^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * linux/fs/file.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Copyright (C) 1998-1999, Stephen Tweedie and Bill Hawes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * Manage the dynamic fd arrays in the process files_struct.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/syscalls.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/export.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/sched/signal.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/file.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/fdtable.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/bitops.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/spinlock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/rcupdate.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/close_range.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <net/sock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) unsigned int sysctl_nr_open __read_mostly = 1024*1024;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) unsigned int sysctl_nr_open_min = BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) /* our min() is unusable in constant expressions ;-/ */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #define __const_min(x, y) ((x) < (y) ? (x) : (y))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) unsigned int sysctl_nr_open_max =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) __const_min(INT_MAX, ~(size_t)0/sizeof(void *)) & -BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) static void __free_fdtable(struct fdtable *fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) kvfree(fdt->fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) kvfree(fdt->open_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) kfree(fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) static void free_fdtable_rcu(struct rcu_head *rcu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) __free_fdtable(container_of(rcu, struct fdtable, rcu));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) #define BITBIT_NR(nr) BITS_TO_LONGS(BITS_TO_LONGS(nr))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) #define BITBIT_SIZE(nr) (BITBIT_NR(nr) * sizeof(long))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) * Copy 'count' fd bits from the old table to the new table and clear the extra
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) * space if any. This does not copy the file pointers. Called with the files
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) * spinlock held for write.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) static void copy_fd_bitmaps(struct fdtable *nfdt, struct fdtable *ofdt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) unsigned int count)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) unsigned int cpy, set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) cpy = count / BITS_PER_BYTE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) set = (nfdt->max_fds - count) / BITS_PER_BYTE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) memcpy(nfdt->open_fds, ofdt->open_fds, cpy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) memset((char *)nfdt->open_fds + cpy, 0, set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) memcpy(nfdt->close_on_exec, ofdt->close_on_exec, cpy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) memset((char *)nfdt->close_on_exec + cpy, 0, set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) cpy = BITBIT_SIZE(count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) set = BITBIT_SIZE(nfdt->max_fds) - cpy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) memcpy(nfdt->full_fds_bits, ofdt->full_fds_bits, cpy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) memset((char *)nfdt->full_fds_bits + cpy, 0, set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) * Copy all file descriptors from the old table to the new, expanded table and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) * clear the extra space. Called with the files spinlock held for write.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) static void copy_fdtable(struct fdtable *nfdt, struct fdtable *ofdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) size_t cpy, set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) BUG_ON(nfdt->max_fds < ofdt->max_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) cpy = ofdt->max_fds * sizeof(struct file *);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) set = (nfdt->max_fds - ofdt->max_fds) * sizeof(struct file *);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) memcpy(nfdt->fd, ofdt->fd, cpy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) memset((char *)nfdt->fd + cpy, 0, set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) copy_fd_bitmaps(nfdt, ofdt, ofdt->max_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) * Note how the fdtable bitmap allocations very much have to be a multiple of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) * BITS_PER_LONG. This is not only because we walk those things in chunks of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) * 'unsigned long' in some places, but simply because that is how the Linux
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) * kernel bitmaps are defined to work: they are not "bits in an array of bytes",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) * they are very much "bits in an array of unsigned long".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) * The ALIGN(nr, BITS_PER_LONG) here is for clarity: since we just multiplied
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) * by that "1024/sizeof(ptr)" before, we already know there are sufficient
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) * clear low bits. Clang seems to realize that, gcc ends up being confused.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) * On a 128-bit machine, the ALIGN() would actually matter. In the meantime,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) * let's consider it documentation (and maybe a test-case for gcc to improve
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) * its code generation ;)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) static struct fdtable * alloc_fdtable(unsigned int nr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) void *data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) * Figure out how many fds we actually want to support in this fdtable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) * Allocation steps are keyed to the size of the fdarray, since it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) * grows far faster than any of the other dynamic data. We try to fit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) * the fdarray into comfortable page-tuned chunks: starting at 1024B
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) * and growing in powers of two from there on.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) nr /= (1024 / sizeof(struct file *));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) nr = roundup_pow_of_two(nr + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) nr *= (1024 / sizeof(struct file *));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) nr = ALIGN(nr, BITS_PER_LONG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) * Note that this can drive nr *below* what we had passed if sysctl_nr_open
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) * had been set lower between the check in expand_files() and here. Deal
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) * with that in caller, it's cheaper that way.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) * We make sure that nr remains a multiple of BITS_PER_LONG - otherwise
