linux-kernel - Re: [PATCH v4 2/4] tpm: ignore burstcount to improve tpm

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:   Mon, 27 Nov 2017 08:22:28 -0500
From:   Mimi Zohar <zohar@...ux.vnet.ibm.com>
To:     Leendert van Doorn <leendert@...amecium.org>,
        Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>,
        Nayna Jain <nayna@...ux.vnet.ibm.com>
Cc:     "Alexander.Steffen@...ineon.com" <Alexander.Steffen@...ineon.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>,
        "peterhuewe@....de" <peterhuewe@....de>,
        "tpmdd@...horst.net" <tpmdd@...horst.net>,
        "patrickc@...ibm.com" <patrickc@...ibm.com>,
        "Safford, David (GE Global Research, US)" <david.safford@...com>
Subject: Re: [PATCH v4 2/4] tpm: ignore burstcount to improve tpm_tis send()
 performance

Hi, Leendert!

On Mon, 2017-11-27 at 07:08 +0000, Leendert van Doorn wrote:
> Hmm, this is almost 20 years old code (
> 
> I think the original code did a burst write and didn't check for
> error conditions until the very last byte write. I seem to remember
> that there was some text in the original standard to that effect
> (this may have gone back as far as IBM's ESS spec).

I really appreciate your responding with your recollections.

> The current code does check for error conditions after each write
> byte(s) so I don't think there is any reason for this anymore.
> Changing the while condition to count < len and setting burstcnt =
> min_t(int, burstcnt, len - count) and remove the
> tpm_tis_write8/wait_for_tpm_stat/tpm_tis_status clauses after the
> while loop should be sufficient.

The context for my question was about removing the while loop and
sending all of the data at once, relying on the bus wait states.
 Before making the change, I wanted to make sure there wasn't anything
special about the last byte.

thanks,

Mimi