1Intro(2) System Calls Intro(2)
2
6 Intro, intro - introduction to system calls and error numbers
7
9 #include <errno.h>
10
13 A system call is a C library function that requests a service from the
14 system, such as getting the time of day. This request is performed in
15 the kernel. The library interface executes a trap into the kernel,
16 which actually executes the system call code.
17 Most system calls return one or more error conditions. An error condi‐
20 tion is indicated by an otherwise impossible return value. This is
21 almost always −1 or the null pointer; the individual descriptions spec‐
22 ify the details. An error number is also made available in the external
23 variable errno, which is not cleared on successful calls, so it should
24 be tested only after an error has been indicated.
25 In the case of multithreaded applications, the -mt option must be spec‐
28 ified on the command line at compilation time (see threads(5)). When
29 the -mt option is specified, errno becomes a macro that enables each
30 thread to have its own errno. This errno macro can be used on either
31 side of the assignment as though it were a variable.
32 An error value listed as "will fail" describes a condition whose detec‐
35 tion and reporting is mandatory for an implementation that conforms to
36 the Single UNIX Specification (SUS). An application can rely on this
37 condition being detected and reported. An error value listed as "may
38 fail" describes a condition whose detection and reporting is optional
39 for an implementation that conforms to the SUS. An application should
40 not rely this condition being detected and reported. An application
41 that relies on such behavior cannot be assured to be portable across
42 conforming implementations. If more than one error occurs in processing
43 a function call, any one of the possible errors might may be returned,
44 as the order of detection is undefined. See standards(5) for additional
45 information regarding the Single UNIX Specification.
46 Each system call description attempts to list all possible error num‐
49 bers. The following is a complete list of the error numbers and their
50 names as defined in <errno.h>.
51 1 EPERM Lacking appropriate privileges
53 Typically this error indicates an attempt to
55 modify a file in some way forbidden except to
56 its owner or an appropriately privileged
57 process. It is also returned for attempts by
58 ordinary users to perform operations allowed
59 only to processes with certain privileges.
60 The manual pages for individual functions docu‐
62 ment which privileges are needed to override the
63 restriction.
64 2 ENOENT No such file or directory
67 A file name is specified and the file should
69 exist but doesn't, or one of the directories in
70 a path name does not exist.
71 3 ESRCH No such process, LWP, or thread
74 No process can be found in the system that cor‐
76 responds to the specified PID, LWPID_t, or
77 thread_t.
78 4 EINTR Interrupted system call
81 An asynchronous signal (such as interrupt or
83 quit), which the user has elected to catch,
84 occurred during a system service function. If
85 execution is resumed after processing the sig‐
86 nal, it will appear as if the interrupted func‐
87 tion call returned this error condition.
88 In a multithreaded application, EINTR may be
90 returned whenever another thread or LWP calls
91 fork(2).
92 5 EIO I/O error
95 Some physical I/O error has occurred. This error
97 may in some cases occur on a call following the
98 one to which it actually applies.
99 6 ENXIO No such device or address
102 I/O on a special file refers to a subdevice
104 which does not exist, or exists beyond the limit
105 of the device. It may also occur when, for exam‐
106 ple, a tape drive is not on-line or no disk pack
107 is loaded on a drive.
108 7 E2BIG Arg list too long
111 An argument list longer than ARG_MAX bytes is
113 presented to a member of the exec family of
114 functions (see exec(2)). The argument list limit
115 is the sum of the size of the argument list plus
116 the size of the environment's exported shell
117 variables.
118 8 ENOEXEC Exec format error
121 A request is made to execute a file which,
123 although it has the appropriate permissions,
124 does not start with a valid format (see
125 a.out(4)).
126 9 EBADF Bad file number
129 Either a file descriptor refers to no open file,
131 or a read(2) (respectively, write(2)) request
132 is made to a file that is open only for writing
133 (respectively, reading).
134 10 ECHILD No child processes
137 A wait(3C) function call was executed by a
139 process that had no existing or unwaited-for
140 child processes.
141 11 EAGAIN No more processes, or no more LWPs
144 For example, the fork(2) function failed because
146 the system's process table is full or the user
147 is not allowed to create any more processes, or
148 a call failed because of insufficient memory or
149 swap space.
150 12 ENOMEM Not enough space
153 During execution of brk() or sbrk() (see
155 brk(2)), or one of the exec family of functions,
156 a program asks for more space than the system is
157 able to supply. This is not a temporary condi‐
158 tion; the maximum size is a system parameter. On
159 some architectures, the error may also occur if
160 the arrangement of text, data, and stack seg‐
161 ments requires too many segmentation registers,
162 or if there is not enough swap space during the
163 fork(2) function.
164 13 EACCES Permission denied
167 An attempt was made to access a file in a way
169 forbidden by the protection system.
170 The manual pages for individual functions docu‐
172 ment which privileges are needed to override the
173 protection system.
174 14 EFAULT Bad address
177 The system encountered a hardware fault in
179 attempting to use an argument of a routine. For
180 example, errno potentially may be set to EFAULT
181 any time a routine that takes a pointer argument
182 is passed an invalid address, if the system can
183 detect the condition. Because systems will dif‐
184 fer in their ability to reliably detect a bad
185 address, on some implementations passing a bad
186 address to a routine will result in undefined
187 behavior.
188 15 ENOTBLK Block device required
191 A non-block device or file was mentioned where a
193 block device was required (for example, in a
194 call to the mount(2) function).
