Re: Re: [tpmdd-devel] [PATCH] tpm: improve tpm_tis send() performance by ignoring burstcount

From: Mimi Zohar <zohar@linux.vnet.ibm.com>
To: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc: Peter Huewe <PeterHuewe@gmx.de>,
	Ken Goldman <kgold@linux.vnet.ibm.com>,
	linux-ima-devel@lists.sourceforge.net,
	linux-security-module@vger.kernel.org,
	tpmdd-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org
Subject: Re: Re: [tpmdd-devel] [PATCH] tpm: improve tpm_tis send() performance by ignoring burstcount
Date: Mon, 14 Aug 2017 08:03:05 -0400	[thread overview]
Message-ID: <1502712185.6179.20.camel@linux.vnet.ibm.com> (raw)
In-Reply-To: <20170814105651.eo3e7tokt7mujeba@linux.intel.com>

On Mon, 2017-08-14 at 13:56 +0300, Jarkko Sakkinen wrote:

> > > > I would like to see tpm_msleep() wrapper to replace current msleep()
> > > > usage across the subsystem before considering this. I.e. wrapper that
> > > > internally uses usleep_range(). This way we can mechanically convert
> > > > everything to a more low latency option.
> > > 
> > > Fine.  I assume you meant tpm_sleep(), not tpm_msleep().
> > 
> > I think it would sense to have a function that takes msecs because msecs
> > are mostly used everywhere in the subsystem. This way we don't have to
> > change any of the existing constants.

For now converting from msleep() to tpm_msleep() will be straight
forward.  Internally we would just use usleep_range().

Going forward, my concern is that even 1 msec might be too long for
some of these sleeps.

Mimi