xref: /kernel/linux/linux-6.6/fs/fuse/dir.c (revision 62306a36)
1/*
2  FUSE: Filesystem in Userspace
3  Copyright (C) 2001-2008  Miklos Szeredi <miklos@szeredi.hu>
4
5  This program can be distributed under the terms of the GNU GPL.
6  See the file COPYING.
7*/
8
9#include "fuse_i.h"
10
11#include <linux/pagemap.h>
12#include <linux/file.h>
13#include <linux/fs_context.h>
14#include <linux/moduleparam.h>
15#include <linux/sched.h>
16#include <linux/namei.h>
17#include <linux/slab.h>
18#include <linux/xattr.h>
19#include <linux/iversion.h>
20#include <linux/posix_acl.h>
21#include <linux/security.h>
22#include <linux/types.h>
23#include <linux/kernel.h>
24
25static bool __read_mostly allow_sys_admin_access;
26module_param(allow_sys_admin_access, bool, 0644);
27MODULE_PARM_DESC(allow_sys_admin_access,
28		 "Allow users with CAP_SYS_ADMIN in initial userns to bypass allow_other access check");
29
30static void fuse_advise_use_readdirplus(struct inode *dir)
31{
32	struct fuse_inode *fi = get_fuse_inode(dir);
33
34	set_bit(FUSE_I_ADVISE_RDPLUS, &fi->state);
35}
36
37#if BITS_PER_LONG >= 64
38static inline void __fuse_dentry_settime(struct dentry *entry, u64 time)
39{
40	entry->d_fsdata = (void *) time;
41}
42
43static inline u64 fuse_dentry_time(const struct dentry *entry)
44{
45	return (u64)entry->d_fsdata;
46}
47
48#else
49union fuse_dentry {
50	u64 time;
51	struct rcu_head rcu;
52};
53
54static inline void __fuse_dentry_settime(struct dentry *dentry, u64 time)
55{
56	((union fuse_dentry *) dentry->d_fsdata)->time = time;
57}
58
59static inline u64 fuse_dentry_time(const struct dentry *entry)
60{
61	return ((union fuse_dentry *) entry->d_fsdata)->time;
62}
63#endif
64
65static void fuse_dentry_settime(struct dentry *dentry, u64 time)
66{
67	struct fuse_conn *fc = get_fuse_conn_super(dentry->d_sb);
68	bool delete = !time && fc->delete_stale;
69	/*
70	 * Mess with DCACHE_OP_DELETE because dput() will be faster without it.
71	 * Don't care about races, either way it's just an optimization
72	 */
73	if ((!delete && (dentry->d_flags & DCACHE_OP_DELETE)) ||
74	    (delete && !(dentry->d_flags & DCACHE_OP_DELETE))) {
75		spin_lock(&dentry->d_lock);
76		if (!delete)
77			dentry->d_flags &= ~DCACHE_OP_DELETE;
78		else
79			dentry->d_flags |= DCACHE_OP_DELETE;
80		spin_unlock(&dentry->d_lock);
81	}
82
83	__fuse_dentry_settime(dentry, time);
84}
85
86/*
87 * FUSE caches dentries and attributes with separate timeout.  The
88 * time in jiffies until the dentry/attributes are valid is stored in
89 * dentry->d_fsdata and fuse_inode->i_time respectively.
90 */
91
92/*
93 * Calculate the time in jiffies until a dentry/attributes are valid
94 */
95u64 fuse_time_to_jiffies(u64 sec, u32 nsec)
96{
97	if (sec || nsec) {
98		struct timespec64 ts = {
99			sec,
100			min_t(u32, nsec, NSEC_PER_SEC - 1)
101		};
102
103		return get_jiffies_64() + timespec64_to_jiffies(&ts);
104	} else
105		return 0;
106}
107
108/*
109 * Set dentry and possibly attribute timeouts from the lookup/mk*
110 * replies
111 */
112void fuse_change_entry_timeout(struct dentry *entry, struct fuse_entry_out *o)
113{
114	fuse_dentry_settime(entry,
115		fuse_time_to_jiffies(o->entry_valid, o->entry_valid_nsec));
116}
117
118void fuse_invalidate_attr_mask(struct inode *inode, u32 mask)
119{
120	set_mask_bits(&get_fuse_inode(inode)->inval_mask, 0, mask);
121}
122
123/*
124 * Mark the attributes as stale, so that at the next call to
125 * ->getattr() they will be fetched from userspace
126 */
127void fuse_invalidate_attr(struct inode *inode)
128{
129	fuse_invalidate_attr_mask(inode, STATX_BASIC_STATS);
130}
131
132static void fuse_dir_changed(struct inode *dir)
133{
134	fuse_invalidate_attr(dir);
135	inode_maybe_inc_iversion(dir, false);
136}
137
138/*
139 * Mark the attributes as stale due to an atime change.  Avoid the invalidate if
140 * atime is not used.
141 */
142void fuse_invalidate_atime(struct inode *inode)
143{
144	if (!IS_RDONLY(inode))
145		fuse_invalidate_attr_mask(inode, STATX_ATIME);
146}
147
148/*
149 * Just mark the entry as stale, so that a next attempt to look it up
150 * will result in a new lookup call to userspace
151 *
152 * This is called when a dentry is about to become negative and the
153 * timeout is unknown (unlink, rmdir, rename and in some cases
154 * lookup)
155 */
156void fuse_invalidate_entry_cache(struct dentry *entry)
157{
158	fuse_dentry_settime(entry, 0);
159}
160
161/*
162 * Same as fuse_invalidate_entry_cache(), but also try to remove the
163 * dentry from the hash
164 */
165static void fuse_invalidate_entry(struct dentry *entry)
166{
167	d_invalidate(entry);
168	fuse_invalidate_entry_cache(entry);
169}
170
171static void fuse_lookup_init(struct fuse_conn *fc, struct fuse_args *args,
172			     u64 nodeid, const struct qstr *name,
173			     struct fuse_entry_out *outarg)
174{
175	memset(outarg, 0, sizeof(struct fuse_entry_out));
176	args->opcode = FUSE_LOOKUP;
177	args->nodeid = nodeid;
178	args->in_numargs = 1;
179	args->in_args[0].size = name->len + 1;
180	args->in_args[0].value = name->name;
181	args->out_numargs = 1;
182	args->out_args[0].size = sizeof(struct fuse_entry_out);
183	args->out_args[0].value = outarg;
184}
185
186/*
187 * Check whether the dentry is still valid
188 *
189 * If the entry validity timeout has expired and the dentry is
190 * positive, try to redo the lookup.  If the lookup results in a
191 * different inode, then let the VFS invalidate the dentry and redo
192 * the lookup once more.  If the lookup results in the same inode,
193 * then refresh the attributes, timeouts and mark the dentry valid.
194 */
195static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags)
196{
197	struct inode *inode;
198	struct dentry *parent;
199	struct fuse_mount *fm;
200	struct fuse_inode *fi;
201	int ret;
202
203	inode = d_inode_rcu(entry);
204	if (inode && fuse_is_bad(inode))
205		goto invalid;
206	else if (time_before64(fuse_dentry_time(entry), get_jiffies_64()) ||
207		 (flags & (LOOKUP_EXCL | LOOKUP_REVAL | LOOKUP_RENAME_TARGET))) {
208		struct fuse_entry_out outarg;
209		FUSE_ARGS(args);
210		struct fuse_forget_link *forget;
211		u64 attr_version;
212
213		/* For negative dentries, always do a fresh lookup */
214		if (!inode)
215			goto invalid;
216
217		ret = -ECHILD;
218		if (flags & LOOKUP_RCU)
219			goto out;
220
221		fm = get_fuse_mount(inode);
222
223		forget = fuse_alloc_forget();
224		ret = -ENOMEM;
225		if (!forget)
226			goto out;
227
228		attr_version = fuse_get_attr_version(fm->fc);
229
230		parent = dget_parent(entry);
231		fuse_lookup_init(fm->fc, &args, get_node_id(d_inode(parent)),
232				 &entry->d_name, &outarg);
233		ret = fuse_simple_request(fm, &args);
234		dput(parent);
235		/* Zero nodeid is same as -ENOENT */
236		if (!ret && !outarg.nodeid)
237			ret = -ENOENT;
238		if (!ret) {
239			fi = get_fuse_inode(inode);
240			if (outarg.nodeid != get_node_id(inode) ||
241			    (bool) IS_AUTOMOUNT(inode) != (bool) (outarg.attr.flags & FUSE_ATTR_SUBMOUNT)) {
242				fuse_queue_forget(fm->fc, forget,
243						  outarg.nodeid, 1);
244				goto invalid;
245			}
246			spin_lock(&fi->lock);
247			fi->nlookup++;
248			spin_unlock(&fi->lock);
249		}
250		kfree(forget);
251		if (ret == -ENOMEM || ret == -EINTR)
252			goto out;
253		if (ret || fuse_invalid_attr(&outarg.attr) ||
254		    fuse_stale_inode(inode, outarg.generation, &outarg.attr))
255			goto invalid;
256
257		forget_all_cached_acls(inode);
258		fuse_change_attributes(inode, &outarg.attr, NULL,
259				       ATTR_TIMEOUT(&outarg),
260				       attr_version);
261		fuse_change_entry_timeout(entry, &outarg);
262	} else if (inode) {
263		fi = get_fuse_inode(inode);
264		if (flags & LOOKUP_RCU) {
265			if (test_bit(FUSE_I_INIT_RDPLUS, &fi->state))
266				return -ECHILD;
267		} else if (test_and_clear_bit(FUSE_I_INIT_RDPLUS, &fi->state)) {
268			parent = dget_parent(entry);
269			fuse_advise_use_readdirplus(d_inode(parent));
270			dput(parent);
271		}
272	}
273	ret = 1;
274out:
275	return ret;
276
277invalid:
278	ret = 0;
279	goto out;
280}
281
282#if BITS_PER_LONG < 64
283static int fuse_dentry_init(struct dentry *dentry)
284{
285	dentry->d_fsdata = kzalloc(sizeof(union fuse_dentry),
286				   GFP_KERNEL_ACCOUNT | __GFP_RECLAIMABLE);
287
288	return dentry->d_fsdata ? 0 : -ENOMEM;
289}
290static void fuse_dentry_release(struct dentry *dentry)
291{
292	union fuse_dentry *fd = dentry->d_fsdata;
293
294	kfree_rcu(fd, rcu);
295}
296#endif
297
298static int fuse_dentry_delete(const struct dentry *dentry)
299{
300	return time_before64(fuse_dentry_time(dentry), get_jiffies_64());
301}
302
303/*
304 * Create a fuse_mount object with a new superblock (with path->dentry
305 * as the root), and return that mount so it can be auto-mounted on
306 * @path.