195 16 EBUSY Device busy
198 An attempt was made to mount a device that was
200 already mounted or an attempt was made to
201 unmount a device on which there is an active
202 file (open file, current directory, mounted-on
203 file, active text segment). It will also occur
204 if an attempt is made to enable accounting when
205 it is already enabled. The device or resource is
206 currently unavailable. EBUSY is also used by
207 mutexes, semaphores, condition variables, and
208 r/w locks, to indicate that a lock is held,
209 and by the processor control function P_ONLINE.
210 17 EEXIST File exists
213 An existing file was mentioned in an inappropri‐
215 ate context (for example, call to the link(2)
216 function).
217 18 EXDEV Cross-device link
220 A hard link to a file on another device was
222 attempted.
223 19 ENODEV No such device
226 An attempt was made to apply an inappropriate
228 operation to a device (for example, read a
229 write-only device).
230 20 ENOTDIR Not a directory
233 A non-directory was specified where a directory
235 is required (for example, in a path prefix or as
236 an argument to the chdir(2) function).
237 21 EISDIR Is a directory
240 An attempt was made to write on a directory.
242 22 EINVAL Invalid argument
245 An invalid argument was specified (for example,
247 unmounting a non-mounted device), mentioning an
248 undefined signal in a call to the signal(3C) or
249 kill(2) function, or an unsupported operation
250 related to extended attributes was attempted.
251 23 ENFILE File table overflow
254 The system file table is full (that is,
256 SYS_OPEN files are open, and temporarily no more
257 files can be opened).
258 24 EMFILE Too many open files
261 No process may have more than OPEN_MAX file
263 descriptors open at a time.
264 25 ENOTTY Inappropriate ioctl for device
267 A call was made to the ioctl(2) function speci‐
269 fying a file that is not a special character
270 device.
271 26 ETXTBSY Text file busy (obsolete)
274 An attempt was made to execute a pure-procedure
276 program that is currently open for writing. Also
277 an attempt to open for writing or to remove a
278 pure-procedure program that is being executed.
279 (This message is obsolete.)
280 27 EFBIG File too large
283 The size of the file exceeded the limit speci‐
285 fied by resource RLIMIT_FSIZEn; the file size
286 exceeds the maximum supported by the file sys‐
287 tem; or the file size exceeds the offset maximum
288 of the file descriptor. See the File Descriptor
289 subsection of the DEFINITIONS section below.
290 28 ENOSPC No space left on device
293 While writing an ordinary file or creating a
295 directory entry, there is no free space left on
296 the device. In the fcntl(2) function, the set‐
297 ting or removing of record locks on a file can‐
298 not be accomplished because there are no more
299 record entries left on the system.
300 29 ESPIPE Illegal seek
303 A call to the lseek(2) function was issued to a
305 pipe.
306 30 EROFS Read-only file system
309 An attempt to modify a file or directory was
311 made on a device mounted read-only.
312 31 EMLINK Too many links
315 An attempt to make more than the maximum number
317 of links, LINK_MAX, to a file.
318 32 EPIPE Broken pipe
321 A write on a pipe for which there is no process
323 to read the data. This condition normally gener‐
324 ates a signal; the error is returned if the sig‐
325 nal is ignored.
326 33 EDOM Math argument out of domain of function
329 The argument of a function in the math package
331 (3M) is out of the domain of the function.
332 34 ERANGE Math result not representable
335 The value of a function in the math package (3M)
337 is not representable within machine precision.
338 35 ENOMSG No message of desired type
341 An attempt was made to receive a message of a
343 type that does not exist on the specified mes‐
344 sage queue (see msgrcv(2)).
345 36 EIDRM Identifier removed
348 This error is returned to processes that resume
350 execution due to the removal of an identifier
351 from the file system's name space (see
352 msgctl(2), semctl(2), and shmctl(2)).
353 37 ECHRNG Channel number out of range
356 38 EL2NSYNC Level 2 not synchronized
359 39 EL3HLT Level 3 halted
362 40 EL3RST Level 3 reset
365 41 ELNRNG Link number out of range
368 42 EUNATCH Protocol driver not attached
371 43 ENOCSI No CSI structure available
374 44 EL2HLT Level 2 halted
377 45 EDEADLK Deadlock condition
380 A deadlock situation was detected and avoided.
382 This error pertains to file and record locking,
383 and also applies to mutexes, semaphores, condi‐
384 tion variables, and r/w locks.
385 46 ENOLCK No record locks available
388 There are no more locks available. The system
390 lock table is full (see fcntl(2)).
391 47 ECANCELED Operation canceled
394 The associated asynchronous operation was can‐
396 celed before completion.
397 48 ENOTSUP Not supported
400 This version of the system does not support this
402 feature. Future versions of the system may pro‐
403 vide support.
404 49 EDQUOT Disc quota exceeded
407 A write(2) to an ordinary file, the creation of
409 a directory or symbolic link, or the creation of
410 a directory entry failed because the user's
411 quota of disk blocks was exhausted, or the allo‐
412 cation of an inode for a newly created file
413 failed because the user's quota of inodes was
414 exhausted.
415 58-59 Reserved
418 60 ENOSTR Device not a stream
421 A putmsg(2) or getmsg(2) call was attempted on a
423 file descriptor that is not a STREAMS device.
