summaryrefslogtreecommitdiff
path: root/src/basic/mount-util.c
blob: bcb7e43288b7b8ebcaec124f529f246efb97aa5e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
/* SPDX-License-Identifier: LGPL-2.1+ */
/***
  Copyright 2010 Lennart Poettering
***/

#include <errno.h>
#include <stdio_ext.h>
#include <stdlib.h>
#include <string.h>
#include <sys/mount.h>
#include <sys/stat.h>
#include <sys/statvfs.h>
#include <unistd.h>

/* Include later */
#include <libmount.h>

#include "alloc-util.h"
#include "escape.h"
#include "extract-word.h"
#include "fd-util.h"
#include "fileio.h"
#include "fs-util.h"
#include "hashmap.h"
#include "mount-util.h"
#include "parse-util.h"
#include "path-util.h"
#include "set.h"
#include "stdio-util.h"
#include "string-util.h"
#include "strv.h"

/* This is the original MAX_HANDLE_SZ definition from the kernel, when the API was introduced. We use that in place of
 * any more currently defined value to future-proof things: if the size is increased in the API headers, and our code
 * is recompiled then it would cease working on old kernels, as those refuse any sizes larger than this value with
 * EINVAL right-away. Hence, let's disconnect ourselves from any such API changes, and stick to the original definition
 * from when it was introduced. We use it as a start value only anyway (see below), and hence should be able to deal
 * with large file handles anyway. */
#define ORIGINAL_MAX_HANDLE_SZ 128

int name_to_handle_at_loop(
                int fd,
                const char *path,
                struct file_handle **ret_handle,
                int *ret_mnt_id,
                int flags) {

        _cleanup_free_ struct file_handle *h = NULL;
        size_t n = ORIGINAL_MAX_HANDLE_SZ;

        /* We need to invoke name_to_handle_at() in a loop, given that it might return EOVERFLOW when the specified
         * buffer is too small. Note that in contrast to what the docs might suggest, MAX_HANDLE_SZ is only good as a
         * start value, it is not an upper bound on the buffer size required.
         *
         * This improves on raw name_to_handle_at() also in one other regard: ret_handle and ret_mnt_id can be passed
         * as NULL if there's no interest in either. */

        for (;;) {
                int mnt_id = -1;

                h = malloc0(offsetof(struct file_handle, f_handle) + n);
                if (!h)
                        return -ENOMEM;

                h->handle_bytes = n;

                if (name_to_handle_at(fd, path, h, &mnt_id, flags) >= 0) {

                        if (ret_handle)
                                *ret_handle = TAKE_PTR(h);

                        if (ret_mnt_id)
                                *ret_mnt_id = mnt_id;

                        return 0;
                }
                if (errno != EOVERFLOW)
                        return -errno;

                if (!ret_handle && ret_mnt_id && mnt_id >= 0) {

                        /* As it appears, name_to_handle_at() fills in mnt_id even when it returns EOVERFLOW when the
                         * buffer is too small, but that's undocumented. Hence, let's make use of this if it appears to
                         * be filled in, and the caller was interested in only the mount ID an nothing else. */

                        *ret_mnt_id = mnt_id;
                        return 0;
                }

                /* If name_to_handle_at() didn't increase the byte size, then this EOVERFLOW is caused by something
                 * else (apparently EOVERFLOW is returned for untriggered nfs4 mounts sometimes), not by the too small
                 * buffer. In that case propagate EOVERFLOW */
                if (h->handle_bytes <= n)
                        return -EOVERFLOW;