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) * bitmaps handling below becomes unpleasant, to put it mildly...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) if (unlikely(nr > sysctl_nr_open))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) nr = ((sysctl_nr_open - 1) | (BITS_PER_LONG - 1)) + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) fdt = kmalloc(sizeof(struct fdtable), GFP_KERNEL_ACCOUNT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) if (!fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) fdt->max_fds = nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) data = kvmalloc_array(nr, sizeof(struct file *), GFP_KERNEL_ACCOUNT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) if (!data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) goto out_fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) fdt->fd = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) data = kvmalloc(max_t(size_t,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) 2 * nr / BITS_PER_BYTE + BITBIT_SIZE(nr), L1_CACHE_BYTES),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) GFP_KERNEL_ACCOUNT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) if (!data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) goto out_arr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) fdt->open_fds = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) data += nr / BITS_PER_BYTE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) fdt->close_on_exec = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) data += nr / BITS_PER_BYTE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) fdt->full_fds_bits = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) return fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) out_arr:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) kvfree(fdt->fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) out_fdt:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) kfree(fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) * Expand the file descriptor table.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) * This function will allocate a new fdtable and both fd array and fdset, of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) * the given size.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) * Return <0 error code on error; 1 on successful completion.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) * The files->file_lock should be held on entry, and will be held on exit.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) static int expand_fdtable(struct files_struct *files, unsigned int nr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) __releases(files->file_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) __acquires(files->file_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) struct fdtable *new_fdt, *cur_fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) new_fdt = alloc_fdtable(nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) /* make sure all __fd_install() have seen resize_in_progress
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) * or have finished their rcu_read_lock_sched() section.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) if (atomic_read(&files->count) > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) synchronize_rcu();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) if (!new_fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) * extremely unlikely race - sysctl_nr_open decreased between the check in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) * caller and alloc_fdtable(). Cheaper to catch it here...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) if (unlikely(new_fdt->max_fds <= nr)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) __free_fdtable(new_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) return -EMFILE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) cur_fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) BUG_ON(nr < cur_fdt->max_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) copy_fdtable(new_fdt, cur_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) rcu_assign_pointer(files->fdt, new_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) if (cur_fdt != &files->fdtab)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) call_rcu(&cur_fdt->rcu, free_fdtable_rcu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) /* coupled with smp_rmb() in __fd_install() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) smp_wmb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) * Expand files.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) * This function will expand the file structures, if the requested size exceeds
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) * the current capacity and there is room for expansion.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) * Return <0 error code on error; 0 when nothing done; 1 when files were
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) * expanded and execution may have blocked.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) * The files->file_lock should be held on entry, and will be held on exit.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) static int expand_files(struct files_struct *files, unsigned int nr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) __releases(files->file_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) __acquires(files->file_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) int expanded = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) repeat:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) /* Do we need to expand? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) if (nr < fdt->max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) return expanded;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) /* Can we expand? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) if (nr >= sysctl_nr_open)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) return -EMFILE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) if (unlikely(files->resize_in_progress)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) expanded = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) wait_event(files->resize_wait, !files->resize_in_progress);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) goto repeat;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) /* All good, so we try */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) files->resize_in_progress = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) expanded = expand_fdtable(files, nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) files->resize_in_progress = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) wake_up_all(&files->resize_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) return expanded;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) static inline void __set_close_on_exec(unsigned int fd, struct fdtable *fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) __set_bit(fd, fdt->close_on_exec);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) static inline void __clear_close_on_exec(unsigned int fd, struct fdtable *fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) if (test_bit(fd, fdt->close_on_exec))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) __clear_bit(fd, fdt->close_on_exec);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) static inline void __set_open_fd(unsigned int fd, struct fdtable *fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) __set_bit(fd, fdt->open_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) fd /= BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) if (!~fdt->open_fds[fd])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) __set_bit(fd, fdt->full_fds_bits);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) static inline void __clear_open_fd(unsigned int fd, struct fdtable *fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) __clear_bit(fd, fdt->open_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) __clear_bit(fd / BITS_PER_LONG, fdt->full_fds_bits);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) static unsigned int count_open_files(struct fdtable *fdt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) unsigned int size = fdt->max_fds;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) /* Find the last open fd */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) for (i = size / BITS_PER_LONG; i > 0; ) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) if (fdt->open_fds[--i])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) i = (i + 1) * BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) return i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) * Note that a sane fdtable size always has to be a multiple of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) * BITS_PER_LONG, since we have bitmaps that are sized by this.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) * 'max_fds' will normally already be properly aligned, but it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) * turns out that in the close_range() -> __close_range() ->
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) * unshare_fd() -> dup_fd() -> sane_fdtable_size() we can end
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) * up having a 'max_fds' value that isn't already aligned.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) * Rather than make close_range() have to worry about this,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) * just make that BITS_PER_LONG alignment be part of a sane