307 */
308static struct vfsmount *fuse_dentry_automount(struct path *path)
309{
310	struct fs_context *fsc;
311	struct vfsmount *mnt;
312	struct fuse_inode *mp_fi = get_fuse_inode(d_inode(path->dentry));
313
314	fsc = fs_context_for_submount(path->mnt->mnt_sb->s_type, path->dentry);
315	if (IS_ERR(fsc))
316		return ERR_CAST(fsc);
317
318	/* Pass the FUSE inode of the mount for fuse_get_tree_submount() */
319	fsc->fs_private = mp_fi;
320
321	/* Create the submount */
322	mnt = fc_mount(fsc);
323	if (!IS_ERR(mnt))
324		mntget(mnt);
325
326	put_fs_context(fsc);
327	return mnt;
328}
329
330const struct dentry_operations fuse_dentry_operations = {
331	.d_revalidate	= fuse_dentry_revalidate,
332	.d_delete	= fuse_dentry_delete,
333#if BITS_PER_LONG < 64
334	.d_init		= fuse_dentry_init,
335	.d_release	= fuse_dentry_release,
336#endif
337	.d_automount	= fuse_dentry_automount,
338};
339
340const struct dentry_operations fuse_root_dentry_operations = {
341#if BITS_PER_LONG < 64
342	.d_init		= fuse_dentry_init,
343	.d_release	= fuse_dentry_release,
344#endif
345};
346
347int fuse_valid_type(int m)
348{
349	return S_ISREG(m) || S_ISDIR(m) || S_ISLNK(m) || S_ISCHR(m) ||
350		S_ISBLK(m) || S_ISFIFO(m) || S_ISSOCK(m);
351}
352
353static bool fuse_valid_size(u64 size)
354{
355	return size <= LLONG_MAX;
356}
357
358bool fuse_invalid_attr(struct fuse_attr *attr)
359{
360	return !fuse_valid_type(attr->mode) || !fuse_valid_size(attr->size);
361}
362
363int fuse_lookup_name(struct super_block *sb, u64 nodeid, const struct qstr *name,
364		     struct fuse_entry_out *outarg, struct inode **inode)
365{
366	struct fuse_mount *fm = get_fuse_mount_super(sb);
367	FUSE_ARGS(args);
368	struct fuse_forget_link *forget;
369	u64 attr_version;
370	int err;
371
372	*inode = NULL;
373	err = -ENAMETOOLONG;
374	if (name->len > FUSE_NAME_MAX)
375		goto out;
376
377
378	forget = fuse_alloc_forget();
379	err = -ENOMEM;
380	if (!forget)
381		goto out;
382
383	attr_version = fuse_get_attr_version(fm->fc);
384
385	fuse_lookup_init(fm->fc, &args, nodeid, name, outarg);
386	err = fuse_simple_request(fm, &args);
387	/* Zero nodeid is same as -ENOENT, but with valid timeout */
388	if (err || !outarg->nodeid)
389		goto out_put_forget;
390
391	err = -EIO;
392	if (fuse_invalid_attr(&outarg->attr))
393		goto out_put_forget;
394	if (outarg->nodeid == FUSE_ROOT_ID && outarg->generation != 0) {
395		pr_warn_once("root generation should be zero\n");
396		outarg->generation = 0;
397	}
398
399	*inode = fuse_iget(sb, outarg->nodeid, outarg->generation,
400			   &outarg->attr, ATTR_TIMEOUT(outarg),
401			   attr_version);
402	err = -ENOMEM;
403	if (!*inode) {
404		fuse_queue_forget(fm->fc, forget, outarg->nodeid, 1);
405		goto out;
406	}
407	err = 0;
408
409 out_put_forget:
410	kfree(forget);
411 out:
412	return err;
413}
414
415static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
416				  unsigned int flags)
417{
418	int err;
419	struct fuse_entry_out outarg;
420	struct inode *inode;
421	struct dentry *newent;
422	bool outarg_valid = true;
423	bool locked;
424
425	if (fuse_is_bad(dir))
426		return ERR_PTR(-EIO);
427
428	locked = fuse_lock_inode(dir);
429	err = fuse_lookup_name(dir->i_sb, get_node_id(dir), &entry->d_name,
430			       &outarg, &inode);
431	fuse_unlock_inode(dir, locked);
432	if (err == -ENOENT) {
433		outarg_valid = false;
434		err = 0;
435	}
436	if (err)
437		goto out_err;
438
439	err = -EIO;
440	if (inode && get_node_id(inode) == FUSE_ROOT_ID)
441		goto out_iput;
442
443	newent = d_splice_alias(inode, entry);
444	err = PTR_ERR(newent);
445	if (IS_ERR(newent))
446		goto out_err;
447
448	entry = newent ? newent : entry;
449	if (outarg_valid)
450		fuse_change_entry_timeout(entry, &outarg);
451	else
452		fuse_invalidate_entry_cache(entry);
453
454	if (inode)
455		fuse_advise_use_readdirplus(dir);
456	return newent;
457
458 out_iput:
459	iput(inode);
460 out_err:
461	return ERR_PTR(err);
462}
463
464static int get_security_context(struct dentry *entry, umode_t mode,
465				struct fuse_in_arg *ext)
466{
467	struct fuse_secctx *fctx;
468	struct fuse_secctx_header *header;
469	void *ctx = NULL, *ptr;
470	u32 ctxlen, total_len = sizeof(*header);
471	int err, nr_ctx = 0;
472	const char *name;
473	size_t namelen;
474
475	err = security_dentry_init_security(entry, mode, &entry->d_name,
476					    &name, &ctx, &ctxlen);
477	if (err) {
478		if (err != -EOPNOTSUPP)
479			goto out_err;
480		/* No LSM is supporting this security hook. Ignore error */
481		ctxlen = 0;
482		ctx = NULL;
483	}
484
485	if (ctxlen) {
486		nr_ctx = 1;
487		namelen = strlen(name) + 1;
488		err = -EIO;
489		if (WARN_ON(namelen > XATTR_NAME_MAX + 1 || ctxlen > S32_MAX))
490			goto out_err;
491		total_len += FUSE_REC_ALIGN(sizeof(*fctx) + namelen + ctxlen);
492	}
493
494	err = -ENOMEM;
495	header = ptr = kzalloc(total_len, GFP_KERNEL);
496	if (!ptr)
497		goto out_err;
498
499	header->nr_secctx = nr_ctx;
500	header->size = total_len;
501	ptr += sizeof(*header);
502	if (nr_ctx) {
503		fctx = ptr;
504		fctx->size = ctxlen;
505		ptr += sizeof(*fctx);
506
507		strcpy(ptr, name);
508		ptr += namelen;
509
510		memcpy(ptr, ctx, ctxlen);
511	}
512	ext->size = total_len;
513	ext->value = header;
514	err = 0;
515out_err:
516	kfree(ctx);
517	return err;
518}
519
520static void *extend_arg(struct fuse_in_arg *buf, u32 bytes)
521{
522	void *p;
523	u32 newlen = buf->size + bytes;
524
525	p = krealloc(buf->value, newlen, GFP_KERNEL);
526	if (!p) {
527		kfree(buf->value);
528		buf->size = 0;
529		buf->value = NULL;
530		return NULL;
531	}
532
533	memset(p + buf->size, 0, bytes);
534	buf->value = p;
535	buf->size = newlen;
536
537	return p + newlen - bytes;
538}
539
540static u32 fuse_ext_size(size_t size)
541{
542	return FUSE_REC_ALIGN(sizeof(struct fuse_ext_header) + size);
543}
544
545/*
546 * This adds just a single supplementary group that matches the parent's group.
547 */
548static int get_create_supp_group(struct inode *dir, struct fuse_in_arg *ext)
549{
550	struct fuse_conn *fc = get_fuse_conn(dir);
551	struct fuse_ext_header *xh;
552	struct fuse_supp_groups *sg;
553	kgid_t kgid = dir->i_gid;
554	gid_t parent_gid = from_kgid(fc->user_ns, kgid);
555	u32 sg_len = fuse_ext_size(sizeof(*sg) + sizeof(sg->groups[0]));
556
557	if (parent_gid == (gid_t) -1 || gid_eq(kgid, current_fsgid()) ||
558	    !in_group_p(kgid))
559		return 0;
560
561	xh = extend_arg(ext, sg_len);
562	if (!xh)
563		return -ENOMEM;
564
565	xh->size = sg_len;
566	xh->type = FUSE_EXT_GROUPS;
567
568	sg = (struct fuse_supp_groups *) &xh[1];
569	sg->nr_groups = 1;
570	sg->groups[0] = parent_gid;
571
572	return 0;
573}
574
575static int get_create_ext(struct fuse_args *args,
576			  struct inode *dir, struct dentry *dentry,
577			  umode_t mode)
578{
579	struct fuse_conn *fc = get_fuse_conn_super(dentry->d_sb);
580	struct fuse_in_arg ext = { .size = 0, .value = NULL };
581	int err = 0;
582
583	if (fc->init_security)
584		err = get_security_context(dentry, mode, &ext);
585	if (!err && fc->create_supp_group)
586		err = get_create_supp_group(dir, &ext);
587
588	if (!err && ext.size) {
589		WARN_ON(args->in_numargs >= ARRAY_SIZE(args->in_args));
590		args->is_ext = true;
591		args->ext_idx = args->in_numargs++;
592		args->in_args[args->ext_idx] = ext;
593	} else {
594		kfree(ext.value);
595	}
596
597	return err;
598}
599
600static void free_ext_value(struct fuse_args *args)
601{
602	if (args->is_ext)
603		kfree(args->in_args[args->ext_idx].value);
604}
605
606/*
607 * Atomic create+open operation
608 *
609 * If the filesystem doesn't support this, then fall back to separate
610 * 'mknod' + 'open' requests.