                /* The buffer was too small. Size the new buffer by what name_to_handle_at() returned. */
                n = h->handle_bytes;
                if (offsetof(struct file_handle, f_handle) + n < n) /* check for addition overflow */
                        return -EOVERFLOW;

                h = mfree(h);
        }
}

static int fd_fdinfo_mnt_id(int fd, const char *filename, int flags, int *mnt_id) {
        char path[STRLEN("/proc/self/fdinfo/") + DECIMAL_STR_MAX(int)];
        _cleanup_free_ char *fdinfo = NULL;
        _cleanup_close_ int subfd = -1;
        char *p;
        int r;

        if ((flags & AT_EMPTY_PATH) && isempty(filename))
                xsprintf(path, "/proc/self/fdinfo/%i", fd);
        else {
                subfd = openat(fd, filename, O_CLOEXEC|O_PATH);
                if (subfd < 0)
                        return -errno;

                xsprintf(path, "/proc/self/fdinfo/%i", subfd);
        }

        r = read_full_file(path, &fdinfo, NULL);
        if (r == -ENOENT) /* The fdinfo directory is a relatively new addition */
                return -EOPNOTSUPP;
        if (r < 0)
                return r;

        p = startswith(fdinfo, "mnt_id:");
        if (!p) {
                p = strstr(fdinfo, "\nmnt_id:");
                if (!p) /* The mnt_id field is a relatively new addition */
                        return -EOPNOTSUPP;

                p += 8;
        }

        p += strspn(p, WHITESPACE);
        p[strcspn(p, WHITESPACE)] = 0;

        return safe_atoi(p, mnt_id);
}

int fd_is_mount_point(int fd, const char *filename, int flags) {
        _cleanup_free_ struct file_handle *h = NULL, *h_parent = NULL;
        int mount_id = -1, mount_id_parent = -1;
        bool nosupp = false, check_st_dev = true;
        struct stat a, b;
        int r;

        assert(fd >= 0);
        assert(filename);

        /* First we will try the name_to_handle_at() syscall, which
         * tells us the mount id and an opaque file "handle". It is
         * not supported everywhere though (kernel compile-time
         * option, not all file systems are hooked up). If it works
         * the mount id is usually good enough to tell us whether
         * something is a mount point.
         *
         * If that didn't work we will try to read the mount id from
         * /proc/self/fdinfo/<fd>. This is almost as good as
         * name_to_handle_at(), however, does not return the
         * opaque file handle. The opaque file handle is pretty useful
         * to detect the root directory, which we should always
         * consider a mount point. Hence we use this only as
         * fallback. Exporting the mnt_id in fdinfo is a pretty recent
         * kernel addition.
         *
         * As last fallback we do traditional fstat() based st_dev
         * comparisons. This is how things were traditionally done,
         * but unionfs breaks this since it exposes file
         * systems with a variety of st_dev reported. Also, btrfs
         * subvolumes have different st_dev, even though they aren't
         * real mounts of their own. */

        r = name_to_handle_at_loop(fd, filename, &h, &mount_id, flags);
        if (IN_SET(r, -ENOSYS, -EACCES, -EPERM, -EOVERFLOW, -EINVAL))
                /* This kernel does not support name_to_handle_at() at all (ENOSYS), or the syscall was blocked
                 * (EACCES/EPERM; maybe through seccomp, because we are running inside of a container?), or the mount
                 * point is not triggered yet (EOVERFLOW, think nfs4), or some general name_to_handle_at() flakiness
                 * (EINVAL): fall back to simpler logic. */
                goto fallback_fdinfo;
        else if (r == -EOPNOTSUPP)
                /* This kernel or file system does not support name_to_handle_at(), hence let's see if the upper fs
                 * supports it (in which case it is a mount point), otherwise fallback to the traditional stat()
                 * logic */
                nosupp = true;
        else if (r < 0)
                return r;

        r = name_to_handle_at_loop(fd, "", &h_parent, &mount_id_parent, AT_EMPTY_PATH);
        if (r == -EOPNOTSUPP) {
                if (nosupp)
                        /* Neither parent nor child do name_to_handle_at()?  We have no choice but to fall back. */
                        goto fallback_fdinfo;
                else
                        /* The parent can't do name_to_handle_at() but the directory we are interested in can?  If so,
                         * it must be a mount point. */
                        return 1;
        } else if (r < 0)
                return r;

