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=us-ascii
Mime-Version: 1.0 (1.0)
Subject: Re: [POC][RFC][PATCH 1/2] jump_function: Addition of new feature "jump_function"
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <20181011125258.ihsdmonepaoc5xtb@treble>
Date:   Thu, 11 Oct 2018 09:20:40 -0700
Cc:     Steven Rostedt <rostedt@goodmis.org>,
        Peter Zijlstra <peterz@infradead.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Ingo Molnar <mingo@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Masami Hiramatsu <mhiramat@kernel.org>,
        Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
        Matthew Helsley <mhelsley@vmware.com>,
        "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
        David Woodhouse <dwmw2@infradead.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Jason Baron <jbaron@akamai.com>, Jiri Kosina <jkosina@suse.cz>,
        Ard Biesheuvel <ard.biesheuvel@linaro.org>,
        Andrew Lutomirski <luto@kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <B7EBEF5D-443E-4764-8589-DE470F93422D@amacapital.net>
References: <20181008155757.GC5663@hirez.programming.kicks-ass.net> <F36C6419-7D92-4357-BDB1-D8B90DE09889@amacapital.net> <20181009021710.qwt5hpntyeps44h3@treble> <20181008235750.59da83ae@gandalf.local.home> <20181010175237.e7m3sldcu2maoqcq@treble> <CALCETrUOaJ1dAsuX9bQO_+B4G6y5-FAb0=ZYA8QbtAEBZ6MxKg@mail.gmail.com> <20181010181605.arsyjxwdztztrjih@treble> <20181010181741.lc4l7kjvcmoxrw5b@treble> <CALCETrUZMmTDOh5-Mzyj1rk7Q82DgdQLinOfbzF4Sb7OC2aYMQ@mail.gmail.com> <20181011030738.wwy6grnss67txd2l@treble> <20181011125258.ihsdmonepaoc5xtb@treble>
To:     Josh Poimboeuf <jpoimboe@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Oct 11, 2018, at 5:52 AM, Josh Poimboeuf <jpoimboe@redhat.com> wrote:
>=20
>> On Wed, Oct 10, 2018 at 10:07:38PM -0500, Josh Poimboeuf wrote:
>>> On Wed, Oct 10, 2018 at 02:13:22PM -0700, Andy Lutomirski wrote:
>>>> On Wed, Oct 10, 2018 at 11:17 AM Josh Poimboeuf <jpoimboe@redhat.com> w=
rote:
>>>>=20
>>>>> On Wed, Oct 10, 2018 at 01:16:05PM -0500, Josh Poimboeuf wrote:
>>>>> On Wed, Oct 10, 2018 at 11:03:43AM -0700, Andy Lutomirski wrote:
>>>>>>> +#define DECLARE_STATIC_CALL(tramp, func)                           =
    \
>>>>>>> +       extern typeof(func) tramp;                                  =
    \
>>>>>>> +       static void __used __section(.discard.static_call_tramps)   =
    \
>>>>>>> +               *__static_call_tramp_##tramp =3D tramp
>>>>>>> +
>>>>>>=20
>>>>>> Confused.  What's the __static_call_tramp_##tramp variable for?  And
>>>>>> why is a DECLARE_ macro defining a variable?
>>>>>=20
>>>>> This is the magic needed for objtool to find all the call sites.
>>>>>=20
>>>>> The variable itself isn't needed, but the .discard.static_call_tramps
>>>>> entry is.  Objtool reads that section to find out which function call
>>>>> sites are targeted to a static call trampoline.
>>>>=20
>>>> To clarify: objtool reads that section to find out which functions are
>>>> really static call trampolines.  Then it annotates all the instructions=

>>>> which call/jmp to those trampolines.  Those annotations are then read b=
y
>>>> the kernel.
>>>>=20
>>>=20
>>> Ah, right, and objtool runs on a per-object basis so it has no other
>>> way to know what symbols are actually static calls.
>>>=20
>>> There's another way to skin this cat, though:
>>>=20
>>> extern typeof(func) __static_call_trampoline_##tramp;
>>> #define tramp __static_call_trampoline_##tramp
>>>=20
>>> And objtool could recognize it by name.  But, of course, you can't put
>>> a #define in a macro.  But maybe there's a way to hack it up with a
>>> static inline?
>=20
> I went with something similar in the latest version.  It's less
> surprising in a couple of ways:
>=20
> - DECLARE_STATIC_CALL doesn't have the magic objtool definition.
>=20
> - Call sites use the static_call() wrapper, which makes static calls
>  clearly visible.

Seems reasonable. Also, for a real patch, it should be straightforward to ha=
ve a fallback implementation in include/linux/static_call.h that just derefe=
rences the pointer.