Skip to content
[Jenkins]Jenkins
log in
Failed

Changes

Summary

  1. Introduce constructor for multi-GPU support. (details)
  2. Fix typo (details)
  3. Add test (details)
  4. Fix typo. (details)
  5. Explicitly check for valid device id (details)
  6. Set the device id in cuda_kernel_arch (details)
  7. Check for default device (details)
  8. Check that device associated with stream matches requested device (details)
  9. Remove extra constructor (details)
  10. Address reviewer comments (details)
  11. m_cudaDev isn't static anymore (details)
  12. Set the device id explicitly for CUDA API calls in impl_initialize (details)
  13. Fixup test math functions ulp should double -> int (details)
  14. Drop DualView converting copy assignment operator (details)
  15. Don't use rocm-docker for clang-format (details)
  16. Diable HIP CI (details)
  17. Do not negate the dependent true traits helper (details)
  18. Add missing gfx940 (details)
  19. Add Impl::always_false type-dendent false trait (details)
  20. Per review prefer always_false<Arg>::value to is_void_v<Arg> (details)
  21. Improve "no copy mechanism" exception message (details)
  22. Add a unit test for new deep_copy exception msg (details)
  23. Add missing include sstream (details)
  24. src->source, dst->destination (details)
  25. Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support (details)
  26. SYCL: Force inlining of Kokkos::printf (#6650) (details)
  27. Improve handling of printf in OMPT on Intel GPUs (details)
  28. OpenMP: Use `omp_get_nested` for older gcc versions (#6685) (details)
  29. Drop unnecessary guarding for a tool library being loaded in ProfilingSection (details)
  30. Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp (details)
Commit b8482317e26c3ae0eb421eb4b5b95fe0d16a0743 by Daniel Arndt
Introduce constructor for multi-GPU support.
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 64a9b3d854521b7b6dabf7fd9e162c959045b2e3 by noreply
Fix typo

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was addedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/unit_test/CMakeLists.txt (diff)
Commit 1683786110806f22fb14b2c9e065ee8350cec8ac by noreply
Fix typo.

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp (diff)
Commit fa1aaa712a36394c5c1f1508771cbf13160353b7 by Daniel Arndt
Explicitly check for valid device id
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 41253bd55d7fad8c72f5f3756792705b66bdc954 by Daniel Arndt
Set the device id in cuda_kernel_arch
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit f6977cf431cf245341e6dd2f744c79c5a62a315e by Daniel Arndt
Check for default device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit e156d5859e36fbc3c4f84575a0c8734689c1b3eb by Daniel Arndt
Check that device associated with stream matches requested device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 1fcce6936bf667339350381c8b265b5d82d810da by Daniel Arndt
Remove extra constructor
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was removedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit a07c7a2b6c9999020859a7de86004e447790d0c4 by Daniel Arndt
Address reviewer comments
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 403c34f304077eeb92dc8fd8cb2fd24ae6c35688 by Daniel Arndt
m_cudaDev isn't static anymore
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit d4a517f82d73a780d03ab06ece627b47d7aad43f by Daniel Arndt
Set the device id explicitly for CUDA API calls in impl_initialize
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 71729af719b2eb20b512efad05ae609e74e1dd7a by Damien L-G
Fixup test math functions ulp should double -> int
The file was modified core/unit_test/TestMathematicalFunctions.hpp (diff)
Commit b9b63dfd89fb670d9e18309ad25ef74c6bdd9400 by Damien L-G
Drop DualView converting copy assignment operator

It is unnecessary, we already have a converting constructor and a
regular copy assignment operator.
The file was modified containers/src/Kokkos_DualView.hpp (diff)
Commit 9fd95ebcb1d2f590d376e42ca993068138c0c829 by Daniel Arndt
Don't use rocm-docker for clang-format
The file was modified .jenkins (diff)
The file was modified .jenkins (diff)
Commit 76ea3a3a95e4619b80e3adf7edda6db05d4ed32d by Damien L-G
Do not negate the dependent true traits helper
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified cmake/KokkosCore_config.h.in (diff)
The file was modified cmake/kokkos_arch.cmake (diff)
The file was modified core/src/HIP/Kokkos_HIP_Instance.hpp (diff)
The file was modified generate_makefile.bash (diff)
The file was modified Makefile.kokkos (diff)
Commit 33db3046a173b5e96d1dff62c8555a8da5cea9bb by Damien L-G
Add Impl::always_false type-dendent false trait
The file was modified core/src/impl/Kokkos_Utilities.hpp (diff)
Commit e4a7cfc78f0564f2aefb8a98cab3144baffd15b6 by Damien L-G
Per review prefer always_false<Arg>::value to is_void_v<Arg>
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
Commit 316ceac58eedf97fec5ee9f2fb87018a83fb97ec by bmkelle
Improve "no copy mechanism" exception message
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 838f8938e9aaf1ac78027575743bb95b442470a9 by bmkelle
Add a unit test for new deep_copy exception msg
The file was modified core/unit_test/TestViewCopy_a.hpp (diff)
Commit 72bc7ed42f0f658ebc3b158520feba086111fc3b by bmkelle
Add missing include sstream
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 154a57df8c53bff88dc4806967cb8a49d8ed1c40 by bmkelle
src->source, dst->destination

and reformat a bit
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 5df22b87b046290e84c1c6a2d4d3bcbbc47d1e5c by Bruno Turcksin
Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support
The file was modified simd/src/Kokkos_SIMD_AVX2.hpp (diff)
Commit 52e44d6cfe35dcfc55c3d91ddff70cc63f8d514c by noreply
SYCL: Force inlining of Kokkos::printf (#6650)