611 */
612static int fuse_create_open(struct inode *dir, struct dentry *entry,
613			    struct file *file, unsigned int flags,
614			    umode_t mode, u32 opcode)
615{
616	int err;
617	struct inode *inode;
618	struct fuse_mount *fm = get_fuse_mount(dir);
619	FUSE_ARGS(args);
620	struct fuse_forget_link *forget;
621	struct fuse_create_in inarg;
622	struct fuse_open_out outopen;
623	struct fuse_entry_out outentry;
624	struct fuse_inode *fi;
625	struct fuse_file *ff;
626	bool trunc = flags & O_TRUNC;
627
628	/* Userspace expects S_IFREG in create mode */
629	BUG_ON((mode & S_IFMT) != S_IFREG);
630
631	forget = fuse_alloc_forget();
632	err = -ENOMEM;
633	if (!forget)
634		goto out_err;
635
636	err = -ENOMEM;
637	ff = fuse_file_alloc(fm);
638	if (!ff)
639		goto out_put_forget_req;
640
641	if (!fm->fc->dont_mask)
642		mode &= ~current_umask();
643
644	flags &= ~O_NOCTTY;
645	memset(&inarg, 0, sizeof(inarg));
646	memset(&outentry, 0, sizeof(outentry));
647	inarg.flags = flags;
648	inarg.mode = mode;
649	inarg.umask = current_umask();
650
651	if (fm->fc->handle_killpriv_v2 && trunc &&
652	    !(flags & O_EXCL) && !capable(CAP_FSETID)) {
653		inarg.open_flags |= FUSE_OPEN_KILL_SUIDGID;
654	}
655
656	args.opcode = opcode;
657	args.nodeid = get_node_id(dir);
658	args.in_numargs = 2;
659	args.in_args[0].size = sizeof(inarg);
660	args.in_args[0].value = &inarg;
661	args.in_args[1].size = entry->d_name.len + 1;
662	args.in_args[1].value = entry->d_name.name;
663	args.out_numargs = 2;
664	args.out_args[0].size = sizeof(outentry);
665	args.out_args[0].value = &outentry;
666	args.out_args[1].size = sizeof(outopen);
667	args.out_args[1].value = &outopen;
668
669	err = get_create_ext(&args, dir, entry, mode);
670	if (err)
671		goto out_put_forget_req;
672
673	err = fuse_simple_request(fm, &args);
674	free_ext_value(&args);
675	if (err)
676		goto out_free_ff;
677
678	err = -EIO;
679	if (!S_ISREG(outentry.attr.mode) || invalid_nodeid(outentry.nodeid) ||
680	    fuse_invalid_attr(&outentry.attr))
681		goto out_free_ff;
682
683	ff->fh = outopen.fh;
684	ff->nodeid = outentry.nodeid;
685	ff->open_flags = outopen.open_flags;
686	inode = fuse_iget(dir->i_sb, outentry.nodeid, outentry.generation,
687			  &outentry.attr, ATTR_TIMEOUT(&outentry), 0);
688	if (!inode) {
689		flags &= ~(O_CREAT | O_EXCL | O_TRUNC);
690		fuse_sync_release(NULL, ff, flags);
691		fuse_queue_forget(fm->fc, forget, outentry.nodeid, 1);
692		err = -ENOMEM;
693		goto out_err;
694	}
695	kfree(forget);
696	d_instantiate(entry, inode);
697	fuse_change_entry_timeout(entry, &outentry);
698	fuse_dir_changed(dir);
699	err = finish_open(file, entry, generic_file_open);
700	if (err) {
701		fi = get_fuse_inode(inode);
702		fuse_sync_release(fi, ff, flags);
703	} else {
704		file->private_data = ff;
705		fuse_finish_open(inode, file);
706		if (fm->fc->atomic_o_trunc && trunc)
707			truncate_pagecache(inode, 0);
708		else if (!(ff->open_flags & FOPEN_KEEP_CACHE))
709			invalidate_inode_pages2(inode->i_mapping);
710	}
711	return err;
712
713out_free_ff:
714	fuse_file_free(ff);
715out_put_forget_req:
716	kfree(forget);
717out_err:
718	return err;
719}
720
721static int fuse_mknod(struct mnt_idmap *, struct inode *, struct dentry *,
722		      umode_t, dev_t);
723static int fuse_atomic_open(struct inode *dir, struct dentry *entry,
724			    struct file *file, unsigned flags,
725			    umode_t mode)
726{
727	int err;
728	struct fuse_conn *fc = get_fuse_conn(dir);
729	struct dentry *res = NULL;
730
731	if (fuse_is_bad(dir))
732		return -EIO;
733
734	if (d_in_lookup(entry)) {
735		res = fuse_lookup(dir, entry, 0);
736		if (IS_ERR(res))
737			return PTR_ERR(res);
738
739		if (res)
740			entry = res;
741	}
742
743	if (!(flags & O_CREAT) || d_really_is_positive(entry))
744		goto no_open;
745
746	/* Only creates */
747	file->f_mode |= FMODE_CREATED;
748
749	if (fc->no_create)
750		goto mknod;
751
752	err = fuse_create_open(dir, entry, file, flags, mode, FUSE_CREATE);
753	if (err == -ENOSYS) {
754		fc->no_create = 1;
755		goto mknod;
756	} else if (err == -EEXIST)
757		fuse_invalidate_entry(entry);
758out_dput:
759	dput(res);
760	return err;
761
762mknod:
763	err = fuse_mknod(&nop_mnt_idmap, dir, entry, mode, 0);
764	if (err)
765		goto out_dput;
766no_open:
767	return finish_no_open(file, res);
768}
769
770/*
771 * Code shared between mknod, mkdir, symlink and link
772 */
773static int create_new_entry(struct fuse_mount *fm, struct fuse_args *args,
774			    struct inode *dir, struct dentry *entry,
775			    umode_t mode)
776{
777	struct fuse_entry_out outarg;
778	struct inode *inode;
779	struct dentry *d;
780	int err;
781	struct fuse_forget_link *forget;
782
783	if (fuse_is_bad(dir))
784		return -EIO;
785
786	forget = fuse_alloc_forget();
787	if (!forget)
788		return -ENOMEM;
789
790	memset(&outarg, 0, sizeof(outarg));
791	args->nodeid = get_node_id(dir);
792	args->out_numargs = 1;
793	args->out_args[0].size = sizeof(outarg);
794	args->out_args[0].value = &outarg;
795
796	if (args->opcode != FUSE_LINK) {
797		err = get_create_ext(args, dir, entry, mode);
798		if (err)
799			goto out_put_forget_req;
800	}
801
802	err = fuse_simple_request(fm, args);
803	free_ext_value(args);
804	if (err)
805		goto out_put_forget_req;
806
807	err = -EIO;
808	if (invalid_nodeid(outarg.nodeid) || fuse_invalid_attr(&outarg.attr))
809		goto out_put_forget_req;
810
811	if ((outarg.attr.mode ^ mode) & S_IFMT)
812		goto out_put_forget_req;
813
814	inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation,
815			  &outarg.attr, ATTR_TIMEOUT(&outarg), 0);
816	if (!inode) {
817		fuse_queue_forget(fm->fc, forget, outarg.nodeid, 1);
818		return -ENOMEM;
819	}
820	kfree(forget);
821
822	d_drop(entry);
823	d = d_splice_alias(inode, entry);
824	if (IS_ERR(d))
825		return PTR_ERR(d);
826
827	if (d) {
828		fuse_change_entry_timeout(d, &outarg);
829		dput(d);
830	} else {
831		fuse_change_entry_timeout(entry, &outarg);
832	}
833	fuse_dir_changed(dir);
834	return 0;
835
836 out_put_forget_req:
837	if (err == -EEXIST)
838		fuse_invalidate_entry(entry);
839	kfree(forget);
840	return err;
841}
842
843static int fuse_mknod(struct mnt_idmap *idmap, struct inode *dir,
844		      struct dentry *entry, umode_t mode, dev_t rdev)
845{
846	struct fuse_mknod_in inarg;
847	struct fuse_mount *fm = get_fuse_mount(dir);
848	FUSE_ARGS(args);
849
850	if (!fm->fc->dont_mask)
851		mode &= ~current_umask();
852
853	memset(&inarg, 0, sizeof(inarg));
854	inarg.mode = mode;
855	inarg.rdev = new_encode_dev(rdev);
856	inarg.umask = current_umask();
857	args.opcode = FUSE_MKNOD;
858	args.in_numargs = 2;
859	args.in_args[0].size = sizeof(inarg);
860	args.in_args[0].value = &inarg;
861	args.in_args[1].size = entry->d_name.len + 1;
862	args.in_args[1].value = entry->d_name.name;
863	return create_new_entry(fm, &args, dir, entry, mode);
864}
865
866static int fuse_create(struct mnt_idmap *idmap, struct inode *dir,
867		       struct dentry *entry, umode_t mode, bool excl)
868{
869	return fuse_mknod(&nop_mnt_idmap, dir, entry, mode, 0);
870}
871
872static int fuse_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
873			struct file *file, umode_t mode)
874{
875	struct fuse_conn *fc = get_fuse_conn(dir);
876	int err;
877
878	if (fc->no_tmpfile)
879		return -EOPNOTSUPP;
880
881	err = fuse_create_open(dir, file->f_path.dentry, file, file->f_flags, mode, FUSE_TMPFILE);
882	if (err == -ENOSYS) {
883		fc->no_tmpfile = 1;
884		err = -EOPNOTSUPP;
885	}
886	return err;
887}
888
889static int fuse_mkdir(struct mnt_idmap *idmap, struct inode *dir,
890		      struct dentry *entry, umode_t mode)
891{
892	struct fuse_mkdir_in inarg;
893	struct fuse_mount *fm = get_fuse_mount(dir);
894	FUSE_ARGS(args);
895
896	if (!fm->fc->dont_mask)
897		mode &= ~current_umask();
898
899	memset(&inarg, 0, sizeof(inarg));
900	inarg.mode = mode;
901	inarg.umask = current_umask();
902	args.opcode = FUSE_MKDIR;
903	args.in_numargs = 2;
904	args.in_args[0].size = sizeof(inarg);
905	args.in_args[0].value = &inarg;
906	args.in_args[1].size = entry->d_name.len + 1;
