diff --git a/CHANGELOG.md b/CHANGELOG.md
index a990471daa..9f72a99a22 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -21,6 +21,9 @@ Increment the:
   cmake when using C++20 [#1852](https://github.com/open-telemetry/opentelemetry-cpp/pull/1852)
 * [SEMANTIC CONVENTIONS] Upgrade to version 1.16.0
   [#1854](https://github.com/open-telemetry/opentelemetry-cpp/pull/1854)
+* [SDK] BatchSpanProcessor now logs a warning when dropping a span because the
+  queue is full
+  [1871](https://github.com/open-telemetry/opentelemetry-cpp/pull/1871)
 
 ## [1.8.1] 2022-12-04
 
diff --git a/api/include/opentelemetry/common/spin_lock_mutex.h b/api/include/opentelemetry/common/spin_lock_mutex.h
index 764b5fc6b8..1877c8eb40 100644
--- a/api/include/opentelemetry/common/spin_lock_mutex.h
+++ b/api/include/opentelemetry/common/spin_lock_mutex.h
@@ -18,6 +18,8 @@
 #elif defined(__i386__) || defined(__x86_64__)
 #  if defined(__clang__)
 #    include <emmintrin.h>
+#  elif defined(__INTEL_COMPILER)
+#    include <immintrin.h>
 #  endif
 #endif
 
@@ -65,7 +67,7 @@ class SpinLockMutex
 #if defined(_MSC_VER)
     YieldProcessor();
 #elif defined(__i386__) || defined(__x86_64__)
-#  if defined(__clang__)
+#  if defined(__clang__) || defined(__INTEL_COMPILER)
     _mm_pause();
 #  else
     __builtin_ia32_pause();
diff --git a/api/test/common/spinlock_benchmark.cc b/api/test/common/spinlock_benchmark.cc
index 50a8bade73..53ea633d16 100644
--- a/api/test/common/spinlock_benchmark.cc
+++ b/api/test/common/spinlock_benchmark.cc
@@ -87,7 +87,7 @@ static void BM_ProcYieldSpinLockThrashing(benchmark::State &s)
 #if defined(_MSC_VER)
           YieldProcessor();
 #elif defined(__i386__) || defined(__x86_64__)
-#  if defined(__clang__)
+#  if defined(__clang__) || defined(__INTEL_COMPILER)
           _mm_pause();
 #  else
           __builtin_ia32_pause();
diff --git a/sdk/src/trace/batch_span_processor.cc b/sdk/src/trace/batch_span_processor.cc
index 4609eae95f..de6c457da7 100644
--- a/sdk/src/trace/batch_span_processor.cc
+++ b/sdk/src/trace/batch_span_processor.cc
@@ -3,6 +3,7 @@
 
 #include "opentelemetry/sdk/trace/batch_span_processor.h"
 #include "opentelemetry/common/spin_lock_mutex.h"
+#include "opentelemetry/sdk/common/global_log_handler.h"
 
 #include <vector>
 using opentelemetry::sdk::common::AtomicUniquePtr;
@@ -51,6 +52,7 @@ void BatchSpanProcessor::OnEnd(std::unique_ptr<Recordable> &&span) noexcept
 
   if (buffer_.Add(span) == false)
   {
+    OTEL_INTERNAL_LOG_WARN("BatchSpanProcessor queue is full - dropping span.");
     return;
   }
 
diff --git a/sdk/test/trace/batch_span_processor_test.cc b/sdk/test/trace/batch_span_processor_test.cc
index ad981cafb6..744d80775d 100644
--- a/sdk/test/trace/batch_span_processor_test.cc
+++ b/sdk/test/trace/batch_span_processor_test.cc
@@ -2,10 +2,12 @@
 // SPDX-License-Identifier: Apache-2.0
 
 #include "opentelemetry/sdk/trace/batch_span_processor.h"
+#include "opentelemetry/sdk/common/global_log_handler.h"
 #include "opentelemetry/sdk/trace/span_data.h"
 #include "opentelemetry/sdk/trace/tracer.h"
 
 #include <gtest/gtest.h>
+#include <algorithm>
 #include <chrono>
 #include <list>
 #include <memory>
@@ -182,10 +184,31 @@ TEST_F(BatchSpanProcessorTestPeer, TestForceFlush)
   }
 }
 
+// A mock log handler to check whether log messages with a specific level were emitted.
+struct MockLogHandler : public sdk::common::internal_log::LogHandler
+{
+  using Message = std::pair<sdk::common::internal_log::LogLevel, std::string>;
+
+  void Handle(sdk::common::internal_log::LogLevel level,
+              const char * /*file*/,
+              int /*line*/,
+              const char *msg,
+              const sdk::common::AttributeMap & /*attributes*/) noexcept override
+  {
+    messages.emplace_back(level, msg);
+  }
+
+  std::vector<Message> messages;
+};
+
 TEST_F(BatchSpanProcessorTestPeer, TestManySpansLoss)
 {
   /* Test that when exporting more than max_queue_size spans, some are most likely lost*/
 
+  // Set up a log handler to verify a warning is generated.
+  auto log_handler = nostd::shared_ptr<sdk::common::internal_log::LogHandler>(new MockLogHandler());
+  sdk::common::internal_log::GlobalLogHandler::SetLogHandler(log_handler);
+
   std::shared_ptr<std::atomic<bool>> is_shutdown(new std::atomic<bool>(false));
   std::shared_ptr<std::vector<std::unique_ptr<sdk::trace::SpanData>>> spans_received(
       new std::vector<std::unique_ptr<sdk::trace::SpanData>>);
@@ -211,6 +234,25 @@ TEST_F(BatchSpanProcessorTestPeer, TestManySpansLoss)
 
   // Span should be exported by now
   EXPECT_GE(max_queue_size, spans_received->size());
+
+  // If we haven't received all spans, some must have dropped, verify a warning was logged.
+  // Only do this when the log level is warning or above.
+#if OTEL_INTERNAL_LOG_LEVEL >= OTEL_INTERNAL_LOG_LEVEL_WARN
+  if (max_queue_size > spans_received->size())
+  {
+    auto &messages = static_cast<MockLogHandler *>(log_handler.get())->messages;
+    EXPECT_TRUE(
+        std::find(messages.begin(), messages.end(),
+                  MockLogHandler::Message(sdk::common::internal_log::LogLevel::Warning,
+                                          "BatchSpanProcessor queue is full - dropping span.")) !=
+        messages.end());
+  }
+#endif
+
+  // Reinstate the default log handler.
+  sdk::common::internal_log::GlobalLogHandler::SetLogHandler(
+      nostd::shared_ptr<sdk::common::internal_log::LogHandler>(
+          new sdk::common::internal_log::DefaultLogHandler()));
 }
 
 TEST_F(BatchSpanProcessorTestPeer, TestManySpansLossLess)