Make sccache read-only outside main by ludfjig · Pull Request #1909 · hyperlight-dev/hyperlight

ludfjig · 2026-10-08T20:31:59Z

Make sccache read-only on every ref except main, so only main writes sccache entries to the GitHub Actions cache. PR runs still read main's entries.

This fixes two problems caused by PR runs writing sccache entries, one per compiled crate:

They filled the repository cache past its 10 GB limit and evicted the main caches that every PR reads.
They exceeded the cache rate limit of 200 uploads per minute per repository (HTTP 429). When that happens, rust-cache reports "No cache found" and the job builds from scratch. In run 37820817731, 15 of 39 rust-cache restores failed this way.

With these changes, PRs lose sccache reuse between pushes, but this is fine since PR entries were evicted within an hour anyway (due to other prs evicting them), so no big deal. sccache lookups still count toward the download limit, so some rate limiting may remain.

PR runs filled the cache past its limit and evicted the main caches that every PR reads. PRs lose sccache reuse between pushes, but this is fine since PR entries were evicted within an hour anyway, so no big deal. Signed-off-by: Ludvig Liljenberg <4257730+ludfjig@users.noreply.github.com>

Copilot

🟢 Approval recommended

The focused configuration change correctly applies sccache’s supported read-only mode to non-main refs.

0 open findings

What changed in this PR

Makes sccache read-only outside main to protect shared cache entries from eviction.

Changes:

Sets SCCACHE_GHA_RW_MODE=READ_ONLY for non-main refs.
Documents the cache-write policy.

File	Description
`.github/actions/setup-sccache/action.yml`	Restricts sccache writes to `main`.

🧠 Review effort: Balanced

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

hyperlight-gh-bot · 2026-10-08T21:15:19Z

Benchmark Results

Measured commit: b097b7e9dede
Baseline commit: 73f8f3ef9d3d

kvm / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`814.80 ns` (➖ 1.08x slower)
`vec_bytes`	`590.09 ns` (➖ 1.01x slower)
		`373.83 µs` (➖ 1.01x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`533.89 ns` (➖ 1.01x slower)
`65536`	`151.59 ns` (➖ 1.07x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`88.45 ms` (➖ 1.10x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`8.10 ns` (➖ 1.01x slower)	`8.45 ns` (➖ 1.09x slower)	`10.29 ns` (➖ 1.30x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`92.58 µs` (➖ 1.01x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.65 µs` (➖ 1.07x slower)	`7.66 µs` (➖ 1.06x slower)
`65536`	`2.15 µs` (➖ 1.11x slower)	`2.08 µs` (➖ 1.00x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`6.28 µs` (➖ 1.01x faster)	`6.98 µs` (➖ 1.11x slower)
`8192`	`1.08 µs` (➖ 1.03x slower)	`1.08 µs` (➖ 1.03x slower)
`262144`		`28.01 µs` (➖ 1.03x slower)

kvm / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`717.83 ns` (➖ 1.04x slower)
`vec_bytes`	`528.24 ns` (➖ 1.02x faster)
		`679.90 µs` (➖ 1.01x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`506.14 ns` (➖ 1.02x faster)
`65536`	`136.85 ns` (➖ 1.00x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`70.93 ms` (➖ 1.08x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`7.04 ns` (➖ 1.06x faster)	`6.99 ns` (➖ 1.02x faster)	`6.98 ns` (➖ 1.04x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`45.14 µs` (➖ 1.01x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.56 µs` (➖ 1.00x slower)	`7.59 µs` (➖ 1.00x slower)
`65536`	`2.14 µs` (➖ 1.00x slower)	`2.16 µs` (➖ 1.04x slower)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`7.21 µs` (➖ 1.02x slower)	`7.11 µs` (➖ 1.00x faster)
`8192`	`743.53 ns` (➖ 1.02x faster)	`728.30 ns` (➖ 1.01x slower)
`262144`		`29.87 µs` (➖ 1.05x faster)

mshv3 / amd (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`985.30 ns` (➖ 1.05x slower)
`vec_bytes`	`705.57 ns` (➖ 1.02x faster)
		`315.17 µs` (➖ 1.02x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`721.79 ns` (➖ 1.07x slower)
`65536`	`183.05 ns` (➖ 1.01x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`61.52 ms` (➖ 1.02x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`10.55 ns` (➖ 1.06x slower)	`9.91 ns` (➖ 1.00x faster)	`10.00 ns` (➖ 1.01x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`86.23 µs` (➖ 1.00x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`9.25 µs` (➖ 1.01x faster)	`9.97 µs` (➖ 1.11x slower)
`65536`	`2.64 µs` (➖ 1.01x faster)	`2.47 µs` (➖ 1.01x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.12 µs` (➖ 1.00x faster)	`8.11 µs` (➖ 1.03x faster)
`8192`	`1.43 µs` (➖ 1.10x slower)	`1.32 µs` (➖ 1.06x faster)
`262144`		`35.07 µs` (➖ 1.00x faster)

mshv3 / intel (Linux) (➖ stable)