* SYCL: Force inlining of Kokkos::printf

* Always force inline Kokkos::printf
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit 79164a43adc5da6e3eee12f3cfc66e3e1402ed3f by Daniel Arndt
Improve handling of printf in OMPT on Intel GPUs
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit cbbe09b93db49e081ab4d400975aca8c007c4223 by noreply
OpenMP: Use `omp_get_nested` for older gcc versions (#6685)

* OpenMP: fix for issue 6670.

* OpenMP: Update gcc version when using max_active_level.

* OpenMP: Edit execute_in_serial.

* Apply suggestions from code review

Co-authored-by: Daniel Arndt <arndtd@ornl.gov>

* clang-format

---------

Co-authored-by: Rahulkumar Gayatri <rgayatri@lbl.gov>
Co-authored-by: Damien L-G <dalg24+github@gmail.com>
Co-authored-by: Daniel Arndt <arndtd@ornl.gov>
The file was modified core/src/OpenMP/Kokkos_OpenMP_Instance.hpp (diff)
Commit 5aa0ceee4b0132ec411f7fc1e11b4b9a1039cde3 by Damien L-G
Drop unnecessary guarding for a tool library being loaded in ProfilingSection
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)
Commit 73c75075536f905818560f0bc415348b462fe38f by Damien L-G
Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)

Summary

  1. Introduce constructor for multi-GPU support. (details)
  2. Fix typo (details)
  3. Add test (details)
  4. Fix typo. (details)
  5. Explicitly check for valid device id (details)
  6. Set the device id in cuda_kernel_arch (details)
  7. Check for default device (details)
  8. Check that device associated with stream matches requested device (details)
  9. Remove extra constructor (details)
  10. Address reviewer comments (details)
  11. m_cudaDev isn't static anymore (details)
  12. Set the device id explicitly for CUDA API calls in impl_initialize (details)
  13. Fixup test math functions ulp should double -> int (details)
  14. Drop DualView converting copy assignment operator (details)
  15. Don't use rocm-docker for clang-format (details)
  16. Diable HIP CI (details)
  17. Do not negate the dependent true traits helper (details)
  18. Add missing gfx940 (details)
  19. Add Impl::always_false type-dendent false trait (details)
  20. Per review prefer always_false<Arg>::value to is_void_v<Arg> (details)
  21. Improve "no copy mechanism" exception message (details)
  22. Add a unit test for new deep_copy exception msg (details)
  23. Add missing include sstream (details)
  24. src->source, dst->destination (details)
  25. Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support (details)
  26. SYCL: Force inlining of Kokkos::printf (#6650) (details)
  27. Improve handling of printf in OMPT on Intel GPUs (details)
  28. OpenMP: Use `omp_get_nested` for older gcc versions (#6685) (details)
  29. Drop unnecessary guarding for a tool library being loaded in ProfilingSection (details)
  30. Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp (details)
Commit b8482317e26c3ae0eb421eb4b5b95fe0d16a0743 by Daniel Arndt
Introduce constructor for multi-GPU support.
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 64a9b3d854521b7b6dabf7fd9e162c959045b2e3 by noreply
Fix typo

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was addedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
Commit 1683786110806f22fb14b2c9e065ee8350cec8ac by noreply
Fix typo.

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp (diff)
Commit fa1aaa712a36394c5c1f1508771cbf13160353b7 by Daniel Arndt
Explicitly check for valid device id
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 41253bd55d7fad8c72f5f3756792705b66bdc954 by Daniel Arndt
Set the device id in cuda_kernel_arch
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit f6977cf431cf245341e6dd2f744c79c5a62a315e by Daniel Arndt
Check for default device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit e156d5859e36fbc3c4f84575a0c8734689c1b3eb by Daniel Arndt
Check that device associated with stream matches requested device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 1fcce6936bf667339350381c8b265b5d82d810da by Daniel Arndt
Remove extra constructor
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was removedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit a07c7a2b6c9999020859a7de86004e447790d0c4 by Daniel Arndt
Address reviewer comments
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 403c34f304077eeb92dc8fd8cb2fd24ae6c35688 by Daniel Arndt
m_cudaDev isn't static anymore
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit d4a517f82d73a780d03ab06ece627b47d7aad43f by Daniel Arndt
Set the device id explicitly for CUDA API calls in impl_initialize
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 71729af719b2eb20b512efad05ae609e74e1dd7a by Damien L-G
Fixup test math functions ulp should double -> int
The file was modified core/unit_test/TestMathematicalFunctions.hpp (diff)
Commit b9b63dfd89fb670d9e18309ad25ef74c6bdd9400 by Damien L-G
Drop DualView converting copy assignment operator

