Conversation
Contributor
|
@JehandadKhan could you help to review this? And maybe ask @atamazov |
Contributor
|
@pfultz2 I am unable to run the tests in the host directory # make check
[ 96%] Built target embed_lib_ck_headers
[ 98%] Built target ck_host
[ 99%] Built target ck_rtc
[ 99%] Building CXX object test/CMakeFiles/test_host_gemm_multiple_d.dir/gemm_multiple_d.cpp.o
/data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:64: error: call to 'isnan' is ambiguous
if(std::any_of(x.begin(), x.end(), [](auto y) { return std::isnan(y); }))
^~~~~~~~~~
/usr/lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/predefined_ops.h:283:16: note: in instantiation of function template specialization 'check_all<_Float16>::operator()(const rtc::buffer<_Float16> &)::(anonymous class)::operator()<_Float16>' requested here
{ return bool(_M_pred(*__it)); }
^
/usr/lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/stl_algo.h:120:8: note: in instantiation of function template specialization '__gnu_cxx::__ops::_Iter_pred<(lambda at /data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:44)>::operator()<const _Float16 *>' requested here
if (__pred(__first))
^
/usr/lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/stl_algo.h:161:14: note: in instantiation of function template specialization 'std::__find_if<const _Float16 *, __gnu_cxx::__ops::_Iter_pred<(lambda at /data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:44)>>' requested here
return __find_if(__first, __last, __pred,
^
/usr/lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/stl_algo.h:3969:19: note: in instantiation of function template specialization 'std::__find_if<const _Float16 *, __gnu_cxx::__ops::_Iter_pred<(lambda at /data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:44)>>' requested here
return std::__find_if(__first, __last,
^
/usr/lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/stl_algo.h:526:40: note: in instantiation of function template specialization 'std::find_if<const _Float16 *, (lambda at /data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:44)>' requested here
{ return __last == _GLIBCXX_STD_A::find_if(__first, __last, __pred); }
^
/usr/lib/gcc/x86_64-linux-gnu/9/../../../../include/c++/9/bits/stl_algo.h:544:20: note: in instantiation of function template specialization 'std::none_of<const _Float16 *, (lambda at /data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:44)>' requested here
{ return !std::none_of(__first, __last, __pred); }
^
/data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:17: note: in instantiation of function template specialization 'std::any_of<const _Float16 *, (lambda at /data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:115:44)>' requested here
if(std::any_of(x.begin(), x.end(), [](auto y) { return std::isnan(y); }))
^
/data/repos/composable_kernel/host/test/gemm_multiple_d.cpp:181:37: note: in instantiation of member function 'check_all<_Float16>::operator()' requested here
CHECK(report(solution, check(rtc::from_gpu(c))));
^
/opt/rocm-5.7.0/llvm/lib/clang/17.0.0/include/__clang_cuda_math_forward_declares.h:110:17: note: candidate function
__DEVICE__ bool isnan(double);
^
/opt/rocm-5.7.0/llvm/lib/clang/17.0.0/include/__clang_cuda_math_forward_declares.h:111:17: note: candidate function
__DEVICE__ bool isnan(float);
^
1 error generated when compiling for gfx90a.
make[3]: *** [test/CMakeFiles/test_host_gemm_multiple_d.dir/build.make:76: test/CMakeFiles/test_host_gemm_multiple_d.dir/gemm_multiple_d.cpp.o] Error 1
make[2]: *** [CMakeFiles/Makefile2:1009: test/CMakeFiles/test_host_gemm_multiple_d.dir/all] Error 2
make[1]: *** [CMakeFiles/Makefile2:885: CMakeFiles/check.dir/rule] Error 2
make: *** [Makefile:544: check] Error 2
If you can help test the branch as well as provide some example on how to run the driver that would be helpful as well |
Contributor
Author
|
@JehandadKhan It should be fixed now. |
Contributor
|
The PR looks good, but we have the following items to do:
|
aosewski
reviewed
Feb 28, 2024
include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp
Outdated
Show resolved
Hide resolved
Collaborator
|
Moved the code from "host" to "codegen" folder, synced with the develop branch, fixed the item pointed by Adam. |
zjing14
approved these changes
Mar 6, 2024
This was referenced Mar 17, 2024
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This adds the host lib which can generate instances and can be used for runtime compilation.