        /* The parent can do name_to_handle_at() but the
         * directory we are interested in can't? If so, it
         * must be a mount point. */
        if (nosupp)
                return 1;

        /* If the file handle for the directory we are
         * interested in and its parent are identical, we
         * assume this is the root directory, which is a mount
         * point. */

        if (h->handle_bytes == h_parent->handle_bytes &&
            h->handle_type == h_parent->handle_type &&
            memcmp(h->f_handle, h_parent->f_handle, h->handle_bytes) == 0)
                return 1;

        return mount_id != mount_id_parent;

fallback_fdinfo:
        r = fd_fdinfo_mnt_id(fd, filename, flags, &mount_id);
        if (IN_SET(r, -EOPNOTSUPP, -EACCES, -EPERM))
                goto fallback_fstat;
        if (r < 0)
                return r;

        r = fd_fdinfo_mnt_id(fd, "", AT_EMPTY_PATH, &mount_id_parent);
        if (r < 0)
                return r;

        if (mount_id != mount_id_parent)
                return 1;

        /* Hmm, so, the mount ids are the same. This leaves one
         * special case though for the root file system. For that,
         * let's see if the parent directory has the same inode as we
         * are interested in. Hence, let's also do fstat() checks now,
         * too, but avoid the st_dev comparisons, since they aren't
         * that useful on unionfs mounts. */
        check_st_dev = false;

fallback_fstat:
        /* yay for fstatat() taking a different set of flags than the other
         * _at() above */
        if (flags & AT_SYMLINK_FOLLOW)
                flags &= ~AT_SYMLINK_FOLLOW;
        else
                flags |= AT_SYMLINK_NOFOLLOW;
        if (fstatat(fd, filename, &a, flags) < 0)
                return -errno;

        if (fstatat(fd, "", &b, AT_EMPTY_PATH) < 0)
                return -errno;

        /* A directory with same device and inode as its parent? Must
         * be the root directory */
        if (a.st_dev == b.st_dev &&
            a.st_ino == b.st_ino)
                return 1;

        return check_st_dev && (a.st_dev != b.st_dev);
}

/* flags can be AT_SYMLINK_FOLLOW or 0 */
int path_is_mount_point(const char *t, const char *root, int flags) {
        _cleanup_free_ char *canonical = NULL, *parent = NULL;
        _cleanup_close_ int fd = -1;
        int r;

        assert(t);
        assert((flags & ~AT_SYMLINK_FOLLOW) == 0);

        if (path_equal(t, "/"))
                return 1;

        /* we need to resolve symlinks manually, we can't just rely on
         * fd_is_mount_point() to do that for us; if we have a structure like
         * /bin -> /usr/bin/ and /usr is a mount point, then the parent that we
         * look at needs to be /usr, not /. */
        if (flags & AT_SYMLINK_FOLLOW) {
                r = chase_symlinks(t, root, CHASE_TRAIL_SLASH, &canonical);
                if (r < 0)
                        return r;

                t = canonical;
        }

        parent = dirname_malloc(t);
        if (!parent)
                return -ENOMEM;

        fd = openat(AT_FDCWD, parent, O_DIRECTORY|O_CLOEXEC|O_PATH);
        if (fd < 0)
                return -errno;

        return fd_is_mount_point(fd, last_path_component(t), flags);
}

int path_get_mnt_id(const char *path, int *ret) {
        int r;

        r = name_to_handle_at_loop(AT_FDCWD, path, NULL, ret, 0);
        if (IN_SET(r, -EOPNOTSUPP, -ENOSYS, -EACCES, -EPERM, -EOVERFLOW, -EINVAL)) /* kernel/fs don't support this, or seccomp blocks access, or untriggered mount, or name_to_handle_at() is flaky */
                return fd_fdinfo_mnt_id(AT_FDCWD, path, 0, ret);

        return r;
}

int umount_recursive(const char *prefix, int flags) {
        bool again;
        int n = 0, r;

