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] [day] [month] [year] [list]
Message-ID: <CA+X5Wn7BBAc8dFE8GVS=bUUerFzW6_w0f178vKH-sRQ6qQQgWw@mail.gmail.com>
Date:	Tue, 9 Aug 2016 06:14:39 -0400
From:	james harvey <jamespharvey20@...il.com>
To:	linux-kernel@...r.kernel.org
Subject: Re: Kernel panic - Fails to load BIOS-E820 memory map booted through
 iPXE, uses BIOS-88

On Mon, Aug 8, 2016 at 8:33 PM, james harvey <jamespharvey20@...il.com> wrote:
> <<< How does booting through iPXE disrupt the BIOS E820 map? >>>
>
> Booting 4.6.4 from the Arch 2016.08.01 ISO on a DVD properly sees
> 64GB.  (See quote 1.)
>
> Booting the same ISO through iPXE only sees about 50*M*B usable, and
> the detected memory map sections are reported as "BIOS-88" rather than
> "BIOS-e820".  (See quote 2.)  I have seen discussions that say the
> 0x88 BIOS memory function may limit reporting to 15MB, or sometimes up
> to 64MB.
>
> This gives a kernel panic for "initrd too large to handle, disabling
> initrd (27,426,236 needed, 26,398,720 available)" (commas added.)
>
> How would booting through iPXE disrupt kernel's ability to obtain the
> BIOS-supplied E820 map?  This is happening in
> linux/arch/x86/kernel/e820.c::default_machine_specific_memory_setup().
>
> What are my avenues forward?
>
> (... snip ...)

Resolved.

Was hitting the card's CTRL+B hotkey during POST, since my BIOS wasn't
recognizing it as a bootable device.  Tracked down using iPXE
debugging that the BIOS E820 function isn't available during the POST
stage.

Luckily for me, my motherboard released a BIOS update that partially
fixes recognizing it as a bootable device in the past month or so, so
I've been able to move past the issue.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