kernel: validate learned semaphore maps and post guest GPU events

- admit {GPU VA -> PA} pairs only when the preceding bytes form a valid
  client memory descriptor header (type 0/1, CPU+GPU mapped bits), prefer
  validated pairs over headerless candidates under first-wins dedup, and
  invalidate the translation cache on every rebuild; fixes stale identity
  entries shadowing live pool pages and lets the guest UVM channel
  bring-up complete on its own
- capture OS event allocations (class 0x79) and post event records
  (function 0x1003) into the guest status queue when a channel's busy
  notification is consumed, with a broadcast fallback, serialized with
  replies via a queue lock
- add BAR0 shadow introspection (dump_bar0/scan_bar0) and guest-side
  diagnostic scripts for the VM cycle

1 files changed, 3 insertions, 3 deletions

diff --git a/scripts/create_mdev.sh b/scripts/create_mdev.sh
index a6ea91e..4f1cc1c 100755
--- a/scripts/create_mdev.sh
+++ b/scripts/create_mdev.sh
@@ -5,10 +5,10 @@
 set -euo pipefail
 
 PCI_ADDR="${1:-}"
-VGPU_TYPE="${2:-pci-nvidia-vgpu-turing}"
+VGPU_TYPE="${2:-nvidia-nvidia-vgpu-turing}"
 
-# Supported vGPU types (matching the mdev type names)
-VALID_TYPES=("pci-nvidia-vgpu-ampere" "pci-nvidia-vgpu-hopper" "pci-nvidia-vgpu-turing" "pci-nvidia-vgpu-pascal")
+# Supported vGPU types (matching the mdev type names the driver registers)
+VALID_TYPES=("nvidia-nvidia-vgpu-ampere" "nvidia-nvidia-vgpu-hopper" "nvidia-nvidia-vgpu-turing" "nvidia-nvidia-vgpu-pascal")
 
 # Find NVIDIA GPU if not specified
 if [[ -z "$PCI_ADDR" ]]; then