On 26/06/2015 16:47, fred.kon...@greensocs.com wrote:
> + CPU_FOREACH(cpu) {
> + if (qemu_cpu_is_self(cpu)) {
> + /* async_run_on_cpu handle this case but this just avoid a malloc
> + * here.
> + */
> + tlb_flush(cpu, flush_global);
> + } else {
> + params = g_malloc(sizeof(struct TLBFlushParams));
> + params->cpu = cpu;
> + params->flush_global = flush_global;
> + async_run_on_cpu(cpu, tlb_flush_async_work, params);
Shouldn't this be synchronous (which you cannot do straightforwardly
because of deadlocks---hence the need to hook cpu_has_work as discussed
earlier)?
Paolo
> + }
> + }
