http://dokuwiki.reauktion.de/ 2026-05-26T21:29:17+00:00 http://dokuwiki.reauktion.de/ http://dokuwiki.reauktion.de/lib/exe/fetch.php?media=wiki:dokuwiki.svg text/html 2026-05-21T13:59:21+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=fourier:brave_arm64_vaapi_wall&rev=1779371961&do=diff Brave 148 ARM64 + Mali-Bifrost / hantro VPU + libva-v4l2-request-fourier — definitive HW video decode wall Verdict: Brave-bin on PineTab2 (and structurally on any RK3566/RK3568 Mali-Bifrost + hantro stack) cannot engage hardware video decode. The wall is text/html 2026-05-04T07:24:51+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=fourier:attribution_2026-05-03&rev=1777879491&do=diff fourier_attribution — Phase 5 review (2026-05-03) Single-question matrix campaign. Operator: mfritsche. Reviewer: handover to fresh model instance per 8(+1)-phase loop Phase 5. Local artifacts: /home/mfritsche/src/fourier_attribution/ on noether. Predecessors (substrate, NOT data): text/html 2026-05-02T20:16:13+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:metrics&rev=1777752973&do=diff metrics.csv — locked Phase 1 cells Locked binding cells for the campaign. Raw CSV attachment: metrics.csv (inlined). Inherits the success-criteria shape from ohm_gl_fix Phase 1r (drops_post_warmup, drops sanity cap) plus a new cell kwin_wayland_cpu_steady_state that locks the intent that the fix must move work, not hide it. text/html 2026-05-02T20:16:13+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:evidence:metrics_csv&rev=1777752973&do=diff metrics.csv — raw Raw evidence file inlined. Source path on boltzmann: /home/mfritsche/src/kwin_overlay_subsurface/metrics.csv. # kwin_overlay_subsurface metrics — locked Phase 1 cells # # Inherits the success-criteria shape from ohm_gl_fix Phase 1r # (drops_post_warmup, drops sanity cap) plus a new cell # kwin_wayland_cpu_steady_state that locks the intent that the # fix must move work, not hide it. # # Schema: # phase phase0_handover | phase1_baseline_kwin # … text/html 2026-05-02T20:16:13+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:start&rev=1777752973&do=diff kwin_overlay_subsurface — KWin per-frame compositor cost on parent + wp_subsurface A scope-narrow campaign to identify and remove the per-frame cost in KWin's overlay-subsurface composite path on Mali-G52 / panfrost / KDE Plasma 6.6.4 Wayland (PineTab2, RK3568). Spun off from text/html 2026-05-02T20:16:12+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:worklist&rev=1777752972&do=diff Work items — kwin_overlay_subsurface Phase 1 — leading research question Status: LOCKED 2026-05-02. Answer in phase2_source_findings lead section. Outcome: NO on both promotion paths for windowed Brave on rockchip-drm — Phase 4 design space narrows to architect's hypothesis (a), dmabuf-to-GL-texture import-caching. text/html 2026-05-02T20:16:12+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:phase2_source_findings&rev=1777752972&do=diff Phase 2 — KWin source archaeology This file is the synthesised source-read for the kwin_overlay_subsurface campaign. It opens with the Phase 1 leading question answer (per worklist), follows with the architectural diagram of the per-frame path, and ends with file-level findings ordered by the priority list in text/html 2026-05-02T20:16:12+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:phase3_findings&rev=1777752972&do=diff Phase 3 — measurement findings Status: LOCKED 2026-05-02. Single-pass measurement (1 replicate per condition; pass-2 averaging deferred per §11). Verdict applies the §2 falsification table from phase3_protocol to N=1 with explicit IQR-deferred caveat. TL;DR — H1 rejected, H1' (working) emerges, drops are not CPU-caused text/html 2026-05-02T18:55:38+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=kwin_overlay_subsurface:phase3_protocol&rev=1777748138&do=diff Phase 3 — measurement protocol Status: LOCKED 2026-05-02. Pre-registered hypothesis and decision rules locked here before data collection. Conditions, falsification criteria, and the “what would change my mind” set are defined upfront so the analysis cannot quietly drift toward confirming the prior. Per architect's framing: this is text/html 2026-05-02T18:03:26+00:00 markus_fritsche (markus_fritsche@undisclosed.example.com) http://dokuwiki.reauktion.de/doku.php?id=start&rev=1777745006&do=diff Hallo, dass Sie hier gelandet sind heißt vermutlich, dass Sie sich verirrt haben. Wie wäre es mit http://www.google.de/? * Rock5 ITX+ * keebs * Claude-assisted dev process — 8-phase loop * RK3588 DDR Init Blob Reverse Engineering * RK3588 LPDDR5 overclock via ddrbin_tool.py * Bin — GenBook u-boot eDP (upstream) * MegabitChip * fourier — V4L2 hardware-decode unlock for ARM Wayland * Project Volta — Radxa Rock 5 ITX+ / boltzmann stack * Neutron — RK3588 mainline kernel campai…