diff --git a/tests/compiler/CMakeLists.txt b/tests/compiler/CMakeLists.txt
index 86b93f60f..c52625c76 100644
--- a/tests/compiler/CMakeLists.txt
+++ b/tests/compiler/CMakeLists.txt
@@ -119,6 +119,7 @@ add_hipcc_test(TestLdg.hip HIPCC_OPTIONS -fsyntax-only)
 add_hipcc_test(TestSwitchCase.hip HIPCC_OPTIONS -O1 -c)
 add_hipcc_test(TestHostSideHIPVectors.hip HIPCC_OPTIONS -fsyntax-only)
 add_hipcc_test(TestAlignAttr.hip HIPCC_OPTIONS -fsyntax-only)
+add_hipcc_test(TestHipccInvalidBitWidth.cpp HIPCC_OPTIONS -O2)
 
 # Check __FAST_MATH__ is set for -ffast-math and preprocessor guards
 # using it are not hiding errors.
diff --git a/tests/compiler/TestHipccInvalidBitWidth.cpp b/tests/compiler/TestHipccInvalidBitWidth.cpp
new file mode 100644
index 000000000..194f40cb6
--- /dev/null
+++ b/tests/compiler/TestHipccInvalidBitWidth.cpp
@@ -0,0 +1,59 @@
+#include <hip/hip_runtime.h>
+#include <stdio.h>
+
+__global__ void testWarpCalc(int* debug) {
+    int tid = threadIdx.x;
+    int bid = blockIdx.x;
+    int globalIdx = bid * blockDim.x + tid;
+    
+    // Do some computation to prevent optimization
+    int result = 0;
+    for(int i = 0; i < tid + 1; i++) {
+        result += i * globalIdx;
+    }
+    
+    // Store using atomic operation
+    atomicExch(&debug[globalIdx], result);
+}
+
+int main() {
+    const int gridSize = 4;
+    const int blockSize = 64;
+    const int numThreads = gridSize * blockSize;
+
+    // Allocate pinned memory
+    int* h_debug;
+    hipHostMalloc(&h_debug, numThreads * sizeof(int));
+    memset(h_debug, 0, numThreads * sizeof(int));
+
+    // Allocate device memory
+    int* d_debug;
+    hipMalloc(&d_debug, numThreads * sizeof(int));
+    hipMemset(d_debug, 0, numThreads * sizeof(int));
+
+    dim3 grid(gridSize);
+    dim3 block(blockSize);
+    
+    printf("Launching kernel with grid=%d, block=%d\n\n", gridSize, blockSize);
+    
+    // Use triple angle bracket syntax
+    testWarpCalc<<<grid, block>>>(d_debug);
+    hipDeviceSynchronize();
+    
+    // Copy results back
+    hipMemcpy(h_debug, d_debug, numThreads * sizeof(int), hipMemcpyDeviceToHost);
+    
+    printf("Results for first few threads:\n");
+    printf("GlobalIdx\tValue\n");
+    
+    // Print first few entries
+    for (int i = 0; i < 8; i++) {
+        printf("%d\t\t%d\n", i, h_debug[i]);
+    }
+
+    // Cleanup
+    hipHostFree(h_debug);
+    hipFree(d_debug);
+    
+    return 0;
+} 
\ No newline at end of file