It is unnecessary, we already have a converting constructor and a
regular copy assignment operator.
The file was modified containers/src/Kokkos_DualView.hpp (diff)
Commit 9fd95ebcb1d2f590d376e42ca993068138c0c829 by Daniel Arndt
Don't use rocm-docker for clang-format
The file was modified .jenkins (diff)
The file was modified .jenkins (diff)
Commit 76ea3a3a95e4619b80e3adf7edda6db05d4ed32d by Damien L-G
Do not negate the dependent true traits helper
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified Makefile.kokkos (diff)
The file was modified generate_makefile.bash (diff)
The file was modified cmake/kokkos_arch.cmake (diff)
The file was modified core/src/HIP/Kokkos_HIP_Instance.hpp (diff)
The file was modified cmake/KokkosCore_config.h.in (diff)
Commit 33db3046a173b5e96d1dff62c8555a8da5cea9bb by Damien L-G
Add Impl::always_false type-dendent false trait
The file was modified core/src/impl/Kokkos_Utilities.hpp (diff)
Commit e4a7cfc78f0564f2aefb8a98cab3144baffd15b6 by Damien L-G
Per review prefer always_false<Arg>::value to is_void_v<Arg>
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
Commit 316ceac58eedf97fec5ee9f2fb87018a83fb97ec by bmkelle
Improve "no copy mechanism" exception message
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 838f8938e9aaf1ac78027575743bb95b442470a9 by bmkelle
Add a unit test for new deep_copy exception msg
The file was modified core/unit_test/TestViewCopy_a.hpp (diff)
Commit 72bc7ed42f0f658ebc3b158520feba086111fc3b by bmkelle
Add missing include sstream
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 154a57df8c53bff88dc4806967cb8a49d8ed1c40 by bmkelle
src->source, dst->destination

and reformat a bit
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 5df22b87b046290e84c1c6a2d4d3bcbbc47d1e5c by Bruno Turcksin
Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support
The file was modified simd/src/Kokkos_SIMD_AVX2.hpp (diff)
Commit 52e44d6cfe35dcfc55c3d91ddff70cc63f8d514c by noreply
SYCL: Force inlining of Kokkos::printf (#6650)

* SYCL: Force inlining of Kokkos::printf

* Always force inline Kokkos::printf
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit 79164a43adc5da6e3eee12f3cfc66e3e1402ed3f by Daniel Arndt
Improve handling of printf in OMPT on Intel GPUs
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit cbbe09b93db49e081ab4d400975aca8c007c4223 by noreply
OpenMP: Use `omp_get_nested` for older gcc versions (#6685)

* OpenMP: fix for issue 6670.

* OpenMP: Update gcc version when using max_active_level.

* OpenMP: Edit execute_in_serial.

* Apply suggestions from code review

Co-authored-by: Daniel Arndt <arndtd@ornl.gov>

* clang-format

---------

Co-authored-by: Rahulkumar Gayatri <rgayatri@lbl.gov>
Co-authored-by: Damien L-G <dalg24+github@gmail.com>
Co-authored-by: Daniel Arndt <arndtd@ornl.gov>
The file was modified core/src/OpenMP/Kokkos_OpenMP_Instance.hpp (diff)
Commit 5aa0ceee4b0132ec411f7fc1e11b4b9a1039cde3 by Damien L-G
Drop unnecessary guarding for a tool library being loaded in ProfilingSection
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)
Commit 73c75075536f905818560f0bc415348b462fe38f by Damien L-G
Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)

Summary

  1. Introduce constructor for multi-GPU support. (details)
  2. Fix typo (details)
  3. Add test (details)
  4. Fix typo. (details)
  5. Explicitly check for valid device id (details)
  6. Set the device id in cuda_kernel_arch (details)
  7. Check for default device (details)
  8. Check that device associated with stream matches requested device (details)
  9. Remove extra constructor (details)
  10. Address reviewer comments (details)
  11. m_cudaDev isn't static anymore (details)
  12. Set the device id explicitly for CUDA API calls in impl_initialize (details)
  13. Fixup test math functions ulp should double -> int (details)
  14. Drop DualView converting copy assignment operator (details)
  15. Don't use rocm-docker for clang-format (details)
  16. Diable HIP CI (details)
  17. Do not negate the dependent true traits helper (details)
  18. Add missing gfx940 (details)
  19. Add Impl::always_false type-dendent false trait (details)
  20. Per review prefer always_false<Arg>::value to is_void_v<Arg> (details)
  21. Improve "no copy mechanism" exception message (details)
  22. Add a unit test for new deep_copy exception msg (details)
  23. Add missing include sstream (details)
  24. src->source, dst->destination (details)
  25. Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support (details)
  26. SYCL: Force inlining of Kokkos::printf (#6650) (details)
  27. Improve handling of printf in OMPT on Intel GPUs (details)
  28. OpenMP: Use `omp_get_nested` for older gcc versions (#6685) (details)
  29. Drop unnecessary guarding for a tool library being loaded in ProfilingSection (details)
  30. Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp (details)
Commit b8482317e26c3ae0eb421eb4b5b95fe0d16a0743 by Daniel Arndt
Introduce constructor for multi-GPU support.
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
Commit 64a9b3d854521b7b6dabf7fd9e162c959045b2e3 by noreply
Fix typo

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was addedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
Commit 1683786110806f22fb14b2c9e065ee8350cec8ac by noreply
Fix typo.

