diff --git a/opencl/source/command_queue/enqueue_common.h b/opencl/source/command_queue/enqueue_common.h
index ccb132dacb..dbbfa5a08b 100644
--- a/opencl/source/command_queue/enqueue_common.h
+++ b/opencl/source/command_queue/enqueue_common.h
@@ -236,6 +236,10 @@ void CommandQueueHw<GfxFamily>::enqueueHandler(Surface **surfacesForResidency,
             }
         }
 
+        if (isMarkerWithProfiling) {
+            flushDependenciesForNonKernelCommand = true;
+        }
+
         if (flushDependenciesForNonKernelCommand) {
             TimestampPacketHelper::programCsrDependenciesForTimestampPacketContainer<GfxFamily>(commandStream, csrDeps);
         }
diff --git a/opencl/test/unit_test/command_queue/blit_enqueue_tests.cpp b/opencl/test/unit_test/command_queue/blit_enqueue_tests.cpp
index 700344f00b..e7e9d10507 100644
--- a/opencl/test/unit_test/command_queue/blit_enqueue_tests.cpp
+++ b/opencl/test/unit_test/command_queue/blit_enqueue_tests.cpp
@@ -1541,6 +1541,33 @@ HWTEST_TEMPLATED_F(BlitEnqueueTaskCountTests, givenEventFromCpuCopyWhenWaitingFo
     clReleaseEvent(outEvent2);
 }
 
+HWTEST_TEMPLATED_F(BlitEnqueueTaskCountTests, givenMarkerThatFollowsCopyOperationWhenItIsWaitedItHasProperDependencies) {
+    auto buffer = createBuffer(1, false);
+    int hostPtr = 0;
+
+    using MI_SEMAPHORE_WAIT = typename FamilyType::MI_SEMAPHORE_WAIT;
+
+    auto mockCmdQueue = static_cast<MockCommandQueueHw<FamilyType> *>(commandQueue.get());
+    mockCmdQueue->commandQueueProperties |= CL_QUEUE_PROFILING_ENABLE;
+
+    cl_event outEvent1;
+    commandQueue->enqueueKernel(mockKernel->mockKernel, 1, nullptr, gws, nullptr, 0, nullptr, nullptr);
+
+    commandQueue->enqueueWriteBuffer(buffer.get(), false, 0, 1, &hostPtr, nullptr, 0, nullptr, nullptr);
+
+    auto offset = mockCmdQueue->getCS(0).getUsed();
+
+    //marker needs to program semaphore
+    commandQueue->enqueueMarkerWithWaitList(0, nullptr, &outEvent1);
+
+    auto cmdListQueue = getCmdList<FamilyType>(mockCmdQueue->getCS(0), offset);
+    expectCommand<MI_SEMAPHORE_WAIT>(cmdListQueue.begin(), cmdListQueue.end());
+
+    clWaitForEvents(1, &outEvent1);
+
+    clReleaseEvent(outEvent1);
+}
+
 using BlitEnqueueWithDisabledGpgpuSubmissionTests = BlitEnqueueTests<1>;
 
 HWTEST_TEMPLATED_F(BlitEnqueueWithDisabledGpgpuSubmissionTests, givenCacheFlushRequiredWhenDoingBcsCopyThenSubmitToGpgpuOnlyIfPreviousEnqueueWasGpgpu) {