[PATCH 1/2] [lsm] introduce a new hook to query LSM for functionality

[PATCH 1/2] [lsm] introduce a new hook to query LSM for functionality

[Olga Kornievskaia]

From: Olga Kornievskaia <kolga@netapp.com>

Add a new hook func_query_vfs to query an LSM module (such as
SELinux) with the intention of finding whether or not it is enabled
or perhaps supports some functionality.

NFS stores security labels for file system objects and SElinux
or any other LSM module can query those objects. But it's
wasteful to do so when no security enforcement is taking place.
Using a new API call of security_func_query_vfs() and asking if

Suggested-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
---
 include/linux/lsm_hook_defs.h | 1 +
 include/linux/security.h      | 4 ++++
 security/security.c           | 6 ++++++
 security/selinux/hooks.c      | 7 +++++++
 4 files changed, 18 insertions(+)

diff --git a/include/linux/lsm_hook_defs.h b/include/linux/lsm_hook_defs.h
index 32a940117e7a..df3454a1fcac 100644
--- a/include/linux/lsm_hook_defs.h
+++ b/include/linux/lsm_hook_defs.h
@@ -257,6 +257,7 @@ LSM_HOOK(int, 0, inode_notifysecctx, struct inode *inode, void *ctx, u32 ctxlen)
 LSM_HOOK(int, 0, inode_setsecctx, struct dentry *dentry, void *ctx, u32 ctxlen)
 LSM_HOOK(int, 0, inode_getsecctx, struct inode *inode, void **ctx,
 	 u32 *ctxlen)
+LSM_HOOK(int, 0, func_query_vfs, unsigned int)
 
 #if defined(CONFIG_SECURITY) && defined(CONFIG_WATCH_QUEUE)
 LSM_HOOK(int, 0, post_notification, const struct cred *w_cred,
diff --git a/include/linux/security.h b/include/linux/security.h
index bc2725491560..e15964059fa4 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -456,6 +456,10 @@ int security_inode_notifysecctx(struct inode *inode, void *ctx, u32 ctxlen);
 int security_inode_setsecctx(struct dentry *dentry, void *ctx, u32 ctxlen);
 int security_inode_getsecctx(struct inode *inode, void **ctx, u32 *ctxlen);
 int security_locked_down(enum lockdown_reason what);
+#define LSM_FQUERY_VFS_NONE     0x00000000
+#define LSM_FQUERY_VFS_XATTRS   0x00000001
+int security_func_query_vfs(unsigned int flags);
+
 #else /* CONFIG_SECURITY */
 
 static inline int call_blocking_lsm_notifier(enum lsm_event event, void *data)
diff --git a/security/security.c b/security/security.c
index a28045dc9e7f..502b33865238 100644
--- a/security/security.c
+++ b/security/security.c
@@ -2067,6 +2067,12 @@ int security_inode_getsecctx(struct inode *inode, void **ctx, u32 *ctxlen)
 }
 EXPORT_SYMBOL(security_inode_getsecctx);
 
+int security_func_query_vfs(unsigned int flags)
+{
+	return call_int_hook(func_query_vfs, 0, flags);
+}
+EXPORT_SYMBOL(security_func_query_vfs);
+
 #ifdef CONFIG_WATCH_QUEUE
 int security_post_notification(const struct cred *w_cred,
 			       const struct cred *cred,
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 6b1826fc3658..38f47570e214 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -92,6 +92,7 @@
 #include <uapi/linux/mount.h>
 #include <linux/fsnotify.h>
 #include <linux/fanotify.h>
+#include <linux/security.h>
 
 #include "avc.h"
 #include "objsec.h"
@@ -6502,6 +6503,11 @@ static void selinux_inode_invalidate_secctx(struct inode *inode)
 	spin_unlock(&isec->lock);
 }
 
+static int selinux_func_query_vfs(unsigned int flags)
+{
+	return !!(flags & LSM_FQUERY_VFS_XATTRS);
+}
+
 /*
  *	called with inode->i_mutex locked
  */
@@ -7085,6 +7091,7 @@ static struct security_hook_list selinux_hooks[] __lsm_ro_after_init = {
 	LSM_HOOK_INIT(inode_invalidate_secctx, selinux_inode_invalidate_secctx),
 	LSM_HOOK_INIT(inode_notifysecctx, selinux_inode_notifysecctx),
 	LSM_HOOK_INIT(inode_setsecctx, selinux_inode_setsecctx),
+	LSM_HOOK_INIT(func_query_vfs, selinux_func_query_vfs),
 
 	LSM_HOOK_INIT(unix_stream_connect, selinux_socket_unix_stream_connect),
 	LSM_HOOK_INIT(unix_may_send, selinux_socket_unix_may_send),
-- 
2.18.2

[PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Olga Kornievskaia]

From: Olga Kornievskaia <kolga@netapp.com>

Currently, the client will always ask for security_labels if the server
returns that it supports that feature regardless of any LSM modules
(such as Selinux) enforcing security policy. This adds performance
penalty to the READDIR operation.

