^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0-only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * linux/kernel/fork.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Copyright (C) 1991, 1992 Linus Torvalds
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) * 'fork.c' contains the help-routines for the 'fork' system call
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) * (see also entry.S and others).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) * Fork is rather simple, once you get the hang of it, but the memory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) * management can be a bitch. See 'mm/memory.c': 'copy_page_range()'
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/anon_inodes.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/sched/autogroup.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/sched/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/sched/coredump.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/sched/user.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/sched/numa_balancing.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/sched/stat.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/sched/task.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/sched/task_stack.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/sched/cputime.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <linux/seq_file.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <linux/rtmutex.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include <linux/unistd.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include <linux/vmalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #include <linux/completion.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) #include <linux/personality.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) #include <linux/mempolicy.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) #include <linux/sem.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) #include <linux/file.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) #include <linux/fdtable.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) #include <linux/iocontext.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) #include <linux/key.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #include <linux/binfmts.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) #include <linux/mman.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) #include <linux/mmu_notifier.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) #include <linux/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) #include <linux/vmacache.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) #include <linux/nsproxy.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) #include <linux/capability.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) #include <linux/cpu.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) #include <linux/cgroup.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) #include <linux/security.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) #include <linux/hugetlb.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) #include <linux/seccomp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) #include <linux/swap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) #include <linux/syscalls.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) #include <linux/jiffies.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) #include <linux/futex.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) #include <linux/compat.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) #include <linux/kthread.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) #include <linux/task_io_accounting_ops.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) #include <linux/rcupdate.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) #include <linux/ptrace.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) #include <linux/mount.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) #include <linux/audit.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) #include <linux/memcontrol.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) #include <linux/ftrace.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) #include <linux/proc_fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) #include <linux/profile.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) #include <linux/rmap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) #include <linux/ksm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) #include <linux/acct.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) #include <linux/userfaultfd_k.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) #include <linux/tsacct_kern.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) #include <linux/cn_proc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) #include <linux/freezer.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) #include <linux/delayacct.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) #include <linux/taskstats_kern.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) #include <linux/random.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) #include <linux/tty.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) #include <linux/blkdev.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) #include <linux/fs_struct.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) #include <linux/magic.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) #include <linux/perf_event.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) #include <linux/posix-timers.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) #include <linux/user-return-notifier.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) #include <linux/oom.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) #include <linux/khugepaged.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) #include <linux/signalfd.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) #include <linux/uprobes.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) #include <linux/aio.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) #include <linux/compiler.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) #include <linux/sysctl.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) #include <linux/kcov.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) #include <linux/livepatch.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) #include <linux/thread_info.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) #include <linux/stackleak.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) #include <linux/kasan.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) #include <linux/scs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) #include <linux/io_uring.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) #include <linux/cpufreq_times.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) #include <asm/pgalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) #include <linux/uaccess.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) #include <asm/mmu_context.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) #include <asm/cacheflush.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) #include <asm/tlbflush.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) #include <trace/events/sched.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) #define CREATE_TRACE_POINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) #include <trace/events/task.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) #undef CREATE_TRACE_POINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) #include <trace/hooks/sched.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) * Minimum number of threads to boot the kernel
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) #define MIN_THREADS 20
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) * Maximum number of threads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) #define MAX_THREADS FUTEX_TID_MASK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) EXPORT_TRACEPOINT_SYMBOL_GPL(task_newtask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) * Protected counters by write_lock_irq(&tasklist_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) unsigned long total_forks; /* Handle normal Linux uptimes. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) int nr_threads; /* The idle threads do not count.. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) static int max_threads; /* tunable limit on nr_threads */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) #define NAMED_ARRAY_INDEX(x) [x] = __stringify(x)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) static const char * const resident_page_types[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) NAMED_ARRAY_INDEX(MM_FILEPAGES),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) NAMED_ARRAY_INDEX(MM_ANONPAGES),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) NAMED_ARRAY_INDEX(MM_SWAPENTS),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) NAMED_ARRAY_INDEX(MM_SHMEMPAGES),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) DEFINE_PER_CPU(unsigned long, process_counts) = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) __cacheline_aligned DEFINE_RWLOCK(tasklist_lock); /* outer */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) EXPORT_SYMBOL_GPL(tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) #ifdef CONFIG_PROVE_RCU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) int lockdep_tasklist_lock_is_held(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) return lockdep_is_held(&tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) EXPORT_SYMBOL_GPL(lockdep_tasklist_lock_is_held);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) #endif /* #ifdef CONFIG_PROVE_RCU */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) int nr_processes(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) int cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) int total = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) for_each_possible_cpu(cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) total += per_cpu(process_counts, cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) return total;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) void __weak arch_release_task_struct(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) static struct kmem_cache *task_struct_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) static inline struct task_struct *alloc_task_struct_node(int node)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) return kmem_cache_alloc_node(task_struct_cachep, GFP_KERNEL, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) static inline void free_task_struct(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) kmem_cache_free(task_struct_cachep, tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) #ifndef CONFIG_ARCH_THREAD_STACK_ALLOCATOR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) * kmemcache based allocator.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) # if THREAD_SIZE >= PAGE_SIZE || defined(CONFIG_VMAP_STACK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) * vmalloc() is a bit slow, and calling vfree() enough times will force a TLB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) * flush. Try to minimize the number of calls by caching stacks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) #define NR_CACHED_STACKS 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) static DEFINE_PER_CPU(struct vm_struct *, cached_stacks[NR_CACHED_STACKS]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) static int free_vm_stack_cache(unsigned int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) struct vm_struct **cached_vm_stacks = per_cpu_ptr(cached_stacks, cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) for (i = 0; i < NR_CACHED_STACKS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) struct vm_struct *vm_stack = cached_vm_stacks[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) if (!vm_stack)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) vfree(vm_stack->addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) cached_vm_stacks[i] = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) static unsigned long *alloc_thread_stack_node(struct task_struct *tsk, int node)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) void *stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) for (i = 0; i < NR_CACHED_STACKS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) struct vm_struct *s;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) s = this_cpu_xchg(cached_stacks[i], NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) if (!s)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) /* Mark stack accessible for KASAN. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) kasan_unpoison_range(s->addr, THREAD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) /* Clear stale pointers from reused stack. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) memset(s->addr, 0, THREAD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) tsk->stack_vm_area = s;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) tsk->stack = s->addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) return s->addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) * Allocated stacks are cached and later reused by new threads,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) * so memcg accounting is performed manually on assigning/releasing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) * stacks to tasks. Drop __GFP_ACCOUNT.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) stack = __vmalloc_node_range(THREAD_SIZE, THREAD_ALIGN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) VMALLOC_START, VMALLOC_END,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) THREADINFO_GFP & ~__GFP_ACCOUNT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) PAGE_KERNEL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) 0, node, __builtin_return_address(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) * We can't call find_vm_area() in interrupt context, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) * free_thread_stack() can be called in interrupt context,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) * so cache the vm_struct.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) if (stack) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) tsk->stack_vm_area = find_vm_area(stack);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) tsk->stack = stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) return stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) struct page *page = alloc_pages_node(node, THREADINFO_GFP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) THREAD_SIZE_ORDER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) if (likely(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) tsk->stack = kasan_reset_tag(page_address(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) return tsk->stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) static inline void free_thread_stack(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) struct vm_struct *vm = task_stack_vm_area(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) if (vm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) for (i = 0; i < THREAD_SIZE / PAGE_SIZE; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) memcg_kmem_uncharge_page(vm->pages[i], 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) for (i = 0; i < NR_CACHED_STACKS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) if (this_cpu_cmpxchg(cached_stacks[i],
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) NULL, tsk->stack_vm_area) != NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) vfree_atomic(tsk->stack);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) __free_pages(virt_to_page(tsk->stack), THREAD_SIZE_ORDER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) # else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) static struct kmem_cache *thread_stack_cache;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) static unsigned long *alloc_thread_stack_node(struct task_struct *tsk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) int node)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) unsigned long *stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) stack = kmem_cache_alloc_node(thread_stack_cache, THREADINFO_GFP, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) stack = kasan_reset_tag(stack);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) tsk->stack = stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) return stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) static void free_thread_stack(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) kmem_cache_free(thread_stack_cache, tsk->stack);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) void thread_stack_cache_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) thread_stack_cache = kmem_cache_create_usercopy("thread_stack",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) THREAD_SIZE, THREAD_SIZE, 0, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) THREAD_SIZE, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) BUG_ON(thread_stack_cache == NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) # endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) /* SLAB cache for signal_struct structures (tsk->signal) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) static struct kmem_cache *signal_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) /* SLAB cache for sighand_struct structures (tsk->sighand) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) struct kmem_cache *sighand_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) /* SLAB cache for files_struct structures (tsk->files) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) struct kmem_cache *files_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) /* SLAB cache for fs_struct structures (tsk->fs) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) struct kmem_cache *fs_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) /* SLAB cache for vm_area_struct structures */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) static struct kmem_cache *vm_area_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) /* SLAB cache for mm_struct structures (tsk->mm) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) static struct kmem_cache *mm_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) struct vm_area_struct *vm_area_alloc(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) struct vm_area_struct *vma;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) if (vma)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) vma_init(vma, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) return vma;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) struct vm_area_struct *new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) if (new) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) ASSERT_EXCLUSIVE_WRITER(orig->vm_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) ASSERT_EXCLUSIVE_WRITER(orig->vm_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) * orig->shared.rb may be modified concurrently, but the clone
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) * will be reinitialized.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) *new = data_race(*orig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) INIT_VMA(new);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) new->vm_next = new->vm_prev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) return new;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) void vm_area_free(struct vm_area_struct *vma)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) kmem_cache_free(vm_area_cachep, vma);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) static void account_kernel_stack(struct task_struct *tsk, int account)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) void *stack = task_stack_page(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) struct vm_struct *vm = task_stack_vm_area(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) /* All stack pages are in the same node. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) if (vm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) mod_lruvec_page_state(vm->pages[0], NR_KERNEL_STACK_KB,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) account * (THREAD_SIZE / 1024));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) mod_lruvec_slab_state(stack, NR_KERNEL_STACK_KB,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) account * (THREAD_SIZE / 1024));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) static int memcg_charge_kernel_stack(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) struct vm_struct *vm = task_stack_vm_area(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) BUILD_BUG_ON(IS_ENABLED(CONFIG_VMAP_STACK) && PAGE_SIZE % 1024 != 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) if (vm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) BUG_ON(vm->nr_pages != THREAD_SIZE / PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) for (i = 0; i < THREAD_SIZE / PAGE_SIZE; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) * If memcg_kmem_charge_page() fails, page->mem_cgroup
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) * pointer is NULL, and memcg_kmem_uncharge_page() in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) * free_thread_stack() will ignore this page.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) ret = memcg_kmem_charge_page(vm->pages[i], GFP_KERNEL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) static void release_task_stack(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) if (WARN_ON(tsk->state != TASK_DEAD))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) return; /* Better to leak the stack than to free prematurely */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) account_kernel_stack(tsk, -1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) free_thread_stack(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) tsk->stack = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) tsk->stack_vm_area = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) #ifdef CONFIG_THREAD_INFO_IN_TASK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) void put_task_stack(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) if (refcount_dec_and_test(&tsk->stack_refcount))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) release_task_stack(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) EXPORT_SYMBOL_GPL(put_task_stack);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) void free_task(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) cpufreq_task_times_exit(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) scs_release(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) trace_android_vh_free_task(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) #ifndef CONFIG_THREAD_INFO_IN_TASK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) * The task is finally done with both the stack and thread_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) * so free both.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) release_task_stack(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) * If the task had a separate stack allocation, it should be gone
