perf(codegen/runtime): #6812 — spill lanes make append-past-capacity arrays clone-eligible (w13 beats node)#6856
Merged
Merged
Conversation
added 2 commits
July 26, 2026 17:31
The guard classifies each lane as INLINE or SPILL from the first receiver (spill = slot past the inline alloc_limit but covered by the object-owned meta.spill buffer: live non-forwarded array, length high-water > slot); every later receiver must sit on the SAME side of its own alloc_limit and pass the same coverage proof, so the emitter's per-lane store sequence is uniform across the proven prefix. Spill lanes set bit 15 of their u16 result lane — the +1 packing cannot carry into it (find_slot caps slots at 4096). Typed-layout-intact receivers reject spill lanes conservatively. The fast nest branches per lane on the loop-invariant flag (LLVM unswitches it): inline lanes store as before; spill lanes store through obj → meta(word header-1) → spill(word 4) → elements(word 1+slot). Both paths stay call-free raw numeric stores, preserving the guard's no-GC interval; const assertions beside the runtime structs lock the codegen offsets. The store chain ends in per-lane done blocks, so the inner back-edge gained a dedicated latch and the counter phi names its true predecessor. This makes append-past-capacity arrays (canonical w13: a 6th key on 5-field literals) clone-eligible after the #6841 peel primes their shapes. Claude-Session: https://claude.ai/code/session_01QJ5mwMDPc63tNLAFPdthAG
Contributor
Author
|
Follow-up evidence on the final tip:
|
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughWalkthroughObject-array numeric write cloning now classifies lanes as inline or spill, validates spill coverage across receivers, encodes lane modes, and emits branch-local stores to object fields or object-owned spill buffers. Runtime layout assertions and documentation describe the updated path. ChangesSpill lane clone path
Estimated code review effort: 4 (Complex) | ~45 minutes Sequence Diagram(s)sequenceDiagram
participant object_array_numeric_write_slots
participant lower_object_array_write_versioned_for
participant ObjectMeta_spill_buffer
object_array_numeric_write_slots->>object_array_numeric_write_slots: classify and encode spill lanes
object_array_numeric_write_slots->>lower_object_array_write_versioned_for: pass packed lane values
lower_object_array_write_versioned_for->>ObjectMeta_spill_buffer: store spill lanes through meta
lower_object_array_write_versioned_for-->>lower_object_array_write_versioned_for: continue through inner latch
Possibly related issues
Possibly related PRs
Suggested labels: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@changelog.d/6812-spill-lanes.md`:
- Line 1: Align the w13 benchmark baseline across
changelog.d/6812-spill-lanes.md at line 1 and docs/object-write-matrix.md at
line 50 by using one authoritative timing, or explicitly document the distinct
measurement protocols and commits for the approximately 155 ms and 160 ms
results; preserve the reported optimized timing and surrounding benchmark
context.
In `@crates/perry-codegen/src/stmt/loops.rs`:
- Around line 2650-2654: Update the meta_word calculation in the loop emitter to
use the ILP32 ObjectHeader byte layout: derive the meta field’s byte offset as
header size minus the 4-byte field width, matching the meta_ptr_size computation
in new.rs, rather than converting the header size to 8-byte words. Preserve
header_words for its existing uses.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: 53671d6a-a8bc-40b4-9512-7d4b8ac07c62
📒 Files selected for processing (5)
changelog.d/6812-spill-lanes.mdcrates/perry-codegen/src/stmt/loops.rscrates/perry-runtime/src/object/mod.rscrates/perry-runtime/src/proxy/put_value.rsdocs/object-write-matrix.md
…eline The spill store path addressed 'meta' as 8-byte word (header_words - 1), which on ILP32 (arm64_32: 24-byte header, 4-byte meta at offset 20) reads the wrong pointer. Address it by byte offset with a pointer-width load (zext to i64 on ILP32), mirroring the new.rs allocator's meta store. Changelog now cites the same w13 baseline lineage as the matrix doc. Claude-Session: https://claude.ai/code/session_01QJ5mwMDPc63tNLAFPdthAG
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
The final piece of the w13 ladder: the whole-loop write clone learns spill lanes, so arrays whose written property lives past the inline slot capacity — the append-then-update pattern (
objs[i][7] = vwhere7was added to 5-field literals) — run in the call-free clone instead of the generic per-write path.Canonical w13: ~155 ms → 7 ms (node ~13 ms) — beats node. Guard trace shows zero rejections; checksums identical.
Guard (
object_array_numeric_write_slots):max(field_count, floor)) or spill (slot past it but covered by the object-ownedmeta.spillbuffer — validated as a live, non-forwarded plain array whose length high-water exceeds the slot, mirroring the shared-keys validation).+1packing cannot carry into it (find_slotcaps slots at 4096).Emitter (
lower_object_array_write_versioned_forfast nest):obj → meta(wordheader-1, the documented last header field)→ spill(ObjectMeta word 4)→ elements(one word past the 8-byte ArrayHeader) — two dependent loads, still call-free, preserving the guard's no-GC interval. Numeric stores create no references, so no barriers or layout notes (the same audit argument as the inline lanes).offset_of!(ObjectMeta, spill) == 32,size_of::<ArrayHeader>() == 8) lock the codegen offsets.doneblocks, so the inner back-edge gained a dedicated latch block and the counter phi names its true predecessor.Validation
spill_lanes_sanity.ts(5 cases): canonical append-past-capacity, two spill lanes + one inline lane in a single nest, the mode-mismatch fallback (learned-resize builders), post-clone readback through get/entries/values/JSON, dynamic inner bound + spill lane — byte-identical vs node v26.3.0 in all four runtime modes (default;PERRY_GC_FORCE_EVACUATE=1 PERRY_GC_VERIFY_EVACUATION=1;PERRY_OBJECT_SPILL=0where spill lanes correctly self-reject since legacy objects have no buffers;PERRY_GEN_GC=0).Refs #6812.
https://claude.ai/code/session_01QJ5mwMDPc63tNLAFPdthAG
Summary by CodeRabbit
Performance
Bug Fixes
Documentation