summaryrefslogtreecommitdiffstats
path: root/wiki/Execution-Plane.md
diff options
context:
space:
mode:
authorArthur Hanson Rasmusson <arthur@vgpu.io>2026-11-10 16:42:39 -0500
committerArthur Hanson Rasmusson <arthur@vgpu.io>2026-11-10 16:42:39 -0500
commit25e3e800df8e5caf18d7e846a0cd36df4f5834f9 (patch)
treeaa39f53f4b136bf422994af90680191fa0ec743e /wiki/Execution-Plane.md
parentdf6e71bb5188bdc762ca0fbc73352f9067572589 (diff)
downloadopen-iov-25e3e800df8e5caf18d7e846a0cd36df4f5834f9.tar.gz
open-iov-25e3e800df8e5caf18d7e846a0cd36df4f5834f9.zip
wiki: 2026-11-10 afternoon refresh - first real GPU readback; direct-GL/window walls
Refresh STATE-CURRENT, the Main Page and the status/frontier pages: the DDX 3D replay executes and a guest pbuffer glClear+glReadPixels returns a red pixel (ff0000ff); the RM write-lock holder (Xorg in the DDX scrub wait) is fixed in df6e71bb. The direct-GL client channel still emits only setup/notify kicks (ops_runs=0), a window draw wedges X, and X startup/RM stability plus the DDX 4 GB sysmem allocation remain. Published to open-iov.org.
Diffstat (limited to 'wiki/Execution-Plane.md')
-rw-r--r--wiki/Execution-Plane.md14
1 files changed, 8 insertions, 6 deletions
diff --git a/wiki/Execution-Plane.md b/wiki/Execution-Plane.md
index a48af30..03a739a 100644
--- a/wiki/Execution-Plane.md
+++ b/wiki/Execution-Plane.md
@@ -8,7 +8,7 @@
> when the two disagree, the docs tree wins. Front door:
> [`../docs/README.md`](../docs/README.md).
-Status: 2026-11-09, tree `master @ bb802909`. Tags: `[live]` proven with the
+Status: 2026-11-09, tree `master @ df6e71bb`. Tags: `[live]` proven with the
stock guest driver; `[gated]` implemented, default-off or parameter-gated;
`[wip]` partial; `[wall]` current blocker; `[target]` named success latch;
`[planned]` designed, not built.
@@ -149,11 +149,13 @@ guest: push + QMD + GPFIFO entry + doorbell (BAR0 0x88704)
buffers plus the notifier/`UPDATE` completion, class-aware arm), so NvKms
registers, X settles and GL 4.6 is live, the X DDX waits are fixed, and
the guest's real GL draws now forward and execute on the physical GPU
- (`backend_ok=5-7`, `backend_timeout=0`); X is reliable and the 2D path is
- green. No rendered (non-black) frame has been produced through the NVIDIA
- path yet: the beef GR ring resolution race, the fixed-CB image's zero MVP
- block and the render-target bind are the named blockers. The lanes and the
- ladder are on [`Graphics-Execution.md`](Graphics-Execution.md).
+ (`backend_ok=5-7`, `backend_timeout=0`); the DDX 3D replay executes, the 2D
+ path is green, and the first real GPU readback landed (pbuffer red pixel).
+ No non-black **screen** frame has been produced through the NVIDIA path
+ yet: the direct-GL client emits only setup/notify kicks and a window draw
+ wedges X, so the direct-GL draw capture and the window/screen present are
+ the named blockers. The lanes and the ladder are on
+ [`Graphics-Execution.md`](Graphics-Execution.md).
## 6. Portability