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) * by now.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) WARN_ON_ONCE(refcount_read(&tsk->stack_refcount) != 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) rt_mutex_debug_task_free(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) ftrace_graph_exit_task(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) arch_release_task_struct(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) if (tsk->flags & PF_KTHREAD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) free_kthread_struct(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) free_task_struct(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) EXPORT_SYMBOL(free_task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) #ifdef CONFIG_MMU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) static __latent_entropy int dup_mmap(struct mm_struct *mm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) struct mm_struct *oldmm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) struct vm_area_struct *mpnt, *tmp, *prev, **pprev, *last = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) struct rb_node **rb_link, *rb_parent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) int retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) unsigned long charge;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) LIST_HEAD(uf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) uprobe_start_dup_mmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) if (mmap_write_lock_killable(oldmm)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) retval = -EINTR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) goto fail_uprobe_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) flush_cache_dup_mm(oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) uprobe_dup_mmap(oldmm, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) * Not linked in yet - no deadlock potential:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) mmap_write_lock_nested(mm, SINGLE_DEPTH_NESTING);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) /* No ordering required: file already has been exposed. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) RCU_INIT_POINTER(mm->exe_file, get_mm_exe_file(oldmm));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) mm->total_vm = oldmm->total_vm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) mm->data_vm = oldmm->data_vm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) mm->exec_vm = oldmm->exec_vm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) mm->stack_vm = oldmm->stack_vm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) rb_link = &mm->mm_rb.rb_node;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) rb_parent = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) pprev = &mm->mmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) retval = ksm_fork(mm, oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) retval = khugepaged_fork(mm, oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) prev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) if (mpnt->vm_flags & VM_DONTCOPY) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) vm_stat_account(mm, mpnt->vm_flags, -vma_pages(mpnt));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) charge = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) * Don't duplicate many vmas if we've been oom-killed (for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) * example)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) if (fatal_signal_pending(current)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) retval = -EINTR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) if (mpnt->vm_flags & VM_ACCOUNT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) unsigned long len = vma_pages(mpnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) if (security_vm_enough_memory_mm(oldmm, len)) /* sic */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) goto fail_nomem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) charge = len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) tmp = vm_area_dup(mpnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) if (!tmp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) goto fail_nomem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) retval = vma_dup_policy(mpnt, tmp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) goto fail_nomem_policy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) tmp->vm_mm = mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) retval = dup_userfaultfd(tmp, &uf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) goto fail_nomem_anon_vma_fork;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) if (tmp->vm_flags & VM_WIPEONFORK) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) * VM_WIPEONFORK gets a clean slate in the child.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) * Don't prepare anon_vma until fault since we don't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) * copy page for current vma.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) tmp->anon_vma = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) } else if (anon_vma_fork(tmp, mpnt))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) goto fail_nomem_anon_vma_fork;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) tmp->vm_flags &= ~(VM_LOCKED | VM_LOCKONFAULT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) file = tmp->vm_file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) if (file) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) struct inode *inode = file_inode(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) struct address_space *mapping = file->f_mapping;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) get_file(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) if (tmp->vm_flags & VM_DENYWRITE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) put_write_access(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) i_mmap_lock_write(mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) if (tmp->vm_flags & VM_SHARED)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) mapping_allow_writable(mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) flush_dcache_mmap_lock(mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) /* insert tmp into the share list, just after mpnt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) vma_interval_tree_insert_after(tmp, mpnt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) &mapping->i_mmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) flush_dcache_mmap_unlock(mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) i_mmap_unlock_write(mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) * Clear hugetlb-related page reserves for children. This only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) * affects MAP_PRIVATE mappings. Faults generated by the child
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) * are not guaranteed to succeed, even if read-only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) if (is_vm_hugetlb_page(tmp))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) reset_vma_resv_huge_pages(tmp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) * Link in the new vma and copy the page table entries.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) *pprev = tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) pprev = &tmp->vm_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) tmp->vm_prev = prev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) prev = tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) __vma_link_rb(mm, tmp, rb_link, rb_parent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) rb_link = &tmp->vm_rb.rb_right;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) rb_parent = &tmp->vm_rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) mm->map_count++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) if (!(tmp->vm_flags & VM_WIPEONFORK)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) if (IS_ENABLED(CONFIG_SPECULATIVE_PAGE_FAULT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) * Mark this VMA as changing to prevent the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) * speculative page fault hanlder to process
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) * it until the TLB are flushed below.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) last = mpnt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) vm_write_begin(mpnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) retval = copy_page_range(tmp, mpnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) if (tmp->vm_ops && tmp->vm_ops->open)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) tmp->vm_ops->open(tmp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) /* a new mm has just been created */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) retval = arch_dup_mmap(oldmm, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) mmap_write_unlock(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) flush_tlb_mm(oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) if (IS_ENABLED(CONFIG_SPECULATIVE_PAGE_FAULT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) * Since the TLB has been flush, we can safely unmark the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) * copied VMAs and allows the speculative page fault handler to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) * process them again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) * Walk back the VMA list from the last marked VMA.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) for (; last; last = last->vm_prev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) if (last->vm_flags & VM_DONTCOPY)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) if (!(last->vm_flags & VM_WIPEONFORK))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) vm_write_end(last);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) mmap_write_unlock(oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) dup_userfaultfd_complete(&uf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) fail_uprobe_end:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) uprobe_end_dup_mmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) return retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) fail_nomem_anon_vma_fork:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) mpol_put(vma_policy(tmp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) fail_nomem_policy:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) vm_area_free(tmp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) fail_nomem:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) retval = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) vm_unacct_memory(charge);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) static inline int mm_alloc_pgd(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) mm->pgd = pgd_alloc(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) if (unlikely(!mm->pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) static inline void mm_free_pgd(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) pgd_free(mm, mm->pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) mmap_write_lock(oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) RCU_INIT_POINTER(mm->exe_file, get_mm_exe_file(oldmm));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) mmap_write_unlock(oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) #define mm_alloc_pgd(mm) (0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) #define mm_free_pgd(mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) #endif /* CONFIG_MMU */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) static void check_mm(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) BUILD_BUG_ON_MSG(ARRAY_SIZE(resident_page_types) != NR_MM_COUNTERS,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) "Please make sure 'struct resident_page_types[]' is updated as well");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) for (i = 0; i < NR_MM_COUNTERS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) long x = atomic_long_read(&mm->rss_stat.count[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) if (unlikely(x))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) pr_alert("BUG: Bad rss-counter state mm:%p type:%s val:%ld\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) mm, resident_page_types[i], x);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) if (mm_pgtables_bytes(mm))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) pr_alert("BUG: non-zero pgtables_bytes on freeing mm: %ld\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) mm_pgtables_bytes(mm));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) #if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) VM_BUG_ON_MM(mm->pmd_huge_pte, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) #define allocate_mm() (kmem_cache_alloc(mm_cachep, GFP_KERNEL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) #define free_mm(mm) (kmem_cache_free(mm_cachep, (mm)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) * Called when the last reference to the mm
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) * is dropped: either by a lazy thread or by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) * mmput. Free the page directory and the mm.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) void __mmdrop(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) BUG_ON(mm == &init_mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) WARN_ON_ONCE(mm == current->mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) WARN_ON_ONCE(mm == current->active_mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) mm_free_pgd(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) destroy_context(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) mmu_notifier_subscriptions_destroy(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) check_mm(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) put_user_ns(mm->user_ns);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) free_mm(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) EXPORT_SYMBOL_GPL(__mmdrop);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) static void mmdrop_async_fn(struct work_struct *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) struct mm_struct *mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) mm = container_of(work, struct mm_struct, async_put_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) __mmdrop(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) static void mmdrop_async(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) if (unlikely(atomic_dec_and_test(&mm->mm_count))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) INIT_WORK(&mm->async_put_work, mmdrop_async_fn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) schedule_work(&mm->async_put_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) static inline void free_signal_struct(struct signal_struct *sig)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) taskstats_tgid_free(sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) sched_autogroup_exit(sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) * __mmdrop is not safe to call from softirq context on x86 due to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) * pgd_dtor so postpone it to the async context
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) if (sig->oom_mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) mmdrop_async(sig->oom_mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) kmem_cache_free(signal_cachep, sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) static inline void put_signal_struct(struct signal_struct *sig)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) if (refcount_dec_and_test(&sig->sigcnt))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) free_signal_struct(sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) void __put_task_struct(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) WARN_ON(!tsk->exit_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) WARN_ON(refcount_read(&tsk->usage));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) WARN_ON(tsk == current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) io_uring_free(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) cgroup_free(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) task_numa_free(tsk, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) security_task_free(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) exit_creds(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) delayacct_tsk_free(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) put_signal_struct(tsk->signal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) if (!profile_handoff_task(tsk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) free_task(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) EXPORT_SYMBOL_GPL(__put_task_struct);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) void __init __weak arch_task_cache_init(void) { }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) * set_max_threads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) static void set_max_threads(unsigned int max_threads_suggested)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) u64 threads;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) unsigned long nr_pages = totalram_pages();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) * The number of threads shall be limited such that the thread
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) * structures may only consume a small part of the available memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) if (fls64(nr_pages) + fls64(PAGE_SIZE) > 64)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) threads = MAX_THREADS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) threads = div64_u64((u64) nr_pages * (u64) PAGE_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) (u64) THREAD_SIZE * 8UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) if (threads > max_threads_suggested)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) threads = max_threads_suggested;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) max_threads = clamp_t(u64, threads, MIN_THREADS, MAX_THREADS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) #ifdef CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) /* Initialized by the architecture: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) int arch_task_struct_size __read_mostly;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) static void task_struct_whitelist(unsigned long *offset, unsigned long *size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) /* Fetch thread_struct whitelist for the architecture. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) arch_thread_struct_whitelist(offset, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) * Handle zero-sized whitelist or empty thread_struct, otherwise
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) * adjust offset to position of thread_struct in task_struct.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) if (unlikely(*size == 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) *offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) *offset += offsetof(struct task_struct, thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) #endif /* CONFIG_ARCH_TASK_STRUCT_ALLOCATOR */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) void __init fork_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) #ifndef ARCH_MIN_TASKALIGN
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) #define ARCH_MIN_TASKALIGN 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) int align = max_t(int, L1_CACHE_BYTES, ARCH_MIN_TASKALIGN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) unsigned long useroffset, usersize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) /* create a slab on which task_structs can be allocated */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) task_struct_whitelist(&useroffset, &usersize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) task_struct_cachep = kmem_cache_create_usercopy("task_struct",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) arch_task_struct_size, align,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) SLAB_PANIC|SLAB_ACCOUNT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) useroffset, usersize, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) /* do the arch specific task caches init */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) arch_task_cache_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) set_max_threads(MAX_THREADS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) init_task.signal->rlim[RLIMIT_NPROC].rlim_cur = max_threads/2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) init_task.signal->rlim[RLIMIT_NPROC].rlim_max = max_threads/2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) init_task.signal->rlim[RLIMIT_SIGPENDING] =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) init_task.signal->rlim[RLIMIT_NPROC];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) for (i = 0; i < UCOUNT_COUNTS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) init_user_ns.ucount_max[i] = max_threads/2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) cpuhp_setup_state(CPUHP_BP_PREPARE_DYN, "fork:vm_stack_cache",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) NULL, free_vm_stack_cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) scs_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) lockdep_init_task(&init_task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) uprobes_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) int __weak arch_dup_task_struct(struct task_struct *dst,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) struct task_struct *src)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) *dst = *src;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) void set_task_stack_end_magic(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) unsigned long *stackend;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) stackend = end_of_stack(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) *stackend = STACK_END_MAGIC; /* for overflow detection */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) struct task_struct *tsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) unsigned long *stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) struct vm_struct *stack_vm_area __maybe_unused;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) if (node == NUMA_NO_NODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) node = tsk_fork_get_node(orig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) tsk = alloc_task_struct_node(node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) if (!tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) stack = alloc_thread_stack_node(tsk, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) if (!stack)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) goto free_tsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) if (memcg_charge_kernel_stack(tsk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) goto free_stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) stack_vm_area = task_stack_vm_area(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) err = arch_dup_task_struct(tsk, orig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) * arch_dup_task_struct() clobbers the stack-related fields. Make
