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;
Date:   Thu, 5 Dec 2019 18:55:45 +0800
From:   Dave Young <dyoung@redhat.com>
To:     Ingo Molnar <mingo@kernel.org>
Cc:     linux-efi@vger.kernel.org, x86@kernel.org,
        linux-kernel@vger.kernel.org,
        Michael Weiser <michael@weiser.dinsnail.net>,
        Ard Biesheuvel <ard.biesheuvel@linaro.org>,
        kexec@lists.infradead.org, Ingo Molnar <mingo@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        "Eric W. Biederman" <ebiederm@xmission.com>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH] x86/efi: update e820 about reserved EFI boot services
 data to fix kexec breakage
Message-ID: <20191205105545.GA6710@dhcp-128-65.nay.redhat.com>
References: <20191204075233.GA10520@dhcp-128-65.nay.redhat.com>
 <20191204075917.GA10587@dhcp-128-65.nay.redhat.com>
 <20191204101412.GD114697@gmail.com>
MIME-Version: 1.0
In-Reply-To: <20191204101412.GD114697@gmail.com>
User-Agent: Mutt/1.12.1 (2019-06-15)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 12/04/19 at 11:14am, Ingo Molnar wrote:
>=20
> * Dave Young <dyoung@redhat.com> wrote:
>=20
> > On 12/04/19 at 03:52pm, Dave Young wrote:
> > > Michael Weiser reported he got below error during a kexec rebooting:
> > > esrt: Unsupported ESRT version 2904149718861218184.
> > >=20
> > > The ESRT memory stays in EFI boot services data, and it was reserved
> > > in kernel via efi_mem_reserve().  The initial purpose of the reservat=
ion
> > > is to reuse the EFI boot services data across kexec reboot. For examp=
le
> > > the BGRT image data and some ESRT memory like Michael reported.=20
> > >=20
> > > But although the memory is reserved it is not updated in X86 e820 tab=
le.
> > > And kexec_file_load iterate system ram in io resource list to find pl=
aces
> > > for kernel, initramfs and other stuff. In Michael's case the kexec lo=
aded
> > > initramfs overwritten the ESRT memory and then the failure happened.
> >=20
> > s/overwritten/overwrote :)  If need a repost please let me know..
> >=20
> > >=20
> > > Since kexec_file_load depends on the e820 to be updated, just fix thi=
s
> > > by updating the reserved EFI boot services memory as reserved type in=
 e820.
> > >=20
> > > Originally any memory descriptors with EFI_MEMORY_RUNTIME attribute a=
re
> > > bypassed in the reservation code path because they are assumed as res=
erved.
> > > But the reservation is still needed for multiple kexec reboot.
> > > And it is the only possible case we come here thus just drop the code
> > > chunk then everything works without side effects.=20
> > >=20
> > > On my machine the ESRT memory sits in an EFI runtime data range, it d=
oes
> > > not trigger the problem, but I successfully tested with BGRT instead.
> > > both kexec_load and kexec_file_load work and kdump works as well.
> > >=20
> > > Signed-off-by: Dave Young <dyoung@redhat.com>
>=20
>=20
> So I edited this to:
>=20
>  From: Dave Young <dyoung@redhat.com>
>=20
>  Michael Weiser reported he got this error during a kexec rebooting:
>=20
>    esrt: Unsupported ESRT version 2904149718861218184.
>=20
>  The ESRT memory stays in EFI boot services data, and it was reserved
>  in kernel via efi_mem_reserve().  The initial purpose of the reservation
>  is to reuse the EFI boot services data across kexec reboot. For example
>  the BGRT image data and some ESRT memory like Michael reported.
>=20
>  But although the memory is reserved it is not updated in the X86 E820 ta=
ble,
>  and kexec_file_load() iterates system RAM in the IO resource list to fin=
d places
>  for kernel, initramfs and other stuff. In Michael's case the kexec loade=
d
>  initramfs overwrote the ESRT memory and then the failure happened.
>=20
>  Since kexec_file_load() depends on the E820 table being updated, just fi=
x this
>  by updating the reserved EFI boot services memory as reserved type in E8=
20.
>=20
>  Originally any memory descriptors with EFI_MEMORY_RUNTIME attribute are
>  bypassed in the reservation code path because they are assumed as reserv=
ed.
>=20
>  But the reservation is still needed for multiple kexec reboots,
>  and it is the only possible case we come here thus just drop the code
>  chunk, then everything works without side effects.
>=20
>  On my machine the ESRT memory sits in an EFI runtime data range, it does
>  not trigger the problem, but I successfully tested with BGRT instead.
>  both kexec_load() and kexec_file_load() work and kdump works as well.
>=20

Thanks for the amending, also thank all for the review and test.

Dave