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) * fdtable size. Becuase that's really what it is.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) static unsigned int sane_fdtable_size(struct fdtable *fdt, unsigned int max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) unsigned int count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) count = count_open_files(fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) if (max_fds < NR_OPEN_DEFAULT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) max_fds = NR_OPEN_DEFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) return ALIGN(min(count, max_fds), BITS_PER_LONG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) * Allocate a new files structure and copy contents from the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) * passed in files structure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) * errorp will be valid only when the returned files_struct is NULL.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) struct files_struct *dup_fd(struct files_struct *oldf, unsigned int max_fds, int *errorp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) struct files_struct *newf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) struct file **old_fds, **new_fds;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) unsigned int open_files, i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) struct fdtable *old_fdt, *new_fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) *errorp = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) newf = kmem_cache_alloc(files_cachep, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) if (!newf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) atomic_set(&newf->count, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) spin_lock_init(&newf->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) newf->resize_in_progress = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) init_waitqueue_head(&newf->resize_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) newf->next_fd = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) new_fdt = &newf->fdtab;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) new_fdt->max_fds = NR_OPEN_DEFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) new_fdt->close_on_exec = newf->close_on_exec_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) new_fdt->open_fds = newf->open_fds_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) new_fdt->full_fds_bits = newf->full_fds_bits_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) new_fdt->fd = &newf->fd_array[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) spin_lock(&oldf->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) old_fdt = files_fdtable(oldf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) open_files = sane_fdtable_size(old_fdt, max_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) * Check whether we need to allocate a larger fd array and fd set.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) while (unlikely(open_files > new_fdt->max_fds)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) spin_unlock(&oldf->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) if (new_fdt != &newf->fdtab)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) __free_fdtable(new_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) new_fdt = alloc_fdtable(open_files - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) if (!new_fdt) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) *errorp = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) goto out_release;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) /* beyond sysctl_nr_open; nothing to do */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) if (unlikely(new_fdt->max_fds < open_files)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) __free_fdtable(new_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) *errorp = -EMFILE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) goto out_release;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) * Reacquire the oldf lock and a pointer to its fd table
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) * who knows it may have a new bigger fd table. We need
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) * the latest pointer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) spin_lock(&oldf->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) old_fdt = files_fdtable(oldf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) open_files = sane_fdtable_size(old_fdt, max_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) copy_fd_bitmaps(new_fdt, old_fdt, open_files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) old_fds = old_fdt->fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) new_fds = new_fdt->fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) for (i = open_files; i != 0; i--) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) struct file *f = *old_fds++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) if (f) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) get_file(f);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) * The fd may be claimed in the fd bitmap but not yet
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) * instantiated in the files array if a sibling thread
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) * is partway through open(). So make sure that this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) * fd is available to the new process.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) __clear_open_fd(open_files - i, new_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) rcu_assign_pointer(*new_fds++, f);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) spin_unlock(&oldf->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) /* clear the remainder */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) memset(new_fds, 0, (new_fdt->max_fds - open_files) * sizeof(struct file *));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) rcu_assign_pointer(newf->fdt, new_fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) return newf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) out_release:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) kmem_cache_free(files_cachep, newf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) static struct fdtable *close_files(struct files_struct * files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) * It is safe to dereference the fd table without RCU or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) * ->file_lock because this is the last reference to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) * files structure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) struct fdtable *fdt = rcu_dereference_raw(files->fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) unsigned int i, j = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) for (;;) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) unsigned long set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) i = j * BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) if (i >= fdt->max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) set = fdt->open_fds[j++];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) while (set) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) if (set & 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) struct file * file = xchg(&fdt->fd[i], NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) if (file) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) filp_close(file, files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) i++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) set >>= 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) return fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) struct files_struct *get_files_struct(struct task_struct *task)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) struct files_struct *files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) task_lock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) files = task->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) if (files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) atomic_inc(&files->count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) task_unlock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) return files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) void put_files_struct(struct files_struct *files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) if (atomic_dec_and_test(&files->count)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) struct fdtable *fdt = close_files(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) /* free the arrays if they are not embedded */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) if (fdt != &files->fdtab)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) __free_fdtable(fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) kmem_cache_free(files_cachep, files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) void reset_files_struct(struct files_struct *files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) struct task_struct *tsk = current;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) struct files_struct *old;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) old = tsk->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) task_lock(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) tsk->files = files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) task_unlock(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) put_files_struct(old);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) void exit_files(struct task_struct *tsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) struct files_struct * files = tsk->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) if (files) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) task_lock(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) tsk->files = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) task_unlock(tsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) put_files_struct(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) struct files_struct init_files = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) .count = ATOMIC_INIT(1),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) .fdt = &init_files.fdtab,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) .fdtab = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) .max_fds = NR_OPEN_DEFAULT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) .fd = &init_files.fd_array[0],
