Skip to content

[MiniMax-M3.1] NVFP4 KV cache and index cache (--kv-cache-dtype nvfp4, --index-cache-dtype nvfp4) - #3310

Open
ganyi1996ppo wants to merge 6 commits into
mainfrom
ganyi/m31-nvfp4-kv-index
Open

ganyi1996ppo wants to merge 6 commits into
mainfrom
ganyi/m31-nvfp4-kv-index

Conversation

@ganyi1996ppo

@ganyi1996ppo ganyi1996ppo commented Nov 13, 2026 •

Copy link
Copy Markdown
Contributor

Motivation

#2962 added the MXFP4 KV and index caches for MiniMax-M3.1. This adds the NVFP4 pair (e2m1 data, one e4m3 scale per 16 elements, one fp32 global scale per layer for K and V) as a second FP4 option: --kv-cache-dtype nvfp4 --index-cache-dtype nvfp4. The kernels live in aiter (five open PRs, below); this PR is the ATOM side: flags and validation, pool and index-cache layout, the write call, and the selector / sparse-PA wiring. Draft until the aiter PRs land.

Technical Details

Hard dependencies (aiter; all five are now in aiter main, see the status below; the page format is the contract between the writer #7824 and the readers):

aiter PR provides
#7813 minimax_m3_index_score_nvfp4_flydsl (decode index scorer)
#7814 minimax_m3_index_score_prefill_nvfp4_flydsl (prefill index scorer; stacked on #7813)
#7824 NVFP4 mode of fused_qknorm_idxrqknorm (kv_cache_dtype="nvfp4" + index_cache_dtype="nvfp4"): norm / RoPE and quantization of K, V and index-K into the pages in one pass
#7825 sparse_pa_nvfp4_decode
#7826 sparse_pa_nvfp4_prefill (stacked on #7825)

Status 2026-11-16: all five are in aiter main. #7824 landed as 47d23011bd. #7826 landed as 8f574bdf51 and, being stacked on #7825, carries the decode kernel too (#7825 was closed as landed). #7814 landed as dad1b66282 and carries #7813's decode scorer (#7813 itself is still open for review nits; its code is on main). Checked with git grep on main f858b48e68 for the symbols this PR imports (score_flydsl_nvfp4, score_prefill_flydsl_nvfp4, sparse_pa_nvfp4_decode, sparse_pa_nvfp4_prefill); the Python signatures this PR calls are unchanged by the merged versions [read]. The merged code went through further review rounds after the heads listed under Test Result (for example the decode split scratch is now a DecodeWorkspace, allocated on first use and refused during graph capture); the GPU results below were not re-run on aiter main [not measured]. They are imported where the NVFP4 path is first reached, so the other caches never load them.

ATOM changes (14 files, +1001 / -72 against #2962; commits: config and validation, writer + selector + sparse PA wiring, read-path tests, docs):

  • atom/config.py, arg_utils.py: the two flags take nvfp4; FP4_CACHE_DTYPES = ("fp4", "nvfp4") replaces the "fp4" tests of [MiniMax-M3.1] MXFP4 index cache and MXFP4 KV cache (--index-cache-dtype fp4, --kv-cache-dtype fp4) #2962 (draft dtype, scorer allocation, CP gate). The index dtype follows the KV dtype when omitted.
  • atom/models/minimax_m3/config.py (startup validation, not the model definition): every [MiniMax-M3.1] MXFP4 index cache and MXFP4 KV cache (--index-cache-dtype fp4, --kv-cache-dtype fp4) #2962 refusal applies to nvfp4 through the same code (vLLM/SGLang bridges, DCP, PCP, sequence parallelism, --block-size != 128, non-gfx950, non-bf16 model, KV offload / --kv-transfer-config, a draft left with FP4 KV, layers that are not all indexed sparse). New: MXFP4 and NVFP4 never pair, an NVFP4 index needs an NVFP4 KV, nvfp4 is refused off MiniMax-M3.
  • nvfp4_pages.py (new): page geometry, Nvfp4KvLayout (one 2 * Hkv * 9216 B field a page), index_page_views. aiter_attention.py: the 9216 B index-cache spec and the layer's fp32 [1] K / V scales (1.0: the checkpoint carries none).
  • attention_mha.py: rope_cache calls the fused op with the nvfp4 dtypes (same call as MXFP4); the selector emits logical block ids (no page-16 table) and the NVFP4 sparse PA reads them. index_topk.py / index_buffers.py: IndexScoreBuffers.nvfp4 picks the NVFP4 page views and scorers. sparse_attn_nvfp4.py (new): thin wrappers and the prefill row positions.
  • Not changed: the M3.1 model definition files.
  • Fail closed: an unsupported combination is refused at startup with a message, never run as FP8. Mono decode refuses non-fp8 caches (existing mono/dispatch.py), so nvfp4 runs the regular decode path. The indexer-CP flag is cleared with the existing warning under any FP4 cache (the gate argument now covers nvfp4).