907	args.in_args[1].value = entry->d_name.name;
908	return create_new_entry(fm, &args, dir, entry, S_IFDIR);
909}
910
911static int fuse_symlink(struct mnt_idmap *idmap, struct inode *dir,
912			struct dentry *entry, const char *link)
913{
914	struct fuse_mount *fm = get_fuse_mount(dir);
915	unsigned len = strlen(link) + 1;
916	FUSE_ARGS(args);
917
918	args.opcode = FUSE_SYMLINK;
919	args.in_numargs = 2;
920	args.in_args[0].size = entry->d_name.len + 1;
921	args.in_args[0].value = entry->d_name.name;
922	args.in_args[1].size = len;
923	args.in_args[1].value = link;
924	return create_new_entry(fm, &args, dir, entry, S_IFLNK);
925}
926
927void fuse_flush_time_update(struct inode *inode)
928{
929	int err = sync_inode_metadata(inode, 1);
930
931	mapping_set_error(inode->i_mapping, err);
932}
933
934static void fuse_update_ctime_in_cache(struct inode *inode)
935{
936	if (!IS_NOCMTIME(inode)) {
937		inode_set_ctime_current(inode);
938		mark_inode_dirty_sync(inode);
939		fuse_flush_time_update(inode);
940	}
941}
942
943void fuse_update_ctime(struct inode *inode)
944{
945	fuse_invalidate_attr_mask(inode, STATX_CTIME);
946	fuse_update_ctime_in_cache(inode);
947}
948
949static void fuse_entry_unlinked(struct dentry *entry)
950{
951	struct inode *inode = d_inode(entry);
952	struct fuse_conn *fc = get_fuse_conn(inode);
953	struct fuse_inode *fi = get_fuse_inode(inode);
954
955	spin_lock(&fi->lock);
956	fi->attr_version = atomic64_inc_return(&fc->attr_version);
957	/*
958	 * If i_nlink == 0 then unlink doesn't make sense, yet this can
959	 * happen if userspace filesystem is careless.  It would be
960	 * difficult to enforce correct nlink usage so just ignore this
961	 * condition here
962	 */
963	if (S_ISDIR(inode->i_mode))
964		clear_nlink(inode);
965	else if (inode->i_nlink > 0)
966		drop_nlink(inode);
967	spin_unlock(&fi->lock);
968	fuse_invalidate_entry_cache(entry);
969	fuse_update_ctime(inode);
970}
971
972static int fuse_unlink(struct inode *dir, struct dentry *entry)
973{
974	int err;
975	struct fuse_mount *fm = get_fuse_mount(dir);
976	FUSE_ARGS(args);
977
978	if (fuse_is_bad(dir))
979		return -EIO;
980
981	args.opcode = FUSE_UNLINK;
982	args.nodeid = get_node_id(dir);
983	args.in_numargs = 1;
984	args.in_args[0].size = entry->d_name.len + 1;
985	args.in_args[0].value = entry->d_name.name;
986	err = fuse_simple_request(fm, &args);
987	if (!err) {
988		fuse_dir_changed(dir);
989		fuse_entry_unlinked(entry);
990	} else if (err == -EINTR || err == -ENOENT)
991		fuse_invalidate_entry(entry);
992	return err;
993}
994
995static int fuse_rmdir(struct inode *dir, struct dentry *entry)
996{
997	int err;
998	struct fuse_mount *fm = get_fuse_mount(dir);
999	FUSE_ARGS(args);
1000
1001	if (fuse_is_bad(dir))
1002		return -EIO;
1003
1004	args.opcode = FUSE_RMDIR;
1005	args.nodeid = get_node_id(dir);
1006	args.in_numargs = 1;
1007	args.in_args[0].size = entry->d_name.len + 1;
1008	args.in_args[0].value = entry->d_name.name;
1009	err = fuse_simple_request(fm, &args);
1010	if (!err) {
1011		fuse_dir_changed(dir);
1012		fuse_entry_unlinked(entry);
1013	} else if (err == -EINTR || err == -ENOENT)
1014		fuse_invalidate_entry(entry);
1015	return err;
1016}
1017
1018static int fuse_rename_common(struct inode *olddir, struct dentry *oldent,
1019			      struct inode *newdir, struct dentry *newent,
1020			      unsigned int flags, int opcode, size_t argsize)
1021{
1022	int err;
1023	struct fuse_rename2_in inarg;
1024	struct fuse_mount *fm = get_fuse_mount(olddir);
1025	FUSE_ARGS(args);
1026
1027	memset(&inarg, 0, argsize);
1028	inarg.newdir = get_node_id(newdir);
1029	inarg.flags = flags;
1030	args.opcode = opcode;
1031	args.nodeid = get_node_id(olddir);
1032	args.in_numargs = 3;
1033	args.in_args[0].size = argsize;
1034	args.in_args[0].value = &inarg;
1035	args.in_args[1].size = oldent->d_name.len + 1;
1036	args.in_args[1].value = oldent->d_name.name;
1037	args.in_args[2].size = newent->d_name.len + 1;
1038	args.in_args[2].value = newent->d_name.name;
1039	err = fuse_simple_request(fm, &args);
1040	if (!err) {
1041		/* ctime changes */
1042		fuse_update_ctime(d_inode(oldent));
1043
1044		if (flags & RENAME_EXCHANGE)
1045			fuse_update_ctime(d_inode(newent));
1046
1047		fuse_dir_changed(olddir);
1048		if (olddir != newdir)
1049			fuse_dir_changed(newdir);
1050
1051		/* newent will end up negative */
1052		if (!(flags & RENAME_EXCHANGE) && d_really_is_positive(newent))
1053			fuse_entry_unlinked(newent);
1054	} else if (err == -EINTR || err == -ENOENT) {
1055		/* If request was interrupted, DEITY only knows if the
1056		   rename actually took place.  If the invalidation
1057		   fails (e.g. some process has CWD under the renamed
1058		   directory), then there can be inconsistency between
1059		   the dcache and the real filesystem.  Tough luck. */
1060		fuse_invalidate_entry(oldent);
1061		if (d_really_is_positive(newent))
1062			fuse_invalidate_entry(newent);
1063	}
1064
1065	return err;
1066}
1067
1068static int fuse_rename2(struct mnt_idmap *idmap, struct inode *olddir,
1069			struct dentry *oldent, struct inode *newdir,
1070			struct dentry *newent, unsigned int flags)
1071{
1072	struct fuse_conn *fc = get_fuse_conn(olddir);
1073	int err;
1074
1075	if (fuse_is_bad(olddir))
1076		return -EIO;
1077
1078	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
1079		return -EINVAL;
1080
1081	if (flags) {
1082		if (fc->no_rename2 || fc->minor < 23)
1083			return -EINVAL;
1084
1085		err = fuse_rename_common(olddir, oldent, newdir, newent, flags,
1086					 FUSE_RENAME2,
1087					 sizeof(struct fuse_rename2_in));
1088		if (err == -ENOSYS) {
1089			fc->no_rename2 = 1;
1090			err = -EINVAL;
1091		}
1092	} else {
1093		err = fuse_rename_common(olddir, oldent, newdir, newent, 0,
1094					 FUSE_RENAME,
1095					 sizeof(struct fuse_rename_in));
1096	}
1097
1098	return err;
1099}
1100
1101static int fuse_link(struct dentry *entry, struct inode *newdir,
1102		     struct dentry *newent)
1103{
1104	int err;
1105	struct fuse_link_in inarg;
1106	struct inode *inode = d_inode(entry);
1107	struct fuse_mount *fm = get_fuse_mount(inode);
1108	FUSE_ARGS(args);
1109
1110	memset(&inarg, 0, sizeof(inarg));
1111	inarg.oldnodeid = get_node_id(inode);
1112	args.opcode = FUSE_LINK;
1113	args.in_numargs = 2;
1114	args.in_args[0].size = sizeof(inarg);
1115	args.in_args[0].value = &inarg;
1116	args.in_args[1].size = newent->d_name.len + 1;
1117	args.in_args[1].value = newent->d_name.name;
1118	err = create_new_entry(fm, &args, newdir, newent, inode->i_mode);
1119	if (!err)
1120		fuse_update_ctime_in_cache(inode);
1121	else if (err == -EINTR)
1122		fuse_invalidate_attr(inode);
1123
1124	return err;
1125}
1126
1127static void fuse_fillattr(struct inode *inode, struct fuse_attr *attr,
1128			  struct kstat *stat)
1129{
1130	unsigned int blkbits;
1131	struct fuse_conn *fc = get_fuse_conn(inode);
1132
1133	stat->dev = inode->i_sb->s_dev;
1134	stat->ino = attr->ino;
1135	stat->mode = (inode->i_mode & S_IFMT) | (attr->mode & 07777);
1136	stat->nlink = attr->nlink;
1137	stat->uid = make_kuid(fc->user_ns, attr->uid);
1138	stat->gid = make_kgid(fc->user_ns, attr->gid);
1139	stat->rdev = inode->i_rdev;
1140	stat->atime.tv_sec = attr->atime;
1141	stat->atime.tv_nsec = attr->atimensec;
1142	stat->mtime.tv_sec = attr->mtime;
1143	stat->mtime.tv_nsec = attr->mtimensec;
1144	stat->ctime.tv_sec = attr->ctime;
1145	stat->ctime.tv_nsec = attr->ctimensec;
1146	stat->size = attr->size;
1147	stat->blocks = attr->blocks;
1148
1149	if (attr->blksize != 0)
1150		blkbits = ilog2(attr->blksize);
1151	else
1152		blkbits = inode->i_sb->s_blocksize_bits;
1153
1154	stat->blksize = 1 << blkbits;
1155}
1156
1157static void fuse_statx_to_attr(struct fuse_statx *sx, struct fuse_attr *attr)
1158{
1159	memset(attr, 0, sizeof(*attr));
1160	attr->ino = sx->ino;
1161	attr->size = sx->size;
1162	attr->blocks = sx->blocks;
1163	attr->atime = sx->atime.tv_sec;
1164	attr->mtime = sx->mtime.tv_sec;
1165	attr->ctime = sx->ctime.tv_sec;
1166	attr->atimensec = sx->atime.tv_nsec;
