- Clean up and standardization of MIT licenses after discussion with legal team.
- Update README.md with blurb for top-level files.
- MIT License explicitly mentioned for relevant projects.
- Removal of years.
- Copyright attribution should be to `Advanced Micro Devices, Inc.` and not `AMD ROCm(TM) Software`
- Removal of `All rights reserved.`
- Reduce line width of the text for readability.
- Add clear visual separators for additional licenses.
- Convert text files to markdown format for aforementioned separators.
- Update build scripts to point to renamed files.
- Fixed SMI doc references
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>
- Update minimum_cmake_required to match version used in CI
- We should match the minimum version that we test against
- Ensure ".S" files are treated as assembly.
- Refactor elfutils build folders. Move build artifacts to the "external" subfolders
- Remove reference to unused DYNINST configuration
- Declare Dyninst interfaces with other external declarations
- Updated cmake warning message
- Update the deprecation warning for "DYNINST_BUILD_${dep}" to make it clearer that the option is being translated to the new option, "ROCPROFSYS_BUILD_${dep}"
[ROCm/rocprofiler-systems commit: eaaec2cc3b]
Update Dyninst submodule
Refactoring of build scripts to build TBB, Boost, ElfUtils, and LibIberty, since Dyninst build scripts no longer do.
Workflows are now building Dyninst and its dependencies.
---------
Co-authored-by: marantic-amd <marantic@amd.com>
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 96df9b6d3e]
- Add support for RCCL API tracing through rocprofiler-sdk.
- Refactored the comm_data code to use the SDK RCCL_API callbacks.
- Add a runtime version check for SDK to gate callback enablement, rather than just the compile-time check.
- Fixed: SAMPLING_TIMEOUT was not being handled correctly in add_test.
[ROCm/rocprofiler-systems commit: af77d93f75]
- Move the MPI gotcha functionality from Timemory to the repo.
- Add the PMPI Fortran MPI functions to the existing mpi gotcha handle.
[ROCm/rocprofiler-systems commit: 4fcd8cc78d]
- Path to merge script not found unless user explicitly sources "share/rocprofiler-systems/setup-env.sh" to setup PATHs.
- Instead, let's derive the path when the application loads and use it when executing the helper script
- Rename script to rocprof-sys-merge-output.sh.
- Change install folder to <prefix>/libexec/rocprofiler-systems based on dev-ops feedback.
- Updated PATH variable in the modulefile and source scrtipt.
- For SWDEV-528101
[ROCm/rocprofiler-systems commit: adc66956b0]
CPACK is converting /usr/bin/env python3 to /usr/libexec/platform-python in RHEL8.
This is resulting in dependency to platform-python
Undefining __brp_mangle_shebangs will prevent the same
[ROCm/rocprofiler-systems commit: 1c3616fcc2]
* Set version for RPM_PACKAGE_OBSOLETES
* Not setting a version of omnitrace which rocprofiler-systems obsoletes will result in a warning
* Patch RPATH in libunwind files
* Update Timemory pointer
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 75916f8876]
- The amd-smi-lib package is not providing libamd_smi.so in its specs. We need to add a package dependency instead.
- For ticket: SWDEV-525268
- Add support for `ROCM_DEP_ROCMCORE` in CMake
[ROCm/rocprofiler-systems commit: eaad295cfc]
- Add JPEG activity track in perfetto trace
- Add JPEG decode tests to the examples
- Change existing videodecode test to include JPEG testing
- Rename videodecode test file to decode to include jpeg tests too
- Fix a bug in the test which checks for total activity of 0
- Disable rocDecode and rocJPEG samples from the github image files
[ROCm/rocprofiler-systems commit: 59d3399901]
- VA API tracing using Timemory gotcha wrappers.
- rocDecode API tracing integration using callback to ROCPROFILER_CALLBACK_TRACING_ROCDECODE_API
- Updated videodecode ctest to validate rocDecode APIs in perfetto trace.
[ROCm/rocprofiler-systems commit: 697d1ac02f]
* Integrating amd-smi into rocprofiler-systems due to rocm-smi deprecation.
* No functionality changes to users other than naming conventions.
* New tracks available in perfetto- gpu busy percentage metrics now splits gfx busy into separate gfx, umc, and mm engine measurements.
---------
Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 0c32dfd6bc]
- Added script to merge multiprocess output automatically to one file when there are multiprocess proto files written into output folder
- Execute the merge multiprocess script from the rank 0 process
- Added the scripts folder path to env path, via setup-env.sh
- Installed merge_multiprocess_output.sh to /share/rocprofiler-systems/bin dir
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 0263e951ff]
- Renames the CMake option "ROCPROFSYS_USE_HIP" to "ROCPROFSYS_USE_ROCM"
- Remove the "ROCPROFSYS_USE_ROCM_SMI option. Controlled with the "ROCPROFSYS_USE_ROCM" option, instead.
