Re: [PATCH v1.3 2/4] key: add tpm_send command

From: Rajiv Andrade <srajiv@linux.vnet.ibm.com>
To: David Howells <dhowells@redhat.com>
Cc: David Safford <safford@watson.ibm.com>,
	Mimi Zohar <zohar@linux.vnet.ibm.com>,
	linux-kernel@vger.kernel.org,
	linux-security-module@vger.kernel.org, keyrings@linux-nfs.org,
	linux-crypto@vger.kernel.org,
	Jason Gunthorpe <jgunthorpe@obsidianresearch.com>,
	James Morris <jmorris@namei.org>
Subject: Re: [PATCH v1.3 2/4] key: add tpm_send command
Date: Wed, 17 Nov 2010 11:12:51 -0200	[thread overview]
Message-ID: <4CE3D4D3.7020306@linux.vnet.ibm.com> (raw)
In-Reply-To: <28655.1289599899@redhat.com>

On 12/11/10 20:11, David Howells wrote:
> David Safford<safford@watson.ibm.com>  wrote:
>
>> David, does this look ok to you? If so, I will do two patches, one to
>> fix the helper name throughout the existing tpm.c, and then a new
>> version of the tpm_send patch which uses the new name.
> I prefer my suggestion: Wrapping the module_put() up so that you don't see it
> directly.  Then you don't need to alter tpm_chip_find_get().  I'll argue that
> you don't need to know how tpm_chips are got/put, except in the code that
> wraps it.
>
tpm_chip_find_get() not only gets the tpm_chip, but also searches for it 
given an index in a tpm_chip_list.
tpm_chip_put() is then the name that fits the argument here, given it'll 
only be a wrapper of the tpm_chip put
functionality, not the searching one I assume.

I'll ack any of the two approaches (tpm_chip_put or tpm_chip_find_get 
renaming) in any case.

Rajiv