linux-kernel - RE: [PATCH 1/3 RESEND] tpm: add longer timeouts for creation commands.

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: <5B8DA87D05A7694D9FA63FD143655C1B9422F018@hasmsx108.ger.corp.intel.com>
Date:   Tue, 6 Mar 2018 08:09:26 +0000
From:   "Winkler, Tomas" <tomas.winkler@...el.com>
To:     Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
CC:     Jason Gunthorpe <jgg@...pe.ca>,
        "Usyskin, Alexander" <alexander.usyskin@...el.com>,
        "linux-integrity@...r.kernel.org" <linux-integrity@...r.kernel.org>,
        "linux-security-module@...r.kernel.org" 
        <linux-security-module@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH 1/3 RESEND] tpm: add longer timeouts for creation
 commands.



> 
> On Mon, Mar 05, 2018 at 06:04:56PM +0000, Winkler, Tomas wrote:
> > >
> > > On Mon, Mar 05, 2018 at 01:09:09PM +0000, Winkler, Tomas wrote:
> > > > > enum tpm_duration {
> > > > > 	TPM_DURATION_DEFAULT = 2000,
> > > > > 	TPM_DURATION_LONG = 300000,
> > > > > };
> > > > >
> > > > How is this aligned with the spec PTP spec?
> > >
> > > For TPM 2.0 that spec only partially defines durations for CCs and
> > > thus our look up table is already kind "flakky". In a sense that the
> > > default duration is upper limit for spec defined durations.
> >
> > The timeouts for LONG and MEDIUM is defined by the  PTP spec,  we need
> to maintain those as those effect the system.
> > The UNDEFINED and LONG LONG is the implementation choice we driver
> from empirical data we have so far.
> 
> Where can be get this empirical data?
>From testing the HW.
> 
> You are not only adding 30s delay but also turning the 2s delay to 12s delay.


I'm adding 3 min, no other changes.  Where is 12s?

> IMHO we could very well use PTP LONG for all commands as the timeout.
> Why that would not work?


Empirically it doesn't go test it you have the HW.

Thanks
Tomas