Skip to content

GPU: Compatibility to ROCm 7.14 #15627

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
davidrohr wants to merge 2 commits into
base: dev
Choose a base branch
from
+7 −3
Open

GPU: Compatibility to ROCm 7.14 #15627

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
10cd3af
GPU: Fix CUDA/HIP API incompatibility with new ROCm
davidrohr Jul 21, 2026
13da97b
GPU CMake: Improve ROCm path detection
davidrohr Jul 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion GPU/GPUTracking/Base/cuda/GPUReconstructionCUDA.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -594,7 +594,11 @@ void GPUReconstructionCUDA::PrintKernelOccupancies()
int32_t maxBlocks = 0, threads = 0, suggestedBlocks = 0, nRegs = 0, sMem = 0;
GPUChkErr(cudaSetDevice(mDeviceId));
for (uint32_t i = 0; i < mInternals->kernelFunctions.size(); i++) {
GPUChkErr(cuOccupancyMaxPotentialBlockSize(&suggestedBlocks, &threads, *mInternals->kernelFunctions[i], 0, 0, 0)); // NOLINT: failure in clang-tidy
#if !defined(__HIPCC__) || (defined(__clang_major__) && __clang_major__ < 23) // CUDA
GPUChkErr(cuOccupancyMaxPotentialBlockSize(&suggestedBlocks, &threads, *mInternals->kernelFunctions[i], 0, 0, 0));
#else
GPUChkErr(cuOccupancyMaxPotentialBlockSize(&suggestedBlocks, &threads, *mInternals->kernelFunctions[i], 0, 0));
#endif
GPUChkErr(cuOccupancyMaxActiveBlocksPerMultiprocessor(&maxBlocks, *mInternals->kernelFunctions[i], threads, 0));
GPUChkErr(cuFuncGetAttribute(&nRegs, CU_FUNC_ATTRIBUTE_NUM_REGS, *mInternals->kernelFunctions[i]));
GPUChkErr(cuFuncGetAttribute(&sMem, CU_FUNC_ATTRIBUTE_SHARED_SIZE_BYTES, *mInternals->kernelFunctions[i]));
Expand Down
4 changes: 2 additions & 2 deletions dependencies/FindO2GPU.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@
# or submit itself to any jurisdiction.

# NOTE!!!! - Whenever this file is changed, move it over to alidist/resources
# FindO2GPU.cmake Version 17
# FindO2GPU.cmake Version 18

set(CUDA_COMPUTETARGET_DEFAULT_FULL 80-real 86-real 89-real 120-real 75-virtual)
set(HIP_AMDGPUTARGET_DEFAULT_FULL gfx906;gfx908)
Expand Down Expand Up @@ -318,7 +318,7 @@ if(ENABLE_HIP)
set(CMAKE_HIP_STANDARD_REQUIRED TRUE)
set(TMP_ROCM_DIR_LIST "${CMAKE_PREFIX_PATH}:$ENV{CMAKE_PREFIX_PATH}")
string(REPLACE ":" ";" TMP_ROCM_DIR_LIST "${TMP_ROCM_DIR_LIST}")
list(FILTER TMP_ROCM_DIR_LIST INCLUDE REGEX rocm)
list(FILTER TMP_ROCM_DIR_LIST INCLUDE REGEX /rocm/lib/cmake)
list(POP_FRONT TMP_ROCM_DIR_LIST TMP_ROCM_DIR)
get_filename_component(TMP_ROCM_DIR ${TMP_ROCM_DIR}/../../ ABSOLUTE)
if (NOT DEFINED CMAKE_HIP_COMPILER)
Expand Down