"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nRevert \"tty: n_gsm: fix UAF in gsm_cleanup_mux\"\n\nThis reverts commit 9b9c8195f3f0d74a826077fc1c01b9ee74907239.\n\nThe commit above is reverted as it did not solve the original issue.\n\ngsm_cleanup_mux() tries to free up the virtual ttys by calling\ngsm_dlci_release() for each available DLCI. There, dlci_put() is called to\ndecrease the reference counter for the DLCI via tty_port_put() which\nfinally calls gsm_dlci_free(). This already clears the pointer which is\nbeing checked in gsm_cleanup_mux() before calling gsm_dlci_release().\nTherefore, it is not necessary to clear this pointer in gsm_cleanup_mux()\nas done in the reverted commit. The commit introduces a null pointer\ndereference:\n <TASK>\n ? __die+0x1f/0x70\n ? page_fault_oops+0x156/0x420\n ? search_exception_tables+0x37/0x50\n ? fixup_exception+0x21/0x310\n ? exc_page_fault+0x69/0x150\n ? asm_exc_page_fault+0x26/0x30\n ? tty_port_put+0x19/0xa0\n gsmtty_cleanup+0x29/0x80 [n_gsm]\n release_one_tty+0x37/0xe0\n process_one_work+0x1e6/0x3e0\n worker_thread+0x4c/0x3d0\n ? __pfx_worker_thread+0x10/0x10\n kthread+0xe1/0x110\n ? __pfx_kthread+0x10/0x10\n ret_from_fork+0x2f/0x50\n ? __pfx_kthread+0x10/0x10\n ret_from_fork_asm+0x1b/0x30\n </TASK>\n\nThe actual issue is that nothing guards dlci_put() from being called\nmultiple times while the tty driver was triggered but did not yet finished\ncalling gsm_dlci_free()."
"value":"En el kernel de Linux, se resolvi\u00f3 la siguiente vulnerabilidad: Revertir \"tty: n_gsm: fix UAF in gsm_cleanup_mux\" Esto revierte el commit 9b9c8195f3f0d74a826077fc1c01b9ee74907239. el commit anterior se revierte porque no resolvi\u00f3 el problema original. gsm_cleanup_mux() intenta liberar los ttys virtuales llamando a gsm_dlci_release() para cada DLCI disponible. All\u00ed, se llama a dlci_put() para disminuir el contador de referencia para el DLCI a trav\u00e9s de tty_port_put() que finalmente llama a gsm_dlci_free(). Esto ya borra el puntero que se est\u00e1 verificando en gsm_cleanup_mux() antes de llamar a gsm_dlci_release(). Por lo tanto, no es necesario borrar este puntero en gsm_cleanup_mux() como se hizo en el commit revertida. el commit introduce una desreferencia de puntero nulo: ? __die+0x1f/0x70 ? page_fault_oops+0x156/0x420? search_exception_tables+0x37/0x50? fixup_exception+0x21/0x310? exc_page_fault+0x69/0x150? asm_exc_page_fault+0x26/0x30? tty_port_put+0x19/0xa0 gsmtty_cleanup+0x29/0x80 [n_gsm] release_one_tty+0x37/0xe0 proceso_one_work+0x1e6/0x3e0 trabajador_thread+0x4c/0x3d0 ? __pfx_worker_thread+0x10/0x10 kthread+0xe1/0x110 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x2f/0x50 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1b/0x30 El problema real es que nada protege a dlci_put() de ser llamado varias veces mientras el controlador tty se activ\u00f3 pero a\u00fan no termin\u00f3 de llamar a gsm_dlci_free()."