1167	attr->mtimensec = sx->mtime.tv_nsec;
1168	attr->ctimensec = sx->ctime.tv_nsec;
1169	attr->mode = sx->mode;
1170	attr->nlink = sx->nlink;
1171	attr->uid = sx->uid;
1172	attr->gid = sx->gid;
1173	attr->rdev = new_encode_dev(MKDEV(sx->rdev_major, sx->rdev_minor));
1174	attr->blksize = sx->blksize;
1175}
1176
1177static int fuse_do_statx(struct inode *inode, struct file *file,
1178			 struct kstat *stat)
1179{
1180	int err;
1181	struct fuse_attr attr;
1182	struct fuse_statx *sx;
1183	struct fuse_statx_in inarg;
1184	struct fuse_statx_out outarg;
1185	struct fuse_mount *fm = get_fuse_mount(inode);
1186	u64 attr_version = fuse_get_attr_version(fm->fc);
1187	FUSE_ARGS(args);
1188
1189	memset(&inarg, 0, sizeof(inarg));
1190	memset(&outarg, 0, sizeof(outarg));
1191	/* Directories have separate file-handle space */
1192	if (file && S_ISREG(inode->i_mode)) {
1193		struct fuse_file *ff = file->private_data;
1194
1195		inarg.getattr_flags |= FUSE_GETATTR_FH;
1196		inarg.fh = ff->fh;
1197	}
1198	/* For now leave sync hints as the default, request all stats. */
1199	inarg.sx_flags = 0;
1200	inarg.sx_mask = STATX_BASIC_STATS | STATX_BTIME;
1201	args.opcode = FUSE_STATX;
1202	args.nodeid = get_node_id(inode);
1203	args.in_numargs = 1;
1204	args.in_args[0].size = sizeof(inarg);
1205	args.in_args[0].value = &inarg;
1206	args.out_numargs = 1;
1207	args.out_args[0].size = sizeof(outarg);
1208	args.out_args[0].value = &outarg;
1209	err = fuse_simple_request(fm, &args);
1210	if (err)
1211		return err;
1212
1213	sx = &outarg.stat;
1214	if (((sx->mask & STATX_SIZE) && !fuse_valid_size(sx->size)) ||
1215	    ((sx->mask & STATX_TYPE) && (!fuse_valid_type(sx->mode) ||
1216					 inode_wrong_type(inode, sx->mode)))) {
1217		fuse_make_bad(inode);
1218		return -EIO;
1219	}
1220
1221	fuse_statx_to_attr(&outarg.stat, &attr);
1222	if ((sx->mask & STATX_BASIC_STATS) == STATX_BASIC_STATS) {
1223		fuse_change_attributes(inode, &attr, &outarg.stat,
1224				       ATTR_TIMEOUT(&outarg), attr_version);
1225	}
1226
1227	if (stat) {
1228		stat->result_mask = sx->mask & (STATX_BASIC_STATS | STATX_BTIME);
1229		stat->btime.tv_sec = sx->btime.tv_sec;
1230		stat->btime.tv_nsec = min_t(u32, sx->btime.tv_nsec, NSEC_PER_SEC - 1);
1231		fuse_fillattr(inode, &attr, stat);
1232		stat->result_mask |= STATX_TYPE;
1233	}
1234
1235	return 0;
1236}
1237
1238static int fuse_do_getattr(struct inode *inode, struct kstat *stat,
1239			   struct file *file)
1240{
1241	int err;
1242	struct fuse_getattr_in inarg;
1243	struct fuse_attr_out outarg;
1244	struct fuse_mount *fm = get_fuse_mount(inode);
1245	FUSE_ARGS(args);
1246	u64 attr_version;
1247
1248	attr_version = fuse_get_attr_version(fm->fc);
1249
1250	memset(&inarg, 0, sizeof(inarg));
1251	memset(&outarg, 0, sizeof(outarg));
1252	/* Directories have separate file-handle space */
1253	if (file && S_ISREG(inode->i_mode)) {
1254		struct fuse_file *ff = file->private_data;
1255
1256		inarg.getattr_flags |= FUSE_GETATTR_FH;
1257		inarg.fh = ff->fh;
1258	}
1259	args.opcode = FUSE_GETATTR;
1260	args.nodeid = get_node_id(inode);
1261	args.in_numargs = 1;
1262	args.in_args[0].size = sizeof(inarg);
1263	args.in_args[0].value = &inarg;
1264	args.out_numargs = 1;
1265	args.out_args[0].size = sizeof(outarg);
1266	args.out_args[0].value = &outarg;
1267	err = fuse_simple_request(fm, &args);
1268	if (!err) {
1269		if (fuse_invalid_attr(&outarg.attr) ||
1270		    inode_wrong_type(inode, outarg.attr.mode)) {
1271			fuse_make_bad(inode);
1272			err = -EIO;
1273		} else {
1274			fuse_change_attributes(inode, &outarg.attr, NULL,
1275					       ATTR_TIMEOUT(&outarg),
1276					       attr_version);
1277			if (stat)
1278				fuse_fillattr(inode, &outarg.attr, stat);
1279		}
1280	}
1281	return err;
1282}
1283
1284static int fuse_update_get_attr(struct inode *inode, struct file *file,
1285				struct kstat *stat, u32 request_mask,
1286				unsigned int flags)
1287{
1288	struct fuse_inode *fi = get_fuse_inode(inode);
1289	struct fuse_conn *fc = get_fuse_conn(inode);
1290	int err = 0;
1291	bool sync;
1292	u32 inval_mask = READ_ONCE(fi->inval_mask);
1293	u32 cache_mask = fuse_get_cache_mask(inode);
1294
1295
1296	/* FUSE only supports basic stats and possibly btime */
1297	request_mask &= STATX_BASIC_STATS | STATX_BTIME;
1298retry:
1299	if (fc->no_statx)
1300		request_mask &= STATX_BASIC_STATS;
1301
1302	if (!request_mask)
1303		sync = false;
1304	else if (flags & AT_STATX_FORCE_SYNC)
1305		sync = true;
1306	else if (flags & AT_STATX_DONT_SYNC)
1307		sync = false;
1308	else if (request_mask & inval_mask & ~cache_mask)
1309		sync = true;
1310	else
1311		sync = time_before64(fi->i_time, get_jiffies_64());
1312
1313	if (sync) {
1314		forget_all_cached_acls(inode);
1315		/* Try statx if BTIME is requested */
1316		if (!fc->no_statx && (request_mask & ~STATX_BASIC_STATS)) {
1317			err = fuse_do_statx(inode, file, stat);
1318			if (err == -ENOSYS) {
1319				fc->no_statx = 1;
1320				goto retry;
1321			}
1322		} else {
1323			err = fuse_do_getattr(inode, stat, file);
1324		}
1325	} else if (stat) {
1326		generic_fillattr(&nop_mnt_idmap, request_mask, inode, stat);
1327		stat->mode = fi->orig_i_mode;
1328		stat->ino = fi->orig_ino;
1329		if (test_bit(FUSE_I_BTIME, &fi->state)) {
1330			stat->btime = fi->i_btime;
1331			stat->result_mask |= STATX_BTIME;
1332		}
1333	}
1334
1335	return err;
1336}
1337
1338int fuse_update_attributes(struct inode *inode, struct file *file, u32 mask)
1339{
1340	return fuse_update_get_attr(inode, file, NULL, mask, 0);
1341}
1342
1343int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
1344			     u64 child_nodeid, struct qstr *name, u32 flags)
1345{
1346	int err = -ENOTDIR;
1347	struct inode *parent;
1348	struct dentry *dir;
1349	struct dentry *entry;
1350
1351	parent = fuse_ilookup(fc, parent_nodeid, NULL);
1352	if (!parent)
1353		return -ENOENT;
1354
1355	inode_lock_nested(parent, I_MUTEX_PARENT);
1356	if (!S_ISDIR(parent->i_mode))
1357		goto unlock;
1358
1359	err = -ENOENT;
1360	dir = d_find_alias(parent);
1361	if (!dir)
1362		goto unlock;
1363
1364	name->hash = full_name_hash(dir, name->name, name->len);
1365	entry = d_lookup(dir, name);
1366	dput(dir);
1367	if (!entry)
1368		goto unlock;
1369
1370	fuse_dir_changed(parent);
1371	if (!(flags & FUSE_EXPIRE_ONLY))
1372		d_invalidate(entry);
1373	fuse_invalidate_entry_cache(entry);
1374
1375	if (child_nodeid != 0 && d_really_is_positive(entry)) {
1376		inode_lock(d_inode(entry));
1377		if (get_node_id(d_inode(entry)) != child_nodeid) {
1378			err = -ENOENT;
1379			goto badentry;
1380		}
1381		if (d_mountpoint(entry)) {
1382			err = -EBUSY;
1383			goto badentry;
1384		}
1385		if (d_is_dir(entry)) {
1386			shrink_dcache_parent(entry);
1387			if (!simple_empty(entry)) {
1388				err = -ENOTEMPTY;
1389				goto badentry;
1390			}
1391			d_inode(entry)->i_flags |= S_DEAD;
1392		}
1393		dont_mount(entry);
1394		clear_nlink(d_inode(entry));
1395		err = 0;
1396 badentry:
1397		inode_unlock(d_inode(entry));
1398		if (!err)
1399			d_delete(entry);
1400	} else {
1401		err = 0;
1402	}
1403	dput(entry);
1404
1405 unlock:
1406	inode_unlock(parent);
1407	iput(parent);
1408	return err;
1409}
1410
1411static inline bool fuse_permissible_uidgid(struct fuse_conn *fc)
1412{
1413	const struct cred *cred = current_cred();
1414
1415	return (uid_eq(cred->euid, fc->user_id) &&
1416		uid_eq(cred->suid, fc->user_id) &&
1417		uid_eq(cred->uid,  fc->user_id) &&
1418		gid_eq(cred->egid, fc->group_id) &&
1419		gid_eq(cred->sgid, fc->group_id) &&
1420		gid_eq(cred->gid,  fc->group_id));
1421}
1422
1423/*
1424 * Calling into a user-controlled filesystem gives the filesystem
1425 * daemon ptrace-like capabilities over the current process.  This
1426 * means, that the filesystem daemon is able to record the exact
1427 * filesystem operations performed, and can also control the behavior
1428 * of the requester process in otherwise impossible ways.  For example
1429 * it can delay the operation for arbitrary length of time allowing
1430 * DoS against the requester.
