[SYCL] Add <iostream> headers to tests which assumed "CL/sycl.hpp" would have it included (#1084)

The purpose of this PR is support the changes made in PR intel/llvm#6469 , which gets rid of #include <iostream> from sycl headers and is assumed to be included in the "sycl/sycl.hpp" header for some of the tests present in this repository

Author: raaiq1

SYCL/AtomicRef/atomic_memory_order.cpp

@@ -11,6 +11,7 @@
 
 #include "atomic_memory_order.h"
 #include <cassert>
+#include <iostream>
 using namespace sycl;
 
 int main() {

SYCL/BFloat16/bfloat16_type.hpp

@@ -1,3 +1,4 @@
+#include <iostream>
 #include <sycl/ext/oneapi/experimental/bfloat16.hpp>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/access_to_subset.cpp

@@ -11,6 +11,7 @@
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/Basic/accessor/accessor.cpp

@@ -12,6 +12,7 @@
 //
 //===----------------------------------------------------------------------===//
 #include <cassert>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 struct IdxID1 {

SYCL/Basic/barrier_order.cpp

@@ -4,6 +4,7 @@
 // RUN: %CPU_RUN_PLACEHOLDER %t.out
 // RUN: %GPU_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <stdlib.h>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/boolean.cpp

@@ -3,9 +3,9 @@
 // RUN: %CPU_RUN_PLACEHOLDER %t.out
 // RUN: %GPU_RUN_PLACEHOLDER %t.out
 // RUN: %ACC_RUN_PLACEHOLDER %t.out
-#include <sycl/sycl.hpp>
-
 #include <cassert>
+#include <iostream>
+#include <sycl/sycl.hpp>
 
 using namespace sycl;
 namespace s = sycl;

SYCL/Basic/buffer/buffer.cpp

@@ -14,10 +14,10 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
-
 #include <cassert>
+#include <iostream>
 #include <memory>
+#include <sycl/sycl.hpp>
 
 using namespace sycl;
 

SYCL/Basic/buffer/buffer_create.cpp

@@ -3,6 +3,7 @@
 // RUN: env ZE_DEBUG=1 %GPU_RUN_PLACEHOLDER %t.out 2> %t1.out; cat %t1.out %GPU_CHECK_PLACEHOLDER
 // UNSUPPORTED: ze_debug-1,ze_debug4
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/Basic/buffer/buffer_full_copy.cpp

@@ -15,6 +15,7 @@
 //===----------------------------------------------------------------------===//
 
 #include <cassert>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 void check_copy_device_to_host(sycl::queue &Queue) {

SYCL/Basic/buffer/buffer_migrate.cpp

@@ -8,6 +8,7 @@
 // root-devices)
 //
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 using namespace sycl;
 

SYCL/Basic/buffer/reinterpret.cpp

@@ -13,9 +13,9 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
-
 #include <climits>
+#include <iostream>
+#include <sycl/sycl.hpp>
 
 // This tests verifies basic cases of using sycl::buffer::reinterpret
 // functionality - changing buffer type and range. This test checks that

SYCL/Basic/diagnostics/handler.cpp

@@ -12,6 +12,7 @@
 //===----------------------------------------------------------------------===//
 
 #include <cassert>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 int main() {

SYCL/Basic/fpga_tests/fpga_dsp_control.cpp

@@ -10,6 +10,7 @@
 // RUN: %clangxx -fsycl -fintelfpga %s -o %t.out
 // RUN: %ACC_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <sycl/ext/intel/fpga_extensions.hpp>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/fpga_tests/fpga_latency_control_lsu.cpp

@@ -10,6 +10,7 @@
 // RUN: %clangxx -fsycl -fintelfpga %s -o %t.out
 // RUN: %ACC_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <sycl/ext/intel/fpga_extensions.hpp>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/fpga_tests/fpga_latency_control_pipe.cpp

@@ -10,6 +10,7 @@
 // RUN: %clangxx -fsycl -fintelfpga %s -o %t.out
 // RUN: %ACC_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <sycl/ext/intel/fpga_extensions.hpp>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/fpga_tests/fpga_lsu.cpp

