Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <CAG-UpRQsdL_Fs9HSEv2pDYXehJC+YXcYjiZKFLvkGBTZkkaTcg@mail.gmail.com>
 <20230706120103.GJ2833176@hirez.programming.kicks-ass.net>
 <CAG-UpRTQ-ovVO02HRd5z-9oZGCPvZ0vODJse8oyuA9L-3NV_pQ@mail.gmail.com>
 <CAG-UpRRLFTHg64b0hG4=FbuzhhqNQEU8jGt6TygCVAK1BaT2kQ@mail.gmail.com>
 <20230707125942.GB2883469@hirez.programming.kicks-ass.net> <f4e839ac5ecb285b8a9d666f1a73e0a39b698864.camel@gmx.de>
In-Reply-To: <f4e839ac5ecb285b8a9d666f1a73e0a39b698864.camel@gmx.de>
From:   Henry Wu <triangletrap12@gmail.com>
Date:   Sat, 8 Jul 2023 01:28:25 +0800
Message-ID: <CAG-UpRTur0PHkWNZ0FqLnz5K85P-HQkuO-PCCxOwJ5XVgwgxyQ@mail.gmail.com>
Subject: Re: Fwd: Possible race in rt_mutex_adjust_prio_chain
To:     Mike Galbraith <efault@gmx.de>,
        Peter Zijlstra <peterz@infradead.org>
Cc:     linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi, Peter and Mike.

Mike Galbraith <efault@gmx.de> =E4=BA=8E2023=E5=B9=B47=E6=9C=887=E6=97=A5=
=E5=91=A8=E4=BA=94 23:39=E5=86=99=E9=81=93=EF=BC=9A
>
> On Fri, 2023-07-07 at 14:59 +0200, Peter Zijlstra wrote:
> >
> > The below implements this duplication and seems to not insta-crash.
>
> RT bits of ww_mutex.h needed tree_entry -> tree.entry.  Modulo that, RT
> seems content.
>
>         -Mike

I patched my kernel with Peter's patch and tested it with my test
program. I haven't seen any race so far and I will test more tomorrow.
Fedora's default kernel config doesn't enable CONFIG_PREEMPT_RT so I
didn't come across with compile error. I will test final patch if
available.