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp (diff)
Commit fa1aaa712a36394c5c1f1508771cbf13160353b7 by Daniel Arndt
Explicitly check for valid device id
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 41253bd55d7fad8c72f5f3756792705b66bdc954 by Daniel Arndt
Set the device id in cuda_kernel_arch
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit f6977cf431cf245341e6dd2f744c79c5a62a315e by Daniel Arndt
Check for default device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit e156d5859e36fbc3c4f84575a0c8734689c1b3eb by Daniel Arndt
Check that device associated with stream matches requested device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 1fcce6936bf667339350381c8b265b5d82d810da by Daniel Arndt
Remove extra constructor
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was removedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
Commit a07c7a2b6c9999020859a7de86004e447790d0c4 by Daniel Arndt
Address reviewer comments
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 403c34f304077eeb92dc8fd8cb2fd24ae6c35688 by Daniel Arndt
m_cudaDev isn't static anymore
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit d4a517f82d73a780d03ab06ece627b47d7aad43f by Daniel Arndt
Set the device id explicitly for CUDA API calls in impl_initialize
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 71729af719b2eb20b512efad05ae609e74e1dd7a by Damien L-G
Fixup test math functions ulp should double -> int
The file was modified core/unit_test/TestMathematicalFunctions.hpp (diff)
Commit b9b63dfd89fb670d9e18309ad25ef74c6bdd9400 by Damien L-G
Drop DualView converting copy assignment operator

It is unnecessary, we already have a converting constructor and a
regular copy assignment operator.
The file was modified containers/src/Kokkos_DualView.hpp (diff)
Commit 9fd95ebcb1d2f590d376e42ca993068138c0c829 by Daniel Arndt
Don't use rocm-docker for clang-format
The file was modified .jenkins (diff)
The file was modified .jenkins (diff)
Commit 76ea3a3a95e4619b80e3adf7edda6db05d4ed32d by Damien L-G
Do not negate the dependent true traits helper
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified cmake/kokkos_arch.cmake (diff)
The file was modified Makefile.kokkos (diff)
The file was modified cmake/KokkosCore_config.h.in (diff)
The file was modified core/src/HIP/Kokkos_HIP_Instance.hpp (diff)
The file was modified generate_makefile.bash (diff)
Commit 33db3046a173b5e96d1dff62c8555a8da5cea9bb by Damien L-G
Add Impl::always_false type-dendent false trait
The file was modified core/src/impl/Kokkos_Utilities.hpp (diff)
Commit e4a7cfc78f0564f2aefb8a98cab3144baffd15b6 by Damien L-G
Per review prefer always_false<Arg>::value to is_void_v<Arg>
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
Commit 316ceac58eedf97fec5ee9f2fb87018a83fb97ec by bmkelle
Improve "no copy mechanism" exception message
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 838f8938e9aaf1ac78027575743bb95b442470a9 by bmkelle
Add a unit test for new deep_copy exception msg
The file was modified core/unit_test/TestViewCopy_a.hpp (diff)
Commit 72bc7ed42f0f658ebc3b158520feba086111fc3b by bmkelle
Add missing include sstream
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 154a57df8c53bff88dc4806967cb8a49d8ed1c40 by bmkelle
src->source, dst->destination

and reformat a bit
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 5df22b87b046290e84c1c6a2d4d3bcbbc47d1e5c by Bruno Turcksin
Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support
The file was modified simd/src/Kokkos_SIMD_AVX2.hpp (diff)
Commit 52e44d6cfe35dcfc55c3d91ddff70cc63f8d514c by noreply
SYCL: Force inlining of Kokkos::printf (#6650)

* SYCL: Force inlining of Kokkos::printf

* Always force inline Kokkos::printf
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit 79164a43adc5da6e3eee12f3cfc66e3e1402ed3f by Daniel Arndt
Improve handling of printf in OMPT on Intel GPUs
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit cbbe09b93db49e081ab4d400975aca8c007c4223 by noreply
OpenMP: Use `omp_get_nested` for older gcc versions (#6685)

* OpenMP: fix for issue 6670.

* OpenMP: Update gcc version when using max_active_level.

* OpenMP: Edit execute_in_serial.

* Apply suggestions from code review

Co-authored-by: Daniel Arndt <arndtd@ornl.gov>

* clang-format

---------

Co-authored-by: Rahulkumar Gayatri <rgayatri@lbl.gov>
Co-authored-by: Damien L-G <dalg24+github@gmail.com>
Co-authored-by: Daniel Arndt <arndtd@ornl.gov>
The file was modified core/src/OpenMP/Kokkos_OpenMP_Instance.hpp (diff)
Commit 5aa0ceee4b0132ec411f7fc1e11b4b9a1039cde3 by Damien L-G
Drop unnecessary guarding for a tool library being loaded in ProfilingSection
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)
Commit 73c75075536f905818560f0bc415348b462fe38f by Damien L-G
Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)

