xref: /netbsd-src/sys/sys/filedesc.h (revision 8ad9ae0b438a309371b3dd0b594d801e1e33acb4)
1 /*	$NetBSD: filedesc.h,v 1.71 2024/11/10 16:14:52 riastradh Exp $	*/
2 
3 /*-
4  * Copyright (c) 2008 The NetBSD Foundation, Inc.
5  * All rights reserved.
6  *
7  * Redistribution and use in source and binary forms, with or without
8  * modification, are permitted provided that the following conditions
9  * are met:
10  * 1. Redistributions of source code must retain the above copyright
11  *    notice, this list of conditions and the following disclaimer.
12  * 2. Redistributions in binary form must reproduce the above copyright
13  *    notice, this list of conditions and the following disclaimer in the
14  *    documentation and/or other materials provided with the distribution.
15  *
16  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
17  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
18  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
19  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
20  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
21  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
22  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
23  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
24  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
25  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
26  * POSSIBILITY OF SUCH DAMAGE.
27  */
28 
29 /*
30  * Copyright (c) 1990, 1993
31  *	The Regents of the University of California.  All rights reserved.
32  *
33  * Redistribution and use in source and binary forms, with or without
34  * modification, are permitted provided that the following conditions
35  * are met:
36  * 1. Redistributions of source code must retain the above copyright
37  *    notice, this list of conditions and the following disclaimer.
38  * 2. Redistributions in binary form must reproduce the above copyright
39  *    notice, this list of conditions and the following disclaimer in the
40  *    documentation and/or other materials provided with the distribution.
41  * 3. Neither the name of the University nor the names of its contributors
42  *    may be used to endorse or promote products derived from this software
43  *    without specific prior written permission.
44  *
45  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
46  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
47  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
48  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
49  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
50  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
51  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
52  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
53  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
54  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
55  * SUCH DAMAGE.
56  *
57  *	@(#)filedesc.h	8.1 (Berkeley) 6/2/93
58  */
59 
60 #ifndef _SYS_FILEDESC_H_
61 #define	_SYS_FILEDESC_H_
62 
63 #include <sys/param.h>
64 #include <sys/queue.h>
65 #include <sys/mutex.h>
66 #include <sys/rwlock.h>
67 #include <sys/condvar.h>
68 
69 /*
70  * This structure is used for the management of descriptors.  It may be
71  * shared by multiple processes.
72  *
73  * A process is initially started out with NDFILE descriptors stored within
74  * this structure, selected to be enough for typical applications based on
75  * the historical limit of 20 open files (and the usage of descriptors by
76  * shells).  If these descriptors are exhausted, a larger descriptor table
77  * may be allocated, up to a process' resource limit; the internal arrays
78  * are then unused.  The initial expansion is set to NDEXTENT; each time
79  * it runs out, it is doubled until the resource limit is reached. NDEXTENT
80  * should be selected to be the biggest multiple of OFILESIZE (see below)
81  * that will fit in a power-of-two sized piece of memory.
82  */
83 #define	NDFILE		20
84 #define	NDEXTENT	50		/* 250 bytes in 256-byte alloc */
85 #define	NDENTRIES	32		/* 32 fds per entry */
86 #define	NDENTRYMASK	(NDENTRIES - 1)
87 #define	NDENTRYSHIFT	5		/* bits per entry */
88 #define	NDLOSLOTS(x)	(((x) + NDENTRIES - 1) >> NDENTRYSHIFT)
89 #define	NDHISLOTS(x)	((NDLOSLOTS(x) + NDENTRIES - 1) >> NDENTRYSHIFT)
90 #define	NDFDFILE	6		/* first 6 descriptors are free */
91 
92 /*
93  * Process-private descriptor reference, one for each descriptor slot
94  * in use.  Locks:
95  *
96  * :	unlocked
97  * a	atomic operations + filedesc_t::fd_lock in some cases
98  * d	filedesc_t::fd_lock
99  *
100  * Note that ff_exclose and ff_allocated are likely to be byte sized
101  * (bool).  In general adjacent sub-word sized fields must be locked
102  * the same way, but in this case it's ok: ff_exclose can only be
103  * modified while the descriptor slot is live, and ff_allocated when
104  * it's invalid.
