjig · 2026-09-22T20:21:17Z

Closes #195
Closes #197

weak link default implementation of c_guest_dispatch_function so c guests dont need define one unless they want custom behavior when unknown guest function is called.

Matches implementation to how guest_dispatch_function is implmeented here

hyperlight/src/hyperlight_guest_bin/src/guest_function/call.rs

Lines 17 to 21 in deb063e

    
           core::arch::global_asm!( 
        
               ".weak guest_dispatch_function", 
        
               ".set guest_dispatch_function, {}", 
        
               sym guest_dispatch_function_default, 
        
           );

.

Tested locally. I didn't want to add a new specific c guest just for this test

Copilot

Copilot review overview

🟢 Approval recommended

Only minor documentation and test-coverage nits remain; no blocking issues were identified.

Review effort: Lite
Findings: None

What changed in this PR

Adds a weak default C dispatch function returning NULL, allowing C guests to omit custom unknown-function handling.

Changes:

Added weak fallback dispatch.
Documented C and Rust dispatch behavior.
Updated the changelog.

File	Summary
`src/hyperlight_guest_capi/src/dispatch.rs`	Adds weak fallback dispatch.
`docs/how-to-build-a-hyperlight-guest-binary.md`	Documents default and override behavior.
`CHANGELOG.md`	Records the user-facing change.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

…nt it Signed-off-by: Ludvig Liljenberg <4257730+ludfjig@users.noreply.github.com>

Signed-off-by: Ludvig Liljenberg <4257730+ludfjig@users.noreply.github.com>

hyperlight-gh-bot · 2026-10-02T17:42:03Z

Benchmark Results

Measured commit: a01cd4d80853
Baseline commit: 43cf3539bd10

kvm / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`771.67 ns` (➖ 1.00x faster)
`vec_bytes`	`588.50 ns` (➖ 1.01x slower)
		`374.36 µs` (➖ 1.00x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`550.39 ns` (➖ 1.02x slower)
`65536`	`141.48 ns` (➖ 1.02x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`79.86 ms` (➖ 1.00x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`7.74 ns` (➖ 1.00x faster)	`7.73 ns` (➖ 1.01x slower)	`7.74 ns` (➖ 1.00x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`93.55 µs` (➖ 1.02x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.17 µs` (➖ 1.01x faster)	`7.43 µs` (➖ 1.01x slower)
`65536`	`2.06 µs` (➖ 1.00x slower)	`1.94 µs` (➖ 1.00x slower)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`6.40 µs` (➖ 1.01x slower)	`6.27 µs` (➖ 1.02x faster)
`8192`	`1.06 µs` (➖ 1.00x slower)	`1.05 µs` (➖ 1.00x faster)
`262144`		`27.29 µs` (➖ 1.00x slower)

kvm / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`694.11 ns` (➖ 1.08x faster)
`vec_bytes`	`506.47 ns` (➖ 1.08x faster)
		`664.03 µs` (➖ 1.04x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`505.88 ns` (➖ 1.13x faster)
`65536`	`136.17 ns` (➖ 1.13x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`75.33 ms` (➖ 1.05x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`6.94 ns` (➖ 1.09x faster)	`6.90 ns` (➖ 1.09x faster)	`6.92 ns` (➖ 1.09x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`45.12 µs` (➖ 1.08x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.55 µs` (➖ 1.13x faster)	`7.57 µs` (➖ 1.13x faster)
`65536`	`2.12 µs` (➖ 1.13x faster)	`2.13 µs` (➖ 1.11x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`7.03 µs` (➖ 1.12x faster)	`7.10 µs` (➖ 1.12x faster)
`8192`	`747.00 ns` (➖ 1.12x faster)	`729.39 ns` (➖ 1.08x faster)
`262144`		`30.04 µs` (➖ 1.11x faster)

mshv3 / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`940.14 ns` (➖ 1.01x faster)
`vec_bytes`	`697.08 ns` (➖ 1.02x faster)
		`269.24 µs` (➖ 1.32x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`728.21 ns` (➖ 1.01x faster)
`65536`	`195.90 ns` (➖ 1.01x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`56.40 ms` (➖ 1.01x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`9.98 ns` (➖ 1.04x slower)	`10.00 ns` (➖ 1.05x slower)	`9.67 ns` (➖ 1.06x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`84.19 µs` (➖ 1.01x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`9.58 µs` (➖ 1.04x faster)	`9.99 µs` (➖ 1.11x slower)
`65536`	`2.41 µs` (➖ 1.07x slower)	`2.46 µs` (➖ 1.01x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.08 µs` (➖ 1.00x slower)	`9.03 µs` (➖ 1.13x slower)
`8192`	`1.27 µs` (➖ 1.00x slower)	`1.33 µs` (➖ 1.01x slower)
`262144`		`36.09 µs` (➖ 1.01x faster)

mshv3 / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`943.40 ns` (➖ 1.01x slower)
`vec_bytes`	`660.49 ns` (➖ 1.01x faster)
		`743.66 µs` (➖ 1.11x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`632.94 ns` (➖ 1.01x faster)
`65536`	`165.04 ns` (➖ 1.00x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`65.88 ms` (➖ 1.06x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`8.78 ns` (➖ 1.00x faster)	`9.49 ns` (➖ 1.04x slower)	`8.97 ns` (➖ 1.02x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`44.73 µs` (➖ 1.02x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.77 µs` (➖ 1.00x slower)	`7.75 µs` (➖ 1.04x faster)
`65536`	`2.23 µs` (➖ 1.00x faster)	`2.28 µs` (➖ 1.00x slower)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`7.49 µs` (➖ 1.00x faster)	`7.57 µs` (➖ 1.01x slower)
`8192`	`873.29 ns` (➖ 1.01x slower)	`864.94 ns` (➖ 1.02x faster)
`262144`		`40.71 µs` (➖ 1.07x slower)

