1 files changed, 93 insertions, 0 deletions
diff --git a/tools/cuda-ab/vector_add.cu b/tools/cuda-ab/vector_add.cu
new file mode 100644
index 0000000..2e3f1d6
--- /dev/null
+++ b/tools/cuda-ab/vector_add.cu
@@ -0,0 +1,93 @@
+// SPDX-License-Identifier: BSD-3-Clause
+/*
+ * vector_add.cu - CUDA A/B workload for the Open-IOV harness.
+ *
+ * One deterministic, machine-readable workload that can run on the host (bare
+ * metal) and in a guest (emulated vGPU). The output lines are stable so the
+ * two runs can be diffed:
+ *
+ *   DEVICE <name> cc=<major>.<minor> sm=<count> mem=<bytes>
+ *   RESULT n=<count> bad=<mismatches> checksum=<sum>
+ *   STATUS OK|FAIL
+ *
+ * Build with a static CUDA runtime (--cudart=static) so the guest only needs
+ * the driver's libcuda.so, not a CUDA toolkit installation.
+ */
+
+#include <cstdio>
+#include <cstdlib>
+#include <cuda_runtime.h>
+
+#define CHECK(x)                                                         \
+	do {                                                             \
+		cudaError_t e_ = (x);                                    \
+		if (e_ != cudaSuccess) {                                 \
+			printf("CUDA_ERROR %s at line %d\n",             \
+			       cudaGetErrorString(e_), __LINE__);        \
+			printf("STATUS FAIL\n");                         \
+			return 1;                                        \
+		}                                                        \
+	} while (0)
+
+__global__ void vadd(const float *a, const float *b, float *c, int n)
+{
+	int i = blockIdx.x * blockDim.x + threadIdx.x;
+
+	if (i < n)
+		c[i] = a[i] + b[i];
+}
+
+int main(void)
+{
+	cudaDeviceProp prop;
+	const int n = 1 << 20;
+	size_t bytes = (size_t)n * sizeof(float);
+	float *h_a, *h_b, *h_c;
+	float *d_a, *d_b, *d_c;
+	double sum = 0.0;
+	int bad = 0, i;
+
+	CHECK(cudaGetDeviceProperties(&prop, 0));
+	printf("DEVICE %s cc=%d.%d sm=%d mem=%llu\n", prop.name,
+	       prop.major, prop.minor, prop.multiProcessorCount,
+	       (unsigned long long)prop.totalGlobalMem);
+
+	h_a = (float *)malloc(bytes);
+	h_b = (float *)malloc(bytes);
+	h_c = (float *)malloc(bytes);
+	if (!h_a || !h_b || !h_c)
+		return 1;
+
+	for (i = 0; i < n; i++) {
+		h_a[i] = (float)(i % 1000) * 0.5f;
+		h_b[i] = (float)(i % 997) * 0.25f;
+	}
+
+	CHECK(cudaMalloc(&d_a, bytes));
+	CHECK(cudaMalloc(&d_b, bytes));
+	CHECK(cudaMalloc(&d_c, bytes));
+	CHECK(cudaMemcpy(d_a, h_a, bytes, cudaMemcpyHostToDevice));
+	CHECK(cudaMemcpy(d_b, h_b, bytes, cudaMemcpyHostToDevice));
+
+	vadd<<<(n + 255) / 256, 256>>>(d_a, d_b, d_c, n);
+	CHECK(cudaGetLastError());
+	CHECK(cudaDeviceSynchronize());
+	CHECK(cudaMemcpy(h_c, d_c, bytes, cudaMemcpyDeviceToHost));
+
+	for (i = 0; i < n; i++) {
+		if (h_c[i] != h_a[i] + h_b[i])
+			bad++;
+		sum += h_c[i];
+	}
+
+	printf("RESULT n=%d bad=%d checksum=%.6f\n", n, bad, sum);
+	printf("STATUS %s\n", bad == 0 ? "OK" : "FAIL");
+
+	CHECK(cudaFree(d_a));
+	CHECK(cudaFree(d_b));
+	CHECK(cudaFree(d_c));
+	free(h_a);
+	free(h_b);
+	free(h_c);
+	return bad == 0 ? 0 : 1;
+}