vgsp: stop fabricating 0xFFFF into the nvPush error notifier; add r1780 GL ops tooling

The uncommitted keep-alive change wrote 0xFFFF into slot 0's status field
(u16 at notifier+0x0e) on the X DDX / GLX kick-idle channel families. Ground
truth (570.144 reference sources + stock nvidia_drv.so disassembly): slot 0
of an nvPush notifier block is the ERROR notifier; nvPushCheckChannelError
reads that status and 0xFFFF aborts every nvPushWaitForNotifier/idle wait as
a channel error. There is no pb_kickIfIdle in 570.144 - the DDX DMA-idle
test reads the progress-tracker semaphore word, not this slot. Leave slot 0
untouched; keep the 0x8000->0 completion ack for the remaining slots and the
r864 0xFFFF/0x8000 completion on the RC-watchdog/beef families. Correct the
CONT.408/409 comments accordingly.

New host/guest tooling: gpa.py (QMP gva2gpa), tick_gen.py (guest-RM timer
BO tick), guest_mode.sh (shared nvClip modeNum reader), gl_verify.sh
(X/GLX bring-up with ScreenInit-lottery retries), push_gl_tools.sh, the
gzdoom_pbuf_run.sh guest recipe and the test_tri SDL2 harness validation
program; ignore tools/guest-diag/build/.

1 files changed, 112 insertions, 0 deletions

diff --git a/tools/guest-diag/test_tri.c b/tools/guest-diag/test_tri.c
new file mode 100644
index 0000000..dc42876
--- /dev/null
+++ b/tools/guest-diag/test_tri.c
@@ -0,0 +1,112 @@
+// SPDX-License-Identifier: BSD-3-Clause
+// Copyright (c) 2026 Open-IOV contributors
+// test_tri.c - minimal SDL2 + OpenGL triangle for the offscreen pbuffer
+// harness.  Intended to be run with SDL_VIDEODRIVER=offscreen and the
+// gz_pbuf.so shim preloaded (plus sdl_cap.so to capture the swaps):
+//
+//   SDL_VIDEODRIVER=offscreen DISPLAY=:0 \
+//   LD_PRELOAD="/tmp/sdl_cap.so /tmp/gz_pbuf.so" GZCAP=/tmp/testframes.bin \
+//   ./test_tri 20 640 480
+//
+// It creates an SDL2 GL window/context (the shim turns the context into a
+// GLX pbuffer), clears to a colour, draws a rotating triangle, reads the
+// centre pixel back (non-black proof), and calls SDL_GL_SwapWindow each
+// frame.  The sdl_cap.so hook captures the swap; the gz_pbuf.so hook turns
+// the swap into glFinish.  Build:
+//   gcc -O2 -o test_tri test_tri.c $(pkg-config --cflags --libs sdl2) -lGL
+#include <SDL2/SDL.h>
+#include <GL/gl.h>
+#include <math.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+int main(int argc, char **argv)
+{
+    int frames = argc > 1 ? atoi(argv[1]) : 20;
+    int w = argc > 2 ? atoi(argv[2]) : 640;
+    int h = argc > 3 ? atoi(argv[3]) : 480;
+    SDL_Window *win;
+    SDL_GLContext ctx;
+    int i;
+
+    setvbuf(stdout, NULL, _IONBF, 0);
+
+    if (SDL_Init(SDL_INIT_VIDEO) != 0) {
+        fprintf(stderr, "test_tri: SDL_Init: %s\n", SDL_GetError());
+        return 1;
+    }
+    printf("test_tri: videodriver=%s\n",
+           SDL_GetCurrentVideoDriver() ? SDL_GetCurrentVideoDriver() : "?");
+
+    SDL_GL_SetAttribute(SDL_GL_RED_SIZE, 8);
+    SDL_GL_SetAttribute(SDL_GL_GREEN_SIZE, 8);
+    SDL_GL_SetAttribute(SDL_GL_BLUE_SIZE, 8);
+    SDL_GL_SetAttribute(SDL_GL_ALPHA_SIZE, 8);
+    SDL_GL_SetAttribute(SDL_GL_DOUBLEBUFFER, 1);
+
+    win = SDL_CreateWindow("test_tri", SDL_WINDOWPOS_UNDEFINED,
+                           SDL_WINDOWPOS_UNDEFINED, w, h,
+                           SDL_WINDOW_OPENGL | SDL_WINDOW_HIDDEN);
+    if (!win) {
+        fprintf(stderr, "test_tri: SDL_CreateWindow: %s\n", SDL_GetError());
+        SDL_Quit();
+        return 2;
+    }
+
+    ctx = SDL_GL_CreateContext(win);
+    if (!ctx) {
+        fprintf(stderr, "test_tri: SDL_GL_CreateContext: %s\n",
+                SDL_GetError());
+        SDL_DestroyWindow(win);
+        SDL_Quit();
+        return 3;
+    }
+    if (SDL_GL_MakeCurrent(win, ctx) != 0)
+        fprintf(stderr, "test_tri: SDL_GL_MakeCurrent: %s\n",
+                SDL_GetError());
+    SDL_GL_SetSwapInterval(0);
+
+    printf("test_tri: GL %s | %s | %s\n",
+           (const char *)glGetString(GL_VENDOR),
+           (const char *)glGetString(GL_RENDERER),
+           (const char *)glGetString(GL_VERSION));
+
+    for (i = 0; i < frames; i++) {
+        float a = (float)i * 0.10f;
+        unsigned char px[4] = { 0, 0, 0, 0 };
+
+        glViewport(0, 0, w, h);
+        glClearColor(0.05f, 0.05f, 0.10f, 1.0f);
+        glClear(GL_COLOR_BUFFER_BIT);
+
+        glMatrixMode(GL_PROJECTION);
+        glLoadIdentity();
+        glOrtho(-1.0, 1.0, -1.0, 1.0, -1.0, 1.0);
+        glMatrixMode(GL_MODELVIEW);
+        glLoadIdentity();
+        glRotatef(a * 57.29578f, 0.0f, 0.0f, 1.0f);
+
+        glBegin(GL_TRIANGLES);
+        glColor3f(1.0f, 0.2f, 0.1f);
+        glVertex2f(0.0f, 0.8f);
+        glColor3f(0.1f, 1.0f, 0.3f);
+        glVertex2f(-0.8f, -0.6f);
+        glColor3f(0.2f, 0.3f, 1.0f);
+        glVertex2f(0.8f, -0.6f);
+        glEnd();
+        glFinish();
+
+        glReadPixels(w / 2, h / 2, 1, 1, GL_RGBA, GL_UNSIGNED_BYTE, px);
+        printf("test_tri: frame=%d centre=%02x%02x%02x%02x err=0x%x\n",
+               i, px[0], px[1], px[2], px[3], glGetError());
+
+        SDL_GL_SwapWindow(win);
+        SDL_Delay(20);
+    }
+
+    printf("test_tri: done frames=%d %dx%d\n", frames, w, h);
+    SDL_GL_DeleteContext(ctx);
+    SDL_DestroyWindow(win);
+    SDL_Quit();
+    return 0;
+}