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

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, 7 Dec 2015 10:52:51 +0100
From:	"Wilck, Martin" <martin.wilck@...fujitsu.com>
To:	Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
CC:	Jason Gunthorpe <jgunthorpe@...idianresearch.com>,
	"tpmdd-devel@...ts.sourceforge.net" 
	<tpmdd-devel@...ts.sourceforge.net>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Uwe Kleine-König 
	<u.kleine-koenig@...gutronix.de>
Subject: Re: [tpmdd-devel] [PATCH v2 0/3] tpm_tis: Clean up force module
 parameter

> > > 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.
> > 
> > Repeat, the memory resource DOES exist on my system. Not sure what proof
> > you saw there.
> 
> Ok, lets go this through.
> 
> I deduced this from two facts:
> 
> * It used to have memory resource as conditional and as a fallback use
>   fixed value.
> * Your workaround reverted the situation to this.
> 
> Did I understand something incorrectly?

The problem in my case didn't occur because ACPI was lacking a resource.
It has one "extra" resource that Jason's original code didn't
recognize. 

Jason's code was wrongly assuming that a resource that isn't of type
"IRQ" has to be of type "MEMORY". If I print out the resource types
encountered in tpm_check_resource(), I get
ACPI_RESOURCE_TYPE_FIXED_MEMORY32  (0x0a) first, followed by
ACPI_RESOURCE_TYPE_END_TAG (0x07). The latter was mistakenly used by
Jason't code as a memory resource. This is how ACPI ResourceTemplates
work (a list with an end marker). The correct solution is to always
check the return value of acpi_dev_resource_memory(), as it's currently
implemented in Jason't current "for-jarkko" branch.

Martin

> 
> /Jarkko