Added public API call to setup agent counter collection on a context.
Refactored the return types internally for dispatch counter collection
to use rocprofiler_status_t (allow for more verbose failures to be
surfaced via the API)
Subsequent commits will fill out the sampling functionality for agent
counter collection.
Co-authored-by: Benjamin Welton <ben@amd.com>
[ROCm/rocprofiler-sdk commit: cb3fc070c7]
* public codeobj info
* Restructure code object source code file layout
* Update get_unloaded_code_objects + add iterate_loaded_code_objects
* Remove get_unloaded_code_objects from visible internal API
- iterate_loaded_code_objects + functor which filters on the hsa_executable_t effectively reproduces this behavior
* Whitespace removal
---------
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
[ROCm/rocprofiler-sdk commit: 733aa8e438]
- install cmake config files to CMAKE_INSTALL_LIBDIR instead of CMAKE_INSTALL_DATAROOTDIR per ROCm standards
- Provide setup-env.sh and modulefiles for rocprofiler-sdk and rocprofiler-sdk-roctx
[ROCm/rocprofiler-sdk commit: 67dbe3f47f]
* Sync more often in counter collection samples
* Update samples/counter_collection/main.cpp
- support command line arguments
- number of iterations
- iterations per sync
- number of devices to use
[ROCm/rocprofiler-sdk commit: 8c985543d9]
- LICENSE needs to be installed in share/doc/rocprofiler-sdk not share/doc/rocprofiler
- CPACK_RPM_PACKAGE_LICENSE should be set to license type not the license location
[ROCm/rocprofiler-sdk commit: 5d365bb8b8]
* Remove -Wno-missing-field-initializers
- Compiler errors if missing field initializers
* Update lib/rocprofiler-sdk/counters/evaluate_ast.cpp
- copy over dispatch ID in perform_reduction/evaluate
[ROCm/rocprofiler-sdk commit: 48273d6a65]
* Add default values for kernel struct
* Update hsa-queue-dependency app
- default initializers
- check HSA_AMD_MEMORY_POOL_INFO_RUNTIME_ALLOC_ALLOWED for memory pools
- clang-tidy fixes (member -> static, etc.)
* Update run-ci.py
- add --progress --output-on-failure -V if no other options regarding verbosity are passed
- improve the ability to control the stages
---------
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
[ROCm/rocprofiler-sdk commit: 29bc84ec0c]
* Stabilizing the machines used for testing
* Update .github/workflows/continuous_integration.yml
* Update continuous_integration.yml
* Delete .github/workflows/ci_pc_sampling.yml
* Update continuous_integration.yml
- add mi300
- use if conditions for whether to run "extended" core tests
* Consistency in matrix for each job
* Update continuous_integration.yml
- include runner in core CDash name
* Update continuous_integration.yml
- remove mi300
* Update continuous_integration.yml
- add mi300
* Update CI workflow
- tweak "Install requirements" step
* Update CI workflow
- timeout on install requirements
* Update CI workflow
- revert sanitizers to gcc-12
* Update CI workflow
- remove core installation of clang-tidy (handled by python pip)
* Update CI workflow
- disable fast-fail
* Update CI workflow
- add runner to all build names
- remove mi200 and mi300
* Update CI workflow
- code coverage runs on navi3
* Update CI workflow
- add runner to all build names
* Tweak to CI sanitizer jobs
* Update CI workflow
- Handle excluded tests
* Update CI workflow
- Handle excluded tests (part 2)
* Update CI workflow
- Handle excluded tests (fix quotations)
---------
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
[ROCm/rocprofiler-sdk commit: ae2abb584b]
- install function wrappers around HIP runtime API
- easily correlated to the executable
- safer than HSA runtime due to potential for HSA to get invoked after main returns
[ROCm/rocprofiler-sdk commit: b570ff5273]
* Modified hipMalloc size for main.cpp in sample
* Update samples/counter_collection/main.cpp
---------
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
[ROCm/rocprofiler-sdk commit: edb1883a05]
- formerly, the rocprofiler_callback_tracing_record_t data was stored in itr["record"], e.g. itr["record"]["correlation_id"]
- dropped "record" key, e.g. itr["correlation_id"]
[ROCm/rocprofiler-sdk commit: 32bc339789]
* Fix drm include for OpenSUSE
- uses libdrm/drm.h instead of drm/drm.h
* Fix "List Files" step in CI workflows
* Fix "List Files" step in CI workflows
[ROCm/rocprofiler-sdk commit: 2d5523b420]
* Minor fix
Removal of HSA from counter collection
Tests for AQL
Updated counter collection client to build profiles in tool init
* Rebased
* Debug printing
* Formatting
* More format
* fix shadowing
---------
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
[ROCm/rocprofiler-sdk commit: c2f659ab5c]
* Page migration reporting support
* Page migration: Update parser and reporting
Container does not lave latest KFD header, so CI might fail
* Add kfd_ioctl.h
* Formatting
* Update get_key
- get key was not used (and shouldn't be), so delete it
* clang-tidy fixes
* Tests for page migration
* Apply suggestions from code review
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* Update tests/bin/page-migration/CMakeLists.txt
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* Update page-migration test app
- add hipHostRegister to register mmap'ed allocation with HIP
- misc cleanup and reorg
- remove HSA_XNACK=1 from test env
* Update lib/rocprofiler-sdk/tests/page_migration.cpp
- fix compilation error
* Minor updates (reorg, rename)
* Page migration reporting support
* Page migration: Update parser and reporting
Container does not lave latest KFD header, so CI might fail
* Update page migration tests, fix trigger types
* Page Migration Tracing Support Refactoring (#753)
* Reorganization
* Update page migration init/fini
* Formatting
* Update page_migration.cpp
- change logging severity
* Skip test if KFD does not support page migration reporting
* Rework skipping test if KFD does not support page migration
* Fix event trigger enum values
* Fix clang-diagnostic-unused-const-variable
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
[ROCm/rocprofiler-sdk commit: fd3d97287c]
* Update lib/rocprofiler-sdk/context/*
- create correlation_id.{hpp,cpp} and moved implementation into these files instead of in context.{hpp,cpp}
* Update lib/rocprofiler-sdk/thread_trace/att_core.hpp
- fixed header includes
* Update lib/common/utility.hpp (runtime sizeof)
- added compute_runtime_sizeof<T>() function to set the "size" field to be the offset of the "reserved_padding" field if one exists
* Fix to compute_runtime_sizeof
[ROCm/rocprofiler-sdk commit: d6bb50cae1]
* Update test/tools/json-tool.cpp
- push/pop ppid as external correlation id instead of pid
* Update environment variables for tests and samples
* Revert to old CDash dashboard in run-ci.py
* Revert to new CDash dashboard in run-ci.py
[ROCm/rocprofiler-sdk commit: 3eaa678054]
Benjamin Welton
Jonathan R. Madsen
Vladimir Indic
Ammar ELWazir
Giovanni Lenzi Baraldi
Mythreya