Граф коммитов
64512 Коммитов
Автор SHA1 Сообщение Дата
Jonathan R. Madsen b330e347e8 Fixes missing call to mpi_gotcha::update() (#88)
- mpi_gotcha::shutdown() calls mpi_gotcha::update()
  - fixes the MPI rank not being available to output files when `OMNITRACE_USE_PID=ON` 

[ROCm/rocprofiler-systems commit: 8f8aa43d1a]
2022-07-18 07:00:38 -05:00
Jonathan R. Madsen 8f8aa43d1a Fixes missing call to mpi_gotcha::update() (#88)
- mpi_gotcha::shutdown() calls mpi_gotcha::update()
  - fixes the MPI rank not being available to output files when `OMNITRACE_USE_PID=ON`
2022-07-18 07:00:38 -05:00
Sarbojit Sarkar b11cc11a79 SWDEV-1 - Fixed hipGetErrorString
Change-Id: I7224014b945bafe1cec6d1d0e11184c3459ac1b0


[ROCm/clr commit: 323ed5bd82]
2022-07-18 03:38:28 -04:00
Sarbojit Sarkar 323ed5bd82 SWDEV-1 - Fixed hipGetErrorString
Change-Id: I7224014b945bafe1cec6d1d0e11184c3459ac1b0
2022-07-18 03:38:28 -04:00
Jonathan R. Madsen 41b86132bb Support ACTIVITY_DOMAIN_ROCTX (#87)
- New configuration variable: OMNITRACE_USE_ROCTX
- Enable support for roctxRangePushA, roctxRangePop, roctxRangeStartA, roctxRangeStop

[ROCm/rocprofiler-systems commit: d22725e830]
2022-07-18 02:06:40 -05:00
Jonathan R. Madsen d22725e830 Support ACTIVITY_DOMAIN_ROCTX (#87)
- New configuration variable: OMNITRACE_USE_ROCTX
- Enable support for roctxRangePushA, roctxRangePop, roctxRangeStartA, roctxRangeStop
2022-07-18 02:06:40 -05:00
Jonathan R. Madsen 39adaaef98 Unified setup_environ b/t libomni and libomni-dl (#86)
- omnitrace::common::setup_environ provides a standardized exporting of relevant env variables

[ROCm/rocprofiler-systems commit: a2dcc7381b]
2022-07-18 01:52:21 -05:00
Jonathan R. Madsen a2dcc7381b Unified setup_environ b/t libomni and libomni-dl (#86)
- omnitrace::common::setup_environ provides a standardized exporting of relevant env variables
2022-07-18 01:52:21 -05:00
Jonathan R. Madsen 9e4977baa8 Fix statistics type and use feature name indexes (#85)
- fix reporting units and statistics type for rocm_data_tracker
- use indexes for feature names instead of strings

[ROCm/rocprofiler-systems commit: 7745c12417]
2022-07-18 01:52:03 -05:00
Jonathan R. Madsen 7745c12417 Fix statistics type and use feature name indexes (#85)
- fix reporting units and statistics type for rocm_data_tracker
- use indexes for feature names instead of strings
2022-07-18 01:52:03 -05:00
Jonathan R. Madsen 7d1989f5a4 GPU HW Counters via rocprofiler (#84)
* Initial support for GPU hardware counters

* Update find modules for roctracer and rocprofiler

- /opt/rocm/{rocprofiler,roctracer} path is deprecated so tweak search procedure

* Improve ConfigCPack for MPI

* Update rocprofiler

- rocm_metrics()
- minor cleanup

* Update rocm find modules

* declare rocm_metrics + call in omnitrace-avail

* relocate omnitrace-launch-compiler

* REALPATH and find_modules

* Examples cmake (may drop)

* omnitrace-avail

- hw_counter categories
- init rocm

* setenv updates for rocprofiler in library.cpp and dl.cpp

* get_rocm_events config

* gpu::hip_device_count()

* rocm_metrics returns hardware_counters::info

* - relocated library/components/roctracer_callbacks.* to library/roctracer.*
- relocated library/components/rocprofiler.* to library/rocprofiler.*
- cleaned up rocprofiler.hpp
- added perfetto output of rocprofiler
- added timemory output of rocprofiler
- renamed omni.roctracer thread to roctracer.hip
- added roctracer.hsa thread name
- updated timemory submodule to support std::variant
- updated timemory submodule to support = in config value
- updated timemory submodule to support standalone storage
- updated timemory submodule to support new hw counter apis
- updated timemory submodule to prevent label/description caching in data_tracker

