CVSS: -EPSS: 0%CPEs: 5EXPL: 0CVE-2021-47012 – RDMA/siw: Fix a use after free in siw_alloc_mr
https://notcve.org/view.php?id=CVE-2021-47012
In the Linux kernel, the following vulnerability has been resolved: RDMA/siw: Fix a use after free in siw_alloc_mr Our code analyzer reported a UAF. In siw_alloc_mr(), it calls siw_mr_add_mem(mr,..). In the implementation of siw_mr_add_mem(), mem is assigned to mr->mem and then mem is freed via kfree(mem) if xa_alloc_cyclic() failed. Here, mr->mem still point to a freed object. After, the execution continue up to the err_out branch of siw_alloc_mr, and the freed mr->mem is used in siw_mr_drop_mem(mr). My patch moves "mr->mem = mem" behind the if (xa_alloc_cyclic(..)<0) {} section, to avoid the uaf. En el kernel de Linux se ha resuelto la siguiente vulnerabilidad: RDMA/siw: Corrige un use after free en siw_alloc_mr Nuestro analizador de código informó un UAF. • https://git.kernel.org/stable/c/2251334dcac9eb337575d8767e2a6a7e81848f7f https://git.kernel.org/stable/c/30b9e92d0b5e5d5dc1101ab856c17009537cbca4 https://git.kernel.org/stable/c/608a4b90ece039940e9425ee2b39c8beff27e00c https://git.kernel.org/stable/c/3e22b88e02c194f6c80867abfef5cc09383461f4 https://git.kernel.org/stable/c/ad9ce7188432650469a6c7625bf479f5ed0b6155 https://git.kernel.org/stable/c/3093ee182f01689b89e9f8797b321603e5de4f63 •
CVSS: 7.8EPSS: 0%CPEs: 6EXPL: 0CVE-2021-47010 – net: Only allow init netns to set default tcp cong to a restricted algo
https://notcve.org/view.php?id=CVE-2021-47010
In the Linux kernel, the following vulnerability has been resolved: net: Only allow init netns to set default tcp cong to a restricted algo tcp_set_default_congestion_control() is netns-safe in that it writes to &net->ipv4.tcp_congestion_control, but it also sets ca->flags |= TCP_CONG_NON_RESTRICTED which is not namespaced. This has the unintended side-effect of changing the global net.ipv4.tcp_allowed_congestion_control sysctl, despite the fact that it is read-only: 97684f0970f6 ("net: Make tcp_allowed_congestion_control readonly in non-init netns") Resolve this netns "leak" by only allowing the init netns to set the default algorithm to one that is restricted. This restriction could be removed if tcp_allowed_congestion_control were namespace-ified in the future. This bug was uncovered with https://github.com/JonathonReinhart/linux-netns-sysctl-verify En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: net: solo permite que init netns establezca la cong tcp predeterminada en un algoritmo restringido tcp_set_default_congestion_control() es seguro para netns porque escribe en &net->ipv4.tcp_congestion_control, pero también establece ca->flags |= TCP_CONG_NON_RESTRICTED que no tiene espacio de nombres. Esto tiene el efecto secundario no deseado de cambiar el sistema global net.ipv4.tcp_allowed_congestion_control, a pesar de que es de solo lectura: 97684f0970f6 ("net: Make tcp_allowed_congestion_control readonly in non-init netns") Resuelva esta "fuga" de netns solo permite que las redes de inicio establezcan el algoritmo predeterminado en uno restringido. Esta restricción podría eliminarse si tcp_allowed_congestion_control tuviera un espacio de nombres en el futuro. Este error se descubrió con https://github.com/JonathonReinhart/linux-netns-sysctl-verify • https://git.kernel.org/stable/c/6670e152447732ba90626f36dfc015a13fbf150e https://git.kernel.org/stable/c/992de06308d9a9584d59b96d294ac676f924e437 https://git.kernel.org/stable/c/9884f745108f7d25b189bbcd6754e284fb29ab68 https://git.kernel.org/stable/c/6c1ea8bee75df8fe2184a50fcd0f70bf82986f42 https://git.kernel.org/stable/c/efe1532a6e1a8e3c343d04fff510f0ed80328f9c https://git.kernel.org/stable/c/e7d7bedd507bb732e600403b7a96f9fe48d0ca31 https://git.kernel.org/stable/c/8d432592f30fcc34ef5a10aac4887b4897884493 • CWE-400: Uncontrolled Resource Consumption •
CVSS: -EPSS: 0%CPEs: 4EXPL: 0CVE-2021-47007 – f2fs: fix panic during f2fs_resize_fs()
https://notcve.org/view.php?id=CVE-2021-47007
In the Linux kernel, the following vulnerability has been resolved: f2fs: fix panic during f2fs_resize_fs() f2fs_resize_fs() hangs in below callstack with testcase: - mkfs 16GB image & mount image - dd 8GB fileA - dd 8GB fileB - sync - rm fileA - sync - resize filesystem to 8GB kernel BUG at segment.c:2484! Call Trace: allocate_segment_by_default+0x92/0xf0 [f2fs] f2fs_allocate_data_block+0x44b/0x7e0 [f2fs] do_write_page+0x5a/0x110 [f2fs] f2fs_outplace_write_data+0x55/0x100 [f2fs] f2fs_do_write_data_page+0x392/0x850 [f2fs] move_data_page+0x233/0x320 [f2fs] do_garbage_collect+0x14d9/0x1660 [f2fs] free_segment_range+0x1f7/0x310 [f2fs] f2fs_resize_fs+0x118/0x330 [f2fs] __f2fs_ioctl+0x487/0x3680 [f2fs] __x64_sys_ioctl+0x8e/0xd0 do_syscall_64+0x33/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xa9 The root cause is we forgot to check that whether we have enough space in resized filesystem to store all valid blocks in before-resizing filesystem, then allocator will run out-of-space during block migration in free_segment_range(). En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: f2fs: corrige el pánico durante f2fs_resize_fs() f2fs_resize_fs() se bloquea en la pila de llamadas debajo con el caso de prueba: - imagen mkfs de 16 GB y montaje de imagen - dd archivo A de 8 GB - agrega archivo B de 8 GB - sincronización - archivo rm A - sincronización - cambiar el tamaño del sistema de archivos al kernel de 8 GB ¡ERROR en segment.c:2484! Seguimiento de llamadas: allocate_segment_by_default+0x92/0xf0 [f2fs] f2fs_allocate_data_block+0x44b/0x7e0 [f2fs] do_write_page+0x5a/0x110 [f2fs] f2fs_outplace_write_data+0x55/0x100 [f2fs] f2fs_do_write_data_page +0x392/0x850 [f2fs] mover_página_datos+0x233/0x320 [f2fs] ] do_garbage_collect+0x14d9/0x1660 [f2fs] free_segment_range+0x1f7/0x310 [f2fs] f2fs_resize_fs+0x118/0x330 [f2fs] __f2fs_ioctl+0x487/0x3680 [f2fs] __x64_sys_ioct l+0x8e/0xd0 do_syscall_64+0x33/0x80 Entry_SYSCALL_64_after_hwframe+0x44/0xa9 La raíz Porque olvidamos verificar si tenemos suficiente espacio en el sistema de archivos redimensionado para almacenar todos los bloques válidos en el sistema de archivos antes de cambiar el tamaño, entonces el asignador se quedará sin espacio durante la migración de bloques en free_segment_range(). • https://git.kernel.org/stable/c/b4b10061ef98c583bcf82a4200703fbaa98c18dc https://git.kernel.org/stable/c/1c20a4896409f5ca1c770e1880c33d0a28a8b10f https://git.kernel.org/stable/c/860afd680d9cc1dabd61cda3cd246f60aa1eb705 https://git.kernel.org/stable/c/822054e5026c43b1dd60cf387dd999e95ee2ecc2 https://git.kernel.org/stable/c/3ab0598e6d860ef49d029943ba80f627c15c15d6 •
CVSS: -EPSS: 0%CPEs: 8EXPL: 0CVE-2021-47006 – ARM: 9064/1: hw_breakpoint: Do not directly check the event's overflow_handler hook
https://notcve.org/view.php?id=CVE-2021-47006
In the Linux kernel, the following vulnerability has been resolved: ARM: 9064/1: hw_breakpoint: Do not directly check the event's overflow_handler hook The commit 1879445dfa7b ("perf/core: Set event's default ::overflow_handler()") set a default event->overflow_handler in perf_event_alloc(), and replace the check event->overflow_handler with is_default_overflow_handler(), but one is missing. Currently, the bp->overflow_handler can not be NULL. As a result, enable_single_step() is always not invoked. Comments from Zhen Lei: https://patchwork.kernel.org/project/linux-arm-kernel/patch/20210207105934.2001-1-thunder.leizhen@huawei.com/ En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: ARM: 9064/1: hw_breakpoint: no comprobar directamente el gancho overflow_handler del evento. el commit 1879445dfa7b ("perf/core: establecer el valor predeterminado del evento ::overflow_handler()") establece un valor predeterminado event->overflow_handler en perf_event_alloc(), y reemplace check event->overflow_handler con is_default_overflow_handler(), pero falta uno. Actualmente, bp->overflow_handler no puede ser NULL. Como resultado, enable_single_step() no siempre se invoca. Comentarios de Zhen Lei: https://patchwork.kernel.org/project/linux-arm-kernel/patch/20210207105934.2001-1-thunder.leizhen@huawei.com/ • https://git.kernel.org/stable/c/1879445dfa7bbd6fe21b09c5cc72f4934798afed https://git.kernel.org/stable/c/555a70f7fff03bd669123487905c47ae27dbdaac https://git.kernel.org/stable/c/ed1f67465327cec4457bb988775245b199da86e6 https://git.kernel.org/stable/c/a9938d6d78a238d6ab8de57a4d3dcf77adceb9bb https://git.kernel.org/stable/c/3ed8832aeaa9a37b0fc386bb72ff604352567c80 https://git.kernel.org/stable/c/630146203108bf6b8934eec0dfdb3e46dcb917de https://git.kernel.org/stable/c/7eeacc6728c5478e3c01bc82a1f08958eaa12366 https://git.kernel.org/stable/c/dabe299425b1a53a69461fed7ac8922ea •
CVSS: -EPSS: 0%CPEs: 4EXPL: 0CVE-2021-47005 – PCI: endpoint: Fix NULL pointer dereference for ->get_features()
https://notcve.org/view.php?id=CVE-2021-47005
In the Linux kernel, the following vulnerability has been resolved: PCI: endpoint: Fix NULL pointer dereference for ->get_features() get_features ops of pci_epc_ops may return NULL, causing NULL pointer dereference in pci_epf_test_alloc_space function. Let us add a check for pci_epc_feature pointer in pci_epf_test_bind before we access it to avoid any such NULL pointer dereference and return -ENOTSUPP in case pci_epc_feature is not found. When the patch is not applied and EPC features is not implemented in the platform driver, we see the following dump due to kernel NULL pointer dereference. Call trace: pci_epf_test_bind+0xf4/0x388 pci_epf_bind+0x3c/0x80 pci_epc_epf_link+0xa8/0xcc configfs_symlink+0x1a4/0x48c vfs_symlink+0x104/0x184 do_symlinkat+0x80/0xd4 __arm64_sys_symlinkat+0x1c/0x24 el0_svc_common.constprop.3+0xb8/0x170 el0_svc_handler+0x70/0x88 el0_svc+0x8/0x640 Code: d2800581 b9403ab9 f9404ebb 8b394f60 (f9400400) ---[ end trace a438e3c5a24f9df0 ]--- En el kernel de Linux, se resolvió la siguiente vulnerabilidad: PCI: endpoint: corrigió la desreferencia del puntero NULL para ->get_features() las operaciones get_features de pci_epc_ops pueden devolver NULL, lo que provoca la desreferencia del puntero NULL en la función pci_epf_test_alloc_space. Agreguemos una verificación del puntero pci_epc_feature en pci_epf_test_bind antes de acceder a él para evitar dicha desreferencia del puntero NULL y devolvamos -ENOTSUPP en caso de que no se encuentre pci_epc_feature. Cuando no se aplica el parche y las funciones de EPC no se implementan en el controlador de la plataforma, vemos el siguiente volcado debido a la desreferencia del puntero NULL del kernel. Rastreo de llamadas: pci_epf_test_bind+0xf4/0x388 pci_epf_bind+0x3c/0x80 pci_epc_epf_link+0xa8/0xcc configfs_symlink+0x1a4/0x48c vfs_symlink+0x104/0x184 do_symlinkat+0x80/0xd4 __arm64_sys _symlinkat+0x1c/0x24 el0_svc_common.constprop.3+0xb8/0x170 el0_svc_handler+0x70 /0x88 el0_svc+0x8/0x640 Código: d2800581 b9403ab9 f9404ebb 8b394f60 (f9400400) ---[ final de seguimiento a438e3c5a24f9df0 ]--- • https://git.kernel.org/stable/c/2c04c5b8eef797dca99699cfb55ff42dd3c12c23 https://git.kernel.org/stable/c/bbed83d7060e07a5d309104d25a00f0a24441428 https://git.kernel.org/stable/c/679ebad058b8168f10e63876d63b0877fd2fe784 https://git.kernel.org/stable/c/0169d4f0bee44fdfef908c13ed21fcb326c38695 https://git.kernel.org/stable/c/6613bc2301ba291a1c5a90e1dc24cf3edf223c03 •