CVE-2024-41010
bpf: Fix too early release of tcx_entry
Severity Score
Exploit Likelihood
Affected Versions
Public Exploits
0Exploited in Wild
-Decision
Descriptions
In the Linux kernel, the following vulnerability has been resolved: bpf: Fix too early release of tcx_entry Pedro Pinto and later independently also Hyunwoo Kim and Wongi Lee reported
an issue that the tcx_entry can be released too early leading to a use
after free (UAF) when an active old-style ingress or clsact qdisc with a
shared tc block is later replaced by another ingress or clsact instance. Essentially, the sequence to trigger the UAF (one example) can be as follows: 1. A network namespace is created 2. An ingress qdisc is created. This allocates a tcx_entry, and &tcx_entry->miniq is stored in the qdisc's miniqp->p_miniq. At the same time, a tcf block with index 1 is created. 3. chain0 is attached to the tcf block. chain0 must be connected to the block linked to the ingress qdisc to later reach the function tcf_chain0_head_change_cb_del() which triggers the UAF. 4. Create and graft a clsact qdisc. This causes the ingress qdisc created in step 1 to be removed, thus freeing the previously linked tcx_entry: rtnetlink_rcv_msg() => tc_modify_qdisc() => qdisc_create() => clsact_init() [a] => qdisc_graft() => qdisc_destroy() => __qdisc_destroy() => ingress_destroy() [b] => tcx_entry_free() => kfree_rcu() // tcx_entry freed 5. Finally, the network namespace is closed. This registers the cleanup_net worker, and during the process of releasing the remaining clsact qdisc, it accesses the tcx_entry that was already freed in step 4, causing the UAF to occur: cleanup_net() => ops_exit_list() => default_device_exit_batch() => unregister_netdevice_many() => unregister_netdevice_many_notify() => dev_shutdown() => qdisc_put() => clsact_destroy() [c] => tcf_block_put_ext() => tcf_chain0_head_change_cb_del() => tcf_chain_head_change_item() => clsact_chain_head_change() => mini_qdisc_pair_swap() // UAF There are also other variants, the gist is to add an ingress (or clsact)
qdisc with a specific shared block, then to replace that qdisc, waiting
for the tcx_entry kfree_rcu() to be executed and subsequently accessing
the current active qdisc's miniq one way or another. The correct fix is to turn the miniq_active boolean into a counter. What
can be observed, at step 2 above, the counter transitions from 0->1, at
step [a] from 1->2 (in order for the miniq object to remain active during
the replacement), then in [b] from 2->1 and finally [c] 1->0 with the
eventual release. The reference counter in general ranges from [0,2] and
it does not need to be atomic since all access to the counter is protected
by the rtnl mutex. With this in place, there is no longer a UAF happening
and the tcx_entry is freed at the correct time.
En el kernel de Linux, se resolvió la siguiente vulnerabilidad: bpf: Se solucionó el lanzamiento demasiado temprano de tcx_entry Pedro Pinto y más tarde, de forma independiente, también Hyunwoo Kim y Wongi Lee informaron un problema por el cual tcx_entry se puede lanzar demasiado pronto, lo que lleva a un uso posterior a la liberación (UAF ) cuando una qdisc ingress o clsact antigua activa con un bloque tc compartido se reemplaza posteriormente por otra instancia de ingress o clsact. Esencialmente, la secuencia para activar la UAF (un ejemplo) puede ser la siguiente: 1. Se crea un espacio de nombres de red. 2. Se crea una qdisc de entrada. Esto asigna un tcx_entry, y &tcx_entry->miniq se almacena en el miniqp->p_miniq de la qdisc. Al mismo tiempo, se crea un bloque tcf con índice 1. 3. chain0 está adjunta al bloque tcf. chain0 debe estar conectado al bloque vinculado a la qdisc de ingreso para luego llegar a la función tcf_chain0_head_change_cb_del() que activa la UAF. 4. Cree e injerte una qdisc clsact. Esto hace que se elimine la qdisc de entrada creada en el paso 1, liberando así la tcx_entry previamente vinculada: rtnetlink_rcv_msg() => tc_modify_qdisc() => qdisc_create() => clsact_init() [a] => qdisc_graft() => qdisc_destroy( ) => __qdisc_destroy() => ingress_destroy() [b] => tcx_entry_free() => kfree_rcu() // tcx_entry liberado 5. Finalmente, se cierra el espacio de nombres de la red. Esto registra el trabajador cleanup_net y, durante el proceso de liberación de la qdisc clsact restante, accede a tcx_entry que ya se liberó en el paso 4, lo que provoca que se produzca la UAF: cleanup_net() => ops_exit_list() => default_device_exit_batch() => unregister_netdevice_many() => unregister_netdevice_many_notify() => dev_shutdown() => qdisc_put() => clsact_destroy() [c] => tcf_block_put_ext() => tcf_chain0_head_change_cb_del() => tcf_chain_head_change_item() => clsact_chain_head_change() => mini_qdisc_pair _intercambiar( ) // UAF También hay otras variantes, lo esencial es agregar una qdisc de ingreso (o clsact) con un bloque compartido específico, luego reemplazar esa qdisc, esperar a que se ejecute tcx_entry kfree_rcu() y posteriormente acceder al activo actual miniq de qdisc de una forma u otra. La solución correcta es convertir el booleano miniq_active en un contador. Lo que se puede observar, en el paso 2 anterior, el contador pasa de 0->1, en el paso [a] de 1->2 (para que el objeto miniq permanezca activo durante el reemplazo), luego en [b] de 2->1 y finalmente [c] 1->0 con el eventual lanzamiento. El contador de referencia en general oscila entre [0,2] y no necesita ser atómico ya que todo acceso al contador está protegido por el mutex rtnl. Con esto implementado, ya no ocurre ningún UAF y tcx_entry se libera en el momento correcto.