Instead, query the LSM module to find if anything is enabled and
if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.

Suggested-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 +++++
 fs/nfs/nfs4xdr.c        | 3 ++-
 include/linux/nfs_xdr.h | 1 +
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 9e0ca9b2b210..774bc5e63ca7 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -55,6 +55,7 @@
 #include <linux/utsname.h>
 #include <linux/freezer.h>
 #include <linux/iversion.h>
+#include <linux/security.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -4968,6 +4969,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
 		.count = count,
 		.bitmask = NFS_SERVER(d_inode(dentry))->attr_bitmask,
 		.plus = plus,
+		.labels = true,
 	};
 	struct nfs4_readdir_res res;
 	struct rpc_message msg = {
@@ -4977,10 +4979,13 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
 		.rpc_cred = cred,
 	};
 	int			status;
+	int sec_flags = LSM_FQUERY_VFS_XATTRS;
 
 	dprintk("%s: dentry = %pd2, cookie = %Lu\n", __func__,
 			dentry,
 			(unsigned long long)cookie);
+	if (!security_func_query_vfs(sec_flags))
+		args.labels = false;
 	nfs4_setup_readdir(cookie, NFS_I(dir)->cookieverf, dentry, &args);
 	res.pgbase = args.pgbase;
 	status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index c6dbfcae7517..585d5b5cc3dc 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1605,7 +1605,8 @@ static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg
 			FATTR4_WORD1_OWNER_GROUP|FATTR4_WORD1_RAWDEV|
 			FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS|
 			FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
-		attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
+		if (readdir->labels)
+			attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
 		dircount >>= 1;
 	}
 	/* Use mounted_on_fileid only if the server supports it */
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index d63cb862d58e..95f648b26525 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -1119,6 +1119,7 @@ struct nfs4_readdir_arg {
 	unsigned int			pgbase;	/* zero-copy data */
 	const u32 *			bitmask;
 	bool				plus;
+	bool				labels;
 };
 
 struct nfs4_readdir_res {
-- 
2.18.2

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Ondrej Mosnacek]

On Thu, Nov 5, 2020 at 6:33 PM Olga Kornievskaia
<olga.kornievskaia@gmail.com> wrote:
> From: Olga Kornievskaia <kolga@netapp.com>
>
> Currently, the client will always ask for security_labels if the server
> returns that it supports that feature regardless of any LSM modules
> (such as Selinux) enforcing security policy. This adds performance
> penalty to the READDIR operation.
>
> Instead, query the LSM module to find if anything is enabled and
> if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.

Having spent some time staring at some of the NFS code very recently,
I can't help but suggest: Would it perhaps be enough to decide whether
to ask for labels based on (NFS_SB(dentry->d_sb)->caps &
NFS_CAP_SECURITY_LABEL)? It is set when mounting the FS iff some LSM
confirms via the security_sb_*_mnt_opts() hook that it wants the
filesystem to give it labels (or at least that's how I interpret the
cryptic name) [1]. It's just a shot in the dark, but it seems to fit
this use case.