        /* Try to umount everything recursively below a
         * directory. Also, take care of stacked mounts, and keep
         * unmounting them until they are gone. */

        do {
                _cleanup_fclose_ FILE *proc_self_mountinfo = NULL;

                again = false;
                r = 0;

                proc_self_mountinfo = fopen("/proc/self/mountinfo", "re");
                if (!proc_self_mountinfo)
                        return -errno;

                (void) __fsetlocking(proc_self_mountinfo, FSETLOCKING_BYCALLER);

                for (;;) {
                        _cleanup_free_ char *path = NULL, *p = NULL;
                        int k;

                        k = fscanf(proc_self_mountinfo,
                                   "%*s "       /* (1) mount id */
                                   "%*s "       /* (2) parent id */
                                   "%*s "       /* (3) major:minor */
                                   "%*s "       /* (4) root */
                                   "%ms "       /* (5) mount point */
                                   "%*s"        /* (6) mount options */
                                   "%*[^-]"     /* (7) optional fields */
                                   "- "         /* (8) separator */
                                   "%*s "       /* (9) file system type */
                                   "%*s"        /* (10) mount source */
                                   "%*s"        /* (11) mount options 2 */
                                   "%*[^\n]",   /* some rubbish at the end */
                                   &path);
                        if (k != 1) {
                                if (k == EOF)
                                        break;

                                continue;
                        }

                        r = cunescape(path, UNESCAPE_RELAX, &p);
                        if (r < 0)
                                return r;

                        if (!path_startswith(p, prefix))
                                continue;

                        if (umount2(p, flags) < 0) {
                                r = log_debug_errno(errno, "Failed to umount %s: %m", p);
                                continue;
                        }

                        log_debug("Successfully unmounted %s", p);

                        again = true;
                        n++;

                        break;
                }

        } while (again);

        return r ? r : n;
}

static int get_mount_flags(const char *path, unsigned long *flags) {
        struct statvfs buf;

        if (statvfs(path, &buf) < 0)
                return -errno;
        *flags = buf.f_flag;
        return 0;
}

/* Use this function only if do you have direct access to /proc/self/mountinfo
 * and need the caller to open it for you. This is the case when /proc is
 * masked or not mounted. Otherwise, use bind_remount_recursive. */
int bind_remount_recursive_with_mountinfo(const char *prefix, bool ro, char **blacklist, FILE *proc_self_mountinfo) {
        _cleanup_set_free_free_ Set *done = NULL;
        _cleanup_free_ char *cleaned = NULL;
        int r;

        assert(proc_self_mountinfo);

        /* Recursively remount a directory (and all its submounts) read-only or read-write. If the directory is already
         * mounted, we reuse the mount and simply mark it MS_BIND|MS_RDONLY (or remove the MS_RDONLY for read-write
         * operation). If it isn't we first make it one. Afterwards we apply MS_BIND|MS_RDONLY (or remove MS_RDONLY) to
         * all submounts we can access, too. When mounts are stacked on the same mount point we only care for each
         * individual "top-level" mount on each point, as we cannot influence/access the underlying mounts anyway. We
         * do not have any effect on future submounts that might get propagated, they migt be writable. This includes
         * future submounts that have been triggered via autofs.
         *
         * If the "blacklist" parameter is specified it may contain a list of subtrees to exclude from the
         * remount operation. Note that we'll ignore the blacklist for the top-level path. */

        cleaned = strdup(prefix);
        if (!cleaned)
                return -ENOMEM;

        path_simplify(cleaned, false);

        done = set_new(&path_hash_ops);
        if (!done)
                return -ENOMEM;

        for (;;) {
                _cleanup_set_free_free_ Set *todo = NULL;
                bool top_autofs = false;
                char *x;
                unsigned long orig_flags;

                todo = set_new(&path_hash_ops);
                if (!todo)
                        return -ENOMEM;

                rewind(proc_self_mountinfo);

                for (;;) {
                        _cleanup_free_ char *path = NULL, *p = NULL, *type = NULL;
                        int k;