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

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`1.20 µs` (➖ 1.07x slower)
`vec_bytes`	`820.98 ns` (➖ 1.16x faster)
		`2.14 ms` (➖ 1.17x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`784.62 ns` (➖ 1.02x faster)
`65536`	`242.11 ns` (➖ 1.06x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`92.75 ms` (➖ 1.19x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`10.07 ns` (➖ 1.38x faster)	`10.23 ns` (➖ 1.04x faster)	`10.20 ns` (➖ 1.03x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`757.02 µs` (➖ 1.04x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`9.07 µs` (➖ 1.04x slower)	`9.12 µs` (➖ 1.02x faster)
`65536`	`2.33 µs` (➖ 1.09x faster)	`2.38 µs` (➖ 1.14x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`9.40 µs` (➖ 1.10x slower)	`8.78 µs` (➖ 1.01x slower)
`8192`	`1.34 µs` (➖ 1.07x faster)	`1.29 µs` (➖ 1.09x faster)
`262144`		`46.09 µs` (➖ 1.28x slower)

hyperv-ws2025 / intel (Windows) (❌ 1.51x)

Top regressions

virtq_readwrite/slot_pool_segmented/262144 — ❌ 1.51x slower

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`1.82 µs` (➖ 1.33x slower)
`vec_bytes`	`817.67 ns` (➖ 1.06x slower)
		`3.14 ms` (➖ 1.01x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`741.18 ns` (➖ 1.00x slower)
`65536`	`212.55 ns` (➖ 1.00x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`113.11 ms` (➖ 1.06x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`10.01 ns` (➖ 1.01x slower)	`10.11 ns` (➖ 1.01x slower)	`10.56 ns` (➖ 1.00x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`585.19 µs` (➖ 1.04x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.81 µs` (➖ 1.03x slower)	`7.60 µs` (➖ 1.02x faster)
`65536`	`2.35 µs` (➖ 1.03x slower)	`2.33 µs` (➖ 1.00x slower)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`10.36 µs` (➖ 1.15x slower)	`10.02 µs` (➖ 1.13x slower)
`8192`	`1.14 µs` (➖ 1.10x faster)	`1.20 µs` (➖ 1.07x slower)
`262144`		`69.41 µs` (❌ 1.51x slower)

_{Reported by cargo ci bench-report --candidate run:37039166647 --baseline run:36945564806 --config-file bench_report.toml.}

Copilot AI lite review requested due to automatic review settings September 22, 2026 20:21

ludfjig requested review from andreiltd, danbugs, dblnz, devigned, jprendes, jsturtevant, simongdavies, squillace and syntactically as code owners September 22, 2026 20:21

Copilot started reviewing on behalf of ludfjig September 22, 2026 20:26

View session

Copilot AI reviewed Sep 22, 2026

View reviewed changes

ludfjig added C-enhancement For PRs adding features, improving functionality, docs, tests, etc. S-needs-review PR is ready for (re-)review labels Sep 22, 2026

jsturtevant previously approved these changes Sep 23, 2026

View reviewed changes

ludfjig dismissed jsturtevant’s stale review via 647e601 September 28, 2026 19:08

ludfjig force-pushed the weak_link_c_guest_dispatch_function branch from 93ea147 to 647e601 Compare September 28, 2026 19:08

simongdavies previously approved these changes Sep 30, 2026

View reviewed changes

ludfjig added 2 commits October 2, 2026 10:12

Weakly link c_guest_dispatch_function so guests don't have to impleme…

e7b94ee

…nt it Signed-off-by: Ludvig Liljenberg <4257730+ludfjig@users.noreply.github.com>

Update docs

a01cd4d

Signed-off-by: Ludvig Liljenberg <4257730+ludfjig@users.noreply.github.com>

ludfjig dismissed simongdavies’s stale review via a01cd4d October 2, 2026 17:12

ludfjig force-pushed the weak_link_c_guest_dispatch_function branch from 647e601 to a01cd4d Compare October 2, 2026 17:12

jsturtevant approved these changes Oct 2, 2026

View reviewed changes

ludfjig merged commit 66d5dd3 into hyperlight-dev:main Oct 2, 2026
68 of 76 checks passed

github-actions Bot removed the S-needs-review PR is ready for (re-)review label Oct 2, 2026

	core::arch::global_asm!(
	".weak guest_dispatch_function",
	".set guest_dispatch_function, {}",
	sym guest_dispatch_function_default,
	);

Conversation

ludfjig commented Sep 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Copilot review overview

🟢 Approval recommended

Uh oh!

hyperlight-gh-bot Bot commented Oct 2, 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

Top regressions

Benchmark Results

function_call_codec

payload_allocation

sandboxes

slot_pool

snapshot_files

virtq_readonly

virtq_readwrite

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

ludfjig commented Sep 22, 2026 •

edited

Loading