On an image whose aiter predates those PRs (image aiter 627461276, which has none of the symbols): [measured] with --kv-cache-dtype nvfp4 the server passes config validation and memory sizing, then aborts at CUDA-graph capture with aiter's fused_qknorm_idxrqknorm.cu:1069 sparse insert mode expects index_cache_dtype='auto', fp8 or mxfp4 (TP2, one run). All other caches are untouched; the new tests skip.

Known limitations (read from the code, not measured unless stated):

  • Non-mono decode only; no MTP / speculative draft with FP4 KV (refused), DSpark / EAGLE own-pool drafts are bf16 as for MXFP4 (not run with nvfp4); no KV offload / P-D transfer; no DCP / PCP / sequence parallelism; no indexer CP (flag cleared with a warning); gfx950 only.
  • The global K / V scale is fixed at 1.0.
  • The aiter writer's contract: no two tokens of one call share a slot (V is written by 32-bit atomic and/or).

Test Plan

  • Lint: black and ruff on the changed files.
  • CPU: the new tests (cache geometry, validation, index-scorer record) and the existing MXFP4 / FlyDSL-optional tests; a full tests/ run against main with aiter / triton / flydsl made unimportable (the non-GPU CI job's condition).
  • GPU (gfx950): read-path test (selector + sparse PA vs oracles on dequantized pages), the aiter op tests of the five PRs, a byte comparison of the aiter fused writer against ATOM's earlier Triton writer, TP2 and TP4 server runs with fixed prompts and GSM8K against fp8 and MXFP4 caches.

Test Result

All GPU runs used a local aiter stack: image aiter 627461276 plus the net diff of the five PR heads (#7813 80ff1e7ee4, #7814 44fee0620b, #7824 2f0bda9b0e, #7825 af14f1c96d, #7826 f3d1915843), not aiter main. Model amd/MiniMax-M3.1-Flash-Quark-MXFP4 with the recipe's online ptpc_fp8 config, --block-size 128, --max-model-len 135000, mono off, no speculative decoding, prefix caching at its default. MI355X (gfx950).

Lint / CPU:

  • black (26.10.1) and ruff (0.17.5): clean on the 13 changed Python files.
  • New and related tests (nvfp4 cache / attention, mxfp4 cache, flydsl-optional) with the image aiter and no GPU: 62 passed; the read test skips (5 skipped); with the local aiter stack, same 62 passed, 5 skipped (no GPU).
  • Full tests/ (aiter / triton / flydsl blocked, no GPU, --continue-on-collection-errors): main 9671 passed / 26 failed / 32 collection errors, this branch 9695 passed / 25 failed / 32 collection errors. The 32 errors and 24 of the failures are the same set on both. The remaining ones: on main two test_benchmark_catalog failures (git worktree not visible in the container), on the branch test_atomesh_spur_dispatch::test_dispatch_waits_until_controller_reports_running, which also fails on main when run alone (a 15 s script timeout). The container is not the CI runner (missing optional packages), so these are comparisons against main, not the CI result.