1431 *
1432 * For this reason only those processes can call into the filesystem,
1433 * for which the owner of the mount has ptrace privilege.  This
1434 * excludes processes started by other users, suid or sgid processes.
1435 */
1436bool fuse_allow_current_process(struct fuse_conn *fc)
1437{
1438	bool allow;
1439
1440	if (fc->allow_other)
1441		allow = current_in_userns(fc->user_ns);
1442	else
1443		allow = fuse_permissible_uidgid(fc);
1444
1445	if (!allow && allow_sys_admin_access && capable(CAP_SYS_ADMIN))
1446		allow = true;
1447
1448	return allow;
1449}
1450
1451static int fuse_access(struct inode *inode, int mask)
1452{
1453	struct fuse_mount *fm = get_fuse_mount(inode);
1454	FUSE_ARGS(args);
1455	struct fuse_access_in inarg;
1456	int err;
1457
1458	BUG_ON(mask & MAY_NOT_BLOCK);
1459
1460	if (fm->fc->no_access)
1461		return 0;
1462
1463	memset(&inarg, 0, sizeof(inarg));
1464	inarg.mask = mask & (MAY_READ | MAY_WRITE | MAY_EXEC);
1465	args.opcode = FUSE_ACCESS;
1466	args.nodeid = get_node_id(inode);
1467	args.in_numargs = 1;
1468	args.in_args[0].size = sizeof(inarg);
1469	args.in_args[0].value = &inarg;
1470	err = fuse_simple_request(fm, &args);
1471	if (err == -ENOSYS) {
1472		fm->fc->no_access = 1;
1473		err = 0;
1474	}
1475	return err;
1476}
1477
1478static int fuse_perm_getattr(struct inode *inode, int mask)
1479{
1480	if (mask & MAY_NOT_BLOCK)
1481		return -ECHILD;
1482
1483	forget_all_cached_acls(inode);
1484	return fuse_do_getattr(inode, NULL, NULL);
1485}
1486
1487/*
1488 * Check permission.  The two basic access models of FUSE are:
1489 *
1490 * 1) Local access checking ('default_permissions' mount option) based
1491 * on file mode.  This is the plain old disk filesystem permission
1492 * modell.
1493 *
1494 * 2) "Remote" access checking, where server is responsible for
1495 * checking permission in each inode operation.  An exception to this
1496 * is if ->permission() was invoked from sys_access() in which case an
1497 * access request is sent.  Execute permission is still checked
1498 * locally based on file mode.
1499 */
1500static int fuse_permission(struct mnt_idmap *idmap,
1501			   struct inode *inode, int mask)
1502{
1503	struct fuse_conn *fc = get_fuse_conn(inode);
1504	bool refreshed = false;
1505	int err = 0;
1506
1507	if (fuse_is_bad(inode))
1508		return -EIO;
1509
1510	if (!fuse_allow_current_process(fc))
1511		return -EACCES;
1512
1513	/*
1514	 * If attributes are needed, refresh them before proceeding
1515	 */
1516	if (fc->default_permissions ||
1517	    ((mask & MAY_EXEC) && S_ISREG(inode->i_mode))) {
1518		struct fuse_inode *fi = get_fuse_inode(inode);
1519		u32 perm_mask = STATX_MODE | STATX_UID | STATX_GID;
1520
1521		if (perm_mask & READ_ONCE(fi->inval_mask) ||
1522		    time_before64(fi->i_time, get_jiffies_64())) {
1523			refreshed = true;
1524
1525			err = fuse_perm_getattr(inode, mask);
1526			if (err)
1527				return err;
1528		}
1529	}
1530
1531	if (fc->default_permissions) {
1532		err = generic_permission(&nop_mnt_idmap, inode, mask);
1533
1534		/* If permission is denied, try to refresh file
1535		   attributes.  This is also needed, because the root
1536		   node will at first have no permissions */
1537		if (err == -EACCES && !refreshed) {
1538			err = fuse_perm_getattr(inode, mask);
1539			if (!err)
1540				err = generic_permission(&nop_mnt_idmap,
1541							 inode, mask);
1542		}
1543
1544		/* Note: the opposite of the above test does not
1545		   exist.  So if permissions are revoked this won't be
1546		   noticed immediately, only after the attribute
1547		   timeout has expired */
1548	} else if (mask & (MAY_ACCESS | MAY_CHDIR)) {
1549		err = fuse_access(inode, mask);
1550	} else if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) {
1551		if (!(inode->i_mode & S_IXUGO)) {
1552			if (refreshed)
1553				return -EACCES;
1554
1555			err = fuse_perm_getattr(inode, mask);
1556			if (!err && !(inode->i_mode & S_IXUGO))
1557				return -EACCES;
1558		}
1559	}
1560	return err;
1561}
1562
1563static int fuse_readlink_page(struct inode *inode, struct page *page)
1564{
1565	struct fuse_mount *fm = get_fuse_mount(inode);
1566	struct fuse_page_desc desc = { .length = PAGE_SIZE - 1 };
1567	struct fuse_args_pages ap = {
1568		.num_pages = 1,
1569		.pages = &page,
1570		.descs = &desc,
1571	};
1572	char *link;
1573	ssize_t res;
1574
1575	ap.args.opcode = FUSE_READLINK;
1576	ap.args.nodeid = get_node_id(inode);
1577	ap.args.out_pages = true;
1578	ap.args.out_argvar = true;
1579	ap.args.page_zeroing = true;
1580	ap.args.out_numargs = 1;
1581	ap.args.out_args[0].size = desc.length;
1582	res = fuse_simple_request(fm, &ap.args);
1583
1584	fuse_invalidate_atime(inode);
1585
1586	if (res < 0)
1587		return res;
1588
1589	if (WARN_ON(res >= PAGE_SIZE))
1590		return -EIO;
1591
1592	link = page_address(page);
1593	link[res] = '\0';
1594
1595	return 0;
1596}
1597
1598static const char *fuse_get_link(struct dentry *dentry, struct inode *inode,
1599				 struct delayed_call *callback)
1600{
1601	struct fuse_conn *fc = get_fuse_conn(inode);
1602	struct page *page;
1603	int err;
1604
1605	err = -EIO;
1606	if (fuse_is_bad(inode))
1607		goto out_err;
1608
1609	if (fc->cache_symlinks)
1610		return page_get_link(dentry, inode, callback);
1611
1612	err = -ECHILD;
1613	if (!dentry)
1614		goto out_err;
1615
1616	page = alloc_page(GFP_KERNEL);
1617	err = -ENOMEM;
1618	if (!page)
1619		goto out_err;
1620
1621	err = fuse_readlink_page(inode, page);
1622	if (err) {
1623		__free_page(page);
1624		goto out_err;
1625	}
1626
1627	set_delayed_call(callback, page_put_link, page);
1628
1629	return page_address(page);
1630
1631out_err:
1632	return ERR_PTR(err);
1633}
1634
1635static int fuse_dir_open(struct inode *inode, struct file *file)
1636{
1637	return fuse_open_common(inode, file, true);
1638}
1639
1640static int fuse_dir_release(struct inode *inode, struct file *file)
1641{
1642	fuse_release_common(file, true);
1643
1644	return 0;
1645}
1646
1647static int fuse_dir_fsync(struct file *file, loff_t start, loff_t end,
1648			  int datasync)
1649{
1650	struct inode *inode = file->f_mapping->host;
1651	struct fuse_conn *fc = get_fuse_conn(inode);
1652	int err;
1653
1654	if (fuse_is_bad(inode))
1655		return -EIO;
1656
1657	if (fc->no_fsyncdir)
1658		return 0;
1659
1660	inode_lock(inode);
1661	err = fuse_fsync_common(file, start, end, datasync, FUSE_FSYNCDIR);
1662	if (err == -ENOSYS) {
1663		fc->no_fsyncdir = 1;
1664		err = 0;
1665	}
1666	inode_unlock(inode);
1667
1668	return err;
1669}
1670
1671static long fuse_dir_ioctl(struct file *file, unsigned int cmd,
1672			    unsigned long arg)
1673{
1674	struct fuse_conn *fc = get_fuse_conn(file->f_mapping->host);
1675
1676	/* FUSE_IOCTL_DIR only supported for API version >= 7.18 */
