1proc(4) File Formats proc(4)
2
6 proc - /proc, the process file system
7
9 /proc is a file system that provides access to the state of each
10 process and light-weight process (lwp) in the system. The name of each
11 entry in the /proc directory is a decimal number corresponding to a
12 process-ID. These entries are themselves subdirectories. Access to
13 process state is provided by additional files contained within each
14 subdirectory; the hierarchy is described more completely below. In this
15 document, ``/proc file'' refers to a non-directory file within the
16 hierarchy rooted at /proc. The owner of each /proc file and subdirec‐
17 tory is determined by the user-ID of the process.
18 /proc can be mounted on any mount point, in addition to the standard
21 /proc mount point, and can be mounted several places at once. Such
22 additional mounts are allowed in order to facilitate the confinement of
23 processes to subtrees of the file system via chroot(1M) and yet allow
24 such processes access to commands like ps(1).
25 Standard system calls are used to access /proc files: open(2),
28 close(2), read(2), and write(2) (including readv(2), writev(2),
29 pread(2), and pwrite(2)). Most files describe process state and can
30 only be opened for reading. ctl and lwpctl (control) files permit
31 manipulation of process state and can only be opened for writing. as
32 (address space) files contain the image of the running process and can
33 be opened for both reading and writing. An open for writing allows
34 process control; a read-only open allows inspection but not control. In
35 this document, we refer to the process as open for reading or writing
36 if any of its associated /proc files is open for reading or writing.
37 In general, more than one process can open the same /proc file at the
40 same time. Exclusive open is an advisory mechanism provided to allow
41 controlling processes to avoid collisions with each other. A process
42 can obtain exclusive control of a target process, with respect to other
43 cooperating processes, if it successfully opens any /proc file in the
44 target process for writing (the as or ctl files, or the lwpctl file of
45 any lwp) while specifying O_EXCL in the open(2). Such an open will fail
46 if the target process is already open for writing (that is, if an as,
47 ctl, or lwpctl file is already open for writing). There can be any num‐
48 ber of concurrent read-only opens; O_EXCL is ignored on opens for read‐
49 ing. It is recommended that the first open for writing by a controlling
50 process use the O_EXCL flag; multiple controlling processes usually
51 result in chaos.
52 If a process opens one of its own /proc files for writing, the open
55 succeeds regardless of O_EXCL and regardless of whether some other
56 process has the process open for writing. Self-opens do not count when
57 another process attempts an exclusive open. (A process cannot exclude a
58 debugger by opening itself for writing and the application of a debug‐
59 ger cannot prevent a process from opening itself.) All self-opens for
60 writing are forced to be close-on-exec (see the F_SETFD operation of
61 fcntl(2)).
62 Data may be transferred from or to any locations in the address space
65 of the traced process by applying lseek(2) to position the as file at
66 the virtual address of interest followed by read(2) or write(2) (or by
67 using pread(2) or pwrite(2) for the combined operation). The address-
68 map files /proc/pid/map and /proc/pid/xmap can be read to determine the
69 accessible areas (mappings) of the address space. I/O transfers may
70 span contiguous mappings. An I/O request extending into an unmapped
71 area is truncated at the boundary. A write request beginning at an
72 unmapped virtual address fails with EIO; a read request beginning at an
73 unmapped virtual address returns zero (an end-of-file indication).
74 Information and control operations are provided through additional
77 files. <procfs.h> contains definitions of data structures and message
78 formats used with these files. Some of these definitions involve the
79 use of sets of flags. The set types sigset_t, fltset_t, and sysset_t
80 correspond, respectively, to signal, fault, and system call enumera‐
81 tions defined in <sys/signal.h>, <sys/fault.h>, and <sys/syscall.h>.
82 Each set type is large enough to hold flags for its own enumeration.
83 Although they are of different sizes, they have a common structure and
84 can be manipulated by these macros:
85 prfillset(&set); /* turn on all flags in set */
87 premptyset(&set); /* turn off all flags in set */
88 praddset(&set, flag); /* turn on the specified flag */
89 prdelset(&set, flag); /* turn off the specified flag */
90 r = prismember(&set, flag); /* != 0 iff flag is turned on */
91 One of prfillset() or premptyset() must be used to initialize set
95 before it is used in any other operation. flag must be a member of the
96 enumeration corresponding to set.
97 Every process contains at least one light-weight process, or lwp. Each
100 lwp represents a flow of execution that is independently scheduled by
101 the operating system. All lwps in a process share its address space as
102 well as many other attributes. Through the use of lwpctl and ctl files
103 as described below, it is possible to affect individual lwps in a
104 process or to affect all of them at once, depending on the operation.
105 When the process has more than one lwp, a representative lwp is chosen
108 by the system for certain process status files and control operations.
109 The representative lwp is a stopped lwp only if all of the process's
110 lwps are stopped; is stopped on an event of interest only if all of the
111 lwps are so stopped (excluding PR_SUSPENDED lwps); is in a PR_REQUESTED
112 stop only if there are no other events of interest to be found; or,
113 failing everything else, is in a PR_SUSPENDED stop (implying that the
114 process is deadlocked). See the description of the status file for def‐
115 initions of stopped states. See the PCSTOP control operation for the
116 definition of ``event of interest''.
117 The representative lwp remains fixed (it will be chosen again on the
120 next operation) as long as all of the lwps are stopped on events of
121 interest or are in a PR_SUSPENDED stop and the PCRUN control operation
122 is not applied to any of them.
123 When applied to the process control file, every /proc control operation
126 that must act on an lwp uses the same algorithm to choose which lwp to
127 act upon. Together with synchronous stopping (see PCSET), this enables
128 a debugger to control a multiple-lwp process using only the process-
129 level status and control files if it so chooses. More fine-grained con‐
130 trol can be achieved using the lwp-specific files.
131 The system supports two process data models, the traditional 32-bit
134 data model in which ints, longs and pointers are all 32 bits wide (the
135 ILP32 data model), and on some platforms the 64-bit data model in which
136 longs and pointers, but not ints, are 64 bits in width (the LP64 data
137 model). In the LP64 data model some system data types, notably size_t,
138 off_t, time_t and dev_t, grow from 32 bits to 64 bits as well.
139 The /proc interfaces described here are available to both 32-bit and
142 64-bit controlling processes. However, many operations attempted by a
143 32-bit controlling process on a 64-bit target process will fail with
144 EOVERFLOW because the address space range of a 32-bit process cannot
145 encompass a 64-bit process or because the data in some 64-bit system
146 data type cannot be compressed to fit into the corresponding 32-bit
147 type without loss of information. Operations that fail in this circum‐
148 stance include reading and writing the address space, reading the
149 address-map files, and setting the target process's registers. There is
150 no restriction on operations applied by a 64-bit process to either a
151 32-bit or a 64-bit target processes.
152 The format of the contents of any /proc file depends on the data model
155 of the observer (the controlling process), not on the data model of the
156 target process. A 64-bit debugger does not have to translate the infor‐
157 mation it reads from a /proc file for a 32-bit process from 32-bit for‐
158 mat to 64-bit format. However, it usually has to be aware of the data
159 model of the target process. The pr_dmodel field of the status files
160 indicates the target process's data model.
161 To help deal with system data structures that are read from 32-bit pro‐
164 cesses, a 64-bit controlling program can be compiled with the C pre‐
165 processor symbol _SYSCALL32 defined before system header files are
166 included. This makes explicit 32-bit fixed-width data structures (like
167 cstruct stat32) visible to the 64-bit program. See types32.h(3HEAD).
168
170 At the top level, the directory /proc contains entries each of which
171 names an existing process in the system. These entries are themselves
172 directories. Except where otherwise noted, the files described below
173 can be opened for reading only. In addition, if a process becomes a
174 zombie (one that has exited but whose parent has not yet performed a
175 wait(3C) upon it), most of its associated /proc files disappear from
176 the hierarchy; subsequent attempts to open them, or to read or write
177 files opened before the process exited, will elicit the error ENOENT.
178 Although process state and consequently the contents of /proc files can
181 change from instant to instant, a single read(2) of a /proc file is
182 guaranteed to return a sane representation of state; that is, the read
183 will be atomic with respect to the state of the process. No such guar‐
184 antee applies to successive reads applied to a /proc file for a running
185 process. In addition, atomicity is not guaranteed for I/O applied to
186 the as (address-space) file for a running process or for a process
187 whose address space contains memory shared by another running process.
188 A number of structure definitions are used to describe the files. These
191 structures may grow by the addition of elements at the end in future
192 releases of the system and it is not legitimate for a program to assume
193 that they will not.
194
196 A given directory /proc/pid contains the following entries. A process
197 can use the invisible alias /proc/self if it wishes to open one of its
198 own /proc files (invisible in the sense that the name ``self'' does not
199 appear in a directory listing of /proc obtained from ls(1), get‐
200 dents(2), or readdir(3C)).
201 contracts
203 A directory containing references to the contracts held by the process.
204 Each entry is a symlink to the contract's directory under /system/con‐
205 tract. See contract(4).
206 as
208 Contains the address-space image of the process; it can be opened for
209 both reading and writing. lseek(2) is used to position the file at the
210 virtual address of interest and then the address space can be examined
211 or changed through read(2) or write(2) (or by using pread(2) or
212 pwrite(2) for the combined operation).
213 ctl
215 A write-only file to which structured messages are written directing
216 the system to change some aspect of the process's state or control its
217 behavior in some way. The seek offset is not relevant when writing to
218 this file. Individual lwps also have associated lwpctl files in the lwp
219 subdirectories. A control message may be written either to the
220 process's ctl file or to a specific lwpctl file with operation-specific
221 effects. The effect of a control message is immediately reflected in
222 the state of the process visible through appropriate status and infor‐
223 mation files. The types of control messages are described in detail
224 later. See CONTROL MESSAGES.