GPU:

  • aiter op_tests/test_fused_qknorm_nvfp4.py (writer pages vs a torch reference, byte for byte): exit 0, every case 0 mismatching bytes. test_flydsl_minimax_m3_sparse_pa_nvfp4.py (decode + prefill, correctness cases only), test_flydsl_minimax_m3_index_score.py and test_flydsl_minimax_m3_index_score_prefill_fp4.py (-b 1 --q-len 512 --ctx-len 8192): exit 0, no failing row. These are the aiter PRs' own tests on the local stack, one run each.
  • tests/test_minimax_m3_nvfp4_read.py (3 decode cases, S 1 and 4, Hkv 1 and 2, global scales 1.0 / 0.37 / 1.9; 2 prefill cases): 5 passed. The selected block sets equal the oracle's exactly (the oracle rounds K to e4m3, the scorers' default k_round); attention is checked against an fp64 reference on the selected blocks with a relative-L2 bound of 2.5e-3.
  • Fused writer vs ATOM's earlier Triton writer (nvfp4_write.py of the source branch): same qkv and slots, whole kv and index pages compared, 90 cases (tokens 1/7/64/300/2048 x Hkv 1/2/4 x global scale 1.0/0.37/3.0 x scattered/sequential slots), 0 mismatching bytes. The Triton writer is not part of this PR.
  • Memory per KV block (all 60 layers, K + V + index; from the server log): TP4 nvfp4 1,658,880 B, fp8 3,010,560 B. With the same --gpu-memory-utilization 0.85: 107,301 vs 59,172 blocks at TP4.
  • Fixed prompts, TP4, greedy (temperature 0, 96 tokens, chat API, thinking off, one request at a time, one run per cache): 4 short chat prompts and 3 long ones (8K, 32K, 100K tokens, "summarize the text" questions) with --kv-cache-dtype nvfp4 --index-cache-dtype nvfp4 finish and give coherent text. Against fp8 caches, one prompt (chat0) is identical and the other six diverge after 45 to 193 characters of the reasoning text; this is not an equality claim.
  • GSM8K, lm_eval 0.4.13, 5-shot chat, full 1319 questions, greedy, max_tokens 2048, num_concurrent 32, one run each, offline dataset copy:
TP KV / index cache flexible-extract strict-match
4 fp8 / fp8 0.9469 (+-0.0062) 0.9060 (+-0.0080)
4 nvfp4 / nvfp4 0.9447 (+-0.0063) 0.8961 (+-0.0084)
2 fp8 / fp8 0.9469 (+-0.0062) 0.9136 (+-0.0077)
2 fp4 / fp4 (MXFP4, #2962 code) 0.9462 (+-0.0062) 0.9030 (+-0.0082)
2 nvfp4 / nvfp4 0.9477 (+-0.0061) 0.8916 (+-0.0086)

flexible-extract is within 0.3 points across all rows; strict-match is 1.0 to 2.2 points lower for nvfp4 than for fp8 (about 1 to 2.6 standard errors, single runs, run-to-run noise not measured). The cause of the strict-match gap (answer formatting vs content) was not investigated.

Not measured: throughput / latency of any kind; accuracy beyond GSM8K (no long-context benchmark); TP8 or other TPs; speculative decoding with nvfp4; prefix-cache behaviour with nvfp4 pages specifically (prefix caching was on, no hit-rate check); aiter main (the stack above is the image aiter plus the five PRs); the CI GPU image; repeated runs.

Submission Checklist

  • black and ruff clean on the changed files
  • Docs updated (docs/configuration_guide.md: flags, conditions, minimum aiter PRs)
  • Tests added; the ones that need the new kernels skip without them
  • aiter #7813, #7814, #7824, #7825, #7826 merged (draft until then)
  • CI

Update 2026-11-15 (merge of main and aiter's rename)

Merged main (3298107) into the branch (a merge commit, no rebase) and followed aiter's rename of the sparse-PA ops (aiter #7825/#7826 review): minimax_m3_sparse_pa_nvfp4_{decode,prefill} -> sparse_pa_nvfp4_{decode,prefill}.

  • One conflict area, from feat(sparse-attn): decode-time indexer context parallelism across TP ranks on the MXFP4 index cache (MiniMax-M3.1 first) #3104 (indexer CP on the MXFP4 index cache): IndexScoreBuffers keeps main's cp_world / cp_rank and this PR's nvfp4 flag. The indexer-CP gate keeps refusing NVFP4 (the NVFP4 decode scorer takes no CP shard) but no longer MXFP4, so indexer_cp_unsupported_reason takes nvfp4_caches in place of the removed mxfp4_caches; a new test checks that an NVFP4 cache clears the flag with the warning.
  • Rename: sparse_attn_nvfp4.py imports the aiter ops under the new names; the _KERNELS list of tests/test_minimax_m3_nvfp4_read.py and the docstring of nvfp4_pages.py follow.
  • [measured] black 26.10.1 and ruff 0.17.3 clean on the 13 changed Python files. CPU tests (image aiter, no GPU: tests/test_minimax_m3_*.py, test_minimax_m31_shared_fusion.py, test_indexer_cp_gate.py, test_indexer_cp_gather_ragged.py): branch 401 passed / 4 failed / 108 skipped, main 373 passed / 4 failed / 103 skipped; the same 4 tests fail on both (test_minimax_m31_shared_fusion.py, 'FusedMoE' object has no attribute 'quant_method', container environment); the +28 passed are this PR's NVFP4 tests.
  • [measured] with aiter main plus the five aiter PR heads, the four NVFP4 names resolve by import on CPU and the old names are gone.
  • Not re-run after the merge and the rename (no GPU was free): the GPU read/writer tests, the aiter op tests on the renamed ops, the server runs, GSM8K, the fixed-prompt and memory numbers above. Those numbers were taken on the earlier aiter heads (#7813 80ff1e7ee4, #7814 44fee0620b, #7824 2f0bda9b0e, #7825 af14f1c96d, #7826 f3d1915843) and ATOM 5aebd79; the aiter heads have moved since (now #7813 6e6f677bf0, #7814 dab3ee3c5a, #7824 382b6e5909, #7825 d83ef9e5ee, #7826 23d5a9d6c6), so do not read them as numbers of the merged branch.

🤖 Generated with Claude Code

ganyi1996ppo and others added 4 commits November 13, 2026 01:30
…idation, pool layout

NVFP4 (e2m1 + one e4m3 scale per 16) K/V pages and index pages for MiniMax-M3.1,
next to the MXFP4 ones: the two flags, their validation (fail closed, same refusals
as the MXFP4 caches, plus: MXFP4 and NVFP4 never pair, nvfp4 is M3 only), the KV
pool layout (one 2 * Hkv * 9216 B field a page) and the 9216 B index cache spec.
FP4_CACHE_DTYPES names the two FP4 dtypes where the MXFP4 code tested "fp4".

Co-Authored-By: Claude <noreply@anthropic.com>
…FP4 selector and sparse PA

rope_cache takes the NVFP4 mode of aiter's fused_qknorm_idxrqknorm (kv_cache_dtype and
index_cache_dtype "nvfp4"): norm / RoPE and the quantization of K, V and the index key
into the pages in one pass, the layer's fp32 [1] global K / V scales riding along. The
selector returns the logical block ids (no page-16 table is emitted) and the NVFP4
sparse-PA decode / prefill kernels read them with the request block table. The index
scorer record says which FP4 index scorer to run (`IndexScoreBuffers.nvfp4`).

Needs aiter ROCm/aiter#7813, #7814, #7824, #7825, #7826; their symbols are imported where
the NVFP4 path is first reached, so nothing changes for the other caches.

Co-Authored-By: Claude <noreply@anthropic.com>
…cles on gfx950

Pages are written by aiter's fused NVFP4 writer from random qkv rows; the oracles read
the pages back through a token-major reader built from the documented page offsets, so
the test pins the layout the writer and the readers share. Skips without gfx950 or an
aiter that has the NVFP4 kernels.

Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Claude <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

🏷️ CI Guide

Runs automatically on every eligible PR before approval:

  • ✅ Pre Checkin: Black, Ruff, catalog schema validation, non-GPU unit tests

Heavy model tests:

  • ✅ Run after the PR is approved and Pre Checkin passes
  • ✅ Run immediately when an approval review is submitted
  • ✅ Can be requested before approval with labels
Label Tests
ci:full Run all heavy PR model tests: native ATOM, vLLM, and SGLang
ci:atom Run native ATOM model accuracy tests
ci:vllm Run ATOM vLLM OOT model accuracy tests
ci:sglang Run ATOM SGLang model accuracy tests

Heavy jobs are skipped when the PR is not approved and no matching ci:* label is present.
Add labels via the sidebar or gh pr edit 3310 --add-label <label>

ganyi1996ppo and others added 2 commits November 15, 2026 07:30
Conflicts (#3104 indexer CP on the MXFP4 index cache): keep main's CP shard
fields in IndexScoreBuffers next to the NVFP4 flag, and keep the indexer-CP
gate for NVFP4 only (indexer_cp_unsupported_reason(nvfp4_caches=...)), since
aiter's NVFP4 decode scorer takes no cp_world/cp_rank.

Co-Authored-By: Claude <noreply@anthropic.com>
…vfp4_{decode,prefill}

Co-Authored-By: Claude <noreply@anthropic.com>
@ganyi1996ppo
ganyi1996ppo marked this pull request as ready for review November 16, 2026 08:35
Copilot AI balanced review requested due to automatic review settings November 16, 2026 08:35

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟡 Changes recommended

The fused-writer test skips every failure, potentially hiding writer/read-path regressions in all five end-to-end cases; only the known unsupported-NVFP4-mode error should trigger a skip.

1 open issue
What changed in this PR

Adds NVFP4 KV and index caches to MiniMax-M3.1’s native ATOM inference path, extending the existing MXFP4 support with a second FP4 format backed by AITER kernels.

Changes:

  • Adds nvfp4 cache options, matching-dtype defaults, and startup validation for unsupported configurations.
  • Implements packed page layouts and connects cache writes, index scoring, and sparse attention.
  • Adds cache-layout, validation, and read-path tests, plus configuration and dependency documentation.
File Description
tests/​test_minimax_m3_nvfp4_read.py Checks selection and attention against dequantized-page references.
tests/​test_minimax_m3_nvfp4_cache.py Tests page geometry, pool allocation, and configuration restrictions.
tests/​test_minimax_m3_nvfp4_attention.py Tests prefill row mapping and scorer dispatch.
tests/​test_minimax_m3_mxfp4_cache.py Allows the shared config helper to select architectures.
docs/​configuration_guide.md Documents NVFP4 flags, restrictions, and required AITER kernels.
atom/​models/​minimax_m3/​config.py Extends FP4 validation to NVFP4 and rejects incompatible cache pairings.
atom/​model_ops/​minimax_m3/​sparse_attn_nvfp4.py Adds sparse-attention wrappers and prefill row mapping.
atom/​model_ops/​minimax_m3/​nvfp4_pages.py Defines NVFP4 page geometry, views, and KV pool layout.
atom/​model_ops/​minimax_m3/​index_topk.py Dispatches index scoring to the matching FP4 kernels.
atom/​model_ops/​minimax_m3/​index_buffers.py Allocates and identifies NVFP4 scorer buffers.
atom/​model_ops/​attentions/​aiter_attention.py Allocates index pages and per-layer global K/V scales.
atom/​model_ops/​attention_mha.py Connects NVFP4 writes, logical block selection, and attention.
atom/​model_engine/​arg_utils.py Exposes NVFP4 cache choices in the CLI.
atom/​config.py Resolves NVFP4 defaults, draft cache types, and model/parallelism restrictions.

Review effort: Balanced

Comment on lines +91 to +92
except Exception as e: # noqa: BLE001 - an aiter without the NVFP4 write mode
pytest.skip(f"aiter's fused writer has no NVFP4 mode: {e}")

@valar-cop valar-cop left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

a few small things..

gpu re-run of the read test and one server start with --kv-cache-dtype nvfp4 against aiter main?

ci green.

if spec is not None and spec.draft_kv_dtype == "fp4":
# "fp4" is left only on a draft in the target's pool (MTP), which has no
# MXFP4 attention impl to write its rows; an own-pool draft is bf16 and passes.
if spec is not None and spec.draft_kv_dtype in FP4_CACHE_DTYPES:

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

name and messages still say MXFP4, but now gates nvfp4 too. please update _fp4_kv_refusal, _check_fp4_kv, _declare_kv_pools and index_scorer docstrings.

