ekondis
mixbench-hip does not compile
I have installed the vanillad ROCm from https://github.com/ROCm-Developer-Tools/HIP/blob/master/INSTALL.md
I then set the HIP_PATH to this installation
cmake completes and gives the following build directory: CMakeCache.txt CMakeFiles cmake_install.cmake Makefile
However, running make then gives many errors:
$ make [ 16%] Building HIPCC object CMakeFiles/mixbench-hip-ro.dir/mixbench-hip-ro_generated_mix_kernels_hip_ro.cpp.o /opt/rocm/hip/include/hip/nvcc_detail/hip_runtime_api.h(1355): error: class "cudaPointerAttributes" has no member "memoryType"
/usr/local/cuda/include/cuda_fp16.h(1525): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1525): error: declaration is incompatible with overloaded function "__shfl_sync" /usr/local/cuda/include/sm_30_intrinsics.h(133): here
...
Any suggestions on how to solve this?
I believe that you have set the HIP_PATH env variable, e.g. as below:
export HIP_PATH=/opt/rocm/hip/
Are you building for CUDA target?
Thanks for your response. Indeed I have set the HIP_PATH as indicated by you and I am building for NVIDIA V100. Can the HIP part not be built for CUDA target, or is there something else that I need to change? I am quite interested in comparing the different software ecosystems, incl the benchmarks, for different type of GPU's.
I have not used the CUDA target recently. Which is the version of the underlying CUDA installation?
I suspect that this is a problem with HIP expecting a member of the struct cudaPointerAttributes that no longer exists in recent CUDA versions (or just got renamed to "type"). See the relevant CUDA documentation reference.
I will ask our system administrators whether we can install it and let you know. Thanks for your help.
On Thu, Sep 24, 2020 at 10:18 PM Elias [email protected] wrote:
Since ROCm v3.8 is out, could you try again with this release?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ekondis/mixbench/issues/20#issuecomment-698567205, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEIFZVJ2G57QLG6QIP6KCNLSHOSSZANCNFSM4QJVF3PA .
We tried updating ROCm to 3.8.0 and also the NVIDIA software and drivers to the latest versions. However, it still fails at the make stage. Please see below for the error report.
=== echo ${HIP_PATH} /opt/rocm-3.8.0/hip
[nikheftest-roonk@wn-gp-01 build-hip]$ cmake3 ../mixbench-hip -- The C compiler identification is GNU 4.8.5 -- The CXX compiler identification is GNU 4.8.5 -- Check for working C compiler: /bin/cc -- Check for working C compiler: /bin/cc - works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Check for working CXX compiler: /bin/c++ -- Check for working CXX compiler: /bin/c++ - works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Found HIP: 3.8.20371-d1886b0b -- Configuring done -- Generating done -- Build files have been written to: /home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/build-hip [nikheftest-roonk@wn-gp-01 build-hip]$ ls CMakeCache.txt CMakeFiles cmake_install.cmake Makefile [nikheftest-roonk@wn-gp-01 build-hip]$ make [ 16%] Building HIPCC object CMakeFiles/mixbench-hip-ro.dir/mixbench-hip-ro_generated_mix_kernels_hip_ro.cpp.o /usr/local/cuda/include/cuda_fp16.h(1525): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1525): error: declaration is incompatible with overloaded function "__shfl_sync" /usr/local/cuda/include/sm_30_intrinsics.h(133): here
/usr/local/cuda/include/cuda_fp16.h(1525): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1525): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1525): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1526): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1526): error: declaration is incompatible with overloaded function "__shfl_up_sync" /usr/local/cuda/include/sm_30_intrinsics.h(135): here
/usr/local/cuda/include/cuda_fp16.h(1526): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1526): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1526): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1527): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1527): error: declaration is incompatible with overloaded function "__shfl_down_sync" /usr/local/cuda/include/sm_30_intrinsics.h(137): here
/usr/local/cuda/include/cuda_fp16.h(1527): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1527): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1527): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1528): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1528): error: declaration is incompatible with overloaded function "__shfl_xor_sync" /usr/local/cuda/include/sm_30_intrinsics.h(139): here
/usr/local/cuda/include/cuda_fp16.h(1528): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1528): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1528): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1529): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1529): error: variable "__shfl_sync" has already been defined
/usr/local/cuda/include/cuda_fp16.h(1529): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1529): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1529): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1530): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1530): error: variable "__shfl_up_sync" has already been defined
/usr/local/cuda/include/cuda_fp16.h(1530): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1530): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1530): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1531): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1531): error: variable "__shfl_down_sync" has already been defined
/usr/local/cuda/include/cuda_fp16.h(1531): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1531): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1531): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1532): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.h(1532): error: variable "__shfl_xor_sync" has already been defined
/usr/local/cuda/include/cuda_fp16.h(1532): error: incomplete type is not allowed
/usr/local/cuda/include/cuda_fp16.h(1532): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.h(1532): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.h(1558): error: declaration is
incompatible with "
/usr/local/cuda/include/cuda_fp16.h(1582): error: declaration is
incompatible with "
