- A valid trace is now given even if an individual entry in an API trace
does not contain a parameter list
- Previously an "unknown" record in an API trace would cause the entire
output to be discarded
Change-Id: I29bf6df2cf58a231d836aafb37fbde9c5d85512f
- A valid trace is now given even if an individual entry in an API trace
does not contain a parameter list
- Previously an "unknown" record in an API trace would cause the entire
output to be discarded
Change-Id: I29bf6df2cf58a231d836aafb37fbde9c5d85512f
[ROCm/rocprofiler commit: 4467e50453]
- Fix for proto files not being viewable in Perfetto UI
- Ported from https://github.com/ROCm/omnitrace/pull/411
- Update Workflows
- Use V47 trace_processor_shell for certain OS releases.
- RedHat 8, SUSE 15.5, and Ubuntu 20.04 are no longer compatible with the latest trace_processor_shell.
- Incompatible version of GLIBC.
- Remove notes about Perfetto workaround in documentation.
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-systems commit: 21d9ab79fd]
- Fix for proto files not being viewable in Perfetto UI
- Ported from https://github.com/ROCm/omnitrace/pull/411
- Update Workflows
- Use V47 trace_processor_shell for certain OS releases.
- RedHat 8, SUSE 15.5, and Ubuntu 20.04 are no longer compatible with the latest trace_processor_shell.
- Incompatible version of GLIBC.
- Remove notes about Perfetto workaround in documentation.
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
* SDK: create CMake option for strict checks on CPU vs. GPU timestamps
- Configurating CMake with `ROCPROFILER_BUILD_CI_STRICT_TIMESTAMPS=ON` will enable fatal errors if dispatch/memcpy timestamps on GPU are outside of the start/end time from the CPU
- `ROCPROFIELR_BUILD_CI_STRICT_TIMESTAMPS` defaults to the value of `ROCPROFILER_BUILD_CI`
* Formatting
* Disable async_copy frequency scaling
* Disable profiling dispatch time frequency scaling
* Support runtime configuration via env variables
- ROCPROFILER_CI_FREQ_SCALE_TIMESTAMPS env variable will enable scaling the timestamps based on the hsa timestamp period
- ROCPROFILER_CI_STRICT_TIMESTAMPS env variable will enable strict timestamp checks
- when cmake is configured with ROCPROFILER_BUILD_CI_STRICT_TIMESTAMPS=ON, this env variable defaults to true
* ROCPROFILER_BUILD_CI_STRICT_TIMESTAMPS defaults to OFF
* Update cmake-target
* Common tracing::adjust_profiling_time
---------
Co-authored-by: Gopesh Bhardwaj <gopesh.bhardwaj@amd.com>
[ROCm/rocprofiler-sdk commit: ad48201912]
* SDK: create CMake option for strict checks on CPU vs. GPU timestamps
- Configurating CMake with `ROCPROFILER_BUILD_CI_STRICT_TIMESTAMPS=ON` will enable fatal errors if dispatch/memcpy timestamps on GPU are outside of the start/end time from the CPU
- `ROCPROFIELR_BUILD_CI_STRICT_TIMESTAMPS` defaults to the value of `ROCPROFILER_BUILD_CI`
* Formatting
* Disable async_copy frequency scaling
* Disable profiling dispatch time frequency scaling
* Support runtime configuration via env variables
- ROCPROFILER_CI_FREQ_SCALE_TIMESTAMPS env variable will enable scaling the timestamps based on the hsa timestamp period
- ROCPROFILER_CI_STRICT_TIMESTAMPS env variable will enable strict timestamp checks
- when cmake is configured with ROCPROFILER_BUILD_CI_STRICT_TIMESTAMPS=ON, this env variable defaults to true
* ROCPROFILER_BUILD_CI_STRICT_TIMESTAMPS defaults to OFF
* Update cmake-target
* Common tracing::adjust_profiling_time
---------
Co-authored-by: Gopesh Bhardwaj <gopesh.bhardwaj@amd.com>
* Install LICENSE to ${CMAKE_INSTALL_DOCDIR}
* Update tarball.yml to test for LICENSE file after the install
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-compute commit: 50755de867]
* Install LICENSE to ${CMAKE_INSTALL_DOCDIR}
* Update tarball.yml to test for LICENSE file after the install
---------
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
* Update VERSION and README to 3.0.0
* Update version mismatch of profiling data to a warning.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
[ROCm/rocprofiler-compute commit: 152eff0df1]
- Default values are being assigned causing occupancy calculation to go
wrong without the right values defined for gfx12 ASICs
- Also added the these values for gfx1105
Change-Id: I611cc3a8ed8c57f2def637310ce1c3a48c16a574
- Update filenames.
- Update executable to `rocprof-compute`
- Update update package to `rocprofiler-compute`
- Update name in application output and logs
- Update name in README files
- Update testing and workflows
---------
Signed-off-by: Xuan Chen <xuchen@amd.com>
[ROCm/rocprofiler-compute commit: 31b4de1a38]
- Update filenames.
- Update executable to `rocprof-compute`
- Update update package to `rocprofiler-compute`
- Update name in application output and logs
- Update name in README files
- Update testing and workflows
---------
Signed-off-by: Xuan Chen <xuchen@amd.com>
- Device can have multiple isas as per HSA spec
- First isa is most specific one, so this change is sort of a NOP
Change-Id: Ib332af21745f2e6a7c25db8986bf7717501059bc
After setting the new params in hipDrvGraphExecMemsetNodeSetParams, we
need to update the AQL packet as well, otherwise during the graph launch
it still dispatches the packet which has the original params and not the
updated one.
Change-Id: Ie49a641ba3f66c8085a29f92d88ac6ea6a1c0534
meserve-amd
David Galiffi
Fei Zheng
Benjamin Welton
Peter Park
srawat
Vladana Stojiljkovic
Marko Arandjelovic
Ajay
fei.zheng
Jaydeep Patel
Li, Todd tiantuo
Branislav Brzak
Jonathan R. Madsen
Cole Ramos
Saleel Kudchadker
Satyanvesh Dittakavi
xuchen-amd
Rahul Manocha
Konstantin Zhuravlyov
Tao Sang
adapryor