Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH] coredump: Limit what can interrupt coredumps
To:     Jens Axboe <axboe@kernel.dk>,
        Olivier Langlois <olivier@trillion01.com>,
        "Eric W. Biederman" <ebiederm@xmission.com>,
        Oleg Nesterov <oleg@redhat.com>
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        io-uring <io-uring@vger.kernel.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        "Pavel Begunkov>" <asml.silence@gmail.com>
References: <CAHk-=wjC7GmCHTkoz2_CkgSc_Cgy19qwSQgJGXz+v2f=KT3UOw@mail.gmail.com>
 <198e912402486f66214146d4eabad8cb3f010a8e.camel@trillion01.com>
 <87eeda7nqe.fsf@disp2133>
 <b8434a8987672ab16f9fb755c1fc4d51e0f4004a.camel@trillion01.com>
 <87pmwt6biw.fsf@disp2133> <87czst5yxh.fsf_-_@disp2133>
 <CAHk-=wiax83WoS0p5nWvPhU_O+hcjXwv6q3DXV8Ejb62BfynhQ@mail.gmail.com>
 <87y2bh4jg5.fsf@disp2133>
 <CAHk-=wjPiEaXjUp6PTcLZFjT8RrYX+ExtD-RY3NjFWDN7mKLbw@mail.gmail.com>
 <87sg1p4h0g.fsf_-_@disp2133> <20210614141032.GA13677@redhat.com>
 <87pmwmn5m0.fsf@disp2133>
 <4d93d0600e4a9590a48d320c5a7dd4c54d66f095.camel@trillion01.com>
 <8af373ec-9609-35a4-f185-f9bdc63d39b7@cybernetics.com>
 <9d194813-ecb1-2fe4-70aa-75faf4e144ad@kernel.dk>
From:   Tony Battersby <tonyb@cybernetics.com>
Message-ID: <7b201ca7-dd1d-61be-8586-5dbf7a3c9333@cybernetics.com>
Date:   Thu, 12 Aug 2021 09:53:21 -0400
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.11.0
MIME-Version: 1.0
In-Reply-To: <9d194813-ecb1-2fe4-70aa-75faf4e144ad@kernel.dk>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Precedence: bulk

On 8/11/21 9:55 PM, Jens Axboe wrote:
>
> That is very interesting. Like Olivier mentioned, it's not that actual
> commit, but rather the change of behavior implemented by it. Before that
> commit, we'd hit the async workers more often, whereas after we do the
> correct retry method where it's driven by the wakeup when the page is
> unlocked. This is purely speculation, but perhaps the fact that the
> process changes state potentially mid dump is why the dump ends up being
> truncated?
>
> I'd love to dive into this and try and figure it out. Absent a test
> case, at least the above gives me an idea of what to try out. I'll see
> if it makes it easier for me to create a case that does result in a
> truncated core dump.
>
If it helps, a "good" coredump from my program is about 350 MB
compressed down to about 7 MB by bzip2.  A truncated coredump varies in
size from about 60 KB to about 2 MB before compression.  The program
that receives the coredump uses bzip2 to compress the data before
writing it to disk.

Tony