424 61 ENODATA No data available
427 62 ETIME Timer expired
430 The timer set for a STREAMS ioctl(2) call has
432 expired. The cause of this error is device-spe‐
433 cific and could indicate either a hardware or
434 software failure, or perhaps a timeout value
435 that is too short for the specific operation.
436 The status of the ioctl() operation is indeter‐
437 minate. This is also returned in the case of
438 _lwp_cond_timedwait(2) or cond_timedwait(3C).
439 63 ENOSR Out of stream resources
442 During a STREAMS open(2) call, either no
444 STREAMS queues or no STREAMS head data struc‐
445 tures were available. This is a temporary condi‐
446 tion; one may recover from it if other processes
447 release resources.
448 65 ENOPKG Package not installed
451 This error occurs when users attempt to use a
453 call from a package which has not been
454 installed.
455 71 EPROTO Protocol error
458 Some protocol error occurred. This error is
460 device-specific, but is generally not related to
461 a hardware failure.
462 77 EBADMSG Not a data message
465 During a read(2), getmsg(2), or ioctl(2)
467 I_RECVFD call to a STREAMS device, something has
468 come to the head of the queue that can not be
469 processed. That something depends on the call:
470 read(): control information or passed file
472 descriptor.
473 getmsg(): passed file descriptor.
476 ioctl(): control or data information.
479 78 ENAMETOOLONG File name too long
483 The length of the path argument exceeds
485 PATH_MAX, or the length of a path component
486 exceeds NAME_MAX while _POSIX_NO_TRUNC is in
487 effect; see limits.h(3HEAD).
488 79 EOVERFLOW Value too large for defined data type.
491 80 ENOTUNIQ Name not unique on network
494 Given log name not unique.
496 81 EBADFD File descriptor in bad state
499 Either a file descriptor refers to no open file
501 or a read request was made to a file that is
502 open only for writing.
503 82 EREMCHG Remote address changed
506 83 ELIBACC Cannot access a needed share library
509 Trying to exec an a.out that requires a static
511 shared library and the static shared library
512 does not exist or the user does not have permis‐
513 sion to use it.
514 84 ELIBBAD Accessing a corrupted shared library
517 Trying to exec an a.out that requires a static
519 shared library (to be linked in) and exec could
520 not load the static shared library. The static
521 shared library is probably corrupted.
522 85 ELIBSCN .lib section in a.out corrupted
525 Trying to exec an a.out that requires a static
527 shared library (to be linked in) and there was
528 erroneous data in the .lib section of the a.out.
529 The .lib section tells exec what static shared
530 libraries are needed. The a.out is probably cor‐
531 rupted.
532 86 ELIBMAX Attempting to link in more shared libraries than
535 system limit
536 Trying to exec an a.out that requires more
538 static shared libraries than is allowed on the
539 current configuration of the system. See System
540 Administration Guide: IP Services
541 87 ELIBEXEC Cannot exec a shared library directly
544 Attempting to exec a shared library directly.
546 88 EILSEQ Error 88
549 Illegal byte sequence. Handle multiple charac‐
551 ters as a single character.
552 89 ENOSYS Operation not applicable
555 90 ELOOP Number of symbolic links encountered during path
558 name traversal exceeds MAXSYMLINKS
559 91 ESTART Restartable system call
562 Interrupted system call should be restarted.
564 92 ESTRPIPE If pipe/FIFO, don't sleep in stream head
567 Streams pipe error (not externally visible).
569 93 ENOTEMPTY Directory not empty
572 94 EUSERS Too many users
575 95 ENOTSOCK Socket operation on non-socket
578 96 EDESTADDRREQ Destination address required
581 A required address was omitted from an operation
583 on a transport endpoint. Destination address
584 required.
585 97 EMGSIZE Message too long
588 A message sent on a transport provider was
590 larger than the internal message buffer or some
591 other network limit.
592 98 EPROTOTYPE Protocol wrong type for socket
595 A protocol was specified that does not support
597 the semantics of the socket type requested.
598 99 ENOPROTOOPT Protocol not available
601 A bad option or level was specified when getting
603 or setting options for a protocol.
604 120 EPROTONOSUPPORT Protocol not supported
607 The protocol has not been configured into the
609 system or no implementation for it exists.
610 121 ESOCKTNOSUPPORT Socket type not supported
613 The support for the socket type has not been
615 configured into the system or no implementation
616 for it exists.
617 122 EOPNOTSUPP Operation not supported on transport endpoint
620 For example, trying to accept a connection on a
622 datagram transport endpoint.
623 123 EPFNOSUPPORT Protocol family not supported
626 The protocol family has not been configured into
628 the system or no implementation for it exists.
629 Used for the Internet protocols.
630 124 EAFNOSUPPORT Address family not supported by protocol family
633 An address incompatible with the requested pro‐
635 tocol was used.
636 125 EADDRINUSE Address already in use
639 User attempted to use an address already in use,
641 and the protocol does not allow this.
642 126 EADDRNOTAVAIL Cannot assign requested address
645 Results from an attempt to create a transport
647 endpoint with an address not on the current
648 machine.
649 127 ENETDOWN Network is down
652 Operation encountered a dead network.
654 128 ENETUNREACH Network is unreachable
657 Operation was attempted to an unreachable net‐
659 work.
660 129 ENETRESET Network dropped connection because of reset
663 The host you were connected to crashed and
665 rebooted.
666 130 ECONNABORTED Software caused connection abort
669 A connection abort was caused internal to your
671 host machine.
