linux-kernel - Re: tpm_tis TPM2.0 not detected on cold boot

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] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 01 Jan 2019 17:47:41 +0100
From:   Michael Niewöhner <linux@...ewoehner.de>
To:     Mimi Zohar <zohar@...ux.ibm.com>,
        Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>,
        James Bottomley <James.Bottomley@...senPartnership.com>,
        peterhuewe@....de, jgg@...pe.ca, arnd@...db.de,
        linux-integrity@...r.kernel.org,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Nayna Jain <nayna@...ux.ibm.com>,
        Ken Goldman <kgold@...ux.ibm.com>
Subject: Re: tpm_tis TPM2.0 not detected on cold boot

On Tue, 2019-01-01 at 11:38 -0500, Mimi Zohar wrote:
> On Tue, 2019-01-01 at 17:15 +0100, Michael Niewöhner wrote:
> > On Mon, 2018-12-31 at 16:17 -0500, Mimi Zohar wrote:
> > > On Sun, 2018-12-30 at 14:22 +0100, Michael Niewöhner wrote:
> > > 
> > > > > difference is that on a cold boot, the TPM takes longer to initialize.
> > > > 
> > > > Well, as I said. Waiting for 10, 20 or even 60 seconds in the boot
> > > > manager
> > > > does
> > > > not solve the problem. So the problem is NOT that the TPM takes longer
> > > > to
> > > > initialize. Even adding a delay of 20 seconds before TPM init does not
> > > > solve
> > > > that while that should be more than enough time.
> > > 
> > > The purpose of commenting out the TPM2 selftest was to minimize the
> > > TPM initialization delay, so that the TPM is ready before IMA.  After
> > > James' patch that wasn't needed anymore.
> > > 
> > > Looking back at this thread, I see you're using systemd-boot, not
> > > grub2.  When you commented out the systemd-boot timeout, IMA found the
> > > TPM.  The question is why isn't the TPM ready with the timeout before
> > > IMA (like above)?  Has systemd-boot done the selftest?
> > 
> > I am not sure wether systemd-boot touches TPM at all but I get the same
> > behaviour with syslinux-efi.
> 
> From looking at the source code, it depends on whether systemd was
> compiled with ENABLE_TPM enabled(eg. src/boot/efi/boot.c,
> src/boot/efi/measure.c, src/boot/efi/stub.c).

Debian build it with TPM enabled. I just checked syslinux-efi which does not do
anything TPM-related.

> 
> Mimi
>