diff options
Diffstat (limited to 'docs/03-architecture/04-uvm.md')
| -rw-r--r-- | docs/03-architecture/04-uvm.md | 15 |
1 files changed, 13 insertions, 2 deletions
diff --git a/docs/03-architecture/04-uvm.md b/docs/03-architecture/04-uvm.md index 299f46b..05ec80b 100644 --- a/docs/03-architecture/04-uvm.md +++ b/docs/03-architecture/04-uvm.md @@ -50,13 +50,24 @@ Each stage has a distinct failure signature; our debugging docs map them. | Requirement | Detail | Our status | |---|---|---| | GPU identity | must match the uuid from RM | โ
| -| CE capabilities | per-engine support bits; UVM *requires* `supported && !grce && sysmem` for CPU_TO_GPU/GPU_TO_CPU channels | โ
(we answer CE_GET_PHYSICAL_CAPS with the correct bits) | +| CE capabilities | per-engine support bits; UVM *requires* `supported && !grce && sysmem` for CPU_TO_GPU/GPU_TO_CPU channels | ๐ก the caps we answer are accepted, but the engine list / CE capabilities are missing the async copy engine (`COPY14`) โ the current context-creation blocker | | Channel creation | RM channel objects with GPFIFO/userd/instance parameters | โ
objects created | -| Pushes executed | UVM's first action on a new channel is an *init push* followed by a semaphore-tracked wait | ๐ก we can complete them (see below) | +| Pushes executed | UVM's first action on a new channel is an *init push* followed by a semaphore-tracked wait | ๐ก the host decodes method streams and executes memsets, copies and semaphore releases in software (a bridge); no kernel is forwarded to the physical GPU yet | | Page trees | UVM initializes page trees with its own pushes; the *server-owned* ranges must be populated by us | ๐ก | | Fault delivery | fault buffer + notification; replay/cancel semantics | ๐ด | | Semantics of completions | tracker values must advance exactly | ๐ก | +**Current blocker (2026-09-23):** `cuCtxCreate` returns `rc=999` +(`CUDA_ERROR_UNKNOWN`) because the emulated engine list lacks the async copy +engine (`COPY14`); teardown of the failed context still oopses in the guest +kernel in `uvm_pmm_devmem_page_to_va_space` (twice per boot). The CUDA-side +contract and reply rules are in +[`../../userspace/cuda/docs/03-driver-and-gpu-interface.md` ยง9](../../userspace/cuda/docs/03-driver-and-gpu-interface.md) +and +[`../../userspace/cuda/docs/04-virtualization-and-emulation.md` ยง6](../../userspace/cuda/docs/04-virtualization-and-emulation.md); +start at +[`../../userspace/cuda/READ-ME-FIRST.md`](../../userspace/cuda/READ-ME-FIRST.md). + ## 4. The most important discovery: how UVM waits UVM's tracker (`uvm_tracker.c`) is the single most visible symptom of an |