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) .close_on_exec = init_files.close_on_exec_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) .open_fds = init_files.open_fds_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) .full_fds_bits = init_files.full_fds_bits_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) .file_lock = __SPIN_LOCK_UNLOCKED(init_files.file_lock),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) .resize_wait = __WAIT_QUEUE_HEAD_INITIALIZER(init_files.resize_wait),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) static unsigned int find_next_fd(struct fdtable *fdt, unsigned int start)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) unsigned int maxfd = fdt->max_fds;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) unsigned int maxbit = maxfd / BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) unsigned int bitbit = start / BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) bitbit = find_next_zero_bit(fdt->full_fds_bits, maxbit, bitbit) * BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) if (bitbit > maxfd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) return maxfd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) if (bitbit > start)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) start = bitbit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) return find_next_zero_bit(fdt->open_fds, maxfd, start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) * allocate a file descriptor, mark it busy.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) int __alloc_fd(struct files_struct *files,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) unsigned start, unsigned end, unsigned flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) unsigned int fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) int error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) repeat:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) fd = start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) if (fd < files->next_fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) fd = files->next_fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) if (fd < fdt->max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) fd = find_next_fd(fdt, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) * N.B. For clone tasks sharing a files structure, this test
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) * will limit the total number of files that can be opened.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) error = -EMFILE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) if (fd >= end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) error = expand_files(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) if (error < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) * If we needed to expand the fs array we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) * might have blocked - try again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) if (error)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) goto repeat;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) if (start <= files->next_fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) files->next_fd = fd + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) __set_open_fd(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) if (flags & O_CLOEXEC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) __set_close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) __clear_close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) error = fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) #if 1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) /* Sanity check */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) if (rcu_access_pointer(fdt->fd[fd]) != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) printk(KERN_WARNING "alloc_fd: slot %d not NULL!\n", fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) rcu_assign_pointer(fdt->fd[fd], NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) return error;
^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) static int alloc_fd(unsigned start, unsigned flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) return __alloc_fd(current->files, start, rlimit(RLIMIT_NOFILE), flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) int __get_unused_fd_flags(unsigned flags, unsigned long nofile)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) return __alloc_fd(current->files, 0, nofile, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) int get_unused_fd_flags(unsigned flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) return __get_unused_fd_flags(flags, rlimit(RLIMIT_NOFILE));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) EXPORT_SYMBOL(get_unused_fd_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) static void __put_unused_fd(struct files_struct *files, unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) struct fdtable *fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) __clear_open_fd(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) if (fd < files->next_fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) files->next_fd = fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) void put_unused_fd(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) __put_unused_fd(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) EXPORT_SYMBOL(put_unused_fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) * Install a file pointer in the fd array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) * The VFS is full of places where we drop the files lock between
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) * setting the open_fds bitmap and installing the file in the file
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) * array. At any such point, we are vulnerable to a dup2() race
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) * installing a file in the array before us. We need to detect this and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) * fput() the struct file we are about to overwrite in this case.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) * It should never happen - if we allow dup2() do it, _really_ bad things
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) * will follow.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) * NOTE: __fd_install() variant is really, really low-level; don't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) * use it unless you are forced to by truly lousy API shoved down
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) * your throat. 'files' *MUST* be either current->files or obtained
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) * by get_files_struct(current) done by whoever had given it to you,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) * or really bad things will happen. Normally you want to use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) * fd_install() instead.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) void __fd_install(struct files_struct *files, unsigned int fd,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) struct file *file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) rcu_read_lock_sched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) if (unlikely(files->resize_in_progress)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) rcu_read_unlock_sched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) BUG_ON(fdt->fd[fd] != NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) rcu_assign_pointer(fdt->fd[fd], file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) /* coupled with smp_wmb() in expand_fdtable() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) smp_rmb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) fdt = rcu_dereference_sched(files->fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) BUG_ON(fdt->fd[fd] != NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) rcu_assign_pointer(fdt->fd[fd], file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) rcu_read_unlock_sched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) * This consumes the "file" refcount, so callers should treat it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) * as if they had called fput(file).