105  *
106  * NOTE: ff_exclose should generally be set with fd_set_exclose(), not
107  * written to directly, when implementing flags like O_CLOEXEC or
108  * SOCK_CLOEXEC, so that struct filedesc::fd_exclose is updated as
109  * needed.  See PR kern/58855: close-on-exec is broken for dup3 and
110  * opening cloning devices.
111  */
112 typedef struct fdfile {
113 	bool		ff_exclose;	/* :: close on exec (fd_set_exclose) */
114 	bool		ff_allocated;	/* d: descriptor slot is allocated */
115 	u_int		ff_refcnt;	/* a: reference count on structure */
116 	struct file	*ff_file;	/* d: pointer to file if open */
117 	SLIST_HEAD(,knote) ff_knlist;	/* d: knotes attached to this fd */
118 	kcondvar_t	ff_closing;	/* d: notifier for close */
119 } fdfile_t;
120 
121 #define FDFILE_SIZE ((sizeof(fdfile_t)+CACHE_LINE_SIZE-1)/CACHE_LINE_SIZE*CACHE_LINE_SIZE)
122 
123 /* Reference count */
124 #define	FR_CLOSING	(0x80000000)	/* closing: must interlock */
125 #define	FR_MASK		(~FR_CLOSING)	/* reference count */
126 
127 /*
128  * Open file table, potentially many 'active' tables per filedesc_t
129  * in a multi-threaded process, or with a shared filedesc_t (clone()).
130  * nfiles is first to avoid pointer arithmetic.
131  */
132 typedef struct fdtab {
133 	u_int		dt_nfiles;	/* number of open files allocated */
134 	struct fdtab	*dt_link;	/* for lists of dtab */
135 	fdfile_t	*dt_ff[NDFILE];	/* file structures for open fds */
136 } fdtab_t;
137 
138 typedef struct filedesc {
139 	/*
140 	 * Built-in fdfile_t records first, since they have strict
141 	 * alignment requirements.
142 	 */
143 	uint8_t		fd_dfdfile[NDFDFILE][FDFILE_SIZE];
144 	/*
145 	 * All of the remaining fields are locked by fd_lock.
146 	 */
147 	kmutex_t	fd_lock;	/* lock on structure */
148 	fdtab_t * volatile fd_dt;	/* active descriptor table */
149 	uint32_t	*fd_himap;	/* each bit points to 32 fds */
150 	uint32_t	*fd_lomap;	/* bitmap of free fds */
151 	struct klist	*fd_knhash;	/* hash of attached non-fd knotes */
152 	int		fd_lastkqfile;	/* max descriptor for kqueue */
153 	int		fd_lastfile;	/* high-water mark of fd_ofiles */
154 	int		fd_refcnt;	/* reference count */
155 	u_long		fd_knhashmask;	/* size of fd_knhash */
156 	int		fd_freefile;	/* approx. next free file */
157 	int		fd_unused;	/* unused */
158 	bool		fd_exclose;	/* non-zero if >0 fd with EXCLOSE */
159 	/*
160 	 * This structure is used when the number of open files is
161 	 * <= NDFILE, and are then pointed to by the pointers above.
162 	 */
163 	fdtab_t		fd_dtbuiltin;
164 	/*
165 	 * These arrays are used when the number of open files is
166 	 * <= 1024, and are then pointed to by the pointers above.
