* getting started for amples and tools
* Adding getting started to documentation
* Addressing review comments
* minor update
* renaming rocprofv2 to rocprofv3 in doc
* Apply suggestions from code review
---------
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
- fix logic for using HSA_TOOLS_LIB when rocprofiler-register support is enabled
- report tool load failure for rocprofiler-register
Change-Id: Ife23aa3e6ed19174376cd694764583b73f8976cd
The alternate scratch memory is used for dispatches that have a low
number of waves but relatively large wave size.
This allows us to keep the tmpring_size.bits.WAVES field of the main
scratch to full occupancy.
Change-Id: I32d240fac4b7d38200d1eebc1b0fdc8a823920d3
For devices where the CP FW supports asynchronous scratch reclaim, ROCr
is able to claw-back scratch memory that was assigned to an AQL queue.
With that ability, ROCr does not have to rely on using USO
(use-scratch-once) when assigning large amounts of memory to a queue.
If we reach a situation where we are running low on device memory, ROCr
will attempt to claw-back the scratch memory.
Change-Id: Iddf8ec84e37ab8b9fdc58bafbe2b61fe2acb6eb7
Update queue structure to add members required for asynchronous reclaim
mechanism and dual-scratch. CP will set the AMD_QUEUE_CAPS_ASYNC_RECLAIM
bit on queue-connect to indicate whether the new features are supported.
The new members are ignored by previous versions of CP FW
Change-Id: Ic8e9ef41c5b1d04f09b43bc9b44b31527863d10f
For gfx11, the trap_handler fails to recognize a trap id 3 and report
the exception to the debugger if the debugger is attached.
This is because the 2nd level trap handler looks for the DEBUG_ENABLED
bit in ttmp13 instead of ttmp11. This bit is set by the 1st level trap
handler and is part of the 1st/2nd level trap handler ABI.
Change-Id: Ib36361f53d9bcbbed52320d8c3a9ab2c0b28c7cd
Skip Extended-scope memory pool as allocation is very close to
fine-grain/coarse-grain but with just different PTE flags.
Only test coarse grain on CPU agent other than the first CPU agent.
Stop bisecting the max size once we are withing 5% to total size for
these pool to speed this test on large memory pools.
Change-Id: I77d1b45a1752ef092dda7c7f27723ea0a292a612
SDMA4.4 and SDMA5.2+ has increased it's available copy size to 2^30 bytes
represented by exponent as bits set in the COUNT field of the
linear copy.
Also note that the full 2^22 byte limit is available from SDMA4 onwards
as it has corrected the 0x3fffe0 HW limitation from SDMA3.
As copy limit has increase, this can change system performance
so provide env var HSA_ENABLE_SDMA_COPY_SIZE_OVERRIDE=0 to fall
back to the original 0x3fffe0 limit for debugging purposes.
Change-Id: I0fb6e5378f68e5b8a00ff559271691a943ee06ee
To be able to trace memcpy asynchronously, both dst and src agents need to have profiling enabled and the api for enabling profiling was only enabling for gpu agents. CPU agents didn't have profiling enabled so the signal owner could not be known. hsa_amd_profiling_get_async_copy_time will fail with an HSA status error because it can't read the agent for the given signal.
Change-Id: Ie165e0e39b8fcd6992a55695b9ffcead10a8e812
- Update CMakeLists.txt
- find_package for rocprofiler-register
- this is an optional package until rocprofiler-register is added to the CI
- define HSA_VERSION_{MAJOR,MINOR,PATCH} ppdefs
- Update runtime.cpp
- include <rocprofiler-register/rocprofiler-register.h>
- if rocprofiler-register succeeds, do not support v1 unless explicitly requested
Change-Id: I8f48bbf3f6b52fb91ddade2f198491a1256035fe
Remove override that forces ROCr image blit source and ROCr test to use
code object version 4 now that mainline has been updated to version 5.
Change-Id: I94681e86835c0e382475306ead4cd4132a2ee78f
* Fix support for version in find_package(rocprofiler-register <version>)
* Update lib/rocprofiler-register/rocprofiler-register.cpp
- change rocprofiler library name to librocprofiler-sdk.so
- add helper function rocp_load_rocprofiler_lib for dlopen of rocprofiler library
- support ROCP_TOOL_LIBRARIES environment variable
- reworked rocprofiler_register_error_string
* Update tests/rocprofiler
- rename library to librocprofiler-sdk
- remove rocprofiler_configure within librocprofiler-sdk.so
* Add tests/generic-tool
- provides implementation of rocprofiler_configure
* Update tests/CMakeLists.txt
- use libgeneric-tool.so
- LD_PRELOAD libgeneric-tool.so instead of librocprofiler
- Add tests which use ROCP_TOOL_LIBRARIES
* Update tests/generic-tool/generic-tool.cpp
- include <cstdint>
* Fix find_package(rocprofiler) in build tree
* Move include/rocprofiler to include/rocprofiler-sdk
* Update include/CMakeLists.txt
- add_subdirectory(rocprofiler-sdk)
* Move lib/rocprofiler to lib/rocprofiler-sdk
* Move lib/rocprofiler-tool to lib/rocprofiler-sdk-tool
* Update lib/CMakeLists.txt
- add_subdirectory(rocprofiler-sdk)
- add_subdirectory(rocprofiler-sdk-tool)
* Update lib/rocprofiler-sdk/CMakeLists.txt
* Rename rocprofiler-tool to rocprofiler-sdk-tool
* Replace include rocprofiler/ with include rocprofiler-sdk/
* Replace include lib/rocprofiler/ with include lib/rocprofiler-sdk/
* Set VERSION to 0.0.0 and finish install to rocprofiler-sdk
* More fixes for rocprofiler -> rocprofiler-sdk
- fix issue with rocprofiler-sdk-config.cmake.in
- fix counters xml install path
* Fix documentation generation
* Create rocprofiler_LIB_ROCPROFILER_SDK_DIR for build tree
* cmake formatting (cmake-format) (#264)
Co-authored-by: jrmadsen <jrmadsen@users.noreply.github.com>
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
MCM die check was inconsistent (using avg power).
By using only the energy counter, this provides
a consistent way of checking which die is the MCM node.
Change-Id: I532fa2047706d0f1e92e643ce1e6759e45b65ec0
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
Uses new support for 'gpu_metrics_v1_4'
Code changes related to the following:
* rsmi gpu_metrics APIs
* rsmi gpu_metrics Logs
* new data structure fields added in 1.4
* added APIs for all other existing metrics before 1.4
* added support to older metrics; 1.1, and 1.2
* added support to dump_internal_metrics_table()
* public APIs renamed to start with prefix 'rsmi_dev_metrics_'
* Unit tests updated
* Examples updated
Build changes related to the following: None
Change-Id: I23e59f99d3ed43318cd6bd43bd2f0c5387e9ccb9
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
* Run jobs on vega20 instead of mi200 as per ammar's instructions. mi200 is too unstable for CI/CD
* Remove the options --memory=128g --cpus=32 from docker execution for job vega20-ubuntu
* Fix tab
* Fix tab again
* Forgot to switch code-coverage job to vega20
* Fix workflow file syntax error
* Fix workflow file syntax error
* Remove flags --memory and --cpus
* Remove flags --device=/dev/dri and device: entries
* Update .github/workflows/continuous_integration.yml
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
* Apply suggestions from code review
---------
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
Gopesh Bhardwaj
Galantsev, Dmitrii
Giovanni LB
Jonathan R. Madsen
Jonathan R. Madsen
David Yat Sin
Shweta Khatri
Lancelot SIX
Lang Yu
Jonathan Kim
Youssef Aly
Karl W Schulz
colramos-amd
Benjamin Welton
JoseSantosAMD
Ammar ELWazir
AravindanC
Charis Poag
Oliveira, Daniel
pbhandar-amd