Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (1.0)
Subject: Re: [PATCH v3 2/7] x86/jump_label: Use text_poke_early() during early_init
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <20181105140925.GD22467@hirez.programming.kicks-ass.net>
Date:   Mon, 5 Nov 2018 09:22:32 -0800
Cc:     Nadav Amit <namit@vmware.com>, Ingo Molnar <mingo@redhat.com>,
        linux-kernel@vger.kernel.org, x86@kernel.org,
        "H. Peter Anvin" <hpa@zytor.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Borislav Petkov <bp@alien8.de>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Andy Lutomirski <luto@kernel.org>,
        Kees Cook <keescook@chromium.org>,
        Dave Hansen <dave.hansen@intel.com>,
        Masami Hiramatsu <mhiramat@kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <BB6FC627-C152-4C4C-A6D4-FA7F42F2610D@amacapital.net>
References: <20181102232946.98461-1-namit@vmware.com> <20181102232946.98461-3-namit@vmware.com> <20181105140925.GD22467@hirez.programming.kicks-ass.net>
To:     Peter Zijlstra <peterz@infradead.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Nov 5, 2018, at 6:09 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>=20
>> On Fri, Nov 02, 2018 at 04:29:41PM -0700, Nadav Amit wrote:
>> diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c
>> index aac0c1f7e354..367c1d0c20a3 100644
>> --- a/arch/x86/kernel/jump_label.c
>> +++ b/arch/x86/kernel/jump_label.c
>> @@ -52,7 +52,13 @@ static void __ref __jump_label_transform(struct jump_e=
ntry *entry,
>>    jmp.offset =3D jump_entry_target(entry) -
>>             (jump_entry_code(entry) + JUMP_LABEL_NOP_SIZE);
>>=20
>> -    if (early_boot_irqs_disabled)
>> +    /*
>> +     * As long as we are in early boot, we can use text_poke_early(), wh=
ich
>> +     * is more efficient: the memory was still not marked as read-only (=
it
>> +     * is only marked after poking_init()). This also prevents us from u=
sing
>> +     * text_poke() before poking_init() is called.
>> +     */
>> +    if (!early_boot_done)
>>        poker =3D text_poke_early;
>>=20
>>    if (type =3D=3D JUMP_LABEL_JMP) {
>=20
> It took me a while to untangle init/maze^H^Hin.c... but I think this
> is all we need:
>=20
> diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c
> index aac0c1f7e354..ed5fe274a7d8 100644
> --- a/arch/x86/kernel/jump_label.c
> +++ b/arch/x86/kernel/jump_label.c
> @@ -52,7 +52,12 @@ static void __ref __jump_label_transform(struct jump_en=
try *entry,
>    jmp.offset =3D jump_entry_target(entry) -
>             (jump_entry_code(entry) + JUMP_LABEL_NOP_SIZE);
>=20
> -    if (early_boot_irqs_disabled)
> +    /*
> +     * As long as we're UP and not yet marked RO, we can use
> +     * text_poke_early; SYSTEM_BOOTING guarantees both, as we switch to
> +     * SYSTEM_SCHEDULING before going either.
> +     */
> +    if (system_state =3D=3D SYSTEM_BOOTING)
>        poker =3D text_poke_early;
>=20
>    if (type =3D=3D JUMP_LABEL_JMP) {

Can we move this logic into text_poke() and get rid of text_poke_early()?

FWIW, alternative patching was, at some point, a significant fraction of tot=
al boot time in some cases. This was probably mostly due to unnecessary sync=
_core() calls.  Although I think this was reported on a VM, and sync_core() u=
sed to be *extremely* expensive on a VM, but that=E2=80=99s fixed now, and i=
t even got backported, I think.

(Hmm. Maybe we can also make jump label patching work in early boot, too!)=