diff options
| author | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-08 04:15:24 -0400 |
|---|---|---|
| committer | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-08 04:15:24 -0400 |
| commit | 5bb2d53ec14f0d802726bc879b994c0286f85415 (patch) | |
| tree | a6b75f0fd7ab8dc21420041d282ea61049797713 /tools/cuda-ab/p2/expected/p2_seq5.txt | |
| parent | 11820d73e35511a8dfe25c65ee2f81970f7ac757 (diff) | |
| download | open-iov-5bb2d53ec14f0d802726bc879b994c0286f85415.tar.gz open-iov-5bb2d53ec14f0d802726bc879b994c0286f85415.zip | |
vgsp+guest-diag: kcore-truth pin resolver (pool/pushbuffer/channel ring+userd) + GR ring userd-derived candidates
guest_pin_stream.py now resolves translations entirely from guest truth via
/proc/kcore + a manual page-table walk (no QMP races): the CE semaphore pool
and pushbuffer memory from the UVM rm_mem structures, and each channel's
GPFIFO ring + GPPUT from the tracking semaphore's embedding uvm_channel_t
(channel_info located by its gpFifoGpuVa value). This removes the watcher
pipeline's flakiness on the cudaMalloc stall class (missing pool/pushbuffer
pins).
gsp_sema_service.c: svc_gr_ring_autopin tries the guest-registered USERD
memdesc base as a ring candidate (the fixed FB convention only matches when
the guest's VRAM allocator happens to reproduce the old layout), plus an
optional (gsp_sema_dbg) one-shot FB scan + raw-word diagnostics when every
candidate fails.
tools/cuda-ab/p2: self-checking arbitrary-kernel acceptance suite (10
programs: sequential launches, shapes, params, smem/GEMM, streams, malloc
churn, printf+error, FMA benchmark, mixed, managed probe) with host oracle
expectations and a guest runner.
Diffstat (limited to 'tools/cuda-ab/p2/expected/p2_seq5.txt')
| -rw-r--r-- | tools/cuda-ab/p2/expected/p2_seq5.txt | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/tools/cuda-ab/p2/expected/p2_seq5.txt b/tools/cuda-ab/p2/expected/p2_seq5.txt new file mode 100644 index 0000000..0823ab1 --- /dev/null +++ b/tools/cuda-ab/p2/expected/p2_seq5.txt @@ -0,0 +1,7 @@ +RESULT seq step=1 n=1024 bad=0 +RESULT seq step=2 n=1024 bad=0 +RESULT seq step=3 n=1024 bad=0 +RESULT seq step=4 n=1024 bad=0 +RESULT seq step=5 n=1024 bad=0 +RESULT seq launches=5 n=1024 bad=0 +STATUS OK |