CVSS: 5.5EPSS: 0%CPEs: 2EXPL: 0CVE-2024-27010 – net/sched: Fix mirred deadlock on device recursion
https://notcve.org/view.php?id=CVE-2024-27010
In the Linux kernel, the following vulnerability has been resolved: net/sched: Fix mirred deadlock on device recursion When the mirred action is used on a classful egress qdisc and a packet is mirrored or redirected to self we hit a qdisc lock deadlock. See trace below. [..... other info removed for brevity....] [ 82.890906] [ 82.890906] ============================================ [ 82.890906] WARNING: possible recursive locking detected [ 82.890906] 6.8.0-05205-g77fadd89fe2d-dirty #213 Tainted: G W [ 82.890906] -------------------------------------------- [ 82.890906] ping/418 is trying to acquire lock: [ 82.890906] ffff888006994110 (&sch->q.lock){+.-.}-{3:3}, at: __dev_queue_xmit+0x1778/0x3550 [ 82.890906] [ 82.890906] but task is already holding lock: [ 82.890906] ffff888006994110 (&sch->q.lock){+.-.}-{3:3}, at: __dev_queue_xmit+0x1778/0x3550 [ 82.890906] [ 82.890906] other info that might help us debug this: [ 82.890906] Possible unsafe locking scenario: [ 82.890906] [ 82.890906] CPU0 [ 82.890906] ---- [ 82.890906] lock(&sch->q.lock); [ 82.890906] lock(&sch->q.lock); [ 82.890906] [ 82.890906] *** DEADLOCK *** [ 82.890906] [..... other info removed for brevity....] Example setup (eth0->eth0) to recreate tc qdisc add dev eth0 root handle 1: htb default 30 tc filter add dev eth0 handle 1: protocol ip prio 2 matchall \ action mirred egress redirect dev eth0 Another example(eth0->eth1->eth0) to recreate tc qdisc add dev eth0 root handle 1: htb default 30 tc filter add dev eth0 handle 1: protocol ip prio 2 matchall \ action mirred egress redirect dev eth1 tc qdisc add dev eth1 root handle 1: htb default 30 tc filter add dev eth1 handle 1: protocol ip prio 2 matchall \ action mirred egress redirect dev eth0 We fix this by adding an owner field (CPU id) to struct Qdisc set after root qdisc is entered. When the softirq enters it a second time, if the qdisc owner is the same CPU, the packet is dropped to break the loop. En el kernel de Linux, se resolvió la siguiente vulnerabilidad: net/sched: corrige el punto muerto reflejado en la recursividad del dispositivo Cuando la acción reflejada se utiliza en una qdisc de salida con clase y un paquete se refleja o redirige a uno mismo, llegamos a un punto muerto de bloqueo de qdisc. Vea el rastro a continuación. [..... otra información eliminada por brevedad....] [ 82.890906] [ 82.890906] ============================= ================ [82.890906] ADVERTENCIA: posible bloqueo recursivo detectado [82.890906] 6.8.0-05205-g77fadd89fe2d-dirty #213 Contaminado: GW [82.890906] ----- --------------------------------------- [ 82.890906] ping/418 está intentando adquirir el bloqueo : [ 82.890906] ffff888006994110 (&sch->q.lock){+.-.}-{3:3}, en: __dev_queue_xmit+0x1778/0x3550 [ 82.890906] [ 82.890906] pero la tarea ya mantiene el bloqueo: [ 82.890906] 88006994110 (&sch->q.lock){+.-.