@@ -9,6 +9,7 @@
 // RUN: %clangxx -fsycl -fintelfpga %s -o %t.out
 // RUN: %ACC_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <sycl/ext/intel/fpga_extensions.hpp>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/half_type.cpp

@@ -12,9 +12,9 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <sycl/sycl.hpp>
-
 #include <cmath>
+#include <iostream>
+#include <sycl/sycl.hpp>
 #include <unordered_set>
 
 using namespace sycl;

SYCL/Basic/handler/interop_task.cpp

@@ -16,6 +16,7 @@
 
 #include "../../helpers.hpp"
 #include <chrono>
+#include <iostream>
 #include <thread>
 
 int main() {

SYCL/Basic/handler/run_on_host_intel.cpp

@@ -10,6 +10,7 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 #include "../../helpers.hpp"

SYCL/Basic/handler/run_on_host_intel2.cpp

@@ -2,6 +2,7 @@
 // RUN: %HOST_RUN_PLACEHOLDER %t.out
 // RUN: %CPU_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 // This tests that early free of command (and, hence, the command group) won't

SYCL/Basic/image/image_accessor_readwrite_half.cpp

@@ -17,10 +17,8 @@
 
 #include <cassert>
 #include <iomanip>
-#include <sycl/sycl.hpp>
-#if DEBUG_OUTPUT
 #include <iostream>
-#endif
+#include <sycl/sycl.hpp>
 
 namespace s = sycl;
 

SYCL/Basic/image/image_max_size.cpp

@@ -9,6 +9,7 @@
 
 // The test checks that 'image' with max allowed sizes is handled correctly.
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 using namespace sycl;
 

SYCL/Basic/image/srgba-read.cpp

@@ -6,6 +6,7 @@
 // UNSUPPORTED: cuda
 // UNSUPPORTED: hip
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/Basic/kernel_bundle/kernel_bundle_api.cpp

@@ -7,6 +7,7 @@
 // -fsycl-device-code-split is not supported for cuda
 // UNSUPPORTED: cuda || hip
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 #include <algorithm>

SYCL/Basic/linear-sub_group.cpp

@@ -16,6 +16,7 @@
 #include <algorithm>
 #include <cstdio>
 #include <cstdlib>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/Basic/parallel_for_disable_range_roundup.cpp

@@ -8,8 +8,8 @@
 // RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple -sycl-std=2020 %s -o %t.out
 // RUN: %GPU_RUN_PLACEHOLDER SYCL_PARALLEL_FOR_RANGE_ROUNDING_TRACE=1 %t.out %GPU_CHECK_PLACEHOLDER --check-prefix CHECK-ENABLED
 
+#include <iostream>
 #include <sycl/sycl.hpp>
-
 using namespace sycl;
 
 range<1> Range1 = {0};

SYCL/Basic/parallel_for_range_roundup.cpp

@@ -1,8 +1,8 @@
 // RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple  %s -o %t.out
 // RUN: env SYCL_PARALLEL_FOR_RANGE_ROUNDING_TRACE=1 %GPU_RUN_PLACEHOLDER %t.out %GPU_CHECK_PLACEHOLDER
 
+#include <iostream>
 #include <sycl/sycl.hpp>
-
 using namespace sycl;
 
 range<1> Range1 = {0};

SYCL/Basic/partition_supported.cpp

@@ -15,8 +15,8 @@ not support the info::partition_affinity_domain provided, an exception with the
 **feature_not_supported error code must be thrown**.
 */
 
+#include <iostream>
 #include <sycl/sycl.hpp>
-
 /** returns true if the device supports a particular affinity domain
  */
 static bool

SYCL/Basic/range_offset_fit_in_int.cpp

@@ -2,6 +2,7 @@
 // RUN: %CPU_RUN_PLACEHOLDER %t.out
 
 #include <climits>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 namespace S = sycl;

SYCL/Basic/sampler/sampler.cpp

@@ -13,6 +13,7 @@
 //===----------------------------------------------------------------------===//
 
 #include <cassert>
+#include <iostream>
 #include <sycl/context.hpp>
 #include <sycl/sycl.hpp>
 

SYCL/Basic/stream/auto_flush.cpp

@@ -13,6 +13,7 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/Basic/stream/stream_max_stmt_exceed.cpp

@@ -4,6 +4,7 @@
 // RUN: %GPU_RUN_ON_LINUX_PLACEHOLDER %t.out %GPU_CHECK_ON_LINUX_PLACEHOLDER
 // RUN: %ACC_RUN_PLACEHOLDER %t.out %ACC_CHECK_PLACEHOLDER
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 #include <cassert>

SYCL/Basic/subdevice_pi.cpp

@@ -6,6 +6,7 @@
 // Intel OpenCL CPU Runtime supports device partition on all (multi-core)
 // platforms. Other devices may not support this.
 
+#include <iostream>
 #include <string>
 #include <sycl/sycl.hpp>
 #include <vector>

SYCL/Basic/use_pinned_host_memory.cpp

@@ -3,6 +3,7 @@
 
 #include <sycl/sycl.hpp>
 
+#include <iostream>
 #include <string>
 
 int main() {

SYCL/DeprecatedFeatures/SpecConsts1.2.1/composite-in-functor.cpp

@@ -13,6 +13,7 @@
 // CHECK-NEXT: 3
 // CHECK-NEXT: 4 : 5
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/DeprecatedFeatures/SpecConsts1.2.1/specialization_constants.cpp

@@ -16,6 +16,7 @@
 
 #include <chrono>
 #include <cstdint>
+#include <iostream>
 #include <random>
 #include <sycl/sycl.hpp>
 

SYCL/DeprecatedFeatures/SpecConsts1.2.1/specialization_constants_negative.cpp

@@ -16,6 +16,7 @@
 
 #include <chrono>
 #include <cstdint>
+#include <iostream>
 #include <random>
 #include <sycl/sycl.hpp>
 

SYCL/DeprecatedFeatures/SpecConsts1.2.1/specialization_constants_override.cpp

@@ -16,6 +16,7 @@
 
 #include <chrono>
 #include <cstdint>
+#include <iostream>
 #include <random>
 #include <sycl/sycl.hpp>
 

SYCL/DeprecatedFeatures/opencl-interop.cpp

@@ -17,6 +17,7 @@
 #include <algorithm>
 #include <cstdio>
 #include <cstdlib>
+#include <iostream>
 #include <numeric>
 #include <sycl/sycl.hpp>
 

SYCL/DeprecatedFeatures/subbuffer_interop.cpp

@@ -16,6 +16,7 @@
 #include <sycl/sycl.hpp>
 
 #include <cassert>
+#include <iostream>
 #include <memory>
 #include <numeric>
 

SYCL/DeviceLib/assert-windows.cpp

@@ -26,6 +26,7 @@
 
 #include <array>
 #include <assert.h>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 using namespace sycl;

SYCL/DeviceLib/assert.cpp

@@ -93,6 +93,7 @@
 
 #include <array>
 #include <assert.h>
+#include <iostream>
 #include <stdlib.h>
 #include <sycl/sycl.hpp>
 #include <sys/types.h>

SYCL/DeviceLib/built-ins/nan.cpp

@@ -5,6 +5,7 @@
 // RUN: %GPU_RUN_PLACEHOLDER %t_gpu.out
 // RUN: %ACC_RUN_PLACEHOLDER %t.out
 
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 #include <cassert>

SYCL/DeviceLib/std_complex_math_fp64_test.cpp

@@ -11,6 +11,7 @@
 #include <array>
 #include <cassert>
 #include <complex>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 #include "math_utils.hpp"

SYCL/DeviceLib/std_complex_math_test.cpp

@@ -11,6 +11,7 @@
 #include <array>
 #include <cassert>
 #include <complex>
+#include <iostream>
 #include <sycl/sycl.hpp>
 
 #include "math_utils.hpp"

SYCL/DiscardEvents/discard_events_usm.cpp

@@ -117,7 +117,7 @@
 // CHECK: The test passed.
 
 #include "discard_events_test_queue_ops.hpp"
-
+#include <iostream>
 int main(int Argc, const char *Argv[]) {
 
   sycl::property_list Props{