225 status
227 Contains state information about the process and the representative
228 lwp. The file contains a pstatus structure which contains an embedded
229 lwpstatus structure for the representative lwp, as follows:
230 typedef struct pstatus {
232 int pr_flags; /* flags (see below) */
233 int pr_nlwp; /* number of active lwps in the process */
234 int pr_nzomb; /* number of zombie lwps in the process */
235 pid_tpr_pid; /* process id */
236 pid_tpr_ppid; /* parent process id */
237 pid_tpr_pgid; /* process group id */
238 pid_tpr_sid; /* session id */
239 id_t pr_aslwpid; /* obsolete */
240 id_t pr_agentid; /* lwp-id of the agent lwp, if any */
241 sigset_t pr_sigpend; /* set of process pending signals */
242 uintptr_t pr_brkbase; /* virtual address of the process heap */
243 size_t pr_brksize; /* size of the process heap, in bytes */
244 uintptr_t pr_stkbase; /* virtual address of the process stack */
245 size_tpr_stksize; /* size of the process stack, in bytes */
246 timestruc_t pr_utime; /* process user cpu time */
247 timestruc_t pr_stime; /* process system cpu time */
248 timestruc_t pr_cutime; /* sum of children's user times */
249 timestruc_t pr_cstime; /* sum of children's system times */
250 sigset_t pr_sigtrace; /* set of traced signals */
251 fltset_t pr_flttrace; /* set of traced faults */
252 sysset_t pr_sysentry; /* set of system calls traced on entry */
253 sysset_t pr_sysexit; /* set of system calls traced on exit */
254 char pr_dmodel; /* data model of the process */
255 taskid_t pr_taskid; /* task id */
256 projid_t pr_projid; /* project id */
257 zoneid_t pr_zoneid; /* zone id */
258 lwpstatus_t pr_lwp; /* status of the representative lwp */
259 } pstatus_t;
260 pr_flags is a bit-mask holding the following process flags. For conve‐
264 nience, it also contains the lwp flags for the representative lwp,
265 described later.
266 PR_ISSYS process is a system process (see PCSTOP).
268 PR_VFORKP process is the parent of a vforked child (see PCWATCH).
271 PR_FORK process has its inherit-on-fork mode set (see PCSET).
274 PR_RLC process has its run-on-last-close mode set (see PCSET).
277 PR_KLC process has its kill-on-last-close mode set (see PCSET).
280 PR_ASYNC process has its asynchronous-stop mode set (see PCSET).
283 PR_MSACCT Set by default in all processes to indicate that
286 microstate accounting is enabled. However, this flag has
287 been deprecated and no longer has any effect. Microstate
288 accounting may not be disabled; however, it is still pos‐
289 sible to toggle the flag.
290 PR_MSFORK Set by default in all processes to indicate that
293 microstate accounting will be enabled for processes that
294 this parent forks(). However, this flag has been depre‐
295 cated and no longer has any effect. It is possible to tog‐
296 gle this flag; however, it is not possible to disable
297 microstate accounting.
298 PR_BPTADJ process has its breakpoint adjustment mode set (see
301 PCSET).
302 PR_PTRACE process has its ptrace-compatibility mode set (see PCSET).
305 pr_nlwp is the total number of active lwps in the process. pr_nzomb is
309 the total number of zombie lwps in the process. A zombie lwp is a non-
310 detached lwp that has terminated but has not been reaped with
311 thr_join(3C) or pthread_join(3C).
312 pr_pid, pr_ppid, pr_pgid, and pr_sid are, respectively, the process ID,
315 the ID of the process's parent, the process's process group ID, and the
316 process's session ID.
317 pr_aslwpid is obsolete and is always zero.
320 pr_agentid is the lwp-ID for the /proc agent lwp (see the PCAGENT con‐
323 trol operation). It is zero if there is no agent lwp in the process.
324 pr_sigpend identifies asynchronous signals pending for the process.
327 pr_brkbase is the virtual address of the process heap and pr_brksize is
330 its size in bytes. The address formed by the sum of these values is the
331 process break (see brk(2)). pr_stkbase and pr_stksize are, respec‐
332 tively, the virtual address of the process stack and its size in bytes.
333 (Each lwp runs on a separate stack; the distinguishing characteristic
334 of the process stack is that the operating system will grow it when
335 necessary.)
336 pr_utime, pr_stime, pr_cutime, and pr_cstime are, respectively, the
339 user CPU and system CPU time consumed by the process, and the cumula‐
340 tive user CPU and system CPU time consumed by the process's children,
341 in seconds and nanoseconds.
342 pr_sigtrace and pr_flttrace contain, respectively, the set of signals
345 and the set of hardware faults that are being traced (see PCSTRACE and
346 PCSFAULT).
347 pr_sysentry and pr_sysexit contain, respectively, the sets of system
350 calls being traced on entry and exit (see PCSENTRY and PCSEXIT).
351 pr_dmodel indicates the data model of the process. Possible values are:
354 PR_MODEL_ILP32 process data model is ILP32.
356 PR_MODEL_LP64 process data model is LP64.
359 PR_MODEL_NATIVE process data model is native.
362 The pr_taskid, pr_projid, and pr_zoneid fields contain respectively,
366 the numeric IDs of the task, project, and zone in which the process was
367 running.
368 The constant PR_MODEL_NATIVE reflects the data model of the controlling
371 process, that is, its value is PR_MODEL_ILP32 or PR_MODEL_LP64 accord‐
372 ing to whether the controlling process has been compiled as a 32-bit
373 program or a 64-bit program, respectively.
374 pr_lwp contains the status information for the representative lwp:
377 typedef struct lwpstatus {
379 int pr_flags; /* flags (see below) */
380 id_t pr_lwpid; /* specific lwp identifier */
381 short pr_why; /* reason for lwp stop, if stopped */
382 short pr_what; /* more detailed reason */
383 short pr_cursig; /* current signal, if any */
384 siginfo_t pr_info; /* info associated with signal or fault */
385 sigset_t pr_lwppend; /* set of signals pending to the lwp */
386 sigset_t pr_lwphold; /* set of signals blocked by the lwp */
387 struct sigaction pr_action;/* signal action for current signal */
388 stack_t pr_altstack; /* alternate signal stack info */
389 uintptr_t pr_oldcontext; /* address of previous ucontext */
390 short pr_syscall; /* system call number (if in syscall) */
391 short pr_nsysarg; /* number of arguments to this syscall */
392 int pr_errno; /* errno for failed syscall */
393 long pr_sysarg[PRSYSARGS]; /* arguments to this syscall */
394 long pr_rval1; /* primary syscall return value */
395 long pr_rval2; /* second syscall return value, if any */
396 char pr_clname[PRCLSZ]; /* scheduling class name */
397 timestruc_t pr_tstamp; /* real-time time stamp of stop */
398 timestruc_t pr_utime; /* lwp user cpu time */
399 timestruc_t pr_stime; /* lwp system cpu time */
400 uintptr_t pr_ustack; /* stack boundary data (stack_t) address */
401 ulong_t pr_instr; /* current instruction */
402 prgregset_t pr_reg; /* general registers */
403 prfpregset_t pr_fpreg; /* floating-point registers */
404 } lwpstatus_t;
405 pr_flags is a bit-mask holding the following lwp flags. For conve‐
409 nience, it also contains the process flags, described previously.
410 PR_STOPPED The lwp is stopped.
412 PR_ISTOP The lwp is stopped on an event of interest (see PCSTOP).
415 PR_DSTOP The lwp has a stop directive in effect (see PCSTOP).
418 PR_STEP The lwp has a single-step directive in effect (see
421 PCRUN).
422 PR_ASLEEP The lwp is in an interruptible sleep within a system
425 call.
426 PR_PCINVAL The lwp's current instruction (pr_instr) is undefined.
429 PR_DETACH This is a detached lwp (see pthread_create(3C) and
432 pthread_join(3C)).
433 PR_DAEMON This is a daemon lwp (see pthread_create(3C)).
436 PR_ASLWP This flag is obsolete and is never set.
439 PR_AGENT This is the /proc agent lwp for the process.
442 pr_lwpid names the specific lwp.
446 pr_why and pr_what together describe, for a stopped lwp, the reason for
449 the stop. Possible values of pr_why and the associated pr_what are:
450 PR_REQUESTED indicates that the stop occurred in response to a stop
452 directive, normally because PCSTOP was applied or
453 because another lwp stopped on an event of interest
454 and the asynchronous-stop flag (see PCSET) was not set
455 for the process. pr_what is unused in this case.
456 PR_SIGNALLED indicates that the lwp stopped on receipt of a signal
459 (see PCSTRACE); pr_what holds the signal number that
460 caused the stop (for a newly-stopped lwp, the same
461 value is in pr_cursig).
462 PR_FAULTED indicates that the lwp stopped on incurring a hardware
465 fault (see PCSFAULT); pr_what holds the fault number
466 that caused the stop.
467 PR_SYSENTRY indicate a stop on entry to or exit from a system call
470 PR_SYSEXIT (see PCSENTRY and PCSEXIT); pr_what holds the system
471 call number.
472 PR_JOBCONTROL indicates that the lwp stopped due to the default
475 action of a job control stop signal (see sigac‐
476 tion(2)); pr_what holds the stopping signal number.
477 PR_SUSPENDED indicates that the lwp stopped due to internal syn‐
480 chronization of lwps within the process. pr_what is
481 unused in this case.
482 pr_cursig names the current signal, that is, the next signal to be
486 delivered to the lwp, if any. pr_info, when the lwp is in a PR_SIG‐
487 NALLED or PR_FAULTED stop, contains additional information pertinent to
488 the particular signal or fault (see <sys/siginfo.h>).
489 pr_lwppend identifies any synchronous or directed signals pending for
492 the lwp. pr_lwphold identifies those signals whose delivery is being
493 blocked by the lwp (the signal mask).
494 pr_action contains the signal action information pertaining to the cur‐
497 rent signal (see sigaction(2)); it is undefined if pr_cursig is zero.
498 pr_altstack contains the alternate signal stack information for the lwp
499 (see sigaltstack(2)).
500 pr_oldcontext, if not zero, contains the address on the lwp stack of a
503 ucontext structure describing the previous user-level context (see
504 ucontext.h(3HEAD)). It is non-zero only if the lwp is executing in the
505 context of a signal handler.
506 pr_syscall is the number of the system call, if any, being executed by
509 the lwp; it is non-zero if and only if the lwp is stopped on PR_SYSEN‐
510 TRY or PR_SYSEXIT, or is asleep within a system call ( PR_ASLEEP is
511 set). If pr_syscall is non-zero, pr_nsysarg is the number of arguments
512 to the system call and pr_sysarg contains the actual arguments.
513 pr_rval1, pr_rval2, and pr_errno are defined only if the lwp is stopped
516 on PR_SYSEXIT or if the PR_VFORKP flag is set. If pr_errno is zero,
517 pr_rval1 and pr_rval2 contain the return values from the system call.
518 Otherwise, pr_errno contains the error number for the failing system
519 call (see <sys/errno.h>).
520 pr_clname contains the name of the lwp's scheduling class.
523 pr_tstamp, if the lwp is stopped, contains a time stamp marking when
526 the lwp stopped, in real time seconds and nanoseconds since an arbi‐
527 trary time in the past.
528 pr_utime is the amount of user level CPU time used by this LWP.
531 pr_stime is the amount of system level CPU time used by this LWP.