[1] https://elixir.bootlin.com/linux/v5.10-rc2/source/fs/nfs/getroot.c#L148

>
> Suggested-by: Scott Mayhew <smayhew@redhat.com>
> Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
> ---
>  fs/nfs/nfs4proc.c       | 5 +++++
>  fs/nfs/nfs4xdr.c        | 3 ++-
>  include/linux/nfs_xdr.h | 1 +
>  3 files changed, 8 insertions(+), 1 deletion(-)
>
> diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
> index 9e0ca9b2b210..774bc5e63ca7 100644
> --- a/fs/nfs/nfs4proc.c
> +++ b/fs/nfs/nfs4proc.c
> @@ -55,6 +55,7 @@
>  #include <linux/utsname.h>
>  #include <linux/freezer.h>
>  #include <linux/iversion.h>
> +#include <linux/security.h>
>
>  #include "nfs4_fs.h"
>  #include "delegation.h"
> @@ -4968,6 +4969,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
>                 .count = count,
>                 .bitmask = NFS_SERVER(d_inode(dentry))->attr_bitmask,
>                 .plus = plus,
> +               .labels = true,
>         };
>         struct nfs4_readdir_res res;
>         struct rpc_message msg = {
> @@ -4977,10 +4979,13 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
>                 .rpc_cred = cred,
>         };
>         int                     status;
> +       int sec_flags = LSM_FQUERY_VFS_XATTRS;
>
>         dprintk("%s: dentry = %pd2, cookie = %Lu\n", __func__,
>                         dentry,
>                         (unsigned long long)cookie);
> +       if (!security_func_query_vfs(sec_flags))
> +               args.labels = false;
>         nfs4_setup_readdir(cookie, NFS_I(dir)->cookieverf, dentry, &args);
>         res.pgbase = args.pgbase;
>         status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0);
> diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
> index c6dbfcae7517..585d5b5cc3dc 100644
> --- a/fs/nfs/nfs4xdr.c
> +++ b/fs/nfs/nfs4xdr.c
> @@ -1605,7 +1605,8 @@ static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg
>                         FATTR4_WORD1_OWNER_GROUP|FATTR4_WORD1_RAWDEV|
>                         FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS|
>                         FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
> -               attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
> +               if (readdir->labels)
> +                       attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
>                 dircount >>= 1;
>         }
>         /* Use mounted_on_fileid only if the server supports it */
> diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
> index d63cb862d58e..95f648b26525 100644
> --- a/include/linux/nfs_xdr.h
> +++ b/include/linux/nfs_xdr.h
> @@ -1119,6 +1119,7 @@ struct nfs4_readdir_arg {
>         unsigned int                    pgbase; /* zero-copy data */
>         const u32 *                     bitmask;
>         bool                            plus;
> +       bool                            labels;
>  };
>
>  struct nfs4_readdir_res {
> --
> 2.18.2
>


-- 
Ondrej Mosnacek
Software Engineer, Platform Security - SELinux kernel
Red Hat, Inc.

Re: [PATCH 1/2] [lsm] introduce a new hook to query LSM for functionality

[Casey Schaufler]

On 11/5/2020 9:33 AM, Olga Kornievskaia wrote:
> From: Olga Kornievskaia <kolga@netapp.com>
>
> Add a new hook func_query_vfs to query an LSM module (such as
> SELinux) with the intention of finding whether or not it is enabled
> or perhaps supports some functionality.
>
> NFS stores security labels for file system objects and SElinux
> or any other LSM module can query those objects. But it's
> wasteful to do so when no security enforcement is taking place.
> Using a new API call of security_func_query_vfs() and asking if

As I mentioned earlier, this is a sub-optimal implementation.
Instead of adding a func_query_vfs() hook add a field to
security_hook_list and have the LSM declare the "features"
it supports there. security_add_hooks() can gather the "features"
for the whole system into a single mask. security_func_query_vfs()
can then just return that mask.

> Suggested-by: Paul Moore <paul@paul-moore.com>
> Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
> ---
>  include/linux/lsm_hook_defs.h | 1 +
>  include/linux/security.h      | 4 ++++
>  security/security.c           | 6 ++++++
>  security/selinux/hooks.c      | 7 +++++++
>  4 files changed, 18 insertions(+)
>
> diff --git a/include/linux/lsm_hook_defs.h b/include/linux/lsm_hook_defs.h
> index 32a940117e7a..df3454a1fcac 100644
> --- a/include/linux/lsm_hook_defs.h
> +++ b/include/linux/lsm_hook_defs.h
> @@ -257,6 +257,7 @@ LSM_HOOK(int, 0, inode_notifysecctx, struct inode *inode, void *ctx, u32 ctxlen)
>  LSM_HOOK(int, 0, inode_setsecctx, struct dentry *dentry, void *ctx, u32 ctxlen)
>  LSM_HOOK(int, 0, inode_getsecctx, struct inode *inode, void **ctx,
>  	 u32 *ctxlen)
> +LSM_HOOK(int, 0, func_query_vfs, unsigned int)
>  
>  #if defined(CONFIG_SECURITY) && defined(CONFIG_WATCH_QUEUE)
>  LSM_HOOK(int, 0, post_notification, const struct cred *w_cred,
> diff --git a/include/linux/security.h b/include/linux/security.h
> index bc2725491560..e15964059fa4 100644
> --- a/include/linux/security.h
> +++ b/include/linux/security.h
> @@ -456,6 +456,10 @@ int security_inode_notifysecctx(struct inode *inode, void *ctx, u32 ctxlen);
>  int security_inode_setsecctx(struct dentry *dentry, void *ctx, u32 ctxlen);
>  int security_inode_getsecctx(struct inode *inode, void **ctx, u32 *ctxlen);
>  int security_locked_down(enum lockdown_reason what);
> +#define LSM_FQUERY_VFS_NONE     0x00000000
> +#define LSM_FQUERY_VFS_XATTRS   0x00000001

This is the wrong granularity for the task at hand.
You don't care if an LSM supports vfs xattrs. You care if
the LSM provides an xattr that may be propagated by NFS
for the purpose of mandatory access control. An LSM that
provides time-of-day controls may use xattrs that are
meaningless to NFS.


> +int security_func_query_vfs(unsigned int flags);
> +
>  #else /* CONFIG_SECURITY */
>  
>  static inline int call_blocking_lsm_notifier(enum lsm_event event, void *data)
> diff --git a/security/security.c b/security/security.c
> index a28045dc9e7f..502b33865238 100644
> --- a/security/security.c
> +++ b/security/security.c
> @@ -2067,6 +2067,12 @@ int security_inode_getsecctx(struct inode *inode, void **ctx, u32 *ctxlen)
>  }
>  EXPORT_SYMBOL(security_inode_getsecctx);
>  
> +int security_func_query_vfs(unsigned int flags)
> +{
> +	return call_int_hook(func_query_vfs, 0, flags);
> +}
> +EXPORT_SYMBOL(security_func_query_vfs);
> +
>  #ifdef CONFIG_WATCH_QUEUE
>  int security_post_notification(const struct cred *w_cred,
>  			       const struct cred *cred,
> diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
> index 6b1826fc3658..38f47570e214 100644
> --- a/security/selinux/hooks.c
> +++ b/security/selinux/hooks.c
> @@ -92,6 +92,7 @@
>  #include <uapi/linux/mount.h>
>  #include <linux/fsnotify.h>
>  #include <linux/fanotify.h>
> +#include <linux/security.h>
>  
>  #include "avc.h"
>  #include "objsec.h"
> @@ -6502,6 +6503,11 @@ static void selinux_inode_invalidate_secctx(struct inode *inode)
>  	spin_unlock(&isec->lock);
>  }
>  
> +static int selinux_func_query_vfs(unsigned int flags)
> +{
> +	return !!(flags & LSM_FQUERY_VFS_XATTRS);
> +}
> +
>  /*
>   *	called with inode->i_mutex locked
>   */
> @@ -7085,6 +7091,7 @@ static struct security_hook_list selinux_hooks[] __lsm_ro_after_init = {
>  	LSM_HOOK_INIT(inode_invalidate_secctx, selinux_inode_invalidate_secctx),
>  	LSM_HOOK_INIT(inode_notifysecctx, selinux_inode_notifysecctx),
>  	LSM_HOOK_INIT(inode_setsecctx, selinux_inode_setsecctx),
> +	LSM_HOOK_INIT(func_query_vfs, selinux_func_query_vfs),
>  
>  	LSM_HOOK_INIT(unix_stream_connect, selinux_socket_unix_stream_connect),
>  	LSM_HOOK_INIT(unix_may_send, selinux_socket_unix_may_send),

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Olga Kornievskaia]

On Thu, Nov 5, 2020 at 1:55 PM Ondrej Mosnacek <omosnace@redhat.com> wrote:
>
> On Thu, Nov 5, 2020 at 6:33 PM Olga Kornievskaia
> <olga.kornievskaia@gmail.com> wrote:
> > From: Olga Kornievskaia <kolga@netapp.com>
> >
> > Currently, the client will always ask for security_labels if the server
> > returns that it supports that feature regardless of any LSM modules
> > (such as Selinux) enforcing security policy. This adds performance
> > penalty to the READDIR operation.
> >
> > Instead, query the LSM module to find if anything is enabled and
> > if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.
>
> Having spent some time staring at some of the NFS code very recently,
> I can't help but suggest: Would it perhaps be enough to decide whether
> to ask for labels based on (NFS_SB(dentry->d_sb)->caps &
> NFS_CAP_SECURITY_LABEL)? It is set when mounting the FS iff some LSM
> confirms via the security_sb_*_mnt_opts() hook that it wants the
> filesystem to give it labels (or at least that's how I interpret the
> cryptic name) [1]. It's just a shot in the dark, but it seems to fit
> this use case.
>
> [1] https://elixir.bootlin.com/linux/v5.10-rc2/source/fs/nfs/getroot.c#L148

Very interesting. I was not aware of something like that nor was it
mentioned when I asked on the selinux mailing list. I wonder if this
is a supported feature that will always stay? In that case, NFS
wouldn't need the extra hook that was added for this series. I will
try this out and report back.

>
> >
> > Suggested-by: Scott Mayhew <smayhew@redhat.com>
> > Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
> > ---
> >  fs/nfs/nfs4proc.c       | 5 +++++
> >  fs/nfs/nfs4xdr.c        | 3 ++-
> >  include/linux/nfs_xdr.h | 1 +
> >  3 files changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
> > index 9e0ca9b2b210..774bc5e63ca7 100644
> > --- a/fs/nfs/nfs4proc.c
> > +++ b/fs/nfs/nfs4proc.c
> > @@ -55,6 +55,7 @@
> >  #include <linux/utsname.h>
> >  #include <linux/freezer.h>
> >  #include <linux/iversion.h>
> > +#include <linux/security.h>
> >
> >  #include "nfs4_fs.h"
> >  #include "delegation.h"
> > @@ -4968,6 +4969,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
> >                 .count = count,
> >                 .bitmask = NFS_SERVER(d_inode(dentry))->attr_bitmask,
> >                 .plus = plus,
> > +               .labels = true,
> >         };
> >         struct nfs4_readdir_res res;
> >         struct rpc_message msg = {
> > @@ -4977,10 +4979,13 @@ static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
> >                 .rpc_cred = cred,
> >         };
> >         int                     status;
> > +       int sec_flags = LSM_FQUERY_VFS_XATTRS;
> >
> >         dprintk("%s: dentry = %pd2, cookie = %Lu\n", __func__,
> >                         dentry,
> >                         (unsigned long long)cookie);
> > +       if (!security_func_query_vfs(sec_flags))
> > +               args.labels = false;
> >         nfs4_setup_readdir(cookie, NFS_I(dir)->cookieverf, dentry, &args);
> >         res.pgbase = args.pgbase;
> >         status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0);
> > diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
> > index c6dbfcae7517..585d5b5cc3dc 100644
> > --- a/fs/nfs/nfs4xdr.c
> > +++ b/fs/nfs/nfs4xdr.c
> > @@ -1605,7 +1605,8 @@ static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg
> >                         FATTR4_WORD1_OWNER_GROUP|FATTR4_WORD1_RAWDEV|
> >                         FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS|
> >                         FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
> > -               attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
> > +               if (readdir->labels)
> > +                       attrs[2] |= FATTR4_WORD2_SECURITY_LABEL;
> >                 dircount >>= 1;
> >         }
> >         /* Use mounted_on_fileid only if the server supports it */
> > diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
> > index d63cb862d58e..95f648b26525 100644
> > --- a/include/linux/nfs_xdr.h
> > +++ b/include/linux/nfs_xdr.h
> > @@ -1119,6 +1119,7 @@ struct nfs4_readdir_arg {
> >         unsigned int                    pgbase; /* zero-copy data */
> >         const u32 *                     bitmask;
> >         bool                            plus;
> > +       bool                            labels;
> >  };
> >
> >  struct nfs4_readdir_res {
> > --
> > 2.18.2
> >
>
>
> --
> Ondrej Mosnacek
> Software Engineer, Platform Security - SELinux kernel
> Red Hat, Inc.
>

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Ondrej Mosnacek]

On Thu, Nov 5, 2020 at 8:51 PM Olga Kornievskaia
<olga.kornievskaia@gmail.com> wrote:
>
> On Thu, Nov 5, 2020 at 1:55 PM Ondrej Mosnacek <omosnace@redhat.com> wrote:
> >
> > On Thu, Nov 5, 2020 at 6:33 PM Olga Kornievskaia
> > <olga.kornievskaia@gmail.com> wrote:
> > > From: Olga Kornievskaia <kolga@netapp.com>
> > >
> > > Currently, the client will always ask for security_labels if the server
> > > returns that it supports that feature regardless of any LSM modules
> > > (such as Selinux) enforcing security policy. This adds performance
> > > penalty to the READDIR operation.
> > >
> > > Instead, query the LSM module to find if anything is enabled and
> > > if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.
> >
> > Having spent some time staring at some of the NFS code very recently,
> > I can't help but suggest: Would it perhaps be enough to decide whether
> > to ask for labels based on (NFS_SB(dentry->d_sb)->caps &
> > NFS_CAP_SECURITY_LABEL)? It is set when mounting the FS iff some LSM
> > confirms via the security_sb_*_mnt_opts() hook that it wants the
> > filesystem to give it labels (or at least that's how I interpret the
> > cryptic name) [1]. It's just a shot in the dark, but it seems to fit
> > this use case.
> >
> > [1] https://elixir.bootlin.com/linux/v5.10-rc2/source/fs/nfs/getroot.c#L148
>
> Very interesting. I was not aware of something like that nor was it
> mentioned when I asked on the selinux mailing list. I wonder if this
> is a supported feature that will always stay? In that case, NFS
> wouldn't need the extra hook that was added for this series. I will
> try this out and report back.