}-{3:3}, en: __dev_queue_xmit+0x1778/0x3550 [ 82.890906] [ 82.890906] otra información que podría ayudarnos a depurar esto: [ 82.890906] Posible escenario de bloqueo inseguro : [ 82.890906] [ 82.890906] CPU0 [ 82.890906] ---- [ 82.890906] bloqueo(&sch->q.lock); [ 82.890906] bloqueo(&sch->q.lock); [ 82.890906] [ 82.890906] *** DEADLOCK *** [ 82.890906] [..... otra información eliminada por brevedad....] Ejemplo de configuración (eth0->eth0) para recrear tc qdisc agregar dev eth0 identificador raíz 1 : htb default 30 tc filter add dev eth0 handle 1: protocolo ip prio 2 matchall \ action mirred egress redirigir dev eth0 Otro ejemplo (eth0->eth1->eth0) para recrear tc qdisc add dev eth0 root handle 1: htb default 30 tc filtro agregar dev eth0 identificador 1: protocolo ip prio 2 matchall \ acción redirección de salida reflejada dev eth1 tc qdisc agregar dev eth1 identificador raíz 1: htb predeterminado 30 tc filtro agregar dev eth1 identificador 1: protocolo ip prio 2 matchall \ acción duplicada redirección de salida dev eth0 Solucionamos esto agregando un campo de propietario (ID de CPU) a la estructura Qdisc establecida después de ingresar la qdisc raíz. • https://git.kernel.org/stable/c/e578d9c02587d57bfa7b560767c698a668a468c6 https://git.kernel.org/stable/c/e6b90468da4dae2281a6e381107f411efb48b0ef https://git.kernel.org/stable/c/0f022d32c3eca477fbf79a205243a6123ed0fe11 https://access.redhat.com/security/cve/CVE-2024-27010 https://bugzilla.redhat.com/show_bug.cgi?id=2278279 •
CVSS: -EPSS: 0%CPEs: 8EXPL: 0CVE-2024-27008 – drm: nv04: Fix out of bounds access
https://notcve.org/view.php?id=CVE-2024-27008
In the Linux kernel, the following vulnerability has been resolved: drm: nv04: Fix out of bounds access When Output Resource (dcb->or) value is assigned in fabricate_dcb_output(), there may be out of bounds access to dac_users array in case dcb->or is zero because ffs(dcb->or) is used as index there. The 'or' argument of fabricate_dcb_output() must be interpreted as a number of bit to set, not value. Utilize macros from 'enum nouveau_or' in calls instead of hardcoding. Found by Linux Verification Center (linuxtesting.org) with SVACE. En el kernel de Linux, se resolvió la siguiente vulnerabilidad: drm: nv04: corregir el acceso fuera de los límites Cuando se asigna el valor del recurso de salida (dcb->or) en fabricate_dcb_output(), puede haber acceso fuera de los límites a la matriz dac_users en caso de que dcb->or es cero porque ffs(dcb->or) se usa como índice allí. El argumento 'o' de fabricate_dcb_output() debe interpretarse como un número de bits a configurar, no como un valor. Utilice macros de 'enum nouveau_or' en las llamadas en lugar de codificarlas. Encontrado por el Centro de verificación de Linux (linuxtesting.org) con SVACE. • https://git.kernel.org/stable/c/2e5702aff39532662198459726c624d5eadbdd78 https://git.kernel.org/stable/c/c2b97f26f081ceec3298151481687071075a25cb https://git.kernel.org/stable/c/5050ae879a828d752b439e3827aac126709da6d1 https://git.kernel.org/stable/c/097c7918fcfa1dee233acfd1f3029f00c3bc8062 https://git.kernel.org/stable/c/df0991da7db846f7fa4ec6740350f743d3b69b04 https://git.kernel.org/stable/c/5fd4b090304e450aa0e7cc9cc2b4873285c6face https://git.kernel.org/stable/c/6690cc2732e2a8d0eaca44dcbac032a4b0148042 https://git.kernel.org/stable/c/26212da39ee14a52c76a202c6ae5153a8 •
CVSS: -EPSS: 0%CPEs: 8EXPL: 0CVE-2024-27001 – comedi: vmk80xx: fix incomplete endpoint checking
https://notcve.org/view.php?id=CVE-2024-27001
