Message-ID: <1489527771.19081.49.camel@HansenPartnership.com>
Subject: Re: [tpmdd-devel] [PATCH] tpm: Add sysfs interface to show TPM
 hardware version
From: James Bottomley <James.Bottomley@HansenPartnership.com>
To: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>,
        Peter Huewe <peterhuewe@gmx.de>
Cc: tpmdd-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org
Date: Tue, 14 Mar 2017 14:42:51 -0700
In-Reply-To: <20170314212425.GA29359@obsidianresearch.com>
References: <1489382517-9078-1-git-send-email-Meng.Li@windriver.com>
         <29B2C361-A0C0-410E-ACCA-07AFA382A251@gmx.de>
         <9f67321b-50bd-9e6f-61f8-c29ae14f261f@linux.vnet.ibm.com>
         <074515F2-51E7-485B-B1AB-11B913CC4B46@gmx.de>
         <20170314212425.GA29359@obsidianresearch.com>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 682
Lines: 19

On Tue, 2017-03-14 at 15:24 -0600, Jason Gunthorpe wrote:
> On Tue, Mar 14, 2017 at 07:58:37PM +0100, Peter Huewe wrote:
> 
> > > In practice, I suspect that a single user space application won't
> > > support both TPMs. 
> > Think of init scripts.
> > Which daemon should it start?
> 
> Right, ideally we'd have a udev rule that triggers systemd to start
> the userspace daemons when a TPM is detected, as other hardware does.
> 
> So whatever format we use has to be compatible with udev's matcher..

Technically, we have that already: my TPM2.0 detection stuff for udev
triggers off the presence of the SUBSYSTEM=="tpmrm"  that matches any
TPM2 device but not a 1.x one.

James