Summary

  1. Introduce constructor for multi-GPU support. (details)
  2. Fix typo (details)
  3. Add test (details)
  4. Fix typo. (details)
  5. Explicitly check for valid device id (details)
  6. Set the device id in cuda_kernel_arch (details)
  7. Check for default device (details)
  8. Check that device associated with stream matches requested device (details)
  9. Remove extra constructor (details)
  10. Address reviewer comments (details)
  11. m_cudaDev isn't static anymore (details)
  12. Set the device id explicitly for CUDA API calls in impl_initialize (details)
  13. Fixup test math functions ulp should double -> int (details)
  14. Drop DualView converting copy assignment operator (details)
  15. Don't use rocm-docker for clang-format (details)
  16. Diable HIP CI (details)
  17. Do not negate the dependent true traits helper (details)
  18. Add missing gfx940 (details)
  19. Add Impl::always_false type-dendent false trait (details)
  20. Per review prefer always_false<Arg>::value to is_void_v<Arg> (details)
  21. Improve "no copy mechanism" exception message (details)
  22. Add a unit test for new deep_copy exception msg (details)
  23. Add missing include sstream (details)
  24. src->source, dst->destination (details)
  25. Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support (details)
  26. SYCL: Force inlining of Kokkos::printf (#6650) (details)
  27. Improve handling of printf in OMPT on Intel GPUs (details)
  28. OpenMP: Use `omp_get_nested` for older gcc versions (#6685) (details)
  29. Drop unnecessary guarding for a tool library being loaded in ProfilingSection (details)
  30. Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp (details)
Commit b8482317e26c3ae0eb421eb4b5b95fe0d16a0743 by Daniel Arndt
Introduce constructor for multi-GPU support.
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
Commit 64a9b3d854521b7b6dabf7fd9e162c959045b2e3 by noreply
Fix typo

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was addedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
Commit 1683786110806f22fb14b2c9e065ee8350cec8ac by noreply
Fix typo.

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp (diff)
Commit fa1aaa712a36394c5c1f1508771cbf13160353b7 by Daniel Arndt
Explicitly check for valid device id
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 41253bd55d7fad8c72f5f3756792705b66bdc954 by Daniel Arndt
Set the device id in cuda_kernel_arch
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit f6977cf431cf245341e6dd2f744c79c5a62a315e by Daniel Arndt
Check for default device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit e156d5859e36fbc3c4f84575a0c8734689c1b3eb by Daniel Arndt
Check that device associated with stream matches requested device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 1fcce6936bf667339350381c8b265b5d82d810da by Daniel Arndt
Remove extra constructor
The file was removedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
Commit a07c7a2b6c9999020859a7de86004e447790d0c4 by Daniel Arndt
Address reviewer comments
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 403c34f304077eeb92dc8fd8cb2fd24ae6c35688 by Daniel Arndt
m_cudaDev isn't static anymore
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit d4a517f82d73a780d03ab06ece627b47d7aad43f by Daniel Arndt
Set the device id explicitly for CUDA API calls in impl_initialize
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 71729af719b2eb20b512efad05ae609e74e1dd7a by Damien L-G
Fixup test math functions ulp should double -> int
The file was modified core/unit_test/TestMathematicalFunctions.hpp (diff)
Commit b9b63dfd89fb670d9e18309ad25ef74c6bdd9400 by Damien L-G
Drop DualView converting copy assignment operator

It is unnecessary, we already have a converting constructor and a
regular copy assignment operator.
The file was modified containers/src/Kokkos_DualView.hpp (diff)
Commit 9fd95ebcb1d2f590d376e42ca993068138c0c829 by Daniel Arndt
Don't use rocm-docker for clang-format
The file was modified .jenkins (diff)
The file was modified .jenkins (diff)
Commit 76ea3a3a95e4619b80e3adf7edda6db05d4ed32d by Damien L-G
Do not negate the dependent true traits helper
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified Makefile.kokkos (diff)
The file was modified core/src/HIP/Kokkos_HIP_Instance.hpp (diff)
The file was modified cmake/KokkosCore_config.h.in (diff)
The file was modified cmake/kokkos_arch.cmake (diff)
The file was modified generate_makefile.bash (diff)
Commit 33db3046a173b5e96d1dff62c8555a8da5cea9bb by Damien L-G
Add Impl::always_false type-dendent false trait
The file was modified core/src/impl/Kokkos_Utilities.hpp (diff)
Commit e4a7cfc78f0564f2aefb8a98cab3144baffd15b6 by Damien L-G
Per review prefer always_false<Arg>::value to is_void_v<Arg>
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
Commit 316ceac58eedf97fec5ee9f2fb87018a83fb97ec by bmkelle
Improve "no copy mechanism" exception message
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 838f8938e9aaf1ac78027575743bb95b442470a9 by bmkelle
Add a unit test for new deep_copy exception msg
The file was modified core/unit_test/TestViewCopy_a.hpp (diff)
Commit 72bc7ed42f0f658ebc3b158520feba086111fc3b by bmkelle
Add missing include sstream
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 154a57df8c53bff88dc4806967cb8a49d8ed1c40 by bmkelle
src->source, dst->destination

and reformat a bit
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 5df22b87b046290e84c1c6a2d4d3bcbbc47d1e5c by Bruno Turcksin
Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support
The file was modified simd/src/Kokkos_SIMD_AVX2.hpp (diff)
Commit 52e44d6cfe35dcfc55c3d91ddff70cc63f8d514c by noreply
SYCL: Force inlining of Kokkos::printf (#6650)

* SYCL: Force inlining of Kokkos::printf

* Always force inline Kokkos::printf
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit 79164a43adc5da6e3eee12f3cfc66e3e1402ed3f by Daniel Arndt
Improve handling of printf in OMPT on Intel GPUs
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit cbbe09b93db49e081ab4d400975aca8c007c4223 by noreply
OpenMP: Use `omp_get_nested` for older gcc versions (#6685)

* OpenMP: fix for issue 6670.