167 	 */
168 #define fd_startzero	fd_dhimap	/* area to zero on return to cache */
169 	uint32_t	fd_dhimap[NDENTRIES >> NDENTRYSHIFT];
170 	uint32_t	fd_dlomap[NDENTRIES];
171 } filedesc_t;
172 
173 /*
174  * Working directory, root and umask information.  Serialization:
175  *
176  * a	atomic operations
177  * l	cwdi_lock
178  */
179 typedef struct cwdinfo {
180 	struct vnode	*cwdi_cdir;	/* l: current directory */
181 	struct vnode	*cwdi_rdir;	/* l: root directory */
182 	struct vnode	*cwdi_edir;	/* l: emulation root (if known) */
183 	u_int		cwdi_cmask;	/* a: mask for file creation */
184 	u_int		cwdi_refcnt;	/* a: reference count */
185 
186 	krwlock_t	cwdi_lock	/* :: lock on struct */
187 	    __aligned(COHERENCY_UNIT);	/* -> gets own cache line */
188 } cwdinfo_t;
189 
190 #ifdef _KERNEL
191 
192 struct fileops;
193 struct socket;
194 struct proc;
195 
196 extern struct cwdinfo cwdi0;
197 
198 /*
199  * Kernel global variables and routines.
200  */
201 void	fd_sys_init(void);
202 int	fd_open(const char*, int, int, int*);
203 int	fd_dupopen(int, bool, int, int *);
204 int	fd_alloc(struct proc *, int, int *);
205 void	fd_tryexpand(struct proc *);
206 int	fd_allocfile(file_t **, int *);
207 void	fd_affix(struct proc *, file_t *, unsigned);
208 void	fd_abort(struct proc *, file_t *, unsigned);
209 filedesc_t *fd_copy(void);
210 filedesc_t *fd_init(filedesc_t *);
211 void	fd_share(proc_t *);
212 void	fd_hold(lwp_t *);
213 void	fd_free(void);
214 void	fd_closeexec(void);
215 void	fd_ktrexecfd(void);
216 int	fd_checkstd(void);
217 file_t	*fd_getfile(unsigned);
218 file_t	*fd_getfile2(proc_t *, unsigned);
219 void	fd_putfile(unsigned);
220 int	fd_getvnode(unsigned, file_t **);
221 int	fd_getsock(unsigned, struct socket **);
222 int	fd_getsock1(unsigned, struct socket **, file_t **);
223 void	fd_putvnode(unsigned);
224 void	fd_putsock(unsigned);
225 int	fd_close(unsigned);
226 int	fd_dup(file_t *, int, int *, bool);
227 int	fd_dup2(file_t *, unsigned, int);
228 int	fd_clone(file_t *, unsigned, int, const struct fileops *, void *);
229 void	fd_set_exclose(struct lwp *, int, bool);
230 int	pipe1(struct lwp *, int *, int);
231 int	dodup(struct lwp *, int, int, int, register_t *);
232 
233 void	cwd_sys_init(void);
234 struct cwdinfo *cwdinit(void);
235 void	cwdshare(proc_t *);
236 void	cwdunshare(proc_t *);
237 void	cwdfree(struct cwdinfo *);
238 void	cwdexec(struct proc *);
239 
240 #define GETCWD_CHECK_ACCESS 0x0001
241 int	getcwd_common(struct vnode *, struct vnode *, char **, char *, int,
242     int, struct lwp *);
243 int	vnode_to_path(char *, size_t, struct vnode *, struct lwp *,
244     struct proc *);
245 
246 int	closef(file_t *);
247 file_t *fgetdummy(void);
248 void	fputdummy(file_t *);
249 
250 struct stat;
251 int	do_sys_fstat(int, struct stat *);
252 struct flock;
253 int	do_fcntl_lock(int, int, struct flock *);
254 int	do_posix_fadvise(int, off_t, off_t, int);
255 
256 extern kmutex_t filelist_lock;
257 extern filedesc_t filedesc0;
258 
259 #endif /* _KERNEL */
260 
261 #endif /* !_SYS_FILEDESC_H_ */
262