[PATCH v3 08/10] certs: Check that builtin blacklist hashes are valid

[Mickaël Salaün]

On 21/01/2021 00:53, Jarkko Sakkinen wrote:
> On Wed, Jan 20, 2021 at 12:57:55PM +0100, Mickaël Salaün wrote:
>>
>> On 20/01/2021 06:19, Jarkko Sakkinen wrote:
>>> On Thu, Jan 14, 2021 at 04:19:07PM +0100, Mickaël Salaün wrote:
>>>> From: Mickaël Salaün <mic at linux.microsoft.com>
>>>>
>>>> Add and use a check-blacklist-hashes.awk script to make sure that the
>>>> builtin blacklist hashes will be approved by the run time blacklist
>>>> description checks.  This is useful to debug invalid hash formats, and
>>>> it make sure that previous hashes which could have been loaded in the
>>>> kernel (but ignored) are now noticed and deal with by the user.
>>>>
>>>> Cc: David Howells <dhowells at redhat.com>
>>>> Cc: David Woodhouse <dwmw2 at infradead.org>
>>>> Signed-off-by: Mickaël Salaün <mic at linux.microsoft.com>
>>>> Acked-by: Jarkko Sakkinen <jarkko at kernel.org>
>>>
>>> I get this with a self-signed cert:
>>>
>>> certs/Makefile:18: *** target pattern contains no '%'.  Stop.
>>>
>>> CONFIG_SYSTEM_BLACKLIST_HASH_LIST="tbs:8eed1340eef37c1dc84d996406ad05c7dbb3eade19132d688408ca2f63904869"
>>
>> As said in the Kconfig documentation for
>> CONFIG_SYSTEM_BLACKLIST_HASH_LIST, you need to provide a file with the
>> list, not to set the string directly in the configuration variable. This
>> patch series didn't change this behavior. The same kind of macros are
>> used for CONFIG_MODULE_SIG_KEY.
> 
> OK, the documentation just states that:
> 
> "Hashes to be preloaded into the system blacklist keyring"
> 
> No mention about a file. I'd add a patch to update this documentation.

I was referring to the full description:

config SYSTEM_BLACKLIST_HASH_LIST
	string "Hashes to be preloaded into the system blacklist keyring"
	depends on SYSTEM_BLACKLIST_KEYRING
	help
	  If set, this option should be the filename of a list of hashes in the
	  form "<hash>", "<hash>", ... .  This will be included into a C
	  wrapper to incorporate the list into the kernel.  Each <hash> should
	  be a string of hex digits.

…but the short description doesn't mention filename.

> 
>>
>>>
>>> I used the script in 10/10 to test this, which is another
>>> reamark: the patches are in invalid order, as you need to
>>> apply 10/10 before you can test  8/10.
>>
>> I'll move patch 10/10 earlier but this kind of formatting was already
>> required (but silently ignored) for this option to be really taken into
>> account. Only the kernel code was available to understand how to
>> effectively create such hash.
> 
> Great, thanks.
> 
> 
>>>
>>> /Jarkko
>>>
> 
> 
> /Jarkko
>