Skip to content

LordOS — a free, fast and secure GNU/Linux distribution based on Debian, with its own desktop environment: display server and Wayland compositor in C, a React UI engine on QuickJS, and private native apps.

The 0.8.2 Monitor scene repeated on a running desktop (VM 116, isolated headless session, 22 Terminal minimize/restore cycles over the moving map, menus, 54 full-repaint comparisons per run, before and after, one box and several rectangles): no stale area, the grey rectangle still not reproduced, several rectangles stay off; cost: layout+paint per frame unchanged, lord-server CPU within 2.5 points, LordBench desktop mode unchanged; CHANGELOG, MONITOR, LAYERS, PAPER-VALIDATION, paper (en/es/pt)

Federico Pereira541004c7fe9b
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6c1e008..61cbc46 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -257,8 +257,35 @@ of the matching [GitHub release](https://github.com/lordbasex/LordOS/releases),
then depended on where the cut fell. `paint_out` now applies the blur rule again to damage gathered over more than
one frame, on the tree as it is painted (`paint_damage_settle`).
Which of these was the 0.8.2 report is not proven (the reported scene itself never failed), so several rectangles
- stay **off by default** (`LORD_DAMAGE=multi`); before turning them on, the 0.8.2 integration test is to be repeated
- on a VM and the blur margin's cost measured. Built with `make -j8` (no warnings) and `make test` on VM 116.
+ stay **off by default** (`LORD_DAMAGE=multi`). Built with `make -j8` (no warnings), `make test` and
+ `build/engine-test` on VM 116.
+ **Checked on a running desktop** (VM 116, 2026-10-29; `docs/MONITOR.md`, "0.8.6: the search for the grey
+ rectangle"): an isolated headless session of the real default theme at 1280×800 (CPU composition), driven over its
+ remote desktop socket by a viewer that keeps its own copy of the screen from the tiles the session sends every
+ frame, so the copy holds exactly what the partial repaints left. The 0.8.2 scene: Monitor → Network with ten slow
+ downloads to mirrors in ten countries (the map's dots moving), a Terminal at 30 % transparency with blur over the
+ map, minimized into the Dock and restored 22 times, the Terminal's six menus opened and closed, 54 checks per run.
+ At each check a second connection makes the session repaint the whole screen and the two are compared (pixels that
+ changed within 2.2 s before or after are left out as animation; the downloads are paused for the check so the
+ dots are still). Six runs: this commit and the previous `main`, each with one box (the default) and with
+ `LORD_DAMAGE=multi`, and both builds again in the default mode with the scene still for 7 s before each check.
+ **No stale area in any run, before or after**: with the scene still, 0 differing pixels in 54 checks (this commit)
+ and 3 single pixels on the Dock's corner (the previous build); with the scene moving, only Monitor's live numbers
+ and charts that changed between the last partial frame and the full one, a few antialiased pixels of the Dock's
+ corner, whose blur takes in Monitor's live "Load" line (both builds, 1 to 7 pixels), and, in 5 of 108 checks of
+ this commit, a map arc that faded under the blurred Terminal in that gap (at most 3 levels of 255; gone with the
+ scene still). The 0.8.2 grey rectangle did not appear in either build, so it is still not reproduced. **Cost**
+ (the same sessions, one remote viewer, `lord-server` whole process, % of one core, 5 × 30 s, two batches): Monitor
+ → Network 56.7 / 57.2 % before, 58.7 / 59.7 % after (one box) and 59.2 / 58.1 → 57.2 / 58.0 % (several
+ rectangles); with the blurred Terminal over the map 35.8 / 36.4 → 38.0 / 37.7 % and 36.7 / 36.6 → 37.5 / 35.9 %;
+ layout and paint per frame (`LORD_FRAME_STATS`) the same within 0.1 ms in every case (14.6 to 14.8 ms). In these
+ sessions (`--profile`) every frame repainted one box of 64 to 79 % of the screen, also with `LORD_DAMAGE=multi`
+ (the moving dots, Monitor's live numbers and charts on both sides and its footer next to the blurred Dock never
+ stayed several rectangles), so the blur margin adds next to nothing there; the per-frame difference measured in
+ `test_multi_damage` (24,078 → 27,363 px) was not visible at the process level. Why the 0.8.2 VM 119 scene gave 5
+ to 11 rectangles and this one gives one is not investigated.
