Re: [PATCH-for-11.1 v6 4/6] monitor: Remove last target_long use in get_monitor_def()

[Pierrick Bouvier <pierrick.bouvier@linaro.org>]

On 3/20/26 2:10 AM, Philippe Mathieu-Daudé wrote:
> Truncate manually, removing the need of target-specific
> code in get_monitor_def().
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
>   monitor/hmp-target.c | 3 +--
>   1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/monitor/hmp-target.c b/monitor/hmp-target.c
> index 155caedd784..5e9f1b0db5b 100644
> --- a/monitor/hmp-target.c
> +++ b/monitor/hmp-target.c
> @@ -34,7 +34,6 @@
>   #include "qapi/qapi-commands-control.h"
>   #include "qapi/qapi-commands-misc.h"
>   #include "qapi/qapi-commands-machine.h"
> -#include "exec/target_long.h"
>   
>   #if defined(TARGET_S390X)
>   #include "hw/s390x/storage-keys.h"
> @@ -93,7 +92,7 @@ int get_monitor_def(Monitor *mon, int64_t *pval, const char *name)
>   
>       ret = target_get_monitor_def(cs, name, &tmp);
>       if (!ret) {
> -        *pval = (target_long) tmp;
> +        *pval = target_long_bits() == 32 ? (int32_t)tmp : tmp;
>       }
>   
>       return ret;

Looking a second time at it, I don't think you need it. Same in previous 
commit.

Since pval is a pointer to a (signed) int64, sign will be extended from 
32 bit value automatically. So either, it's a positive value, and will 
be treated as it is. Or it's a negative value, and all upper bits will 
be set to 1.

So the casting in int32_t has no effect, since it will be extended again 
automatically.

Regards,
Pierrick