534 pr_ustack is the virtual address of the stack_t that contains the stack
537 boundaries for this LWP. See getustack(2) and _stack_grow(3C).
538 pr_instr contains the machine instruction to which the lwp's program
541 counter refers. The amount of data retrieved from the process is
542 machine-dependent. On SPARC based machines, it is a 32-bit word. On
543 x86-based machines, it is a single byte. In general, the size is that
544 of the machine's smallest instruction. If PR_PCINVAL is set, pr_instr
545 is undefined; this occurs whenever the lwp is not stopped or when the
546 program counter refers to an invalid virtual address.
547 pr_reg is an array holding the contents of a stopped lwp's general reg‐
550 isters.
551 SPARC On SPARC-based machines, the predefined constants
553 R_G0 ... R_G7, R_O0 ... R_O7, R_L0 ... R_L7, R_I0
554 ... R_I7, R_PC, R_nPC, and R_Y can be used as
555 indices to refer to the corresponding registers;
556 previous register windows can be read from their
557 overflow locations on the stack (however, see the
558 gwindows file in the /proc/pid/lwp/lwpid subdirec‐
559 tory).
560 SPARC V8 (32-bit) For SPARC V8 (32-bit) controlling processes, the
563 predefined constants R_PSR, R_WIM, and R_TBR can
564 be used as indices to refer to the corresponding
565 special registers. For SPARC V9 (64-bit) control‐
566 ling processes, the predefined constants R_CCR,
567 R_ASI, and R_FPRS can be used as indices to refer
568 to the corresponding special registers.
569 x86 (32-bit) For 32-bit x86 processes, the predefined constants
572 listed belowcan be used as indices to refer to the
573 corresponding registers.
574 SS
576 UESP
577 EFL
578 CS
579 EIP
580 ERR
581 TRAPNO
582 EAX
583 ECX
584 EDX
585 EBX
586 ESP
587 EBP
588 ESI
589 EDI
590 DS
591 ES
592 GS
593 The preceding constants are listed in
595 <sys/regset.h>.
596 Note that a 32-bit process can run on an x86
598 64-bit system, using the constants listed above.
599 x86 (64-bit) To read the registers of a 32- or a 64-bit
602 process, a 64-bit x86 process should use the pre‐
603 defined constants listed below.
604 REG_GSBASE
606 REG_FSBASE
607 REG_DS
608 REG_ES
609 REG_GS
610 REG_FS
611 REG_SS
612 REG_RSP
613 REG_RFL
614 REG_CS
615 REG_RIP
616 REG_ERR
617 REG_TRAPNO
618 REG_RAX
619 REG_RCX
620 REG_RDX
621 REG_RBX
622 REG_RBP
623 REG_RSI
624 REG_RDI
625 REG_R8
626 REG_R9
627 REG_R10
628 REG_R11
629 REG_R12
630 REG_R13
631 REG_R14
632 REG_R15
633 The preceding constants are listed in
635 <sys/regset.h>.
636 pr_fpreg is a structure holding the contents of the floating-point reg‐
640 isters.
641 SPARC registers, both general and floating-point, as seen by a 64-bit
644 controlling process are the V9 versions of the registers, even if the
645 target process is a 32-bit (V8) process. V8 registers are a subset of
646 the V9 registers.
647 If the lwp is not stopped, all register values are undefined.
650 psinfo
652 Contains miscellaneous information about the process and the represen‐
653 tative lwp needed by the ps(1) command. psinfo remains accessible after
654 a process becomes a zombie. The file contains a psinfo structure which
655 contains an embedded lwpsinfo structure for the representative lwp, as
656 follows:
657 typedef struct psinfo {
659 int pr_flag; /* process flags (DEPRECATED: see below) */
660 int pr_nlwp; /* number of active lwps in the process */
661 int pr_nzomb; /* number of zombie lwps in the process */
662 pid_t pr_pid; /* process id */
663 pid_t pr_ppid; /* process id of parent */
664 pid_t pr_pgid; /* process id of process group leader */
665 pid_t pr_sid; /* session id */
666 uid_t pr_uid; /* real user id */
667 uid_t pr_euid; /* effective user id */
668 gid_t pr_gid; /* real group id */
669 gid_t pr_egid; /* effective group id */
670 uintptr_t pr_addr; /* address of process */
671 size_t pr_size; /* size of process image in Kbytes */
672 size_t pr_rssize; /* resident set size in Kbytes */
673 dev_t pr_ttydev; /* controlling tty device (or PRNODEV) */
674 ushort_t pr_pctcpu; /* % of recent cpu time used by all lwps */
675 ushort_t pr_pctmem; /* % of system memory used by process */
676 timestruc_t pr_start; /* process start time, from the epoch */
677 timestruc_t pr_time; /* cpu time for this process */
678 timestruc_t pr_ctime; /* cpu time for reaped children */
679 char pr_fname[PRFNSZ]; /* name of exec'ed file */
680 char pr_psargs[PRARGSZ]; /* initial characters of arg list */
681 int pr_wstat; /* if zombie, the wait() status */
682 int pr_argc; /* initial argument count */
683 uintptr_t pr_argv; /* address of initial argument vector */
684 uintptr_t pr_envp; /* address of initial environment vector */
685 char pr_dmodel; /* data model of the process */
686 lwpsinfo_t pr_lwp; /* information for representative lwp */
687 taskid_t pr_taskid; /* task id */
688 projid_t pr_projid; /* project id */
689 poolid_t pr_poolid; /* pool id */
690 zoneid_t pr_zoneid; /* zone id */
691 ctid_t pr_contract; /* process contract id */
692 } psinfo_t;
693 Some of the entries in psinfo, such as pr_addr, refer to internal ker‐
697 nel data structures and should not be expected to retain their meanings
698 across different versions of the operating system.
699 psinfo_t.pr_flag is a deprecated interface that should no longer be
702 used. Applications currently relying on the SSYS bit in pr_flag should
703 migrate to checking PR_ISSYS in the pstatus structure's pr_flags field.
704 pr_pctcpu and pr_pctmem are 16-bit binary fractions in the range 0.0 to
707 1.0 with the binary point to the right of the high-order bit (1.0 ==
708 0x8000). pr_pctcpu is the summation over all lwps in the process.
709 pr_lwp contains the ps(1) information for the representative lwp. If
712 the process is a zombie, pr_nlwp, pr_nzomb, and pr_lwp.pr_lwpid are
713 zero and the other fields of pr_lwp are undefined:
714 typedef struct lwpsinfo {
716 int pr_flag; /* lwp flags (DEPRECATED: see below) */
717 id_t pr_lwpid; /* lwp id */
718 uintptr_t pr_addr; /* internal address of lwp */
719 uintptr_t pr_wchan; /* wait addr for sleeping lwp */
720 char pr_stype; /* synchronization event type */
721 char pr_state; /* numeric lwp state */
722 char pr_sname; /* printable character for pr_state */
723 char pr_nice; /* nice for cpu usage */
724 short pr_syscall; /* system call number (if in syscall) */
725 char pr_oldpri; /* pre-SVR4, low value is high priority */
726 char pr_cpu; /* pre-SVR4, cpu usage for scheduling */
727 int pr_pri; /* priority, high value = high priority */
728 ushort_t pr_pctcpu; /* % of recent cpu time used by this lwp */
729 timestruc_t pr_start; /* lwp start time, from the epoch */
730 timestruc_t pr_time; /* cpu time for this lwp */
731 char pr_clname[PRCLSZ]; /* scheduling class name */
732 char pr_name[PRFNSZ]; /* name of system lwp */
733 processorid_t pr_onpro; /* processor which last ran this lwp */
734 processorid_t pr_bindpro;/* processor to which lwp is bound */
735 psetid_t pr_bindpset; /* processor set to which lwp is bound */
736 lgrp_id_t pr_lgrp /* home lgroup */
737 } lwpsinfo_t;
738 Some of the entries in lwpsinfo, such as pr_addr, pr_wchan, pr_stype,
742 pr_state, and pr_name, refer to internal kernel data structures and
743 should not be expected to retain their meanings across different ver‐
744 sions of the operating system.
745 lwpsinfo_t.pr_flag is a deprecated interface that should no longer be
748 used.
749 pr_pctcpu is a 16-bit binary fraction, as described above. It repre‐
752 sents the CPU time used by the specific lwp. On a multi-processor
753 machine, the maximum value is 1/N, where N is the number of CPUs.
754 pr_contract is the id of the process contract of which the process is a
757 member. See contract(4) and process(4).
758 cred
760 Contains a description of the credentials associated with the process:
761 typedef struct prcred {
763 uid_t pr_euid; /* effective user id */
764 uid_t pr_ruid; /* real user id */
765 uid_t pr_suid; /* saved user id (from exec) */
766 gid_t pr_egid; /* effective group id */
767 gid_t pr_rgid; /* real group id */
768 gid_t pr_sgid; /* saved group id (from exec) */
769 int pr_ngroups; /* number of supplementary groups */
770 gid_t pr_groups[1]; /* array of supplementary groups */
771 } prcred_t;
772 The array of associated supplementary groups in pr_groups is of vari‐
777 able length; the cred file contains all of the supplementary groups.
778 pr_ngroups indicates the number of supplementary groups. (See also the
779 PCSCRED and PCSCREDX control operations.)
780 priv
782 Contains a description of the privileges associated with the process:
783 typedef struct prpriv {
785 uint32_t pr_nsets; /* number of privilege set */
786 uint32_t pr_setsize; /* size of privilege set */
787 uint32_t pr_infosize; /* size of supplementary data */
788 priv_chunk_t pr_sets[1]; /* array of sets */
789 } prpriv_t;
790 The actual dimension of the pr_sets[] field is
794 pr_sets[pr_nsets][pr_setsize]
796 which is followed by additional information about the process state
800 pr_infosize bytes in size.
801 The full size of the structure can be computed using
804 PRIV_PRPRIV_SIZE(prpriv_t *).
805 sigact
807 Contains an array of sigaction structures describing the current dispo‐
808 sitions of all signals associated with the traced process (see sigac‐
809 tion(2)). Signal numbers are displaced by 1 from array indices, so that
810 the action for signal number n appears in position n-1 of the array.
811 auxv
813 Contains the initial values of the process's aux vector in an array of
814 auxv_t structures (see <sys/auxv.h>). The values are those that were
815 passed by the operating system as startup information to the dynamic
816 linker.
817 ldt
819 This file exists only on x86-based machines. It is non-empty only if
820 the process has established a local descriptor table (LDT). If non-
821 empty, the file contains the array of currently active LDT entries in
822 an array of elements of type struct ssd, defined in <sys/sysi86.h>, one
823 element for each active LDT entry.