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) void fd_install(unsigned int fd, struct file *file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) __fd_install(current->files, fd, file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) EXPORT_SYMBOL(fd_install);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) static struct file *pick_file(struct files_struct *files, unsigned fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) struct file *file = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) if (fd >= fdt->max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) file = fdt->fd[fd];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) rcu_assign_pointer(fdt->fd[fd], NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) __put_unused_fd(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) out_unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) return file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) * The same warnings as for __alloc_fd()/__fd_install() apply here...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) int __close_fd(struct files_struct *files, unsigned fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) file = pick_file(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) return -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) return filp_close(file, files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) EXPORT_SYMBOL(__close_fd); /* for ksys_close() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) * __close_range() - Close all file descriptors in a given range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) * @fd: starting file descriptor to close
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) * @max_fd: last file descriptor to close
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) * This closes a range of file descriptors. All file descriptors
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) * from @fd up to and including @max_fd are closed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) int __close_range(unsigned fd, unsigned max_fd, unsigned int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) unsigned int cur_max;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) struct task_struct *me = current;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) struct files_struct *cur_fds = me->files, *fds = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) if (flags & ~CLOSE_RANGE_UNSHARE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) if (fd > max_fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) cur_max = files_fdtable(cur_fds)->max_fds;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) /* cap to last valid index into fdtable */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) cur_max--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) if (flags & CLOSE_RANGE_UNSHARE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) unsigned int max_unshare_fds = NR_OPEN_MAX;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) * If the requested range is greater than the current maximum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) * we're closing everything so only copy all file descriptors
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) * beneath the lowest file descriptor.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) if (max_fd >= cur_max)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) max_unshare_fds = fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) ret = unshare_fd(CLONE_FILES, max_unshare_fds, &fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) * We used to share our file descriptor table, and have now
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) * created a private one, make sure we're using it below.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) if (fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) swap(cur_fds, fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) max_fd = min(max_fd, cur_max);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) while (fd <= max_fd) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) file = pick_file(cur_fds, fd++);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) filp_close(file, cur_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) if (fds) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) * We're done closing the files we were supposed to. Time to install
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) * the new file descriptor table and drop the old one.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) task_lock(me);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) me->files = cur_fds;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) task_unlock(me);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) put_files_struct(fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) }
^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) * variant of __close_fd that gets a ref on the file for later fput.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) * The caller must ensure that filp_close() called on the file, and then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) * an fput().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) int __close_fd_get_file(unsigned int fd, struct file **res)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) if (fd >= fdt->max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) file = fdt->fd[fd];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) rcu_assign_pointer(fdt->fd[fd], NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) __put_unused_fd(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) get_file(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) *res = file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) out_unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) *res = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) return -ENOENT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) void do_close_on_exec(struct files_struct *files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) unsigned i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) /* exec unshares first */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) for (i = 0; ; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) unsigned long set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) unsigned fd = i * BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) if (fd >= fdt->max_fds)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) set = fdt->close_on_exec[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) if (!set)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) fdt->close_on_exec[i] = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) for ( ; set ; fd++, set >>= 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) if (!(set & 1))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) file = fdt->fd[fd];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) rcu_assign_pointer(fdt->fd[fd], NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) __put_unused_fd(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) filp_close(file, files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) static inline struct file *__fget_files_rcu(struct files_struct *files,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) unsigned int fd, fmode_t mask, unsigned int refs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) for (;;) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) struct fdtable *fdt = rcu_dereference_raw(files->fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) struct file __rcu **fdentry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) if (unlikely(fd >= fdt->max_fds))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) fdentry = fdt->fd + array_index_nospec(fd, fdt->max_fds);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) file = rcu_dereference_raw(*fdentry);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) if (unlikely(!file))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) if (unlikely(file->f_mode & mask))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) * Ok, we have a file pointer. However, because we do
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) * this all locklessly under RCU, we may be racing with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) * that file being closed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) * Such a race can take two forms:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) * (a) the file ref already went down to zero,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) * and get_file_rcu_many() fails. Just try
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) * again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) if (unlikely(!get_file_rcu_many(file, refs)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) * (b) the file table entry has changed under us.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) * Note that we don't need to re-check the 'fdt->fd'
