Re: [PATCH bpf-next v7 1/2] bpf: Introduce jit_required flag and refactor kfunc path

[KaFai Wan <kafai.wan@linux.dev>]

On Mon, 2026-07-06 at 11:59 +0800, Leon Hwang wrote:
> On 6/7/26 08:56, Tiezhu Yang wrote:
> > On 2026/7/4 上午9:57, KaFai Wan wrote:
> > > On Sat, 2026-07-04 at 00:14 +0800, Tiezhu Yang wrote:
> > > > On 7/3/26 21:55, KaFai Wan wrote:
> > > > > On Thu, 2026-07-02 at 22:36 +0800, Tiezhu Yang wrote:
> > 
> > ...
> > 
> > > > 
> > > > IIUC, there are two places to modify so far:
> > > > (1) struct bpf_prog
> > > 
> > > yes
> > > 
> > > > (2) bpf_prog_has_kfunc_call()
> > > 
> > > I think we can remove this func.
> > > 
> > > actually, I feed your patch to dsv4-pro and gpt-5.4 yesterday, both
> > > point out the we should do this
> > > and change the log message.
> > > 
> > > diff --git a/kernel/bpf/fixups.c b/kernel/bpf/fixups.c
> > > index 12a8a4eb757f..7fb92b5fa415 100644
> > > --- a/kernel/bpf/fixups.c
> > > +++ b/kernel/bpf/fixups.c
> > > @@ -1378,7 +1378,6 @@ int bpf_fixup_call_args(struct bpf_verifier_env
> > > *env)
> > >   #ifndef CONFIG_BPF_JIT_ALWAYS_ON
> > >       struct bpf_prog *prog = env->prog;
> > >       struct bpf_insn *insn = prog->insnsi;
> > > -    bool has_kfunc_call = bpf_prog_has_kfunc_call(prog);
> > >       int depth;
> > >   #endif
> > >       int i, err = 0;
> > > @@ -1404,7 +1403,7 @@ int bpf_fixup_call_args(struct bpf_verifier_env
> > > *env)
> > >               return err;
> > >       }
> > >   #ifndef CONFIG_BPF_JIT_ALWAYS_ON
> > > -    if (has_kfunc_call) {
> > > +    if (prog->jit_required) {
> 
> 
> Agreed with this change.
> 
> Then, think it harder, pls.
> 
> Can we clean some code of __bpf_prog_select_runtime() btw?
> 
> diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
> index 649cce41e13f..bad55e119cb5 100644
> --- a/kernel/bpf/core.c
> +++ b/kernel/bpf/core.c
> @@ -120,6 +120,9 @@ struct bpf_prog *bpf_prog_alloc_no_stats(unsigned
> int size, gfp_t gfp_extra_flag
>  		return NULL;
>  	}
> 
> +#ifdef CONFIG_BPF_JIT_ALWAYS_ON
> +	fp->jit_required = true;
> +#endif

Don't need #ifdef, fp->jit_required = IS_ENABLED(CONFIG_BPF_JIT_ALWAYS_ON) is fine. :)

>  	fp->pages = size / PAGE_SIZE;
>  	fp->aux = aux;
>  	fp->aux->main_prog_aux = aux;
> @@ -2614,15 +2617,11 @@ struct bpf_prog
> *__bpf_prog_select_runtime(struct bpf_verifier_env *env, struct
>  	/* In case of BPF to BPF calls, verifier did all the prep
>  	 * work with regards to JITing, etc.
>  	 */
> -	bool jit_needed = false;
> +	bool jit_needed = fp->jit_required;
> 
>  	if (fp->bpf_func)
>  		goto finalize;
> 
> -	if (IS_ENABLED(CONFIG_BPF_JIT_ALWAYS_ON) ||
> -	    bpf_prog_has_kfunc_call(fp))
> -		jit_needed = true;
> -
>  	if (!bpf_prog_select_interpreter(fp))
>  		jit_needed = true;
> 
> WDYT?
> 
> Thanks,
> Leon
> 
> > >           verbose(env, "calling kernel functions are not allowed in
> > > non-JITed programs\n");
> > >           return -EINVAL;
> > > [...]

-- 
Thanks,
KaFai