In the Linux kernel, the following vulnerability has been resolved: bpf: Fix too early release of tcx_entry Pedro Pinto and later independently also Hyunwoo Kim and Wongi Lee reported an issue that the tcx_entry can be released too early leading to a use after free (UAF) when an active old-style ingress or clsact qdisc with a shared tc block is later replaced by another ingress or clsact instance. Essentially, the sequence to trigger the UAF (one example) can be as follows: 1. A network namespace is created 2. An ingress qdisc is created. This allocates a tcx_entry, and &tcx_entry->miniq is stored in the qdisc's miniqp->p_miniq. At the same time, a tcf block with index 1 is created. 3. chain0 is attached to the tcf block. chain0 must be connected to the block linked to the ingress qdisc to later reach the function tcf_chain0_head_change_cb_del() which triggers the UAF. 4. Create and graft a clsact qdisc. This causes the ingress qdisc created in step 1 to be removed, thus freeing the previously linked tcx_entry: rtnetlink_rcv_msg() => tc_modify_qdisc() => qdisc_create() => clsact_init() [a] => qdisc_graft() => qdisc_destroy() => __qdisc_destroy() => ingress_destroy() [b] => tcx_entry_free() => kfree_rcu() // tcx_entry freed 5. Finally, the network namespace is closed. This registers the cleanup_net worker, and during the process of releasing the remaining clsact qdisc, it accesses the tcx_entry that was already freed in step 4, causing the UAF to occur: cleanup_net() => ops_exit_list() => default_device_exit_batch() => unregister_netdevice_many() => unregister_netdevice_many_notify() => dev_shutdown() => qdisc_put() => clsact_destroy() [c] => tcf_block_put_ext() => tcf_chain0_head_change_cb_del() => tcf_chain_head_change_item() => clsact_chain_head_change() => mini_qdisc_pair_swap() // UAF There are also other variants, the gist is to add an ingress (or clsact) qdisc with a specific shared block, then to replace that qdisc, waiting for the tcx_entry kfree_rcu() to be executed and subsequently accessing the current active qdisc's miniq one way or another. The correct fix is to turn the miniq_active boolean into a counter. What can be observed, at step 2 above, the counter transitions from 0->1, at step [a] from 1->2 (in order for the miniq object to remain active during the replacement), then in [b] from 2->1 and finally [c] 1->0 with the eventual release. The reference counter in general ranges from [0,2] and it does not need to be atomic since all access to the counter is protected by the rtnl mutex. With this in place, there is no longer a UAF happening and the tcx_entry is freed at the correct time.
Chenyuan Yang discovered that the USB Gadget subsystem in the Linux kernel did not properly check for the device to be enabled before writing. A local attacker could possibly use this to cause a denial of service. Several security issues were discovered in the Linux kernel. An attacker could possibly use these to compromise the system.
CVSS Scores
SSVC
- Decision:Track
Timeline
- 2024-07-12 CVE Reserved
- 2024-07-17 CVE Published
- 2025-05-04 CVE Updated
- 2025-11-27 EPSS Updated
- ---------- Exploited in Wild
- ---------- KEV Due Date
- ---------- First Exploit
CWE
- CWE-416: Use After Free
CAPEC
References (6)
| URL | Tag | Source |
|---|---|---|
| https://git.kernel.org/stable/c/e420bed025071a623d2720a92bc2245c84757ecb | Vuln. Introduced |
| URL | Date | SRC |
|---|
| URL | Date | SRC |
|---|---|---|
| https://access.redhat.com/security/cve/CVE-2024-41010 | 2025-05-13 | |
| https://bugzilla.redhat.com/show_bug.cgi?id=2298411 | 2025-05-13 |
Affected Vendors, Products, and Versions
| Vendor | Product | Version | Other | Status | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Vendor | Product | Version | Other | Status | <-- --> | Vendor | Product | Version | Other | Status |
| Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 6.6 < 6.6.41 Search vendor "Linux" for product "Linux Kernel" and version " >= 6.6 < 6.6.41" | en |
Affected
| ||||||
| Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 6.6 < 6.9.10 Search vendor "Linux" for product "Linux Kernel" and version " >= 6.6 < 6.9.10" | en |
Affected
| ||||||
| Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 6.6 < 6.10 Search vendor "Linux" for product "Linux Kernel" and version " >= 6.6 < 6.10" | en |
Affected
| ||||||
