diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 709f2b4f53a..9db8be5cc88 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -113,6 +113,7 @@ build/cuda101/nompi/clang/cuda_wo_omp/release/shared:
     CUDA_ARCH: 35
 
 # Job with example runs.
+# Also explicitly test PAPI SDE
 build/cuda101/openmpi/gcc/all/debug/shared:
   extends:
     - .build_template
@@ -126,6 +127,7 @@ build/cuda101/openmpi/gcc/all/debug/shared:
     MPI_AS_ROOT: "ON"
     BUILD_HIP: "ON"
     BUILD_TYPE: "Debug"
+    BUILD_PAPI_SDE: "ON"
     RUN_EXAMPLES: "ON"
     CUDA_ARCH: 35
 
diff --git a/.gitlab/scripts.yml b/.gitlab/scripts.yml
index 4f699cb53fc..88372236658 100644
--- a/.gitlab/scripts.yml
+++ b/.gitlab/scripts.yml
@@ -44,6 +44,7 @@
         -DGINKGO_BUILD_HIP=${BUILD_HIP}
         -DGINKGO_BUILD_MPI=${BUILD_MPI} ${MPI_STR}
         -DGINKGO_BUILD_HWLOC=${BUILD_HWLOC}
+        -DGINKGO_BUILD_PAPI_SDE=${BUILD_PAPI_SDE}
         -DGINKGO_BUILD_TESTS=ON -DGINKGO_BUILD_EXAMPLES=ON
         -DGINKGO_FAST_TESTS=${FAST_TESTS}
         -DGINKGO_TEST_NONDEFAULT_STREAM=${NONDEFAULT_STREAM}
@@ -89,6 +90,7 @@
         -DGINKGO_BUILD_HIP=${BUILD_HIP}
         -DGINKGO_BUILD_MPI=${BUILD_MPI} ${MPI_STR}
         -DGINKGO_BUILD_HWLOC=${BUILD_HWLOC}
+        -DGINKGO_BUILD_PAPI_SDE=${BUILD_PAPI_SDE}
         -DGINKGO_BUILD_TESTS=ON -DGINKGO_BUILD_EXAMPLES=ON
         -DGINKGO_FAST_TESTS=${FAST_TESTS}
         -DGINKGO_MIXED_PRECISION=${MIXED_PRECISION}
diff --git a/.gitlab/variables.yml b/.gitlab/variables.yml
index 183bdef9e4e..2316b5abc71 100644
--- a/.gitlab/variables.yml
+++ b/.gitlab/variables.yml
@@ -12,6 +12,7 @@
     BUILD_CUDA: "OFF"
     BUILD_HIP: "OFF"
     BUILD_HWLOC: "ON"
+    BUILD_PAPI_SDE: "OFF"
     BUILD_MPI: "OFF"
     GKO_COMPILER_FLAGS: ""
     MPI_AS_ROOT: "OFF"