/usr/local/cuda/include/cuda_fp16.h(1606): error: declaration is
incompatible with "
/usr/local/cuda/include/cuda_fp16.h(1629): error: declaration is
incompatible with "
/usr/local/cuda/include/cuda_fp16.h(1653): error: declaration is incompatible with previous "__shfl_sync" (1525): here
/usr/local/cuda/include/cuda_fp16.h(1676): error: declaration is incompatible with previous "__shfl_up_sync" (1526): here
/usr/local/cuda/include/cuda_fp16.h(1700): error: declaration is incompatible with previous "__shfl_down_sync" (1527): here
/usr/local/cuda/include/cuda_fp16.h(1723): error: declaration is incompatible with previous "__shfl_xor_sync" (1528): here
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: declaration is incompatible with overloaded function "__shfl_sync" /usr/local/cuda/include/cuda_fp16.h(1558): here
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: expected an identifier
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: expected an identifier
/usr/local/cuda/include/cuda_fp16.hpp(1327): error: inline specifier allowed on function declarations only
/usr/local/cuda/include/cuda_fp16.hpp(1329): error: type name is not allowed
/usr/local/cuda/include/cuda_fp16.hpp(1329): error: expected a ","
/usr/local/cuda/include/cuda_fp16.hpp(1329): error: expected a "}"
/usr/local/cuda/include/cuda_fp16.hpp(1330): error: expected a ")"
/usr/local/cuda/include/cuda_fp16.hpp(1331): error: identifier "width" is undefined
/usr/local/cuda/include/cuda_fp16.hpp(1332): error: expected a declaration
/usr/local/cuda/include/cuda_fp16.hpp(1333): error: expected a declaration
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(31): error: identifier "__hfma2" is undefined
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(37): error: identifier "__hbeq2" is undefined
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(115): error: no instance of constructor "__half2::__half2" matches the argument list argument types are: (float)
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=0U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=0U]" (165): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=0U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=0U]" (165): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=1U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=1U]" (166): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=1U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=1U]" (166): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=2U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=2U]" (167): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=2U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=2U]" (167): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=3U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=3U]" (168): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=3U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=3U]" (168): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=4U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=4U]" (169): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=4U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=4U]" (169): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=5U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=5U]" (170): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=5U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=5U]" (170): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=6U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=6U]" (171): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=6U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=6U]" (171): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=7U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=7U]" (172): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=7U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=7U]" (172): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=8U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=8U]" (173): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=8U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=8U]" (173): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=9U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=9U]" (174): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=9U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=9U]" (174): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=10U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=10U]" (175): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=10U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=10U]" (175): here
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(60):
error: no instance of function template "is_equal" matches the argument list
argument types are: (half2,
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp(55): error: no suitable constructor exists to convert from "int" to "__half2" detected during: instantiation of "void benchmark_func<T,blockSize,granularity,compute_iterations>(T, T *) [with T=half2, blockSize=256, granularity=8U, compute_iterations=11U]" (116): here instantiation of "void runbench<compute_iterations>(double *, long) [with compute_iterations=11U]" (176): here
Error limit reached. 100 errors detected in the compilation of "/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/mixbench-hip/mix_kernels_hip_ro.cpp". Compilation terminated. CMake Error at mixbench-hip-ro_generated_mix_kernels_hip_ro.cpp.o.cmake:192 (message): Error generating file
/home/nikheftest-roonk/RO_GPU/MIXBENCH/mixbench/build-hip/CMakeFiles/mixbench-hip-ro.dir//./mixbench-hip-ro_generated_mix_kernels_hip_ro.cpp.o
make[2]: *** [CMakeFiles/mixbench-hip-ro.dir/mixbench-hip-ro_generated_mix_kernels_hip_ro.cpp.o] Error 1 make[1]: *** [CMakeFiles/mixbench-hip-ro.dir/all] Error 2 make: *** [all] Error 2
On Fri, Sep 25, 2020 at 4:35 PM Johannes Oonk [email protected] wrote:
I will ask our system administrators whether we can install it and let you know. Thanks for your help.
On Thu, Sep 24, 2020 at 10:18 PM Elias [email protected] wrote:
Since ROCm v3.8 is out, could you try again with this release?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ekondis/mixbench/issues/20#issuecomment-698567205, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEIFZVJ2G57QLG6QIP6KCNLSHOSSZANCNFSM4QJVF3PA .