* update omnitrace-avail info_type generation

* Update timemory submodule

* rocprofiler component

* cmake formatting

* omnitrace-avail handle no GPUs

- Add -c command-line option for --categories
- support verbosity

* hsa_rsrc_factory throws exceptions

- throw exceptions to avoid aborting on HSA_STATUS_ERROR_NOT_INITIALIZED when advantageous
- removed duplicate specialization of is_available for component::rocprofiler

* rocprofiler symbols for when disabled

* Fix warning in omnitrace-avail

- std::stringstream from initializer list would use explicit constructor

* Fix finalization after settings are deleted

* Reorganized rocprofiler source

* Updated formatting

* Miscellaneous tweaks

- added using statements from timemory
- tweaked the main and thread bundle names
- fixed timemory header includes

[ROCm/rocprofiler-systems commit: 4208b5654c]
2022-07-17 21:52:09 -05:00
Jonathan R. Madsen 4208b5654c GPU HW Counters via rocprofiler (#84)
* Initial support for GPU hardware counters

* Update find modules for roctracer and rocprofiler

- /opt/rocm/{rocprofiler,roctracer} path is deprecated so tweak search procedure

* Improve ConfigCPack for MPI

* Update rocprofiler

- rocm_metrics()
- minor cleanup

* Update rocm find modules

* declare rocm_metrics + call in omnitrace-avail

* relocate omnitrace-launch-compiler

* REALPATH and find_modules

* Examples cmake (may drop)

* omnitrace-avail

- hw_counter categories
- init rocm

* setenv updates for rocprofiler in library.cpp and dl.cpp

* get_rocm_events config

* gpu::hip_device_count()

* rocm_metrics returns hardware_counters::info

* - relocated library/components/roctracer_callbacks.* to library/roctracer.*
- relocated library/components/rocprofiler.* to library/rocprofiler.*
- cleaned up rocprofiler.hpp
- added perfetto output of rocprofiler
- added timemory output of rocprofiler
- renamed omni.roctracer thread to roctracer.hip
- added roctracer.hsa thread name
- updated timemory submodule to support std::variant
- updated timemory submodule to support = in config value
- updated timemory submodule to support standalone storage
- updated timemory submodule to support new hw counter apis
- updated timemory submodule to prevent label/description caching in data_tracker

* update omnitrace-avail info_type generation

* Update timemory submodule

* rocprofiler component

* cmake formatting

* omnitrace-avail handle no GPUs

- Add -c command-line option for --categories
- support verbosity

* hsa_rsrc_factory throws exceptions

- throw exceptions to avoid aborting on HSA_STATUS_ERROR_NOT_INITIALIZED when advantageous
- removed duplicate specialization of is_available for component::rocprofiler

* rocprofiler symbols for when disabled

* Fix warning in omnitrace-avail

- std::stringstream from initializer list would use explicit constructor

* Fix finalization after settings are deleted

* Reorganized rocprofiler source

* Updated formatting

* Miscellaneous tweaks

- added using statements from timemory
- tweaked the main and thread bundle names
- fixed timemory header includes
2022-07-17 21:52:09 -05:00
Ajay 56a71f18a2 SWDEV-327563 - catch2 deviceLib build failure
Change-Id: I82f95d61f64018a761e45dc8f424af032953fc51


[ROCm/clr commit: 9b18e404ae]
2022-07-15 21:11:20 -04:00
Ajay 9b18e404ae SWDEV-327563 - catch2 deviceLib build failure
Change-Id: I82f95d61f64018a761e45dc8f424af032953fc51
2022-07-15 21:11:20 -04:00
Brian Sumner 1d88a5df2c SWDEV-343278 - fix conversion functions
Change-Id: Ib93f886363314e980462ef562c998981bb958569


[ROCm/clr commit: a47887aa0d]
2022-07-15 16:00:58 -04:00
Brian Sumner a47887aa0d SWDEV-343278 - fix conversion functions
Change-Id: Ib93f886363314e980462ef562c998981bb958569
2022-07-15 16:00:58 -04:00
Ammar ELWazir f72cedce81 SWDEV-345650 - Solving hipKernelNameRefByPtr Stream Issue
Stream is not important to get the kernel
name from Function that can be found in
the functions_ map.

Change-Id: I164bc3ebcc5552359856e76204d8b124ba0d2f34


[ROCm/clr commit: fb8690f812]
2022-07-15 15:38:39 -04:00
Ammar ELWazir fb8690f812 SWDEV-345650 - Solving hipKernelNameRefByPtr Stream Issue
Stream is not important to get the kernel
name from Function that can be found in
the functions_ map.

