In decoder.hpp/cpp we have defined a DecoderLocker class that extends MutexLockerEx and handles the lock to use and whether to use safepoint checks:

DecoderLocker::DecoderLocker() :
  MutexLockerEx(DecoderLocker::is_first_error_thread() ?
                NULL : Decoder::shared_decoder_lock(),
                Mutex::_no_safepoint_check_flag) {
  _decoder = is_first_error_thread() ?
    Decoder::get_error_handler_instance() : Decoder::get_shared_instance();
  assert(_decoder != NULL, "null decoder");
}

but the actual methods still use MutexLockerEx e.g.

bool Decoder::decode(address addr, char* buf, int buflen, int* offset, const char* modulepath, bool demangle) {
  bool error_handling_thread = os::current_thread_id() == VMError::first_error_tid;
  MutexLockerEx locker(error_handling_thread ? NULL : shared_decoder_lock(),
                       Mutex::_no_safepoint_check_flag);
  AbstractDecoder* decoder = error_handling_thread ?
    get_error_handler_instance(): get_shared_instance();
  assert(decoder != NULL, "null decoder");

  return decoder->decode(addr, buf, buflen, offset, modulepath, demangle);
}

We should either use DecoderLocker as intended, or else delete it.