- Runtime configuration can still toggle ROCPROFSYS_USE_ROCM_SMI to disable the sampling.
- Rename ROCPROFSYS_HIP_VERSION macro to ROCPROFSYS_ROCM_VERSION and remove blocks for `ROCPROFSYS_ROCM_VERSION < 60000`
- Remove ROCPROFSYS_USE_ROCTRACER and ROCPROFSYS_USE_ROCPROFILER
- Update test cases
- Update docker files and workflows to install cmake 3.21, which is required for the rocprofiler-sdk findPackage script.
- Removed rocm-6.2 from workflows due to a rocprofiler-sdk API change.
[ROCm/rocprofiler-systems commit: 88aa2d3cbe]
* Add installers for rocm-6.3 and rhel-9.5
* Updated the template "rocprof-sys-install.py.in".
Fixed the installer for the "rocm-x.y.z" style tags.
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 7e2242414c]
Specify that "rocprofiler-systems" replaces / obsoletes the "omnitrace" package in the CPACK configuration.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: eb1af12177]
Limit to single job during the configure step of rocprofiler-systems-papi-build; error due to files not being available in time if number of jobs is too low.
Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
[ROCm/rocprofiler-systems commit: bf1f1e64d6]
Porting https://github.com/ROCm/omnitrace/pull/410.
As eval builtin interprets its arguments in the same way as shell would do, which would need some escape work, otherwise, it won't work if the input arguments contains e.g. a JSON string:
omnitrace-python -- ./test.py --json='{"foo": "bar"}'
[ROCm/rocprofiler-systems commit: ba0efacd3f]
- Porting from https://github.com/ROCm/omnitrace/pull/411
- Improve OMPT support
- Add OpenMP target example to testing
- Update Timemory submodule to use ROCm/Timemory rather than NERSC/Timemory
- Update `actions/upload-artifacts` to v4
- Standardize the `cmake_minimum_required` to 3.18.4 across workflows, project, and examples
- Updated Ubuntu 20.04 workflows
[ROCm/rocprofiler-systems commit: 7dce5926a7]
Updated OS test matrix to match ROCm 6.2.
Update build and CI docker files
Remove the "docs" workflow, because "read-the-docs" is now being used for ROCm documentation
[ROCm/rocprofiler-systems commit: b15c9e94fc]
The Omnitrace program is being renamed.
Full name: "ROCm Systems Profiler"
Package name: "rocprofiler-systems"
Binary / Library names: "rocprof-sys-*"
---------
Co-authored-by: Xuan Chen <xuchen@amd.com>
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: d07bf508a9]
- Current Perfetto source repo is from android.googlesource
which is censored in China and causes an infinite
hang when building Omnitrace
- This commit changes the Perfetto source repo to Github which acts
as a read-only mirror to the googlesource perfetto
repo.
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 6f22b2d2e5]
Fix for "SWDEV-479652" - Perfetto-based tests are failing.
Updated version of perfetto submodule to v46.0.
Modified Omnitrace code that uses Perfetto, so it can compile.
Modified the testing code, so it can run the version of trace_processor_shell provided (v46.0).
---------
Signed-off-by: Aleksandar Janicijevic <Aleksandar.Janicijevic@amd.com>
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 96d7b8f0ab]
* Modified submodules dyninst and timemory.
* Modified PAPI submodule to fix GCC 13 build for Ubuntu 24.04.
* Updated PAPI submodule URL.
* Fixed papi submodule URL.
* Using latest tag (papi-7-1-0-t) for papi submodule.
* Update submodule dyninst to new version
* Added a mirror to elfutils
[ROCm/rocprofiler-systems commit: c11abc4ea1]
* Add CMake CACHE Variable ElfUtils_DOWNLOAD_URL
Used to override the default URL to download ElfUtils from.
Useful for internal builds
Also, include a mirror to fallback to if the override URL fails.
* Update timemory submodule
Updating to include the BINUTIL_DOWNLOAD_URL override cmake
variable.
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: b0b1dca21e]
* Build omnitrace-rt library
- Explicitly build dyninstAPI_RT as omnitrace-rt so that the SONAME in the ELF is omnitrace-rt instead of dyninstAPI_RT
- Create symbolic link lib/omnitrace/libdyninstAPI_RT.so which points to lib/libomnitrace-rt.so
- Simplify build tree location of libomnitrace-rt.so since it is ../lib from the bin directory even in the build tree
- Update dyninst submodule with minor tweaks to dyninstAPI_RT/CMakeLists.txt
* Update source/lib/omnitrace-rt/cmake/platform.cmake
* Use ftpmirror.gnu.org instead of ftp.gnu.org
- in timemory and dyninst submodules
- minor .clang-tidy tweak
[ROCm/rocprofiler-systems commit: 0cf017251e]
* CPACK: Enable features to generate dependency lists
For DEBIAN, enable CPACK_DEBIAN_PACKAGE_SHLIBDEPS to generate a package
dependency list.
For RPM, enable CPACK_RPM_PACKAGE_AUTOREQPROV to automatically generate
lists of shared libraries that this package requires and provides.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
* Manually derive RPM provides list for libunwind.
[Why]
`dnf install` fails dependency resolution
[How]
Auto-generation did not add it to the "provides" list, despite it being
included in the package when OMNITRACE_BUILD_LIBUNWIND is enabled. So,
manually include these in the "CPACK_RPM_PACKAGE_PROVIDES" parameter.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
* Fix cmake-format linting
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 129580b416]
* Update CPack packaging variables.
Look for ROCM_LIBPATH_VERSION environment variable to patch the
CPACK_PACKAGE_VERSION.
Add some status logs to output packaging information.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
* Add CPack variables to "omnitrace_add_feature".
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 6bcd1d36cc]
* Tests for exceeding OMNITRACE_MAX_THREADS
- tests which exceeds OMNITRACE_MAX_THREADS value for thread creation
* CMake Formatting.cmake update
- include source files in /tests/source directory
* Add unknown-hash= to OMNITRACE_ABORT_FAIL_REGEX
- fail if a timemory hash is not resolved to a name
* Tests for exceeding OMNITRACE_MAX_THREADS
- update
* omnitrace-sample update
- remove env disabling of critical-trace and process-sampling
* core library update
- make_unique in concepts.hpp
- add OMNITRACE_USE_ROCM_SMI to "process_sampling" category
- remove forced disabling of critical-trace in sampling mode
- parentheses for OMNITRACE_PREFER
- use tim::get_hash_id instead of tim::get_combined_hash_id
* core library update (containers)
- added aligned_static_vector.hpp
- similar to static_vector.hpp but attempts to align to cache line size
- alignment template parameter for stable_vector
- added missing aliases in static_vector
- consistent with aligned_static_vector aliases
* thread_info update
- track the peak number of threads created
- thread_info::get_peak_num_threads() returns the peak number of threads
* thread_data update
- generic thread_data inherits from base_thread_data
- thread_data reworked to support dynamic expansion
- base_thread_data updated to invoke private_instance() function
- thread_data<optional<T>> uses stable_vector aligned to cache line width
- thread_data<identity<T>> uses stable_vector aligned to cache line width
- thread_data for optional and identity provide private private_instance function + friend to base_thread_data
- component_bundle_cache<T> is now thread_data<component_bundle_cache_impl<T>>
* causal update
- thread_data<T>::instances -> thread_data<T>::instance(construct_on_thread{ ... })
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
- tim::get_combined_hash_id -> tim::get_hash_id
- update progress_bundle usage to new thread_data API
* backtrace/backtrace_metrics component update
- backtrace_metrics update
- update to new thead_data API
- add thread CPU time row in perfetto
- fix potential bug when rusage categories are disabled
- fix bug in operator-= not subtracting cpu time of rhs
- backtrace update
- skip all child call-stack below 'tim::openmp::' if sampling_keep_internal = false
* pthread_gotcha component update
- pthread_gotcha::shutdown() invokes pthread_create_gotcha::shutdown()
* pthread_create_gotcha component update
- minor tweak to {start,stop}_bundle functions: pass in thread id
- update to new thread_data API
- track native handles of internal threads
- implement system with pthread_kill to stop dangling bundles
* rocprofiler/roctracer component update
- update to new thread_data API
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
* critical trace (library) update
- update to new thread_data API
- tim::get_combined_hash_id -> tim::get_hash_id
* coverage update
- update to new thread_data API
* tasking update
- update to new thread_data API
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
* roctracer update
- update to new thread_data API
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
* rocm_smi update
- update to new thread_data API
* runtime.cpp update
- update to new thread_data API
* sampling.cpp update
- update to new thread_data API
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
* ompt.cpp update
- invoke pthread_gotcha::shutdown before invoking OMPT finalize function
- this prevents signals from being delivered to OpenMP threads
* tracing.hpp and tracing.cpp update
- replace get_timemory_hash_{ids,aliases} functions with copy_timemory_hash_ids function
- update to new thread_data API
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
- tim::get_combined_hash_id -> tim::get_hash_id
- improvements to + error checking in thread_init function
* library.cpp update
- move copying timemory hash id/aliases to tracing.cpp
- update to new thread_data API
- loop over max_supported_threads (constexpr) -> loop over thread_info::get_peak_num_threads()
* Update BuildSettings.cmake
- add -Wno-interference-size to suppress warning about use of std::hardware_destructive_interference
* Update fork example
- improve scheme for waiting on child processes via waitpid instead of wait
- support running main routine multiple times
- push/pop regions in child process
* Update lib/common/defines.h.in
- allow use to specify misc values via -D <name>=<value>
- OMNITRACE_CACHELINE_SIZE
- OMNITRACE_CACHELINE_SIZE_MIN
- OMNITRACE_ROCM_MAX_COUNTERS
- remove unused defines
- OMNITRACE_ROCM_LOOK_AHEAD
- OMNITRACE_MAX_ROCM_QUEUES
* Update rocprofiler.hpp
- OMNITRACE_MAX_ROCM_COUNTERS -> OMNITRACE_ROCM_MAX_COUNTERS
* Update aligned_static_vector
- set cacheline_align_v from max of OMNITRACE_CACHELINE_SIZE and OMNITRACE_CACHELINE_SIZE_MIN
* Update tracing.cpp
- acquire locks for updating main hash ids/aliases
- only propagate ids/aliases when finalizing
* Update pthread_create_gotcha.cpp
- make sure hash for "start_thread" exists on main thread
* Update causal end to end tests
- if OMNITRACE_BUILD_NUMBER is 1, set OMNITRACE_VERBOSE=0
[ROCm/rocprofiler-systems commit: 518c83e0f9]
* Bump version to 1.10.3
* Drop releases for ROCm < 5.3
- ROCm is no longer providing release for Ubuntu 18.04 starting with 5.3 so omnitrace is dropping support for Ubuntu 18.04 + ROCm
- Dropping ROCm 5.2 releases for Ubuntu 20.04
- Dropping ROCm 5.2 releases for OpenSUSE 15.4
* Update redhat workflow
- Test RedHat 9.1 + ROCm 5.5
- Test RedHat 9.1 + ROCm 5.6
* Update ubuntu-focal workflow
- drop ROCm 5.2 testing
- add ROCm 5.6 testing
* Update Findroctracer.cmake
- provide /opt/amdgpu to HINTS/PATHS for drm and drm_amdgpu libraries
* Update Findrocprofiler.cmake
- prefer librocprofiler64.so.1
* Update librocprofiler64.so to librocprofiler64.so.1
- search for the SOVERSION 1 library of librocprofiler64.so if ROCm > 5.5.0
* Update Findrocprofiler.cmake
- link to libpciaccess for ROCm 5.5.0
* Update redhat CI workflow
- install libpciaccess for rocm CI
* Update cpack workflow
- Remove all RHEL 9.0 packaging
- Remove all packaging for ROCm 5.3 on OSes supporting where releases are provided for 5.4, 5.5, and 5.6
* Update ubuntu focal workflow
- remove rocm 5.3 jobs
[ROCm/rocprofiler-systems commit: 1216fd99a7]
* Update perfetto args.gn.in
- remove enable_perfetto_tools_trace_to_text (unused)
* core timeout implementation
- requires OMNITRACE_CI=ON
- requires OMNITRACE_CI_TIMEOUT=<sec>
- adds pthread_self and std::this_thread::get_id to thread info
- pthread_create_gotcha stores native handles (pthread_self)
* Testing updates
- improve detection of segfault/failures with PASS_REGEX exists
- add OMNITRACE_CI_TIMEOUT env variable to all tests
* Line-info in releases
- e.g. -g1 + more options to minimize size of debug info
* Fix typo in config exit action message
* OMNITRACE_UNLIKELY around debug/verbose messages
* format fixes
* Overflow tests + capability check
* transpose example update
- link to threads library
* roctracer/rocprofiler update
- in ROCm 5.5.0, cannot include rocprofiler.h and roctracer.h in same file due to conflicting enum defs
- Moved HSA tracing setup/shutdown to component::roctracer
* roctracer update
- fix definition of roctracer::setup when disabled
* Update fork example
- detach threads on main PID
- flush io outputs when printing info
* Update overflow tests
- pass regular expressions
- overflow on PERF_COUNT_SW_CPU_CLOCK event
* fork gotcha update
- use getpid() instead of getppid()
* update fork example
- wait on threads calling fork
* timeout update
- wait on timeout thread to launch before proceeding
[ROCm/rocprofiler-systems commit: 3e2fa69a14]
Joseph Macaranas
Kian Cossettini
David Galiffi
Aleksandar Djordjevic
Pranjal Swarup
Sajina PK
raramakr
arvindcheru
cfallows-amd
ajanicijamd
jamesxu2
Jonathan R. Madsen
Tal Ben-Nun