1677	if (fc->minor < 18)
1678		return -ENOTTY;
1679
1680	return fuse_ioctl_common(file, cmd, arg, FUSE_IOCTL_DIR);
1681}
1682
1683static long fuse_dir_compat_ioctl(struct file *file, unsigned int cmd,
1684				   unsigned long arg)
1685{
1686	struct fuse_conn *fc = get_fuse_conn(file->f_mapping->host);
1687
1688	if (fc->minor < 18)
1689		return -ENOTTY;
1690
1691	return fuse_ioctl_common(file, cmd, arg,
1692				 FUSE_IOCTL_COMPAT | FUSE_IOCTL_DIR);
1693}
1694
1695static bool update_mtime(unsigned ivalid, bool trust_local_mtime)
1696{
1697	/* Always update if mtime is explicitly set  */
1698	if (ivalid & ATTR_MTIME_SET)
1699		return true;
1700
1701	/* Or if kernel i_mtime is the official one */
1702	if (trust_local_mtime)
1703		return true;
1704
1705	/* If it's an open(O_TRUNC) or an ftruncate(), don't update */
1706	if ((ivalid & ATTR_SIZE) && (ivalid & (ATTR_OPEN | ATTR_FILE)))
1707		return false;
1708
1709	/* In all other cases update */
1710	return true;
1711}
1712
1713static void iattr_to_fattr(struct fuse_conn *fc, struct iattr *iattr,
1714			   struct fuse_setattr_in *arg, bool trust_local_cmtime)
1715{
1716	unsigned ivalid = iattr->ia_valid;
1717
1718	if (ivalid & ATTR_MODE)
1719		arg->valid |= FATTR_MODE,   arg->mode = iattr->ia_mode;
1720	if (ivalid & ATTR_UID)
1721		arg->valid |= FATTR_UID,    arg->uid = from_kuid(fc->user_ns, iattr->ia_uid);
1722	if (ivalid & ATTR_GID)
1723		arg->valid |= FATTR_GID,    arg->gid = from_kgid(fc->user_ns, iattr->ia_gid);
1724	if (ivalid & ATTR_SIZE)
1725		arg->valid |= FATTR_SIZE,   arg->size = iattr->ia_size;
1726	if (ivalid & ATTR_ATIME) {
1727		arg->valid |= FATTR_ATIME;
1728		arg->atime = iattr->ia_atime.tv_sec;
1729		arg->atimensec = iattr->ia_atime.tv_nsec;
1730		if (!(ivalid & ATTR_ATIME_SET))
1731			arg->valid |= FATTR_ATIME_NOW;
1732	}
1733	if ((ivalid & ATTR_MTIME) && update_mtime(ivalid, trust_local_cmtime)) {
1734		arg->valid |= FATTR_MTIME;
1735		arg->mtime = iattr->ia_mtime.tv_sec;
1736		arg->mtimensec = iattr->ia_mtime.tv_nsec;
1737		if (!(ivalid & ATTR_MTIME_SET) && !trust_local_cmtime)
1738			arg->valid |= FATTR_MTIME_NOW;
1739	}
1740	if ((ivalid & ATTR_CTIME) && trust_local_cmtime) {
1741		arg->valid |= FATTR_CTIME;
1742		arg->ctime = iattr->ia_ctime.tv_sec;
1743		arg->ctimensec = iattr->ia_ctime.tv_nsec;
1744	}
1745}
1746
1747/*
1748 * Prevent concurrent writepages on inode
1749 *
1750 * This is done by adding a negative bias to the inode write counter
1751 * and waiting for all pending writes to finish.
1752 */
1753void fuse_set_nowrite(struct inode *inode)
1754{
1755	struct fuse_inode *fi = get_fuse_inode(inode);
1756
1757	BUG_ON(!inode_is_locked(inode));
1758
1759	spin_lock(&fi->lock);
1760	BUG_ON(fi->writectr < 0);
1761	fi->writectr += FUSE_NOWRITE;
1762	spin_unlock(&fi->lock);
1763	wait_event(fi->page_waitq, fi->writectr == FUSE_NOWRITE);
1764}
1765
1766/*
1767 * Allow writepages on inode
1768 *
1769 * Remove the bias from the writecounter and send any queued
1770 * writepages.
1771 */
1772static void __fuse_release_nowrite(struct inode *inode)
1773{
1774	struct fuse_inode *fi = get_fuse_inode(inode);
1775
1776	BUG_ON(fi->writectr != FUSE_NOWRITE);
1777	fi->writectr = 0;
1778	fuse_flush_writepages(inode);
1779}
1780
1781void fuse_release_nowrite(struct inode *inode)
1782{
1783	struct fuse_inode *fi = get_fuse_inode(inode);
1784
1785	spin_lock(&fi->lock);
1786	__fuse_release_nowrite(inode);
1787	spin_unlock(&fi->lock);
1788}
1789
1790static void fuse_setattr_fill(struct fuse_conn *fc, struct fuse_args *args,
1791			      struct inode *inode,
1792			      struct fuse_setattr_in *inarg_p,
1793			      struct fuse_attr_out *outarg_p)
1794{
1795	args->opcode = FUSE_SETATTR;
1796	args->nodeid = get_node_id(inode);
1797	args->in_numargs = 1;
1798	args->in_args[0].size = sizeof(*inarg_p);
1799	args->in_args[0].value = inarg_p;
1800	args->out_numargs = 1;
1801	args->out_args[0].size = sizeof(*outarg_p);
1802	args->out_args[0].value = outarg_p;
1803}
1804
1805/*
1806 * Flush inode->i_mtime to the server
1807 */
1808int fuse_flush_times(struct inode *inode, struct fuse_file *ff)
1809{
1810	struct fuse_mount *fm = get_fuse_mount(inode);
1811	FUSE_ARGS(args);
1812	struct fuse_setattr_in inarg;
1813	struct fuse_attr_out outarg;
1814
1815	memset(&inarg, 0, sizeof(inarg));
1816	memset(&outarg, 0, sizeof(outarg));
1817
1818	inarg.valid = FATTR_MTIME;
1819	inarg.mtime = inode->i_mtime.tv_sec;
1820	inarg.mtimensec = inode->i_mtime.tv_nsec;
1821	if (fm->fc->minor >= 23) {
1822		inarg.valid |= FATTR_CTIME;
1823		inarg.ctime = inode_get_ctime(inode).tv_sec;
1824		inarg.ctimensec = inode_get_ctime(inode).tv_nsec;
1825	}
1826	if (ff) {
1827		inarg.valid |= FATTR_FH;
1828		inarg.fh = ff->fh;
1829	}
1830	fuse_setattr_fill(fm->fc, &args, inode, &inarg, &outarg);
1831
1832	return fuse_simple_request(fm, &args);
1833}
1834
1835/*
1836 * Set attributes, and at the same time refresh them.
1837 *
1838 * Truncation is slightly complicated, because the 'truncate' request
1839 * may fail, in which case we don't want to touch the mapping.
1840 * vmtruncate() doesn't allow for this case, so do the rlimit checking
1841 * and the actual truncation by hand.
1842 */
1843int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
1844		    struct file *file)
1845{
1846	struct inode *inode = d_inode(dentry);
1847	struct fuse_mount *fm = get_fuse_mount(inode);
1848	struct fuse_conn *fc = fm->fc;
1849	struct fuse_inode *fi = get_fuse_inode(inode);
1850	struct address_space *mapping = inode->i_mapping;
1851	FUSE_ARGS(args);
1852	struct fuse_setattr_in inarg;
1853	struct fuse_attr_out outarg;
1854	bool is_truncate = false;
1855	bool is_wb = fc->writeback_cache && S_ISREG(inode->i_mode);
1856	loff_t oldsize;
1857	int err;
1858	bool trust_local_cmtime = is_wb;
1859	bool fault_blocked = false;
1860
1861	if (!fc->default_permissions)
1862		attr->ia_valid |= ATTR_FORCE;
1863
1864	err = setattr_prepare(&nop_mnt_idmap, dentry, attr);
1865	if (err)
1866		return err;
1867
1868	if (attr->ia_valid & ATTR_SIZE) {
1869		if (WARN_ON(!S_ISREG(inode->i_mode)))
1870			return -EIO;
1871		is_truncate = true;
1872	}
1873
1874	if (FUSE_IS_DAX(inode) && is_truncate) {
1875		filemap_invalidate_lock(mapping);
1876		fault_blocked = true;
1877		err = fuse_dax_break_layouts(inode, 0, 0);
1878		if (err) {
1879			filemap_invalidate_unlock(mapping);
1880			return err;
1881		}
1882	}
1883
1884	if (attr->ia_valid & ATTR_OPEN) {
1885		/* This is coming from open(..., ... | O_TRUNC); */
1886		WARN_ON(!(attr->ia_valid & ATTR_SIZE));
1887		WARN_ON(attr->ia_size != 0);
1888		if (fc->atomic_o_trunc) {
1889			/*
1890			 * No need to send request to userspace, since actual
1891			 * truncation has already been done by OPEN.  But still
1892			 * need to truncate page cache.
