Remove workaround for missing std::atomic_thread_fence

Fix compilation error with gcc Linux. std::memory_order is an enum
class, not an enum, which removes implicit conversions in C++20.
Compilers that support C++20 should contain the fix for the libstdc++
bug (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51038), so removing
the workaround in atomicops_internals_portable.h should be fine.

Bug: 339934812
Change-Id: Ie4f6f4de26b084ff33330db7d87c7881db223a62
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/mini_chromium/+/5514410
Reviewed-by: danakj <danakj@chromium.org>
Reviewed-by: Mark Mentovai <mark@chromium.org>

base/atomicops_internals_portable.h

@@ -51,13 +51,7 @@ static_assert(sizeof(*(AtomicLocation32) nullptr) == sizeof(Atomic32),
               "incompatible 32-bit atomic layout");
 
 inline void MemoryBarrier() {
-#if defined(__GLIBCXX__)
-  // Work around libstdc++ bug 51038 where atomic_thread_fence was declared but
-  // not defined, leading to the linker complaining about undefined references.
-  __atomic_thread_fence(std::memory_order_seq_cst);
-#else
   std::atomic_thread_fence(std::memory_order_seq_cst);
-#endif
 }
 
 inline Atomic32 NoBarrier_CompareAndSwap(volatile Atomic32* ptr,