                        k = fscanf(proc_self_mountinfo,
                                   "%*s "       /* (1) mount id */
                                   "%*s "       /* (2) parent id */
                                   "%*s "       /* (3) major:minor */
                                   "%*s "       /* (4) root */
                                   "%ms "       /* (5) mount point */
                                   "%*s"        /* (6) mount options (superblock) */
                                   "%*[^-]"     /* (7) optional fields */
                                   "- "         /* (8) separator */
                                   "%ms "       /* (9) file system type */
                                   "%*s"        /* (10) mount source */
                                   "%*s"        /* (11) mount options (bind mount) */
                                   "%*[^\n]",   /* some rubbish at the end */
                                   &path,
                                   &type);
                        if (k != 2) {
                                if (k == EOF)
                                        break;

                                continue;
                        }

                        r = cunescape(path, UNESCAPE_RELAX, &p);
                        if (r < 0)
                                return r;

                        if (!path_startswith(p, cleaned))
                                continue;

                        /* Ignore this mount if it is blacklisted, but only if it isn't the top-level mount we shall
                         * operate on. */
                        if (!path_equal(cleaned, p)) {
                                bool blacklisted = false;
                                char **i;

                                STRV_FOREACH(i, blacklist) {

                                        if (path_equal(*i, cleaned))
                                                continue;

                                        if (!path_startswith(*i, cleaned))
                                                continue;

                                        if (path_startswith(p, *i)) {
                                                blacklisted = true;
                                                log_debug("Not remounting %s, because blacklisted by %s, called for %s", p, *i, cleaned);
                                                break;
                                        }
                                }
                                if (blacklisted)
                                        continue;
                        }

                        /* Let's ignore autofs mounts.  If they aren't
                         * triggered yet, we want to avoid triggering
                         * them, as we don't make any guarantees for
                         * future submounts anyway.  If they are
                         * already triggered, then we will find
                         * another entry for this. */
                        if (streq(type, "autofs")) {
                                top_autofs = top_autofs || path_equal(cleaned, p);
                                continue;
                        }

                        if (!set_contains(done, p)) {
                                r = set_consume(todo, p);
                                p = NULL;
                                if (r == -EEXIST)
                                        continue;
                                if (r < 0)
                                        return r;
                        }
                }

                /* If we have no submounts to process anymore and if
                 * the root is either already done, or an autofs, we
                 * are done */
                if (set_isempty(todo) &&
                    (top_autofs || set_contains(done, cleaned)))
                        return 0;

                if (!set_contains(done, cleaned) &&
                    !set_contains(todo, cleaned)) {
                        /* The prefix directory itself is not yet a mount, make it one. */
                        if (mount(cleaned, cleaned, NULL, MS_BIND|MS_REC, NULL) < 0)
                                return -errno;

                        orig_flags = 0;
                        (void) get_mount_flags(cleaned, &orig_flags);
                        orig_flags &= ~MS_RDONLY;

                        if (mount(NULL, prefix, NULL, orig_flags|MS_BIND|MS_REMOUNT|(ro ? MS_RDONLY : 0), NULL) < 0)
                                return -errno;

                        log_debug("Made top-level directory %s a mount point.", prefix);

                        x = strdup(cleaned);
                        if (!x)
                                return -ENOMEM;

                        r = set_consume(done, x);
                        if (r < 0)
                                return r;
                }

                while ((x = set_steal_first(todo))) {

                        r = set_consume(done, x);
                        if (IN_SET(r, 0, -EEXIST))
                                continue;
                        if (r < 0)
                                return r;

                        /* Deal with mount points that are obstructed by a later mount */
                        r = path_is_mount_point(x, NULL, 0);
                        if (IN_SET(r, 0, -ENOENT))
                                continue;
                        if (r < 0)
                                return r;

                        /* Try to reuse the original flag set */
                        orig_flags = 0;
                        (void) get_mount_flags(x, &orig_flags);
                        orig_flags &= ~MS_RDONLY;

