mirror of https://github.com/wolfpld/tracy.git synced 2026-08-08 06:19:20 +00:00

Files

Basil Milanich d9a1cc06c1 Fix repro build: add -arch=native to use correct GPU architecture

NVCC 13.1 defaults to a PTX version incompatible with the installed
driver (580.105.08), causing kernels to silently fail with "provided
PTX was compiled with an unsupported toolchain". Use -arch=native so
NVCC auto-detects the target GPU (H100, sm_90) at build time.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-06 13:33:05 -05:00

Makefile

Fix repro build: add -arch=native to use correct GPU architecture

2026-04-06 13:33:05 -05:00

README.md

Add CUDA Graph GPU zone repro

2026-03-24 09:37:50 -05:00

repro.cu

Expand repro: multiple graphs, multiple kernels, interleaved launches

2026-04-06 11:54:08 -05:00

README.md

Tracy CUDA Graph GPU Zone Repro

Demonstrates that unpatched Tracy fails to show GPU zones for kernels launched via CUDA Graphs (cudaGraphLaunch).

Root cause

When kernels are launched through CUDA Graphs, CUPTI delivers CONCURRENT_KERNEL and MEMCPY activity records but no corresponding API callback fires for the individual kernel launches. Tracy's matchActivityToAPICall() always fails, and matchError() silently drops every GPU zone.

Build and run

make
./repro

What to expect

Tracy version	GPU zones shown
Unpatched	0
Patched (cuda-graph-gpu-zones.patch)	~30 (10 launches x 3 ops)

The graph structure

Each graph launch contains:

vector_add kernel (c = a + b)
Device-to-device memcpy
vector_add kernel (c = a + c)

The graph is launched 10 times, so 30 GPU operations total.