672 131 ECONNRESET Connection reset by peer
675 A connection was forcibly closed by a peer. This
677 normally results from a loss of the connection
678 on the remote host due to a timeout or a reboot.
679 132 ENOBUFS No buffer space available
682 An operation on a transport endpoint or pipe was
684 not performed because the system lacked suffi‐
685 cient buffer space or because a queue was full.
686 133 EISCONN Transport endpoint is already connected
689 A connect request was made on an already con‐
691 nected transport endpoint; or, a sendto(3SOCKET)
692 or sendmsg(3SOCKET) request on a connected
693 transport endpoint specified a destination when
694 already connected.
695 134 ENOTCONN Transport endpoint is not connected
698 A request to send or receive data was disallowed
700 because the transport endpoint is not connected
701 and (when sending a datagram) no address was
702 supplied.
703 143 ESHUTDOWN Cannot send after transport endpoint shutdown
706 A request to send data was disallowed because
708 the transport endpoint has already been shut
709 down.
710 144 ETOOMANYREFS Too many references: cannot splice
713 145 ETIMEDOUT Connection timed out
716 A connect(3SOCKET) or send(3SOCKET) request
718 failed because the connected party did not prop‐
719 erly respond after a period of time; or a
720 write(2) or fsync(3C) request failed because a
721 file is on an NFS file system mounted with the
722 soft option.
723 146 ECONNREFUSED Connection refused
726 No connection could be made because the target
728 machine actively refused it. This usually
729 results from trying to connect to a service that
730 is inactive on the remote host.
731 147 EHOSTDOWN Host is down
734 A transport provider operation failed because
736 the destination host was down.
737 148 EHOSTUNREACH No route to host
740 A transport provider operation was attempted to
742 an unreachable host.
743 149 EALREADY Operation already in progress
746 An operation was attempted on a non-blocking
748 object that already had an operation in
749 progress.
750 150 EINPROGRESS Operation now in progress
753 An operation that takes a long time to complete
755 (such as a connect()) was attempted on a non-
756 blocking object.
757 151 ESTALE Stale NFS file handle
760
763 Background Process Group
764 Any process group that is not the foreground process group of a ses‐
765 sion that has established a connection with a controlling terminal.
766 Controlling Process
768 A session leader that established a connection to a controlling termi‐
769 nal.
770 Controlling Terminal
772 A terminal that is associated with a session. Each session may have,
773 at most, one controlling terminal associated with it and a controlling
774 terminal may be associated with only one session. Certain input
775 sequences from the controlling terminal cause signals to be sent to
776 process groups in the session associated with the controlling terminal;
777 see termio(7I).
778 Directory
780 Directories organize files into a hierarchical system where directories
781 are the nodes in the hierarchy. A directory is a file that catalogs the
782 list of files, including directories (sub-directories), that are
783 directly beneath it in the hierarchy. Entries in a directory file are
784 called links. A link associates a file identifier with a filename. By
785 convention, a directory contains at least two links, . (dot) and ..
786 (dot-dot). The link called dot refers to the directory itself while
787 dot-dot refers to its parent directory. The root directory, which is
788 the top-most node of the hierarchy, has itself as its parent directory.
789 The pathname of the root directory is / and the parent directory of the
790 root directory is /.
791 Downstream
793 In a stream, the direction from stream head to driver.
794 Driver
796 In a stream, the driver provides the interface between peripheral hard‐
797 ware and the stream. A driver can also be a pseudo-driver, such as a
798 multiplexor or log driver (see log(7D)), which is not associated with a
799 hardware device.
800 Effective User ID and Effective Group ID
802 An active process has an effective user ID and an effective group ID
803 that are used to determine file access permissions (see below). The
804 effective user ID and effective group ID are equal to the process's
805 real user ID and real group ID, respectively, unless the process or one
806 of its ancestors evolved from a file that had the set-user-ID bit or
807 set-group-ID bit set (see exec(2)).
808 File Access Permissions
810 Read, write, and execute/search permissions for a file are granted to a
811 process if one or more of the following are true:
812 o The effective user ID of the process matches the user ID of
814 the owner of the file and the appropriate access bit of the
815 "owner" portion (0700) of the file mode is set.
816 o The effective user ID of the process does not match the
818 user ID of the owner of the file, but either the effective
819 group ID or one of the supplementary group IDs of the
820 process match the group ID of the file and the appropriate
821 access bit of the "group" portion (0070) of the file mode is
822 set.
823 o The effective user ID of the process does not match the user
825 ID of the owner of the file, and neither the effective group
826 ID nor any of the supplementary group IDs of the process
827 match the group ID of the file, but the appropriate access
828 bit of the "other" portion (0007) of the file mode is set.
829 o The read, write, or execute mode bit is not set but the
831 process has the discretionary file access override privilege
832 for the corresponding mode bit: {PRIV_FILE_DAC_READ} for the
833 read bit {PRIV_FILE_DAC_WRITE} for the write bit,
834 {PRIV_FILE_DAC_SEARCH} for the execute bit on directories,
835 and {PRIV_FILE_DAC_EXECUTE} for the executable bit on plain
836 files.
837 Otherwise, the corresponding permissions are denied.
840 File Descriptor
842 A file descriptor is a small integer used to perform I/O on a file. The
843 value of a file descriptor is from 0 to (NOFILES−1). A process may have
844 no more than NOFILES file descriptors open simultaneously. A file
845 descriptor is returned by calls such as open(2) or pipe(2). The file
846 descriptor is used as an argument by calls such as read(2), write(2),
847 ioctl(2), and close(2).