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) * pointer having changed, because it always goes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) * hand-in-hand with 'fdt'.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) * If so, we need to put our refs and try again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) if (unlikely(rcu_dereference_raw(files->fdt) != fdt) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) unlikely(rcu_dereference_raw(*fdentry) != file)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) fput_many(file, refs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) * Ok, we have a ref to the file, and checked that it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) * still exists.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) return file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) static struct file *__fget_files(struct files_struct *files, unsigned int fd,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) fmode_t mask, unsigned int refs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) file = __fget_files_rcu(files, fd, mask, refs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) return file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) static inline struct file *__fget(unsigned int fd, fmode_t mask,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) unsigned int refs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) return __fget_files(current->files, fd, mask, refs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) struct file *fget_many(unsigned int fd, unsigned int refs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) return __fget(fd, FMODE_PATH, refs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) struct file *fget(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) return __fget(fd, FMODE_PATH, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) EXPORT_SYMBOL(fget);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) struct file *fget_raw(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) return __fget(fd, 0, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) EXPORT_SYMBOL(fget_raw);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) struct file *fget_task(struct task_struct *task, unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) struct file *file = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) task_lock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) if (task->files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) file = __fget_files(task->files, fd, 0, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) task_unlock(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) return file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) }
^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) * Lightweight file lookup - no refcnt increment if fd table isn't shared.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) * You can use this instead of fget if you satisfy all of the following
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) * conditions:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) * 1) You must call fput_light before exiting the syscall and returning control
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) * to userspace (i.e. you cannot remember the returned struct file * after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) * returning to userspace).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) * 2) You must not call filp_close on the returned struct file * in between
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) * calls to fget_light and fput_light.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) * 3) You must not clone the current task in between the calls to fget_light
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) * and fput_light.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) * The fput_needed flag returned by fget_light should be passed to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) * corresponding fput_light.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) static unsigned long __fget_light(unsigned int fd, fmode_t mask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) if (atomic_read(&files->count) == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) file = __fcheck_files(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) if (!file || unlikely(file->f_mode & mask))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) return (unsigned long)file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) file = __fget(fd, mask, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) return FDPUT_FPUT | (unsigned long)file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) unsigned long __fdget(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) return __fget_light(fd, FMODE_PATH);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) EXPORT_SYMBOL(__fdget);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) unsigned long __fdget_raw(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) return __fget_light(fd, 0);
^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) unsigned long __fdget_pos(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) unsigned long v = __fdget(fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) struct file *file = (struct file *)(v & ~3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) if (file && (file->f_mode & FMODE_ATOMIC_POS)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) if (file_count(file) > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) v |= FDPUT_POS_UNLOCK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) mutex_lock(&file->f_pos_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) return v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) void __f_unlock_pos(struct file *f)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) mutex_unlock(&f->f_pos_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) * We only lock f_pos if we have threads or if the file might be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) * shared with another process. In both cases we'll have an elevated
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) * file count (done either by fdget() or by fork()).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) void set_close_on_exec(unsigned int fd, int flag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) if (flag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) __set_close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) __clear_close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) bool get_close_on_exec(unsigned int fd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) bool res;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) res = close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) return res;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) static int do_dup2(struct files_struct *files,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) struct file *file, unsigned fd, unsigned flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) __releases(&files->file_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) struct file *tofree;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) * We need to detect attempts to do dup2() over allocated but still
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) * not finished descriptor. NB: OpenBSD avoids that at the price of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) * extra work in their equivalent of fget() - they insert struct
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) * file immediately after grabbing descriptor, mark it larval if
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) * more work (e.g. actual opening) is needed and make sure that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) * fget() treats larval files as absent. Potentially interesting,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) * but while extra work in fget() is trivial, locking implications
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) * and amount of surgery on open()-related paths in VFS are not.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) * FreeBSD fails with -EBADF in the same situation, NetBSD "solution"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) * deadlocks in rather amusing ways, AFAICS. All of that is out of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) * scope of POSIX or SUS, since neither considers shared descriptor
