[PATCH net v4 0/3] vsock: Fix transport_{h2g,g2h,dgram,local} TOCTOU issues

[PATCH net v4 0/3] vsock: Fix transport_{h2g,g2h,dgram,local} TOCTOU issues

[Michal Luczaj]

transport_{h2g,g2h,dgram,local} may become NULL on vsock_core_unregister().
Make sure a poorly timed `rmmod transport` won't lead to a NULL/stale
pointer dereference.

Note that these oopses are pretty unlikely to happen in the wild. Splats
were collected after sprinkling kernel with mdelay()s.

Signed-off-by: Michal Luczaj <mhal@rbox.co>
---
Changes in v4:
- Fix a typo in a comment [Stefano]
- Link to v3: https://lore.kernel.org/r/20250702-vsock-transports-toctou-v3-0-0a7e2e692987@rbox.co

Changes in v3:
- Static transport_* CID getter rename and comment [Stefano]
- Link to v2: https://lore.kernel.org/r/20250620-vsock-transports-toctou-v2-0-02ebd20b1d03@rbox.co

Changes in v2:
- Introduce a helper function to get local CIDs safely [Stefano]
- Rename goto label to indicate an error path, explain why releasing
  vsock_register_mutex after try_module_get() is safe [Stefano]
- Link to v1: https://lore.kernel.org/r/20250618-vsock-transports-toctou-v1-0-dd2d2ede9052@rbox.co

---
Michal Luczaj (3):
      vsock: Fix transport_{g2h,h2g} TOCTOU
      vsock: Fix transport_* TOCTOU
      vsock: Fix IOCTL_VM_SOCKETS_GET_LOCAL_CID to check also `transport_local`

 net/vmw_vsock/af_vsock.c | 57 ++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 46 insertions(+), 11 deletions(-)
---
base-commit: 223e2288f4b8c262a864e2c03964ffac91744cd5
change-id: 20250523-vsock-transports-toctou-4b75d9c2a805

Best regards,
-- 
Michal Luczaj <mhal@rbox.co>

[PATCH net v4 1/3] vsock: Fix transport_{g2h,h2g} TOCTOU

[Michal Luczaj]

vsock_find_cid() and vsock_dev_do_ioctl() may race with module unload.
transport_{g2h,h2g} may become NULL after the NULL check.

Introduce vsock_transport_local_cid() to protect from a potential
null-ptr-deref.

KASAN: null-ptr-deref in range [0x0000000000000118-0x000000000000011f]
RIP: 0010:vsock_find_cid+0x47/0x90
Call Trace:
 __vsock_bind+0x4b2/0x720
 vsock_bind+0x90/0xe0
 __sys_bind+0x14d/0x1e0
 __x64_sys_bind+0x6e/0xc0
 do_syscall_64+0x92/0x1c0
 entry_SYSCALL_64_after_hwframe+0x4b/0x53

KASAN: null-ptr-deref in range [0x0000000000000118-0x000000000000011f]
RIP: 0010:vsock_dev_do_ioctl.isra.0+0x58/0xf0
Call Trace:
 __x64_sys_ioctl+0x12d/0x190
 do_syscall_64+0x92/0x1c0
 entry_SYSCALL_64_after_hwframe+0x4b/0x53

Fixes: c0cfa2d8a788 ("vsock: add multi-transports support")
Suggested-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Michal Luczaj <mhal@rbox.co>
---
 net/vmw_vsock/af_vsock.c | 27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 2e7a3034e965db30b6ee295370d866e6d8b1c341..39473b9e0829f240045262aef00cbae82a425dcc 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -531,9 +531,25 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 }
 EXPORT_SYMBOL_GPL(vsock_assign_transport);
 
+/*
+ * Provide safe access to static transport_{h2g,g2h,dgram,local} callbacks.
+ * Otherwise we may race with module removal. Do not use on `vsk->transport`.
+ */
+static u32 vsock_registered_transport_cid(const struct vsock_transport **transport)
+{
+	u32 cid = VMADDR_CID_ANY;
+
+	mutex_lock(&vsock_register_mutex);
+	if (*transport)
+		cid = (*transport)->get_local_cid();
+	mutex_unlock(&vsock_register_mutex);
+
+	return cid;
+}
+
 bool vsock_find_cid(unsigned int cid)
 {
-	if (transport_g2h && cid == transport_g2h->get_local_cid())
+	if (cid == vsock_registered_transport_cid(&transport_g2h))
 		return true;
 
 	if (transport_h2g && cid == VMADDR_CID_HOST)
@@ -2536,18 +2552,17 @@ static long vsock_dev_do_ioctl(struct file *filp,
 			       unsigned int cmd, void __user *ptr)
 {
 	u32 __user *p = ptr;
-	u32 cid = VMADDR_CID_ANY;
 	int retval = 0;
+	u32 cid;
 
 	switch (cmd) {
 	case IOCTL_VM_SOCKETS_GET_LOCAL_CID:
 		/* To be compatible with the VMCI behavior, we prioritize the
 		 * guest CID instead of well-know host CID (VMADDR_CID_HOST).
 		 */
-		if (transport_g2h)
-			cid = transport_g2h->get_local_cid();
-		else if (transport_h2g)
-			cid = transport_h2g->get_local_cid();
+		cid = vsock_registered_transport_cid(&transport_g2h);
+		if (cid == VMADDR_CID_ANY)
+			cid = vsock_registered_transport_cid(&transport_h2g);
 
 		if (put_user(cid, p) != 0)
 			retval = -EFAULT;

-- 
2.50.0

[PATCH net v4 2/3] vsock: Fix transport_* TOCTOU

[Michal Luczaj]

Transport assignment may race with module unload. Protect new_transport
from becoming a stale pointer.