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) * sure they're properly initialized before using any stack-related
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) * functions again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) tsk->stack = stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) #ifdef CONFIG_VMAP_STACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) tsk->stack_vm_area = stack_vm_area;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) #ifdef CONFIG_THREAD_INFO_IN_TASK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) refcount_set(&tsk->stack_refcount, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) goto free_stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) err = scs_prepare(tsk, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) goto free_stack;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) #ifdef CONFIG_SECCOMP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) * We must handle setting up seccomp filters once we're under
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) * the sighand lock in case orig has changed between now and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) * then. Until then, filter must be NULL to avoid messing up
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) * the usage counts on the error path calling free_task.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) tsk->seccomp.filter = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) setup_thread_stack(tsk, orig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) clear_user_return_notifier(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) clear_tsk_need_resched(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) set_task_stack_end_magic(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) #ifdef CONFIG_STACKPROTECTOR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) tsk->stack_canary = get_random_canary();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) if (orig->cpus_ptr == &orig->cpus_mask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) tsk->cpus_ptr = &tsk->cpus_mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) * One for the user space visible state that goes away when reaped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) * One for the scheduler.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) refcount_set(&tsk->rcu_users, 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) /* One for the rcu users */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) refcount_set(&tsk->usage, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) #ifdef CONFIG_BLK_DEV_IO_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) tsk->btrace_seq = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) tsk->splice_pipe = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) tsk->task_frag.page = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) tsk->wake_q.next = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) account_kernel_stack(tsk, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) kcov_task_init(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) #ifdef CONFIG_FAULT_INJECTION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) tsk->fail_nth = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) #ifdef CONFIG_BLK_CGROUP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) tsk->throttle_queue = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) tsk->use_memdelay = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) #ifdef CONFIG_MEMCG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) tsk->active_memcg = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) android_init_vendor_data(tsk, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) android_init_oem_data(tsk, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) trace_android_vh_dup_task_struct(tsk, orig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) return tsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) free_stack:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) free_thread_stack(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) free_tsk:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) free_task_struct(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) __cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) static unsigned long default_dump_filter = MMF_DUMP_FILTER_DEFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) static int __init coredump_filter_setup(char *s)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) default_dump_filter =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) (simple_strtoul(s, NULL, 0) << MMF_DUMP_FILTER_SHIFT) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) MMF_DUMP_FILTER_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) __setup("coredump_filter=", coredump_filter_setup);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) #include <linux/init_task.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) static void mm_init_aio(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) #ifdef CONFIG_AIO
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) spin_lock_init(&mm->ioctx_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) mm->ioctx_table = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) static __always_inline void mm_clear_owner(struct mm_struct *mm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) struct task_struct *p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) #ifdef CONFIG_MEMCG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) if (mm->owner == p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) WRITE_ONCE(mm->owner, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) static void mm_init_owner(struct mm_struct *mm, struct task_struct *p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) #ifdef CONFIG_MEMCG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) mm->owner = p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) static void mm_init_pasid(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) #ifdef CONFIG_IOMMU_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) mm->pasid = INIT_PASID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) static void mm_init_uprobes_state(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) #ifdef CONFIG_UPROBES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) mm->uprobes_state.xol_area = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) struct user_namespace *user_ns)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) mm->mmap = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) mm->mm_rb = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) mm->vmacache_seqnum = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) #ifdef CONFIG_SPECULATIVE_PAGE_FAULT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) rwlock_init(&mm->mm_rb_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) atomic_set(&mm->mm_users, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) atomic_set(&mm->mm_count, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) seqcount_init(&mm->write_protect_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) mmap_init_lock(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) INIT_LIST_HEAD(&mm->mmlist);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) mm->core_state = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) mm_pgtables_bytes_init(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) mm->map_count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) mm->locked_vm = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) atomic_set(&mm->has_pinned, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) atomic64_set(&mm->pinned_vm, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) memset(&mm->rss_stat, 0, sizeof(mm->rss_stat));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) spin_lock_init(&mm->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) spin_lock_init(&mm->arg_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) mm_init_cpumask(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) mm_init_aio(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) mm_init_owner(mm, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) mm_init_pasid(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) RCU_INIT_POINTER(mm->exe_file, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) if (!mmu_notifier_subscriptions_init(mm))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) goto fail_nopgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) init_tlb_flush_pending(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) #if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) mm->pmd_huge_pte = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) mm_init_uprobes_state(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) hugetlb_count_init(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) if (current->mm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) mm->flags = current->mm->flags & MMF_INIT_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) mm->def_flags = current->mm->def_flags & VM_INIT_DEF_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) mm->flags = default_dump_filter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) mm->def_flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) if (mm_alloc_pgd(mm))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) goto fail_nopgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) if (init_new_context(p, mm))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) goto fail_nocontext;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) mm->user_ns = get_user_ns(user_ns);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) return mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) fail_nocontext:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) mm_free_pgd(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) fail_nopgd:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) free_mm(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) * Allocate and initialize an mm_struct.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) struct mm_struct *mm_alloc(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) struct mm_struct *mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) mm = allocate_mm();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) if (!mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) memset(mm, 0, sizeof(*mm));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) return mm_init(mm, current, current_user_ns());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) static inline void __mmput(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) VM_BUG_ON(atomic_read(&mm->mm_users));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) uprobe_clear_state(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) exit_aio(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) ksm_exit(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) khugepaged_exit(mm); /* must run before exit_mmap */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) exit_mmap(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) mm_put_huge_zero_page(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) set_mm_exe_file(mm, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) if (!list_empty(&mm->mmlist)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) spin_lock(&mmlist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) list_del(&mm->mmlist);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) spin_unlock(&mmlist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) if (mm->binfmt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) module_put(mm->binfmt->module);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) mmdrop(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) * Decrement the use count and release all resources for an mm.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) void mmput(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) might_sleep();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) if (atomic_dec_and_test(&mm->mm_users))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) __mmput(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) EXPORT_SYMBOL_GPL(mmput);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) #ifdef CONFIG_MMU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) static void mmput_async_fn(struct work_struct *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) struct mm_struct *mm = container_of(work, struct mm_struct,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) async_put_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) __mmput(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) void mmput_async(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) if (atomic_dec_and_test(&mm->mm_users)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) INIT_WORK(&mm->async_put_work, mmput_async_fn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) schedule_work(&mm->async_put_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) * set_mm_exe_file - change a reference to the mm's executable file
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) * This changes mm's executable file (shown as symlink /proc/[pid]/exe).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181) * Main users are mmput() and sys_execve(). Callers prevent concurrent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) * invocations: in mmput() nobody alive left, in execve task is single
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) * threaded. sys_prctl(PR_SET_MM_MAP/EXE_FILE) also needs to set the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) * mm->exe_file, but does so without using set_mm_exe_file() in order
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) * to do avoid the need for any locks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) struct file *old_exe_file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) * It is safe to dereference the exe_file without RCU as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) * this function is only called if nobody else can access
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) * this mm -- see comment above for justification.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) old_exe_file = rcu_dereference_raw(mm->exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) if (new_exe_file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) get_file(new_exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) rcu_assign_pointer(mm->exe_file, new_exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) if (old_exe_file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) fput(old_exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) * get_mm_exe_file - acquire a reference to the mm's executable file
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) * Returns %NULL if mm has no associated executable file.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) * User must release file via fput().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) struct file *get_mm_exe_file(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) struct file *exe_file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) exe_file = rcu_dereference(mm->exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) if (exe_file && !get_file_rcu(exe_file))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) exe_file = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220) return exe_file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) EXPORT_SYMBOL(get_mm_exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) * get_task_exe_file - acquire a reference to the task's executable file
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) * Returns %NULL if task's mm (if any) has no associated executable file or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) * this is a kernel thread with borrowed mm (see the comment above get_task_mm).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) * User must release file via fput().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) struct file *get_task_exe_file(struct task_struct *task)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) struct file *exe_file = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) struct mm_struct *mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) task_lock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) mm = task->mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) if (mm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) if (!(task->flags & PF_KTHREAD))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) exe_file = get_mm_exe_file(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) task_unlock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) return exe_file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) EXPORT_SYMBOL(get_task_exe_file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) * get_task_mm - acquire a reference to the task's mm
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) * Returns %NULL if the task has no mm. Checks PF_KTHREAD (meaning
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) * this kernel workthread has transiently adopted a user mm with use_mm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) * to do its AIO) is not set and if so returns a reference to it, after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) * bumping up the use count. User must release the mm via mmput()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) * after use. Typically used by /proc and ptrace.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) struct mm_struct *get_task_mm(struct task_struct *task)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) struct mm_struct *mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) task_lock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) mm = task->mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) if (mm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) if (task->flags & PF_KTHREAD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) mm = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) mmget(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) task_unlock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) return mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) EXPORT_SYMBOL_GPL(get_task_mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) struct mm_struct *mm_access(struct task_struct *task, unsigned int mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275) struct mm_struct *mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) err = down_read_killable(&task->signal->exec_update_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) return ERR_PTR(err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) mm = get_task_mm(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) if (mm && mm != current->mm &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) !ptrace_may_access(task, mode)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) mmput(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) mm = ERR_PTR(-EACCES);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) up_read(&task->signal->exec_update_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) return mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) static void complete_vfork_done(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295) struct completion *vfork;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) task_lock(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) vfork = tsk->vfork_done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) if (likely(vfork)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) tsk->vfork_done = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) complete(vfork);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) task_unlock(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) static int wait_for_vfork_done(struct task_struct *child,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) struct completion *vfork)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309) int killed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) freezer_do_not_count();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) cgroup_enter_frozen();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) killed = wait_for_completion_killable(vfork);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) cgroup_leave_frozen(false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) freezer_count();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) if (killed) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) task_lock(child);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) child->vfork_done = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) task_unlock(child);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) put_task_struct(child);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) return killed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) /* Please note the differences between mmput and mm_release.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) * mmput is called whenever we stop holding onto a mm_struct,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) * error success whatever.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) * mm_release is called after a mm_struct has been removed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) * from the current process.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334) * This difference is important for error handling, when we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) * only half set up a mm_struct for a new process and need to restore
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) * the old one. Because we mmput the new mm_struct before
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) * restoring the old one. . .