weight(), weight(), 1, slots, kv, kv, idx, 128, q_out, iq_out, slots,
kv_cache_dtype="nvfp4", index_cache_dtype="nvfp4", k_scale=ks, v_scale=vs,
) # fmt: skip
except Exception as e: # noqa: BLE001 - an aiter without the NVFP4 write mode

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

catching every exception as skip hides real failures. gate on the nvfp4 write mode up front, or catch only the unsupported-mode error?


from atom.model_ops.attentions.pool_layout.entry_arena import EntryField

PAGE_TOKENS = 128

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

PAGE_TOKENS duplicates geometry.SPARSE_BLOCK_SIZE, import it instead of hard coding 128?

@valarLip valarLip left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Action required before merge (6)

  1. Re-run the GPU evidence on aiter main and this head, and add perf numbers. Every GPU result in the description (read test, server runs, GSM8K, memory) was taken on the pre-merge aiter heads + ATOM 5aebd79; aiter main has since changed the API under you (ops renamed, sparse_pa_nvfp4_decode now uses a lazily allocated DecodeWorkspace that is refused inside graph capture). Post tests/test_minimax_m3_nvfp4_read.py + one TP4 server start (CUDA-graph capture) + GSM8K on aiter main 0b0381a1a8 or later, plus decode TPOT / output throughput vs fp8 and MXFP4 (fp4) at the same TP/concurrency. Also state the reason in the description: M3.1's native fp4 KV/index format is NVFP4 and external benchmark rules reject non-native fp4 caches.
  2. Stop turning writer failures into skips. tests/test_minimax_m3_nvfp4_read.py:84-92 catches every Exception around fused_qknorm_idxrqknorm and skips, so a wrong layout or a launch error skips all 5 cases. All the NVFP4 symbols are on aiter main now, so drop that try and the _missing_kernels getattr probe; keep only the gfx950 skip (Copilot + valar-cop raised this too; not addressed).
  3. Declare state instead of stashing attributes. attention_mha.py:1771-1775 stashes prefill_md.nvfp4_row_info through getattr(..., None). Make it a field of MiniMaxM3SparsePrefillMetadata built once in make_sparse_prefill_metadata. aiter_attention.py:647-656 (build_kv_cache_tensor) sets impl.nvfp4_k_scale/v_scale from the builder, but the impl's __init__ never declares them; declare them there, next to self.kv_layout = Nvfp4KvLayout().
  4. Fix the names, messages and docs that still say MXFP4 (valar-cop asked, not addressed): rename _mxfp4_kv_refusal (models/minimax_m3/config.py:10); _check_fp4_kv docstring (config.py:1987); the user-visible "MXFP4 index decode: ..." error and comments that now fire for NVFP4 (aiter_attention.py:1125-1133, 467); the index_scorer docstrings in sparse_attn.py:103,121,172; the module docstring of index_topk.py:11-18. docs/configuration_guide.md: the index_cache_dtype row still says "anything else is refused" on M3, and the minimum-aiter list names #7813 (still open) and #7825 (closed). Cite what landed instead (#7814 dad1b66282, #7824 47d23011bd, #7826 8f574bdf51).
  5. Delete duplication. nvfp4_pages.PAGE_TOKENS = 128 duplicates geometry.SPARSE_BLOCK_SIZE (valar-cop); sparse_pa_nvfp4_decode/prefill in sparse_attn_nvfp4.py are 1:1 pass-throughs of the aiter ops, so call aiter directly and keep only prefill_row_positions; the (selected, None, None) if nvfp4 else selected unpack is duplicated in _sparse_prefill/_sparse_decode and should be one helper (or always return the 3-tuple).
  6. Refresh the PR description. "Draft until the aiter PRs land" and the unchecked aiter box are stale; also list what is untested (DSpark/EAGLE own-pool draft with nvfp4, prefix-cache hits).

Reviewed at 2421500. aiter deps checked: all 4 FlyDSL symbols + the fused writer's nvfp4 mode are on aiter main 0b0381a1a8 with matching signatures. Light CI green; heavy CI not run yet. Full findings list kept with the reviewer.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants