[PATCH net-next 0/4] net: Mark the sk parameter of routing functions as 'const'.

[PATCH net-next 0/4] net: Mark the sk parameter of routing functions as 'const'.

[Guillaume Nault]

The sk_getsecid security hook prevents the use of a const sk pointer in
several routing functions. Since this hook should only read sk data,
make its sk argument const (patch 1), then constify the sk parameter of
various routing functions (patches 2-4).

Build-tested with make allmodconfig.

Guillaume Nault (4):
  security: Constify sk in the sk_getsecid hook.
  ipv4: Constify the sk parameter of ip_route_output_*().
  ipv6: Constify the sk parameter of several helper functions.
  pptp: Constify the po parameter of pptp_route_output().

 drivers/net/ppp/pptp.c        |  4 ++--
 include/linux/icmpv6.h        | 10 ++++------
 include/linux/lsm_hook_defs.h |  2 +-
 include/linux/security.h      |  5 +++--
 include/net/route.h           |  6 +++---
 net/ipv6/datagram.c           |  7 ++++---
 net/ipv6/icmp.c               |  6 ++----
 net/ipv6/mcast.c              |  8 +++-----
 security/security.c           |  2 +-
 security/selinux/hooks.c      |  4 ++--
 10 files changed, 25 insertions(+), 29 deletions(-)

-- 
2.39.2

[PATCH net-next 1/4] security: Constify sk in the sk_getsecid hook.

[Guillaume Nault]

The sk_getsecid hook shouldn't need to modify its socket argument.
Make it const so that callers of security_sk_classify_flow() can use a
const struct sock *.

Signed-off-by: Guillaume Nault <gnault@redhat.com>
---
 include/linux/lsm_hook_defs.h | 2 +-
 include/linux/security.h      | 5 +++--
 security/security.c           | 2 +-
 security/selinux/hooks.c      | 4 ++--
 4 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/include/linux/lsm_hook_defs.h b/include/linux/lsm_hook_defs.h
index 7308a1a7599b..4f2621e87634 100644
--- a/include/linux/lsm_hook_defs.h
+++ b/include/linux/lsm_hook_defs.h
@@ -316,7 +316,7 @@ LSM_HOOK(int, 0, sk_alloc_security, struct sock *sk, int family, gfp_t priority)
 LSM_HOOK(void, LSM_RET_VOID, sk_free_security, struct sock *sk)
 LSM_HOOK(void, LSM_RET_VOID, sk_clone_security, const struct sock *sk,
 	 struct sock *newsk)
-LSM_HOOK(void, LSM_RET_VOID, sk_getsecid, struct sock *sk, u32 *secid)
+LSM_HOOK(void, LSM_RET_VOID, sk_getsecid, const struct sock *sk, u32 *secid)
 LSM_HOOK(void, LSM_RET_VOID, sock_graft, struct sock *sk, struct socket *parent)
 LSM_HOOK(int, 0, inet_conn_request, const struct sock *sk, struct sk_buff *skb,
 	 struct request_sock *req)
diff --git a/include/linux/security.h b/include/linux/security.h
index 32828502f09e..994cf099d9ac 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1439,7 +1439,8 @@ int security_socket_getpeersec_dgram(struct socket *sock, struct sk_buff *skb, u
 int security_sk_alloc(struct sock *sk, int family, gfp_t priority);
 void security_sk_free(struct sock *sk);
 void security_sk_clone(const struct sock *sk, struct sock *newsk);
-void security_sk_classify_flow(struct sock *sk, struct flowi_common *flic);
+void security_sk_classify_flow(const struct sock *sk,
+			       struct flowi_common *flic);
 void security_req_classify_flow(const struct request_sock *req,
 				struct flowi_common *flic);
 void security_sock_graft(struct sock*sk, struct socket *parent);
@@ -1597,7 +1598,7 @@ static inline void security_sk_clone(const struct sock *sk, struct sock *newsk)
 {
 }
 
-static inline void security_sk_classify_flow(struct sock *sk,
+static inline void security_sk_classify_flow(const struct sock *sk,
 					     struct flowi_common *flic)
 {
 }
diff --git a/security/security.c b/security/security.c
index b720424ca37d..2dfc7b9f6ed9 100644
--- a/security/security.c
+++ b/security/security.c
@@ -4396,7 +4396,7 @@ void security_sk_clone(const struct sock *sk, struct sock *newsk)
 }
 EXPORT_SYMBOL(security_sk_clone);
 
-void security_sk_classify_flow(struct sock *sk, struct flowi_common *flic)
+void security_sk_classify_flow(const struct sock *sk, struct flowi_common *flic)
 {
 	call_void_hook(sk_getsecid, sk, &flic->flowic_secid);
 }
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index d06e350fedee..2bdc48dd8670 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -5167,12 +5167,12 @@ static void selinux_sk_clone_security(const struct sock *sk, struct sock *newsk)
 	selinux_netlbl_sk_security_reset(newsksec);
 }
 