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) * Eric Biederman 10 January 1998
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) static void mm_release(struct task_struct *tsk, struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) uprobe_free_utask(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344) /* Get rid of any cached register state */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) deactivate_mm(tsk, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) * Signal userspace if we're not exiting with a core dump
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) * because we want to leave the value intact for debugging
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) * purposes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) if (tsk->clear_child_tid) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) if (!(tsk->signal->flags & SIGNAL_GROUP_COREDUMP) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) atomic_read(&mm->mm_users) > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) * We don't check the error code - if userspace has
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) * not set up a proper pointer then tough luck.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) put_user(0, tsk->clear_child_tid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) do_futex(tsk->clear_child_tid, FUTEX_WAKE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) 1, NULL, NULL, 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) tsk->clear_child_tid = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) * All done, finally we can wake up parent and return this mm to him.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) * Also kthread_stop() uses this completion for synchronization.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) if (tsk->vfork_done)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) complete_vfork_done(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374) void exit_mm_release(struct task_struct *tsk, struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) futex_exit_release(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) mm_release(tsk, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) void exec_mm_release(struct task_struct *tsk, struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) futex_exec_release(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383) mm_release(tsk, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) * dup_mm() - duplicates an existing mm structure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) * @tsk: the task_struct with which the new mm will be associated.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) * @oldmm: the mm to duplicate.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) * Allocates a new mm structure and duplicates the provided @oldmm structure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) * content into it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394) * Return: the duplicated mm or NULL on failure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) static struct mm_struct *dup_mm(struct task_struct *tsk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) struct mm_struct *oldmm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) struct mm_struct *mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) mm = allocate_mm();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) if (!mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) goto fail_nomem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) memcpy(mm, oldmm, sizeof(*mm));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) if (!mm_init(mm, tsk, mm->user_ns))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409) goto fail_nomem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) err = dup_mmap(mm, oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) goto free_pt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) mm->hiwater_rss = get_mm_rss(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) mm->hiwater_vm = mm->total_vm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) if (mm->binfmt && !try_module_get(mm->binfmt->module))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) goto free_pt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) return mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) free_pt:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) /* don't put binfmt in mmput, we haven't got module yet */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) mm->binfmt = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) mm_init_owner(mm, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) mmput(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) fail_nomem:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) static int copy_mm(unsigned long clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) struct mm_struct *mm, *oldmm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) int retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) tsk->min_flt = tsk->maj_flt = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439) tsk->nvcsw = tsk->nivcsw = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) #ifdef CONFIG_DETECT_HUNG_TASK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) tsk->last_switch_count = tsk->nvcsw + tsk->nivcsw;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) tsk->last_switch_time = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) tsk->mm = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) tsk->active_mm = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) * Are we cloning a kernel thread?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) * We need to steal a active VM for that..
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) oldmm = current->mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) if (!oldmm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) /* initialize the new vmacache entries */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) vmacache_flush(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) if (clone_flags & CLONE_VM) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) mmget(oldmm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) mm = oldmm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) goto good_mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) retval = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) mm = dup_mm(tsk, current->mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) if (!mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) goto fail_nomem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) good_mm:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) tsk->mm = mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473) tsk->active_mm = mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) fail_nomem:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) return retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) static int copy_fs(unsigned long clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) struct fs_struct *fs = current->fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) if (clone_flags & CLONE_FS) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) /* tsk->fs is already what we want */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) spin_lock(&fs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) if (fs->in_exec) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) spin_unlock(&fs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) return -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) fs->users++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491) spin_unlock(&fs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) tsk->fs = copy_fs_struct(fs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) if (!tsk->fs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) static int copy_files(unsigned long clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) struct files_struct *oldf, *newf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) int error = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) * A background process may not have any files ...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) oldf = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) if (!oldf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) if (clone_flags & CLONE_FILES) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) atomic_inc(&oldf->count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) newf = dup_fd(oldf, NR_OPEN_MAX, &error);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) if (!newf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521) tsk->files = newf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) error = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) return error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) static int copy_io(unsigned long clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) #ifdef CONFIG_BLOCK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) struct io_context *ioc = current->io_context;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) struct io_context *new_ioc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) if (!ioc)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) * Share io context with parent, if CLONE_IO is set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) if (clone_flags & CLONE_IO) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) ioc_task_link(ioc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) tsk->io_context = ioc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) } else if (ioprio_valid(ioc->ioprio)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) new_ioc = get_task_io_context(tsk, GFP_KERNEL, NUMA_NO_NODE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) if (unlikely(!new_ioc))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) new_ioc->ioprio = ioc->ioprio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) put_io_context(new_ioc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) struct sighand_struct *sig;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) if (clone_flags & CLONE_SIGHAND) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) refcount_inc(¤t->sighand->count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) sig = kmem_cache_alloc(sighand_cachep, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) RCU_INIT_POINTER(tsk->sighand, sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) if (!sig)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) refcount_set(&sig->count, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) spin_lock_irq(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) memcpy(sig->action, current->sighand->action, sizeof(sig->action));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) spin_unlock_irq(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) /* Reset all signal handler not set to SIG_IGN to SIG_DFL. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) if (clone_flags & CLONE_CLEAR_SIGHAND)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) flush_signal_handlers(tsk, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) void __cleanup_sighand(struct sighand_struct *sighand)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) if (refcount_dec_and_test(&sighand->count)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) signalfd_cleanup(sighand);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) * sighand_cachep is SLAB_TYPESAFE_BY_RCU so we can free it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) * without an RCU grace period, see __lock_task_sighand().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) kmem_cache_free(sighand_cachep, sighand);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) * Initialize POSIX timer handling for a thread group.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) static void posix_cpu_timers_init_group(struct signal_struct *sig)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) struct posix_cputimers *pct = &sig->posix_cputimers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) unsigned long cpu_limit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) cpu_limit = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) posix_cputimers_group_init(pct, cpu_limit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604) struct signal_struct *sig;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) if (clone_flags & CLONE_THREAD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) sig = kmem_cache_zalloc(signal_cachep, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) tsk->signal = sig;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) if (!sig)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) sig->nr_threads = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) atomic_set(&sig->live, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) refcount_set(&sig->sigcnt, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) /* list_add(thread_node, thread_head) without INIT_LIST_HEAD() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) sig->thread_head = (struct list_head)LIST_HEAD_INIT(tsk->thread_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) tsk->thread_node = (struct list_head)LIST_HEAD_INIT(sig->thread_head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) init_waitqueue_head(&sig->wait_chldexit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) sig->curr_target = tsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) init_sigpending(&sig->shared_pending);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) INIT_HLIST_HEAD(&sig->multiprocess);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) seqlock_init(&sig->stats_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) prev_cputime_init(&sig->prev_cputime);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) #ifdef CONFIG_POSIX_TIMERS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) INIT_LIST_HEAD(&sig->posix_timers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) hrtimer_init(&sig->real_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) sig->real_timer.function = it_real_fn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) task_lock(current->group_leader);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) task_unlock(current->group_leader);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) posix_cpu_timers_init_group(sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) tty_audit_fork(sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) sched_autogroup_fork(sig);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) sig->oom_score_adj = current->signal->oom_score_adj;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) sig->oom_score_adj_min = current->signal->oom_score_adj_min;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) mutex_init(&sig->cred_guard_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648) init_rwsem(&sig->exec_update_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) static void copy_seccomp(struct task_struct *p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) #ifdef CONFIG_SECCOMP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657) * Must be called with sighand->lock held, which is common to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) * all threads in the group. Holding cred_guard_mutex is not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) * needed because this new task is not yet running and cannot
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) * be racing exec.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) assert_spin_locked(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) /* Ref-count the new filter user, and assign it. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665) get_seccomp_filter(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) p->seccomp = current->seccomp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) * Explicitly enable no_new_privs here in case it got set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670) * between the task_struct being duplicated and holding the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671) * sighand lock. The seccomp state and nnp must be in sync.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673) if (task_no_new_privs(current))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) task_set_no_new_privs(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677) * If the parent gained a seccomp mode after copying thread
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) * flags and between before we held the sighand lock, we have
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) * to manually enable the seccomp thread flag here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681) if (p->seccomp.mode != SECCOMP_MODE_DISABLED)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682) set_tsk_thread_flag(p, TIF_SECCOMP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) SYSCALL_DEFINE1(set_tid_address, int __user *, tidptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) current->clear_child_tid = tidptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690) return task_pid_vnr(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693) static void rt_mutex_init_task(struct task_struct *p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) raw_spin_lock_init(&p->pi_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) #ifdef CONFIG_RT_MUTEXES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697) p->pi_waiters = RB_ROOT_CACHED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) p->pi_top_task = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699) p->pi_blocked_on = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) static inline void init_task_pid_links(struct task_struct *task)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) enum pid_type type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) for (type = PIDTYPE_PID; type < PIDTYPE_MAX; ++type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708) INIT_HLIST_NODE(&task->pid_links[type]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712) static inline void
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713) init_task_pid(struct task_struct *task, enum pid_type type, struct pid *pid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715) if (type == PIDTYPE_PID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) task->thread_pid = pid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718) task->signal->pids[type] = pid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721) static inline void rcu_copy_process(struct task_struct *p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723) #ifdef CONFIG_PREEMPT_RCU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724) p->rcu_read_lock_nesting = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725) p->rcu_read_unlock_special.s = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) p->rcu_blocked_node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727) INIT_LIST_HEAD(&p->rcu_node_entry);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728) #endif /* #ifdef CONFIG_PREEMPT_RCU */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729) #ifdef CONFIG_TASKS_RCU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730) p->rcu_tasks_holdout = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731) INIT_LIST_HEAD(&p->rcu_tasks_holdout_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) p->rcu_tasks_idle_cpu = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733) #endif /* #ifdef CONFIG_TASKS_RCU */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734) #ifdef CONFIG_TASKS_TRACE_RCU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) p->trc_reader_nesting = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736) p->trc_reader_special.s = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737) INIT_LIST_HEAD(&p->trc_holdout_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738) #endif /* #ifdef CONFIG_TASKS_TRACE_RCU */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741) struct pid *pidfd_pid(const struct file *file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743) if (file->f_op == &pidfd_fops)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744) return file->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) return ERR_PTR(-EBADF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) static int pidfd_release(struct inode *inode, struct file *file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) struct pid *pid = file->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) file->private_data = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754) put_pid(pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) #ifdef CONFIG_PROC_FS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760) * pidfd_show_fdinfo - print information about a pidfd
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) * @m: proc fdinfo file
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) * @f: file referencing a pidfd
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764) * Pid:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) * This function will print the pid that a given pidfd refers to in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766) * pid namespace of the procfs instance.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767) * If the pid namespace of the process is not a descendant of the pid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) * namespace of the procfs instance 0 will be shown as its pid. This is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769) * similar to calling getppid() on a process whose parent is outside of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) * its pid namespace.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772) * NSpid:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773) * If pid namespaces are supported then this function will also print
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774) * the pid of a given pidfd refers to for all descendant pid namespaces
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775) * starting from the current pid namespace of the instance, i.e. the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776) * Pid field and the first entry in the NSpid field will be identical.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777) * If the pid namespace of the process is not a descendant of the pid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) * namespace of the procfs instance 0 will be shown as its first NSpid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) * entry and no others will be shown.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780) * Note that this differs from the Pid and NSpid fields in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) * /proc/<pid>/status where Pid and NSpid are always shown relative to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782) * the pid namespace of the procfs instance. The difference becomes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783) * obvious when sending around a pidfd between pid namespaces from a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784) * different branch of the tree, i.e. where no ancestoral relation is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785) * present between the pid namespaces:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786) * - create two new pid namespaces ns1 and ns2 in the initial pid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787) * namespace (also take care to create new mount namespaces in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788) * new pid namespace and mount procfs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) * - create a process with a pidfd in ns1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) * - send pidfd from ns1 to ns2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791) * - read /proc/self/fdinfo/<pidfd> and observe that both Pid and NSpid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792) * have exactly one entry, which is 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794) static void pidfd_show_fdinfo(struct seq_file *m, struct file *f)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796) struct pid *pid = f->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) struct pid_namespace *ns;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) pid_t nr = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800) if (likely(pid_has_task(pid, PIDTYPE_PID))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801) ns = proc_pid_ns(file_inode(m->file)->i_sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802) nr = pid_nr_ns(pid, ns);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) seq_put_decimal_ll(m, "Pid:\t", nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807) #ifdef CONFIG_PID_NS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) seq_put_decimal_ll(m, "\nNSpid:\t", nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) if (nr > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) /* If nr is non-zero it means that 'pid' is valid and that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813) * ns, i.e. the pid namespace associated with the procfs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814) * instance, is in the pid namespace hierarchy of pid.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) * Start at one below the already printed level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817) for (i = ns->level + 1; i <= pid->level; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818) seq_put_decimal_ll(m, "\t", pid->numbers[i].nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821) seq_putc(m, '\n');
