From 872846bcdcb0010eb60882fcb8057970d8dda8d6 Mon Sep 17 00:00:00 2001 From: David Galiffi Date: Tue, 26 Aug 2025 22:03:51 -0400 Subject: [PATCH] Updated Group By HIP Stream documentation (#717) Based on feedback from https://github.com/ROCm/rocprofiler-systems/pull/306 --- projects/rocprofiler-systems/CHANGELOG.md | 18 ++++++++++++++---- .../how-to/configuring-runtime-options.rst | 8 ++++---- 2 files changed, 18 insertions(+), 8 deletions(-) diff --git a/projects/rocprofiler-systems/CHANGELOG.md b/projects/rocprofiler-systems/CHANGELOG.md index dea0591f5c..3cd0470434 100644 --- a/projects/rocprofiler-systems/CHANGELOG.md +++ b/projects/rocprofiler-systems/CHANGELOG.md @@ -1,7 +1,20 @@ + + # Changelog for ROCm Systems Profiler Full documentation for ROCm Systems Profiler is available at [https://rocm.docs.amd.com/projects/rocprofiler-systems/en/latest/](https://rocm.docs.amd.com/projects/rocprofiler-systems/en/latest/). +## ROCm Systems Profiler 1.2.0 for ROCm 7.1 + +### Added + +- ``ROCPROFSYS_ROCM_GROUP_BY_QUEUE`` configuration setting to allow grouping of events by hardware queue, instead of the default grouping. +- Support for rocPD database output with the `ROCPROFSYS_USE_ROCPD` configuration setting. + +## Changed + +- Updated the grouping of "kernel dispatch" and "memory copy" events in Perfetto traces. They are now grouped together by HIP Stream rather than separately and by hardware queue. + ## ROCm Systems Profiler 1.1.0 for ROCm 7.0 ### Added @@ -10,9 +23,6 @@ Full documentation for ROCm Systems Profiler is available at [https://rocm.docs. - How-to document for VCN and JPEG activity sampling and tracing. - Support for tracing Fortran applications. - Support for tracing MPI API in Fortran. -- Initial support for rocPD database output with the `ROCPROFSYS_USE_ROCPD` configuration setting. -- By default, group "kernel dispatch" and "memory copy" events by HIP stream ID in Perfetto traces. - - Add the "ROCPROFSYS_ROCM_GROUP_BY_QUEUE" configuration setting to group events by queue, instead. ### Changed @@ -111,4 +121,4 @@ Full documentation for ROCm Systems Profiler is available at [https://rocm.docs. - Perfetto can no longer open Omnitrace proto files. Loading the Perfetto trace output `.proto` file in `ui.perfetto.dev` can result in a dialog with the message, "Oops, something went wrong! Please file a bug." The information in the dialog will refer to an "Unknown field type." The workaround is to open the files with the previous version of the Perfetto UI found - at https://ui.perfetto.dev/v46.0-35b3d9845/#!/. + at . diff --git a/projects/rocprofiler-systems/docs/how-to/configuring-runtime-options.rst b/projects/rocprofiler-systems/docs/how-to/configuring-runtime-options.rst index 7d5420f06f..da35a9cd4b 100644 --- a/projects/rocprofiler-systems/docs/how-to/configuring-runtime-options.rst +++ b/projects/rocprofiler-systems/docs/how-to/configuring-runtime-options.rst @@ -220,10 +220,10 @@ The following example: ROCPROFSYS_ROCM_GROUP_BY_QUEUE ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -By default, Perfetto trace will show the HIP streams to which kernel -and memory copy operations submitted. With the -``ROCPROFSYS_ROCM_GROUP_BY_QUEUE=ON`` setting, the trace will display HSA queues -to which these kernel and memory operations were submitted. +By default, the Perfetto trace groups the kernel dispatch and memory copy +operations to tracks corresponding to their HIP Stream ID. However, with the +``ROCPROFSYS_ROCM_GROUP_BY_QUEUE=ON`` setting, the events are on separate tracks +and grouped by hardware queue. ROCPROFSYS_USE_RCCLP ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^