848 Each file descriptor has a corresponding offset maximum. For regular
851 files that were opened without setting the O_LARGEFILE flag, the offset
852 maximum is 2 Gbyte − 1 byte (2^31 −1 bytes). For regular files that
853 were opened with the O_LARGEFILE flag set, the offset maximum is 2^63
854 −1 bytes.
855 File Name
857 Names consisting of 1 to NAME_MAX characters may be used to name an
858 ordinary file, special file or directory.
859 These characters may be selected from the set of all character values
862 excluding \0 (null) and the ASCII code for / (slash).
863 Note that it is generally unwise to use *, ?, [, or ] as part of file
866 names because of the special meaning attached to these characters by
867 the shell (see sh(1), csh(1), and ksh(1)). Although permitted, the use
868 of unprintable characters in file names should be avoided.
869 A file name is sometimes referred to as a pathname component. The
872 interpretation of a pathname component is dependent on the values of
873 NAME_MAX and _POSIX_NO_TRUNC associated with the path prefix of that
874 component. If any pathname component is longer than NAME_MAX and
875 _POSIX_NO_TRUNC is in effect for the path prefix of that component (see
876 fpathconf(2) and limits.h(3HEAD)), it shall be considered an error con‐
877 dition in that implementation. Otherwise, the implementation shall use
878 the first NAME_MAX bytes of the pathname component.
879 Foreground Process Group
881 Each session that has established a connection with a controlling ter‐
882 minal will distinguish one process group of the session as the fore‐
883 ground process group of the controlling terminal. This group has cer‐
884 tain privileges when accessing its controlling terminal that are denied
885 to background process groups.
886 {IOV_MAX}
888 Maximum number of entries in a struct iovec array.
889 {LIMIT}
891 The braces notation, {LIMIT}, is used to denote a magnitude limitation
892 imposed by the implementation. This indicates a value which may be
893 defined by a header file (without the braces), or the actual value may
894 be obtained at runtime by a call to the configuration inquiry path‐
895 conf(2) with the name argument _PC_LIMIT.
896 Masks
898 The file mode creation mask of the process used during any create func‐
899 tion calls to turn off permission bits in the mode argument supplied.
900 Bit positions that are set in umask(cmask) are cleared in the mode of
901 the created file.
902 Message
904 In a stream, one or more blocks of data or information, with associated
905 STREAMS control structures. Messages can be of several defined types,
906 which identify the message contents. Messages are the only means of
907 transferring data and communicating within a stream.
908 Message Queue
910 In a stream, a linked list of messages awaiting processing by a module
911 or driver.
912 Message Queue Identifier
914 A message queue identifier (msqid) is a unique positive integer created
915 by a msgget(2) call. Each msqid has a message queue and a data struc‐
916 ture associated with it. The data structure is referred to as msqid_ds
917 and contains the following members:
918 struct ipc_perm msg_perm;
920 struct msg *msg_first;
921 struct msg *msg_last;
922 ulong_t msg_cbytes;
923 ulong_t msg_qnum;
924 ulong_t msg_qbytes;
925 pid_t msg_lspid;
926 pid_t msg_lrpid;
927 time_t msg_stime;
928 time_t msg_rtime;
929 time_t msg_ctime;
930 The following are descriptions of the msqid_ds structure members:
934 The msg_perm member is an ipc_perm structure that specifies the message
937 operation permission (see below). This structure includes the following
938 members:
939 uid_t cuid; /* creator user id */
941 gid_t cgid; /* creator group id */
942 uid_t uid; /* user id */
943 gid_t gid; /* group id */
944 mode_t mode; /* r/w permission */
945 ulong_t seq; /* slot usage sequence # */
946 key_t key; /* key */
947 The *msg_first member is a pointer to the first message on the queue.
951 The *msg_last member is a pointer to the last message on the queue.
954 The msg_cbytes member is the current number of bytes on the queue.
957 The msg_qnum member is the number of messages currently on the queue.
960 The msg_qbytes member is the maximum number of bytes allowed on the
963 queue.
964 The msg_lspid member is the process ID of the last process that per‐
967 formed a msgsnd() operation.
968 The msg_lrpid member is the process id of the last process that per‐
971 formed a msgrcv() operation.
972 The msg_stime member is the time of the last msgsnd() operation.
975 The msg_rtime member is the time of the last msgrcv() operation.
978 The msg_ctime member is the time of the last msgctl() operation that
981 changed a member of the above structure.
982 Message Operation Permissions
984 In the msgctl(2), msgget(2), msgrcv(2), and msgsnd(2) function descrip‐
985 tions, the permission required for an operation is given as {token},
986 where token is the type of permission needed, interpreted as follows:
987 00400 READ by user
989 00200 WRITE by user
990 00040 READ by group
991 00020 WRITE by group
992 00004 READ by others
993 00002 WRITE by others
994 Read and write permissions for a msqid are granted to a process if one
998 or more of the following are true:
999 o The {PRIV_IPC_DAC_READ} or {PRIV_IPC_DAC_WRITE} privilege is
1001 present in the effective set.
1002 o The effective user ID of the process matches msg_perm.cuid
1004 or msg_perm.uid in the data structure associated with msqid
1005 and the appropriate bit of the "user" portion (0600) of
1006 msg_perm.mode is set.