824 map, xmap
826 Contain information about the virtual address map of the process. The
827 map file contains an array of prmap structures while the xmap file con‐
828 tains an array of prxmap structures. Each structure describes a con‐
829 tiguous virtual address region in the address space of the traced
830 process:
831 typedef struct prmap {
833 uintptr_tpr_vaddr; /* virtual address of mapping */
834 size_t pr_size; /* size of mapping in bytes */
835 char pr_mapname[PRMAPSZ]; /* name in /proc/pid/object */
836 offset_t pr_offset; /* offset into mapped object, if any */
837 int pr_mflags; /* protection and attribute flags */
838 int pr_pagesize; /* pagesize for this mapping in bytes */
839 int pr_shmid; /* SysV shared memory identifier */
840 } prmap_t;
841 typedef struct prxmap {
845 uintptr_t pr_vaddr; /* virtual address of mapping */
846 size_t pr_size; /* size of mapping in bytes */
847 char pr_mapname[PRMAPSZ]; /* name in /proc/pid/object */
848 offset_t pr_offset; /* offset into mapped object, if any */
849 int pr_mflags; /* protection and attribute flags */
850 int pr_pagesize; /* pagesize for this mapping in bytes */
851 int pr_shmid; /* SysV shared memory identifier */
852 dev_t pr_dev; /* device of mapped object, if any */
853 uint64_t pr_ino; /* inode of mapped object, if any */
854 size_t pr_rss; /* pages of resident memory */
855 size_t pr_anon; /* pages of resident anonymous memory */
856 size_t pr_locked; /* pages of locked memory */
857 uint64_t pr_hatpagesize; /* pagesize of mapping */
858 } prxmap_t;
859 pr_vaddr is the virtual address of the mapping within the traced
864 process and pr_size is its size in bytes. pr_mapname, if it does not
865 contain a null string, contains the name of a file in the object direc‐
866 tory (see below) that can be opened read-only to obtain a file descrip‐
867 tor for the mapped file associated with the mapping. This enables a
868 debugger to find object file symbol tables without having to know the
869 real path names of the executable file and shared libraries of the
870 process. pr_offset is the 64-bit offset within the mapped file (if any)
871 to which the virtual address is mapped.
872 pr_mflags is a bit-mask of protection and attribute flags:
875 MA_READ mapping is readable by the traced process.
877 MA_WRITE mapping is writable by the traced process.
880 MA_EXEC mapping is executable by the traced process.
883 MA_SHARED mapping changes are shared by the mapped object.
886 MA_ISM mapping is intimate shared memory (shared MMU
889 resources)
890 MAP_NORESERVE mapping does not have swap space reserved (mapped with
893 MAP_NORESERVE)
894 MA_SHM mapping System V shared memory
897 A contiguous area of the address space having the same underlying
901 mapped object may appear as multiple mappings due to varying read,
902 write, and execute attributes. The underlying mapped object does not
903 change over the range of a single mapping. An I/O operation to a map‐
904 ping marked MA_SHARED fails if applied at a virtual address not corre‐
905 sponding to a valid page in the underlying mapped object. A write to a
906 MA_SHARED mapping that is not marked MA_WRITE fails. Reads and writes
907 to private mappings always succeed. Reads and writes to unmapped
908 addresses fail.
909 pr_pagesize is the page size for the mapping, currently always the sys‐
912 tem pagesize.
913 pr_shmid is the shared memory identifier, if any, for the mapping. Its
916 value is −1 if the mapping is not System V shared memory. See
917 shmget(2).
918 pr_dev is the device of the mapped object, if any, for the mapping. Its
921 value is PRNODEV (-1) if the mapping does not have a device.
922 pr_ino is the inode of the mapped object, if any, for the mapping. Its
925 contents are only valid if pr_dev is not PRNODEV.
926 pr_rss is the number of resident pages of memory for the mapping. The
929 number of resident bytes for the mapping may be determined by multiply‐
930 ing pr_rss by the page size given by pr_pagesize.
931 pr_anon is the number of resident anonymous memory pages (pages which
934 are private to this process) for the mapping.
935 pr_locked is the number of locked pages for the mapping. Pages which
938 are locked are always resident in memory.
939 pr_hatpagesize is the size, in bytes, of the HAT (MMU) translation for
942 the mapping. pr_hatpagesize may be different than pr_pagesize. The pos‐
943 sible values are hardware architecture specific, and may change over a
944 mapping's lifetime.
945 rmap
947 Contains information about the reserved address ranges of the process.
948 The file contains an array of prmap structures, as defined above for
949 the map file. Each structure describes a contiguous virtual address
950 region in the address space of the traced process that is reserved by
951 the system in the sense that an mmap(2) system call that does not spec‐
952 ify MAP_FIXED will not use any part of it for the new mapping. Examples
953 of such reservations include the address ranges reserved for the
954 process stack and the individual thread stacks of a multi-threaded
955 process.
956 cwd
958 A symbolic link to the process's current working directory. See
959 chdir(2). A readlink(2) of /proc/pid/cwd yields a null string. However,
960 it can be opened, listed, and searched as a directory, and can be the
961 target of chdir(2).
962 root
964 A symbolic link to the process's root directory. /proc/pid/root can
965 differ from the system root directory if the process or one of its
966 ancestors executed chroot(2) as super user. It has the same semantics
967 as /proc/pid/cwd.
968 fd
970 A directory containing references to the open files of the process.
971 Each entry is a decimal number corresponding to an open file descriptor
972 in the process.
973 If an entry refers to a regular file, it can be opened with normal file
976 system semantics but, to ensure that the controlling process cannot
977 gain greater access than the controlled process, with no file access
978 modes other than its read/write open modes in the controlled process.
979 If an entry refers to a directory, it can be accessed with the same
980 semantics as /proc/pid/cwd. An attempt to open any other type of entry
981 fails with EACCES.
982 object
984 A directory containing read-only files with names corresponding to the
985 pr_mapname entries in the map and pagedata files. Opening such a file
986 yields a file descriptor for the underlying mapped file associated with
987 an address-space mapping in the process. The file name a.out appears in
988 the directory as an alias for the process's executable file.
989 The object directory makes it possible for a controlling process to
992 gain access to the object file and any shared libraries (and conse‐
993 quently the symbol tables) without having to know the actual path names
994 of the executable files.
995 path
997 A directory containing symbolic links to files opened by the process.
998 The directory includes one entry for cwd and root. The directory also
999 contains a numerical entry for each file descriptor in the fd direc‐
1000 tory, and entries matching those in the object directory. If this
1001 information is not available, any attempt to read the contents of the
1002 symbolic link will fail. This is most common for files that do not
1003 exist in the filesystem namespace (such as FIFOs and sockets), but can
1004 also happen for regular files. For the file descriptor entries, the
1005 path may be different from the one used by the process to open the
1006 file.
1007 pagedata
1009 Opening the page data file enables tracking of address space references
1010 and modifications on a per-page basis.
1011 A read(2) of the page data file descriptor returns structured page data
1014 and atomically clears the page data maintained for the file by the sys‐
1015 tem. That is to say, each read returns data collected since the last
1016 read; the first read returns data collected since the file was opened.
1017 When the call completes, the read buffer contains the following struc‐
1018 ture as its header and thereafter contains a number of section header
1019 structures and associated byte arrays that must be accessed by walking
1020 linearly through the buffer.
1021 typedef struct prpageheader {
1023 timestruc_t pr_tstamp; /* real time stamp, time of read() */
1024 ulong_t pr_nmap; /* number of address space mappings */
1025 ulong_t pr_npage; /* total number of pages */
1026 } prpageheader_t;
1027 The header is followed by pr_nmap prasmap structures and associated
1031 data arrays. The prasmap structure contains the following elements:
1032 typedef struct prasmap {
1034 uintptr_t pr_vaddr; /* virtual address of mapping */
1035 ulong_t pr_npage; /* number of pages in mapping */
1036 char pr_mapname[PRMAPSZ]; /* name in /proc/pid/object */
1037 offset_t pr_offset; /* offset into mapped object, if any */
1038 int pr_mflags; /* protection and attribute flags */
1039 int pr_pagesize; /* pagesize for this mapping in bytes */
1040 int pr_shmid; /* SysV shared memory identifier */
1041 } prasmap_t;
1042 Each section header is followed by pr_npage bytes, one byte for each
1046 page in the mapping, plus 0-7 null bytes at the end so that the next
1047 prasmap structure begins on an eight-byte aligned boundary. Each data
1048 byte may contain these flags:
1049 PG_REFERENCED page has been referenced.
1051 PG_MODIFIED page has been modified.
1054 If the read buffer is not large enough to contain all of the page data,
1058 the read fails with E2BIG and the page data is not cleared. The
1059 required size of the read buffer can be determined through fstat(2).
1060 Application of lseek(2) to the page data file descriptor is ineffec‐
1061 tive; every read starts from the beginning of the file. Closing the
1062 page data file descriptor terminates the system overhead associated
1063 with collecting the data.
1064 More than one page data file descriptor for the same process can be
1067 opened, up to a system-imposed limit per traced process. A read of one
1068 does not affect the data being collected by the system for the others.
1069 An open of the page data file will fail with ENOMEM if the system-
1070 imposed limit would be exceeded.
1071 watch
1073 Contains an array of prwatch structures, one for each watched area
1074 established by the PCWATCH control operation. See PCWATCH for details.
1075 usage
1077 Contains process usage information described by a prusage structure
1078 which contains at least the following fields:
1079 typedef struct prusage {
1081 id_t pr_lwpid; /* lwp id. 0: process or defunct */
1082 int pr_count; /* number of contributing lwps */
1083 timestruc_t pr_tstamp; /* real time stamp, time of read() */
1084 timestruc_t pr_create; /* process/lwp creation time stamp */
1085 timestruc_t pr_term; /* process/lwp termination time stamp */
1086 timestruc_t pr_rtime; /* total lwp real (elapsed) time */
1087 timestruc_t pr_utime; /* user level CPU time */
1088 timestruc_t pr_stime; /* system call CPU time */
1089 timestruc_t pr_ttime; /* other system trap CPU time */
1090 timestruc_t pr_tftime; /* text page fault sleep time */
1091 timestruc_t pr_dftime; /* data page fault sleep time */
1092 timestruc_t pr_kftime; /* kernel page fault sleep time */
1093 timestruc_t pr_ltime; /* user lock wait sleep time */
1094 timestruc_t pr_slptime; /* all other sleep time */
1095 timestruc_t pr_wtime; /* wait-cpu (latency) time */
1096 timestruc_t pr_stoptime; /* stopped time */
1097 ulong_t pr_minf; /* minor page faults */
1098 ulong_t pr_majf; /* major page faults */
1099 ulong_t pr_nswap; /* swaps */
1100 ulong_t pr_inblk; /* input blocks */
1101 ulong_t pr_oublk; /* output blocks */
1102 ulong_t pr_msnd; /* messages sent */
1103 ulong_t pr_mrcv; /* messages received */
1104 ulong_t pr_sigs; /* signals received */
1105 ulong_t pr_vctx; /* voluntary context switches */
1106 ulong_t pr_ictx; /* involuntary context switches */
1107 ulong_t pr_sysc; /* system calls */
1108 ulong_t pr_ioch; /* chars read and written */
1109 } prusage_t;
1110 Microstate accounting is now continuously enabled. While this informa‐
1114 tion was previously an estimate, if microstate accounting were not
1115 enabled, the current information is now never an estimate represents
1116 time the process has spent in various states.
