cpus-common: lock-free fast path for cpu_exec_start/end

Set cpu->running without taking the cpu_list lock, only requiring it if
there is a concurrent exclusive section.  This requires adding a new
field to CPUState, which records whether a running CPU is being counted
in pending_cpus.

When an exclusive section is started concurrently with cpu_exec_start,
cpu_exec_start can use the new field to determine if it has to wait for
the end of the exclusive section.  Likewise, cpu_exec_end can use it to
see if start_exclusive is waiting for that CPU.

This a separate patch for easier bisection of issues.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

1 files changed, 3 insertions, 2 deletions

diff --git a/include/qom/cpu.h b/include/qom/cpu.h
index 5dfe74a0e7..22b54d6d93 100644
--- a/include/qom/cpu.h
+++ b/include/qom/cpu.h
@@ -242,7 +242,8 @@ struct qemu_work_item;
  * @nr_threads: Number of threads within this CPU.
  * @numa_node: NUMA node this CPU is belonging to.
  * @host_tid: Host thread ID.
- * @running: #true if CPU is currently running;
+ * @running: #true if CPU is currently running (lockless).
+ * @has_waiter: #true if a CPU is currently waiting for the cpu_exec_end;
  * valid under cpu_list_lock.
  * @created: Indicates whether the CPU thread has been successfully created.
  * @interrupt_request: Indicates a pending interrupt request.
@@ -296,7 +297,7 @@ struct CPUState {
 #endif
     int thread_id;
     uint32_t host_tid;
-    bool running;
+    bool running, has_waiter;
     struct QemuCond *halt_cond;
     bool thread_kicked;
     bool created;