                        if (mount(NULL, x, NULL, orig_flags|MS_BIND|MS_REMOUNT|(ro ? MS_RDONLY : 0), NULL) < 0)
                                return -errno;

                        log_debug("Remounted %s read-only.", x);
                }
        }
}

int bind_remount_recursive(const char *prefix, bool ro, char **blacklist) {
        _cleanup_fclose_ FILE *proc_self_mountinfo = NULL;

        proc_self_mountinfo = fopen("/proc/self/mountinfo", "re");
        if (!proc_self_mountinfo)
                return -errno;

        (void) __fsetlocking(proc_self_mountinfo, FSETLOCKING_BYCALLER);

        return bind_remount_recursive_with_mountinfo(prefix, ro, blacklist, proc_self_mountinfo);
}

int mount_move_root(const char *path) {
        assert(path);

        if (chdir(path) < 0)
                return -errno;

        if (mount(path, "/", NULL, MS_MOVE, NULL) < 0)
                return -errno;

        if (chroot(".") < 0)
                return -errno;

        if (chdir("/") < 0)
                return -errno;

        return 0;
}

bool fstype_is_network(const char *fstype) {
        const char *x;

        x = startswith(fstype, "fuse.");
        if (x)
                fstype = x;

        return STR_IN_SET(fstype,
                          "afs",
                          "cifs",
                          "smbfs",
                          "sshfs",
                          "ncpfs",
                          "ncp",
                          "nfs",
                          "nfs4",
                          "gfs",
                          "gfs2",
                          "glusterfs",
                          "pvfs2", /* OrangeFS */
                          "ocfs2",
                          "lustre");
}

bool fstype_is_api_vfs(const char *fstype) {
        return STR_IN_SET(fstype,
                          "autofs",
                          "bpf",
                          "cgroup",
                          "cgroup2",
                          "configfs",
                          "cpuset",
                          "debugfs",
                          "devpts",
                          "devtmpfs",
                          "efivarfs",
                          "fusectl",
                          "hugetlbfs",
                          "mqueue",
                          "proc",
                          "pstore",
                          "ramfs",
                          "securityfs",
                          "sysfs",
                          "tmpfs",
                          "tracefs");
}

bool fstype_is_ro(const char *fstype) {
        /* All Linux file systems that are necessarily read-only */
        return STR_IN_SET(fstype,
                          "DM_verity_hash",
                          "iso9660",
                          "squashfs");
}

bool fstype_can_discard(const char *fstype) {
        return STR_IN_SET(fstype,
                          "btrfs",
                          "ext4",
                          "vfat",
                          "xfs");
}

bool fstype_can_uid_gid(const char *fstype) {

        /* All file systems that have a uid=/gid= mount option that fixates the owners of all files and directories,
         * current and future. */

        return STR_IN_SET(fstype,
                          "adfs",
                          "fat",
                          "hfs",
                          "hpfs",
                          "iso9660",
                          "msdos",
                          "ntfs",
                          "vfat");
}

int repeat_unmount(const char *path, int flags) {
        bool done = false;

        assert(path);

        /* If there are multiple mounts on a mount point, this
         * removes them all */

        for (;;) {
                if (umount2(path, flags) < 0) {

                        if (errno == EINVAL)
                                return done;

                        return -errno;
                }

                done = true;
        }
}

const char* mode_to_inaccessible_node(mode_t mode) {
        /* This function maps a node type to a corresponding inaccessible file node. These nodes are created during
         * early boot by PID 1. In some cases we lacked the privs to create the character and block devices (maybe
         * because we run in an userns environment, or miss CAP_SYS_MKNOD, or run with a devices policy that excludes
         * device nodes with major and minor of 0), but that's fine, in that case we use an AF_UNIX file node instead,
         * which is not the same, but close enough for most uses. And most importantly, the kernel allows bind mounts
         * from socket nodes to any non-directory file nodes, and that's the most important thing that matters. */

        switch(mode & S_IFMT) {
                case S_IFREG:
                        return "/run/systemd/inaccessible/reg";