1007 o Any group ID in the process credentials from the set
1009 (cr_gid, cr_groups) matches msg_perm.cgid or msg_perm.gid
1010 and the appropriate bit of the "group" portion (060) of
1011 msg_perm.mode is set.
1012 o The appropriate bit of the "other" portion (006) of
1014 msg_perm.mode is set."
1015 Otherwise, the corresponding permissions are denied.
1018 Module
1020 A module is an entity containing processing routines for input and out‐
1021 put data. It always exists in the middle of a stream, between the
1022 stream's head and a driver. A module is the STREAMS counterpart to the
1023 commands in a shell pipeline except that a module contains a pair of
1024 functions which allow independent bidirectional (downstream and
1025 upstream) data flow and processing.
1026 Multiplexor
1028 A multiplexor is a driver that allows streams associated with several
1029 user processes to be connected to a single driver, or several drivers
1030 to be connected to a single user process. STREAMS does not provide a
1031 general multiplexing driver, but does provide the facilities for con‐
1032 structing them and for connecting multiplexed configurations of
1033 streams.
1034 Offset Maximum
1036 An offset maximum is an attribute of an open file description repre‐
1037 senting the largest value that can be used as a file offset.
1038 Orphaned Process Group
1040 A process group in which the parent of every member in the group is
1041 either itself a member of the group, or is not a member of the process
1042 group's session.
1043 Path Name
1045 A path name is a null-terminated character string starting with an
1046 optional slash (/), followed by zero or more directory names separated
1047 by slashes, optionally followed by a file name.
1048 If a path name begins with a slash, the path search begins at the root
1051 directory. Otherwise, the search begins from the current working direc‐
1052 tory.
1053 A slash by itself names the root directory.
1056 Unless specifically stated otherwise, the null path name is treated as
1059 if it named a non-existent file.
1060 Privileged User
1062 Solaris software implements a set of privileges that provide fine-
1063 grained control over the actions of processes. The possession of of a
1064 certain privilege allows a process to perform a specific set of
1065 restricted operations. Prior to the Solaris 10 release, a process run‐
1066 ning with uid 0 was granted all privileges. See privileges(5) for the
1067 semantics and the degree of backward compatibility awarded to processes
1068 with an effective uid of 0.
1069 Process ID
1071 Each process in the system is uniquely identified during its lifetime
1072 by a positive integer called a process ID. A process ID cannot be
1073 reused by the system until the process lifetime, process group life‐
1074 time, and session lifetime ends for any process ID, process group ID,
1075 and session ID equal to that process ID. There are threads within a
1076 process with thread IDs thread_t and LWPID_t. These threads are not
1077 visible to the outside process.
1078 Parent Process ID
1080 A new process is created by a currently active process (see fork(2)).
1081 The parent process ID of a process is the process ID of its creator.
1082 Privilege
1084 Having appropriate privilege means having the capability to override
1085 system restrictions.
1086 Process Group
1088 Each process in the system is a member of a process group that is iden‐
1089 tified by a process group ID. Any process that is not a process group
1090 leader may create a new process group and become its leader. Any
1091 process that is not a process group leader may join an existing
1092 process group that shares the same session as the process. A newly
1093 created process joins the process group of its parent.
1094 Process Group Leader
1096 A process group leader is a process whose process ID is the same as its
1097 process group ID.
1098 Process Group ID
1100 Each active process is a member of a process group and is identified by
1101 a positive integer called the process group ID. This ID is the process
1102 ID of the group leader. This grouping permits the signaling of related
1103 processes (see kill(2)).
1104 Process Lifetime
1106 A process lifetime begins when the process is forked and ends after it
1107 exits, when its termination has been acknowledged by its parent
1108 process. See wait(3C).
1109 Process Group Lifetime
1111 A process group lifetime begins when the process group is created by
1112 its process group leader, and ends when the lifetime of the last
1113 process in the group ends or when the last process in the group leaves
1114 the group.
1115 Processor Set ID
1117 The processors in a system may be divided into subsets, known as pro‐
1118 cessor sets. A process bound to one of these sets will run only on pro‐
1119 cessors in that set, and the processors in the set will normally run
1120 only processes that have been bound to the set. Each active processor
1121 set is identified by a positive integer. See pset_create(2).
1122 Read Queue
1124 In a stream, the message queue in a module or driver containing mes‐
1125 sages moving upstream.
1126 Real User ID and Real Group ID
1128 Each user allowed on the system is identified by a positive integer (0
1129 to MAXUID) called a real user ID.
1130 Each user is also a member of a group. The group is identified by a
1133 positive integer called the real group ID.
1134 An active process has a real user ID and real group ID that are set to
1137 the real user ID and real group ID, respectively, of the user responsi‐
1138 ble for the creation of the process.
1139 Root Directory and Current Working Directory
1141 Each process has associated with it a concept of a root directory and a
1142 current working directory for the purpose of resolving path name
1143 searches. The root directory of a process need not be the root direc‐
1144 tory of the root file system.
1145 Saved Resource Limits
1147 Saved resource limits is an attribute of a process that provides some
1148 flexibility in the handling of unrepresentable resource limits, as
1149 described in the exec family of functions and setrlimit(2).
