linux-kernel - Re: [PATCH v7 2/2] tracing: Add support for preempt and irq enable/disable events

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAJWu+ormUaN-vBsXwshs7N=jnofqd-OGnEpE1_Zcc68dsot3Tg@mail.gmail.com>
Date:   Wed, 4 Oct 2017 09:56:33 -0700
From:   Joel Fernandes <joelaf@...gle.com>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        LKML <linux-kernel@...r.kernel.org>, kernel-team@...roid.com
Subject: Re: [PATCH v7 2/2] tracing: Add support for preempt and irq
 enable/disable events

Hi Steven,

On Wed, Oct 4, 2017 at 9:44 AM, Steven Rostedt <rostedt@...dmis.org> wrote:
> On Wed, 4 Oct 2017 18:04:34 +0200
> Peter Zijlstra <peterz@...radead.org> wrote:
>
>> On Fri, Sep 29, 2017 at 02:22:45PM -0700, Joel Fernandes wrote:
>> > +++ b/kernel/trace/trace_irqsoff.c
>>
>> >  EXPORT_SYMBOL(trace_hardirqs_on);
>> >  EXPORT_SYMBOL(trace_hardirqs_off);
>> >  EXPORT_SYMBOL(trace_hardirqs_on_caller);
>> >  EXPORT_SYMBOL(trace_hardirqs_off_caller);
>>
>> Steve, how does this compiler with lockdep enabled? Because:
>>
>> kernel/locking/lockdep.c:EXPORT_SYMBOL(trace_hardirqs_on_caller);
>> kernel/locking/lockdep.c:EXPORT_SYMBOL(trace_hardirqs_on);
>> kernel/locking/lockdep.c:EXPORT_SYMBOL(trace_hardirqs_off_caller);
>> kernel/locking/lockdep.c:EXPORT_SYMBOL(trace_hardirqs_off);
>
> I'll have to see if it does, I haven't applied them yet. Looks like
> they were removed from the:
>
> #ifdef CONFIG_PROVE_LOCKING
>
> [..]
>
> #else /* !CONFIG_PROVE_LOCKING */
>
> <here>
>
> #endif
>
>
> I don't see the protection in the patches. I will definitely test that
> before pushing it anywhere.

The protection added by CONFIG_PROVE_LOCKING was already there before
my patches. I just moved the code around in patch 1/2, nothing changed
in that patch. I already tested the combination of PROVE_LOCKING
enabled and disabled and several other combinations on your -next
tree. It all compiled fine. Here is my test script FWIW:
(By the way, do you mean that another tree/branch than your -next or
Linus's master has protection removed? If so, could you let me know
which tree/branch?)

#!/bin/bash -x

set -e

function disable_all() {
        make x86_64_defconfig
        ./scripts/config --disable CONFIG_PREEMPTIRQ_EVENTS
        ./scripts/config --disable CONFIG_IRQSOFF_TRACER
        ./scripts/config --disable CONFIG_PREEMPT_TRACER
        ./scripts/config --disable CONFIG_PREEMPT
        ./scripts/config --disable CONFIG_PROVE_LOCKING
}

function build_all() {
        make olddefconfig
        make -j8
}

disable_all
./scripts/config --enable CONFIG_IRQSOFF_TRACER
build_all

disable_all
./scripts/config --enable CONFIG_PREEMPT
./scripts/config --enable CONFIG_PREEMPTIRQ_EVENTS
build_all

disable_all
./scripts/config --enable CONFIG_PROVE_LOCKING
build_all

disable_all
./scripts/config --enable CONFIG_PROVE_LOCKING
./scripts/config --enable CONFIG_PREEMPT
./scripts/config --enable CONFIG_PREEMPT_TRACER
build_all

disable_all
./scripts/config --enable CONFIG_PROVE_LOCKING
./scripts/config --enable CONFIG_IRQSOFF_TRACER
build_all

disable_all
./scripts/config --enable CONFIG_PROVE_LOCKING
./scripts/config --enable CONFIG_PREEMPT
./scripts/config --enable CONFIG_PREEMPTIRQ_EVENTS
build_all

disable_all
./scripts/config --enable CONFIG_PROVE_LOCKING
./scripts/config --enable CONFIG_PREEMPT
./scripts/config --enable CONFIG_PREEMPTIRQ_EVENTS
./scripts/config --enable CONFIG_PREEMPT_TRACER
./scripts/config --enable CONFIG_IRQSOFF_TRACER
build_all