1117 lstatus
1119 Contains a prheader structure followed by an array of lwpstatus struc‐
1120 tures, one for each active lwp in the process (see also
1121 /proc/pid/lwp/lwpid/lwpstatus, below). The prheader structure describes
1122 the number and size of the array entries that follow.
1123 typedef struct prheader {
1125 long pr_nent; /* number of entries */
1126 size_t pr_entsize; /* size of each entry, in bytes */
1127 } prheader_t;
1128 The lwpstatus structure may grow by the addition of elements at the end
1132 in future releases of the system. Programs must use pr_entsize in the
1133 file header to index through the array. These comments apply to all
1134 /proc files that include a prheader structure (lpsinfo and lusage,
1135 below).
1136 lpsinfo
1138 Contains a prheader structure followed by an array of lwpsinfo struc‐
1139 tures, one for eachactive and zombie lwp in the process. See also
1140 /proc/pid/lwp/lwpid/lwpsinfo, below.
1141 lusage
1143 Contains a prheader structure followed by an array of prusage struc‐
1144 tures, one for each active lwp in the process, plus an additional ele‐
1145 ment at the beginning that contains the summation over all defunct lwps
1146 (lwps that once existed but no longer exist in the process). Excluding
1147 the pr_lwpid, pr_tstamp, pr_create, and pr_term entries, the entry-by-
1148 entry summation over all these structures is the definition of the
1149 process usage information obtained from the usage file. (See also
1150 /proc/pid/lwp/lwpid/lwpusage, below.)
1151 lwp
1153 A directory containing entries each of which names an active or zombie
1154 lwp within the process. These entries are themselves directories con‐
1155 taining additional files as described below. Only the lwpsinfo file
1156 exists in the directory of a zombie lwp.
1157
1159 A given directory /proc/pid/lwp/lwpid contains the following entries:
1160 lwpctl
1162 Write-only control file. The messages written to this file affect the
1163 specific lwp rather than the representative lwp, as is the case for the
1164 process's ctl file.
1165 lwpstatus
1167 lwp-specific state information. This file contains the lwpstatus struc‐
1168 ture for the specific lwp as described above for the representative lwp
1169 in the process's status file.
1170 lwpsinfo
1172 lwp-specific ps(1) information. This file contains the lwpsinfo struc‐
1173 ture for the specific lwp as described above for the representative lwp
1174 in the process's psinfo file. The lwpsinfo file remains accessible
1175 after an lwp becomes a zombie.
1176 lwpusage
1178 This file contains the prusage structure for the specific lwp as
1179 described above for the process's usage file.
1180 gwindows
1182 This file exists only on SPARC based machines. If it is non-empty, it
1183 contains a gwindows_t structure, defined in <sys/regset.h>, with the
1184 values of those SPARC register windows that could not be stored on the
1185 stack when the lwp stopped. Conditions under which register windows are
1186 not stored on the stack are: the stack pointer refers to nonexistent
1187 process memory or the stack pointer is improperly aligned. If the lwp
1188 is not stopped or if there are no register windows that could not be
1189 stored on the stack, the file is empty (the usual case).
1190 xregs
1192 Extra state registers. The extra state register set is architecture
1193 dependent; this file is empty if the system does not support extra
1194 state registers. If the file is non-empty, it contains an architecture
1195 dependent structure of type prxregset_t, defined in <procfs.h>, with
1196 the values of the lwp's extra state registers. If the lwp is not
1197 stopped, all register values are undefined. See also the PCSXREG con‐
1198 trol operation, below.
1199 asrs
1201 This file exists only for 64-bit SPARC V9 processes. It contains an
1202 asrset_t structure, defined in <sys/regset.h>, containing the values of
1203 the lwp's platform-dependent ancillary state registers. If the lwp is
1204 not stopped, all register values are undefined. See also the PCSASRS
1205 control operation, below.
1206 templates
1208 A directory which contains references to the active templates for the
1209 lwp, named by the contract type. Changes made to an active template
1210 descriptor do not affect the original template which was activated,
1211 though they do affect the active template. It is not possible to acti‐
1212 vate an active template descriptor. See contract(4).
1213
1215 Process state changes are effected through messages written to a
1216 process's ctl file or to an individual lwp's lwpctl file. All control
1217 messages consist of a long that names the specific operation followed
1218 by additional data containing the operand, if any.
1219 Multiple control messages may be combined in a single write(2) (or
1222 writev(2)) to a control file, but no partial writes are permitted. That
1223 is, each control message, operation code plus operand, if any, must be
1224 presented in its entirety to the write(2) and not in pieces over sev‐
1225 eral system calls. If a control operation fails, no subsequent opera‐
1226 tions contained in the same write(2) are attempted.
1227 Descriptions of the allowable control messages follow. In all cases,
1230 writing a message to a control file for a process or lwp that has ter‐
1231 minated elicits the error ENOENT.
1232 PCSTOP PCDSTOP PCWSTOP PCTWSTOP
1234 When applied to the process control file, PCSTOP directs all lwps to
1235 stop and waits for them to stop, PCDSTOP directs all lwps to stop with‐
1236 out waiting for them to stop, and PCWSTOP simply waits for all lwps to
1237 stop. When applied to an lwp control file, PCSTOP directs the specific
1238 lwp to stop and waits until it has stopped, PCDSTOP directs the spe‐
1239 cific lwp to stop without waiting for it to stop, and PCWSTOP simply
1240 waits for the specific lwp to stop. When applied to an lwp control
1241 file, PCSTOP and PCWSTOP complete when the lwp stops on an event of
1242 interest, immediately if already so stopped; when applied to the
1243 process control file, they complete when every lwp has stopped either
1244 on an event of interest or on a PR_SUSPENDED stop.
1245 PCTWSTOP is identical to PCWSTOP except that it enables the operation
1248 to time out, to avoid waiting forever for a process or lwp that may
1249 never stop on an event of interest. PCTWSTOP takes a long operand spec‐
1250 ifying a number of milliseconds; the wait will terminate successfully
1251 after the specified number of milliseconds even if the process or lwp
1252 has not stopped; a timeout value of zero makes the operation identical
1253 to PCWSTOP.
1254 An ``event of interest'' is either a PR_REQUESTED stop or a stop that
1257 has been specified in the process's tracing flags (set by PCSTRACE,
1258 PCSFAULT, PCSENTRY, and PCSEXIT). PR_JOBCONTROL and PR_SUSPENDED stops
1259 are specifically not events of interest. (An lwp may stop twice due to
1260 a stop signal, first showing PR_SIGNALLED if the signal is traced and
1261 again showing PR_JOBCONTROL if the lwp is set running without clearing
1262 the signal.) If PCSTOP or PCDSTOP is applied to an lwp that is stopped,
1263 but not on an event of interest, the stop directive takes effect when
1264 the lwp is restarted by the competing mechanism. At that time, the lwp
1265 enters a PR_REQUESTED stop before executing any user-level code.
1266 A write of a control message that blocks is interruptible by a signal
1269 so that, for example, an alarm(2) can be set to avoid waiting forever
1270 for a process or lwp that may never stop on an event of interest. If
1271 PCSTOP is interrupted, the lwp stop directives remain in effect even
1272 though the write(2) returns an error. (Use of PCTWSTOP with a non-zero
1273 timeout is recommended over PCWSTOP with an alarm(2).)
1274 A system process (indicated by the PR_ISSYS flag) never executes at
1277 user level, has no user-level address space visible through /proc, and
1278 cannot be stopped. Applying one of these operations to a system process
1279 or any of its lwps elicits the error EBUSY.
1280 PCRUN
1282 Make an lwp runnable again after a stop. This operation takes a long
1283 operand containing zero or more of the following flags:
1284 PRCSIG clears the current signal, if any (see PCCSIG).
1286 PRCFAULT clears the current fault, if any (see PCCFAULT).
1289 PRSTEP directs the lwp to execute a single machine instruction. On
1292 completion of the instruction, a trace trap occurs. If FLT‐
1293 TRACE is being traced, the lwp stops; otherwise, it is sent
1294 SIGTRAP. If SIGTRAP is being traced and is not blocked, the
1295 lwp stops. When the lwp stops on an event of interest, the
1296 single-step directive is cancelled, even if the stop occurs
1297 before the instruction is executed. This operation requires
1298 hardware and operating system support and may not be imple‐
1299 mented on all processors. It is implemented on SPARC and
1300 x86-based machines.
1301 PRSABORT is meaningful only if the lwp is in a PR_SYSENTRY stop or
1304 is marked PR_ASLEEP; it instructs the lwp to abort execu‐
1305 tion of the system call (see PCSENTRY and PCSEXIT).
1306 PRSTOP directs the lwp to stop again as soon as possible after
1309 resuming execution (see PCDSTOP). In particular, if the lwp
1310 is stopped on PR_SIGNALLED or PR_FAULTED, the next stop
1311 will show PR_REQUESTED, no other stop will have intervened,
1312 and the lwp will not have executed any user-level code.
1313 When applied to an lwp control file, PCRUN clears any outstanding
1317 directed-stop request and makes the specific lwp runnable. The opera‐
1318 tion fails with EBUSY if the specific lwp is not stopped on an event of
1319 interest or has not been directed to stop or if the agent lwp exists
1320 and this is not the agent lwp (see PCAGENT).
1321 When applied to the process control file, a representative lwp is cho‐
1324 sen for the operation as described for /proc/pid/status. The operation
1325 fails with EBUSY if the representative lwp is not stopped on an event
1326 of interest or has not been directed to stop or if the agent lwp
1327 exists. If PRSTEP or PRSTOP was requested, the representative lwp is
1328 made runnable and its outstanding directed-stop request is cleared;
1329 otherwise all outstanding directed-stop requests are cleared and, if it
1330 was stopped on an event of interest, the representative lwp is marked
1331 PR_REQUESTED. If, as a consequence, all lwps are in the PR_REQUESTED or
1332 PR_SUSPENDED stop state, all lwps showing PR_REQUESTED are made
1333 runnable.
