From 3a0a1eba404919951ca3333ae9492f7213e60f75 Mon Sep 17 00:00:00 2001 From: Arthur Hanson Rasmusson Date: Tue, 17 Nov 2026 01:19:29 -0500 Subject: docs(wiki): 2026-11-17 status refresh - real GPU-GL draw admitted/replayed; per-draw MME macro body is the gate - 11 wiki pages refreshed: the offscreen EGL/GBM shim gives the guest a real NVIDIA GL context without Xorg/DDX; the real GZDoom draw window (12068 B at push 0x402f9d8, op=4 CALL(0) 0x3800 imm 0xd48 plus slots 44/57/58) is recognised, admitted bind-free, chunked and replayed verbatim. - CALL(0) executes without trapping (eng89t resolve through client SA[0] then golden, not-resident calls elided); c1026/c1027 entered the captured macro and c1027 reached the first DS with real stage data. Trap 0x80001528 decoded as a sticky TRAPPED_ADDR latch, not a cause/PC. - Remaining gate recorded precisely: the client's per-draw MME macro body was never captured (the prelude carries only the 12 setup macros), with the synthetic golden body the surviving wedge; stores still wrote=0, no px. - STATE-CURRENT.md snapshot prepended; wiki published to open-iov.org (revs 24221-24231; 68 pages audited, 0 problems). --- STATE-CURRENT.md | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 63 insertions(+) (limited to 'STATE-CURRENT.md') diff --git a/STATE-CURRENT.md b/STATE-CURRENT.md index f18d3be..e1275c7 100644 --- a/STATE-CURRENT.md +++ b/STATE-CURRENT.md @@ -1,5 +1,68 @@ # Open-IOV Current State +> **Snapshot: 2026-11-17 ~01:20 (SESSION 157 continued; HEAD `1db3831e` == +> `vgpu/master`; the working tree carries only this docs refresh; supersedes +> the 2026-11-16 ~22:55 entry below).** +> The graphics front crossed the **real GPU-GL gate**: the guest now renders +> through the real NVIDIA driver without Xorg/DDX, and the host admits, +> chunks and replays the **real client draw**. The remaining wall is one +> missing input — the client's per-draw MME macro body. +> +> - **Real NVIDIA GL in the guest, no Xorg/DDX.** An offscreen SDL EGL/GBM +> shim (`GZ_EGL`, `tools/guest-diag/gz_es.c`) gives GZDoom a real NVIDIA +> context: `GL_RENDERER: NVIDIA GeForce RTX 2080 SUPER/PCIe/SSE2`, +> `GL_VERSION: 4.6.0 NVIDIA 570.144` (not llvmpipe). +> - **The real draw is decoded, admitted and replayed.** The GZDoom draw +> window is 12068 B at push `0x402f9d8` (3017 words) and has no +> `0x3910/0x3914`: its draw is an op=4 IMMD `CALL(0)` to method `0x3800` +> (`imm=0xd48=3400`) plus CALL slots 44/57/58 (`0x3960/0x39c8/0x39d0`) and +> 7x `0x1b00` inline uploads. It carries no colour-target bind, so +> RT-gated admission had refused it; recognition now covers the real +> marker, admission is bind-free, the CALL-bearing chunk is forced first, +> and every chunk replays verbatim (c1025: `GL-FLUSH admitted=1`, +> `DRAW-WRITE m=0x3800 op=4`, `3D chunk launch ... draw=1`, +> `MME-CALL-ARGS slot=0 data=0xd48`, `MME-EXEC-CALL slot=0`). +> - **`CALL(0)` executes without wedging.** eng89t resolves the immediate as +> an IR address through client SA[0] (then golden) and elides a +> not-resident op=4 CALL(0) to a NOP (`MME-CALL-UNMAPPED +> reason=not-resident`). c1026/c1027 entered the captured macro +> (`macro-status=00000004`, `mpc` 0->4) with no trap; c1027 reached the +> first DS with real stage data (`DS-FAST2 ds_after=83044206 +> stage=015e0900,015e0a00,015c0400`). Trap `0x80001528` is decoded as a +> sticky `NV_PGRAPH_TRAPPED_ADDR` latch (bit31 INVALID, subch 0, method +> `0x1528` = `SET_SHADER_EXCEPTIONS`), not a cause/PC. +> - **Remaining gap (precise): the per-draw MME macro body was never +> captured.** The 59580-byte prelude carries only the 12 setup macros +> (573 code words, pc 0..0x98); the draw push has no `0x114/0x11c`; and +> `0xd48` is a macro length/argument, not an IR entry. The surviving wedge +> is the synthetic golden body (`badf1201 pc=513 mac=0xf1`) on a +> non-CALL(0) launch. Next: capture the per-draw `0x114` upload (every +> beef push plus the X/2D channel) or implement/verify the MME64 body, +> install it before the draw launch, then check `DS-FAST2 ds_after=8[01]`, +> `GUEST-WIT va=0x15d`, `wrote>0`, `px0!=0`. No vGPU surface store or +> pixel exists yet (`wrote=0`, no `px_nonblack`). +> - **GZDoom software-GL video unchanged and verified** (Mesa llvmpipe; four +> files, 22 s flight + 18 s orbit, 640x480/1280x960, 30 fps) at +> `https://files.vgpu.io/files/open-iov-gzdoom-20261116` — a +> guaranteed-quality deliverable, not a vGPU rendering claim. +> - Commits since the last snapshot: `9386fe4b` (`phys_gr_dump_push`), +> `10e03cd7` (eng89s: real `CALL(0)` `0x3800` draw + GZ_EGL harness), +> `1db3831e` (eng89t: CALL(0) resolve + no-wedge fallback); HEAD +> `1db3831e` == `vgpu/master`, working tree clean before this refresh. +> +> CUDA front unchanged (backgrounded while graphics holds the device; its +> 10-30 status stands): `41343c63` green (`victory_mail` `VADD bad=0` + +> `STATUS OK`; `vecadd_loop` 100 launches PASS); the fresh-boot battery's red +> class is D2H-stage with release-discipline fixes staged; the SESSION 143 +> Vulkan fence regression is not re-measured. +> +> Live at check (read-only): host uptime 16 h 03 m (boot 2026-11-16 ~09:13); +> the campaign finished its last cycle and the module was unloaded cleanly at +> 00:09:59 (`nv_vgpu_exit` completed; no mdev, no QEMU; host idle); `nvidia` +> stays loaded for the host's own GPU; pstore empty; no Xid/Oops; no reboot +> initiated during the check. The open-iov.org wiki was refreshed alongside +> this snapshot (11 pages, revs 24221-24231; 68 pages audited, 0 problems). +> > **Snapshot: 2026-11-16 ~22:55 (SESSION 157 continued; HEAD `a06d4b13` == > `vgpu/master`; the working tree carries only this docs refresh; supersedes > the 2026-11-16 ~00:35 entry below).** -- cgit v1.2.3