CVE-2023-52621

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> bpf: Check rcu_read_lock_trace_held() before calling bpf map helpers<br /> <br /> These three bpf_map_{lookup,update,delete}_elem() helpers are also<br /> available for sleepable bpf program, so add the corresponding lock<br /> assertion for sleepable bpf program, otherwise the following warning<br /> will be reported when a sleepable bpf program manipulates bpf map under<br /> interpreter mode (aka bpf_jit_enable=0):<br /> <br /> WARNING: CPU: 3 PID: 4985 at kernel/bpf/helpers.c:40 ......<br /> CPU: 3 PID: 4985 Comm: test_progs Not tainted 6.6.0+ #2<br /> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) ......<br /> RIP: 0010:bpf_map_lookup_elem+0x54/0x60<br /> ......<br /> Call Trace:<br /> <br /> ? __warn+0xa5/0x240<br /> ? bpf_map_lookup_elem+0x54/0x60<br /> ? report_bug+0x1ba/0x1f0<br /> ? handle_bug+0x40/0x80<br /> ? exc_invalid_op+0x18/0x50<br /> ? asm_exc_invalid_op+0x1b/0x20<br /> ? __pfx_bpf_map_lookup_elem+0x10/0x10<br /> ? rcu_lockdep_current_cpu_online+0x65/0xb0<br /> ? rcu_is_watching+0x23/0x50<br /> ? bpf_map_lookup_elem+0x54/0x60<br /> ? __pfx_bpf_map_lookup_elem+0x10/0x10<br /> ___bpf_prog_run+0x513/0x3b70<br /> __bpf_prog_run32+0x9d/0xd0<br /> ? __bpf_prog_enter_sleepable_recur+0xad/0x120<br /> ? __bpf_prog_enter_sleepable_recur+0x3e/0x120<br /> bpf_trampoline_6442580665+0x4d/0x1000<br /> __x64_sys_getpgid+0x5/0x30<br /> ? do_syscall_64+0x36/0xb0<br /> entry_SYSCALL_64_after_hwframe+0x6e/0x76<br />