+ LordBench's desktop mode in the headless harness (three runs each): 58.9-60.0 fps, build p50 1.25-1.26 ms and p99
+ 5.9-7.6 ms before; 60.0 fps, p50 1.24-1.27 ms and p99 5.95-6.32 ms after.
- **The desktop under load** (found by LordBench's desktop test; `docs/BENCHMARK.md`, "The desktop under load"): the
compositor's cgroup (`server`) had the same `cpu.weight` as every app scope, and nice values don't count between
cgroups, so an app keeping every core busy took half of the compositor's CPU even at nice 19. On a Raspberry Pi 5
diff --git a/docs/LAYERS.md b/docs/LAYERS.md
index 80f0c9e..578ff10 100644
--- a/docs/LAYERS.md
+++ b/docs/LAYERS.md
@@ -116,7 +116,9 @@ blurred into the surface's edges, while a full repaint blurs only what lies belo
difference stayed until the next full repaint. `test_damage_fuzz` in `build/engine-test` found it
(random trees through lord-server's whole damage pipeline, every painted frame compared with a full
repaint). The cost: a repaint that touches the Dock or the menu bar also repaints the band of twice
-the blur radius around it (about 41 px for a 20 px blur; the screen edges cut it).
+the blur radius around it (about 41 px for a 20 px blur; the screen edges cut it). On VM 116 with
+Monitor's moving map, with and without a blurred Terminal over it, layout and paint per frame stayed
+the same within 0.1 ms (MONITOR.md, "The 0.8.2 integration test, repeated").
| | |
|---|---|
diff --git a/docs/MONITOR.md b/docs/MONITOR.md
index c635812..948807e 100644
--- a/docs/MONITOR.md
+++ b/docs/MONITOR.md
@@ -849,11 +849,55 @@ transform; a raster grid that moved with the subtree's extent; a raster changing
coming in) outside the damage; blurs taking in what is painted above them from outside the
repainted area; and several frames' rectangles cutting through a blurred node. Which one the 0.8.2
report was is not proven (the reported scene itself never failed), so `LORD_DAMAGE=multi` is still
-needed to turn the several-rectangles path on. Before turning it on by default: repeat the 0.8.2
-integration test on a VM with `LORD_DAMAGE=multi` (minimize a Terminal over the animating map at
-least 20 times, screenshots after each), and measure the cost of the blur fix (the margin of a
-blurred surface is now repainted with it: in `test_multi_damage` the pixels repainted per frame
-went from 24,078 to 27,363).
+needed to turn the several-rectangles path on. The margin of a blurred surface is now repainted
+with it: in `test_multi_damage` the pixels repainted per frame went from 24,078 to 27,363.
+
+**The 0.8.2 integration test, repeated (VM 116, 2026-10-29).** An isolated headless session of the
+real default theme (own home, runtime folder and bus; 1280×800, CPU composition), driven over its
+remote desktop socket. The viewer subscribes to the video with lossless tiles and keeps its own copy
+of the screen from them; for the test, the VM copy of `mirror.c` sends the changed tiles every frame
+instead of only once the screen is still (`LORD_TEST_TILES`, never committed), so the copy is exactly
+what the partial repaints left. At each check a second connection makes the session repaint the whole
+screen (`cb.refresh`), and the copy before is compared with the copy after; pixels that changed within
+2.2 s before or after are left out as animation. The scene: Monitor → Network with ten slow downloads
+to mirrors in ten countries, a Terminal at 30 % transparency with blur over the map, minimized into the
+Dock and restored 22 times while the dots move, then its six menus; the downloads pause for each check
+(2.5 s, or 7 s for the still runs) so the map stops. 54 checks per run, six runs:
+
+| Build | Mode | Scene at the check | Differing pixels outside Monitor's live numbers |
+|---|---|---|---|
+| previous `main` | one box | still 7 s | 3 (one pixel each, the Dock's corner) |
+| this commit | one box | still 7 s | **0** |
+| previous `main` | one box | still 2.5 s | 1 |
+| this commit | one box | still 2.5 s | 389 (an arc fading under the blurred Terminal, ≤ 3 levels; 2 pixels on the Dock's corner) |
+| previous `main` | several (`multi`) | still 2.5 s | 6 (the Dock's corner) |
+| this commit | several (`multi`) | still 2.5 s | 505 (the arc, ≤ 3 levels; 7 on the Dock's corner) |
+
+No stale area in any run. The Dock's corner pixels (1 to 7, antialiased) change with Monitor's
+"Load" line, which the Dock's blur takes in; the arc is a connection whose line faded between the
+last partial frame and the full one; both are the scene changing between the two copies, and neither
+appears when the scene is still for 7 s. The grey rectangle did not appear in either build, so the
+0.8.2 report is still not reproduced, and several rectangles stay off by default.