In the Linux kernel, the following vulnerability has been resolved: comedi: vmk80xx: fix incomplete endpoint checking While vmk80xx does have endpoint checking implemented, some things can fall through the cracks. Depending on the hardware model, URBs can have either bulk or interrupt type, and current version of vmk80xx_find_usb_endpoints() function does not take that fully into account. While this warning does not seem to be too harmful, at the very least it will crash systems with 'panic_on_warn' set on them. Fix the issue found by Syzkaller [1] by somewhat simplifying the endpoint checking process with usb_find_common_endpoints() and ensuring that only expected endpoint types are present. This patch has not been tested on real hardware. [1] Syzkaller report: usb 1-1: BOGUS urb xfer, pipe 1 != type 3 WARNING: CPU: 0 PID: 781 at drivers/usb/core/urb.c:504 usb_submit_urb+0xc4e/0x18c0 drivers/usb/core/urb.c:503 ... Call Trace: <TASK> usb_start_wait_urb+0x113/0x520 drivers/usb/core/message.c:59 vmk80xx_reset_device drivers/comedi/drivers/vmk80xx.c:227 [inline] vmk80xx_auto_attach+0xa1c/0x1a40 drivers/comedi/drivers/vmk80xx.c:818 comedi_auto_config+0x238/0x380 drivers/comedi/drivers.c:1067 usb_probe_interface+0x5cd/0xb00 drivers/usb/core/driver.c:399 ... Similar issue also found by Syzkaller: En el kernel de Linux, se resolvió la siguiente vulnerabilidad: comedi: vmk80xx: corrige la verificación de endpoints incompleta Si bien vmk80xx tiene implementada la verificación de endpoints, algunas cosas pueden pasar desapercibidas. Dependiendo del modelo de hardware, las URB pueden tener un tipo masivo o de interrupción, y la versión actual de la función vmk80xx_find_usb_endpoints() no lo tiene completamente en cuenta. • https://git.kernel.org/stable/c/49253d542cc0f5f771dc254d248162a2a666649d https://git.kernel.org/stable/c/3a63ae0348d990e137cca04eced5b08379969ea9 https://git.kernel.org/stable/c/a3b8ae7e9297dd453f2977b011c5bc75eb20e71b https://git.kernel.org/stable/c/f15370e315976198f338b41611f37ce82af6cf54 https://git.kernel.org/stable/c/b0b268eeb087e324ef3ea71f8e6cabd07630517f https://git.kernel.org/stable/c/ac882d6b21bffecb57bcc4486701239eef5aa67b https://git.kernel.org/stable/c/59f33af9796160f851641d960bd93937f282c696 https://git.kernel.org/stable/c/6ec3514a7d35ad9cfab600187612c29f6 •
CVSS: -EPSS: 0%CPEs: 8EXPL: 0CVE-2024-27000 – serial: mxs-auart: add spinlock around changing cts state
https://notcve.org/view.php?id=CVE-2024-27000
In the Linux kernel, the following vulnerability has been resolved: serial: mxs-auart: add spinlock around changing cts state The uart_handle_cts_change() function in serial_core expects the caller to hold uport->lock. For example, I have seen the below kernel splat, when the Bluetooth driver is loaded on an i.MX28 board. [ 85.119255] ------------[ cut here ]------------ [ 85.124413] WARNING: CPU: 0 PID: 27 at /drivers/tty/serial/serial_core.c:3453 uart_handle_cts_change+0xb4/0xec [ 85.134694] Modules linked in: hci_uart bluetooth ecdh_generic ecc wlcore_sdio configfs [ 85.143314] CPU: 0 PID: 27 Comm: kworker/u3:0 Not tainted 6.6.3-00021-gd62a2f068f92 #1 [ 85.151396] Hardware name: Freescale MXS (Device Tree) [ 85.156679] Workqueue: hci0 hci_power_on [bluetooth] (...) [ 85.191765] uart_handle_cts_change from mxs_auart_irq_handle+0x380/0x3f4 [ 85.198787] mxs_auart_irq_handle