1334 PCSTRACE
1336 Define a set of signals to be traced in the process. The receipt of one
1337 of these signals by an lwp causes the lwp to stop. The set of signals
1338 is defined using an operand sigset_t contained in the control message.
1339 Receipt of SIGKILL cannot be traced; if specified, it is silently
1340 ignored.
1341 If a signal that is included in an lwp's held signal set (the signal
1344 mask) is sent to the lwp, the signal is not received and does not cause
1345 a stop until it is removed from the held signal set, either by the lwp
1346 itself or by setting the held signal set with PCSHOLD.
1347 PCCSIG
1349 The current signal, if any, is cleared from the specific or representa‐
1350 tive lwp.
1351 PCSSIG
1353 The current signal and its associated signal information for the spe‐
1354 cific or representative lwp are set according to the contents of the
1355 operand siginfo structure (see <sys/siginfo.h>). If the specified sig‐
1356 nal number is zero, the current signal is cleared. The semantics of
1357 this operation are different from those of kill(2) in that the signal
1358 is delivered to the lwp immediately after execution is resumed (even if
1359 it is being blocked) and an additional PR_SIGNALLED stop does not
1360 intervene even if the signal is traced. Setting the current signal to
1361 SIGKILL terminates the process immediately.
1362 PCKILL
1364 If applied to the process control file, a signal is sent to the process
1365 with semantics identical to those of kill(2). If applied to an lwp con‐
1366 trol file, a directed signal is sent to the specific lwp. The signal is
1367 named in a long operand contained in the message. Sending SIGKILL ter‐
1368 minates the process immediately.
1369 PCUNKILL
1371 A signal is deleted, that is, it is removed from the set of pending
1372 signals. If applied to the process control file, the signal is deleted
1373 from the process's pending signals. If applied to an lwp control file,
1374 the signal is deleted from the lwp's pending signals. The current sig‐
1375 nal (if any) is unaffected. The signal is named in a long operand in
1376 the control message. It is an error (EINVAL) to attempt to delete
1377 SIGKILL.
1378 PCSHOLD
1380 Set the set of held signals for the specific or representative lwp
1381 (signals whose delivery will be blocked if sent to the lwp). The set of
1382 signals is specified with a sigset_t operand. SIGKILL and SIGSTOP can‐
1383 not be held; if specified, they are silently ignored.
1384 PCSFAULT
1386 Define a set of hardware faults to be traced in the process. On incur‐
1387 ring one of these faults, an lwp stops. The set is defined via the op‐
1388 erand fltset_t structure. Fault names are defined in <sys/fault.h> and
1389 include the following. Some of these may not occur on all processors;
1390 there may be processor-specific faults in addition to these.
1391 FLTILL illegal instruction
1393 FLTPRIV privileged instruction
1396 FLTBPT breakpoint trap
1399 FLTTRACE trace trap (single-step)
1402 FLTWATCH watchpoint trap
1405 FLTACCESS memory access fault (bus error)
1408 FLTBOUNDS memory bounds violation
1411 FLTIOVF integer overflow
1414 FLTIZDIV integer zero divide
1417 FLTFPE floating-point exception
1420 FLTSTACK unrecoverable stack fault
1423 FLTPAGE recoverable page fault
1426 When not traced, a fault normally results in the posting of a signal to
1430 the lwp that incurred the fault. If an lwp stops on a fault, the signal
1431 is posted to the lwp when execution is resumed unless the fault is
1432 cleared by PCCFAULT or by the PRCFAULT option of PCRUN. FLTPAGE is an
1433 exception; no signal is posted. The pr_info field in the lwpstatus
1434 structure identifies the signal to be sent and contains machine-spe‐
1435 cific information about the fault.
1436 PCCFAULT
1438 The current fault, if any, is cleared; the associated signal will not
1439 be sent to the specific or representative lwp.
1440 PCSENTRY PCSEXIT
1442 These control operations instruct the process's lwps to stop on entry
1443 to or exit from specified system calls. The set of system calls to be
1444 traced is defined via an operand sysset_t structure.
1445 When entry to a system call is being traced, an lwp stops after having
1448 begun the call to the system but before the system call arguments have
1449 been fetched from the lwp. When exit from a system call is being
1450 traced, an lwp stops on completion of the system call just prior to
1451 checking for signals and returning to user level. At this point, all
1452 return values have been stored into the lwp's registers.
1453 If an lwp is stopped on entry to a system call (PR_SYSENTRY) or when
1456 sleeping in an interruptible system call (PR_ASLEEP is set), it may be
1457 instructed to go directly to system call exit by specifying the
1458 PRSABORT flag in a PCRUN control message. Unless exit from the system
1459 call is being traced, the lwp returns to user level showing EINTR.
1460 PCWATCH
1462 Set or clear a watched area in the controlled process from a prwatch
1463 structure operand:
1464 typedef struct prwatch {
1466 uintptr_t pr_vaddr; /* virtual address of watched area */
1467 size_t pr_size; /* size of watched area in bytes */
1468 int pr_wflags; /* watch type flags */
1469 } prwatch_t;
1470 pr_vaddr specifies the virtual address of an area of memory to be
1474 watched in the controlled process. pr_size specifies the size of the
1475 area, in bytes. pr_wflags specifies the type of memory access to be
1476 monitored as a bit-mask of the following flags:
1477 WA_READ read access
1479 WA_WRITE write access
1482 WA_EXEC execution access
1485 WA_TRAPAFTER trap after the instruction completes
1488 If pr_wflags is non-empty, a watched area is established for the vir‐
1492 tual address range specified by pr_vaddr and pr_size. If pr_wflags is
1493 empty, any previously-established watched area starting at the speci‐
1494 fied virtual address is cleared; pr_size is ignored.
1495 A watchpoint is triggered when an lwp in the traced process makes a
1498 memory reference that covers at least one byte of a watched area and
1499 the memory reference is as specified in pr_wflags. When an lwp triggers
1500 a watchpoint, it incurs a watchpoint trap. If FLTWATCH is being traced,
1501 the lwp stops; otherwise, it is sent a SIGTRAP signal; if SIGTRAP is
1502 being traced and is not blocked, the lwp stops.
1503 The watchpoint trap occurs before the instruction completes unless
1506 WA_TRAPAFTER was specified, in which case it occurs after the instruc‐
1507 tion completes. If it occurs before completion, the memory is not modi‐
1508 fied. If it occurs after completion, the memory is modified (if the
1509 access is a write access).
1510 Physical i/o is an exception for watchpoint traps. In this instance,
1513 there is no guarantee that memory before the watched area has already
1514 been modified (or in the case of WA_TRAPAFTER, that the memory follow‐
1515 ing the watched area has not been modified) when the watchpoint trap
1516 occurs and the lwp stops.
1517 pr_info in the lwpstatus structure contains information pertinent to
1520 the watchpoint trap. In particular, the si_addr field contains the vir‐
1521 tual address of the memory reference that triggered the watchpoint, and
1522 the si_code field contains one of TRAP_RWATCH, TRAP_WWATCH, or
1523 TRAP_XWATCH, indicating read, write, or execute access, respectively.
1524 The si_trapafter field is zero unless WA_TRAPAFTER is in effect for
1525 this watched area; non-zero indicates that the current instruction is
1526 not the instruction that incurred the watchpoint trap. The si_pc field
1527 contains the virtual address of the instruction that incurred the trap.
1528 A watchpoint trap may be triggered while executing a system call that
1531 makes reference to the traced process's memory. The lwp that is execut‐
1532 ing the system call incurs the watchpoint trap while still in the sys‐
1533 tem call. If it stops as a result, the lwpstatus structure contains the
1534 system call number and its arguments. If the lwp does not stop, or if
1535 it is set running again without clearing the signal or fault, the sys‐
1536 tem call fails with EFAULT. If WA_TRAPAFTER was specified, the memory
1537 reference will have completed and the memory will have been modified
1538 (if the access was a write access) when the watchpoint trap occurs.
1539 If more than one of WA_READ, WA_WRITE, and WA_EXEC is specified for a
1542 watched area, and a single instruction incurs more than one of the
1543 specified types, only one is reported when the watchpoint trap occurs.
1544 The precedence is WA_EXEC, WA_READ, WA_WRITE (WA_EXEC and WA_READ take
1545 precedence over WA_WRITE), unless WA_TRAPAFTER was specified, in which
1546 case it is WA_WRITE, WA_READ, WA_EXEC (WA_WRITE takes precedence).
1547 PCWATCH fails with EINVAL if an attempt is made to specify overlapping
1550 watched areas or if pr_wflags contains flags other than those specified
1551 above. It fails with ENOMEM if an attempt is made to establish more
1552 watched areas than the system can support (the system can support thou‐
1553 sands).
1554 The child of a vfork(2) borrows the parent's address space. When a
1557 vfork(2) is executed by a traced process, all watched areas established
1558 for the parent are suspended until the child terminates or performs an
1559 exec(2). Any watched areas established independently in the child are
1560 cancelled when the parent resumes after the child's termination or
1561 exec(2). PCWATCH fails with EBUSY if applied to the parent of a
1562 vfork(2) before the child has terminated or performed an exec(2). The
1563 PR_VFORKP flag is set in the pstatus structure for such a parent
1564 process.
1565 Certain accesses of the traced process's address space by the operating
1568 system are immune to watchpoints. The initial construction of a signal
1569 stack frame when a signal is delivered to an lwp will not trigger a
1570 watchpoint trap even if the new frame covers watched areas of the
1571 stack. Once the signal handler is entered, watchpoint traps occur nor‐
1572 mally. On SPARC based machines, register window overflow and underflow
1573 will not trigger watchpoint traps, even if the register window save
1574 areas cover watched areas of the stack.
1575 Watched areas are not inherited by child processes, even if the traced
1578 process's inherit-on-fork mode, PR_FORK, is set (see PCSET, below). All
1579 watched areas are cancelled when the traced process performs a success‐
1580 ful exec(2).
1581 PCSET PCUNSET
1583 PCSET sets one or more modes of operation for the traced process. PCUN‐
1584 SET unsets these modes. The modes to be set or unset are specified by
1585 flags in an operand long in the control message:
1586 PR_FORK (inherit-on-fork): When set, the process's tracing flags
1588 and its inherit-on-fork mode are inherited by the child of
1589 a fork(2), fork1(2), or vfork(2). When unset, child pro‐
1590 cesses start with all tracing flags cleared.