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) * Poll support for process exit notification.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828) static __poll_t pidfd_poll(struct file *file, struct poll_table_struct *pts)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830) struct pid *pid = file->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831) __poll_t poll_flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833) poll_wait(file, &pid->wait_pidfd, pts);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836) * Inform pollers only when the whole thread group exits.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) * If the thread group leader exits before all other threads in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838) * group, then poll(2) should block, similar to the wait(2) family.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840) if (thread_group_exited(pid))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841) poll_flags = EPOLLIN | EPOLLRDNORM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843) return poll_flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846) const struct file_operations pidfd_fops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847) .release = pidfd_release,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848) .poll = pidfd_poll,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849) #ifdef CONFIG_PROC_FS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) .show_fdinfo = pidfd_show_fdinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1852) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1853)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1854) static void __delayed_free_task(struct rcu_head *rhp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1855) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1856) struct task_struct *tsk = container_of(rhp, struct task_struct, rcu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1857)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1858) free_task(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1859) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1860)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1861) static __always_inline void delayed_free_task(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1862) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1863) if (IS_ENABLED(CONFIG_MEMCG))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1864) call_rcu(&tsk->rcu, __delayed_free_task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1865) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1866) free_task(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1867) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1869) static void copy_oom_score_adj(u64 clone_flags, struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1870) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1871) /* Skip if kernel thread */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1872) if (!tsk->mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1873) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1874)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1875) /* Skip if spawning a thread or using vfork */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1876) if ((clone_flags & (CLONE_VM | CLONE_THREAD | CLONE_VFORK)) != CLONE_VM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1877) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1878)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1879) /* We need to synchronize with __set_oom_adj */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1880) mutex_lock(&oom_adj_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1881) set_bit(MMF_MULTIPROCESS, &tsk->mm->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1882) /* Update the values in case they were changed after copy_signal */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1883) tsk->signal->oom_score_adj = current->signal->oom_score_adj;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1884) tsk->signal->oom_score_adj_min = current->signal->oom_score_adj_min;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1885) mutex_unlock(&oom_adj_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1886) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1887)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1888) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1889) * This creates a new process as a copy of the old one,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1890) * but does not actually start it yet.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1891) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1892) * It copies the registers, and all the appropriate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1893) * parts of the process environment (as per the clone
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1894) * flags). The actual kick-off is left to the caller.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1895) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1896) static __latent_entropy struct task_struct *copy_process(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1897) struct pid *pid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1898) int trace,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1899) int node,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1900) struct kernel_clone_args *args)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1901) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1902) int pidfd = -1, retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1903) struct task_struct *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1904) struct multiprocess_signals delayed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1905) struct file *pidfile = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1906) u64 clone_flags = args->flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1907) struct nsproxy *nsp = current->nsproxy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1908)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1909) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1910) * Don't allow sharing the root directory with processes in a different
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1911) * namespace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1912) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1913) if ((clone_flags & (CLONE_NEWNS|CLONE_FS)) == (CLONE_NEWNS|CLONE_FS))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1914) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1915)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1916) if ((clone_flags & (CLONE_NEWUSER|CLONE_FS)) == (CLONE_NEWUSER|CLONE_FS))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1917) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1918)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1919) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1920) * Thread groups must share signals as well, and detached threads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1921) * can only be started up within the thread group.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1922) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1923) if ((clone_flags & CLONE_THREAD) && !(clone_flags & CLONE_SIGHAND))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1924) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1925)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1926) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1927) * Shared signal handlers imply shared VM. By way of the above,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1928) * thread groups also imply shared VM. Blocking this case allows
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1929) * for various simplifications in other code.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1930) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1931) if ((clone_flags & CLONE_SIGHAND) && !(clone_flags & CLONE_VM))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1932) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1933)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1934) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1935) * Siblings of global init remain as zombies on exit since they are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1936) * not reaped by their parent (swapper). To solve this and to avoid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1937) * multi-rooted process trees, prevent global and container-inits
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1938) * from creating siblings.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1939) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1940) if ((clone_flags & CLONE_PARENT) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1941) current->signal->flags & SIGNAL_UNKILLABLE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1942) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1943)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1944) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1945) * If the new process will be in a different pid or user namespace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1946) * do not allow it to share a thread group with the forking task.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1947) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1948) if (clone_flags & CLONE_THREAD) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1949) if ((clone_flags & (CLONE_NEWUSER | CLONE_NEWPID)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1950) (task_active_pid_ns(current) != nsp->pid_ns_for_children))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1951) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1952) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1953)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1954) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1955) * If the new process will be in a different time namespace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1956) * do not allow it to share VM or a thread group with the forking task.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1957) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1958) if (clone_flags & (CLONE_THREAD | CLONE_VM)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1959) if (nsp->time_ns != nsp->time_ns_for_children)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1960) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1961) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1962)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1963) if (clone_flags & CLONE_PIDFD) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1964) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1965) * - CLONE_DETACHED is blocked so that we can potentially
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1966) * reuse it later for CLONE_PIDFD.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1967) * - CLONE_THREAD is blocked until someone really needs it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1968) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1969) if (clone_flags & (CLONE_DETACHED | CLONE_THREAD))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1970) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1971) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1972)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1973) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1974) * Force any signals received before this point to be delivered
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1975) * before the fork happens. Collect up signals sent to multiple
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1976) * processes that happen during the fork and delay them so that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1977) * they appear to happen after the fork.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1978) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1979) sigemptyset(&delayed.signal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1980) INIT_HLIST_NODE(&delayed.node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1981)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1982) spin_lock_irq(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1983) if (!(clone_flags & CLONE_THREAD))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1984) hlist_add_head(&delayed.node, ¤t->signal->multiprocess);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1985) recalc_sigpending();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1986) spin_unlock_irq(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1987) retval = -ERESTARTNOINTR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1988) if (signal_pending(current))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1989) goto fork_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1990)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1991) retval = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1992) p = dup_task_struct(current, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1993) if (!p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1994) goto fork_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1996) cpufreq_task_times_init(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1997)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1998) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1999) * This _must_ happen before we call free_task(), i.e. before we jump
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2000) * to any of the bad_fork_* labels. This is to avoid freeing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2001) * p->set_child_tid which is (ab)used as a kthread's data pointer for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2002) * kernel threads (PF_KTHREAD).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2003) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2004) p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? args->child_tid : NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2005) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2006) * Clear TID on mm_release()?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2007) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2008) p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? args->child_tid : NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2009)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2010) ftrace_graph_init_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2011)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2012) rt_mutex_init_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2013)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2014) lockdep_assert_irqs_enabled();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2015) #ifdef CONFIG_PROVE_LOCKING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2016) DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2017) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2018) retval = -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2019) if (atomic_read(&p->real_cred->user->processes) >=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2020) task_rlimit(p, RLIMIT_NPROC)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2021) if (p->real_cred->user != INIT_USER &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2022) !capable(CAP_SYS_RESOURCE) && !capable(CAP_SYS_ADMIN))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2023) goto bad_fork_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2024) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2025) current->flags &= ~PF_NPROC_EXCEEDED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2026)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2027) retval = copy_creds(p, clone_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2028) if (retval < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2029) goto bad_fork_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2030)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2031) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2032) * If multiple threads are within copy_process(), then this check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2033) * triggers too late. This doesn't hurt, the check is only there
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2034) * to stop root fork bombs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2035) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2036) retval = -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2037) if (data_race(nr_threads >= max_threads))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2038) goto bad_fork_cleanup_count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2039)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2040) delayacct_tsk_init(p); /* Must remain after dup_task_struct() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2041) p->flags &= ~(PF_SUPERPRIV | PF_WQ_WORKER | PF_IDLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2042) p->flags |= PF_FORKNOEXEC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2043) INIT_LIST_HEAD(&p->children);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2044) INIT_LIST_HEAD(&p->sibling);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2045) rcu_copy_process(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2046) p->vfork_done = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2047) spin_lock_init(&p->alloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2048)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2049) init_sigpending(&p->pending);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2050)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2051) p->utime = p->stime = p->gtime = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2052) #ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2053) p->utimescaled = p->stimescaled = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2054) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2055) prev_cputime_init(&p->prev_cputime);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2056)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2057) #ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2058) seqcount_init(&p->vtime.seqcount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2059) p->vtime.starttime = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2060) p->vtime.state = VTIME_INACTIVE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2061) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2062)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2063) #ifdef CONFIG_IO_URING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2064) p->io_uring = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2065) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2066)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2067) #if defined(SPLIT_RSS_COUNTING)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2068) memset(&p->rss_stat, 0, sizeof(p->rss_stat));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2069) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2070)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2071) p->default_timer_slack_ns = current->timer_slack_ns;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2072)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2073) #ifdef CONFIG_PSI
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2074) p->psi_flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2075) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2076)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2077) task_io_accounting_init(&p->ioac);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2078) acct_clear_integrals(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2079)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2080) posix_cputimers_init(&p->posix_cputimers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2082) p->io_context = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2083) audit_set_context(p, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2084) cgroup_fork(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2085) #ifdef CONFIG_NUMA
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2086) p->mempolicy = mpol_dup(p->mempolicy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2087) if (IS_ERR(p->mempolicy)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2088) retval = PTR_ERR(p->mempolicy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2089) p->mempolicy = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2090) goto bad_fork_cleanup_threadgroup_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2091) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2092) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2093) #ifdef CONFIG_CPUSETS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2094) p->cpuset_mem_spread_rotor = NUMA_NO_NODE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2095) p->cpuset_slab_spread_rotor = NUMA_NO_NODE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2096) seqcount_spinlock_init(&p->mems_allowed_seq, &p->alloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2097) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2098) #ifdef CONFIG_TRACE_IRQFLAGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2099) memset(&p->irqtrace, 0, sizeof(p->irqtrace));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2100) p->irqtrace.hardirq_disable_ip = _THIS_IP_;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2101) p->irqtrace.softirq_enable_ip = _THIS_IP_;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2102) p->softirqs_enabled = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2103) p->softirq_context = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2104) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2106) p->pagefault_disabled = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2108) #ifdef CONFIG_LOCKDEP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2109) lockdep_init_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2110) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2112) #ifdef CONFIG_DEBUG_MUTEXES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2113) p->blocked_on = NULL; /* not blocked yet */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2114) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2115) #ifdef CONFIG_BCACHE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2116) p->sequential_io = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2117) p->sequential_io_avg = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2118) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2120) /* Perform scheduler related setup. Assign this task to a CPU. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2121) retval = sched_fork(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2122) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2123) goto bad_fork_cleanup_policy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2124)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2125) retval = perf_event_init_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2126) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2127) goto bad_fork_cleanup_policy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2128) retval = audit_alloc(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2129) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2130) goto bad_fork_cleanup_perf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2131) /* copy all the process information */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2132) shm_init_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2133) retval = security_task_alloc(p, clone_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2134) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2135) goto bad_fork_cleanup_audit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2136) retval = copy_semundo(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2137) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2138) goto bad_fork_cleanup_security;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2139) retval = copy_files(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2140) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2141) goto bad_fork_cleanup_semundo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2142) retval = copy_fs(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2143) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2144) goto bad_fork_cleanup_files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2145) retval = copy_sighand(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2146) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2147) goto bad_fork_cleanup_fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2148) retval = copy_signal(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2149) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2150) goto bad_fork_cleanup_sighand;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2151) retval = copy_mm(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2152) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2153) goto bad_fork_cleanup_signal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2154) retval = copy_namespaces(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2155) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2156) goto bad_fork_cleanup_mm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2157) retval = copy_io(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2158) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2159) goto bad_fork_cleanup_namespaces;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2160) retval = copy_thread(clone_flags, args->stack, args->stack_size, p, args->tls);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2161) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2162) goto bad_fork_cleanup_io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2164) stackleak_task_init(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2165)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2166) if (pid != &init_struct_pid) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2167) pid = alloc_pid(p->nsproxy->pid_ns_for_children, args->set_tid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2168) args->set_tid_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2169) if (IS_ERR(pid)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2170) retval = PTR_ERR(pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2171) goto bad_fork_cleanup_thread;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2172) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2173) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2174)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2175) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2176) * This has to happen after we've potentially unshared the file
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2177) * descriptor table (so that the pidfd doesn't leak into the child
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2178) * if the fd table isn't shared).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2179) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2180) if (clone_flags & CLONE_PIDFD) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2181) retval = get_unused_fd_flags(O_RDWR | O_CLOEXEC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2182) if (retval < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2183) goto bad_fork_free_pid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2185) pidfd = retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2187) pidfile = anon_inode_getfile("[pidfd]", &pidfd_fops, pid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2188) O_RDWR | O_CLOEXEC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2189) if (IS_ERR(pidfile)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2190) put_unused_fd(pidfd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2191) retval = PTR_ERR(pidfile);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2192) goto bad_fork_free_pid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2193) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2194) get_pid(pid); /* held by pidfile now */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2195)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2196) retval = put_user(pidfd, args->pidfd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2197) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2198) goto bad_fork_put_pidfd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2199) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2200)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2201) #ifdef CONFIG_BLOCK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2202) p->plug = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2203) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2204) futex_init_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2206) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2207) * sigaltstack should be cleared when sharing the same VM
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2208) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2209) if ((clone_flags & (CLONE_VM|CLONE_VFORK)) == CLONE_VM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2210) sas_ss_reset(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2212) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2213) * Syscall tracing and stepping should be turned off in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2214) * child regardless of CLONE_PTRACE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2215) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2216) user_disable_single_step(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2217) clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2218) #ifdef TIF_SYSCALL_EMU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2219) clear_tsk_thread_flag(p, TIF_SYSCALL_EMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2220) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2221) clear_tsk_latency_tracing(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2223) /* ok, now we should be set up.. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2224) p->pid = pid_nr(pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2225) if (clone_flags & CLONE_THREAD) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2226) p->group_leader = current->group_leader;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2227) p->tgid = current->tgid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2228) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2229) p->group_leader = p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2230) p->tgid = p->pid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2231) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2232)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2233) p->nr_dirtied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2234) p->nr_dirtied_pause = 128 >> (PAGE_SHIFT - 10);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2235) p->dirty_paused_when = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2237) p->pdeath_signal = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2238) INIT_LIST_HEAD(&p->thread_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2239) p->task_works = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2240) clear_posix_cputimers_work(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2241)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2242) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2243) * Ensure that the cgroup subsystem policies allow the new process to be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2244) * forked. It should be noted that the new process's css_set can be changed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2245) * between here and cgroup_post_fork() if an organisation operation is in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2246) * progress.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2247) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2248) retval = cgroup_can_fork(p, args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2249) if (retval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2250) goto bad_fork_put_pidfd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2252) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2253) * Now that the cgroups are pinned, re-clone the parent cgroup and put
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2254) * the new task on the correct runqueue. All this *before* the task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2255) * becomes visible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2256) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2257) * This isn't part of ->can_fork() because while the re-cloning is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2258) * cgroup specific, it unconditionally needs to place the task on a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2259) * runqueue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2260) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2261) sched_cgroup_fork(p, args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2263) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2264) * From this point on we must avoid any synchronous user-space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2265) * communication until we take the tasklist-lock. In particular, we do
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2266) * not want user-space to be able to predict the process start-time by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2267) * stalling fork(2) after we recorded the start_time but before it is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2268) * visible to the system.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2269) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2271) p->start_time = ktime_get_ns();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2272) p->start_boottime = ktime_get_boottime_ns();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2274) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2275) * Make it visible to the rest of the system, but dont wake it up yet.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2276) * Need tasklist lock for parent etc handling!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2277) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2278) write_lock_irq(&tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2279)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2280) /* CLONE_PARENT re-uses the old parent */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2281) if (clone_flags & (CLONE_PARENT|CLONE_THREAD)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2282) p->real_parent = current->real_parent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2283) p->parent_exec_id = current->parent_exec_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2284) if (clone_flags & CLONE_THREAD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2285) p->exit_signal = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2286) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2287) p->exit_signal = current->group_leader->exit_signal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2288) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2289) p->real_parent = current;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2290) p->parent_exec_id = current->self_exec_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2291) p->exit_signal = args->exit_signal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2292) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2294) klp_copy_process(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2296) spin_lock(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2298) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2299) * Copy seccomp details explicitly here, in case they were changed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2300) * before holding sighand lock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2301) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2302) copy_seccomp(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2304) rseq_fork(p, clone_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2306) /* Don't start children in a dying pid namespace */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2307) if (unlikely(!(ns_of_pid(pid)->pid_allocated & PIDNS_ADDING))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2308) retval = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2309) goto bad_fork_cancel_cgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2310) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2311)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2312) /* Let kill terminate clone/fork in the middle */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2313) if (fatal_signal_pending(current)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2314) retval = -EINTR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2315) goto bad_fork_cancel_cgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2316) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2318) init_task_pid_links(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2319) if (likely(p->pid)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2320) ptrace_init_task(p, (clone_flags & CLONE_PTRACE) || trace);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2322) init_task_pid(p, PIDTYPE_PID, pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2323) if (thread_group_leader(p)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2324) init_task_pid(p, PIDTYPE_TGID, pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2325) init_task_pid(p, PIDTYPE_PGID, task_pgrp(current));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2326) init_task_pid(p, PIDTYPE_SID, task_session(current));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2328) if (is_child_reaper(pid)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2329) ns_of_pid(pid)->child_reaper = p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2330) p->signal->flags |= SIGNAL_UNKILLABLE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2331) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2332) p->signal->shared_pending.signal = delayed.signal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2333) p->signal->tty = tty_kref_get(current->signal->tty);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2334) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2335) * Inherit has_child_subreaper flag under the same
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2336) * tasklist_lock with adding child to the process tree
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2337) * for propagate_has_child_subreaper optimization.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2338) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2339) p->signal->has_child_subreaper = p->real_parent->signal->has_child_subreaper ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2340) p->real_parent->signal->is_child_subreaper;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2341) list_add_tail(&p->sibling, &p->real_parent->children);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2342) list_add_tail_rcu(&p->tasks, &init_task.tasks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2343) attach_pid(p, PIDTYPE_TGID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2344) attach_pid(p, PIDTYPE_PGID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2345) attach_pid(p, PIDTYPE_SID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2346) __this_cpu_inc(process_counts);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2347) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2348) current->signal->nr_threads++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2349) atomic_inc(¤t->signal->live);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2350) refcount_inc(¤t->signal->sigcnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2351) task_join_group_stop(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2352) list_add_tail_rcu(&p->thread_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2353) &p->group_leader->thread_group);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2354) list_add_tail_rcu(&p->thread_node,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2355) &p->signal->thread_head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2356) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2357) attach_pid(p, PIDTYPE_PID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2358) nr_threads++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2359) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2360) total_forks++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2361) hlist_del_init(&delayed.node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2362) spin_unlock(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2363) syscall_tracepoint_update(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2364) write_unlock_irq(&tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2365)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2366) if (pidfile)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2367) fd_install(pidfd, pidfile);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2369) proc_fork_connector(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2370) sched_post_fork(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2371) cgroup_post_fork(p, args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2372) perf_event_fork(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2374) trace_task_newtask(p, clone_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2375) uprobe_copy_process(p, clone_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2377) copy_oom_score_adj(clone_flags, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2379) return p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2380)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2381) bad_fork_cancel_cgroup:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2382) spin_unlock(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2383) write_unlock_irq(&tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2384) cgroup_cancel_fork(p, args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2385) bad_fork_put_pidfd:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2386) if (clone_flags & CLONE_PIDFD) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2387) fput(pidfile);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2388) put_unused_fd(pidfd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2389) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2390) bad_fork_free_pid:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2391) if (pid != &init_struct_pid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2392) free_pid(pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2393) bad_fork_cleanup_thread:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2394) exit_thread(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2395) bad_fork_cleanup_io:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2396) if (p->io_context)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2397) exit_io_context(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2398) bad_fork_cleanup_namespaces:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2399) exit_task_namespaces(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2400) bad_fork_cleanup_mm:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2401) if (p->mm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2402) mm_clear_owner(p->mm, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2403) mmput(p->mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2404) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2405) bad_fork_cleanup_signal:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2406) if (!(clone_flags & CLONE_THREAD))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2407) free_signal_struct(p->signal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2408) bad_fork_cleanup_sighand:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2409) __cleanup_sighand(p->sighand);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2410) bad_fork_cleanup_fs:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2411) exit_fs(p); /* blocking */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2412) bad_fork_cleanup_files:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2413) exit_files(p); /* blocking */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2414) bad_fork_cleanup_semundo:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2415) exit_sem(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2416) bad_fork_cleanup_security:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2417) security_task_free(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2418) bad_fork_cleanup_audit:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2419) audit_free(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2420) bad_fork_cleanup_perf:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2421) perf_event_free_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2422) bad_fork_cleanup_policy:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2423) lockdep_free_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2424) #ifdef CONFIG_NUMA
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2425) mpol_put(p->mempolicy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2426) bad_fork_cleanup_threadgroup_lock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2427) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2428) delayacct_tsk_free(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2429) bad_fork_cleanup_count:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2430) atomic_dec(&p->cred->user->processes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2431) exit_creds(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2432) bad_fork_free:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2433) p->state = TASK_DEAD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2434) put_task_stack(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2435) delayed_free_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2436) fork_out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2437) spin_lock_irq(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2438) hlist_del_init(&delayed.node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2439) spin_unlock_irq(¤t->sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2440) return ERR_PTR(retval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2441) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2442)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2443) static inline void init_idle_pids(struct task_struct *idle)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2444) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2445) enum pid_type type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2447) for (type = PIDTYPE_PID; type < PIDTYPE_MAX; ++type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2448) INIT_HLIST_NODE(&idle->pid_links[type]); /* not really needed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2449) init_task_pid(idle, type, &init_struct_pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2450) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2451) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2453) struct task_struct * __init fork_idle(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2454) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2455) struct task_struct *task;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2456) struct kernel_clone_args args = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2457) .flags = CLONE_VM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2458) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2460) task = copy_process(&init_struct_pid, 0, cpu_to_node(cpu), &args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2461) if (!IS_ERR(task)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2462) init_idle_pids(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2463) init_idle(task, cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2464) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2466) return task;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2467) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2468)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2469) struct mm_struct *copy_init_mm(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2470) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2471) return dup_mm(NULL, &init_mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2472) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2473)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2474) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2475) * Ok, this is the main fork-routine.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2476) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2477) * It copies the process, and if successful kick-starts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2478) * it and waits for it to finish using the VM if required.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2479) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2480) * args->exit_signal is expected to be checked for sanity by the caller.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2481) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2482) pid_t kernel_clone(struct kernel_clone_args *args)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2483) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2484) u64 clone_flags = args->flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2485) struct completion vfork;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2486) struct pid *pid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2487) struct task_struct *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2488) int trace = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2489) pid_t nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2491) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2492) * For legacy clone() calls, CLONE_PIDFD uses the parent_tid argument
