On 10/03/14 12:23, Stefano Stabellini wrote:
On Fri, 3 Oct 2014, Don Slutz wrote:
On 10/03/14 05:52, Stefano Stabellini wrote:
On Thu, 2 Oct 2014, Don Slutz wrote:
This adds synchronisation of the 6 vcpu registers (only 32bits of
them) that vmport.c needs between Xen and QEMU.
...
}
-static void handle_ioreq(ioreq_t *req)
+static void regs_to_cpu(XenIOState *state, vmware_ioreq_t *vmport_req)
+{
+ X86CPU *cpu;
+ CPUX86State *env;
+
+ if (!state->cpu_by_ioreq_id[0]) {
+ CPUState *cpu_state;
+
+ CPU_FOREACH(cpu_state) {
+ state->cpu_by_ioreq_id[cpu_state->cpu_index] = cpu_state;
+ }
+ }
This is just the initialization, isn't it?
It would be best to move it to an initialization function then.
A new initialization function would need to be added. A new call to it would
need
to be added (not sure where the best place is). Since the overhead here is
small I went with the less intrusive change.
I'd prefer the initialization function if it is possible.
Ok, how does:
@@ -1023,6 +1028,11 @@ static void xen_main_loop_prepare(XenIOState *state)
state);
if (evtchn_fd != -1) {
+ CPUState *cpu_state;
+
+ CPU_FOREACH(cpu_state) {
+ state->cpu_by_ioreq_id[cpu_state->cpu_index] = cpu_state;
+ }
qemu_set_fd_handler(evtchn_fd, cpu_handle_ioreq, NULL, state);
}
}
Look?
-Don Slutz
