Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20211110010906.1923210-1-eric.dumazet@gmail.com>
 <20211110010906.1923210-3-eric.dumazet@gmail.com> <YYuEXQ7Ur9f88pCw@hirez.programming.kicks-ass.net>
 <CANn89iLSvNzgDc4s5rv65eELx2fGVzb6Nj=8D7T3Up8Wgsm_SA@mail.gmail.com>
In-Reply-To: <CANn89iLSvNzgDc4s5rv65eELx2fGVzb6Nj=8D7T3Up8Wgsm_SA@mail.gmail.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Wed, 10 Nov 2021 07:41:22 -0800
Message-ID: <CANn89iLZr6Dg0ON=RcZsH=Yd=+qdKNEuAgzOj6hu=f-y=7ngfw@mail.gmail.com>
Subject: Re: [PATCH 2/2] jump_label: refine placement of static_keys
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Eric Dumazet <eric.dumazet@gmail.com>,
        linux-kernel <linux-kernel@vger.kernel.org>, jpoimboe@redhat.com,
        jbaron@akamai.com, rostedt@goodmis.org, ardb@kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Nov 10, 2021 at 7:30 AM Eric Dumazet <edumazet@google.com> wrote:
>

>
> Sure, I can make sure to include all cases. I was hoping
> DEFINE_STATIC_KEY_TRUE()/DEFINE_STATIC_KEY_FALSE()
> would catch the majority of uses.
>
> We also can add a new DEFINE_STATIC_KEY() macro to ease these cases.

Although use of 'struct static_key' is marked deprecated.

I guess that we can leave legacy uses be converted by maintainers if they care.
(They do not have to, __static_key section is only a hint, not a requirement)

For instance these ones had no __read_mostly and nobody cared.

arch/x86/kernel/paravirt.c:124:struct static_key paravirt_steal_enabled;
arch/x86/kernel/paravirt.c:125:struct static_key paravirt_steal_rq_enabled;


>
>
>
> >
> > Would something like:
> >
> >         typedef struct static_key __static_key static_key_t;
> >
> > work? I forever seem to forget the exact things you can make a typedef
> > do :/
>
> I tried, but a typedef was not working. A macro would work.