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`952.09 ns` (➖ 1.03x slower)
`vec_bytes`	`657.27 ns` (➖ 1.01x slower)
		`686.36 µs` (➖ 1.03x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`636.75 ns` (➖ 1.03x slower)
`65536`	`179.18 ns` (➖ 1.10x slower)

sandboxes

	`create_initialized_and_drop`
`medium`	`61.96 ms` (➖ 1.08x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`9.66 ns` (➖ 1.00x faster)	`8.79 ns` (➖ 1.02x faster)	`8.80 ns` (➖ 1.00x slower)

snapshot_files

	`load_snapshot_unverified`
`small`	`44.47 µs` (➖ 1.02x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`8.00 µs` (➖ 1.01x slower)	`7.96 µs` (➖ 1.01x slower)
`65536`	`2.30 µs` (➖ 1.00x slower)	`2.28 µs` (➖ 1.02x slower)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`7.59 µs` (➖ 1.01x slower)	`7.52 µs` (➖ 1.00x faster)
`8192`	`910.73 ns` (➖ 1.02x faster)	`910.08 ns` (➖ 1.01x slower)
`262144`		`39.09 µs` (➖ 1.00x 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` (➖ 1.10x faster)
`vec_bytes`	`742.94 ns` (➖ 1.01x faster)
		`2.21 ms` (➖ 1.01x slower)

payload_allocation

	`slot_pool_segmented`
`262144`	`840.64 ns` (➖ 1.07x slower)
`65536`	`230.12 ns` (➖ 1.00x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`106.08 ms` (➖ 1.01x slower)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`11.72 ns` (➖ 1.10x slower)	`10.62 ns` (➖ 1.00x slower)	`10.54 ns` (➖ 1.00x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`777.55 µs` (➖ 1.19x slower)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`8.60 µs` (➖ 1.00x slower)	`8.66 µs` (➖ 1.00x slower)
`65536`	`2.53 µs` (➖ 1.01x faster)	`2.53 µs` (➖ 1.01x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.77 µs` (➖ 1.01x slower)	`8.91 µs` (➖ 1.02x slower)
`8192`	`1.38 µs` (➖ 1.04x faster)	`1.40 µs` (➖ 1.03x slower)
`262144`		`37.36 µs` (➖ 1.04x faster)

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

No benchmark improved or regressed.

Benchmark Results

function_call_codec

	`encode_control`	`decode_vec_bytes_copy`
`byte_chunks`	`1.22 µs` (➖ 1.02x slower)
`vec_bytes`	`787.83 ns` (➖ 1.07x faster)
		`3.06 ms` (➖ 1.00x faster)

payload_allocation

	`slot_pool_segmented`
`262144`	`745.56 ns` (➖ 1.00x slower)
`65536`	`211.15 ns` (➖ 1.02x faster)

sandboxes

	`create_initialized_and_drop`
`medium`	`107.13 ms` (➖ 1.11x faster)

slot_pool

	`alloc_dealloc_1500`	`alloc_dealloc_4096`	`alloc_dealloc_128`
	`10.57 ns` (➖ 1.02x slower)	`10.24 ns` (➖ 1.01x slower)	`10.69 ns` (➖ 1.02x faster)

snapshot_files

	`load_snapshot_unverified`
`small`	`539.29 µs` (➖ 1.52x faster)

virtq_readonly

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`262144`	`7.83 µs` (➖ 1.01x slower)	`7.76 µs` (➖ 1.12x faster)
`65536`	`2.33 µs` (➖ 1.14x faster)	`2.32 µs` (➖ 1.01x faster)

virtq_readwrite

	`slot_pool_segmented_fragmented`	`slot_pool_segmented`
`65536`	`8.37 µs` (➖ 1.03x faster)	`8.36 µs` (➖ 1.06x faster)
`8192`	`1.12 µs` (➖ 1.21x faster)	`1.11 µs` (➖ 1.09x faster)
`262144`		`41.05 µs` (➖ 1.06x faster)

_{Reported by cargo ci bench-report --candidate run:37840164996 --config-file bench_report.toml.}

Copilot AI balanced review requested due to automatic review settings October 8, 2026 20:32

ludfjig requested review from andreiltd, danbugs, dblnz, devigned, jprendes, jsturtevant, simongdavies, squillace and syntactically as code owners October 8, 2026 20:32

Copilot started reviewing on behalf of ludfjig October 8, 2026 20:32

View session

Copilot AI reviewed Oct 8, 2026

View reviewed changes

ludfjig added the C-bug For PRs that fix bugs label Oct 8, 2026

jprendes approved these changes Oct 8, 2026

View reviewed changes

ludfjig merged commit 9980dd3 into hyperlight-dev:main Oct 8, 2026
62 of 63 checks passed

Conversation

ludfjig commented Oct 8, 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

🟢 Approval recommended

Uh oh!

hyperlight-gh-bot Bot commented Oct 8, 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!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ludfjig commented Oct 8, 2026 •

edited

Loading