This also takes care of an insecure call in vsock_use_local_transport();
add a lockdep assert.

BUG: unable to handle page fault for address: fffffbfff8056000
Oops: Oops: 0000 [#1] SMP KASAN
RIP: 0010:vsock_assign_transport+0x366/0x600
Call Trace:
 vsock_connect+0x59c/0xc40
 __sys_connect+0xe8/0x100
 __x64_sys_connect+0x6e/0xc0
 do_syscall_64+0x92/0x1c0
 entry_SYSCALL_64_after_hwframe+0x4b/0x53

Fixes: c0cfa2d8a788 ("vsock: add multi-transports support")
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Michal Luczaj <mhal@rbox.co>
---
 net/vmw_vsock/af_vsock.c | 28 +++++++++++++++++++++++-----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 39473b9e0829f240045262aef00cbae82a425dcc..66404c06bdaa0dc453a625c08a04c7eb14a95498 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -407,6 +407,8 @@ EXPORT_SYMBOL_GPL(vsock_enqueue_accept);
 
 static bool vsock_use_local_transport(unsigned int remote_cid)
 {
+	lockdep_assert_held(&vsock_register_mutex);
+
 	if (!transport_local)
 		return false;
 
@@ -464,6 +466,8 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 
 	remote_flags = vsk->remote_addr.svm_flags;
 
+	mutex_lock(&vsock_register_mutex);
+
 	switch (sk->sk_type) {
 	case SOCK_DGRAM:
 		new_transport = transport_dgram;
@@ -479,12 +483,15 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 			new_transport = transport_h2g;
 		break;
 	default:
-		return -ESOCKTNOSUPPORT;
+		ret = -ESOCKTNOSUPPORT;
+		goto err;
 	}
 
 	if (vsk->transport) {
-		if (vsk->transport == new_transport)
-			return 0;
+		if (vsk->transport == new_transport) {
+			ret = 0;
+			goto err;
+		}
 
 		/* transport->release() must be called with sock lock acquired.
 		 * This path can only be taken during vsock_connect(), where we
@@ -508,8 +515,16 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	/* We increase the module refcnt to prevent the transport unloading
 	 * while there are open sockets assigned to it.
 	 */
-	if (!new_transport || !try_module_get(new_transport->module))
-		return -ENODEV;
+	if (!new_transport || !try_module_get(new_transport->module)) {
+		ret = -ENODEV;
+		goto err;
+	}
+
+	/* It's safe to release the mutex after a successful try_module_get().
+	 * Whichever transport `new_transport` points at, it won't go away until
+	 * the last module_put() below or in vsock_deassign_transport().
+	 */
+	mutex_unlock(&vsock_register_mutex);
 
 	if (sk->sk_type == SOCK_SEQPACKET) {
 		if (!new_transport->seqpacket_allow ||
@@ -528,6 +543,9 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk)
 	vsk->transport = new_transport;
 
 	return 0;
+err:
+	mutex_unlock(&vsock_register_mutex);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(vsock_assign_transport);
 

-- 
2.50.0

[PATCH net v4 3/3] vsock: Fix IOCTL_VM_SOCKETS_GET_LOCAL_CID to check also `transport_local`

[Michal Luczaj]

Support returning VMADDR_CID_LOCAL in case no other vsock transport is
available.

Fixes: 0e12190578d0 ("vsock: add local transport support in the vsock core")
Suggested-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Michal Luczaj <mhal@rbox.co>
---
 net/vmw_vsock/af_vsock.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 66404c06bdaa0dc453a625c08a04c7eb14a95498..1053662725f8f03b78dc6ef80d46c31167ba0055 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -2581,6 +2581,8 @@ static long vsock_dev_do_ioctl(struct file *filp,
 		cid = vsock_registered_transport_cid(&transport_g2h);
 		if (cid == VMADDR_CID_ANY)
 			cid = vsock_registered_transport_cid(&transport_h2g);
+		if (cid == VMADDR_CID_ANY)
+			cid = vsock_registered_transport_cid(&transport_local);
 
 		if (put_user(cid, p) != 0)
 			retval = -EFAULT;

-- 
2.50.0

Re: [PATCH net v4 0/3] vsock: Fix transport_{h2g,g2h,dgram,local} TOCTOU issues

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Thu, 03 Jul 2025 17:18:17 +0200 you wrote:
> transport_{h2g,g2h,dgram,local} may become NULL on vsock_core_unregister().
> Make sure a poorly timed `rmmod transport` won't lead to a NULL/stale
> pointer dereference.
> 
> Note that these oopses are pretty unlikely to happen in the wild. Splats
> were collected after sprinkling kernel with mdelay()s.
> 
> [...]

Here is the summary with links:
  - [net,v4,1/3] vsock: Fix transport_{g2h,h2g} TOCTOU
    https://git.kernel.org/netdev/net/c/209fd720838a
  - [net,v4,2/3] vsock: Fix transport_* TOCTOU
    https://git.kernel.org/netdev/net/c/687aa0c5581b
  - [net,v4,3/3] vsock: Fix IOCTL_VM_SOCKETS_GET_LOCAL_CID to check also `transport_local`
    https://git.kernel.org/netdev/net/c/1e7d9df379a0

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