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) * tables and this condition does not arise without those.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) fdt = files_fdtable(files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) tofree = fdt->fd[fd];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) if (!tofree && fd_is_open(fd, fdt))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) goto Ebusy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) get_file(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) rcu_assign_pointer(fdt->fd[fd], file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) __set_open_fd(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) if (flags & O_CLOEXEC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) __set_close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) __clear_close_on_exec(fd, fdt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) if (tofree)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) filp_close(tofree, files);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) return fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) Ebusy:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) return -EBUSY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) int replace_fd(unsigned fd, struct file *file, unsigned flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) return __close_fd(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) if (fd >= rlimit(RLIMIT_NOFILE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) return -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) err = expand_files(files, fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) if (unlikely(err < 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) return do_dup2(files, file, fd, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) out_unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) return err;
^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) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) * __receive_fd() - Install received file into file descriptor table
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) * @fd: fd to install into (if negative, a new fd will be allocated)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) * @file: struct file that was received from another process
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) * @ufd: __user pointer to write new fd number to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) * @o_flags: the O_* flags to apply to the new fd entry
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) * Installs a received file into the file descriptor table, with appropriate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) * checks and count updates. Optionally writes the fd number to userspace, if
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) * @ufd is non-NULL.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) * This helper handles its own reference counting of the incoming
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) * struct file.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) * Returns newly install fd or -ve on error.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) int __receive_fd(int fd, struct file *file, int __user *ufd, unsigned int o_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) int new_fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) int error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) error = security_file_receive(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) if (error)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) return error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) if (fd < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) new_fd = get_unused_fd_flags(o_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) if (new_fd < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) return new_fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) new_fd = fd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) if (ufd) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) error = put_user(new_fd, ufd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) if (error) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) if (fd < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) put_unused_fd(new_fd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) return error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) if (fd < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) fd_install(new_fd, get_file(file));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) error = replace_fd(new_fd, file, o_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) if (error)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) return error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) /* Bump the sock usage counts, if any. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) __receive_sock(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) return new_fd;
^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) static int ksys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) int err = -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) if ((flags & ~O_CLOEXEC) != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) if (unlikely(oldfd == newfd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) if (newfd >= rlimit(RLIMIT_NOFILE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) return -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) err = expand_files(files, newfd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) file = fcheck(oldfd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) if (unlikely(!file))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) goto Ebadf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) if (unlikely(err < 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) if (err == -EMFILE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) goto Ebadf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) return do_dup2(files, file, newfd, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) Ebadf:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) err = -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) out_unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) SYSCALL_DEFINE3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) return ksys_dup3(oldfd, newfd, flags);
^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) SYSCALL_DEFINE2(dup2, unsigned int, oldfd, unsigned int, newfd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) if (unlikely(newfd == oldfd)) { /* corner case */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) struct files_struct *files = current->files;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) int retval = oldfd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) rcu_read_lock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) if (!fcheck_files(files, oldfd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) retval = -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) rcu_read_unlock();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) return retval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) return ksys_dup3(oldfd, newfd, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) SYSCALL_DEFINE1(dup, unsigned int, fildes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) int ret = -EBADF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) struct file *file = fget_raw(fildes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) if (file) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) ret = get_unused_fd_flags(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) if (ret >= 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) fd_install(ret, file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) fput(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) int f_dupfd(unsigned int from, struct file *file, unsigned flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) if (from >= rlimit(RLIMIT_NOFILE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) err = alloc_fd(from, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) if (err >= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) get_file(file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) fd_install(err, file);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) int iterate_fd(struct files_struct *files, unsigned n,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) int (*f)(const void *, struct file *, unsigned),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) const void *p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) struct fdtable *fdt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) int res = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) if (!files)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) spin_lock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) for (fdt = files_fdtable(files); n < fdt->max_fds; n++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) struct file *file;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) file = rcu_dereference_check_fdtable(files, fdt->fd[n]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) if (!file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) res = f(p, file, n);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) if (res)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) spin_unlock(&files->file_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) return res;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) EXPORT_SYMBOL(iterate_fd);
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags