[ESIMD] Add ahead-of-time compilation test. (#523)

Signed-off-by: Konstantin S Bobrovsky <[email protected]>

Author: kbobrovs

SYCL/ESIMD/aot_mixed.cpp

@@ -0,0 +1,173 @@
+//==---------------- aot_mixed.cpp  - DPC++ ESIMD on-device test -----------==//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// REQUIRES: gpu
+// UNSUPPORTED: cuda || hip
+// RUN: %clangxx -fsycl -fsycl-targets=spir64_gen -Xsycl-target-backend=spir64_gen "-device gen9" -o %t.sycl.out -DENABLE_SYCL=0 %s
+// RUN: %GPU_RUN_PLACEHOLDER %t.sycl.out
+// RUN: %clangxx -fsycl -fsycl-targets=spir64_gen -Xsycl-target-backend=spir64_gen "-device gen9" -o %t.out %s
+// RUN: %GPU_RUN_PLACEHOLDER %t.out
+// XFAIL: *
+// TODO: remove XFAIL once ocloc support for automatic scalar/vector SPIRV
+// module dispatching is available in public drivers. Also change 'gen9' (safe
+// variant to reliably get unexpected PASS when ocloc is fixed) to '*' (which
+// stresses ocloc).
+
+// This test checks ESIMD ahead-of-time compilation scenarios:
+// 1) When the application contains both SYCL and ESIMD kernel, thus requiring
+//    different GPU back-ends (scalar and vector) to kick-in at compile-time.
+// 2) When the application contains only ESIMD kernel.
+
+#include "esimd_test_utils.hpp"
+
+#include <CL/sycl.hpp>
+#include <iostream>
+#include <sycl/ext/intel/experimental/esimd.hpp>
+
+using namespace cl::sycl;
+
+#ifndef ENABLE_SIMD
+#define ENABLE_SIMD 1
+#endif
+
+#ifndef ENABLE_SYCL
+#define ENABLE_SYCL 1
+#endif
+
+bool verify(float *A, float *B, float *C, size_t Size) {
+  int err_cnt = 0;
+
+  for (unsigned i = 0; i < Size; ++i) {
+    if (A[i] + B[i] != C[i]) {
+      if (++err_cnt < 10) {
+        std::cout << "failed at index " << i << ", " << C[i] << " != " << A[i]
+                  << " + " << B[i] << "\n";
+      }
+    }
+  }
+  if (err_cnt > 0) {
+    std::cout << "  pass rate: "
+              << ((float)(Size - err_cnt) / (float)Size) * 100.0f << "% ("
+              << (Size - err_cnt) << "/" << Size << ")\n";
+  }
+  return err_cnt == 0;
+}
+
+constexpr unsigned Size = 1024 * 128;
+constexpr unsigned VL = 16;
+
+#if ENABLE_SIMD
+bool test_esimd(queue q) {
+  std::cout << "Running ESIMD kernel...\n";
+  float *A = new float[Size];
+  float *B = new float[Size];
+  float *C = new float[Size];
+
+  for (unsigned i = 0; i < Size; ++i) {
+    A[i] = B[i] = i;
+    C[i] = 0.0f;
+  }
+
+  try {
+    buffer<float, 1> bufa(A, range<1>(Size));
+    buffer<float, 1> bufb(B, range<1>(Size));
+    buffer<float, 1> bufc(C, range<1>(Size));
+
+    auto e = q.submit([&](handler &cgh) {
+      auto PA = bufa.get_access<access::mode::read>(cgh);
+      auto PB = bufb.get_access<access::mode::read>(cgh);
+      auto PC = bufc.get_access<access::mode::write>(cgh);
+      cgh.parallel_for<class TestESIMD>(
+          Size / VL, [=](id<1> i) SYCL_ESIMD_KERNEL {
+            using namespace sycl::ext::intel::experimental::esimd;
+            unsigned int offset = i * VL * sizeof(float);
+            simd<float, VL> va;
+            va.copy_from(PA, offset);
+            simd<float, VL> vb;
+            vb.copy_from(PB, offset);
+            simd<float, VL> vc = va + vb;
+            vc.copy_to(PC, offset);
+          });
+    });
+    e.wait();
+  } catch (cl::sycl::exception const &e) {
+    std::cout << "SYCL exception caught: " << e.what() << '\n';
+
+    delete[] A;
+    delete[] B;
+    delete[] C;
+
+    return false;
+  }
+  bool passed = verify(A, B, C, Size);
+
+  delete[] A;
+  delete[] B;
+  delete[] C;
+  return passed;
+}
+#endif
+
+#if ENABLE_SYCL
+bool test_sycl(queue q) {
+  std::cout << "Running SYCL kernel...\n";
+  float *A = new float[Size];
+  float *B = new float[Size];
+  float *C = new float[Size];
+
+  for (unsigned i = 0; i < Size; ++i) {
+    A[i] = B[i] = i;
+    C[i] = 0.0f;
+  }
+
+  try {
+    buffer<float, 1> bufa(A, range<1>(Size));
+    buffer<float, 1> bufb(B, range<1>(Size));
+    buffer<float, 1> bufc(C, range<1>(Size));
+
+    auto e = q.submit([&](handler &cgh) {
+      auto PA = bufa.get_access<access::mode::read>(cgh);
+      auto PB = bufb.get_access<access::mode::read>(cgh);
+      auto PC = bufc.get_access<access::mode::write>(cgh);
+      cgh.parallel_for<class TestSYCL>(Size,
+                                       [=](id<1> i) { PC[i] = PA[i] + PB[i]; });
+    });
+    e.wait();
+  } catch (cl::sycl::exception const &e) {
+    std::cout << "SYCL exception caught: " << e.what() << '\n';
+
+    delete[] A;
+    delete[] B;
+    delete[] C;
+
+    return false;
+  }
+  bool passed = verify(A, B, C, Size);
+
+  delete[] A;
+  delete[] B;
+  delete[] C;
+  return passed;
+}
+#endif
+
+int main(void) {
+  queue q(esimd_test::ESIMDSelector{}, esimd_test::createExceptionHandler());
+  auto dev = q.get_device();
+  std::cout << "Running on " << dev.get_info<info::device::name>() << "\n";
+
+  bool passed = true;
+#if ENABLE_SIMD
+  passed &= test_esimd(q);
+#endif
+#if ENABLE_SYCL
+  passed &= test_sycl(q);
+#endif
+
+  std::cout << (passed ? "TEST Passed\n" : "TEST FAILED\n");
+  return passed ? 0 : 1;
+}