Re: [Qemu-devel] [PATCH 4/6] Handle CPU interrupts by inline checking of a flag

[Lluís Vilanova]

Peter Maydell writes:
[...]
> diff --git a/include/exec/gen-icount.h b/include/exec/gen-icount.h
> index c858a73..384153b 100644
> --- a/include/exec/gen-icount.h
> +++ b/include/exec/gen-icount.h
> @@ -7,10 +7,19 @@
 
>  static TCGArg *icount_arg;
>  static int icount_label;
> +static int exitreq_label;
 
>  static inline void gen_icount_start(void)
>  {
>      TCGv_i32 count;
> +    TCGv_i32 flag;
> +
> +    exitreq_label = gen_new_label();
> +    flag = tcg_temp_local_new_i32();
> +    tcg_gen_ld_i32(flag, cpu_env,
> +                   offsetof(CPUState, tcg_exit_req) - ENV_OFFSET);
> +    tcg_gen_brcondi_i32(TCG_COND_NE, flag, 0, exitreq_label);
> +    tcg_temp_free_i32(flag);

It feels weird to me that this code is in 'gen_icount_start' as opposed that
some other separate function like 'gen_tb_exit_check'.

 
>      if (!use_icount)
>          return;
> @@ -29,6 +38,9 @@ static inline void gen_icount_start(void)
 
>  static void gen_icount_end(TranslationBlock *tb, int num_insns)
>  {
> +    gen_set_label(exitreq_label);
> +    tcg_gen_exit_tb((tcg_target_long)tb + TB_EXIT_REQUESTED);
> +

This too feels weird to me.


Lluis

-- 
 "And it's much the same thing with knowledge, for whenever you learn
 something new, the whole world becomes that much richer."
 -- The Princess of Pure Reason, as told by Norton Juster in The Phantom
 Tollbooth