[ESIMD] Add smoke test for simd_obj_impl::any/all.

SYCL/ESIMD/api/simd_any_all.cpp

@@ -0,0 +1,153 @@
+//==------- esimd_any_all.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 %s -fsycl-device-code-split=per_kernel -o %t.out
+// RUN: %GPU_RUN_PLACEHOLDER %t.out
+//
+// Smoke test for esimd any/all operations APIs.
+
+#include "../esimd_test_utils.hpp"
+
+#include <CL/sycl.hpp>
+#include <sycl/ext/intel/experimental/esimd.hpp>
+
+#include <iostream>
+
+using namespace cl::sycl;
+using namespace sycl::ext::intel::experimental;
+using namespace sycl::ext::intel::experimental::esimd;
+
+struct bit_op {
+  enum { any, all, num_ops };
+};
+
+template <typename T, int N, int Op> struct test_id;
+
+template <typename T> struct char_to_int {
+  using type = typename std::conditional<
+      sizeof(T) == 1,
+      typename std::conditional<std::is_signed<T>::value, int, unsigned>::type,
+      T>::type;
+};
+
+// The main test routine.
+template <typename T, int N, int Op> bool test_impl(queue q) {
+  const char *ops[bit_op::num_ops] = {"any", "all"};
+  std::cout << "Testing op=" << ops[Op] << " T=" << typeid(T).name()
+            << ", N=" << N << "...\n";
+
+  simd<T, N> all_zero((T)0);
+  simd<T, N> all_one((T)1);
+  if (std::is_signed_v<T>) {
+    all_one[0] = -1;
+  }
+  simd<T, N> all_two((T)2); // check that non-zero with LSB=0 counts as 'set'
+  if (std::is_signed_v<T>) {
+    all_two[N - 1] = -2;
+  }
+  simd<T, N> zero_two((T)0);
+
+  if (N > 1) {
+    zero_two[1] = 2;
+  }
+
+  simd<T, N> test_vals_arr[] = {all_zero, all_one, all_two, zero_two};
+  constexpr size_t num_vals = sizeof(test_vals_arr) / sizeof(test_vals_arr[0]);
+  T *test_vals = sycl::malloc_shared<T>(num_vals * N, q);
+  uint16_t *res = sycl::malloc_shared<uint16_t>(num_vals, q);
+
+  for (unsigned int i = 0; i < num_vals; ++i) {
+    res[i] = 0xFFff;
+  }
+  memcpy(test_vals, test_vals_arr, sizeof(test_vals_arr));
+
+  try {
+    auto e = q.submit([&](handler &cgh) {
+      cgh.single_task<test_id<T, N, Op>>([=]() SYCL_ESIMD_KERNEL {
+        for (int i = 0; i < num_vals; i++) {
+          simd<T, N> src(test_vals + i * N);
+
+          if constexpr (Op == bit_op::any) {
+            res[i] = src.any();
+          } else if constexpr (Op == bit_op::all) {
+            res[i] = src.all();
+          }
+        }
+      });
+    });
+    e.wait_and_throw();
+  } catch (cl::sycl::exception const &e) {
+    std::cout << "  SYCL exception caught: " << e.what() << '\n';
+    sycl::free(res, q);
+    sycl::free(test_vals, q);
+    return false;
+  }
+
+  unsigned int Gold[num_vals * bit_op::num_ops] = {
+      // any:
+      0, // all zero
+      1, // all one
+      1, // all two
+      1, // zero, two
+      // all:
+      0, // all zero
+      1, // all one
+      1, // all two
+      0, // zero, two
+  };
+  int err_cnt = 0;
+
+  using ValTy = typename char_to_int<T>::type;
+
+  for (unsigned i = 0; i < num_vals; ++i) {
+    if ((N == 1) && (i == 3)) {
+      continue; // (zero, two) testcase not available for single element
+    }
+    T gold = Gold[Op * num_vals + i];
+    T val = res[i];
+    std::cout << "  " << ops[Op] << "(" << (simd<ValTy, N>)test_vals_arr[i]
+              << ") = " << (ValTy)val;
+
+    if (val != gold) {
+      ++err_cnt;
+      std::cout << " ERROR. " << (ValTy)val << " != " << (ValTy)gold
+                << "(gold)\n";
+    } else {
+      std::cout << " (ok)\n";
+    }
+  }
+  std::cout << (err_cnt > 0 ? "  FAILED\n" : "  Passed\n");
+  sycl::free(res, q);
+  sycl::free(test_vals, q);
+  return err_cnt > 0 ? false : true;
+}
+
+template <class T1, class T2> bool test(queue q) {
+  bool passed = true;
+  passed &= test_impl<T1, 32, bit_op::any>(q);
+  passed &= test_impl<T1, 8, bit_op::all>(q);
+  passed &= test_impl<T2, 3, bit_op::any>(q);
+  passed &= test_impl<T2, 1, bit_op::all>(q);
+  return passed;
+}
+
+int main(int argc, char **argv) {
+  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;
+  passed &= test<int8_t, uint8_t>(q);
+  passed &= test<int16_t, uint16_t>(q);
+  passed &= test<int32_t, uint32_t>(q);
+  passed &= test<int64_t, uint64_t>(q);
+
+  std::cout << (passed ? "=== Test passed\n" : "=== Test FAILED\n");
+  return passed ? 0 : 1;
+}