linux-kernel - Re: [PATCH 6/6] tpm2: add session handle context saving and restoring to the space code'

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]

Message-ID: <20170210190727.abaxbbpqq45kb2mm@intel.com>
Date:   Fri, 10 Feb 2017 21:07:27 +0200
From:   Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
To:     James Bottomley <James.Bottomley@...senPartnership.com>
Cc:     tpmdd-devel@...ts.sourceforge.net,
        linux-security-module@...r.kernel.org,
        Peter Huewe <peterhuewe@....de>,
        Marcel Selhorst <tpmdd@...horst.net>,
        Jason Gunthorpe <jgunthorpe@...idianresearch.com>,
        open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 6/6] tpm2: add session handle context saving and
 restoring to the space code'

On Fri, Feb 10, 2017 at 08:11:18AM -0800, James Bottomley wrote:
> On Fri, 2017-02-10 at 10:52 +0200, Jarkko Sakkinen wrote:
> > On Wed, Feb 08, 2017 at 01:07:08PM +0200, Jarkko Sakkinen wrote:
> > > +		rc = tpm2_load_context(chip, space->session_buf,
> > > +				       &offset, &handle);
> > > +		if (rc == -ENOENT) {
> > > +			/* load failed, just forget session */
> > > +			space->session_tbl[i] = 0;
> > 
> > This is my only concern in this commit. Should we also in this case 
> > just flush the space or not?
> 
> I elected not to.  If the handle is flushed by an external resource
> manager, we get this event.  If the RM and the app agreed to release
> the session handle, then flushing the space would be overkill because
> it would destroy the client session, so simply removing the handle
> works.  If the client tries to use the session again, it gets an error
> and if it doesn't everything just works, which seems to be optimal.
> 
> James

Makes sense. Just wanted the check the logic you had in this decision.

/Jarkko