Message-passing is an increasingly popular design style in MPSoCs. This usually results in systems that perform better compared to external shared-memory designs performance and power-wise, this because of much decreased data transfers with external memory. This scheme relies on explicit communications between processing tasks that participate in the application, which is usually described as a task graph. Contrarily to shared-memory multiprocessors, tasks usually get assigned to processors at design-time. In order to cope with transient performance losses originating from various phenomena such as increased processing workload or peak traffic in the communication subsystem, various adaptation mechanisms based on task migration have been proposed in the literature. These often imply migrating application code from processor to processor, which incurs penalty in latency and power consumption. This paper proposes a local shared-memory strategy in which processors execute code hosted in a remote processor. Besides the benefits in term of migration latency, this approach enables local symmetric multi-threading (SMT) which broadens the scope of supported applications. Results are given in term of latency, performance and scalability both using benchmarks and realistic applications.