linux-kernel - Re: of_reserved_mem()/kexec interaction

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: <f48b0010-a136-48a4-bfc7-7bbaf91b4da9@arm.com>
Date:   Tue, 2 Apr 2019 11:14:25 +0100
From:   James Morse <james.morse@....com>
To:     Roy Pledge <roy.pledge@....com>
Cc:     Rob Herring <robh@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>
Subject: Re: of_reserved_mem()/kexec interaction

Hi Roy,

On 01/04/2019 20:09, Roy Pledge wrote:
> I'm trying to understand if memory reserved in the device tree via the
> "reserved-memory" facility is preserved during a kexec system call,
> i.e., is the memory at the same location with the contents undisturbed
> when the new kernel starts?

If the reservation is a static-allocation (so the address appears in the DT) then the new
kernel should know this is reserved too, and not touch it.

If its a dynamic-allocation, the address isn't in the DT, so the new kernel can't know. It
will dynamically allocate a new reservation, which may be in a different place. If the
data didn't matter at the first-boot, it probably doesn't matter over subsequent kexec either.

As an outlier: the gic has some funny requirements around this. It needs to dynamically
allocate a page during first-boot that is preserved over kexec. It does this using
efi_mem_reserve_persistent().

Thanks,

James