Re: [PATCH v12 04/10] KEYS: Move KEY_LOOKUP_ to include/linux/key.h

From: Jarkko Sakkinen <jarkko@kernel.org>
To: Roberto Sassu <roberto.sassu@huaweicloud.com>
Cc: ast@kernel.org, daniel@iogearbox.net, andrii@kernel.org,
	martin.lau@linux.dev, song@kernel.org, yhs@fb.com,
	john.fastabend@gmail.com, kpsingh@kernel.org, sdf@google.com,
	haoluo@google.com, jolsa@kernel.org, mykolal@fb.com,
	corbet@lwn.net, dhowells@redhat.com, rostedt@goodmis.org,
	mingo@redhat.com, paul@paul-moore.com, jmorris@namei.org,
	serge@hallyn.com, shuah@kernel.org, bpf@vger.kernel.org,
	linux-doc@vger.kernel.org, keyrings@vger.kernel.org,
	linux-security-module@vger.kernel.org,
	linux-kselftest@vger.kernel.org, linux-kernel@vger.kernel.org,
	deso@posteo.net, Roberto Sassu <roberto.sassu@huawei.com>
Subject: Re: [PATCH v12 04/10] KEYS: Move KEY_LOOKUP_ to include/linux/key.h
Date: Sun, 28 Aug 2022 06:57:41 +0300	[thread overview]
Message-ID: <YwrntbSJUs2cA7Xn@kernel.org> (raw)
In-Reply-To: <acae432697e854748d9a44c732ec8cab807d9d46.camel@huaweicloud.com>

On Fri, Aug 26, 2022 at 09:14:09AM +0200, Roberto Sassu wrote:
> On Fri, 2022-08-26 at 08:42 +0300, Jarkko Sakkinen wrote:
> > On Thu, Aug 18, 2022 at 05:29:23PM +0200, 
> > roberto.sassu@huaweicloud.com wrote:
> > > From: Roberto Sassu <roberto.sassu@huawei.com>
> > > 
> > > In preparation for the patch that introduces the
> > > bpf_lookup_user_key() eBPF
> > > kfunc, move KEY_LOOKUP_ definitions to include/linux/key.h, to be
> > > able to
> > > validate the kfunc parameters.
> > > 
> > > Also, introduce key_lookup_flags_check() directly in
> > > include/linux/key.h,
> > > to reduce the risk that the check is not in sync with currently
> > > defined
> > > flags.
> > 
> > Missing the description what the heck this function even is.
> > 
> > Please, explain that.
> 
> Hi Jarkko
> 
> sorry, forgot to update the commit description. Will do it.
> 
> > Also, the short subject is misleading because this *just*
> > does not move flags.
> > 
> > > Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
> > > Reviewed-by: KP Singh <kpsingh@kernel.org>
> > > ---
> > >  include/linux/key.h      | 11 +++++++++++
> > >  security/keys/internal.h |  2 --
> > >  2 files changed, 11 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/include/linux/key.h b/include/linux/key.h
> > > index 7febc4881363..b5bbae77a9e7 100644
> > > --- a/include/linux/key.h
> > > +++ b/include/linux/key.h
> > > @@ -88,6 +88,17 @@ enum key_need_perm {
> > >  	KEY_DEFER_PERM_CHECK,	/* Special: permission check is
> > > deferred */
> > >  };
> > >  
> > > +#define KEY_LOOKUP_CREATE	0x01
> > > +#define KEY_LOOKUP_PARTIAL	0x02
> > > +
> > 
> > /*
> >  * Explain what the heck this function is.
> >  */
> > > +static inline int key_lookup_flags_check(u64 flags)
> > > +{
> > > +	if (flags & ~(KEY_LOOKUP_CREATE | KEY_LOOKUP_PARTIAL))
> > > +		return -EINVAL;
> > > +
> > > +	return 0;
> > > +}
> > 
> > This is essentially a boolean function, right?
> > 
> > I.e. the implementation can be just:
> > 
> > !!(flags & ~(KEY_LOOKUP_CREATE | KEY_LOOKUP_PARTIAL))
> 
> Absolutely fine with that, if you prefer.

It can be either, it more depends on if a new function
is needed in the first place.

E.g. if you are worried about maintaining you could just
as well define a constant containing the mask, right?

> 
> > Not even sure if this is needed in the first place, or
> > would it be better just to open code it. How many call
> > sites does it have anyway?
> > 
> 
> Daniel preferred to have this check here.

How many call sites?

BR, Jarkko