I wish I could have suggested it sooner, but I only learned about it
in recent days while investigating a bug that involves SELinux and NFS
:) And when I saw your patch, it immediately reminded me of that flag.

I don't think it's going away, at least as long as NFS depends on it.
If someone were to remove it, they would have to provide something
equivalent to make sure no existing users get broken, as with any
in-kernel change.

-- 
Ondrej Mosnacek
Software Engineer, Platform Security - SELinux kernel
Red Hat, Inc.

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Trond Myklebust]

On Thu, 2020-11-05 at 14:51 -0500, Olga Kornievskaia wrote:
> On Thu, Nov 5, 2020 at 1:55 PM Ondrej Mosnacek <omosnace@redhat.com>
> wrote:
> > 
> > On Thu, Nov 5, 2020 at 6:33 PM Olga Kornievskaia
> > <olga.kornievskaia@gmail.com> wrote:
> > > From: Olga Kornievskaia <kolga@netapp.com>
> > > 
> > > Currently, the client will always ask for security_labels if the
> > > server
> > > returns that it supports that feature regardless of any LSM
> > > modules
> > > (such as Selinux) enforcing security policy. This adds
> > > performance
> > > penalty to the READDIR operation.
> > > 
> > > Instead, query the LSM module to find if anything is enabled and
> > > if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.
> > 
> > Having spent some time staring at some of the NFS code very
> > recently,
> > I can't help but suggest: Would it perhaps be enough to decide
> > whether
> > to ask for labels based on (NFS_SB(dentry->d_sb)->caps &
> > NFS_CAP_SECURITY_LABEL)? It is set when mounting the FS iff some
> > LSM
> > confirms via the security_sb_*_mnt_opts() hook that it wants the
> > filesystem to give it labels (or at least that's how I interpret
> > the
> > cryptic name) [1]. It's just a shot in the dark, but it seems to
> > fit
> > this use case.
> > 
> > [1]
> > https://elixir.bootlin.com/linux/v5.10-rc2/source/fs/nfs/getroot.c#L148
> 
> Very interesting. I was not aware of something like that nor was it
> mentioned when I asked on the selinux mailing list. I wonder if this
> is a supported feature that will always stay? In that case, NFS
> wouldn't need the extra hook that was added for this series. I will
> try this out and report back.

NFS_CAP_SECURITY_LABEL is just the NFS server capability flag. It tells
you whether or not the client believes that the server might support
NFSv4.2 requests for labelled NFS metadata.

My understanding of Olga's requirement is that she needs to be able to
ignore that flag and simply not query for labelled NFS metadata if the
NFS client is not configured to enforce the LSM policy. The objective
is to avoid unnecessary RPC traffic to the server to query for data
that is unused.

-- 
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@hammerspace.com

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Olga Kornievskaia]

On Thu, Nov 5, 2020 at 4:18 PM Trond Myklebust <trondmy@hammerspace.com> wrote:
>
> On Thu, 2020-11-05 at 14:51 -0500, Olga Kornievskaia wrote:
> > On Thu, Nov 5, 2020 at 1:55 PM Ondrej Mosnacek <omosnace@redhat.com>
> > wrote:
> > >
> > > On Thu, Nov 5, 2020 at 6:33 PM Olga Kornievskaia
> > > <olga.kornievskaia@gmail.com> wrote:
> > > > From: Olga Kornievskaia <kolga@netapp.com>
> > > >
> > > > Currently, the client will always ask for security_labels if the
> > > > server
> > > > returns that it supports that feature regardless of any LSM
> > > > modules
> > > > (such as Selinux) enforcing security policy. This adds
> > > > performance
> > > > penalty to the READDIR operation.
> > > >
> > > > Instead, query the LSM module to find if anything is enabled and
> > > > if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.
> > >
> > > Having spent some time staring at some of the NFS code very
> > > recently,
> > > I can't help but suggest: Would it perhaps be enough to decide
> > > whether
> > > to ask for labels based on (NFS_SB(dentry->d_sb)->caps &
> > > NFS_CAP_SECURITY_LABEL)? It is set when mounting the FS iff some
> > > LSM
> > > confirms via the security_sb_*_mnt_opts() hook that it wants the
> > > filesystem to give it labels (or at least that's how I interpret
> > > the
> > > cryptic name) [1]. It's just a shot in the dark, but it seems to
> > > fit
> > > this use case.
> > >
> > > [1]
> > > https://elixir.bootlin.com/linux/v5.10-rc2/source/fs/nfs/getroot.c#L148
> >
> > Very interesting. I was not aware of something like that nor was it
> > mentioned when I asked on the selinux mailing list. I wonder if this
> > is a supported feature that will always stay? In that case, NFS
> > wouldn't need the extra hook that was added for this series. I will
> > try this out and report back.
>
> NFS_CAP_SECURITY_LABEL is just the NFS server capability flag. It tells
> you whether or not the client believes that the server might support
> NFSv4.2 requests for labelled NFS metadata.
>
> My understanding of Olga's requirement is that she needs to be able to
> ignore that flag and simply not query for labelled NFS metadata if the
> NFS client is not configured to enforce the LSM policy. The objective
> is to avoid unnecessary RPC traffic to the server to query for data
> that is unused.

Actually that seems to be working. I think it's because while the
server returns that it supports sec_labels, after calling
security_sb_set_mnt_opts() the kflags_out don't have this
SECURITY_LSM_NATIVE_LABELS set (assuming this happens because selinux
isn't enabled) then we turned server's sec_labl cap off.

I'm about to send v2 without relying on modifications to selinux.

>
> --
> Trond Myklebust
> Linux NFS client maintainer, Hammerspace
> trond.myklebust@hammerspace.com
>
>

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 3845 bytes --]

Hi Olga,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on nfs/linux-next]
[also build test ERROR on pcmoore-selinux/next linus/master security/next-testing v5.10-rc2 next-20201105]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Olga-Kornievskaia/introduce-a-new-hook-to-query-LSM-for-functionality/20201106-013417
base:   git://git.linux-nfs.org/projects/trondmy/linux-nfs.git linux-next
config: openrisc-randconfig-r002-20201104 (attached as .config)
compiler: or1k-linux-gcc (GCC) 9.3.0
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # https://github.com/0day-ci/linux/commit/d765c00ede01a334b7a3f995ab27b8d4ebd5ea38
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Olga-Kornievskaia/introduce-a-new-hook-to-query-LSM-for-functionality/20201106-013417
        git checkout d765c00ede01a334b7a3f995ab27b8d4ebd5ea38
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross ARCH=openrisc 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

   fs/nfs/nfs4proc.c: In function '_nfs4_proc_readdir':
>> fs/nfs/nfs4proc.c:4982:18: error: 'LSM_FQUERY_VFS_XATTRS' undeclared (first use in this function)
    4982 |  int sec_flags = LSM_FQUERY_VFS_XATTRS;
         |                  ^~~~~~~~~~~~~~~~~~~~~
   fs/nfs/nfs4proc.c:4982:18: note: each undeclared identifier is reported only once for each function it appears in
>> fs/nfs/nfs4proc.c:4987:7: error: implicit declaration of function 'security_func_query_vfs' [-Werror=implicit-function-declaration]
    4987 |  if (!security_func_query_vfs(sec_flags))
         |       ^~~~~~~~~~~~~~~~~~~~~~~
   cc1: some warnings being treated as errors

vim +/LSM_FQUERY_VFS_XATTRS +4982 fs/nfs/nfs4proc.c

  4960	
  4961	static int _nfs4_proc_readdir(struct dentry *dentry, const struct cred *cred,
  4962			u64 cookie, struct page **pages, unsigned int count, bool plus)
  4963	{
  4964		struct inode		*dir = d_inode(dentry);
  4965		struct nfs4_readdir_arg args = {
  4966			.fh = NFS_FH(dir),
  4967			.pages = pages,
  4968			.pgbase = 0,
  4969			.count = count,
  4970			.bitmask = NFS_SERVER(d_inode(dentry))->attr_bitmask,
  4971			.plus = plus,
  4972			.labels = true,
  4973		};
  4974		struct nfs4_readdir_res res;
  4975		struct rpc_message msg = {
  4976			.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READDIR],
  4977			.rpc_argp = &args,
  4978			.rpc_resp = &res,
  4979			.rpc_cred = cred,
  4980		};
  4981		int			status;
> 4982		int sec_flags = LSM_FQUERY_VFS_XATTRS;
  4983	
  4984		dprintk("%s: dentry = %pd2, cookie = %Lu\n", __func__,
  4985				dentry,
  4986				(unsigned long long)cookie);
> 4987		if (!security_func_query_vfs(sec_flags))
  4988			args.labels = false;
  4989		nfs4_setup_readdir(cookie, NFS_I(dir)->cookieverf, dentry, &args);
  4990		res.pgbase = args.pgbase;
  4991		status = nfs4_call_sync(NFS_SERVER(dir)->client, NFS_SERVER(dir), &msg, &args.seq_args, &res.seq_res, 0);
  4992		if (status >= 0) {
  4993			memcpy(NFS_I(dir)->cookieverf, res.verifier.data, NFS4_VERIFIER_SIZE);
  4994			status += args.pgbase;
  4995		}
  4996	
  4997		nfs_invalidate_atime(dir);
  4998	
  4999		dprintk("%s: returns %d\n", __func__, status);
  5000		return status;
  5001	}
  5002	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 31537 bytes --]

