batched/dense: Reduce register pressure #1588

e10harvey · 2022-11-15T00:08:54Z

Fixes #1547.

kokkos-devops-admin · 2022-11-15T02:53:27Z

Status Flag 'Pre-Test Inspection' - Auto Inspected - Inspection Is Not Necessary for this Pull Request.

kokkos-devops-admin · 2022-11-15T03:22:49Z

Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects:

Pull Request Auto Testing STARTING (click to expand)

Build Information

Test Name: KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930

Build Num: 154
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight

Build Num: 162
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020

Build Num: 115
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020_Light_LayoutRight

Build Num: 114
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_GCC1020

Build Num: 77
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_INTEL19

Build Num: 164
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG1001

Build Num: 213
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG13CUDA10

Build Num: 99
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020

Build Num: 103
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Using Repos:

Repo: KOKKOSKERNELS (e10harvey/kokkos-kernels)

Pull Request Author: e10harvey

kokkos-devops-admin · 2022-11-15T04:20:32Z

Status Flag 'Pull Request AutoTester' - Jenkins Testing: 1 or more Jobs FAILED

Note: Testing will normally be attempted again in approx. 2 Hrs 30 Mins. If a change to the PR source branch occurs, the testing will be attempted again on next available autotester run.

Pull Request Auto Testing has FAILED (click to expand)

Build Information

Test Name: KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930

Build Num: 154
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight

Build Num: 162
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020

Build Num: 115
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020_Light_LayoutRight

Build Num: 114
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_GCC1020

Build Num: 77
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_INTEL19

Build Num: 164
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG1001

Build Num: 213
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG13CUDA10

Build Num: 99
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020

Build Num: 103
Status: FAILED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`9bdbe69`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Console Output (last 100 lines) : KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930 # 154 (click to expand)


make: *** [all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
gcc-9.3.0-OpenMP-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.19.3 gcc/9.3.0
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP --arch=Power8,Pascal60 --compiler=/home/projects/ppc64le/gcc/9.3.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930/KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930.154/TestAll_2022-11-14_19.53.48/gcc/9.3.0/OpenMP-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

gcc-9.3.0-OpenMP_Serial-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    source /etc/profile.d/modules.sh
    module purge
    module load cmake/3.19.3 gcc/9.3.0

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP,Serial --arch=Power8,Pascal60 --compiler=/home/projects/ppc64le/gcc/9.3.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930/KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930.154/TestAll_2022-11-14_19.53.48/gcc/9.3.0/OpenMP_Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

gcc-9.3.0-Serial-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    source /etc/profile.d/modules.sh
    module purge
    module load cmake/3.19.3 gcc/9.3.0

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Serial --arch=Power8,Pascal60 --compiler=/home/projects/ppc64le/gcc/9.3.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930/KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930.154/TestAll_2022-11-14_19.53.48/gcc/9.3.0/Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight # 162 (click to expand)


/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp(306): here
            instantiation of "void test_batched_gemm_with_layout(int) [with ViewType=Kokkos::View, DeviceType=Kokkos::Cuda, ValueType=double, ScalarType=double, ParamTagType=Test::SharedParamTag]" 
/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp(369): here
            instantiation of "int test_batched_gemm() [with DeviceType=Kokkos::Cuda, ValueType=double, ScalarType=double, ParamTagType=Test::SharedParamTag]" 
/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm_Real.hpp(247): here
/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp(528): error: expression must be a modifiable lvalue

detected during:

instantiation of "int KokkosBatched::BatchedGemm<ArgTransA,ArgTransB,ArgBatchSzDim,BatchedGemmHandleType,ScalarType,AViewType,BViewType,CViewType>(BatchedGemmHandleType *, ScalarType, const AViewType &, const BViewType &, ScalarType, const CViewType &) [with ArgTransA=KokkosBlas::Trans::NoTranspose, ArgTransB=KokkosBlas::Trans::NoTranspose, ArgBatchSzDim=KokkosBatched::BatchLayout::Left, BatchedGemmHandleType=KokkosBatched::BatchedGemmHandle, ScalarType=double, AViewType=Kokkos::View<double ***, Kokkos::LayoutLeft, Kokkos::Cuda>, BViewType=Kokkos::View<double ***, Kokkos::LayoutLeft, Kokkos::Cuda>, CViewType=Kokkos::View<double ***, Kokkos::LayoutLeft, Kokkos::Cuda>]"

/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp(113): here

instantiation of "void Test::impl_test_batched_gemm_with_handle<DeviceType,ViewType,ScalarType,ParamTagType>(KokkosBatched::BatchedGemmHandle *, int, int, int, int, int, int, int, ScalarType, ScalarType) [with DeviceType=Kokkos::Cuda, ViewType=Kokkos::View<double ***, Kokkos::LayoutLeft, Kokkos::Cuda>, ScalarType=double, ParamTagType=Test::SharedParamTag<KokkosBlas::Trans::NoTranspose, KokkosBlas::Trans::NoTranspose, KokkosBatched::BatchLayout::Left>]"

