Re: [PATCH v1 4/8] LSM: Maintain a table of LSM attribute data

From: Greg KH <gregkh@linuxfoundation.org>
To: Casey Schaufler <casey@schaufler-ca.com>
Cc: casey.schaufler@intel.com, paul@paul-moore.com,
	linux-security-module@vger.kernel.org, jmorris@namei.org,
	keescook@chromium.org, john.johansen@canonical.com,
	penguin-kernel@i-love.sakura.ne.jp,
	stephen.smalley.work@gmail.com, linux-kernel@vger.kernel.org,
	linux-api@vger.kernel.org, mic@digikod.net
Subject: Re: [PATCH v1 4/8] LSM: Maintain a table of LSM attribute data
Date: Thu, 27 Oct 2022 19:13:47 +0200	[thread overview]
Message-ID: <Y1q8SzpcdWgm/fLq@kroah.com> (raw)
In-Reply-To: <d545ef2a-5cc5-2848-e699-ff791d34d7c7@schaufler-ca.com>

On Thu, Oct 27, 2022 at 10:08:23AM -0700, Casey Schaufler wrote:
> On 10/26/2022 11:29 PM, Greg KH wrote:
> > On Wed, Oct 26, 2022 at 05:38:21PM -0700, Casey Schaufler wrote:
> >> On 10/25/2022 11:00 PM, Greg KH wrote:
> >>> On Tue, Oct 25, 2022 at 11:45:15AM -0700, Casey Schaufler wrote:
> >>>> As LSMs are registered add their lsm_id pointers to a table.
> >>>> This will be used later for attribute reporting.
> >>>>
> >>>> Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
> >>>> ---
> >>>>  include/linux/security.h | 17 +++++++++++++++++
> >>>>  security/security.c      | 18 ++++++++++++++++++
> >>>>  2 files changed, 35 insertions(+)
> >>>>
> >>>> diff --git a/include/linux/security.h b/include/linux/security.h
> >>>> index ca1b7109c0db..e1678594d983 100644
> >>>> --- a/include/linux/security.h
> >>>> +++ b/include/linux/security.h
> >>>> @@ -138,6 +138,23 @@ enum lockdown_reason {
> >>>>  
> >>>>  extern const char *const lockdown_reasons[LOCKDOWN_CONFIDENTIALITY_MAX+1];
> >>>>  
> >>>> +#define LSMID_ENTRIES ( \
> >>>> +	1 + /* capabilities */ \
> >>> No #define for capabilities?
> >> Nope. There isn't one. CONFIG_SECURITY takes care of it.
> >>
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_SELINUX) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_SMACK) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_TOMOYO) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_IMA) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_APPARMOR) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_YAMA) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_LOADPIN) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_SAFESETID) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_LOCKDOWN) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_BPF_LSM) ? 1 : 0) + \
> >>>> +	(IS_ENABLED(CONFIG_SECURITY_LANDLOCK) ? 1 : 0))
> >>>> +
> >>>> +extern int lsm_id;
> >>> u64?
> >> u32. I doubt we'll get more than 32K security modules.
> > These should be bits, not values, right?
> 
> lsm_id is the count of security modules that are registered.
> It seemed like a good name for the value at the time, but as
> it's causing confusion I should probably change it.

Yeah, that's confusing.  "lsm_num_availble" might be better.

> > Wait, this magic entry value is going to change depeneding on what is,
> > or is not, enabled.  How is that a stable user/kernel api at all?
> >
> > confused.
> 
> I'll clarify.
> 
> This patch isn't implementing an API, but is required by subsequent
> patches that do. Does linux-api want to see patches that are in support
> of APIs, or just those with actual API implementation?

There's nothing wrong with seeing this patch, I was just confused as it
seemed to be a user facing api.  It wasn't obvious to me, sorry.

greg k-h