Fixes reading pp_od_clk_voltage new variable format and size.
Code changes related to the following:
* get_od_clk_volt_info()
* get_od_clk_volt_curve_regions()
* Unit tests
* CLI options restored: --showclkvolt, --showvc, --showvoltagerange, --setvc
* Rework: 48ddd9ab
* Bump CLI version
* CHANGELOG.md
Change-Id: I817ca224de923fdaa992df84592d63b4d5a12b22
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
provide a kinder error message in the case where python dependencies
are not available locally. This is motivated for future execution by
users who are running from rocm-based binary packaging instead of using
normal cmake build system which would have verified the dependencies.
Signed-off-by: Karl W. Schulz <karl.schulz@amd.com>
For new ASIC, the RDC_EVNT_XGMI, RDC_FI_PCIE_RX and RDC_FI_PCIE_TX
are not supported. New fileds RDC_FI_XGMI and RDC_FI_PCIE_BANDWIDTH
should be used.
Change-Id: Iff5bbef4c07994090fa7c4e9b319966215525283
This commit aligns the rsmiBindings.py.in file's
"notification_type_names" & "rsmi_evt_notification_type_t" with
those found in the rsmiBindings.py file.
Change-Id: I67f36606c505992fb98495651310bd70a1755033
Signed-off-by: Ori Messinger <Ori.Messinger@amd.com>
Fixes reading pp_od_clk_voltage new variable format and size.
Code changes related to the following:
* get_od_clk_volt_info()
* get_od_clk_volt_curve_regions()
* Unit tests
* CLI options removed: --showclkvolt, --showvc, --showvoltagerange, --setvc
Change-Id: Ieedb845eeadcea2f2e447ec576c253ad2a814176
Signed-off-by: Oliveira, Daniel <daniel.oliveira@amd.com>
* 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>
This patch adds 'ring hang' enums to ROCM SMI LIB.
This event type name is KFD_SMI_EVENT_RING_HANG.
Signed-off-by: Ori Messinger <Ori.Messinger@amd.com>
Change-Id: I9b886eb1fc027f03bcca1e5d1a89a2a186b64bf5
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>
This test may fail when run on non-upstream versions of KFD as this
feature will not be upstreamed.
Change-Id: I7131e1f50984739c0df12e4c9afe790bd7e4cdfa
It seesm the Release() function is not reliable and can cause segfaults.
This is a temporary work-around until the Release() function is fixed.
Change-Id: I95470a800c6153673e4b8f4fe46a646903325074
Modifying the /opt/rocm/etc/rdc file modifies RDC launch options. If
the file doesn't exist, the service should still launch (though a new
file should likely be included with the next released package of 'rdc'.
Change-Id: I1a1891e9c5c3e6048754eb555779a97a170754c0
- Add CPACK_RPM_PACKAGE_RELEASE_DIST to set RPM distro.
- Indent packaging logs for easier reading.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
If pthread_attr_setaffinity_np function exists use it instead of
pthread_setaffinity_np as pthread_setaffinity_np seems to fail to set
the affinity settings on some systems.
Change-Id: Icd8b17039699ac10d9cd5c4dbb6ac44630673949
Bumping HSA_AMD_INTERFACE_VERSION_MINOR version to 5 to account for
previously added GPU agent query: HSA_AMD_AGENT_INFO_MEMORY_PROPERTIES
Change-Id: Ic8cfdcfb7bad6f3d1e0b3d68f505a62074fc26b9
- move install command for license to rocprofiler_register_config_install.cmake
- CPACK_RPM_PACKAGE_LICENSE should be set to license type (i.e. "MIT"), not the path to the license file
* SWDEV-448381 - CPACK dependency list should be comma separated
* Remove redundant code and update the dependency list in first place
* Update cmake/rocprofiler_register_config_packaging.cmake
---------
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
Perfetto added more restrictions on the types, so "1" is different from 1 alone, which wasn't the case before. Thats why we are seeing this issue recently
Change-Id: I69835c8f61bb9168fdfcdcd0821c1ba35d0369fe
* 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>
Maisam Arif
Oliveira, Daniel
Jonathan R. Madsen
Galantsev, Dmitrii
guanyu12
Ammar ELWazir
Giovanni Lenzi Baraldi
Karl W. Schulz
coleramos425
Bill(Shuzhou) Liu
Ori Messinger
Chris Freehill
David Galiffi
amd-jmacaran
AravindanC
Benjamin Welton
David Yat Sin
Brandon Bagwell
Kent Russell
raramakr
Ammar ELWazir
Vladimir Indic