/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp(254): here

instantiation of "void Test::impl_test_batched_gemm<DeviceType,ViewType,ScalarType,ParamTagType>(int, int, int, int, int, int, int) [with DeviceType=Kokkos::Cuda, ViewType=Kokkos::View<double ***, Kokkos::LayoutLeft, Kokkos::Cuda>, ScalarType=double, ParamTagType=Test::SharedParamTag<KokkosBlas::Trans::NoTranspose, KokkosBlas::Trans::NoTranspose, KokkosBatched::BatchLayout::Left>]"

/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp(306): here

instantiation of "void test_batched_gemm_with_layout<ViewType,DeviceType,ValueType,ScalarType,ParamTagType>(int) [with ViewType=Kokkos::View<double ***, Kokkos::LayoutLeft, Kokkos::Cuda>, DeviceType=Kokkos::Cuda, ValueType=double, ScalarType=double, ParamTagType=Test::SharedParamTag<KokkosBlas::Trans::NoTranspose, KokkosBlas::Trans::NoTranspose, KokkosBatched::BatchLayout::Left>]"

/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp(369): here

instantiation of "int test_batched_gemm<DeviceType,ValueType,ScalarType,ParamTagType>() [with DeviceType=Kokkos::Cuda, ValueType=double, ScalarType=double, ParamTagType=Test::SharedParamTag<KokkosBlas::Trans::NoTranspose, KokkosBlas::Trans::NoTranspose, KokkosBatched::BatchLayout::Left>]"

/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm_Real.hpp(247): here
Error limit reached.

100 errors detected in the compilation of "/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos-kernels/batched/dense/unit_test/backends/Test_Cuda_Batched_Dense.cpp".

Compilation terminated.

make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_cuda.dir/build.make:90: batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_cuda.dir/backends/Test_Cuda_Batched_Dense.cpp.o] Error 1

make[1]: *** [CMakeFiles/Makefile2:1567: batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_cuda.dir/all] Error 2

[ 76%] Linking CXX executable KokkosKernels_batched_sla_openmp

[ 76%] Built target KokkosKernels_batched_sla_openmp

[ 76%] Linking CXX executable KokkosKernels_common_openmp

[ 76%] Built target KokkosKernels_common_openmp

[ 77%] Linking CXX executable KokkosKernels_batched_sla_cuda

[ 77%] Built target KokkosKernels_batched_sla_cuda

[ 77%] Linking CXX executable KokkosKernels_common_cuda

[ 77%] Built target KokkosKernels_common_cuda

[ 78%] Linking CXX executable KokkosKernels_blas_openmp

[ 78%] Built target KokkosKernels_blas_openmp

[ 78%] Linking CXX executable KokkosKernels_graph_openmp

[ 78%] Built target KokkosKernels_graph_openmp

[ 78%] Linking CXX executable KokkosKernels_blas_cuda

[ 78%] Built target KokkosKernels_blas_cuda

[ 78%] Linking CXX executable KokkosKernels_graph_cuda

[ 78%] Built target KokkosKernels_graph_cuda

[ 78%] Linking CXX executable KokkosKernels_sparse_openmp

[ 78%] Built target KokkosKernels_sparse_openmp

[ 79%] Linking CXX executable KokkosKernels_sparse_cuda

[ 79%] Built target KokkosKernels_sparse_cuda

make: *** [Makefile:146: all] Error 2

#######################################################

PASSED TESTS

#######################################################

#######################################################

FAILED TESTS

#######################################################

cuda-11.2.2-Cuda_OpenMP-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    source /etc/profile.d/modules.sh
    module purge
    module load cmake/3.21.2 cuda/11.2.2 openblas/0.3.20/gcc/9.3.0

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Cuda,OpenMP --arch=Volta70 --compiler=/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/kokkos/bin/nvcc_wrapper --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wuninitialized " --cxxstandard="17" --ldflags="" --with-cuda=/home/projects/ppc64le-pwr9-nvidia/spack-installs/cuda/11.2.2/gcc/8.3.1/base/uflzqtv  --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight/KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight.162/TestAll_2022-11-14_20.16.25/cuda/11.2.2/Cuda_OpenMP-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_GCC1020 # 115 (click to expand)


[ 70%] Built target sparse_pcg
[ 71%] Linking CXX executable sparse_sptrsv_superlu
[ 71%] Built target sparse_sptrsv_superlu
[ 71%] Linking CXX executable KokkosKernels_batched_sla_openmp
[ 71%] Built target KokkosKernels_batched_sla_openmp
[ 71%] Linking CXX executable sparse_block_pcg
[ 72%] Built target sparse_block_pcg
[ 72%] Linking CXX executable sparse_spgemm
[ 72%] Linking CXX executable KokkosBatched_Test_GMRES
[ 72%] Built target sparse_spgemm
[ 72%] Built target KokkosBatched_Test_GMRES
[ 72%] Linking CXX executable KokkosKernels_common_openmp
[ 72%] Built target KokkosKernels_common_openmp
[ 73%] Linking CXX executable sparse_gs
[ 73%] Built target sparse_gs
[ 74%] Linking CXX executable KokkosKernels_graph_openmp
[ 74%] Built target KokkosKernels_graph_openmp
[ 74%] Linking CXX executable KokkosKernels_blas_openmp
[ 74%] Built target KokkosKernels_blas_openmp
[ 75%] Linking CXX executable KokkosKernels_sparse_openmp
[ 75%] Built target KokkosKernels_sparse_openmp
make: *** [all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
gcc-10.2.0-OpenMP-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.19.3 gcc/10.2.0
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP --arch=SKX --compiler=/home/projects/x86-64/gcc/10.2.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_GCC1020/KokkosKernels_PullRequest_GCC1020.115/TestAll_2022-11-14_20.16.38/gcc/10.2.0/OpenMP-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

gcc-10.2.0-Threads_Serial-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    source /etc/profile.d/modules.sh
    module purge
    module load cmake/3.19.3 gcc/10.2.0

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Threads,Serial --arch=SKX --compiler=/home/projects/x86-64/gcc/10.2.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_GCC1020/KokkosKernels_PullRequest_GCC1020.115/TestAll_2022-11-14_20.16.38/gcc/10.2.0/Threads_Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

srun: error: blake10: task 0: Exited with exit code 2

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_GCC1020_Light_LayoutRight # 114 (click to expand)


[ 70%] Built target sparse_pcg
[ 71%] Linking CXX executable sparse_sptrsv_superlu
[ 71%] Built target sparse_sptrsv_superlu
[ 72%] Linking CXX executable sparse_spgemm
[ 72%] Linking CXX executable KokkosKernels_batched_sla_openmp
[ 72%] Built target sparse_spgemm
[ 72%] Built target KokkosKernels_batched_sla_openmp
[ 72%] Linking CXX executable KokkosBatched_Test_GMRES
[ 72%] Linking CXX executable sparse_block_pcg
[ 72%] Built target KokkosBatched_Test_GMRES
[ 72%] Built target sparse_block_pcg
[ 72%] Linking CXX executable KokkosKernels_common_openmp
[ 72%] Built target KokkosKernels_common_openmp
[ 73%] Linking CXX executable sparse_gs
[ 73%] Built target sparse_gs
[ 74%] Linking CXX executable KokkosKernels_graph_openmp
[ 74%] Built target KokkosKernels_graph_openmp
[ 74%] Linking CXX executable KokkosKernels_blas_openmp
[ 74%] Built target KokkosKernels_blas_openmp
[ 75%] Linking CXX executable KokkosKernels_sparse_openmp
[ 75%] Built target KokkosKernels_sparse_openmp
make: *** [all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
gcc-10.2.0-OpenMP-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.19.3 gcc/10.2.0
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP --arch=SKX --compiler=/home/projects/x86-64/gcc/10.2.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutRight --with-tpls=    --with-options= --with-cuda-options=  --with-spaces=hostspace --no-examples   --no-default-eti
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_GCC1020_Light_LayoutRight/KokkosKernels_PullRequest_GCC1020_Light_LayoutRight.114/TestAll_2022-11-14_20.16.48/gcc/10.2.0/OpenMP-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

gcc-10.2.0-Threads_Serial-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    source /etc/profile.d/modules.sh
    module purge
    module load cmake/3.19.3 gcc/10.2.0

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Threads,Serial --arch=SKX --compiler=/home/projects/x86-64/gcc/10.2.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutRight --with-tpls=    --with-options= --with-cuda-options=  --with-spaces=hostspace --no-examples   --no-default-eti
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_GCC1020_Light_LayoutRight/KokkosKernels_PullRequest_GCC1020_Light_LayoutRight.114/TestAll_2022-11-14_20.16.48/gcc/10.2.0/Threads_Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

srun: error: blake12: task 0: Exited with exit code 2

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_Tpls_GCC1020 # 77 (click to expand)


/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:44: error: assignment of read-only variable tile_k
  528 |           tile_m = 32, tile_n = 32, tile_k = 16;
      |                                     ~~~~~~~^~~~
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp: In instantiation of int KokkosBatched::BatchedGemm(BatchedGemmHandleType*, ScalarType, const AViewType&, const BViewType&, ScalarType, const CViewType&) [with ArgTransA = KokkosBlas::Trans::Transpose; ArgTransB = KokkosBlas::Trans::Transpose; ArgBatchSzDim = KokkosBatched::BatchLayout::Right; BatchedGemmHandleType = KokkosBatched::BatchedGemmHandle; ScalarType = Kokkos::complex; AViewType = Kokkos::View***, Kokkos::LayoutLeft, Kokkos::Serial>; BViewType = Kokkos::View***, Kokkos::LayoutLeft, Kokkos::Serial>; CViewType = Kokkos::View***, Kokkos::LayoutLeft, Kokkos::Serial>]:
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp:111:52:   required from void Test::impl_test_batched_gemm_with_handle(KokkosBatched::BatchedGemmHandle*, int, int, int, int, int, int, int, ScalarType, ScalarType) [with DeviceType = Kokkos::Serial; ViewType = Kokkos::View***, Kokkos::LayoutLeft, Kokkos::Serial>; ScalarType = Kokkos::complex; ParamTagType = Test::SharedParamTag]
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp:252:59:   required from void Test::impl_test_batched_gemm(int, int, int, int, int, int, int) [with DeviceType = Kokkos::Serial; ViewType = Kokkos::View***, Kokkos::LayoutLeft, Kokkos::Serial>; ScalarType = Kokkos::complex; ParamTagType = Test::SharedParamTag]
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp:306:47:   required from void test_batched_gemm_with_layout(int) [with ViewType = Kokkos::View***, Kokkos::LayoutLeft, Kokkos::Serial>; DeviceType = Kokkos::Serial; ValueType = Kokkos::complex; ScalarType = Kokkos::complex; ParamTagType = Test::SharedParamTag]
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm.hpp:369:46:   required from int test_batched_gemm() [with DeviceType = Kokkos::Serial; ValueType = Kokkos::complex; ScalarType = Kokkos::complex; ParamTagType = Test::SharedParamTag]
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/unit_test/Test_Batched_BatchedGemm_Complex.hpp:167:62:   required from here
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:18: error: assignment of read-only variable tile_m
  528 |           tile_m = 32, tile_n = 32, tile_k = 16;
      |           ~~~~~~~^~~~
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:31: error: assignment of read-only variable tile_n
  528 |           tile_m = 32, tile_n = 32, tile_k = 16;
      |                        ~~~~~~~^~~~
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:44: error: assignment of read-only variable tile_k
  528 |           tile_m = 32, tile_n = 32, tile_k = 16;
      |                                     ~~~~~~~^~~~
make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_openmp.dir/backends/Test_OpenMP_Batched_Dense.cpp.o] Error 1
make[1]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_openmp.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/backends/Test_Serial_Batched_Dense.cpp.o] Error 1
make[1]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/all] Error 2
[ 77%] Linking CXX executable KokkosKernels_batched_sla_openmp
[ 77%] Linking CXX executable KokkosKernels_batched_sla_serial
[ 77%] Built target KokkosKernels_batched_sla_openmp
[ 77%] Built target KokkosKernels_batched_sla_serial
[ 77%] Linking CXX executable KokkosKernels_common_serial
[ 77%] Built target KokkosKernels_common_serial
[ 77%] Linking CXX executable KokkosKernels_common_openmp
[ 77%] Built target KokkosKernels_common_openmp
[ 77%] Linking CXX executable KokkosKernels_graph_serial
[ 77%] Built target KokkosKernels_graph_serial
[ 77%] Linking CXX executable KokkosKernels_graph_openmp
[ 77%] Built target KokkosKernels_graph_openmp
[ 78%] Linking CXX executable KokkosKernels_blas_serial
[ 78%] Built target KokkosKernels_blas_serial
[ 78%] Linking CXX executable KokkosKernels_blas_openmp
[ 78%] Built target KokkosKernels_blas_openmp
[ 79%] Linking CXX executable KokkosKernels_sparse_openmp
[ 79%] Built target KokkosKernels_sparse_openmp
[ 79%] Linking CXX executable KokkosKernels_sparse_serial
[ 79%] Built target KokkosKernels_sparse_serial
make: *** [all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
gcc-10.2.0-OpenMP_Serial-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.19.3 gcc/10.2.0 openblas/0.3.21/gcc/10.2.0
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP,Serial --arch=SKX --compiler=/home/projects/x86-64/gcc/10.2.0/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=blas --user-blas-path=/home/projects/x86-64-skylake/spack/opt/spack/linux-rhel7-skylake_avx512/gcc-10.2.0/openblas-0.3.21-k3hscel2el5bpov4ejvsn5yl5atsezr6/lib --user-lapack-path=/home/projects/x86-64-skylake/spack/opt/spack/linux-rhel7-skylake_avx512/gcc-10.2.0/openblas-0.3.21-k3hscel2el5bpov4ejvsn5yl5atsezr6/lib --user-blas-lib=blas --user-lapack-lib=lapack --extra-linker-flags=-lgfortran,-lm --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_GCC1020/KokkosKernels_PullRequest_Tpls_GCC1020.77/TestAll_2022-11-14_20.16.58/gcc/10.2.0/OpenMP_Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

srun: error: blake13: task 0: Exited with exit code 1

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_Tpls_INTEL19 # 164 (click to expand)


[ 90%] Built target KokkosKernels_wiki_coarsening
[ 90%] Linking CXX executable KokkosKernels_wiki_coloring
[ 90%] Built target KokkosKernels_wiki_coloring
[ 91%] Linking CXX executable KokkosKernels_gmres_test_real_A
[ 91%] Built target KokkosKernels_gmres_test_real_A
[ 91%] Linking CXX executable KokkosKernels_batched_sla_threads
[ 92%] Linking CXX executable KokkosKernels_gmres_test_prec
[ 93%] Linking CXX executable gmres_ex_real_A
[ 93%] Built target KokkosKernels_batched_sla_threads
[ 93%] Built target KokkosKernels_gmres_test_prec
[ 93%] Built target gmres_ex_real_A
[ 94%] Linking CXX executable gmres_test_cmplx_A
[ 94%] Built target gmres_test_cmplx_A
[ 94%] Linking CXX executable KokkosKernels_common_threads
[ 94%] Built target KokkosKernels_common_threads
[ 94%] Linking CXX executable KokkosKernels_graph_threads
[ 94%] Built target KokkosKernels_graph_threads
[ 95%] Linking CXX executable KokkosKernels_blas_threads
[ 95%] Built target KokkosKernels_blas_threads
[ 96%] Linking CXX executable KokkosKernels_sparse_threads
[ 96%] Built target KokkosKernels_sparse_threads
make: *** [all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
intel-19.5.281-OpenMP-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.19.3 intel/compilers/19.5.281
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP --arch=SKX --compiler=/home/projects/x86-64/intel/compilers/2019/compilers_and_libraries_2019.5.281/linux/bin/intel64/icpc --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wuninitialized -diag-disable=1011 -diag-disable=869 -diag-disable=1011 -diag-disable=869" --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=mkl    --with-options= --with-cuda-options=   --no-examples   --disable-perftests
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_INTEL19/KokkosKernels_PullRequest_Tpls_INTEL19.164/TestAll_2022-11-14_20.17.09/intel/19.5.281/OpenMP-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

intel-19.5.281-Threads-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    source /etc/profile.d/modules.sh
    module purge
    module load cmake/3.19.3 intel/compilers/19.5.281

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Threads --arch=SKX --compiler=/home/projects/x86-64/intel/compilers/2019/compilers_and_libraries_2019.5.281/linux/bin/intel64/icpc --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wuninitialized -diag-disable=1011 -diag-disable=869 -diag-disable=1011 -diag-disable=869" --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=mkl    --with-options= --with-cuda-options=   --no-examples   --disable-perftests
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_Tpls_INTEL19/KokkosKernels_PullRequest_Tpls_INTEL19.164/TestAll_2022-11-14_20.17.09/intel/19.5.281/Threads-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

srun: error: blake14: task 0: Exited with exit code 1

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_CLANG1001 # 213 (click to expand)


          ~~~~~~ ^
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_CLANG1001/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:524:23: note: variable 'tile_m' declared const here
        constexpr int tile_m = 32, tile_n = 32, tile_k = 8;
        ~~~~~~~~~~~~~~^~~~~~~~~~~
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_CLANG1001/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:31: error: cannot assign to variable 'tile_n' with const-qualified type 'const int'
          tile_m = 32, tile_n = 32, tile_k = 16;
                       ~~~~~~ ^
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_CLANG1001/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:524:36: note: variable 'tile_n' declared const here
        constexpr int tile_m = 32, tile_n = 32, tile_k = 8;
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_CLANG1001/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:44: error: cannot assign to variable 'tile_k' with const-qualified type 'const int'
          tile_m = 32, tile_n = 32, tile_k = 16;
                                    ~~~~~~ ^
/home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_CLANG1001/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:524:49: note: variable 'tile_k' declared const here
        constexpr int tile_m = 32, tile_n = 32, tile_k = 8;
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~
4 errors generated.
make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_threads.dir/backends/Test_Threads_Batched_Dense.cpp.o] Error 1
make[1]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_threads.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
4 errors generated.
make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/backends/Test_Serial_Batched_Dense.cpp.o] Error 1
make[1]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/all] Error 2
[ 76%] Linking CXX executable KokkosKernels_common_threads
[ 76%] Linking CXX executable KokkosKernels_batched_sla_threads
[ 77%] Linking CXX executable KokkosKernels_batched_sla_serial
[ 77%] Linking CXX executable KokkosKernels_common_serial
[ 77%] Built target KokkosKernels_batched_sla_threads
[ 77%] Built target KokkosKernels_batched_sla_serial
[ 77%] Built target KokkosKernels_common_threads
[ 77%] Built target KokkosKernels_common_serial
[ 77%] Linking CXX executable KokkosKernels_graph_threads
[ 77%] Built target KokkosKernels_graph_threads
[ 77%] Linking CXX executable KokkosKernels_graph_serial
[ 77%] Built target KokkosKernels_graph_serial
[ 78%] Linking CXX executable KokkosKernels_blas_threads
[ 78%] Built target KokkosKernels_blas_threads
[ 78%] Linking CXX executable KokkosKernels_blas_serial
[ 78%] Built target KokkosKernels_blas_serial
[ 78%] Linking CXX executable KokkosKernels_sparse_threads
[ 78%] Built target KokkosKernels_sparse_threads
[ 79%] Linking CXX executable KokkosKernels_sparse_serial
[ 79%] Built target KokkosKernels_sparse_serial
make: *** [all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
clang-10.0.1-Threads_Serial-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.19.3 clang/10.0.1
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Threads,Serial --arch=SKX --compiler=/home/projects/x86-64/spack/opt/spack/linux-rhel7-x86_64/gcc-10.2.0/llvm-10.0.1-nldepabrxk2zcykcq7rlky5sr53hnm63/bin/clang++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/blake-new/workspace/KokkosKernels_PullRequest_CLANG1001/KokkosKernels_PullRequest_CLANG1001.213/TestAll_2022-11-14_20.17.20/clang/10.0.1/Threads_Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

srun: error: blake15: task 0: Exited with exit code 1

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_CLANG13CUDA10 # 99 (click to expand)


        ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~
/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CLANG13CUDA10/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:528:44: error: cannot assign to variable 'tile_k' with const-qualified type 'const int'
          tile_m = 32, tile_n = 32, tile_k = 16;
                                    ~~~~~~ ^
/home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CLANG13CUDA10/kokkos-kernels/batched/dense/src/KokkosBatched_Gemm_Decl.hpp:524:49: note: variable 'tile_k' declared const here
        constexpr int tile_m = 32, tile_n = 32, tile_k = 8;
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~
4 errors generated when compiling for sm_70.
make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/build.make:90: batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/backends/Test_Serial_Batched_Dense.cpp.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:1567: batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_serial.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
4 errors generated when compiling for sm_70.
make[2]: *** [batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_cuda.dir/build.make:90: batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_cuda.dir/backends/Test_Cuda_Batched_Dense.cpp.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:1540: batched/dense/unit_test/CMakeFiles/KokkosKernels_batched_dla_cuda.dir/all] Error 2
[ 76%] Linking CXX executable KokkosKernels_batched_sla_serial
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 76%] Built target KokkosKernels_batched_sla_serial
[ 76%] Linking CXX executable KokkosKernels_common_serial
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 76%] Built target KokkosKernels_common_serial
[ 77%] Linking CXX executable KokkosKernels_batched_sla_cuda
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 77%] Built target KokkosKernels_batched_sla_cuda
[ 77%] Linking CXX executable KokkosKernels_common_cuda
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 77%] Built target KokkosKernels_common_cuda
[ 77%] Linking CXX executable KokkosKernels_graph_serial
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 77%] Built target KokkosKernels_graph_serial
[ 78%] Linking CXX executable KokkosKernels_blas_serial
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 78%] Built target KokkosKernels_blas_serial
[ 78%] Linking CXX executable KokkosKernels_blas_cuda
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 78%] Built target KokkosKernels_blas_cuda
[ 78%] Linking CXX executable KokkosKernels_graph_cuda
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 78%] Built target KokkosKernels_graph_cuda
[ 78%] Linking CXX executable KokkosKernels_sparse_serial
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 78%] Built target KokkosKernels_sparse_serial
[ 79%] Linking CXX executable KokkosKernels_sparse_cuda
/home/projects/ppc64le/binutils/2.30.0/bin/ld: warning: libgfortran.so.5, needed by /home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib/liblapack.so, may conflict with libgfortran.so.4
[ 79%] Built target KokkosKernels_sparse_cuda
make: *** [Makefile:146: all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
clang-13.0.0-Cuda-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        source /etc/profile.d/modules.sh
        module purge
        module load cmake/3.21.2 clang/13.0.0 openblas/0.3.20/gcc/9.3.0 cuda/10.1.243
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Cuda --arch=Power9,Volta70 --compiler=/home/projects/ppc64le-pwr9-nvidia/spack/opt/spack/linux-rhel7-power9le/gcc-7.4.0/llvm-13.0.0-t6hzufjroylzhs7hg3dvmhrrcsvhygzv/bin/clang++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wuninitialized " --cxxstandard="17" --ldflags="" --with-cuda=/home/projects/ppc64le-pwr9-nvidia/cuda/10.1.243  --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=blas --user-blas-path=/home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib --user-lapack-path=/home/projects/ppc64le-pwr9/spack/opt/spack/linux-rhel7-power9le/gcc-9.3.0/openblas-0.3.20-wt32he2mqdzpqfzdbyhiwaqibx6j6s3l/lib --user-blas-lib=blas --user-lapack-lib=lapack --extra-linker-flags=-lgfortran,-lm --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/weaver-new/workspace/KokkosKernels_PullRequest_CLANG13CUDA10/KokkosKernels_PullRequest_CLANG13CUDA10.99/TestAll_2022-11-14_20.20.59/clang/13.0.0/Cuda-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

Build step 'Execute shell' marked build as failure

Finished: FAILURE

Console Output (last 100 lines) : KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020 # 103 (click to expand)


[ 72%] Linking CXX executable sparse_spgemm
[ 72%] Built target sparse_spgemm
[ 72%] Linking CXX executable sparse_block_pcg
[ 72%] Built target sparse_block_pcg
[ 72%] Linking CXX executable KokkosKernels_batched_sla_serial
[ 72%] Built target KokkosKernels_batched_sla_serial
[ 72%] Linking CXX executable KokkosBatched_Test_GMRES
[ 72%] Built target KokkosBatched_Test_GMRES
[ 73%] Linking CXX executable sparse_gs
[ 73%] Built target sparse_gs
[ 73%] Linking CXX executable KokkosKernels_common_serial
[ 73%] Built target KokkosKernels_common_serial
[ 73%] Linking CXX executable sparse_sptrsv_supernode
[ 73%] Built target sparse_sptrsv_supernode
[ 74%] Linking CXX executable KokkosKernels_graph_serial
[ 74%] Built target KokkosKernels_graph_serial
[ 74%] Linking CXX executable KokkosKernels_blas_serial
[ 74%] Built target KokkosKernels_blas_serial
[ 75%] Linking CXX executable KokkosKernels_sparse_serial
[ 75%] Built target KokkosKernels_sparse_serial
make: *** [Makefile:158: all] Error 2
#######################################################
PASSED TESTS
#######################################################
#######################################################
FAILED TESTS
#######################################################
armpl-21.1.0-OpenMP-release (build failed)
#######################################################
  # Reproducer instructions:
  #   Load modules:
        module purge
        module purge
        module load cmake/3.17.0 gcc/10.2.0 armpl/21.1.0
$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=OpenMP --arch=A64FX --compiler=/opt/spatse/gcc/2020-09-17/spack/opt/spack/linux-rhel8-a64fx/gcc-8.2.1/gcc-10.2.0-f73mwr3ryd77o37a5jyofxet6nk7xowg/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=armpl    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/inouye/workspace/workspace/KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020/KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020.103/TestAll_2022-11-14_20.23.03/armpl/21.1.0/OpenMP-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

armpl-21.1.0-Serial-release (build failed)

#######################################################
Reproducer instructions:
Load modules:
    module purge
    module purge
    module load cmake/3.17.0 gcc/10.2.0 armpl/21.1.0

$KOKKOSKERNELS_PATH/cm_generate_makefile.bash --with-devices=Serial --arch=A64FX --compiler=/opt/spatse/gcc/2020-09-17/spack/opt/spack/linux-rhel8-a64fx/gcc-8.2.1/gcc-10.2.0-f73mwr3ryd77o37a5jyofxet6nk7xowg/bin/g++ --cxxflags="-O3 -Wall -Wunused-parameter -Wshadow -pedantic -Werror -Wsign-compare -Wtype-limits -Wignored-qualifiers -Wempty-body -Wclobbered -Wuninitialized " --cxxstandard="17" --ldflags=""   --kokkos-path=$KOKKOS_PATH --kokkoskernels-path=$KOKKOSKERNELS_PATH --with-scalars='double,complex_double' --with-ordinals=int --with-offsets=int,size_t --with-layouts=LayoutLeft --with-tpls=armpl    --with-options= --with-cuda-options=   --no-examples
To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:
  # Move to the build directory
    cd /home/jenkins/inouye/workspace/workspace/KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020/KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020.103/TestAll_2022-11-14_20.23.03/armpl/21.1.0/Serial-release
  # To reload modules
    source ./reload_modules.sh
  # To reconfigure
    ./call_generate_makefile.sh
  # To rebuild
    make -j
  # To retest
    ctest -V

#######################################################

salloc: Relinquishing job allocation 2546

salloc: Job allocation 2546 has been revoked.

Build step 'Execute shell' marked build as failure

Finished: FAILURE

vqd8a · 2022-11-15T22:06:10Z

batched/dense/src/KokkosBatched_Gemm_Decl.hpp

+template <>
+constexpr KOKKOS_INLINE_FUNCTION int
+kk_gemm_dlb_buf_tile_k<Kokkos::Experimental::HIP>() {
+  return 16;


@e10harvey I don't really follow. Please remind me. Why can tile_k=16 for HIP reduce the register allocation as compared to tile_k=8? Isn't that with tile_k=16 we have larger tile which needs more register? Am I missing something?

tile_k = 16 results in larger shared memory allocations but smaller REG_M and REG_N sizes:

kokkos-kernels/batched/dense/impl/KokkosBatched_Gemm_DblBuf_Impl.hpp

Lines 385 to 386 in f3556c5

view_type_2d_scratch svA_scr(member.team_scratch(0), TILE_M, TILE_K);

view_type_2d_scratch svB_scr(member.team_scratch(0), TILE_K, TILE_N);

kokkos-kernels/batched/dense/impl/KokkosBatched_Gemm_DblBuf_Impl.hpp

Lines 146 to 147 in f3556c5

constexpr int reg_m = TILE_M / TILE_K;

constexpr int reg_n = TILE_N / TILE_K + 2 * !!(TILE_N % TILE_K);

Thanks @e10harvey. Now I remember. :-)

vqd8a

Thanks @e10harvey. LGTM.

lucbv

Looks good to me and yeah register pressure on MI100 is a known issue... thanks for fixing this!

e10harvey · 2022-11-16T22:57:24Z

Looks good to me and yeah register pressure on MI100 is a known issue...

@lucbv: Ah, is there somewhere I can read up on this?

kokkos-devops-admin · 2022-11-17T02:30:58Z

Status Flag 'Pre-Test Inspection' - Auto Inspected - Inspection Is Not Necessary for this Pull Request.

kokkos-devops-admin · 2022-11-17T03:44:55Z

Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects:

Pull Request Auto Testing STARTING (click to expand)

Build Information

Test Name: KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930

Build Num: 159
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight

Build Num: 167
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020

Build Num: 120
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020_Light_LayoutRight

Build Num: 119
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_GCC1020

Build Num: 82
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_INTEL19

Build Num: 169
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG1001

Build Num: 218
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG13CUDA10

Build Num: 104
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020

Build Num: 108
Status: STARTED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Using Repos:

Repo: KOKKOSKERNELS (e10harvey/kokkos-kernels)

Pull Request Author: e10harvey

kokkos-devops-admin · 2022-11-17T06:46:02Z

Status Flag 'Pull Request AutoTester' - Jenkins Testing: all Jobs PASSED

Pull Request Auto Testing has PASSED (click to expand)

Build Information

Test Name: KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930

Build Num: 159
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight

Build Num: 167
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020

Build Num: 120
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020_Light_LayoutRight

Build Num: 119
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_GCC1020

Build Num: 82
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_INTEL19

Build Num: 169
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG1001

Build Num: 218
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_CLANG13CUDA10

Build Num: 104
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020

Build Num: 108
Status: PASSED

Jenkins Parameters

Parameter Name	Value
KOKKOSKERNELS_SOURCE_BRANCH	issue1547
KOKKOSKERNELS_SOURCE_REPO	https://github.com/e10harvey/kokkos-kernels
KOKKOSKERNELS_SOURCE_SHA	`95c88e9`
KOKKOSKERNELS_TARGET_BRANCH	develop
KOKKOSKERNELS_TARGET_REPO	https://github.com/kokkos/kokkos-kernels
KOKKOSKERNELS_TARGET_SHA	`f3556c5`
PR_LABELS
PULLREQUESTNUM	1588
TEST_REPO_ALIAS	KOKKOSKERNELS

kokkos-devops-admin · 2022-11-17T06:46:20Z

Status Flag 'Pre-Merge Inspection' - SUCCESS: The last commit to this Pull Request has been INSPECTED AND APPROVED by [ vqd8a lucbv ]!

kokkos-devops-admin · 2022-11-17T06:46:27Z

Status Flag 'Pull Request AutoTester' - Pull Request MUST BE MERGED MANUALLY BY Project Team - This Repo does not support Automerge

lucbv · 2022-11-17T14:45:29Z

@e10harvey re-launching the github actions so we can have all the checks passing.
Regarding the register pressure, I do not know if you can easily read up on it but it is something that AMD mentioned more than once during hackathon and presentations about the MI100, I think it is a bit better on MI200 and higher...

ndellingwood · 2022-11-29T20:41:36Z

Candidate cherry-pick for release-candidate-3.7.01

batched/dense: Reduce register pressure (cherry picked from commit a131d8b)

ndellingwood · 2022-11-29T21:53:41Z

Cherry-picked to release-candidate-3.7.01 SHA 9b3b045

e10harvey requested review from ndellingwood and lucbv November 15, 2022 00:08

e10harvey added the AT: WIP label Nov 15, 2022

e10harvey force-pushed the issue1547 branch from 2e1a6fb to 05c6a9b Compare November 15, 2022 21:32

batched/dense: Reduce register pressure on MI100

95c88e9

e10harvey force-pushed the issue1547 branch from 05c6a9b to 95c88e9 Compare November 15, 2022 21:41

e10harvey requested a review from vqd8a November 15, 2022 21:42

vqd8a reviewed Nov 15, 2022

View reviewed changes

vqd8a approved these changes Nov 15, 2022

View reviewed changes

lucbv approved these changes Nov 16, 2022

View reviewed changes

e10harvey removed the AT: WIP label Nov 16, 2022

e10harvey merged commit a131d8b into kokkos:develop Nov 17, 2022

ndellingwood added the patch-release-cherry-pick label Nov 29, 2022

ndellingwood pushed a commit that referenced this pull request Nov 29, 2022

Merge pull request #1588 from e10harvey/issue1547

9b3b045

batched/dense: Reduce register pressure (cherry picked from commit a131d8b)

	view_type_2d_scratch svA_scr(member.team_scratch(0), TILE_M, TILE_K);
	view_type_2d_scratch svB_scr(member.team_scratch(0), TILE_K, TILE_N);

	constexpr int reg_m = TILE_M / TILE_K;
	constexpr int reg_n = TILE_N / TILE_K + 2 * !!(TILE_N % TILE_K);

batched/dense: Reduce register pressure #1588

batched/dense: Reduce register pressure #1588

Conversation

e10harvey commented Nov 15, 2022

kokkos-devops-admin commented Nov 15, 2022

kokkos-devops-admin commented Nov 15, 2022

Build Information

Test Name: KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020_Light_LayoutRight

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_GCC1020

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_INTEL19

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_CLANG1001

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_CLANG13CUDA10

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020

Jenkins Parameters

Using Repos:

kokkos-devops-admin commented Nov 15, 2022

Build Information

Test Name: KokkosKernels_PullRequest_GCC930_Light_Tpls_GCC930

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_CUDA11_CUDA11_LayoutRight

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_GCC1020_Light_LayoutRight

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_GCC1020

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_INTEL19

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_CLANG1001

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_CLANG13CUDA10

Jenkins Parameters

Build Information

Test Name: KokkosKernels_PullRequest_Tpls_ARMPL2110_Tpls_ARMPL2030_GCC1020

Jenkins Parameters

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

Reproducer instructions:

Load modules:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

Reproducer instructions:

Load modules:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

Reproducer instructions:

Load modules:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

Reproducer instructions:

Load modules:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following:

Reproducer instructions:

Load modules:

To reload modules, reconfigure, rebuild, and retest directly from this failing build do the following: