[SYCL] Use raw pointers in enqueue flow (#18306)

vinser52 · web-flow · commit fc7429e7d04c · 2025-05-06T08:58:10.000-07:00
For `kernel_impl` parameter - it's an optional input only parameter that
is owned by the caller for the duration of the call and we don't extend
its lifetime/create copies as can be seen by the absence of changes in
the callee (all we need is to access its data).

For `event_impl` "out" parameter the semantics is such that the caller
creates an object, and the modified interfaces update that object with
the UR handles once enqueue/launch is done. The ownership remains
controlled by the caller.

---------

Signed-off-by: Sergei Vinogradov &lt;sergey.vinogradov@intel.com&gt;
diff --git a/sycl/source/detail/kernel_impl.hpp b/sycl/source/detail/kernel_impl.hpp
@@ -231,7 +231,7 @@ class kernel_impl {
   ur_program_handle_t getProgramRef() const { return MProgram; }
   ContextImplPtr getContextImplPtr() const { return MContext; }
 
-  std::mutex &getNoncacheableEnqueueMutex() {
+  std::mutex &getNoncacheableEnqueueMutex() const {
     return MNoncacheableEnqueueMutex;
   }
 
@@ -247,7 +247,7 @@ class kernel_impl {
   const DeviceImageImplPtr MDeviceImageImpl;
   const KernelBundleImplPtr MKernelBundleImpl;
   bool MIsInterop = false;
-  std::mutex MNoncacheableEnqueueMutex;
+  mutable std::mutex MNoncacheableEnqueueMutex;
   const KernelArgMask *MKernelArgMaskPtr;
   std::mutex *MCacheMutex = nullptr;
   mutable std::string MName;
diff --git a/sycl/source/detail/scheduler/commands.cpp b/sycl/source/detail/scheduler/commands.cpp
@@ -2380,8 +2380,7 @@ static ur_result_t SetKernelParamsAndLaunch(
     const QueueImplPtr &Queue, std::vector<ArgDesc> &Args,
     const std::shared_ptr<device_image_impl> &DeviceImageImpl,
     ur_kernel_handle_t Kernel, NDRDescT &NDRDesc,
-    std::vector<ur_event_handle_t> &RawEvents,
-    const detail::EventImplPtr &OutEventImpl,
+    std::vector<ur_event_handle_t> &RawEvents, detail::event_impl *OutEventImpl,
     const KernelArgMask *EliminatedArgMask,
     const std::function<void *(Requirement *Req)> &getMemAllocationFunc,
     bool IsCooperative, bool KernelUsesClusterLaunch,
@@ -2651,9 +2650,8 @@ ur_result_t enqueueImpCommandBufferKernel(
 void enqueueImpKernel(
     const QueueImplPtr &Queue, NDRDescT &NDRDesc, std::vector<ArgDesc> &Args,
     const std::shared_ptr<detail::kernel_bundle_impl> &KernelBundleImplPtr,
-    const std::shared_ptr<detail::kernel_impl> &MSyclKernel,
-    KernelNameStrRefT KernelName, std::vector<ur_event_handle_t> &RawEvents,
-    const detail::EventImplPtr &OutEventImpl,
+    const detail::kernel_impl *MSyclKernel, KernelNameStrRefT KernelName,
+    std::vector<ur_event_handle_t> &RawEvents, detail::event_impl *OutEventImpl,
     const std::function<void *(Requirement *Req)> &getMemAllocationFunc,
     ur_kernel_cache_config_t KernelCacheConfig, const bool KernelIsCooperative,
     const bool KernelUsesClusterLaunch, const size_t WorkGroupMemorySize,
@@ -2761,7 +2759,7 @@ ur_result_t enqueueReadWriteHostPipe(const QueueImplPtr &Queue,
                                      const std::string &PipeName, bool blocking,
                                      void *ptr, size_t size,
                                      std::vector<ur_event_handle_t> &RawEvents,
-                                     const detail::EventImplPtr &OutEventImpl,
+                                     detail::event_impl *OutEventImpl,
                                      bool read) {
   assert(Queue &&
          "ReadWrite host pipe submissions should have an associated queue");
@@ -3122,7 +3120,7 @@ ur_result_t ExecCGCommand::enqueueImpQueue() {
 
   ur_event_handle_t UREvent = nullptr;
   ur_event_handle_t *Event = DiscardUrEvent ? nullptr : &UREvent;
-  detail::EventImplPtr EventImpl = DiscardUrEvent ? nullptr : MEvent;
+  detail::event_impl *EventImpl = DiscardUrEvent ? nullptr : MEvent.get();
 
   auto SetEventHandleOrDiscard = [&]() {
     if (Event)
@@ -3237,7 +3235,7 @@ ur_result_t ExecCGCommand::enqueueImpQueue() {
           (!SyclKernel || SyclKernel->hasSYCLMetadata()) &&
           ProgramManager::getInstance().kernelUsesAssert(KernelName);
       if (KernelUsesAssert) {
-        EventImpl = MEvent;
+        EventImpl = MEvent.get();
       }
     }
 
@@ -3248,7 +3246,7 @@ ur_result_t ExecCGCommand::enqueueImpQueue() {
       assert(BinImage && "Failed to obtain a binary image.");
     }
     enqueueImpKernel(MQueue, NDRDesc, Args, ExecKernel->getKernelBundle(),
-                     SyclKernel, KernelName, RawEvents, EventImpl,
+                     SyclKernel.get(), KernelName, RawEvents, EventImpl,
                      getMemAllocationFunc, ExecKernel->MKernelCacheConfig,
                      ExecKernel->MKernelIsCooperative,
                      ExecKernel->MKernelUsesClusterLaunch,
@@ -3645,7 +3643,7 @@ ur_result_t ExecCGCommand::enqueueImpQueue() {
     bool read = ExecReadWriteHostPipe->isReadHostPipe();
 
     if (!EventImpl) {
-      EventImpl = MEvent;
+      EventImpl = MEvent.get();
     }
     return enqueueReadWriteHostPipe(MQueue, pipeName, blocking, hostPtr,
                                     typeSize, RawEvents, EventImpl, read);
diff --git a/sycl/source/detail/scheduler/commands.hpp b/sycl/source/detail/scheduler/commands.hpp
@@ -614,15 +614,14 @@ ur_result_t enqueueReadWriteHostPipe(const QueueImplPtr &Queue,
                                      const std::string &PipeName, bool blocking,
                                      void *ptr, size_t size,
                                      std::vector<ur_event_handle_t> &RawEvents,
-                                     const detail::EventImplPtr &OutEventImpl,
+                                     detail::event_impl *OutEventImpl,
                                      bool read);
 
 void enqueueImpKernel(
     const QueueImplPtr &Queue, NDRDescT &NDRDesc, std::vector<ArgDesc> &Args,
     const std::shared_ptr<detail::kernel_bundle_impl> &KernelBundleImplPtr,
-    const std::shared_ptr<detail::kernel_impl> &MSyclKernel,
-    KernelNameStrRefT KernelName, std::vector<ur_event_handle_t> &RawEvents,
-    const detail::EventImplPtr &Event,
+    const detail::kernel_impl *MSyclKernel, KernelNameStrRefT KernelName,
+    std::vector<ur_event_handle_t> &RawEvents, detail::event_impl *OutEventImpl,
     const std::function<void *(Requirement *Req)> &getMemAllocationFunc,
     ur_kernel_cache_config_t KernelCacheConfig, bool KernelIsCooperative,
     const bool KernelUsesClusterLaunch, const size_t WorkGroupMemorySize,
diff --git a/sycl/source/handler.cpp b/sycl/source/handler.cpp
@@ -537,13 +537,13 @@ event handler::finalize() {
               detail::retrieveKernelBinary(MQueue, MKernelName.data());
           assert(BinImage && "Failed to obtain a binary image.");
         }
-        enqueueImpKernel(
-            MQueue, impl->MNDRDesc, impl->MArgs, KernelBundleImpPtr, MKernel,
-            MKernelName.data(), RawEvents,
-            DiscardEvent ? detail::EventImplPtr{} : LastEventImpl, nullptr,
-            impl->MKernelCacheConfig, impl->MKernelIsCooperative,
-            impl->MKernelUsesClusterLaunch, impl->MKernelWorkGroupMemorySize,
-            BinImage);
+        enqueueImpKernel(MQueue, impl->MNDRDesc, impl->MArgs,
+                         KernelBundleImpPtr, MKernel.get(), MKernelName.data(),
+                         RawEvents,
+                         DiscardEvent ? nullptr : LastEventImpl.get(), nullptr,
+                         impl->MKernelCacheConfig, impl->MKernelIsCooperative,
+                         impl->MKernelUsesClusterLaunch,
+                         impl->MKernelWorkGroupMemorySize, BinImage);
 #ifdef XPTI_ENABLE_INSTRUMENTATION
         // Emit signal only when event is created
         if (!DiscardEvent) {
