diff options
Diffstat (limited to 'kernel/signal.c')
-rw-r--r-- | kernel/signal.c | 346 |
1 files changed, 217 insertions, 129 deletions
diff --git a/kernel/signal.c b/kernel/signal.c index edf8915ddd54..c4da1ef56fdf 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -45,6 +45,7 @@ #include <linux/posix-timers.h> #include <linux/livepatch.h> #include <linux/cgroup.h> +#include <linux/audit.h> #define CREATE_TRACE_POINTS #include <trace/events/signal.h> @@ -54,7 +55,6 @@ #include <asm/unistd.h> #include <asm/siginfo.h> #include <asm/cacheflush.h> -#include "audit.h" /* audit_signal_info() */ /* * SLAB caches for signal bits. @@ -90,6 +90,11 @@ static bool sig_task_ignored(struct task_struct *t, int sig, bool force) handler == SIG_DFL && !(force && sig_kernel_only(sig))) return true; + /* Only allow kernel generated signals to this kthread */ + if (unlikely((t->flags & PF_KTHREAD) && + (handler == SIG_KTHREAD_KERNEL) && !force)) + return true; + return sig_handler_ignored(handler, sig); } @@ -349,7 +354,7 @@ void task_clear_jobctl_pending(struct task_struct *task, unsigned long mask) * @task has %JOBCTL_STOP_PENDING set and is participating in a group stop. * Group stop states are cleared and the group stop count is consumed if * %JOBCTL_STOP_CONSUME was set. If the consumption completes the group - * stop, the appropriate %SIGNAL_* flags are set. + * stop, the appropriate `SIGNAL_*` flags are set. * * CONTEXT: * Must be called with @task->sighand->siglock held. @@ -1057,29 +1062,8 @@ static inline bool legacy_queue(struct sigpending *signals, int sig) return (sig < SIGRTMIN) && sigismember(&signals->signal, sig); } -#ifdef CONFIG_USER_NS -static inline void userns_fixup_signal_uid(struct kernel_siginfo *info, struct task_struct *t) -{ - if (current_user_ns() == task_cred_xxx(t, user_ns)) - return; - - if (SI_FROMKERNEL(info)) - return; - - rcu_read_lock(); - info->si_uid = from_kuid_munged(task_cred_xxx(t, user_ns), - make_kuid(current_user_ns(), info->si_uid)); - rcu_read_unlock(); -} -#else -static inline void userns_fixup_signal_uid(struct kernel_siginfo *info, struct task_struct *t) -{ - return; -} -#endif - static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struct *t, - enum pid_type type, int from_ancestor_ns) + enum pid_type type, bool force) { struct sigpending *pending; struct sigqueue *q; @@ -1089,8 +1073,7 @@ static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struc assert_spin_locked(&t->sighand->siglock); result = TRACE_SIGNAL_IGNORED; - if (!prepare_signal(sig, t, - from_ancestor_ns || (info == SEND_SIG_PRIV))) + if (!prepare_signal(sig, t, force)) goto ret; pending = (type != PIDTYPE_PID) ? &t->signal->shared_pending : &t->pending; @@ -1135,7 +1118,11 @@ static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struc q->info.si_code = SI_USER; q->info.si_pid = task_tgid_nr_ns(current, task_active_pid_ns(t)); - q->info.si_uid = from_kuid_munged(current_user_ns(), current_uid()); + rcu_read_lock(); + q->info.si_uid = + from_kuid_munged(task_cred_xxx(t, user_ns), + current_uid()); + rcu_read_unlock(); break; case (unsigned long) SEND_SIG_PRIV: clear_siginfo(&q->info); @@ -1147,30 +1134,24 @@ static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struc break; default: copy_siginfo(&q->info, info); - if (from_ancestor_ns) - q->info.si_pid = 0; break; } - - userns_fixup_signal_uid(&q->info, t); - - } else if (!is_si_special(info)) { - if (sig >= SIGRTMIN && info->si_code != SI_USER) { - /* - * Queue overflow, abort. We may abort if the - * signal was rt and sent by user using something - * other than kill(). - */ - result = TRACE_SIGNAL_OVERFLOW_FAIL; - ret = -EAGAIN; - goto ret; - } else { - /* - * This is a silent loss of information. We still - * send the signal, but the *info bits are lost. - */ - result = TRACE_SIGNAL_LOSE_INFO; - } + } else if (!is_si_special(info) && + sig >= SIGRTMIN && info->si_code != SI_USER) { + /* + * Queue overflow, abort. We may abort if the + * signal was rt and sent by user using something + * other than kill(). + */ + result = TRACE_SIGNAL_OVERFLOW_FAIL; + ret = -EAGAIN; + goto ret; + } else { + /* + * This is a silent loss of information. We still + * send the signal, but the *info bits are lost. + */ + result = TRACE_SIGNAL_LOSE_INFO; } out_set: @@ -1197,17 +1178,62 @@ ret: return ret; } +static inline bool has_si_pid_and_uid(struct kernel_siginfo *info) +{ + bool ret = false; + switch (siginfo_layout(info->si_signo, info->si_code)) { + case SIL_KILL: + case SIL_CHLD: + case SIL_RT: + ret = true; + break; + case SIL_TIMER: + case SIL_POLL: + case SIL_FAULT: + case SIL_FAULT_MCEERR: + case SIL_FAULT_BNDERR: + case SIL_FAULT_PKUERR: + case SIL_SYS: + ret = false; + break; + } + return ret; +} + static int send_signal(int sig, struct kernel_siginfo *info, struct task_struct *t, enum pid_type type) { - int from_ancestor_ns = 0; + /* Should SIGKILL or SIGSTOP be received by a pid namespace init? */ + bool force = false; -#ifdef CONFIG_PID_NS - from_ancestor_ns = si_fromuser(info) && - !task_pid_nr_ns(current, task_active_pid_ns(t)); -#endif + if (info == SEND_SIG_NOINFO) { + /* Force if sent from an ancestor pid namespace */ + force = !task_pid_nr_ns(current, task_active_pid_ns(t)); + } else if (info == SEND_SIG_PRIV) { + /* Don't ignore kernel generated signals */ + force = true; + } else if (has_si_pid_and_uid(info)) { + /* SIGKILL and SIGSTOP is special or has ids */ + struct user_namespace *t_user_ns; - return __send_signal(sig, info, t, type, from_ancestor_ns); + rcu_read_lock(); + t_user_ns = task_cred_xxx(t, user_ns); + if (current_user_ns() != t_user_ns) { + kuid_t uid = make_kuid(current_user_ns(), info->si_uid); + info->si_uid = from_kuid_munged(t_user_ns, uid); + } + rcu_read_unlock(); + + /* A kernel generated signal? */ + force = (info->si_code == SI_KERNEL); + + /* From an ancestor pid namespace? */ + if (!task_pid_nr_ns(current, task_active_pid_ns(t))) { + info->si_pid = 0; + force = true; + } + } + return __send_signal(sig, info, t, type, force); } static void print_fatal_signal(int signr) @@ -1274,12 +1300,13 @@ int do_send_sig_info(int sig, struct kernel_siginfo *info, struct task_struct *p * We don't want to have recursive SIGSEGV's etc, for example, * that is why we also clear SIGNAL_UNKILLABLE. */ -int -force_sig_info(int sig, struct kernel_siginfo *info, struct task_struct *t) +static int +force_sig_info_to_task(struct kernel_siginfo *info, struct task_struct *t) { unsigned long int flags; int ret, blocked, ignored; struct k_sigaction *action; + int sig = info->si_signo; spin_lock_irqsave(&t->sighand->siglock, flags); action = &t->sighand->action[sig-1]; @@ -1304,6 +1331,11 @@ force_sig_info(int sig, struct kernel_siginfo *info, struct task_struct *t) return ret; } +int force_sig_info(struct kernel_siginfo *info) +{ + return force_sig_info_to_task(info, current); +} + /* * Nuke all other threads in the group. */ @@ -1440,13 +1472,44 @@ static inline bool kill_as_cred_perm(const struct cred *cred, uid_eq(cred->uid, pcred->uid); } -/* like kill_pid_info(), but doesn't use