                case S_IFDIR:
                        return "/run/systemd/inaccessible/dir";

                case S_IFCHR:
                        if (access("/run/systemd/inaccessible/chr", F_OK) == 0)
                                return "/run/systemd/inaccessible/chr";
                        return "/run/systemd/inaccessible/sock";

                case S_IFBLK:
                        if (access("/run/systemd/inaccessible/blk", F_OK) == 0)
                                return "/run/systemd/inaccessible/blk";
                        return "/run/systemd/inaccessible/sock";

                case S_IFIFO:
                        return "/run/systemd/inaccessible/fifo";

                case S_IFSOCK:
                        return "/run/systemd/inaccessible/sock";
        }
        return NULL;
}

#define FLAG(name) (flags & name ? STRINGIFY(name) "|" : "")
static char* mount_flags_to_string(long unsigned flags) {
        char *x;
        _cleanup_free_ char *y = NULL;
        long unsigned overflow;

        overflow = flags & ~(MS_RDONLY |
                             MS_NOSUID |
                             MS_NODEV |
                             MS_NOEXEC |
                             MS_SYNCHRONOUS |
                             MS_REMOUNT |
                             MS_MANDLOCK |
                             MS_DIRSYNC |
                             MS_NOATIME |
                             MS_NODIRATIME |
                             MS_BIND |
                             MS_MOVE |
                             MS_REC |
                             MS_SILENT |
                             MS_POSIXACL |
                             MS_UNBINDABLE |
                             MS_PRIVATE |
                             MS_SLAVE |
                             MS_SHARED |
                             MS_RELATIME |
                             MS_KERNMOUNT |
                             MS_I_VERSION |
                             MS_STRICTATIME |
                             MS_LAZYTIME);

        if (flags == 0 || overflow != 0)
                if (asprintf(&y, "%lx", overflow) < 0)
                        return NULL;

        x = strjoin(FLAG(MS_RDONLY),
                    FLAG(MS_NOSUID),
                    FLAG(MS_NODEV),
                    FLAG(MS_NOEXEC),
                    FLAG(MS_SYNCHRONOUS),
                    FLAG(MS_REMOUNT),
                    FLAG(MS_MANDLOCK),
                    FLAG(MS_DIRSYNC),
                    FLAG(MS_NOATIME),
                    FLAG(MS_NODIRATIME),
                    FLAG(MS_BIND),
                    FLAG(MS_MOVE),
                    FLAG(MS_REC),
                    FLAG(MS_SILENT),
                    FLAG(MS_POSIXACL),
                    FLAG(MS_UNBINDABLE),
                    FLAG(MS_PRIVATE),
                    FLAG(MS_SLAVE),
                    FLAG(MS_SHARED),
                    FLAG(MS_RELATIME),
                    FLAG(MS_KERNMOUNT),
                    FLAG(MS_I_VERSION),
                    FLAG(MS_STRICTATIME),
                    FLAG(MS_LAZYTIME),
                    y);
        if (!x)
                return NULL;
        if (!y)
                x[strlen(x) - 1] = '\0'; /* truncate the last | */
        return x;
}

int mount_verbose(
                int error_log_level,
                const char *what,
                const char *where,
                const char *type,
                unsigned long flags,
                const char *options) {

        _cleanup_free_ char *fl = NULL, *o = NULL;
        unsigned long f;
        int r;

        r = mount_option_mangle(options, flags, &f, &o);
        if (r < 0)
                return log_full_errno(error_log_level, r,
                                      "Failed to mangle mount options %s: %m",
                                      strempty(options));

        fl = mount_flags_to_string(f);