+
+Cost, the same sessions with one viewer (`lord-server` whole process, % of one core, 5 × 30 s, two
+batches, the build alternated):
+
+| Scene | Mode | Before | After |
+|---|---|---|---|
+| Monitor → Network, dots moving | one box | 56.7 / 57.2 % | 58.7 / 59.7 % |
+| Monitor → Network, dots moving | several | 59.2 / 58.1 % | 57.2 / 58.0 % |
+| the same with the blurred Terminal over the map | one box | 35.8 / 36.4 % | 38.0 / 37.7 % |
+| the same with the blurred Terminal over the map | several | 36.7 / 36.6 % | 37.5 / 35.9 % |
+
+Layout and paint per frame (`LORD_FRAME_STATS`) were the same within 0.1 ms in every row (14.6 to
+14.8 ms). `--profile` shows why: in these sessions every frame repainted one box of 64 to 79 % of the
+screen, also with `LORD_DAMAGE=multi` (the dots, the live numbers and charts on both sides of the
+window and its footer next to the blurred Dock never stayed several rectangles), so the blur margin
+adds next to nothing; why VM 119's scene gave 5 to 11 rectangles and this one gives one was not
+looked into. Most of the process's CPU here is the remote desktop's frame comparison and the session
+itself, not paint. LordBench's desktop mode in the headless harness (`perf-theme.js`, three runs each):
+58.9–60.0 fps, build p50 1.25–1.26 ms, p99 5.9–7.6 ms before; 60.0 fps, p50 1.24–1.27 ms, p99
+5.95–6.32 ms after.
VM 119: 4 vCPU Xeon Silver 4108 at 1.8 GHz, `virtio` GPU without 3D, CPU compositing, 1280×800.
Network tab in front, ten slow downloads to mirrors in ten countries (arcs to 8–11 countries, about
diff --git a/docs/PAPER-VALIDATION.md b/docs/PAPER-VALIDATION.md
index a95339f..6ef3797 100644
--- a/docs/PAPER-VALIDATION.md
+++ b/docs/PAPER-VALIDATION.md
@@ -118,6 +118,7 @@ memory from a process's RSS (§7.1) and points to versioned results (§11.5, `do
| 10.2 | Untested on real hardware: AMD/Nvidia GPUs, monitors on PCs (one 1920×1080 monitor tested on the Pi 5), laptops, Bluetooth, printers, scanners, FIDO2, mail providers | System Information reports (0.8.7: community reports) | **Untested** |
| 10.2 | Secure Boot must be off | System Information shows Secure Boot | Holds |
| 11.2 | `make test` 23 passed, 1 skipped | Today (2026-10-27, VM 116) 23 passed, 0 failed, 12 skipped (the node suites; they pass on a machine with node) | Holds |
+| 11.2 | (0.8.6, in development) every frame compared with a full repaint; seven partial-repaint errors fixed; the 0.8.2 scene on a running desktop found no stale area | `build/engine-test` (`test_damage_pipeline`, `test_damage_fuzz`); VM 116 isolated headless session, six runs of 54 checks (`docs/MONITOR.md`, "The 0.8.2 integration test, repeated") | Holds (2026-10-29) |
| 11.2 | 30,953 parser checks with mutation fuzzing | Tests (`remote-fuzz`, mail, doc, imager fuzzers) | Holds; count to refresh |
| 11.4 | Port 5673 closed until enabled | Check `remoteClosed` (effective) | Holds |
diff --git a/server/src/ui/paint.c b/server/src/ui/paint.c
index c54470b..c33c938 100644
--- a/server/src/ui/paint.c
+++ b/server/src/ui/paint.c
@@ -855,7 +855,9 @@ static void blur_walk(Node *n, int *grew) {
* blurs taking in what is painted above them from outside the repainted area, and several frames'
* rectangles cutting through a blurred node (paint_damage_settle). Each one could leave a stale
* area until something repainted it, with one box or several; which one the 0.8.2 report was is
- * not proven, so the switch stays off until it is tried again on the VM. */
+ * not proven. The 0.8.2 scene was tried again on VM 116 (docs/MONITOR.md) with both modes and no
+ * stale area appeared, before or after these fixes: the report is still not reproduced, so the
+ * switch stays off. */
int paint_multi_damage = 0;
static Box drects[DAMAGE_MAX];
static int ndrects;
diff --git a/web/content/paper/paper.en.md b/web/content/paper/paper.en.md
index 8d6ed3e..c457d77 100644
--- a/web/content/paper/paper.en.md
+++ b/web/content/paper/paper.en.md
@@ -585,7 +585,7 @@ make -C server test-full # adds the helpers, the sandbox and the HarfBuzz comp
- **Terminal and text:** the VT emulator, OpenType shaping against HarfBuzz, Unicode bidi against the official test file, input methods, Thai word breaking against libthai.
- **Security:** the sandbox (app, network, helper), signing (NIST and RFC 8032 vectors), package state, archive path rules against malicious archives, `lord.fs` protections, the firewall rules (with 20,000 JSON mutations and 30,000 random operations against a linear reference), the DNS reader (200,000 random packets), the kernel-map compiler (480,000 connections), and the kernel integration test (18 of 18 on a VM).
- **Parsers:** mail, MIME, HTML sanitizer, iCalendar, vCard and XML (30,953 checks with mutation fuzzing), the vault format, importers, QR, CBOR, D-Bus marshalling, the country database reader, journal, TZif and IPP.
-- **Engine and hardware:** layout and painting, NEON vs. C, the touchpad driver with recorded gestures, fans on fake sysfs trees, display arrangement, sound, and the WebAssembly specification suite.
+- **Engine and hardware:** layout and painting, NEON vs. C, the touchpad driver with recorded gestures, fans on fake sysfs trees, display arrangement, sound, and the WebAssembly specification suite. Since 0.8.6 (in development; the latest release is 0.8.5 Beta), every frame of random scenes, and of a Monitor scene with a translucent, blurred Terminal minimized into the Dock, is compared with a full repaint through a copy of the compositor's damage and double-buffer bookkeeping; that comparison found and fixed seven partial-repaint errors, and the same check on a running desktop on a virtual machine (the 0.8.2 scene, 22 minimize and restore cycles, 54 comparisons per run, before and after, with one damaged box and with several) found no stale area.
- **System Information (0.8.5):** the collector over five fake machines (an AMD desktop, an Intel laptop, an Nvidia desktop, a Raspberry Pi 5, a virtual machine) filled with serial numbers, UUIDs, MAC addresses and names that must not reach a report (none does, 460 checks), the SMBIOS memory reader against crafted and malformed tables, and the Checks and Sandbox Probe evaluation (56 checks).
On the final ISO build: `make -C server test` reported 23 passed, 0 failed, 1 skipped (bidi data not installed).
diff --git a/web/content/paper/paper.es.md b/web/content/paper/paper.es.md
index db6447f..f9a67a0 100644
--- a/web/content/paper/paper.es.md
+++ b/web/content/paper/paper.es.md
@@ -585,7 +585,7 @@ make -C server test-full # adds the helpers, the sandbox and the HarfBuzz comp
- **Terminal y texto:** el emulador VT, el shaping OpenType contra HarfBuzz, el bidi Unicode contra el archivo de prueba oficial, los métodos de entrada y la separación de palabras en tailandés contra libthai.
- **Seguridad:** la caja de arena (app, red, ayudante), la firma (vectores del NIST y del RFC 8032), el estado de los paquetes, las reglas de rutas de archivos comprimidos contra archivos maliciosos, las protecciones de `lord.fs`, las reglas del firewall (con 20.000 mutaciones de JSON y 30.000 operaciones aleatorias contra una referencia lineal), el lector DNS (200.000 paquetes aleatorios), el compilador de mapas del kernel (480.000 conexiones) y la prueba de integración con el kernel (18 de 18 en una VM).
- **Lectores:** correo, MIME, sanitizador de HTML, iCalendar, vCard y XML (30.953 verificaciones con fuzzing por mutación), el formato de la bóveda, los importadores, QR, CBOR, la serialización de D-Bus, el lector de la base de países, el journal, TZif e IPP.
-- **Motor y hardware:** layout y pintado, NEON vs. C, el driver de touchpad con gestos grabados, ventiladores sobre árboles sysfs falsos, la disposición de pantallas, el sonido y la suite de la especificación de WebAssembly.
+- **Motor y hardware:** layout y pintado, NEON vs. C, el driver de touchpad con gestos grabados, ventiladores sobre árboles sysfs falsos, la disposición de pantallas, el sonido y la suite de la especificación de WebAssembly. Desde la 0.8.6 (en desarrollo; la última versión publicada es la 0.8.5 Beta), cada cuadro de escenas aleatorias, y de una escena de Monitor con una Terminal translúcida y desenfocada minimizada al Dock, se compara con un repintado completo a través de una copia de la contabilidad de daño y doble búfer del compositor; esa comparación encontró y corrigió siete errores de repintado parcial, y la misma prueba sobre un escritorio en marcha en una máquina virtual (la escena de la 0.8.2, 22 ciclos de minimizar y restaurar, 54 comparaciones por corrida, antes y después, con un solo rectángulo dañado y con varios) no encontró ninguna zona desactualizada.
- **Información del sistema (0.8.5):** el recolector sobre cinco máquinas falsas (un escritorio AMD, una notebook Intel, un escritorio Nvidia, una Raspberry Pi 5, una máquina virtual) llenas de números de serie, UUID, direcciones MAC y nombres que no deben llegar a un informe (no llega ninguno, 460 verificaciones), el lector de memoria SMBIOS contra tablas armadas a propósito y malformadas, y la evaluación de las Verificaciones y de la Prueba del aislamiento (56 verificaciones).
En la compilación de la ISO final: `make -C server test` informó 23 pasaron, 0 fallaron, 1 omitida (faltaban los datos de bidi).
diff --git a/web/content/paper/paper.pt.md b/web/content/paper/paper.pt.md
index 1221079..15f3a25 100644
--- a/web/content/paper/paper.pt.md
+++ b/web/content/paper/paper.pt.md
@@ -585,7 +585,7 @@ O `make test` falha se alguma suíte falhar; uma suíte cuja ferramenta ou cujos
- **Terminal e texto:** o emulador VT, o shaping OpenType comparado ao HarfBuzz, o bidi Unicode comparado ao arquivo de teste oficial, os métodos de entrada, a separação de palavras em tailandês comparada à libthai.
- **Segurança:** o sandbox (app, rede, ajudante), a assinatura (vetores do NIST e da RFC 8032), o estado dos pacotes, as regras de caminhos de extração contra arquivos compactados maliciosos, as proteções do `lord.fs`, as regras do firewall (com 20.000 mutações de JSON e 30.000 operações aleatórias contra uma referência linear), o leitor de DNS (200.000 pacotes aleatórios), o compilador dos mapas do kernel (480.000 conexões) e o teste de integração com o kernel (18 de 18 numa VM).
- **Leitores:** e-mail, MIME, sanitizador de HTML, iCalendar, vCard e XML (30.953 verificações com fuzzing por mutação), o formato do cofre, os importadores, QR, CBOR, serialização D-Bus, o leitor da base de países, journal, TZif e IPP.
-- **Motor e hardware:** layout e pintura, NEON vs. C, o driver de touchpad com gestos gravados, as ventoinhas em árvores sysfs simuladas, o arranjo de telas, o som e a suíte de especificação do WebAssembly.
+- **Motor e hardware:** layout e pintura, NEON vs. C, o driver de touchpad com gestos gravados, as ventoinhas em árvores sysfs simuladas, o arranjo de telas, o som e a suíte de especificação do WebAssembly. Desde a 0.8.6 (em desenvolvimento; a última versão publicada é a 0.8.5 Beta), cada quadro de cenas aleatórias, e de uma cena do Monitor com um Terminal translúcido e desfocado minimizado para o Dock, é comparado com uma repintura completa por meio de uma cópia da contabilidade de dano e buffer duplo do compositor; essa comparação encontrou e corrigiu sete erros de repintura parcial, e o mesmo teste em um desktop em execução numa máquina virtual (a cena da 0.8.2, 22 ciclos de minimizar e restaurar, 54 comparações por execução, antes e depois, com um único retângulo danificado e com vários) não encontrou nenhuma área desatualizada.
- **Informações do sistema (0.8.5):** o coletor sobre cinco máquinas falsas (um desktop AMD, um notebook Intel, um desktop Nvidia, um Raspberry Pi 5, uma máquina virtual) cheias de números de série, UUIDs, endereços MAC e nomes que não podem chegar a um relatório (nenhum chega, 460 verificações), o leitor de memória SMBIOS contra tabelas forjadas e malformadas, e a avaliação das Verificações e do Teste do isolamento (56 verificações).
Na geração da ISO final, o `make -C server test` informou 23 aprovados, 0 falhas, 1 pulado (dados de bidi não instalados).