Change-Id: I164bc3ebcc5552359856e76204d8b124ba0d2f34
2022-07-15 15:38:39 -04:00
Ammar ELWazir b2b252ba31 SWDEV-344206: Fixing Roctracer Tests
Making sure not to count duplicates for load_unload_reload_trace and
fixed the ignore-count option in check_trace.py.

Change-Id: I9e674aa624ec3b473bb7c6cc95260e240204627f


[ROCm/roctracer commit: c588c49743]
2022-07-15 12:39:55 -04:00
Ammar ELWazir c588c49743 SWDEV-344206: Fixing Roctracer Tests
Making sure not to count duplicates for load_unload_reload_trace and
fixed the ignore-count option in check_trace.py.

Change-Id: I9e674aa624ec3b473bb7c6cc95260e240204627f
2022-07-15 12:39:55 -04:00
Ammar ELWazir dc269b6c7e Temporary: Disabling Roctracer test package
When separate debug info is requested, the test package
generation fails because /usr/bin/objcopy does not understand
the HSA code object format. We need a workaround to get
past this issue.

Change-Id: I9a307fcf532ce8219a9301850aae972303d19990


[ROCm/roctracer commit: c2ecd15197]
2022-07-15 09:33:08 -07:00
Ammar ELWazir c2ecd15197 Temporary: Disabling Roctracer test package
When separate debug info is requested, the test package
generation fails because /usr/bin/objcopy does not understand
the HSA code object format. We need a workaround to get
past this issue.

Change-Id: I9a307fcf532ce8219a9301850aae972303d19990
2022-07-15 09:33:08 -07:00
Laurent Morichetti 68daae8c75 Fix a SEGV in std::atoi
When ROCP_TRUNCATE_NAMES is not set, getenv returns NULL and std::atoi
crashes. Check that getenv returns a non-NULL string before calling
std::atoi.

Change-Id: Ie479a481f8d23f034b425d14e3cfefb3d62c84e8


[ROCm/roctracer commit: a7cac4b4c9]
2022-07-15 01:06:46 -07:00
Laurent Morichetti a7cac4b4c9 Fix a SEGV in std::atoi
When ROCP_TRUNCATE_NAMES is not set, getenv returns NULL and std::atoi
crashes. Check that getenv returns a non-NULL string before calling
std::atoi.

Change-Id: Ie479a481f8d23f034b425d14e3cfefb3d62c84e8
2022-07-15 01:06:46 -07:00
Chauncey Hui a78bdbd384 SWDEV-2 - Change OpenCL version number from 3485 to 3486
[ROCm/clr commit: f8627e6121]
2022-07-15 03:02:41 -04:00
Chauncey Hui f8627e6121 SWDEV-2 - Change OpenCL version number from 3485 to 3486 2022-07-15 03:02:41 -04:00
Laurent Morichetti 71469dfc97 Split hsa_prof_str.h
Split the public and private HSA profiler/tracer interfaces. Only the
public interface should be exposed in include/roctracer.

Change-Id: I7e4424cd90023693350c31e6b02caca8c984ba84


[ROCm/roctracer commit: 96bcac0f47]
2022-07-14 21:20:30 -04:00
Laurent Morichetti 96bcac0f47 Split hsa_prof_str.h
Split the public and private HSA profiler/tracer interfaces. Only the
public interface should be exposed in include/roctracer.

Change-Id: I7e4424cd90023693350c31e6b02caca8c984ba84
2022-07-14 21:20:30 -04:00
Ranjith Ramakrishnan e7eb195924 SWDEV-345870 - Correct include paths for new directory layout
Use hsa header files from /opt/rocm-ver/include rather than using wrapper files from /opt/rocm-ver/hsa/include/hsa

Change-Id: Id7a9bde19447cd2a0fd6e03b11c08471f09c2a46
2022-07-14 16:08:41 -07:00
Ranjith Ramakrishnan a5a941cbfc SWDEV-345870 - Correct include paths for new directory layout
Use hsa header files from /opt/rocm-ver/include rather than using wrapper files from /opt/rocm-ver/hsa/include/hsa

Change-Id: Id7a9bde19447cd2a0fd6e03b11c08471f09c2a46


[ROCm/rocprofiler commit: e7eb195924]
2022-07-14 16:08:41 -07:00
sdashmiz 3389e6077a SWDEV-334233 - add support for p2p in windows
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I9109120b5444c400e65cfff869cb36e876ffd1fc


[ROCm/clr commit: e176e27bf7]
2022-07-14 15:07:33 -04:00
sdashmiz e176e27bf7 SWDEV-334233 - add support for p2p in windows
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I9109120b5444c400e65cfff869cb36e876ffd1fc
2022-07-14 15:07:33 -04:00
Jeremy Newton aa25cb1acc Fix numa linking
We should link against numa without hardcoding the path to it.
CMake should determine how to link numa automatically, similar to how rt
and pthread is linked.