from __handle_irq_event_percpu+0x88/0x210 (...) En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: serial: mxs-auart: agrega spinlock para cambiar el estado de cts. La función uart_handle_cts_change() en serial_core espera que la persona que llama mantenga uport->lock. Por ejemplo, he visto el siguiente símbolo del kernel, cuando el controlador Bluetooth está cargado en una placa i.MX28. [85.119255] ------------[ cortar aquí ]------------ [ 85.124413] ADVERTENCIA: CPU: 0 PID: 27 en /drivers/tty/serial/ serial_core.c:3453 uart_handle_cts_change+0xb4/0xec [85.134694] Módulos vinculados en: hci_uart bluetooth ecdh_generic ecc wlcore_sdio configfs [85.143314] CPU: 0 PID: 27 Comm: kworker/u3:0 No contaminado 6.6.3-00021-gd6 2a2f068f92 #1 [85.151396] Nombre de hardware: Freescale MXS (árbol de dispositivos) [85.156679] Cola de trabajo: hci0 hci_power_on [bluetooth] (...) [85.191765] uart_handle_cts_change from mxs_auart_irq_handle+0x380/0x3f4 [ 85.198787] q_handle de __handle_irq_event_percpu+0x88/0x210 (.. .) • https://git.kernel.org/stable/c/4d90bb147ef6b91f529a21b498ff2b5fdc6785b4 https://git.kernel.org/stable/c/56434e295bd446142025913bfdf1587f5e1970ad https://git.kernel.org/stable/c/21535ef0ac1945080198fe3e4347ea498205c99a https://git.kernel.org/stable/c/0dc0637e6b16158af85945425821bfd0151adb37 https://git.kernel.org/stable/c/479244d68f5d94f3903eced52b093c1e01ddb495 https://git.kernel.org/stable/c/2c9b943e9924cf1269e44289bc5e60e51b0f5270 https://git.kernel.org/stable/c/5f40fd6ca2cf0bfbc5a5c9e403dfce8ca899ba37 https://git.kernel.org/stable/c/94b0e65c75f4af888ab2dd6c90f060f76 •
CVSS: -EPSS: 0%CPEs: 8EXPL: 0CVE-2024-26999 – serial/pmac_zilog: Remove flawed mitigation for rx irq flood
https://notcve.org/view.php?id=CVE-2024-26999
In the Linux kernel, the following vulnerability has been resolved: serial/pmac_zilog: Remove flawed mitigation for rx irq flood The mitigation was intended to stop the irq completely. That may be better than a hard lock-up but it turns out that you get a crash anyway if you're using pmac_zilog as a serial console: ttyPZ0: pmz: rx irq flood ! BUG: spinlock recursion on CPU#0, swapper/0 That's because the pr_err() call in pmz_receive_chars() results in pmz_console_write() attempting to lock a spinlock already locked in pmz_interrupt(). With CONFIG_DEBUG_SPINLOCK=y, this produces a fatal BUG splat. The spinlock in question is the one in struct uart_port. Even when it's not fatal, the serial port rx function ceases to work. Also, the iteration limit doesn't play nicely with QEMU, as can be seen in the bug report linked below. A web search for other reports of the error message "pmz: rx irq flood" didn't produce anything. • https://git.kernel.org/stable/c/1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 https://git.kernel.org/stable/c/69a02273e288011b521ee7c1f3ab2c23fda633ce https://git.kernel.org/stable/c/d679c816929d62af51c8e6d7fc0e165c9412d2f3 https://git.kernel.org/stable/c/ab86cf6f8d24e63e9aca23da5108af1aa5483928 https://git.kernel.org/stable/c/7a3bbe41efa55323b6ea3c35fa15941d4dbecdef https://git.kernel.org/stable/c/bbaafbb4651fede8d3c3881601ecaa4f834f9d3f https://git.kernel.org/stable/c/52aaf1ff14622a04148dbb9ccce6d9de5d534ea7 https://git.kernel.org/stable/c/ca09dfc3cfdf89e6af3ac24e1c6c0be5c •