1893			 */
1894			i_size_write(inode, 0);
1895			truncate_pagecache(inode, 0);
1896			goto out;
1897		}
1898		file = NULL;
1899	}
1900
1901	/* Flush dirty data/metadata before non-truncate SETATTR */
1902	if (is_wb &&
1903	    attr->ia_valid &
1904			(ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_MTIME_SET |
1905			 ATTR_TIMES_SET)) {
1906		err = write_inode_now(inode, true);
1907		if (err)
1908			return err;
1909
1910		fuse_set_nowrite(inode);
1911		fuse_release_nowrite(inode);
1912	}
1913
1914	if (is_truncate) {
1915		fuse_set_nowrite(inode);
1916		set_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
1917		if (trust_local_cmtime && attr->ia_size != inode->i_size)
1918			attr->ia_valid |= ATTR_MTIME | ATTR_CTIME;
1919	}
1920
1921	memset(&inarg, 0, sizeof(inarg));
1922	memset(&outarg, 0, sizeof(outarg));
1923	iattr_to_fattr(fc, attr, &inarg, trust_local_cmtime);
1924	if (file) {
1925		struct fuse_file *ff = file->private_data;
1926		inarg.valid |= FATTR_FH;
1927		inarg.fh = ff->fh;
1928	}
1929
1930	/* Kill suid/sgid for non-directory chown unconditionally */
1931	if (fc->handle_killpriv_v2 && !S_ISDIR(inode->i_mode) &&
1932	    attr->ia_valid & (ATTR_UID | ATTR_GID))
1933		inarg.valid |= FATTR_KILL_SUIDGID;
1934
1935	if (attr->ia_valid & ATTR_SIZE) {
1936		/* For mandatory locking in truncate */
1937		inarg.valid |= FATTR_LOCKOWNER;
1938		inarg.lock_owner = fuse_lock_owner_id(fc, current->files);
1939
1940		/* Kill suid/sgid for truncate only if no CAP_FSETID */
1941		if (fc->handle_killpriv_v2 && !capable(CAP_FSETID))
1942			inarg.valid |= FATTR_KILL_SUIDGID;
1943	}
1944	fuse_setattr_fill(fc, &args, inode, &inarg, &outarg);
1945	err = fuse_simple_request(fm, &args);
1946	if (err) {
1947		if (err == -EINTR)
1948			fuse_invalidate_attr(inode);
1949		goto error;
1950	}
1951
1952	if (fuse_invalid_attr(&outarg.attr) ||
1953	    inode_wrong_type(inode, outarg.attr.mode)) {
1954		fuse_make_bad(inode);
1955		err = -EIO;
1956		goto error;
1957	}
1958
1959	spin_lock(&fi->lock);
1960	/* the kernel maintains i_mtime locally */
1961	if (trust_local_cmtime) {
1962		if (attr->ia_valid & ATTR_MTIME)
1963			inode->i_mtime = attr->ia_mtime;
1964		if (attr->ia_valid & ATTR_CTIME)
1965			inode_set_ctime_to_ts(inode, attr->ia_ctime);
1966		/* FIXME: clear I_DIRTY_SYNC? */
1967	}
1968
1969	fuse_change_attributes_common(inode, &outarg.attr, NULL,
1970				      ATTR_TIMEOUT(&outarg),
1971				      fuse_get_cache_mask(inode));
1972	oldsize = inode->i_size;
1973	/* see the comment in fuse_change_attributes() */
1974	if (!is_wb || is_truncate)
1975		i_size_write(inode, outarg.attr.size);
1976
1977	if (is_truncate) {
1978		/* NOTE: this may release/reacquire fi->lock */
1979		__fuse_release_nowrite(inode);
1980	}
1981	spin_unlock(&fi->lock);
1982
1983	/*
1984	 * Only call invalidate_inode_pages2() after removing
1985	 * FUSE_NOWRITE, otherwise fuse_launder_folio() would deadlock.
1986	 */
1987	if ((is_truncate || !is_wb) &&
1988	    S_ISREG(inode->i_mode) && oldsize != outarg.attr.size) {
1989		truncate_pagecache(inode, outarg.attr.size);
1990		invalidate_inode_pages2(mapping);
1991	}
1992
1993	clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
1994out:
1995	if (fault_blocked)
1996		filemap_invalidate_unlock(mapping);
1997
1998	return 0;
1999
2000error:
2001	if (is_truncate)
2002		fuse_release_nowrite(inode);
2003
2004	clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
2005
2006	if (fault_blocked)
2007		filemap_invalidate_unlock(mapping);
2008	return err;
2009}
2010
2011static int fuse_setattr(struct mnt_idmap *idmap, struct dentry *entry,
2012			struct iattr *attr)
2013{
2014	struct inode *inode = d_inode(entry);
2015	struct fuse_conn *fc = get_fuse_conn(inode);
2016	struct file *file = (attr->ia_valid & ATTR_FILE) ? attr->ia_file : NULL;
2017	int ret;
2018
2019	if (fuse_is_bad(inode))
2020		return -EIO;
2021
2022	if (!fuse_allow_current_process(get_fuse_conn(inode)))
2023		return -EACCES;
2024
2025	if (attr->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID)) {
2026		attr->ia_valid &= ~(ATTR_KILL_SUID | ATTR_KILL_SGID |
2027				    ATTR_MODE);
2028
2029		/*
2030		 * The only sane way to reliably kill suid/sgid is to do it in
2031		 * the userspace filesystem
2032		 *
2033		 * This should be done on write(), truncate() and chown().
2034		 */
2035		if (!fc->handle_killpriv && !fc->handle_killpriv_v2) {
2036			/*
2037			 * ia_mode calculation may have used stale i_mode.
2038			 * Refresh and recalculate.
2039			 */
2040			ret = fuse_do_getattr(inode, NULL, file);
2041			if (ret)
2042				return ret;
2043
2044			attr->ia_mode = inode->i_mode;
2045			if (inode->i_mode & S_ISUID) {
2046				attr->ia_valid |= ATTR_MODE;
2047				attr->ia_mode &= ~S_ISUID;
2048			}
2049			if ((inode->i_mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) {
2050				attr->ia_valid |= ATTR_MODE;
2051				attr->ia_mode &= ~S_ISGID;
2052			}
2053		}
2054	}
2055	if (!attr->ia_valid)
2056		return 0;
2057
2058	ret = fuse_do_setattr(entry, attr, file);
2059	if (!ret) {
2060		/*
2061		 * If filesystem supports acls it may have updated acl xattrs in
2062		 * the filesystem, so forget cached acls for the inode.
2063		 */
2064		if (fc->posix_acl)
2065			forget_all_cached_acls(inode);
2066
2067		/* Directory mode changed, may need to revalidate access */
2068		if (d_is_dir(entry) && (attr->ia_valid & ATTR_MODE))
2069			fuse_invalidate_entry_cache(entry);
2070	}
2071	return ret;
2072}
2073
2074static int fuse_getattr(struct mnt_idmap *idmap,
2075			const struct path *path, struct kstat *stat,
2076			u32 request_mask, unsigned int flags)
2077{
2078	struct inode *inode = d_inode(path->dentry);
2079	struct fuse_conn *fc = get_fuse_conn(inode);
2080
2081	if (fuse_is_bad(inode))
2082		return -EIO;
2083
2084	if (!fuse_allow_current_process(fc)) {
2085		if (!request_mask) {
2086			/*
2087			 * If user explicitly requested *nothing* then don't
2088			 * error out, but return st_dev only.
2089			 */
2090			stat->result_mask = 0;
2091			stat->dev = inode->i_sb->s_dev;
2092			return 0;
2093		}
2094		return -EACCES;
2095	}
2096
2097	return fuse_update_get_attr(inode, NULL, stat, request_mask, flags);
2098}
2099
2100static const struct inode_operations fuse_dir_inode_operations = {
2101	.lookup		= fuse_lookup,
2102	.mkdir		= fuse_mkdir,
2103	.symlink	= fuse_symlink,
2104	.unlink		= fuse_unlink,
2105	.rmdir		= fuse_rmdir,
2106	.rename		= fuse_rename2,
2107	.link		= fuse_link,
2108	.setattr	= fuse_setattr,
2109	.create		= fuse_create,
2110	.atomic_open	= fuse_atomic_open,
2111	.tmpfile	= fuse_tmpfile,
2112	.mknod		= fuse_mknod,
2113	.permission	= fuse_permission,
2114	.getattr	= fuse_getattr,
2115	.listxattr	= fuse_listxattr,
2116	.get_inode_acl	= fuse_get_inode_acl,
2117	.get_acl	= fuse_get_acl,
2118	.set_acl	= fuse_set_acl,
2119	.fileattr_get	= fuse_fileattr_get,
2120	.fileattr_set	= fuse_fileattr_set,
2121};
2122
2123static const struct file_operations fuse_dir_operations = {
2124	.llseek		= generic_file_llseek,
2125	.read		= generic_read_dir,
2126	.iterate_shared	= fuse_readdir,
2127	.open		= fuse_dir_open,
2128	.release	= fuse_dir_release,
2129	.fsync		= fuse_dir_fsync,
2130	.unlocked_ioctl	= fuse_dir_ioctl,
2131	.compat_ioctl	= fuse_dir_compat_ioctl,
2132};
2133
2134static const struct inode_operations fuse_common_inode_operations = {
2135	.setattr	= fuse_setattr,
2136	.permission	= fuse_permission,
2137	.getattr	= fuse_getattr,
2138	.listxattr	= fuse_listxattr,
2139	.get_inode_acl	= fuse_get_inode_acl,
2140	.get_acl	= fuse_get_acl,
2141	.set_acl	= fuse_set_acl,
2142	.fileattr_get	= fuse_fileattr_get,
2143	.fileattr_set	= fuse_fileattr_set,
2144};
2145
2146static const struct inode_operations fuse_symlink_inode_operations = {
2147	.setattr	= fuse_setattr,
2148	.get_link	= fuse_get_link,
2149	.getattr	= fuse_getattr,
2150	.listxattr	= fuse_listxattr,
2151};
2152
2153void fuse_init_common(struct inode *inode)
2154{
2155	inode->i_op = &fuse_common_inode_operations;
2156}
2157
2158void fuse_init_dir(struct inode *inode)
2159{
2160	struct fuse_inode *fi = get_fuse_inode(inode);
2161
2162	inode->i_op = &fuse_dir_inode_operations;
2163	inode->i_fop = &fuse_dir_operations;
2164
2165	spin_lock_init(&fi->rdc.lock);
2166	fi->rdc.cached = false;
2167	fi->rdc.size = 0;
2168	fi->rdc.pos = 0;
2169	fi->rdc.version = 0;
2170}
2171
2172static int fuse_symlink_read_folio(struct file *null, struct folio *folio)
2173{
2174	int err = fuse_readlink_page(folio->mapping->host, &folio->page);
2175
2176	if (!err)
2177		folio_mark_uptodate(folio);
2178
2179	folio_unlock(folio);
2180
2181	return err;
2182}
2183
2184static const struct address_space_operations fuse_symlink_aops = {
2185	.read_folio	= fuse_symlink_read_folio,
2186};
2187
2188void fuse_init_symlink(struct inode *inode)
2189{
2190	inode->i_op = &fuse_symlink_inode_operations;
2191	inode->i_data.a_ops = &fuse_symlink_aops;
2192	inode_nohighmem(inode);
2193}
2194