* OpenMP: Update gcc version when using max_active_level.

* OpenMP: Edit execute_in_serial.

* Apply suggestions from code review

Co-authored-by: Daniel Arndt <arndtd@ornl.gov>

* clang-format

---------

Co-authored-by: Rahulkumar Gayatri <rgayatri@lbl.gov>
Co-authored-by: Damien L-G <dalg24+github@gmail.com>
Co-authored-by: Daniel Arndt <arndtd@ornl.gov>
The file was modified core/src/OpenMP/Kokkos_OpenMP_Instance.hpp (diff)
Commit 5aa0ceee4b0132ec411f7fc1e11b4b9a1039cde3 by Damien L-G
Drop unnecessary guarding for a tool library being loaded in ProfilingSection
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)
Commit 73c75075536f905818560f0bc415348b462fe38f by Damien L-G
Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)

Summary

  1. Introduce constructor for multi-GPU support. (details)
  2. Fix typo (details)
  3. Add test (details)
  4. Fix typo. (details)
  5. Explicitly check for valid device id (details)
  6. Set the device id in cuda_kernel_arch (details)
  7. Check for default device (details)
  8. Check that device associated with stream matches requested device (details)
  9. Remove extra constructor (details)
  10. Address reviewer comments (details)
  11. m_cudaDev isn't static anymore (details)
  12. Set the device id explicitly for CUDA API calls in impl_initialize (details)
  13. Fixup test math functions ulp should double -> int (details)
  14. Drop DualView converting copy assignment operator (details)
  15. Don't use rocm-docker for clang-format (details)
  16. Diable HIP CI (details)
  17. Do not negate the dependent true traits helper (details)
  18. Add missing gfx940 (details)
  19. Add Impl::always_false type-dendent false trait (details)
  20. Per review prefer always_false<Arg>::value to is_void_v<Arg> (details)
  21. Improve "no copy mechanism" exception message (details)
  22. Add a unit test for new deep_copy exception msg (details)
  23. Add missing include sstream (details)
  24. src->source, dst->destination (details)
  25. Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support (details)
  26. SYCL: Force inlining of Kokkos::printf (#6650) (details)
  27. Improve handling of printf in OMPT on Intel GPUs (details)
  28. OpenMP: Use `omp_get_nested` for older gcc versions (#6685) (details)
  29. Drop unnecessary guarding for a tool library being loaded in ProfilingSection (details)
  30. Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp (details)
Commit b8482317e26c3ae0eb421eb4b5b95fe0d16a0743 by Daniel Arndt
Introduce constructor for multi-GPU support.
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
Commit 64a9b3d854521b7b6dabf7fd9e162c959045b2e3 by noreply
Fix typo

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was addedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
Commit 1683786110806f22fb14b2c9e065ee8350cec8ac by noreply
Fix typo.

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp (diff)
Commit fa1aaa712a36394c5c1f1508771cbf13160353b7 by Daniel Arndt
Explicitly check for valid device id
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 41253bd55d7fad8c72f5f3756792705b66bdc954 by Daniel Arndt
Set the device id in cuda_kernel_arch
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit f6977cf431cf245341e6dd2f744c79c5a62a315e by Daniel Arndt
Check for default device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit e156d5859e36fbc3c4f84575a0c8734689c1b3eb by Daniel Arndt
Check that device associated with stream matches requested device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 1fcce6936bf667339350381c8b265b5d82d810da by Daniel Arndt
Remove extra constructor
The file was removedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
Commit a07c7a2b6c9999020859a7de86004e447790d0c4 by Daniel Arndt
Address reviewer comments
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 403c34f304077eeb92dc8fd8cb2fd24ae6c35688 by Daniel Arndt
m_cudaDev isn't static anymore
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit d4a517f82d73a780d03ab06ece627b47d7aad43f by Daniel Arndt
Set the device id explicitly for CUDA API calls in impl_initialize
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 71729af719b2eb20b512efad05ae609e74e1dd7a by Damien L-G
Fixup test math functions ulp should double -> int
The file was modified core/unit_test/TestMathematicalFunctions.hpp (diff)
Commit b9b63dfd89fb670d9e18309ad25ef74c6bdd9400 by Damien L-G
Drop DualView converting copy assignment operator

It is unnecessary, we already have a converting constructor and a
regular copy assignment operator.
The file was modified containers/src/Kokkos_DualView.hpp (diff)
Commit 9fd95ebcb1d2f590d376e42ca993068138c0c829 by Daniel Arndt
Don't use rocm-docker for clang-format
The file was modified .jenkins (diff)
The file was modified .jenkins (diff)
Commit 76ea3a3a95e4619b80e3adf7edda6db05d4ed32d by Damien L-G
Do not negate the dependent true traits helper
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified cmake/KokkosCore_config.h.in (diff)
The file was modified Makefile.kokkos (diff)
The file was modified generate_makefile.bash (diff)
The file was modified cmake/kokkos_arch.cmake (diff)
The file was modified core/src/HIP/Kokkos_HIP_Instance.hpp (diff)
Commit 33db3046a173b5e96d1dff62c8555a8da5cea9bb by Damien L-G
Add Impl::always_false type-dendent false trait
The file was modified core/src/impl/Kokkos_Utilities.hpp (diff)
Commit e4a7cfc78f0564f2aefb8a98cab3144baffd15b6 by Damien L-G
Per review prefer always_false<Arg>::value to is_void_v<Arg>
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
Commit 316ceac58eedf97fec5ee9f2fb87018a83fb97ec by bmkelle
Improve "no copy mechanism" exception message
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 838f8938e9aaf1ac78027575743bb95b442470a9 by bmkelle
Add a unit test for new deep_copy exception msg
The file was modified core/unit_test/TestViewCopy_a.hpp (diff)
Commit 72bc7ed42f0f658ebc3b158520feba086111fc3b by bmkelle
Add missing include sstream
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 154a57df8c53bff88dc4806967cb8a49d8ed1c40 by bmkelle
src->source, dst->destination

and reformat a bit
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 5df22b87b046290e84c1c6a2d4d3bcbbc47d1e5c by Bruno Turcksin
Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support
The file was modified simd/src/Kokkos_SIMD_AVX2.hpp (diff)
Commit 52e44d6cfe35dcfc55c3d91ddff70cc63f8d514c by noreply
SYCL: Force inlining of Kokkos::printf (#6650)

* SYCL: Force inlining of Kokkos::printf

* Always force inline Kokkos::printf
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit 79164a43adc5da6e3eee12f3cfc66e3e1402ed3f by Daniel Arndt
Improve handling of printf in OMPT on Intel GPUs
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit cbbe09b93db49e081ab4d400975aca8c007c4223 by noreply
OpenMP: Use `omp_get_nested` for older gcc versions (#6685)

* OpenMP: fix for issue 6670.

* OpenMP: Update gcc version when using max_active_level.

* OpenMP: Edit execute_in_serial.

* Apply suggestions from code review

Co-authored-by: Daniel Arndt <arndtd@ornl.gov>

* clang-format

---------

Co-authored-by: Rahulkumar Gayatri <rgayatri@lbl.gov>
Co-authored-by: Damien L-G <dalg24+github@gmail.com>
Co-authored-by: Daniel Arndt <arndtd@ornl.gov>
The file was modified core/src/OpenMP/Kokkos_OpenMP_Instance.hpp (diff)
Commit 5aa0ceee4b0132ec411f7fc1e11b4b9a1039cde3 by Damien L-G
Drop unnecessary guarding for a tool library being loaded in ProfilingSection
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)
Commit 73c75075536f905818560f0bc415348b462fe38f by Damien L-G
Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)

Summary

  1. Introduce constructor for multi-GPU support. (details)
  2. Fix typo (details)
  3. Add test (details)
  4. Fix typo. (details)
  5. Explicitly check for valid device id (details)
  6. Set the device id in cuda_kernel_arch (details)
  7. Check for default device (details)
  8. Check that device associated with stream matches requested device (details)
  9. Remove extra constructor (details)
  10. Address reviewer comments (details)
  11. m_cudaDev isn't static anymore (details)
  12. Set the device id explicitly for CUDA API calls in impl_initialize (details)
  13. Fixup test math functions ulp should double -> int (details)
  14. Drop DualView converting copy assignment operator (details)
  15. Don't use rocm-docker for clang-format (details)
  16. Diable HIP CI (details)
  17. Do not negate the dependent true traits helper (details)
  18. Add missing gfx940 (details)
  19. Add Impl::always_false type-dendent false trait (details)
  20. Per review prefer always_false<Arg>::value to is_void_v<Arg> (details)
  21. Improve "no copy mechanism" exception message (details)
  22. Add a unit test for new deep_copy exception msg (details)
  23. Add missing include sstream (details)
  24. src->source, dst->destination (details)
  25. Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support (details)
  26. SYCL: Force inlining of Kokkos::printf (#6650) (details)
  27. Improve handling of printf in OMPT on Intel GPUs (details)
  28. OpenMP: Use `omp_get_nested` for older gcc versions (#6685) (details)
  29. Drop unnecessary guarding for a tool library being loaded in ProfilingSection (details)
  30. Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp (details)
Commit b8482317e26c3ae0eb421eb4b5b95fe0d16a0743 by Daniel Arndt
Introduce constructor for multi-GPU support.
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 64a9b3d854521b7b6dabf7fd9e162c959045b2e3 by noreply
Fix typo

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was addedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
The file was modified core/unit_test/CMakeLists.txt (diff)
Commit 1683786110806f22fb14b2c9e065ee8350cec8ac by noreply
Fix typo.