        if ((f & MS_REMOUNT) && !what && !type)
                log_debug("Remounting %s (%s \"%s\")...",
                          where, strnull(fl), strempty(o));
        else if (!what && !type)
                log_debug("Mounting %s (%s \"%s\")...",
                          where, strnull(fl), strempty(o));
        else if ((f & MS_BIND) && !type)
                log_debug("Bind-mounting %s on %s (%s \"%s\")...",
                          what, where, strnull(fl), strempty(o));
        else if (f & MS_MOVE)
                log_debug("Moving mount %s → %s (%s \"%s\")...",
                          what, where, strnull(fl), strempty(o));
        else
                log_debug("Mounting %s on %s (%s \"%s\")...",
                          strna(type), where, strnull(fl), strempty(o));
        if (mount(what, where, type, f, o) < 0)
                return log_full_errno(error_log_level, errno,
                                      "Failed to mount %s on %s (%s \"%s\"): %m",
                                      strna(type), where, strnull(fl), strempty(o));
        return 0;
}

int umount_verbose(const char *what) {
        log_debug("Umounting %s...", what);
        if (umount(what) < 0)
                return log_error_errno(errno, "Failed to unmount %s: %m", what);
        return 0;
}

const char *mount_propagation_flags_to_string(unsigned long flags) {

        switch (flags & (MS_SHARED|MS_SLAVE|MS_PRIVATE)) {
        case 0:
                return "";
        case MS_SHARED:
                return "shared";
        case MS_SLAVE:
                return "slave";
        case MS_PRIVATE:
                return "private";
        }

        return NULL;
}

int mount_propagation_flags_from_string(const char *name, unsigned long *ret) {

        if (isempty(name))
                *ret = 0;
        else if (streq(name, "shared"))
                *ret = MS_SHARED;
        else if (streq(name, "slave"))
                *ret = MS_SLAVE;
        else if (streq(name, "private"))
                *ret = MS_PRIVATE;
        else
                return -EINVAL;
        return 0;
}

int mount_option_mangle(
                const char *options,
                unsigned long mount_flags,
                unsigned long *ret_mount_flags,
                char **ret_remaining_options) {

        const struct libmnt_optmap *map;
        _cleanup_free_ char *ret = NULL;
        const char *p;
        int r;

        /* This extracts mount flags from the mount options, and store
         * non-mount-flag options to '*ret_remaining_options'.
         * E.g.,
         * "rw,nosuid,nodev,relatime,size=1630748k,mode=700,uid=1000,gid=1000"
         * is split to MS_NOSUID|MS_NODEV|MS_RELATIME and
         * "size=1630748k,mode=700,uid=1000,gid=1000".
         * See more examples in test-mount-utils.c.
         *
         * Note that if 'options' does not contain any non-mount-flag options,
         * then '*ret_remaining_options' is set to NULL instread of empty string.
         * Note that this does not check validity of options stored in
         * '*ret_remaining_options'.
         * Note that if 'options' is NULL, then this just copies 'mount_flags'
         * to '*ret_mount_flags'. */

        assert(ret_mount_flags);
        assert(ret_remaining_options);

        map = mnt_get_builtin_optmap(MNT_LINUX_MAP);
        if (!map)
                return -EINVAL;

        p = options;
        for (;;) {
                _cleanup_free_ char *word = NULL;
                const struct libmnt_optmap *ent;

                r = extract_first_word(&p, &word, ",", EXTRACT_QUOTES);
                if (r < 0)
                        return r;
                if (r == 0)
                        break;

                for (ent = map; ent->name; ent++) {
                        /* All entries in MNT_LINUX_MAP do not take any argument.
                         * Thus, ent->name does not contain "=" or "[=]". */
                        if (!streq(word, ent->name))
                                continue;

                        if (!(ent->mask & MNT_INVERT))
                                mount_flags |= ent->id;
                        else if (mount_flags & ent->id)
                                mount_flags ^= ent->id;

                        break;
                }

                /* If 'word' is not a mount flag, then store it in '*ret_remaining_options'. */
                if (!ent->name && !strextend_with_separator(&ret, ",", word, NULL))
                        return -ENOMEM;
        }

        *ret_mount_flags = mount_flags;
        *ret_remaining_options = TAKE_PTR(ret);

        return 0;
}