Fixes 

Change-Id: Ifb9ac30e200c66cbd7f1cf80d25fffef1dcf8d2f
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
2022-07-14 11:04:09 -04:00
Jeremy Newton 7ab88b110f Fix numa linking
We should link against numa without hardcoding the path to it.
CMake should determine how to link numa automatically, similar to how rt
and pthread is linked.

Fixes 

Change-Id: Ifb9ac30e200c66cbd7f1cf80d25fffef1dcf8d2f
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>


[ROCm/ROCR-Runtime commit: aa25cb1acc]
2022-07-14 11:04:09 -04:00
Brian Sumner a369b4730b SWDEV-333033 - add safe and unsafe atomic min and max including gfx940 and add missing nvidia support
Change-Id: I829a0a5fd49c510e77eabbcb92d1a415ef6b5a4c


[ROCm/hipother commit: 13d1a1b0e4]
2022-07-14 10:22:42 -04:00
Brian Sumner 13d1a1b0e4 SWDEV-333033 - add safe and unsafe atomic min and max including gfx940 and add missing nvidia support
Change-Id: I829a0a5fd49c510e77eabbcb92d1a415ef6b5a4c
2022-07-14 10:22:42 -04:00
Brian Sumner 5a402b9c99 SWDEV-333033 - add safe and unsafe atomic min and max including gfx940 and add missing nvidia support
Change-Id: I829a0a5fd49c510e77eabbcb92d1a415ef6b5a4c


[ROCm/clr commit: 4375b9f5b9]
2022-07-14 10:22:42 -04:00
Brian Sumner 4375b9f5b9 SWDEV-333033 - add safe and unsafe atomic min and max including gfx940 and add missing nvidia support
Change-Id: I829a0a5fd49c510e77eabbcb92d1a415ef6b5a4c
2022-07-14 10:22:42 -04:00
Chauncey Hui f69b6922bb SWDEV-2 - Change OpenCL version number from 3484 to 3485
[ROCm/clr commit: 782e0f2a8d]
2022-07-14 03:00:05 -04:00
Chauncey Hui 782e0f2a8d SWDEV-2 - Change OpenCL version number from 3484 to 3485 2022-07-14 03:00:05 -04:00
Maneesh Gupta 782d0b3628 SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: Ibe710eba031ddce646fdae4593615174ee717a29


[ROCm/hip commit: 7cfde0047f]
2022-07-14 05:57:27 +00:00
Maneesh Gupta 7cfde0047f SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: Ibe710eba031ddce646fdae4593615174ee717a29
2022-07-14 05:57:27 +00:00
Maneesh Gupta 9174393aa4 Disable Unit_hipMemVmm_Basic on windows for now
[ROCm/hip-tests commit: 8c5c2991ba]
2022-07-14 11:25:25 +05:30
Maneesh Gupta 8c5c2991ba Disable Unit_hipMemVmm_Basic on windows for now 2022-07-14 11:25:25 +05:30
Maneesh Gupta 551f9b4f76 Disable Unit_hipMemVmm_Basic on windows for now
[ROCm/hip commit: 9e84a11f23]
2022-07-14 11:25:25 +05:30
Maneesh Gupta 9e84a11f23 Disable Unit_hipMemVmm_Basic on windows for now 2022-07-14 11:25:25 +05:30
Jaydeep Patel 40fda3cfe9 SWDEV-343168 - Clean ihipMallocPitch
Change-Id: Iab1c10ac2aea635ca10cc8bc98141df409ec917c


[ROCm/clr commit: 6b3b62f49e]
2022-07-14 01:15:08 -04:00
Jaydeep Patel 6b3b62f49e SWDEV-343168 - Clean ihipMallocPitch
Change-Id: Iab1c10ac2aea635ca10cc8bc98141df409ec917c
2022-07-14 01:15:08 -04:00
Sarbojit Sarkar 06c0775ca6 SWDEV-345741 - hipDeviceSetLimit windows build error
Change-Id: I384541b190f1eb9e045553299215345426c5adc2


[ROCm/clr commit: a4d4289382]
2022-07-13 20:37:32 -04:00
Sarbojit Sarkar a4d4289382 SWDEV-345741 - hipDeviceSetLimit windows build error
Change-Id: I384541b190f1eb9e045553299215345426c5adc2
2022-07-13 20:37:32 -04:00