uid/euid of "current" */ -int kill_pid_info_as_cred(int sig, struct kernel_siginfo *info, struct pid *pid, - const struct cred *cred) +/* + * The usb asyncio usage of siginfo is wrong. The glibc support + * for asyncio which uses SI_ASYNCIO assumes the layout is SIL_RT. + * AKA after the generic fields: + * kernel_pid_t si_pid; + * kernel_uid32_t si_uid; + * sigval_t si_value; + * + * Unfortunately when usb generates SI_ASYNCIO it assumes the layout + * after the generic fields is: + * void __user *si_addr; + * + * This is a practical problem when there is a 64bit big endian kernel + * and a 32bit userspace. As the 32bit address will encoded in the low + * 32bits of the pointer. Those low 32bits will be stored at higher + * address than appear in a 32 bit pointer. So userspace will not + * see the address it was expecting for it's completions. + * + * There is nothing in the encoding that can allow + * copy_siginfo_to_user32 to detect this confusion of formats, so + * handle this by requiring the caller of kill_pid_usb_asyncio to + * notice when this situration takes place and to store the 32bit + * pointer in sival_int, instead of sival_addr of the sigval_t addr + * parameter. + */ +int kill_pid_usb_asyncio(int sig, int errno, sigval_t addr, + struct pid *pid, const struct cred *cred) { - int ret = -EINVAL; + struct kernel_siginfo info; struct task_struct *p; unsigned long flags; + int ret = -EINVAL; + + clear_siginfo(&info); + info.si_signo = sig; + info.si_errno = errno; + info.si_code = SI_ASYNCIO; + *((sigval_t *)&info.si_pid) = addr; if (!valid_signal(sig)) return ret; @@ -1457,17 +1520,17 @@ int kill_pid_info_as_cred(int sig, struct kernel_siginfo *info, struct pid *pid, ret = -ESRCH; goto out_unlock; } - if (si_fromuser(info) && !kill_as_cred_perm(cred, p)) { + if (!kill_as_cred_perm(cred, p)) { ret = -EPERM; goto out_unlock; } - ret = security_task_kill(p, info, sig, cred); + ret = security_task_kill(p, &info, sig, cred); if (ret) goto out_unlock; if (sig) { if (lock_task_sighand(p, &flags)) { - ret = __send_signal(sig, info, p, PIDTYPE_TGID, 0); + ret = __send_signal(sig, &info, p, PIDTYPE_TGID, false); unlock_task_sighand(p, &flags); } else ret = -ESRCH; @@ -1476,7 +1539,7 @@ out_unlock: rcu_read_unlock(); return ret; } -EXPORT_SYMBOL_GPL(kill_pid_info_as_cred); +EXPORT_SYMBOL_GPL(kill_pid_usb_asyncio); /* * kill_something_info() interprets pid in interesting ways just like kill(2). @@ -1552,9 +1615,17 @@ send_sig(int sig, struct task_struct *p, int priv) } EXPORT_SYMBOL(send_sig); -void force_sig(int sig, struct task_struct *p) +void force_sig(int sig) { - force_sig_info(sig, SEND_SIG_PRIV, p); + struct kernel_siginfo info; + + clear_siginfo(&info); + info.si_signo = sig; + info.si_errno = 0; + info.si_code = SI_KERNEL; + info.si_pid = 0; + info.si_uid = 0; + force_sig_info(&info); } EXPORT_SYMBOL(force_sig); @@ -1564,18 +1635,20 @@ EXPORT_SYMBOL(force_sig); * the problem was already a SIGSEGV, we'll want to * make sure we don't even try to deliver the signal.. */ -void force_sigsegv(int sig, struct task_struct *p) +void force_sigsegv(int sig) { + struct task_struct *p = current; + if (sig == SIGSEGV) { unsigned long flags; spin_lock_irqsave(&p->sighand->siglock, flags); p->sighand->action[sig - 1].sa.sa_handler = SIG_DFL; spin_unlock_irqrestore(&p->sighand->siglock, flags); } - force_sig(SIGSEGV, p); + force_sig(SIGSEGV); } -int force_sig_fault(int sig, int code, void __user *addr +int force_sig_fault_to_task(int sig, int code, void __user *addr ___ARCH_SI_TRAPNO(int trapno) ___ARCH_SI_IA64(int imm, unsigned int flags, unsigned long isr) , struct task_struct *t) @@ -1595,7 +1668,16 @@ int force_sig_fault(int sig, int code, void __user *addr info.si_flags = flags; info.si_isr = isr; #endif - return force_sig_info(info.si_signo, &info, t); + return force_sig_info_to_task(&info, t); +} + +int force_sig_fault(int sig, int code, void __user *addr + ___ARCH_SI_TRAPNO(int trapno) + ___ARCH_SI_IA64(int imm, unsigned int flags, unsigned long isr)) +{ + return force_sig_fault_to_task(sig, code, addr + ___ARCH_SI_TRAPNO(trapno) + ___ARCH_SI_IA64(imm, flags, isr), current); } int send_sig_fault(int sig, int code, void __user *addr @@ -1621,7 +1703,7 @@ int send_sig_fault(int sig, int code, void __user *addr return send_sig_info(info.si_signo, &info, t); } -int force_sig_mceerr(int code, void __user *addr, short lsb, struct task_struct *t) +int force_sig_mceerr(int code, void __user *addr, short lsb) { struct kernel_siginfo info; @@ -1632,7 +1714,7 @@ int force_sig_mceerr(int code, void __user *addr, short lsb, struct task_struct info.si_code = code; info.si_addr = addr; info.si_addr_lsb = lsb; - return force_sig_info(info.si_signo, &info, t); + return force_sig_info(&info); } int send_sig_mceerr(int code, void __user *addr, short lsb, struct task_struct *t) @@ -1661,7 +1743,7 @@ int force_sig_bnderr(void __user *addr, void __user *lower, void __user *upper) info.si_addr = addr; info.si_lower = lower; info.si_upper = upper; - return force_sig_info(info.si_signo, &info, current); + return force_sig_info(&info); } #ifdef SEGV_PKUERR @@ -1675,7 +1757,7 @@ int force_sig_pkuerr(void __user *addr, u32 pkey) info.si_code = SEGV_PKUERR; info.si_addr = addr; info.si_pkey = pkey; - return force_sig_info(info.si_signo, &info, current); + return force_sig_info(&info); } #endif @@ -1691,7 +1773,7 @@ int force_sig_ptrace_errno_trap(int errno, void __user *addr) info.si_errno = errno; info.si_code = TRAP_HWBKPT; info.si_addr = addr; - return force_sig_info(info.si_signo, &info, current); + return force_sig_info(&info); } int kill_pgrp(struct pid *pid, int sig, int priv) @@ -1804,6 +1886,15 @@ ret: return ret; } +static void do_notify_pidfd(struct task_struct *task) +{ + struct pid *pid; + + WARN_ON(task->exit_state == 0); + pid = task_pid(task); + wake_up_all(&pid->wait_pidfd); +} + /* * Let a parent know about the death of a child. * For a stopped/continued status change, use do_notify_parent_cldstop instead. @@ -1827,6 +1918,9 @@ bool do_notify_parent(struct task_struct *tsk, int sig) BUG_ON(!tsk->ptrace && (tsk->group_leader != tsk || !thread_group_empty(tsk))); + /* Wake up all pidfd waiters */ + do_notify_pidfd(tsk); + if (sig != SIGCHLD) { /* * This is only possible if parent == real_parent. @@ -2676,7 +2770,7 @@ static void signal_delivered(struct ksignal *ksig, int stepping) void signal_setup_done(int failed, struct ksignal *ksig, int stepping) { if (failed) - force_sigsegv(ksig->sig, current); + force_sigsegv(ksig->sig); else signal_delivered(ksig, stepping); } @@ -2863,80 +2957,49 @@ EXPORT_SYMBOL(sigprocmask); * * This is useful for syscalls such as ppoll, pselect, io_pgetevents and * epoll_pwait where a new sigmask is passed from userland for the syscalls. + * + * Note that it does set_restore_sigmask() in advance, so it must be always + * paired with restore_saved_sigmask_unless() before return from syscall. */ -int set_user_sigmask(const sigset_t __user *usigmask, sigset_t *set, - sigset_t *oldset, size_t sigsetsize) +int set_user_sigmask(const sigset_t __user *umask, size_t sigsetsize) { - if (!usigmask) - return 0; + sigset_t kmask; + if (!umask) + return 0; if (sigsetsize != sizeof(sigset_t)) return -EINVAL; - if (copy_from_user(set, usigmask, sizeof(sigset_t))) + if (copy_from_user(&kmask, umask, sizeof(sigset_t))) return -EFAULT; - *oldset = current->blocked; - set_current_blocked(set); + set_restore_sigmask(); + current->saved_sigmask = current->blocked; + set_current_blocked(&kmask); return 0; } -EXPORT_SYMBOL(set_user_sigmask); #ifdef CONFIG_COMPAT -int set_compat_user_sigmask(const compat_sigset_t __user *usigmask, - sigset_t *set, sigset_t *oldset, +int set_compat_user_sigmask(const compat_sigset_t __user *umask, size_t sigsetsize) { - if (!usigmask) - return 0; + sigset_t kmask; + if (!umask) + return 0; if (sigsetsize != sizeof(compat_sigset_t)) return -EINVAL; - if (get_compat_sigset(set, usigmask)) + if (get_compat_sigset(&kmask, umask)) return -EFAULT; - *oldset = current->blocked; - set_current_blocked(set); + set_restore_sigmask(); + current->saved_sigmask = current->blocked; + set_current_blocked(&kmask); return 0; } -EXPORT_SYMBOL(set_compat_user_sigmask); #endif -/* - * restore_user_sigmask: - * usigmask: sigmask passed in from userland. - * sigsaved: saved sigmask when the syscall started and changed the sigmask to - * usigmask. - * - * This is useful for syscalls such as ppoll, pselect, io_pgetevents and - * epoll_pwait where a new sigmask is passed in from userland for the syscalls. - */ -void restore_user_sigmask(const void __user *usigmask, sigset_t *sigsaved, - bool interrupted) -{ - - if (!usigmask) - return; - /* - * When signals are pending, do not restore them here. - * Restoring sigmask here can lead to delivering signals that the above - * syscalls are intended to block because of the sigmask passed in. - */ - if (interrupted) { - current->saved_sigmask = *sigsaved; - set_restore_sigmask(); - return; - } - - /* - * This is needed because the fast syscall return path does not restore - * saved_sigmask when signals are not pending. - */ - set_current_blocked(sigsaved); -} -EXPORT_SYMBOL(restore_user_sigmask); - /** * sys_rt_sigprocmask - change the list of currently blocked signals * @how: whether to add, remove, or set signals @@ -3615,8 +3678,11 @@ static int copy_siginfo_from_user_any(kernel_siginfo_t *kinfo, siginfo_t *info) static struct pid *pidfd_to_pid(const struct file *file) { - if (file->f_op == &pidfd_fops) - return file->private_data; + struct pid *pid; + + pid = pidfd_pid(file); + if (!IS_ERR(pid)) + return pid; return tgid_pidfd_to_pid(file); } @@ -4477,6 +4543,28 @@ static inline void siginfo_buildtime_checks(void) CHECK_OFFSET(si_syscall); CHECK_OFFSET(si_arch); #undef CHECK_OFFSET + + /* usb asyncio */ + BUILD_BUG_ON(offsetof(struct siginfo, si_pid) != + offsetof(struct siginfo, si_addr)); + if (sizeof(int) == sizeof(void __user *)) { + BUILD_BUG_ON(sizeof_field(struct siginfo, si_pid) != + sizeof(void __user *)); + } else { + BUILD_BUG_ON((sizeof_field(struct siginfo, si_pid) + + sizeof_field(struct siginfo, si_uid)) != + sizeof(void __user *)); + BUILD_BUG_ON(offsetofend(struct siginfo, si_pid) != + offsetof(struct siginfo, si_uid)); + } +#ifdef CONFIG_COMPAT + BUILD_BUG_ON(offsetof(struct compat_siginfo, si_pid) != + offsetof(struct compat_siginfo, si_addr)); + BUILD_BUG_ON(sizeof_field(struct compat_siginfo, si_pid) != + sizeof(compat_uptr_t)); + BUILD_BUG_ON(sizeof_field(struct compat_siginfo, si_pid) != + sizeof_field(struct siginfo, si_pid)); +#endif } void __init signals_init(void) |