diff --git a/libcxxabi/test/test_exception_storage.pass.cpp b/libcxxabi/test/test_exception_storage.pass.cpp
--- a/libcxxabi/test/test_exception_storage.pass.cpp
+++ b/libcxxabi/test/test_exception_storage.pass.cpp
@@ -6,71 +6,19 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include <algorithm>
-#include <cstdio>
-#include <cstdlib>
-#include <__threading_support>
-#include <unistd.h>
+// UNSUPPORTED: c++03
 
+#include "assert_macros.h"
+#include "concat_macros.h"
 #include "../src/cxa_exception.h"
 
-#include "test_macros.h"
+int main(int, char**) {
+  void* globals = __cxxabiv1::__cxa_get_globals();
+  TEST_REQUIRE(globals != nullptr, TEST_WRITE_CONCATENATED("Got null result from __cxa_get_globals"));
 
-typedef __cxxabiv1::__cxa_eh_globals globals_t;
+  void* fast_globals = __cxxabiv1::__cxa_get_globals_fast();
+  TEST_REQUIRE(globals == fast_globals, TEST_WRITE_CONCATENATED("__cxa_get_globals returned ", globals,
+                                                                " but __cxa_get_globals_fast returned ", fast_globals));
 
-void* thread_code(void* parm) {
-  size_t* result = (size_t*)parm;
-  globals_t *glob1, *glob2;
-
-  glob1 = __cxxabiv1::__cxa_get_globals();
-  if (NULL == glob1)
-    std::printf("Got null result from __cxa_get_globals\n");
-
-  glob2 = __cxxabiv1::__cxa_get_globals_fast();
-  if (glob1 != glob2)
-    std::printf("Got different globals!\n");
-
-  *result = (size_t)glob1;
-#ifndef TEST_HAS_NO_THREADS
-  sleep(1);
-#endif
-  return parm;
-}
-
-#ifndef TEST_HAS_NO_THREADS
-#  define NUMTHREADS 10
-size_t thread_globals[NUMTHREADS] = {0};
-std::__libcpp_thread_t threads[NUMTHREADS];
-#endif
-
-int main() {
-#ifndef TEST_HAS_NO_THREADS
-  //  Make the threads, let them run, and wait for them to finish
-  for (int i = 0; i < NUMTHREADS; ++i)
-    std::__libcpp_thread_create(threads + i, thread_code, (void*)(thread_globals + i));
-  for (int i = 0; i < NUMTHREADS; ++i)
-    std::__libcpp_thread_join(&threads[i]);
-
-  int retVal = 0;
-  for (int i = 0; i < NUMTHREADS; ++i) {
-    if (0 == thread_globals[i]) {
-      std::printf("Thread #%d had a zero global\n", i);
-      retVal = 1;
-    }
-  }
-
-  std::sort(thread_globals, thread_globals + NUMTHREADS);
-  for (int i = 1; i < NUMTHREADS; ++i) {
-    if (thread_globals[i - 1] == thread_globals[i]) {
-      std::printf("Duplicate thread globals (%d and %d)\n", i - 1, i);
-      retVal = 2;
-    }
-  }
-  return retVal;
-#else  // TEST_HAS_NO_THREADS
-  size_t thread_globals;
-  thread_code(&thread_globals);
-  // Check that __cxa_get_globals() is not NULL.
-  return (thread_globals == 0) ? 1 : 0;
-#endif // !TEST_HAS_NO_THREADS
+  return 0;
 }
diff --git a/libcxxabi/test/test_exception_storage.threads.pass.cpp b/libcxxabi/test/test_exception_storage.threads.pass.cpp
new file mode 100644
--- /dev/null
+++ b/libcxxabi/test/test_exception_storage.threads.pass.cpp
@@ -0,0 +1,77 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// UNSUPPORTED: c++03, no-threads
+
+#include "test_macros.h"
+
+#include <algorithm>
+#include <condition_variable>
+#include <functional>
+#include <mutex>
+#include <thread>
+#include <utility>
+#include <vector>
+
+#include "assert_macros.h"
+#include "concat_macros.h"
+#include "../src/cxa_exception.h"
+
+static int threads_remaining;
+static std::mutex threads_remaining_lock;
+static std::condition_variable threads_remaining_cv;
+
+static void thread_code(void*& globals) {
+  std::thread::id thread_id = std::this_thread::get_id();
+  (void)thread_id;
+
+  globals = __cxxabiv1::__cxa_get_globals();
+  TEST_REQUIRE(globals != nullptr,
+               TEST_WRITE_CONCATENATED("Got null result from __cxa_get_globals on thread ", thread_id));
+
+  void* fast_globals = __cxxabiv1::__cxa_get_globals_fast();
+  TEST_REQUIRE(globals == fast_globals,
+               TEST_WRITE_CONCATENATED("__cxa_get_globals returned ", globals, " but __cxa_get_globals_fast returned ",
+                                       fast_globals, " on thread ", thread_id));
+
+  // Ensure that all threads are running at the same time, since we check for
+  // duplicate globals below. We do this manually instead of using std::barrier
+  // or std::latch to avoid requiring C++20.
+  std::unique_lock<std::mutex> lock(threads_remaining_lock);
+  --threads_remaining;
+  if (threads_remaining == 0) {
+    lock.unlock();
+    threads_remaining_cv.notify_all();
+  } else {
+    threads_remaining_cv.wait(lock, []() { return threads_remaining == 0; });
+  }
+}
+
+int main(int, char**) {
+  int num_threads = std::thread::hardware_concurrency();
+  if (num_threads == 0)
+    num_threads = 4; // arbitrary fallback value
+
+  std::vector<void*> thread_globals(num_threads);
+  std::vector<std::thread> threads;
+  threads_remaining = num_threads;
+
+  // Make the threads, let them run, and wait for them to finish
+  for (int i = 0; i < num_threads; ++i)
+    threads.emplace_back(thread_code, std::ref(thread_globals[i]));
+  for (std::thread& thread : threads)
+    thread.join();
+
+  std::sort(thread_globals.begin(), thread_globals.end());
+  for (int i = 1; i < num_threads; ++i) {
+    TEST_REQUIRE(thread_globals[i - 1] != thread_globals[i],
+                 TEST_WRITE_CONCATENATED("Duplicate thread globals ", thread_globals[i]));
+  }
+
+  return 0;
+}