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2493) * to return the pidfd. Hence, CLONE_PIDFD and CLONE_PARENT_SETTID are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2494) * mutually exclusive. With clone3() CLONE_PIDFD has grown a separate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2495) * field in struct clone_args and it still doesn't make sense to have
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2496) * them both point at the same memory location. Performing this check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2497) * here has the advantage that we don't need to have a separate helper
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2498) * to check for legacy clone().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2499) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2500) if ((args->flags & CLONE_PIDFD) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2501) (args->flags & CLONE_PARENT_SETTID) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2502) (args->pidfd == args->parent_tid))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2503) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2505) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2506) * Determine whether and which event to report to ptracer. When
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2507) * called from kernel_thread or CLONE_UNTRACED is explicitly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2508) * requested, no event is reported; otherwise, report if the event
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2509) * for the type of forking is enabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2510) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2511) if (!(clone_flags & CLONE_UNTRACED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2512) if (clone_flags & CLONE_VFORK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2513) trace = PTRACE_EVENT_VFORK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2514) else if (args->exit_signal != SIGCHLD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2515) trace = PTRACE_EVENT_CLONE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2516) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2517) trace = PTRACE_EVENT_FORK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2518)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2519) if (likely(!ptrace_event_enabled(current, trace)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2520) trace = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2521) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2522)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2523) p = copy_process(NULL, trace, NUMA_NO_NODE, args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2524) add_latent_entropy();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2526) if (IS_ERR(p))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2527) return PTR_ERR(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2528)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2529) cpufreq_task_times_alloc(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2530)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2531) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2532) * Do this prior waking up the new thread - the thread pointer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2533) * might get invalid after that point, if the thread exits quickly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2534) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2535) trace_sched_process_fork(current, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2537) pid = get_task_pid(p, PIDTYPE_PID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2538) nr = pid_vnr(pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2539)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2540) if (clone_flags & CLONE_PARENT_SETTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2541) put_user(nr, args->parent_tid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2543) if (clone_flags & CLONE_VFORK) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2544) p->vfork_done = &vfork;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2545) init_completion(&vfork);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2546) get_task_struct(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2547) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2548)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2549) wake_up_new_task(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2550)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2551) /* forking complete and child started to run, tell ptracer */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2552) if (unlikely(trace))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2553) ptrace_event_pid(trace, pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2554)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2555) if (clone_flags & CLONE_VFORK) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2556) if (!wait_for_vfork_done(p, &vfork))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2557) ptrace_event_pid(PTRACE_EVENT_VFORK_DONE, pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2558) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2559)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2560) put_pid(pid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2561) return nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2562) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2563)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2564) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2565) * Create a kernel thread.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2566) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2567) pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2568) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2569) struct kernel_clone_args args = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2570) .flags = ((lower_32_bits(flags) | CLONE_VM |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2571) CLONE_UNTRACED) & ~CSIGNAL),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2572) .exit_signal = (lower_32_bits(flags) & CSIGNAL),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2573) .stack = (unsigned long)fn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2574) .stack_size = (unsigned long)arg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2575) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2576)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2577) return kernel_clone(&args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2578) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2579)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2580) #ifdef __ARCH_WANT_SYS_FORK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2581) SYSCALL_DEFINE0(fork)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2582) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2583) #ifdef CONFIG_MMU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2584) struct kernel_clone_args args = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2585) .exit_signal = SIGCHLD,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2586) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2588) return kernel_clone(&args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2589) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2590) /* can not support in nommu mode */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2591) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2592) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2593) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2594) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2596) #ifdef __ARCH_WANT_SYS_VFORK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2597) SYSCALL_DEFINE0(vfork)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2598) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2599) struct kernel_clone_args args = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2600) .flags = CLONE_VFORK | CLONE_VM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2601) .exit_signal = SIGCHLD,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2602) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2603)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2604) return kernel_clone(&args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2605) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2606) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2607)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2608) #ifdef __ARCH_WANT_SYS_CLONE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2609) #ifdef CONFIG_CLONE_BACKWARDS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2610) SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2611) int __user *, parent_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2612) unsigned long, tls,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2613) int __user *, child_tidptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2614) #elif defined(CONFIG_CLONE_BACKWARDS2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2615) SYSCALL_DEFINE5(clone, unsigned long, newsp, unsigned long, clone_flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2616) int __user *, parent_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2617) int __user *, child_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2618) unsigned long, tls)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2619) #elif defined(CONFIG_CLONE_BACKWARDS3)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2620) SYSCALL_DEFINE6(clone, unsigned long, clone_flags, unsigned long, newsp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2621) int, stack_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2622) int __user *, parent_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2623) int __user *, child_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2624) unsigned long, tls)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2625) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2626) SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2627) int __user *, parent_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2628) int __user *, child_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2629) unsigned long, tls)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2630) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2631) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2632) struct kernel_clone_args args = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2633) .flags = (lower_32_bits(clone_flags) & ~CSIGNAL),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2634) .pidfd = parent_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2635) .child_tid = child_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2636) .parent_tid = parent_tidptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2637) .exit_signal = (lower_32_bits(clone_flags) & CSIGNAL),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2638) .stack = newsp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2639) .tls = tls,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2640) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2641)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2642) return kernel_clone(&args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2643) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2644) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2645)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2646) #ifdef __ARCH_WANT_SYS_CLONE3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2647)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2648) noinline static int copy_clone_args_from_user(struct kernel_clone_args *kargs,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2649) struct clone_args __user *uargs,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2650) size_t usize)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2651) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2652) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2653) struct clone_args args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2654) pid_t *kset_tid = kargs->set_tid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2655)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2656) BUILD_BUG_ON(offsetofend(struct clone_args, tls) !=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2657) CLONE_ARGS_SIZE_VER0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2658) BUILD_BUG_ON(offsetofend(struct clone_args, set_tid_size) !=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2659) CLONE_ARGS_SIZE_VER1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2660) BUILD_BUG_ON(offsetofend(struct clone_args, cgroup) !=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2661) CLONE_ARGS_SIZE_VER2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2662) BUILD_BUG_ON(sizeof(struct clone_args) != CLONE_ARGS_SIZE_VER2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2664) if (unlikely(usize > PAGE_SIZE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2665) return -E2BIG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2666) if (unlikely(usize < CLONE_ARGS_SIZE_VER0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2667) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2668)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2669) err = copy_struct_from_user(&args, sizeof(args), uargs, usize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2670) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2671) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2672)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2673) if (unlikely(args.set_tid_size > MAX_PID_NS_LEVEL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2674) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2675)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2676) if (unlikely(!args.set_tid && args.set_tid_size > 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2677) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2678)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2679) if (unlikely(args.set_tid && args.set_tid_size == 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2680) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2681)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2682) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2683) * Verify that higher 32bits of exit_signal are unset and that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2684) * it is a valid signal
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2685) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2686) if (unlikely((args.exit_signal & ~((u64)CSIGNAL)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2687) !valid_signal(args.exit_signal)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2688) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2689)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2690) if ((args.flags & CLONE_INTO_CGROUP) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2691) (args.cgroup > INT_MAX || usize < CLONE_ARGS_SIZE_VER2))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2692) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2693)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2694) *kargs = (struct kernel_clone_args){
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2695) .flags = args.flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2696) .pidfd = u64_to_user_ptr(args.pidfd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2697) .child_tid = u64_to_user_ptr(args.child_tid),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2698) .parent_tid = u64_to_user_ptr(args.parent_tid),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2699) .exit_signal = args.exit_signal,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2700) .stack = args.stack,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2701) .stack_size = args.stack_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2702) .tls = args.tls,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2703) .set_tid_size = args.set_tid_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2704) .cgroup = args.cgroup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2705) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2707) if (args.set_tid &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2708) copy_from_user(kset_tid, u64_to_user_ptr(args.set_tid),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2709) (kargs->set_tid_size * sizeof(pid_t))))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2710) return -EFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2712) kargs->set_tid = kset_tid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2713)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2714) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2715) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2716)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2717) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2718) * clone3_stack_valid - check and prepare stack
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2719) * @kargs: kernel clone args
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2720) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2721) * Verify that the stack arguments userspace gave us are sane.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2722) * In addition, set the stack direction for userspace since it's easy for us to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2723) * determine.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2724) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2725) static inline bool clone3_stack_valid(struct kernel_clone_args *kargs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2726) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2727) if (kargs->stack == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2728) if (kargs->stack_size > 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2729) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2730) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2731) if (kargs->stack_size == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2732) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2733)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2734) if (!access_ok((void __user *)kargs->stack, kargs->stack_size))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2735) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2736)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2737) #if !defined(CONFIG_STACK_GROWSUP) && !defined(CONFIG_IA64)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2738) kargs->stack += kargs->stack_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2739) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2740) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2742) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2743) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2744)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2745) static bool clone3_args_valid(struct kernel_clone_args *kargs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2746) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2747) /* Verify that no unknown flags are passed along. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2748) if (kargs->flags &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2749) ~(CLONE_LEGACY_FLAGS | CLONE_CLEAR_SIGHAND | CLONE_INTO_CGROUP))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2750) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2751)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2752) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2753) * - make the CLONE_DETACHED bit reuseable for clone3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2754) * - make the CSIGNAL bits reuseable for clone3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2755) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2756) if (kargs->flags & (CLONE_DETACHED | CSIGNAL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2757) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2758)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2759) if ((kargs->flags & (CLONE_SIGHAND | CLONE_CLEAR_SIGHAND)) ==
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2760) (CLONE_SIGHAND | CLONE_CLEAR_SIGHAND))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2761) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2762)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2763) if ((kargs->flags & (CLONE_THREAD | CLONE_PARENT)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2764) kargs->exit_signal)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2765) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2766)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2767) if (!clone3_stack_valid(kargs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2768) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2769)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2770) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2771) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2773) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2774) * clone3 - create a new process with specific properties
