linux-kernel - Re: [PATCH v2 0/3] Runtime PM for TPM2 CRB chips

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: <20160619171531.GA25349@intel.com>
Date:	Sun, 19 Jun 2016 19:15:31 +0200
From:	Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
To:	Peter Huewe <peterhuewe@....de>
Cc:	Jason Gunthorpe <jgunthorpe@...idianresearch.com>,
	open list <linux-kernel@...r.kernel.org>,
	"moderated list:TPM DEVICE DRIVER" 
	<tpmdd-devel@...ts.sourceforge.net>
Subject: Re: [PATCH v2 0/3] Runtime PM for TPM2 CRB chips

On Sat, Jun 18, 2016 at 05:10:25PM +0200, Jarkko Sakkinen wrote:
> These patches implement support for runtime PM with TPM2 CRB chips such
> as PTT in Skylake.
> 
> This is very naive implementation since the TPM is resumed and supended
> for every transaction thus causing total 40 ms latency for each
> transaction but for the existing use cases this shouldn't be an issue.
> 
> Jarkko Sakkinen (3):
>   tpm_crb: fix crb_req_canceled behavior
>   tpm, tpm_crb: remove wmb()'s
>   tpm, tpm_crb: runtime power management
> 
>  drivers/char/tpm/tpm-interface.c |  3 ++
>  drivers/char/tpm/tpm_crb.c       | 60 +++++++++++++++++++++++++++++++++++-----
>  2 files changed, 56 insertions(+), 7 deletions(-)

Reviewing myself:

* I fsckd the description. The latency is whopping 400 ms, not 4 ms.
  Sorry about that. Thus it's better to use autosuspend.
* Callbacks should take tpm_mutex before doing anything.

I guess that tpm_tis could later on also take advantage of this work?

/Jarkko