From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: drm/ci: i915: cml: update runner tag Date: Thu, 12 Feb 2026 15:48:41 +1000 Message-ID: In-Reply-To: <20260210071138.2256773-3-vignesh.raman@collabora.com> References: <20260210071138.2256773-1-vignesh.raman@collabora.com> <20260210071138.2256773-3-vignesh.raman@collabora.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-Mailer: Claude Code Patch Reviewer Patch Review **Commit:** `156dc8e49612d` Changes the CML (Comet Lake) i915 CI job from `asus-C436FA-Flip-hatch` to `acer-chromebox-cxi4-puff`, increasing parallel count from 2 to 5 to match the larger device pool. Also increases `i915:amly` parallel count from 2 to 3. ```yaml - parallel: 2 + parallel: 5 variables: - DEVICE_TYPE: asus-C436FA-Flip-hatch + DEVICE_TYPE: acer-chromebox-cxi4-puff GPU_VERSION: cml - RUNNER_TAG: mesa-ci-x86-64-lava-asus-C436FA-Flip-hatch + RUNNER_TAG: mesa-ci-x86-64-lava-acer-chromebox-cxi4-puff ``` **Analysis:** The commit message explains hatch has only 2 DUTs causing timeouts, while puff has 5 DUTs. Switching to puff and setting `parallel: 5` makes sense. The `GPU_VERSION: cml` is retained, confirming both devices have the same CML GPU. The amly parallel increase to 3 is mentioned in the commit message ("Also increase parallel count for amly jobs to 3"). **Note:** The i915-cml-flakes.txt still references "Board Name: asus-C436FA-Flip-hatch" for the `perf_pmu@most-busy-check-all` entry. This is slightly inconsistent since the actual CI will now run on puff, but flake entries are informational metadata and the test name is what matters for matching, not the board name comment. This is cosmetic and not a functional issue. **Verdict:** No issues. The device switch is well-justified. --- --- Generated by Claude Code Patch Reviewer