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2775) * @uargs: argument structure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2776) * @size: size of @uargs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2777) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2778) * clone3() is the extensible successor to clone()/clone2().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2779) * It takes a struct as argument that is versioned by its size.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2780) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2781) * Return: On success, a positive PID for the child process.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2782) * On error, a negative errno number.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2783) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2784) SYSCALL_DEFINE2(clone3, struct clone_args __user *, uargs, size_t, size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2785) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2786) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2787)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2788) struct kernel_clone_args kargs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2789) pid_t set_tid[MAX_PID_NS_LEVEL];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2790)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2791) kargs.set_tid = set_tid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2793) err = copy_clone_args_from_user(&kargs, uargs, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2794) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2795) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2796)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2797) if (!clone3_args_valid(&kargs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2798) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2799)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2800) return kernel_clone(&kargs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2801) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2802) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2803)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2804) void walk_process_tree(struct task_struct *top, proc_visitor visitor, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2805) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2806) struct task_struct *leader, *parent, *child;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2807) int res;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2808)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2809) read_lock(&tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2810) leader = top = top->group_leader;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2811) down:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2812) for_each_thread(leader, parent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2813) list_for_each_entry(child, &parent->children, sibling) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2814) res = visitor(child, data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2815) if (res) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2816) if (res < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2817) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2818) leader = child;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2819) goto down;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2820) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2821) up:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2822) ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2823) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2824) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2825)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2826) if (leader != top) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2827) child = leader;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2828) parent = child->real_parent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2829) leader = parent->group_leader;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2830) goto up;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2831) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2832) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2833) read_unlock(&tasklist_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2834) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2835)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2836) #ifndef ARCH_MIN_MMSTRUCT_ALIGN
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2837) #define ARCH_MIN_MMSTRUCT_ALIGN 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2838) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2839)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2840) static void sighand_ctor(void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2841) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2842) struct sighand_struct *sighand = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2843)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2844) spin_lock_init(&sighand->siglock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2845) init_waitqueue_head(&sighand->signalfd_wqh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2846) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2847)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2848) void __init proc_caches_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2849) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2850) unsigned int mm_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2852) sighand_cachep = kmem_cache_create("sighand_cache",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2853) sizeof(struct sighand_struct), 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2854) SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_TYPESAFE_BY_RCU|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2855) SLAB_ACCOUNT, sighand_ctor);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2856) signal_cachep = kmem_cache_create("signal_cache",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2857) sizeof(struct signal_struct), 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2858) SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2859) NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2860) files_cachep = kmem_cache_create("files_cache",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2861) sizeof(struct files_struct), 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2862) SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2863) NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2864) fs_cachep = kmem_cache_create("fs_cache",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2865) sizeof(struct fs_struct), 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2866) SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2867) NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2869) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2870) * The mm_cpumask is located at the end of mm_struct, and is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2871) * dynamically sized based on the maximum CPU number this system
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2872) * can have, taking hotplug into account (nr_cpu_ids).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2873) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2874) mm_size = sizeof(struct mm_struct) + cpumask_size();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2875)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2876) mm_cachep = kmem_cache_create_usercopy("mm_struct",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2877) mm_size, ARCH_MIN_MMSTRUCT_ALIGN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2878) SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2879) offsetof(struct mm_struct, saved_auxv),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2880) sizeof_field(struct mm_struct, saved_auxv),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2881) NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2882) vm_area_cachep = KMEM_CACHE(vm_area_struct, SLAB_PANIC|SLAB_ACCOUNT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2883) mmap_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2884) nsproxy_cache_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2885) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2886)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2887) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2888) * Check constraints on flags passed to the unshare system call.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2889) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2890) static int check_unshare_flags(unsigned long unshare_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2891) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2892) if (unshare_flags & ~(CLONE_THREAD|CLONE_FS|CLONE_NEWNS|CLONE_SIGHAND|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2893) CLONE_VM|CLONE_FILES|CLONE_SYSVSEM|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2894) CLONE_NEWUTS|CLONE_NEWIPC|CLONE_NEWNET|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2895) CLONE_NEWUSER|CLONE_NEWPID|CLONE_NEWCGROUP|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2896) CLONE_NEWTIME))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2897) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2898) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2899) * Not implemented, but pretend it works if there is nothing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2900) * to unshare. Note that unsharing the address space or the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2901) * signal handlers also need to unshare the signal queues (aka
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2902) * CLONE_THREAD).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2903) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2904) if (unshare_flags & (CLONE_THREAD | CLONE_SIGHAND | CLONE_VM)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2905) if (!thread_group_empty(current))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2906) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2907) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2908) if (unshare_flags & (CLONE_SIGHAND | CLONE_VM)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2909) if (refcount_read(¤t->sighand->count) > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2910) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2911) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2912) if (unshare_flags & CLONE_VM) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2913) if (!current_is_single_threaded())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2914) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2915) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2916)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2917) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2918) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2919)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2920) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2921) * Unshare the filesystem structure if it is being shared
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2922) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2923) static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2924) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2925) struct fs_struct *fs = current->fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2926)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2927) if (!(unshare_flags & CLONE_FS) || !fs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2928) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2929)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2930) /* don't need lock here; in the worst case we'll do useless copy */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2931) if (fs->users == 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2932) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2933)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2934) *new_fsp = copy_fs_struct(fs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2935) if (!*new_fsp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2936) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2937)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2938) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2939) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2940)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2941) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2942) * Unshare file descriptor table if it is being shared
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2943) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2944) int unshare_fd(unsigned long unshare_flags, unsigned int max_fds,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2945) struct files_struct **new_fdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2946) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2947) struct files_struct *fd = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2948) int error = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2949)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2950) if ((unshare_flags & CLONE_FILES) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2951) (fd && atomic_read(&fd->count) > 1)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2952) *new_fdp = dup_fd(fd, max_fds, &error);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2953) if (!*new_fdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2954) return error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2955) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2956)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2957) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2958) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2959)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2960) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2961) * unshare allows a process to 'unshare' part of the process
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2962) * context which was originally shared using clone. copy_*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2963) * functions used by kernel_clone() cannot be used here directly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2964) * because they modify an inactive task_struct that is being
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2965) * constructed. Here we are modifying the current, active,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2966) * task_struct.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2967) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2968) int ksys_unshare(unsigned long unshare_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2969) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2970) struct fs_struct *fs, *new_fs = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2971) struct files_struct *fd, *new_fd = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2972) struct cred *new_cred = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2973) struct nsproxy *new_nsproxy = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2974) int do_sysvsem = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2975) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2976)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2977) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2978) * If unsharing a user namespace must also unshare the thread group
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2979) * and unshare the filesystem root and working directories.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2980) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2981) if (unshare_flags & CLONE_NEWUSER)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2982) unshare_flags |= CLONE_THREAD | CLONE_FS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2983) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2984) * If unsharing vm, must also unshare signal handlers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2985) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2986) if (unshare_flags & CLONE_VM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2987) unshare_flags |= CLONE_SIGHAND;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2988) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2989) * If unsharing a signal handlers, must also unshare the signal queues.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2990) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2991) if (unshare_flags & CLONE_SIGHAND)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2992) unshare_flags |= CLONE_THREAD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2993) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2994) * If unsharing namespace, must also unshare filesystem information.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2995) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2996) if (unshare_flags & CLONE_NEWNS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2997) unshare_flags |= CLONE_FS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2999) err = check_unshare_flags(unshare_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3000) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3001) goto bad_unshare_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3002) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3003) * CLONE_NEWIPC must also detach from the undolist: after switching
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3004) * to a new ipc namespace, the semaphore arrays from the old
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3005) * namespace are unreachable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3006) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3007) if (unshare_flags & (CLONE_NEWIPC|CLONE_SYSVSEM))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3008) do_sysvsem = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3009) err = unshare_fs(unshare_flags, &new_fs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3010) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3011) goto bad_unshare_out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3012) err = unshare_fd(unshare_flags, NR_OPEN_MAX, &new_fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3013) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3014) goto bad_unshare_cleanup_fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3015) err = unshare_userns(unshare_flags, &new_cred);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3016) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3017) goto bad_unshare_cleanup_fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3018) err = unshare_nsproxy_namespaces(unshare_flags, &new_nsproxy,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3019) new_cred, new_fs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3020) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3021) goto bad_unshare_cleanup_cred;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3022)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3023) if (new_fs || new_fd || do_sysvsem || new_cred || new_nsproxy) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3024) if (do_sysvsem) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3025) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3026) * CLONE_SYSVSEM is equivalent to sys_exit().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3027) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3028) exit_sem(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3029) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3030) if (unshare_flags & CLONE_NEWIPC) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3031) /* Orphan segments in old ns (see sem above). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3032) exit_shm(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3033) shm_init_task(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3034) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3035)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3036) if (new_nsproxy)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3037) switch_task_namespaces(current, new_nsproxy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3039) task_lock(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3040)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3041) if (new_fs) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3042) fs = current->fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3043) spin_lock(&fs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3044) current->fs = new_fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3045) if (--fs->users)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3046) new_fs = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3047) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3048) new_fs = fs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3049) spin_unlock(&fs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3050) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3051)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3052) if (new_fd) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3053) fd = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3054) current->files = new_fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3055) new_fd = fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3056) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3057)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3058) task_unlock(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3059)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3060) if (new_cred) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3061) /* Install the new user namespace */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3062) commit_creds(new_cred);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3063) new_cred = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3064) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3065) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3066)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3067) perf_event_namespaces(current);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3068)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3069) bad_unshare_cleanup_cred:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3070) if (new_cred)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3071) put_cred(new_cred);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3072) bad_unshare_cleanup_fd:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3073) if (new_fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3074) put_files_struct(new_fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3076) bad_unshare_cleanup_fs:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3077) if (new_fs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3078) free_fs_struct(new_fs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3079)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3080) bad_unshare_out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3081) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3082) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3083)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3084) SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3085) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3086) return ksys_unshare(unshare_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3087) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3088)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3089) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3090) * Helper to unshare the files of the current task.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3091) * We don't want to expose copy_files internals to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3092) * the exec layer of the kernel.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3093) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3094)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3095) int unshare_files(struct files_struct **displaced)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3096) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3097) struct task_struct *task = current;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3098) struct files_struct *copy = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3099) int error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3101) error = unshare_fd(CLONE_FILES, NR_OPEN_MAX, ©);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3102) if (error || !copy) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3103) *displaced = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3104) return error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3105) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3106) *displaced = task->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3107) task_lock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3108) task->files = copy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3109) task_unlock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3110) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3111) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3112)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3113) int sysctl_max_threads(struct ctl_table *table, int write,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3114) void *buffer, size_t *lenp, loff_t *ppos)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3115) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3116) struct ctl_table t;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3117) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3118) int threads = max_threads;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3119) int min = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3120) int max = MAX_THREADS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3121)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3122) t = *table;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3123) t.data = &threads;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3124) t.extra1 = &min;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3125) t.extra2 = &max;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3127) ret = proc_dointvec_minmax(&t, write, buffer, lenp, ppos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3128) if (ret || !write)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3129) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3131) max_threads = threads;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3133) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3134) }
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags