Vulnerabilidad en kernel de Linux (CVE-2024-26588)

En el kernel de Linux, se resolvió la siguiente vulnerabilidad: LoongArch: BPF: evita el acceso a la memoria fuera de los límites La prueba test_tag desencadena un error de página no controlada: # ./test_tag [130.640218] CPU 0 No se puede manejar la solicitud de paginación del kernel en virtual dirección ffff80001b898004, era == 9000000003137f7c, ra == 9000000003139e70 [ 130.640501] Ups[#3]: [ 130.640553] CPU: 0 PID: 1326 Comm: test_tag Contaminado: GDO 6.7.0-rc4 -loong-devel-gb62ab1a397cf #47 61985c1d94084daa2432f771daa45b56b10d8d2a [130.640764] Nombre de hardware: QEMU QEMU Máquina virtual, BIOS desconocido 2/2/2022 [ 130.640874] pc 9000000003137f7c ra 9000000003139e70 tp 9000000104cb4000 sp 9000000104cb7a40 [ 13 0.641001] a0 ffff80001b894000 a1 ffff80001b897ff8 a2 000000006ba210be a3 0000000000000000 [ 130.641128] a4 000000006ba210be a5 00000000000000 f1 a6 00000000000000b3 a7 0000000000000000 [ 130.641256] t0 00000000000000000 t1 00000000000007f6 t2 00000000000000000 t3 9000000004091b70 [ 130.641387] t4 00 0000006ba210be t5 0000000000000004 t6 ffffffffffffffff0 t7 90000000040913e0 [ 130.641512] t8 00000000000000005 u0 0000000000000dc0 s9 000000000 0000009 s0 9000000104cb7ae0 [ 130.641641] s1 00000000000007f6 s2 0000000000000009 s3 00000000000000095 s4 0000000000000000 [ 130.6 41771] s5 ffff80001b894000 s6 ffff80001b897fb0 s7 9000000004090c50 s8 0000000000000000 [ 130.641900] ra: 9000000003139e70 build_body+0x1fcc/0x4988 [ 130.642007] ERA: 9 000000003137f7c build_body+0xd8/0x4988 [ 130.642112] CRMD: 000000b0 (PLV0 -IE -DA +PG DACF=CC DACM=CC -WE ) [ 130.642261] PRMD: 00000004 (PPLV0 +PIE -PWE) [ 130.642353] EUEN: 00000003 (+FPE +SXE -ASXE -BTE) [ 130.642458] ECFG: 00071c1c (LIE=2-4,10-12 VS=7) [ 130.642554] ESTAT: 00010000 [PIL] (IS= ECode=1 EssubCode=0) [ 130.642658] BADV: ffff80001b898004 [ 130.642719] PRID: 0014c010 (Loongson-64bit, Loongson-3A5000) [ 1 30.642815] Módulos vinculados en: [última descarga : bpf_testmod(O)] [130.642924] Procesar test_tag (pid: 1326, threadinfo=00000000f7f4015f, tarea=000000006499f9fd) [130.643062] Pila: 0000000000000000 900000000338072 4 0000000000000000 0000000104cb7be8 [ 130.643213] 0000000000000000 25af8d9b6e600558 9000000106250ea0 9000000104cb7ae0 [ 130.643378] 0 000000000000000 0000000000000000 9000000104cb7be8 90000000049f6000 [ 130.643538] 0000000000000090 9000000106250ea0 ffff80001b894000 ffff80001b894000 [ 130.643685] 00007ffffb917790 900000000313ca94 00000000000000000 0000000000000000 [ 130.643831] ffff80001b894000 0000000000000ff7 0000000000000000 9000000100468000 [ 130.643983] 00000000000000000 0000000000000000 0000000000000040 25af8d9b6e600558 [ 130.644131] 0000000000000bb7 ffff80001b894048 0000000000000000 00000000000000 000 [ 130.644276] 9000000104cb7be8 90000000049f6000 0000000000000090 9000000104cb7bdc [ 130.644423] ffff80001b894000 0000000000000000 0 00007ffffb917790 90000000032acfb0 [ 130.644572] . .. [ 130.644629] Seguimiento de llamadas: [ 130.644641] [<9000000003137f7c>] build_body+0xd8/0x4988 [ 130.644785] [<900000000313ca94>] bpf_int_jit_compile+0x228/0x4ec [ 1 30.644891] [<90000000032acfb0>] bpf_prog_select_runtime+0x158/0x1b0 [ 130.645003] [<90000000032b3504>] bpf_prog_load+0x760/0xb44 [ 130.645089] [<90000000032b6744>] __sys_bpf+0xbb8/0x2588 [ 130.645175] [<90000000032b838 8>] sys_bpf+0x20/0x2c [ 130.645259] [<9000000003f6ab38>] do_syscall+0x7c/0x94 [ 130.645369] [<9000000003121c5c>] handle_syscall+0xbc/0x158 [ 130.645507] [ 130.645539] Código: 380839f6 380831f9 28412bae <24000ca6> 004081ad 0014 cb50 004083e8 02bff34c 58008e91 [ 130.645729] [ 130.646418] ---[ final de seguimiento 0000000000000000 ]--- En mi máquina, que tiene CONFIG_PAGE_SIZE_16KB=y, la prueba falló al cargar un programa BPF con 2039 instrucciones: prog = (struct bpf_prog *)ffff80001b894000 insn = (struct bpf_insn *)(prog->insnsi)fff ---truncado---