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;
MIME-Version: 1.0
References: <20210126204125.313820-1-pasha.tatashin@soleen.com>
 <20210126204125.313820-2-pasha.tatashin@soleen.com> <87lfcczvab.fsf@x220.int.ebiederm.org>
In-Reply-To: <87lfcczvab.fsf@x220.int.ebiederm.org>
From:   Pavel Tatashin <pasha.tatashin@soleen.com>
Date:   Thu, 28 Jan 2021 16:43:51 -0500
Message-ID: <CA+CK2bBMO+USv3SE25GNQDqpY-+4cEXrKbAmaSmsko=_CS6ZPQ@mail.gmail.com>
Subject: Re: [PATCH v2 1/1] kexec: dump kmessage before machine_kexec
To:     "Eric W. Biederman" <ebiederm@xmission.com>
Cc:     James Morris <jmorris@namei.org>, Sasha Levin <sashal@kernel.org>,
        Tyler Hicks <tyhicks@linux.microsoft.com>,
        Petr Mladek <pmladek@suse.com>,
        Kees Cook <keescook@chromium.org>,
        Anton Vorontsov <anton@enomsg.org>,
        Colin Cross <ccross@android.com>,
        Tony Luck <tony.luck@intel.com>,
        kexec mailing list <kexec@lists.infradead.org>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Jan 28, 2021 at 3:01 PM Eric W. Biederman <ebiederm@xmission.com> wrote:
>
> Pavel Tatashin <pasha.tatashin@soleen.com> writes:
>
> > kmsg_dump(KMSG_DUMP_SHUTDOWN) is called before
> > machine_restart(), machine_halt(), machine_power_off(), the only one that
> > is missing is  machine_kexec().
> >
> > The dmesg output that it contains can be used to study the shutdown
> > performance of both kernel and systemd during kexec reboot.
> >
> > Here is example of dmesg data collected after kexec:
>
> As long was we keep kmsg_dump out of the crash_kexec path where
> it completely breaks kexec on panic this seems a reasonable thing to do.
> On the ordinary kernel_kexec path everything is expected to be working.

This is an ordinary kexec reboot path, not kdump.

>
> Is kmsg_dump expected to work after all of the device drivers
> are shut down?  Otherwise this placement of kmsg_dump is too late.

Yes it is. It is called after device_shutdown(); in all other places.


Thank you,
Pasha