summaryrefslogtreecommitdiffstats
path: root/tools/cuda-ab/streams2.cu
diff options
context:
space:
mode:
authorArthur Hanson Rasmusson <arthur@vgpu.io>2026-10-25 10:08:53 -0400
committerArthur Hanson Rasmusson <arthur@vgpu.io>2026-10-25 10:08:53 -0400
commite4d8ef84f70c084243468a3cae26448d973c841d (patch)
tree1314b602862fa3489c4525cf25454fe13f26ceeb /tools/cuda-ab/streams2.cu
parent98a120b2da9361aa9331776351ef4694fc72ed98 (diff)
downloadopen-iov-e4d8ef84f70c084243468a3cae26448d973c841d.tar.gz
open-iov-e4d8ef84f70c084243468a3cae26448d973c841d.zip
wall-2 session-86: the DEVICE_LOCAL -2 wall CRACKED - the 512 MiB emulated VRAM was the gate (the ICD hooks captured ALLOC size=0x4000 typeIndex=1 valid; vk_memprops shows device-local heap[0]=512 MiB); the nvidia-nvidia-vgpu-turing-2g mdev type (2 GiB heap, VGPU_MDEV_TYPE env var honoured by start_vm_keep.sh) lets the same client with the same recipe proceed past vkAllocateMemory into a deeper wait (os_schedule, no -2); remaining: the later pushes at va=+0x52d4/+0x586c read pa=0x0 (window page list scattered; injections covered page 1 only) - extend the window injections; tools: vk_memprops.py, vk_alloc_test.py (hits -4 without the harness), t91.py
Diffstat (limited to 'tools/cuda-ab/streams2.cu')
0 files changed, 0 insertions, 0 deletions