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: [RFC PATCH 1/5] x86: introduce preemption disable prefix
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <20181019083325.GC3121@hirez.programming.kicks-ass.net>
Date:   Fri, 19 Oct 2018 07:29:45 -0700
Cc:     Nadav Amit <namit@vmware.com>, Ingo Molnar <mingo@redhat.com>,
        Andrew Lutomirski <luto@kernel.org>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        LKML <linux-kernel@vger.kernel.org>, X86 ML <x86@kernel.org>,
        Borislav Petkov <bp@alien8.de>,
        "Woodhouse, David" <dwmw@amazon.co.uk>
Content-Transfer-Encoding: quoted-printable
Message-Id: <F1F3B914-2630-41ED-B205-BBA50F94B64F@amacapital.net>
References: <20181018005420.82993-1-namit@vmware.com> <20181018005420.82993-2-namit@vmware.com> <D247AB7C-2890-49CE-B102-163E5B368FC2@amacapital.net> <07255D2B-0243-4254-B62A-37050C44207E@vmware.com> <CALCETrXEG8OrMpB3VBS9RF41O_MwtPmXBazkdSGE0ohZmXDcBw@mail.gmail.com> <925F22EA-F8CB-4194-B96B-378409ED7918@vmware.com> <2626124E-7344-42F3-AD07-0BB34D62A9EE@amacapital.net> <6F1FD9DA-5E86-42A2-8EAF-05F5D70FE2EF@vmware.com> <20181019083325.GC3121@hirez.programming.kicks-ass.net>
To:     Peter Zijlstra <peterz@infradead.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Oct 19, 2018, at 1:33 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>=20
>> On Fri, Oct 19, 2018 at 01:08:23AM +0000, Nadav Amit wrote:
>> Consider for example do_int3(), and see my inlined comments:
>>=20
>> dotraplinkage void notrace do_int3(struct pt_regs *regs, long error_code)=

>> {
>>    ...
>>    ist_enter(regs);        // =3D> preempt_disable()
>>    cond_local_irq_enable(regs);    // =3D> assume it enables IRQs
>>=20
>>    ...
>>    // resched irq can be delivered here. It will not caused rescheduling
>>    // since preemption is disabled
>>=20
>>    cond_local_irq_disable(regs);    // =3D> assume it disables IRQs
>>    ist_exit(regs);            // =3D> preempt_enable_no_resched()
>> }
>>=20
>> At this point resched will not happen for unbounded length of time (unles=
s
>> there is another point when exiting the trap handler that checks if
>> preemption should take place).
>>=20
>> Another example is __BPF_PROG_RUN_ARRAY(), which also uses
>> preempt_enable_no_resched().
>>=20
>> Am I missing something?
>=20
> Would not the interrupt return then check for TIF_NEED_RESCHED and call
> schedule() ?

The paranoid exit path doesn=E2=80=99t check TIF_NEED_RESCHED because it=E2=80=
=99s fundamentally atomic =E2=80=94 it=E2=80=99s running on a percpu stack a=
nd it can=E2=80=99t schedule. In theory we could do some evil stack switchin=
g, but we don=E2=80=99t.

How does NMI handle this?  If an NMI that hit interruptible kernel code over=
flows a perf counter, how does the wake up work?

(do_int3() is special because it=E2=80=99s not actually IST.  But it can hit=
 in odd places due to kprobes, and I=E2=80=99m nervous about recursing incor=
rectly into RCU and context tracking code if we were to use exception_enter(=
).)

>=20
> I think (and this certainly wants a comment) is that the ist_exit()
> thing hard relies on the interrupt-return path doing the reschedule.