-static void selinux_sk_getsecid(struct sock *sk, u32 *secid)
+static void selinux_sk_getsecid(const struct sock *sk, u32 *secid)
 {
 	if (!sk)
 		*secid = SECINITSID_ANY_SOCKET;
 	else {
-		struct sk_security_struct *sksec = sk->sk_security;
+		const struct sk_security_struct *sksec = sk->sk_security;
 
 		*secid = sksec->sid;
 	}
-- 
2.39.2

Re: [PATCH net-next 1/4] security: Constify sk in the sk_getsecid hook.

[Simon Horman]

On Tue, Jul 11, 2023 at 03:06:08PM +0200, Guillaume Nault wrote:
> The sk_getsecid hook shouldn't need to modify its socket argument.
> Make it const so that callers of security_sk_classify_flow() can use a
> const struct sock *.
> 
> Signed-off-by: Guillaume Nault <gnault@redhat.com>

Reviewed-by: Simon Horman <simon.horman@corigine.com>

Re: [PATCH net-next 0/4] net: Mark the sk parameter of routing functions as 'const'.

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by David S. Miller <davem@davemloft.net>:

On Tue, 11 Jul 2023 15:06:00 +0200 you wrote:
> The sk_getsecid security hook prevents the use of a const sk pointer in
> several routing functions. Since this hook should only read sk data,
> make its sk argument const (patch 1), then constify the sk parameter of
> various routing functions (patches 2-4).
> 
> Build-tested with make allmodconfig.
> 
> [...]

Here is the summary with links:
  - [net-next,1/4] security: Constify sk in the sk_getsecid hook.
    https://git.kernel.org/netdev/net-next/c/5b52ad34f948
  - [net-next,2/4] ipv4: Constify the sk parameter of ip_route_output_*().
    https://git.kernel.org/netdev/net-next/c/8d6eba33a272
  - [net-next,3/4] ipv6: Constify the sk parameter of several helper functions.
    https://git.kernel.org/netdev/net-next/c/5bc67a854cb4
  - [net-next,4/4] pptp: Constify the po parameter of pptp_route_output().
    https://git.kernel.org/netdev/net-next/c/dc4c399d215d

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH net-next 1/4] security: Constify sk in the sk_getsecid hook.

[Paul Moore]

On Tue, Jul 11, 2023 at 9:06 AM Guillaume Nault <gnault@redhat.com> wrote:
>
> The sk_getsecid hook shouldn't need to modify its socket argument.
> Make it const so that callers of security_sk_classify_flow() can use a
> const struct sock *.
>
> Signed-off-by: Guillaume Nault <gnault@redhat.com>
> ---
>  include/linux/lsm_hook_defs.h | 2 +-
>  include/linux/security.h      | 5 +++--
>  security/security.c           | 2 +-
>  security/selinux/hooks.c      | 4 ++--
>  4 files changed, 7 insertions(+), 6 deletions(-)

Thanks Guillaume, this looks good to me.  I had limited network access
last week and was only monitoring my email for urgent issues, but from
what I can tell it looks like this was picked up in the netdev tree so
I'll leave it alone, but if anything changes let me know and I'll
merge it via the LSM tree.

-- 
paul-moore.com

Re: [PATCH net-next 1/4] security: Constify sk in the sk_getsecid hook.

[Guillaume Nault]

On Mon, Jul 17, 2023 at 12:05:44PM -0400, Paul Moore wrote:
> On Tue, Jul 11, 2023 at 9:06 AM Guillaume Nault <gnault@redhat.com> wrote:
> >
> > The sk_getsecid hook shouldn't need to modify its socket argument.
> > Make it const so that callers of security_sk_classify_flow() can use a
> > const struct sock *.
> >
> > Signed-off-by: Guillaume Nault <gnault@redhat.com>
> > ---
> >  include/linux/lsm_hook_defs.h | 2 +-
> >  include/linux/security.h      | 5 +++--
> >  security/security.c           | 2 +-
> >  security/selinux/hooks.c      | 4 ++--
> >  4 files changed, 7 insertions(+), 6 deletions(-)
> 
> Thanks Guillaume, this looks good to me.  I had limited network access
> last week and was only monitoring my email for urgent issues, but from
> what I can tell it looks like this was picked up in the netdev tree so
> I'll leave it alone, but if anything changes let me know and I'll
> merge it via the LSM tree.

Thanks Paul, this series has indeed been applied to the networking tree.
So no special action is needed.

> -- 
> paul-moore.com
>