Re: [tpmdd-devel] [PATCH v2 0/3] tpm_tis: Clean up force module parameter

From: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
To: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Cc: "Martin Wilck" <Martin.Wilck@ts.fujitsu.com>,
	tpmdd-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org,
	"Uwe Kleine-König" <u.kleine-koenig@pengutronix.de>
Subject: Re: [tpmdd-devel] [PATCH v2 0/3] tpm_tis: Clean up force module parameter
Date: Sun, 6 Dec 2015 06:20:31 +0200	[thread overview]
Message-ID: <20151206042031.GA5831@intel.com> (raw)
In-Reply-To: <20151206041544.GA5585@intel.com>

On Sun, Dec 06, 2015 at 06:15:44AM +0200, Jarkko Sakkinen wrote:
> On Sun, Dec 06, 2015 at 06:02:26AM +0200, Jarkko Sakkinen wrote:
> > On Thu, Dec 03, 2015 at 11:19:32AM -0700, Jason Gunthorpe wrote:
> > > On Thu, Dec 03, 2015 at 08:00:42AM +0200, Jarkko Sakkinen wrote:
> > > 
> > > > I guess it'd be more realiable. In my NUC the current fix works and the
> > > > people who tested it. If you supply me a fix that changes it to use that
> > > > I can test it and this will give also coverage to the people who tested
> > > > my original fix.
> > > 
> > > Here is the updated series:
> > > 
> > > https://github.com/jgunthorpe/linux/commits/for-jarkko
> > > 
> > > What does your dmesg say?
> > > 
> > > It really isn't OK to hardwire an address for acpi devices, so I've
> > > added something like this. Just completely guessing that control_pa is
> > > where the BIOS is hiding the base address. Maybe it is cca->cmd_pa ?
> > 
> > I'm a bit confused about the discussion because Martin replied that
> > tpm_tis used to get the address range before applying this series.
> > 
> > And pnp_driver in the backend for TPM 1.x devices grabs the address
> > range from DSDT.
> 
> You can completely ignore this question. I saw Martins reply with a fix for
> "tpm_tis: Use devm_ioremap_resource" that you should squash into that
> change. So it's proved that TPM ACPI device objects do not always have a
> memory resource. Good.
> 
> I think these changes are important but there's no really reason to rush
> them. Maybe, since there's been a lot of commentary, it'd be better to
> resubmit a new revision of the series to the mailing list so that it can
> be peer-reviewed once again.

Maybe even there could be a common tpm_tcg driver once the common code
has been factored out (at some point, lets take this step by step and
fix the issues first). Transmit functions are not heavy and ACPI stuff
is mostly the same.

/Jarkko