1150 Saved User ID and Saved Group ID
1152 The saved user ID and saved group ID are the values of the effective
1153 user ID and effective group ID just after an exec of a file whose set
1154 user or set group file mode bit has been set (see exec(2)).
1155 Semaphore Identifier
1157 A semaphore identifier (semid) is a unique positive integer created by
1158 a semget(2) call. Each semid has a set of semaphores and a data struc‐
1159 ture associated with it. The data structure is referred to as semid_ds
1160 and contains the following members:
1161 struct ipc_perm sem_perm; /* operation permission struct */
1163 struct sem *sem_base; /* ptr to first semaphore in set */
1164 ushort_t sem_nsems; /* number of sems in set */
1165 time_t sem_otime; /* last operation time */
1166 time_t sem_ctime; /* last change time */
1167 /* Times measured in secs since */
1168 /* 00:00:00 GMT, Jan. 1, 1970 */
1169 The following are descriptions of the semid_ds structure members:
1173 The sem_perm member is an ipc_perm structure that specifies the sema‐
1176 phore operation permission (see below). This structure includes the
1177 following members:
1178 uid_t uid; /* user id */
1180 gid_t gid; /* group id */
1181 uid_t cuid; /* creator user id */
1182 gid_t cgid; /* creator group id */
1183 mode_t mode; /* r/a permission */
1184 ulong_t seq; /* slot usage sequence number */
1185 key_t key; /* key */
1186 The sem_nsems member is equal to the number of semaphores in the set.
1190 Each semaphore in the set is referenced by a nonnegative integer
1191 referred to as a sem_num. sem_num values run sequentially from 0 to the
1192 value of sem_nsems minus 1.
1193 The sem_otime member is the time of the last semop(2) operation.
1196 The sem_ctime member is the time of the last semctl(2) operation that
1199 changed a member of the above structure.
1200 A semaphore is a data structure called sem that contains the following
1203 members:
1204 ushort_t semval; /* semaphore value */
1206 pid_t sempid; /* pid of last operation */
1207 ushort_t semncnt; /* # awaiting semval > cval */
1208 ushort_t semzcnt; /* # awaiting semval = 0 */
1209 The following are descriptions of the sem structure members:
1213 The semval member is a non-negative integer that is the actual value of
1216 the semaphore.
1217 The sempid member is equal to the process ID of the last process that
1220 performed a semaphore operation on this semaphore.
1221 The semncnt member is a count of the number of processes that are cur‐
1224 rently suspended awaiting this semaphore's semval to become greater
1225 than its current value.
1226 The semzcnt member is a count of the number of processes that are cur‐
1229 rently suspended awaiting this semaphore's semval to become 0.
1230 Semaphore Operation Permissions
1232 In the semop(2) and semctl(2) function descriptions, the permission
1233 required for an operation is given as {token}, where token is the type
1234 of permission needed interpreted as follows:
1235 00400 READ by user
1237 00200 ALTER by user
1238 00040 READ by group
1239 00020 ALTER by group
1240 00004 READ by others
1241 00002 ALTER by others
1242 Read and alter permissions for a semid are granted to a process if one
1246 or more of the following are true:
1247 o The {PRIV_IPC_DAC_READ} or {PRIV_IPC_DAC_WRITE} privilege is
1249 present in the effective set.
1250 o The effective user ID of the process matches sem_perm.cuid
1252 or sem_perm.uid in the data structure associated with semid
1253 and the appropriate bit of the "user" portion (0600) of
1254 sem_perm.mode is set.
1255 o The effective group ID of the process matches sem_perm.cgid
1257 or sem_perm.gid and the appropriate bit of the "group" por‐
1258 tion (060) of sem_perm.mode is set.
1259 o The appropriate bit of the "other" portion (06) of
1261 sem_perm.mode is set.
1262 Otherwise, the corresponding permissions are denied.
1265 Session
1267 A session is a group of processes identified by a common ID called a
1268 session ID, capable of establishing a connection with a controlling
1269 terminal. Any process that is not a process group leader may create a
1270 new session and process group, becoming the session leader of the ses‐
1271 sion and process group leader of the process group. A newly created
1272 process joins the session of its creator.
1273 Session ID
1275 Each session in the system is uniquely identified during its lifetime
1276 by a positive integer called a session ID, the process ID of its ses‐
1277 sion leader.
1278 Session Leader
1280 A session leader is a process whose session ID is the same as its
1281 process and process group ID.
1282 Session Lifetime
1284 A session lifetime begins when the session is created by its session
1285 leader, and ends when the lifetime of the last process that is a member
1286 of the session ends, or when the last process that is a member in the
1287 session leaves the session.
