perf(cpu): bring up, benchmark, and profile Qwen3.5-2B on Raspberry Pi 5 - #65
Open
richiejp wants to merge 4 commits into
Open
perf(cpu): bring up, benchmark, and profile Qwen3.5-2B on Raspberry Pi 5#65richiejp wants to merge 4 commits into
richiejp wants to merge 4 commits into
Conversation
FOLLOWING_AGENTS_PROTOCOL Assisted-by: Codex:gpt-5 [Codex]
Pin the Q8_K_XL vehicle, observed Pi hardware and PMU surface, recursive optimization method, correctness gates, and same-file llama.cpp floor before implementation. Record Pi evidence as pending without changing the binding 20-core Arm result. FOLLOWING_AGENTS_PROTOCOL Assisted-by: Codex:gpt-5 [Codex]
Add a review-capped vt-op harness with deterministic quant GEMM fixtures, calibrated timing, cache-pressure profiles, JSON output, affinity and system metadata, plus grouped generic and Cortex-A76 perf_event_open counters. Keep unsupported and multiplexed events explicit and preserve the legacy quant benchmark. CPU-gated with GCC warning-as-error, clang-format, CTest CLI/schema checks, deterministic checksums, and direct 1-thread and 4-thread PMU runs. Pi execution and every performance claim remain pending. FOLLOWING_AGENTS_PROTOCOL Assisted-by: Codex:gpt-5 [Codex]
Build ARM64 artifacts under buildx/QEMU, verify the exported binaries and exact Qwen3.5-2B tokens on Cortex-A76, and record the PMU-ranked portable baseline that selects Q8 SDOT as the first assembly candidate. FOLLOWING_AGENTS_PROTOCOL Assisted-by: Codex:gpt-5 [Codex]
richiejp
marked this pull request as ready for review
August 6, 2026 20:45
richiejp
changed the title
5 tasks
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.
Row
BACKEND-CPU— one row per PR.What changed
Claims the Raspberry Pi 5 Cortex-A76 lane. The draft will add the spike, PMU-backed recursive CPU kernel harness, runtime-gated Cortex-A76 implementations, and measured Qwen3.5-2B correctness/performance evidence.
Evidence
scripts/agent-preflight.shpasses (upstream baseline currently has six unrelated inherited failures, recorded on Ordino taskt-83389f8c)docs/STATUS.md,docs/BENCHMARKS.md, anddocs/FEATURES.mdif a feature/model/backend/quant surface moved)Speed claims
Honest gaps
The branch currently contains only the protocol claim commit. No implementation, Pi model run, correctness result, or performance result is claimed yet. vLLM will not run on the Pi; x86_64 supplies the vLLM/current-engine correctness oracle and same-file llama.cpp supplies the Pi performance floor.