Add new environment variable to control the use of copy engine for in-order queues

Author: asudarsa

sycl/doc/EnvironmentVariables.md

@@ -33,6 +33,7 @@ subject to change. Do not rely on these variables in production code.
 | `SYCL_PI_LEVEL_ZERO_FILTER_EVENT_WAIT_LIST` | Integer | When set to 0, disables filtering of signaled events from wait lists when using the Level Zero backend. The default is 1. |
 | `SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE` | Integer | Allows the use of copy engine, if available in the device, in Level Zero plugin to transfer SYCL buffer or image data between the host and/or device(s) and to fill SYCL buffer or image data in device or shared memory. The default is 1. |
 | `SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE_FOR_D2D_COPY` (experimental) | Integer | Allows the use of copy engine, if available in the device, in Level Zero plugin for device to device copy operations. The default is 0. This option is experimental and will be removed once heuristics are added to make a decision about use of copy engine for device to device copy operations. |
+| `SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE_FOR_IN_ORDER_QUEUE` (experimental) | Integer | Allows the use of copy engine, if available in the device, in Level Zero plugin for copy operations submitted to an in-order queue. The default is 0. |
 | `SYCL_PI_LEVEL_ZERO_TRACK_INDIRECT_ACCESS_MEMORY` | Any(\*) | Enable support of the kernels with indirect access and corresponding deferred release of memory allocations in the Level Zero plugin. |
 | `SYCL_PARALLEL_FOR_RANGE_ROUNDING_TRACE` | Any(\*) | Enables tracing of `parallel_for` invocations with rounded-up ranges. |
 | `SYCL_DISABLE_PARALLEL_FOR_RANGE_ROUNDING` | Any(\*) | Disables automatic rounding-up of `parallel_for` invocation ranges. |

sycl/plugins/level_zero/pi_level_zero.cpp

@@ -60,6 +60,15 @@ static const bool UseCopyEngineForD2DCopy = [] {
   return (CopyEngineForD2DCopy && (std::stoi(CopyEngineForD2DCopy) != 0));
 }();
 
+// This is an experimental option to test performance of use copy engine for
+// in order queue copy operations on copy engines (versus compute engine)
+static const bool UseCopyEngineForInOrderQueue = [] {
+  const char *CopyEngineForInOrderQueue =
+      std::getenv("SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE_FOR_IN_ORDER_QUEUE");
+  return (CopyEngineForInOrderQueue &&
+          (std::stoi(CopyEngineForInOrderQueue) != 0));
+}();
+
 static const bool CopyEngineRequested = [] {
   const char *CopyEngine = std::getenv("SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE");
   bool UseCopyEngine = (!CopyEngine || (std::stoi(CopyEngine) != 0));