target/hppa: Fix OS reboot issues

When the OS triggers a reboot, the reset helper function sends a
qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET) together with an
EXCP_HLT exception to halt the CPUs.

So, at reboot when initializing the CPUs again, make sure to set all
instruction pointers to the firmware entry point, disable any interrupts,
disable data and instruction translations, enable PSW_Q bit  and tell qemu
to unhalt (halted=0) the CPUs again.

This fixes the various reboot issues which were seen when rebooting a
Linux VM, including the case where even the monarch CPU has been virtually
halted from the OS (e.g. via "chcpu -d 0" inside the Linux VM).

Signed-off-by: Helge Deller <deller@gmx.de>
(cherry picked from commit 50ba97e)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>

hw/hppa/machine.c

@@ -417,10 +417,16 @@ static void hppa_machine_reset(MachineState *ms, ShutdownCause reason)
 
     /* Start all CPUs at the firmware entry point.
      *  Monarch CPU will initialize firmware, secondary CPUs
-     *  will enter a small idle look and wait for rendevouz. */
+     *  will enter a small idle loop and wait for rendevouz. */
     for (i = 0; i < smp_cpus; i++) {
-        cpu_set_pc(CPU(cpu[i]), firmware_entry);
+        CPUState *cs = CPU(cpu[i]);
+
+        cpu_set_pc(cs, firmware_entry);
+        cpu[i]->env.psw = PSW_Q;
         cpu[i]->env.gr[5] = CPU_HPA + i * 0x1000;
+
+        cs->exception_index = -1;
+        cs->halted = 0;
     }
 
     /* already initialized by machine_hppa_init()? */