1288 Shared Memory Identifier
1290 A shared memory identifier (shmid) is a unique positive integer created
1291 by a shmget(2) call. Each shmid has a segment of memory (referred to as
1292 a shared memory segment) and a data structure associated with it. (Note
1293 that these shared memory segments must be explicitly removed by the
1294 user after the last reference to them is removed.) The data structure
1295 is referred to as shmid_ds and contains the following members:
1296 struct ipc_perm shm_perm; /* operation permission struct */
1298 size_t shm_segsz; /* size of segment */
1299 struct anon_map *shm_amp; /* ptr to region structure */
1300 char pad[4]; /* for swap compatibility */
1301 pid_t shm_lpid; /* pid of last operation */
1302 pid_t shm_cpid; /* creator pid */
1303 shmatt_t shm_nattch; /* number of current attaches */
1304 ulong_t shm_cnattch; /* used only for shminfo */
1305 time_t shm_atime; /* last attach time */
1306 time_t shm_dtime; /* last detach time */
1307 time_t shm_ctime; /* last change time */
1308 /* Times measured in secs since */
1309 /* 00:00:00 GMT, Jan. 1, 1970 */
1310 The following are descriptions of the shmid_ds structure members:
1314 The shm_perm member is an ipc_perm structure that specifies the shared
1317 memory operation permission (see below). This structure includes the
1318 following members:
1319 uid_t cuid; /* creator user id */
1321 gid_t cgid; /* creator group id */
1322 uid_t uid; /* user id */
1323 gid_t gid; /* group id */
1324 mode_t mode; /* r/w permission */
1325 ulong_t seq; /* slot usage sequence # */
1326 key_t key; /* key */
1327 The shm_segsz member specifies the size of the shared memory segment in
1331 bytes.
1332 The shm_cpid member is the process ID of the process that created the
1335 shared memory identifier.
1336 The shm_lpid member is the process ID of the last process that per‐
1339 formed a shmat() or shmdt() operation (see shmop(2)).
1340 The shm_nattch member is the number of processes that currently have
1343 this segment attached.
1344 The shm_atime member is the time of the last shmat() operation (see
1347 shmop(2)).
1348 The shm_dtime member is the time of the last shmdt() operation (see
1351 shmop(2)).
1352 The shm_ctime member is the time of the last shmctl(2) operation that
1355 changed one of the members of the above structure.
1356 Shared Memory Operation Permissions
1358 In the shmctl(2), shmat(), and shmdt() (see shmop(2)) function descrip‐
1359 tions, the permission required for an operation is given as {token},
1360 where token is the type of permission needed interpreted as follows:
1361 00400 READ by user
1363 00200 WRITE by user
1364 00040 READ by group
1365 00020 WRITE by group
1366 00004 READ by others
1367 00002 WRITE by others
1368 Read and write permissions for a shmid are granted to a process if one
1372 or more of the following are true:
1373 o The {PRIV_IPC_DAC_READ} or {PRIV_IPC_DAC_WRITE} privilege is
1375 present in the effective set.
1376 o The effective user ID of the process matches shm_perm.cuid
1378 or shm_perm.uid in the data structure associated with shmid
1379 and the appropriate bit of the "user" portion (0600) of
1380 shm_perm.mode is set.
1381 o The effective group ID of the process matches shm_perm.cgid
1383 or shm_perm.gid and the appropriate bit of the "group" por‐
1384 tion (060) of shm_perm.mode is set.
1385 o The appropriate bit of the "other" portion (06) of
1387 shm_perm.mode is set.
1388 Otherwise, the corresponding permissions are denied.
1391 Special Processes
1393 The process with ID 0 and the process with ID 1 are special processes
1394 referred to as proc0 and proc1; see kill(2). proc0 is the process
1395 scheduler. proc1 is the initialization process (init); proc1 is the
1396 ancestor of every other process in the system and is used to control
1397 the process structure.
1398 STREAMS
1400 A set of kernel mechanisms that support the development of network ser‐
1401 vices and data communication drivers. It defines interface standards
1402 for character input/output within the kernel and between the kernel and
1403 user level processes. The STREAMS mechanism is composed of utility rou‐
1404 tines, kernel facilities and a set of data structures.
1405 Stream
1407 A stream is a full-duplex data path within the kernel between a user
1408 process and driver routines. The primary components are a stream head,
1409 a driver, and zero or more modules between the stream head and driver.
1410 A stream is analogous to a shell pipeline, except that data flow and
1411 processing are bidirectional.
1412 Stream Head
1414 In a stream, the stream head is the end of the stream that provides the
1415 interface between the stream and a user process. The principal func‐
1416 tions of the stream head are processing STREAMS-related system calls
1417 and passing data and information between a user process and the stream.
1418 Upstream
1420 In a stream, the direction from driver to stream head.
1421 Write Queue
1423 In a stream, the message queue in a module or driver containing mes‐
1424 sages moving downstream.
1425
1427 Sun Microsystems, Inc. gratefully acknowledges The Open Group for per‐
1428 mission to reproduce portions of its copyrighted documentation. Origi‐
1429 nal documentation from The Open Group can be obtained online at
1430 http://www.opengroup.org/bookstore/.
1431 The Institute of Electrical and Electronics Engineers and The Open
1434 Group, have given us permission to reprint portions of their documenta‐
1435 tion.
1436 In the following statement, the phrase ``this text'' refers to portions
1439 of the system documentation.
1440 Portions of this text are reprinted and reproduced in electronic form
1443 in the SunOS Reference Manual, from IEEE Std 1003.1, 2004 Edition,
1444 Standard for Information Technology -- Portable Operating System Inter‐
1445 face (POSIX), The Open Group Base Specifications Issue 6, Copyright (C)
1446 2001-2004 by the Institute of Electrical and Electronics Engineers, Inc
1447 and The Open Group. In the event of any discrepancy between these ver‐
1448 sions and the original IEEE and The Open Group Standard, the original
1449 IEEE and The Open Group Standard is the referee document. The original
1450 Standard can be obtained online at http://www.open‐
1451 group.org/unix/online.html.
1452 This notice shall appear on any product containing this material.
1455
1457 standards(5), threads(5)
1458SunOS 5.11 17 Nov 2008 Intro(2)