Re: [PATCH v3 17/51] gdbstub: Report actual watchpoint trigger address

On 7/10/26 22:53, Richard Henderson wrote:
> While gdb doesn't care either way, it seems more correct to
> report the address of the watchpoint hit, rather than the
> beginning of the watchpoint region.
> 
> Adjust all virtualization hosts which construct a dummy
> CPUWatchpoint structure.
> 
> Signed-off-by: Richard Henderson <[email protected]>
> ---
>   gdbstub/system.c       | 4 ++--
>   target/arm/hvf/hvf.c   | 1 +
>   target/arm/kvm.c       | 1 +
>   target/i386/kvm/kvm.c  | 4 ++--
>   target/ppc/kvm.c       | 4 ++--
>   target/s390x/kvm/kvm.c | 2 +-
>   6 files changed, 9 insertions(+), 7 deletions(-)

GDB seems to be copying stopped_data_address around quite a bit, but at
the end of the day it seems like the only thing that matters is whether
it's present or not.


Reviewed-by: Ilya Leoskevich <[email protected]>