hw/qdev: Warn when using pre-qdev/QOM devices

[RFC PATCH 26/35] hw/openrisc/cputimer: Emit warning when old code is used

Posted by Philippe Mathieu-Daudé 5 years, 8 months ago

This code hasn't been QOM'ified yet. Warn the user.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
 hw/openrisc/cputimer.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/hw/openrisc/cputimer.c b/hw/openrisc/cputimer.c
index 93268815d8..60f2c9667f 100644
--- a/hw/openrisc/cputimer.c
+++ b/hw/openrisc/cputimer.c
@@ -22,6 +22,7 @@
 #include "cpu.h"
 #include "migration/vmstate.h"
 #include "qemu/timer.h"
+#include "hw/qdev-deprecated.h"
 
 #define TIMER_PERIOD 50 /* 50 ns period for 20 MHz timer */
 
@@ -135,6 +136,8 @@ static const VMStateDescription vmstate_or1k_timer = {
 
 void cpu_openrisc_clock_init(OpenRISCCPU *cpu)
 {
+    qdev_warn_deprecated_function_used();
+
     cpu->env.timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, &openrisc_timer_cb, cpu);
     cpu->env.ttmr = 0x00000000;
 
-- 
2.21.3

Re: [RFC PATCH 26/35] hw/openrisc/cputimer: Emit warning when old code is used

Posted by Paolo Bonzini 5 years, 8 months ago

On 08/06/20 18:00, Philippe Mathieu-Daudé wrote:
> This code hasn't been QOM'ified yet. Warn the user.
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
>  hw/openrisc/cputimer.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/hw/openrisc/cputimer.c b/hw/openrisc/cputimer.c
> index 93268815d8..60f2c9667f 100644
> --- a/hw/openrisc/cputimer.c
> +++ b/hw/openrisc/cputimer.c
> @@ -22,6 +22,7 @@
>  #include "cpu.h"
>  #include "migration/vmstate.h"
>  #include "qemu/timer.h"
> +#include "hw/qdev-deprecated.h"
>  
>  #define TIMER_PERIOD 50 /* 50 ns period for 20 MHz timer */
>  
> @@ -135,6 +136,8 @@ static const VMStateDescription vmstate_or1k_timer = {
>  
>  void cpu_openrisc_clock_init(OpenRISCCPU *cpu)
>  {
> +    qdev_warn_deprecated_function_used();
> +
>      cpu->env.timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, &openrisc_timer_cb, cpu);
>      cpu->env.ttmr = 0x00000000;
>  
> 


I was about to give this a pass, but if we did so it should be the CPU
itself that calls cpu_openrisc_clock_init (not openrisc_sim_init).

Paolo