Re: [Qemu-devel] [PATCH] Document how x86 gdb_num_core_regs is computed.

[Paolo Bonzini <pbonzini@redhat.com>]

On 03/11/2016 22:48, Doug Evans wrote:
> Hi.
> 
> It helps when reading the code to see how the number is arrived at.
> 
> Signed-off-by: Doug Evans <dje@google.com>
> ---
>  target-i386/cpu.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/target-i386/cpu.c b/target-i386/cpu.c
> index 14c5186..01f1ab0 100644
> --- a/target-i386/cpu.c
> +++ b/target-i386/cpu.c
> @@ -3721,6 +3721,8 @@ static void x86_cpu_common_class_init(ObjectClass
> *oc, void *data)
>      cc->write_elf32_qemunote = x86_cpu_write_elf32_qemunote;
>      cc->vmsd = &vmstate_x86_cpu;
>  #endif
> +    /* CPU_NB_REGS * 2 = general regs + xmm regs
> +       25 = eip, eflags, 6 seg regs, st[0-7], fctrl,...,fop, mxcsr */
>      cc->gdb_num_core_regs = CPU_NB_REGS * 2 + 25;
>  #ifndef CONFIG_USER_ONLY
>      cc->debug_excp_handler = breakpoint_handler;
> -- 
> 
> 

Queued for 2.8, thanks.

Paolo