Re: [PATCH 2/2] NFSv4.2: condition READDIR's mask for security label based on LSM state

[Ondrej Mosnacek]

On Thu, Nov 5, 2020 at 10:43 PM Olga Kornievskaia
<olga.kornievskaia@gmail.com> wrote:
> On Thu, Nov 5, 2020 at 4:18 PM Trond Myklebust <trondmy@hammerspace.com> wrote:
> >
> > On Thu, 2020-11-05 at 14:51 -0500, Olga Kornievskaia wrote:
> > > On Thu, Nov 5, 2020 at 1:55 PM Ondrej Mosnacek <omosnace@redhat.com>
> > > wrote:
> > > >
> > > > On Thu, Nov 5, 2020 at 6:33 PM Olga Kornievskaia
> > > > <olga.kornievskaia@gmail.com> wrote:
> > > > > From: Olga Kornievskaia <kolga@netapp.com>
> > > > >
> > > > > Currently, the client will always ask for security_labels if the
> > > > > server
> > > > > returns that it supports that feature regardless of any LSM
> > > > > modules
> > > > > (such as Selinux) enforcing security policy. This adds
> > > > > performance
> > > > > penalty to the READDIR operation.
> > > > >
> > > > > Instead, query the LSM module to find if anything is enabled and
> > > > > if not, then remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.
> > > >
> > > > Having spent some time staring at some of the NFS code very
> > > > recently,
> > > > I can't help but suggest: Would it perhaps be enough to decide
> > > > whether
> > > > to ask for labels based on (NFS_SB(dentry->d_sb)->caps &
> > > > NFS_CAP_SECURITY_LABEL)? It is set when mounting the FS iff some
> > > > LSM
> > > > confirms via the security_sb_*_mnt_opts() hook that it wants the
> > > > filesystem to give it labels (or at least that's how I interpret
> > > > the
> > > > cryptic name) [1]. It's just a shot in the dark, but it seems to
> > > > fit
> > > > this use case.
> > > >
> > > > [1]
> > > > https://elixir.bootlin.com/linux/v5.10-rc2/source/fs/nfs/getroot.c#L148
> > >
> > > Very interesting. I was not aware of something like that nor was it
> > > mentioned when I asked on the selinux mailing list. I wonder if this
> > > is a supported feature that will always stay? In that case, NFS
> > > wouldn't need the extra hook that was added for this series. I will
> > > try this out and report back.
> >
> > NFS_CAP_SECURITY_LABEL is just the NFS server capability flag. It tells
> > you whether or not the client believes that the server might support
> > NFSv4.2 requests for labelled NFS metadata.
> >
> > My understanding of Olga's requirement is that she needs to be able to
> > ignore that flag and simply not query for labelled NFS metadata if the
> > NFS client is not configured to enforce the LSM policy. The objective
> > is to avoid unnecessary RPC traffic to the server to query for data
> > that is unused.
>
> Actually that seems to be working. I think it's because while the
> server returns that it supports sec_labels, after calling
> security_sb_set_mnt_opts() the kflags_out don't have this
> SECURITY_LSM_NATIVE_LABELS set (assuming this happens because selinux
> isn't enabled) then we turned server's sec_labl cap off.
>
> I'm about to send v2 without relying on modifications to selinux.

Just to keep the LSM and SELinux lists in the loop: a v2 has been
posted to linux-nfs that uses NFS_CAP_SECURITY_LABEL instead of adding
a new hook:
https://lore.kernel.org/linux-nfs/CAFqZXNtJ2fkae7Xt-+nDR79kVs=yY=9vSoZaS-V-5UKSk22s4A@mail.gmail.com/T/

-- 
Ondrej Mosnacek
Software Engineer, Platform Security - SELinux kernel
Red Hat, Inc.

Re: [PATCH 1/2] [lsm] introduce a new hook to query LSM for functionality

[James Morris]

On Thu, 5 Nov 2020, Olga Kornievskaia wrote:

> From: Olga Kornievskaia <kolga@netapp.com>
> 
> Add a new hook func_query_vfs to query an LSM module (such as
> SELinux) with the intention of finding whether or not it is enabled
> or perhaps supports some functionality.
> 
> NFS stores security labels for file system objects and SElinux
> or any other LSM module can query those objects. But it's
> wasteful to do so when no security enforcement is taking place.
> Using a new API call of security_func_query_vfs() and asking if

Seems we lost something here.



-- 
James Morris
<jmorris@namei.org>