derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
50 lines
1.4 KiB
Markdown
50 lines
1.4 KiB
Markdown
# Proxy-trace profiler plugin
|
|
|
|
This shared library implements the former built-in **ProxyTrace** feature as an
|
|
**NCCL/RCCL profiler v6 plugin** (`ncclProfiler_v6`; descriptor includes RCCL proxy-trace fields). RCCL records proxy diagnostics through the
|
|
profiler API (`ncclProfileProxyDiag`); this plugin stores the same maps and
|
|
dump format as the legacy implementation.
|
|
|
|
## Build
|
|
|
|
```bash
|
|
cd plugins/profiler/proxytrace
|
|
make
|
|
```
|
|
|
|
Requires `libfmt` development headers (`fmt/format.h`) and a C++17 compiler. The
|
|
plugin does not use `plugins/profiler/example/`; minimal NCCL plugin types live
|
|
in `proxytrace_plugin_shim.h` next to the source.
|
|
|
|
Output: `librccl-profiler-proxytrace.so`
|
|
|
|
## Usage
|
|
|
|
**Option A — explicit plugin path**
|
|
|
|
```bash
|
|
export NCCL_PROFILER_PLUGIN=/path/to/librccl-profiler-proxytrace.so
|
|
```
|
|
|
|
**Option B — RCCL auto-load (same directory as librccl)**
|
|
|
|
```bash
|
|
export RCCL_ENABLE_PROXY_TRACE=1
|
|
# RCCL tries to dlopen dirname(librccl)/librccl-profiler-proxytrace.so when
|
|
# NCCL_PROFILER_PLUGIN is unset.
|
|
```
|
|
|
|
**Option C — short name (requires `libnccl-profiler-proxytrace.so` on `LD_LIBRARY_PATH`)**
|
|
|
|
```bash
|
|
export NCCL_PROFILER_PLUGIN=proxytrace
|
|
```
|
|
|
|
Copy or symlink `librccl-profiler-proxytrace.so` to `libnccl-profiler-proxytrace.so`
|
|
if you rely on this naming convention.
|
|
|
|
## Dump
|
|
|
|
On communicator destroy, RCCL calls `ncclProfilerProxyTraceDump` if exported by
|
|
the loaded profiler library. `ncclCommDump` uses the same hook.
|