Co-authored-by: Dong Hun Lee <59181952+ldh4@users.noreply.github.com>
The file was modified core/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp (diff)
Commit fa1aaa712a36394c5c1f1508771cbf13160353b7 by Daniel Arndt
Explicitly check for valid device id
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 41253bd55d7fad8c72f5f3756792705b66bdc954 by Daniel Arndt
Set the device id in cuda_kernel_arch
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit f6977cf431cf245341e6dd2f744c79c5a62a315e by Daniel Arndt
Check for default device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit e156d5859e36fbc3c4f84575a0c8734689c1b3eb by Daniel Arndt
Check that device associated with stream matches requested device
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 1fcce6936bf667339350381c8b265b5d82d810da by Daniel Arndt
Remove extra constructor
The file was modified core/src/Cuda/Kokkos_Cuda.hpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.hpp (diff)
The file was modified core/unit_test/CMakeLists.txt (diff)
The file was removedcore/unit_test/cuda/TestCuda_InterOp_StreamsMultiGPU.cpp
Commit a07c7a2b6c9999020859a7de86004e447790d0c4 by Daniel Arndt
Address reviewer comments
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 403c34f304077eeb92dc8fd8cb2fd24ae6c35688 by Daniel Arndt
m_cudaDev isn't static anymore
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit d4a517f82d73a780d03ab06ece627b47d7aad43f by Daniel Arndt
Set the device id explicitly for CUDA API calls in impl_initialize
The file was modified core/src/Cuda/Kokkos_Cuda_Instance.cpp (diff)
Commit 71729af719b2eb20b512efad05ae609e74e1dd7a by Damien L-G
Fixup test math functions ulp should double -> int
The file was modified core/unit_test/TestMathematicalFunctions.hpp (diff)
Commit b9b63dfd89fb670d9e18309ad25ef74c6bdd9400 by Damien L-G
Drop DualView converting copy assignment operator

It is unnecessary, we already have a converting constructor and a
regular copy assignment operator.
The file was modified containers/src/Kokkos_DualView.hpp (diff)
Commit 9fd95ebcb1d2f590d376e42ca993068138c0c829 by Daniel Arndt
Don't use rocm-docker for clang-format
The file was modified .jenkins (diff)
The file was modified .jenkins (diff)
Commit 76ea3a3a95e4619b80e3adf7edda6db05d4ed32d by Damien L-G
Do not negate the dependent true traits helper
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified cmake/KokkosCore_config.h.in (diff)
The file was modified generate_makefile.bash (diff)
The file was modified cmake/kokkos_arch.cmake (diff)
The file was modified Makefile.kokkos (diff)
The file was modified core/src/HIP/Kokkos_HIP_Instance.hpp (diff)
Commit 33db3046a173b5e96d1dff62c8555a8da5cea9bb by Damien L-G
Add Impl::always_false type-dendent false trait
The file was modified core/src/impl/Kokkos_Utilities.hpp (diff)
Commit e4a7cfc78f0564f2aefb8a98cab3144baffd15b6 by Damien L-G
Per review prefer always_false<Arg>::value to is_void_v<Arg>
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_Team.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_MDRange.hpp (diff)
The file was modified core/src/OpenACC/Kokkos_OpenACC_ParallelReduce_Range.hpp (diff)
Commit 316ceac58eedf97fec5ee9f2fb87018a83fb97ec by bmkelle
Improve "no copy mechanism" exception message
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 838f8938e9aaf1ac78027575743bb95b442470a9 by bmkelle
Add a unit test for new deep_copy exception msg
The file was modified core/unit_test/TestViewCopy_a.hpp (diff)
Commit 72bc7ed42f0f658ebc3b158520feba086111fc3b by bmkelle
Add missing include sstream
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 154a57df8c53bff88dc4806967cb8a49d8ed1c40 by bmkelle
src->source, dst->destination

and reformat a bit
The file was modified core/src/Kokkos_CopyViews.hpp (diff)
Commit 5df22b87b046290e84c1c6a2d4d3bcbbc47d1e5c by Bruno Turcksin
Workaround for ROCm 6.0 failing to compile with AVX2 SIMD support
The file was modified simd/src/Kokkos_SIMD_AVX2.hpp (diff)
Commit 52e44d6cfe35dcfc55c3d91ddff70cc63f8d514c by noreply
SYCL: Force inlining of Kokkos::printf (#6650)

* SYCL: Force inlining of Kokkos::printf

* Always force inline Kokkos::printf
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit 79164a43adc5da6e3eee12f3cfc66e3e1402ed3f by Daniel Arndt
Improve handling of printf in OMPT on Intel GPUs
The file was modified core/src/Kokkos_Printf.hpp (diff)
Commit cbbe09b93db49e081ab4d400975aca8c007c4223 by noreply
OpenMP: Use `omp_get_nested` for older gcc versions (#6685)

* OpenMP: fix for issue 6670.

* OpenMP: Update gcc version when using max_active_level.

* OpenMP: Edit execute_in_serial.

* Apply suggestions from code review

Co-authored-by: Daniel Arndt <arndtd@ornl.gov>

* clang-format

---------

Co-authored-by: Rahulkumar Gayatri <rgayatri@lbl.gov>
Co-authored-by: Damien L-G <dalg24+github@gmail.com>
Co-authored-by: Daniel Arndt <arndtd@ornl.gov>
The file was modified core/src/OpenMP/Kokkos_OpenMP_Instance.hpp (diff)
Commit 5aa0ceee4b0132ec411f7fc1e11b4b9a1039cde3 by Damien L-G
Drop unnecessary guarding for a tool library being loaded in ProfilingSection
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)
Commit 73c75075536f905818560f0bc415348b462fe38f by Damien L-G
Drop unnecessary header include in Kokkos_Profiling_ProfileSection.hpp
The file was modified core/src/Kokkos_Profiling_ProfileSection.hpp (diff)