1591 PR_RLC (run-on-last-close): When set and the last writable /proc
1594 file descriptor referring to the traced process or any of
1595 its lwps is closed, all of the process's tracing flags and
1596 watched areas are cleared, any outstanding stop directives
1597 are canceled, and if any lwps are stopped on events of
1598 interest, they are set running as though PCRUN had been
1599 applied to them. When unset, the process's tracing flags
1600 and watched areas are retained and lwps are not set run‐
1601 ning on last close.
1602 PR_KLC (kill-on-last-close): When set and the last writable /proc
1605 file descriptor referring to the traced process or any of
1606 its lwps is closed, the process is terminated with
1607 SIGKILL.
1608 PR_ASYNC (asynchronous-stop): When set, a stop on an event of
1611 interest by one lwp does not directly affect any other lwp
1612 in the process. When unset and an lwp stops on an event of
1613 interest other than PR_REQUESTED, all other lwps in the
1614 process are directed to stop.
1615 PR_MSACCT (microstate accounting): Microstate accounting is now con‐
1618 tinuously enabled. This flag is deprecated and no longer
1619 has any effect upon microstate accounting. Applications
1620 may toggle this flag; however, microstate accounting will
1621 remain enabled regardless.
1622 PR_MSFORK (inherit microstate accounting): All processes now inherit
1625 microstate accounting, as it is continuously enabled. This
1626 flag has been deprecated and its use no longer has any
1627 effect upon the behavior of microstate accounting.
1628 PR_BPTADJ (breakpoint trap pc adjustment): On x86-based machines, a
1631 breakpoint trap leaves the program counter (the EIP)
1632 referring to the breakpointed instruction plus one byte.
1633 When PR_BPTADJ is set, the system will adjust the program
1634 counter back to the location of the breakpointed instruc‐
1635 tion when the lwp stops on a breakpoint. This flag has no
1636 effect on SPARC based machines, where breakpoint traps
1637 leave the program counter referring to the breakpointed
1638 instruction.
1639 PR_PTRACE (ptrace-compatibility): When set, a stop on an event of
1642 interest by the traced process is reported to the parent
1643 of the traced process by wait(3C), SIGTRAP is sent to the
1644 traced process when it executes a successful exec(2),
1645 setuid/setgid flags are not honored for execs performed by
1646 the traced process, any exec of an object file that the
1647 traced process cannot read fails, and the process dies
1648 when its parent dies. This mode is deprecated; it is pro‐
1649 vided only to allow ptrace(3C) to be implemented as a
1650 library function using /proc.
1651 It is an error (EINVAL) to specify flags other than those described
1655 above or to apply these operations to a system process. The current
1656 modes are reported in the pr_flags field of /proc/pid/status and
1657 /proc/pid/lwp/lwp/lwpstatus.
1658 PCSREG
1660 Set the general registers for the specific or representative lwp
1661 according to the operand prgregset_t structure.
1662 On SPARC based systems, only the condition-code bits of the processor-
1665 status register (R_PSR) of SPARC V8 (32-bit) processes can be modified
1666 by PCSREG. Other privileged registers cannot be modified at all.
1667 On x86-based systems, only certain bits of the flags register (EFL) can
1670 be modified by PCSREG: these include the condition codes, direction-
1671 bit, and overflow-bit.
1672 PCSREG fails with EBUSY if the lwp is not stopped on an event of inter‐
1675 est.
1676 PCSVADDR
1678 Set the address at which execution will resume for the specific or rep‐
1679 resentative lwp from the operand long. On SPARC based systems, both %pc
1680 and %npc are set, with %npc set to the instruction following the vir‐
1681 tual address. On x86-based systems, only %eip is set. PCSVADDR fails
1682 with EBUSY if the lwp is not stopped on an event of interest.
1683 PCSFPREG
1685 Set the floating-point registers for the specific or representative lwp
1686 according to the operand prfpregset_t structure. An error (EINVAL) is
1687 returned if the system does not support floating-point operations (no
1688 floating-point hardware and the system does not emulate floating-point
1689 machine instructions). PCSFPREG fails with EBUSY if the lwp is not
1690 stopped on an event of interest.
1691 PCSXREG
1693 Set the extra state registers for the specific or representative lwp
1694 according to the architecture-dependent operand prxregset_t structure.
1695 An error (EINVAL) is returned if the system does not support extra
1696 state registers. PCSXREG fails with EBUSY if the lwp is not stopped on
1697 an event of interest.
1698 PCSASRS
1700 Set the ancillary state registers for the specific or representative
1701 lwp according to the SPARC V9 platform-dependent operand asrset_t
1702 structure. An error (EINVAL) is returned if either the target process
1703 or the controlling process is not a 64-bit SPARC V9 process. Most of
1704 the ancillary state registers are privileged registers that cannot be
1705 modified. Only those that can be modified are set; all others are
1706 silently ignored. PCSASRS fails with EBUSY if the lwp is not stopped on
1707 an event of interest.
1708 PCAGENT
1710 Create an agent lwp in the controlled process with register values from
1711 the operand prgregset_t structure (see PCSREG, above). The agent lwp is
1712 created in the stopped state showing PR_REQUESTED and with its held
1713 signal set (the signal mask) having all signals except SIGKILL and
1714 SIGSTOP blocked.
1715 The PCAGENT operation fails with EBUSY unless the process is fully
1718 stopped via /proc, that is, unless all of the lwps in the process are
1719 stopped either on events of interest or on PR_SUSPENDED, or are stopped
1720 on PR_JOBCONTROL and have been directed to stop via PCDSTOP. It fails
1721 with EBUSY if an agent lwp already exists. It fails with ENOMEM if sys‐
1722 tem resources for creating new lwps have been exhausted.
1723 Any PCRUN operation applied to the process control file or to the con‐
1726 trol file of an lwp other than the agent lwp fails with EBUSY as long
1727 as the agent lwp exists. The agent lwp must be caused to terminate by
1728 executing the SYS_lwp_exit system call trap before the process can be
1729 restarted.
1730 Once the agent lwp is created, its lwp-ID can be found by reading the
1733 process status file. To facilitate opening the agent lwp's control and
1734 status files, the directory name /propc/pid/lwp/agent is accepted for
1735 lookup operations as an invisible alias for /proc/pid/lwp/lwpid, lwpid
1736 being the lwp-ID of the agent lwp (invisible in the sense that the name
1737 ``agent'' does not appear in a directory listing of /proc/pid/lwp
1738 obtained from ls(1), getdents(2), or readdir(3C)).
1739 The purpose of the agent lwp is to perform operations in the controlled
1742 process on behalf of the controlling process: to gather information not
1743 directly available via /proc files, or in general to make the process
1744 change state in ways not directly available via /proc control opera‐
1745 tions. To make use of an agent lwp, the controlling process must be
1746 capable of making it execute system calls (specifically, the
1747 SYS_lwp_exit system call trap). The register values given to the agent
1748 lwp on creation are typically the registers of the representative lwp,
1749 so that the agent lwp can use its stack.
1750 The agent lwp is not allowed to execute any variation of the SYS_fork
1753 or SYS_exec system call traps. Attempts to do so yield ENOTSUP to the
1754 agent lwp.
1755 Symbolic constants for system call trap numbers like SYS_lwp_exit and
1758 SYS_lwp_create can be found in the header file <sys/syscall.h>.
1759 PCREAD PCWRITE
1761 Read or write the target process's address space via a priovec struc‐
1762 ture operand:
1763 typedef struct priovec {
1765 void *pio_base; /* buffer in controlling process */
1766 size_t pio_len; /* size of read/write request in bytes */
1767 off_t pio_offset; /* virtual address in target process */
1768 } priovec_t;
1769 These operations have the same effect as pread(2) and pwrite(2),
1773 respectively, of the target process's address space file. The differ‐
1774 ence is that more than one PCREAD or PCWRITE control operation can be
1775 written to the control file at once, and they can be interspersed with
1776 other control operations in a single write to the control file. This is
1777 useful, for example, when planting many breakpoint instructions in the
1778 process's address space, or when stepping over a breakpointed instruc‐
1779 tion. Unlike pread(2) and pwrite(2), no provision is made for partial
1780 reads or writes; if the operation cannot be performed completely, it
1781 fails with EIO.
1782 PCNICE
1784 The traced process's nice(2) value is incremented by the amount in the
1785 operand long. Only a process with the {PRIV_PROC_PRIOCNTL} privilege
1786 asserted in its effective set can better a process's priority in this
1787 way, but any user may lower the priority. This operation is not mean‐
1788 ingful for all scheduling classes.
1789 PCSCRED
1791 Set the target process credentials to the values contained in the
1792 prcred_t structure operand (see /proc/pid/cred). The effective, real,
1793 and saved user-IDs and group-IDs of the target process are set. The
1794 target process's supplementary groups are not changed; the pr_ngroups
1795 and pr_groups members of the structure operand are ignored. Only the
1796 privileged processes can perform this operation; for all others it
1797 fails with EPERM.
1798 PCSCREDX
1800 Operates like PCSCRED but also sets the supplementary groups; the
1801 length of the data written with this control operation should be
1802 "sizeof (prcred_t) + sizeof (gid_t) * (#groups - 1)".
1803 PCSPRIV
1805 Set the target process privilege to the values contained in the
1806 prpriv_t operand (see /proc/pid/priv). The effective, permitted, inher‐
1807 itable, and limit sets are all changed. Privilege flags can also be
1808 set. The process is made privilege aware unless it can relinquish priv‐
1809 ilege awareness. See privileges(5).
1810 The limit set of the target process cannot be grown. The other privi‐
1813 lege sets must be subsets of the intersection of the effective set of
1814 the calling process with the new limit set of the target process or
1815 subsets of the original values of the sets in the target process.
1816 If any of the above restrictions are not met, EPERM is returned. If the
1819 structure written is improperly formatted, EINVAL is returned.
1820
1822 For security reasons, except for the psinfo, usage, lpsinfo, lusage,
1823 lwpsinfo, and lwpusage files, which are world-readable, and except for
1824 privileged processes, an open of a /proc file fails unless both the
1825 user-ID and group-ID of the caller match those of the traced process
1826 and the process's object file is readable by the caller. The effective
1827 set of the caller is a superset of both the inheritable and the permit‐
1828 ted set of the target process. The limit set of the caller is a super‐
1829 set of the limit set of the target process. Except for the world-read‐
1830 able files just mentioned, files corresponding to setuid and setgid
1831 processes can be opened only by the appropriately privileged process.
1832 A process that is missing the basic privilege {PRIV_PROC_INFO} cannot
1835 see any processes under /proc that it cannot send a signal to.
1836 A process that has {PRIV_PROC_OWNER} asserted in its effective set can
1839 open any file for reading. To manipulate or control a process, the con‐
1840 trolling process must have at least as many privileges in its effective
1841 set as the target process has in its effective, inheritable, and per‐
1842 mitted sets. The limit set of the controlling process must be a super‐
1843 set of the limit set of the target process. Additional restrictions
1844 apply if any of the uids of the target process are 0. See privi‐
1845 leges(5).
1846 Even if held by a privileged process, an open process or lwp file
1849 descriptor (other than file descriptors for the world-readable files)
1850 becomes invalid if the traced process performs an exec(2) of a
1851 setuid/setgid object file or an object file that the traced process
1852 cannot read. Any operation performed on an invalid file descriptor,
1853 except close(2), fails with EAGAIN. In this situation, if any tracing
1854 flags are set and the process or any lwp file descriptor is open for
1855 writing, the process will have been directed to stop and its run-on-
1856 last-close flag will have been set (see PCSET). This enables a control‐
1857 ling process (if it has permission) to reopen the /proc files to get
1858 new valid file descriptors, close the invalid file descriptors, unset
1859 the run-on-last-close flag (if desired), and proceed. Just closing the
1860 invalid file descriptors causes the traced process to resume execution
1861 with all tracing flags cleared. Any process not currently open for
1862 writing via /proc, but that has left-over tracing flags from a previous
1863 open, and that executes a setuid/setgid or unreadable object file, will
1864 not be stopped but will have all its tracing flags cleared.
1865 To wait for one or more of a set of processes or lwps to stop or termi‐
1868 nate, /proc file descriptors (other than those obtained by opening the
1869 cwd or root directories or by opening files in the fd or object direc‐
1870 tories) can be used in a poll(2) system call. When requested and
1871 returned, either of the polling events POLLPRI or POLLWRNORM indicates
1872 that the process or lwp stopped on an event of interest. Although they
1873 cannot be requested, the polling events POLLHUP, POLLERR, and POLLNVAL
1874 may be returned. POLLHUP indicates that the process or lwp has termi‐
1875 nated. POLLERR indicates that the file descriptor has become invalid.
1876 POLLNVAL is returned immediately if POLLPRI or POLLWRNORM is requested
1877 on a file descriptor referring to a system process (see PCSTOP). The
1878 requested events may be empty to wait simply for termination.
1879
1881 /proc
1882 directory (list of processes)
1884 /proc/pid
1887 specific process directory
1889 /proc/self
1892 alias for a process's own directory
1894 /proc/pid/as
1897 address space file
1899 /proc/pid/ctl
1902 process control file
1904 /proc/pid/status
1907 process status
1909 /proc/pid/lstatus
1912 array of lwp status structs
1914 /proc/pid/psinfo
1917 process ps(1) info
1919 /proc/pid/lpsinfo
1922 array of lwp ps(1) info structs
1924 /proc/pid/map
1927 address space map
1929 /proc/pid/xmap
1932 extended address space map
1934 /proc/pid/rmap
1937 reserved address map
1939 /proc/pid/cred
1942 process credentials
1944 /proc/pid/priv
1947 process privileges
1949 /proc/pid/sigact
1952 process signal actions
1954 /proc/pid/auxv
1957 process aux vector
1959 /proc/pid/ldt
1962 process LDT (x86 only)
1964 /proc/pid/usage
1967 process usage
1969 /proc/pid/lusage
1972 array of lwp usage structs
1974 /proc/pid/path
1977 symbolic links to process open files
1979 /proc/pid/pagedata
1982 process page data
1984 /proc/pid/watch
1987 active watchpoints
1989 /proc/pid/cwd
1992 alias for the current working directory
1994 /proc/pid/root
1997 alias for the root directory
1999 /proc/pid/fd
2002 directory (list of open files)
2004 /proc/pid/fd/*
2007 aliases for process's open files
2009 /proc/pid/object
2012 directory (list of mapped files)
2014 /proc/pid/object/a.out
2017 alias for process's executable file
2019 /proc/pid/object/*
2022 aliases for other mapped files
2024 /proc/pid/lwp
2027 directory (list of lwps)
2029 /proc/pid/lwp/lwpid
2032 specific lwp directory
2034 /proc/pid/lwp/agent
2037 alias for the agent lwp directory
2039 /proc/pid/lwp/lwpid/lwpctl
2042 lwp control file
2044 /proc/pid/lwp/lwpid/lwpstatus
2047 lwp status
2049 /proc/pid/lwp/lwpid/lwpsinfo
2052 lwp ps(1) info
2054 /proc/pid/lwp/lwpid/lwpusage
2057 lwp usage
2059 /proc/pid/lwp/lwpid/gwindows
2062 register windows (SPARC only)
2064 /proc/pid/lwp/lwpid/xregs
2067 extra state registers
2069 /proc/pid/lwp/lwpid/asrs
2072 ancillary state registers (SPARC V9 only)
2074
2077 ls(1), ps(1), chroot(1M), alarm(2), brk(2), chdir(2), chroot(2),
2078 close(2), creat(2), dup(2), exec(2), fcntl(2), fork(2), fork1(2),
2079 fstat(2), getdents(2), getustack(2), kill(2), lseek(2), mmap(2),
2080 nice(2), open(2), poll(2), pread(2), ptrace(3C), pwrite(2), read(2),
2081 readlink(2), readv(2), shmget(2), sigaction(2), sigaltstack(2),
2082 vfork(2), write(2), writev(2), _stack_grow(3C), readdir(3C),
2083 pthread_create(3C), pthread_join(3C), siginfo.h(3HEAD), sig‐
2084 nal.h(3HEAD), thr_create(3C), thr_join(3C), types32.h(3HEAD), ucon‐
2085 text.h(3HEAD), wait(3C), contract(4), process(4), lfcompile(5), privi‐
2086 leges(5)
2087
2089 Errors that can occur in addition to the errors normally associated
2090 with file system access:
2091 E2BIG Data to be returned in a read(2) of the page data file
2093 exceeds the size of the read buffer provided by the call‐
2094 er.
2095 EACCES An attempt was made to examine a process that ran under a
2098 different uid than the controlling process and
2099 {PRIV_PROC_OWNER} was not asserted in the effective set.
2100 EAGAIN The traced process has performed an exec(2) of a
2103 setuid/setgid object file or of an object file that it
2104 cannot read; all further operations on the process or lwp
2105 file descriptor (except close(2)) elicit this error.
2106 EBUSY PCSTOP, PCDSTOP, PCWSTOP, or PCTWSTOP was applied to a
2109 system process; an exclusive open(2) was attempted on a
2110 /proc file for a process already open for writing; PCRUN,
2111 PCSREG, PCSVADDR, PCSFPREG, or PCSXREG was applied to a
2112 process or lwp not stopped on an event of interest; an
2113 attempt was made to mount /proc when it was already
2114 mounted; PCAGENT was applied to a process that was not
2115 fully stopped or that already had an agent lwp.
2116 EINVAL In general, this means that some invalid argument was sup‐
2119 plied to a system call. A non-exhaustive list of condi‐
2120 tions eliciting this error includes: a control message
2121 operation code is undefined; an out-of-range signal number
2122 was specified with PCSSIG, PCKILL, or PCUNKILL; SIGKILL
2123 was specified with PCUNKILL; PCSFPREG was applied on a
2124 system that does not support floating-point operations;
2125 PCSXREG was applied on a system that does not support
2126 extra state registers.
2127 EINTR A signal was received by the controlling process while
2130 waiting for the traced process or lwp to stop via PCSTOP,
2131 PCWSTOP, or PCTWSTOP.
2132 EIO A write(2) was attempted at an illegal address in the
2135 traced process.
2136 ENOENT The traced process or lwp has terminated after being
2139 opened. The basic privilege {PRIV_PROC_INFO} is not
2140 asserted in the effective set of the calling process and
2141 the calling process cannot send a signal to the target
2142 process.
2143 ENOMEM The system-imposed limit on the number of page data file
2146 descriptors was reached on an open of /proc/pid/pagedata;
2147 an attempt was made with PCWATCH to establish more watched
2148 areas than the system can support; the PCAGENT operation
2149 was issued when the system was out of resources for creat‐
2150 ing lwps.
2151 ENOSYS An attempt was made to perform an unsupported operation
2154 (such as creat(2), link(2), or unlink(2)) on an entry in
2155 /proc.
2156 EOVERFLOW A 32-bit controlling process attempted to read or write
2159 the as file or attempted to read the map, rmap, or page‐
2160 data file of a 64-bit target process. A 32-bit controlling
2161 process attempted to apply one of the control operations
2162 PCSREG, PCSXREG, PCSVADDR, PCWATCH, PCAGENT, PCREAD,
2163 PCWRITE to a 64-bit target process.
2164 EPERM The process that issued the PCSCRED or PCSCREDX operation
2167 did not have the {PRIV_PROC_SETID} privilege asserted in
2168 its effective set, or the process that issued the PCNICE
2169 operation did not have the {PRIV_PROC_PRIOCNTL} in its
2170 effective set.
2171 An attempt was made to control a process of which the E,
2173 P, and I privilege sets were not a subset of the effective
2174 set of the controlling process or the limit set of the
2175 controlling process is not a superset of limit set of the
2176 controlled process.
2177 Any of the uids of the target process are 0 or an attempt
2179 was made to change any of the uids to 0 using PCSCRED and
2180 the security policy imposed additional restrictions. See
2181 privileges(5).
2182
2185 Descriptions of structures in this document include only interesting
2186 structure elements, not filler and padding fields, and may show ele‐
2187 ments out of order for descriptive clarity. The actual structure defi‐
2188 nitions are contained in <procfs.h>.
2189
2191 Because the old ioctl(2)-based version of /proc is currently supported
2192 for binary compatibility with old applications, the top-level directory
2193 for a process, /proc/pid, is not world-readable, but it is world-
2194 searchable. Thus, anyone can open /proc/pid/psinfo even though ls(1)
2195 applied to /proc/pid will fail for anyone but the owner or an appropri‐
2196 ately privileged process. Support for the old ioctl(2)-based version of
2197 /proc will be dropped in a future release, at which time the top-level
2198 directory for a process will be made world-readable.
2199 On SPARC based machines, the types gregset_t and fpregset_t defined in
2202 <sys/regset.h> are similar to but not the same as the types prgregset_t
2203 and prfpregset_t defined in <procfs.h>.
2204SunOS 5.11 29 Nov 2006 proc(4)