1 .\" Copyright (C) 1998 Andries Brouwer (aeb@cwi.nl)
2 .\" and Copyright (C) 2002, 2006, 2008, 2012, 2013 Michael Kerrisk <mtk.manpages@gmail.com>
3 .\" and Copyright Guillem Jover <guillem@hadrons.org>
4 .\" and Copyright (C) 2014 Dave Hansen / Intel
6 .\" %%%LICENSE_START(VERBATIM)
7 .\" Permission is granted to make and distribute verbatim copies of this
8 .\" manual provided the copyright notice and this permission notice are
9 .\" preserved on all copies.
11 .\" Permission is granted to copy and distribute modified versions of this
12 .\" manual under the conditions for verbatim copying, provided that the
13 .\" entire resulting derived work is distributed under the terms of a
14 .\" permission notice identical to this one.
16 .\" Since the Linux kernel and libraries are constantly changing, this
17 .\" manual page may be incorrect or out-of-date. The author(s) assume no
18 .\" responsibility for errors or omissions, or for damages resulting from
19 .\" the use of the information contained herein. The author(s) may not
20 .\" have taken the same level of care in the production of this manual,
21 .\" which is licensed free of charge, as they might when working
24 .\" Formatted or processed versions of this manual, if unaccompanied by
25 .\" the source, must acknowledge the copyright and authors of this work.
28 .\" Modified Thu Nov 11 04:19:42 MET 1999, aeb: added PR_GET_PDEATHSIG
29 .\" Modified 27 Jun 02, Michael Kerrisk
30 .\" Added PR_SET_DUMPABLE, PR_GET_DUMPABLE,
31 .\" PR_SET_KEEPCAPS, PR_GET_KEEPCAPS
32 .\" Modified 2006-08-30 Guillem Jover <guillem@hadrons.org>
33 .\" Updated Linux versions where the options where introduced.
34 .\" Added PR_SET_TIMING, PR_GET_TIMING, PR_SET_NAME, PR_GET_NAME,
35 .\" PR_SET_UNALIGN, PR_GET_UNALIGN, PR_SET_FPEMU, PR_GET_FPEMU,
36 .\" PR_SET_FPEXC, PR_GET_FPEXC
37 .\" 2008-04-29 Serge Hallyn, Document PR_CAPBSET_READ and PR_CAPBSET_DROP
38 .\" 2008-06-13 Erik Bosman, <ejbosman@cs.vu.nl>
39 .\" Document PR_GET_TSC and PR_SET_TSC.
40 .\" 2008-06-15 mtk, Document PR_SET_SECCOMP, PR_GET_SECCOMP
41 .\" 2009-10-03 Andi Kleen, document PR_MCE_KILL
42 .\" 2012-04 Cyrill Gorcunov, Document PR_SET_MM
43 .\" 2012-04-25 Michael Kerrisk, Document PR_TASK_PERF_EVENTS_DISABLE and
44 .\" PR_TASK_PERF_EVENTS_ENABLE
45 .\" 2012-09-20 Kees Cook, update PR_SET_SECCOMP for mode 2
46 .\" 2012-09-20 Kees Cook, document PR_SET_NO_NEW_PRIVS, PR_GET_NO_NEW_PRIVS
47 .\" 2012-10-25 Michael Kerrisk, Document PR_SET_TIMERSLACK and
49 .\" 2013-01-10 Kees Cook, document PR_SET_PTRACER
50 .\" 2012-02-04 Michael Kerrisk, document PR_{SET,GET}_CHILD_SUBREAPER
51 .\" 2014-11-10 Dave Hansen, document PR_MPX_{EN,DIS}ABLE_MANAGEMENT
54 .TH PRCTL 2 2019-03-06 "Linux" "Linux Programmer's Manual"
56 prctl \- operations on a process
59 .B #include <sys/prctl.h>
61 .BI "int prctl(int " option ", unsigned long " arg2 ", unsigned long " arg3 ,
62 .BI " unsigned long " arg4 ", unsigned long " arg5 );
66 is called with a first argument describing what to do
67 (with values defined in \fI<linux/prctl.h>\fP), and further
68 arguments with a significance depending on the first one.
69 The first argument can be:
72 .BR PR_CAP_AMBIENT " (since Linux 4.3)"
73 .\" commit 58319057b7847667f0c9585b9de0e8932b0fdb08
74 Reads or changes the ambient capability set of the calling thread,
75 according to the value of
77 which must be one of the following:
81 .B PR_CAP_AMBIENT_RAISE
82 The capability specified in
84 is added to the ambient set.
85 The specified capability must already be present in
86 both the permitted and the inheritable sets of the process.
87 This operation is not permitted if the
88 .B SECBIT_NO_CAP_AMBIENT_RAISE
91 .B PR_CAP_AMBIENT_LOWER
92 The capability specified in
94 is removed from the ambient set.
96 .B PR_CAP_AMBIENT_IS_SET
99 call returns 1 if the capability in
101 is in the ambient set and 0 if it is not.
103 .BR PR_CAP_AMBIENT_CLEAR_ALL
104 All capabilities will be removed from the ambient set.
105 This operation requires setting
110 In all of the above operations,
114 must be specified as 0.
116 Higher-level interfaces layered on top of the above operations are
119 library in the form of
120 .BR cap_get_ambient (3),
121 .BR cap_set_ambient (3),
123 .BR cap_reset_ambient (3).
125 .BR PR_CAPBSET_READ " (since Linux 2.6.25)"
126 Return (as the function result) 1 if the capability specified in
128 is in the calling thread's capability bounding set,
130 (The capability constants are defined in
131 .IR <linux/capability.h> .)
132 The capability bounding set dictates
133 whether the process can receive the capability through a
134 file's permitted capability set on a subsequent call to
137 If the capability specified in
139 is not valid, then the call fails with the error
142 A higher-level interface layered on top of this operation is provided in the
144 library in the form of
145 .BR cap_get_bound (3).
147 .BR PR_CAPBSET_DROP " (since Linux 2.6.25)"
148 If the calling thread has the
150 capability within its user namespace, then drop the capability specified by
152 from the calling thread's capability bounding set.
153 Any children of the calling thread will inherit the newly
154 reduced bounding set.
156 The call fails with the error:
158 if the calling thread does not have the
163 does not represent a valid capability; or
165 if file capabilities are not enabled in the kernel,
166 in which case bounding sets are not supported.
168 A higher-level interface layered on top of this operation is provided in the
170 library in the form of
171 .BR cap_drop_bound (3).
173 .BR PR_SET_CHILD_SUBREAPER " (since Linux 3.4)"
174 .\" commit ebec18a6d3aa1e7d84aab16225e87fd25170ec2b
178 set the "child subreaper" attribute of the calling process;
181 is zero, unset the attribute.
183 A subreaper fulfills the role of
185 for its descendant processes.
186 When a process becomes orphaned
187 (i.e., its immediate parent terminates)
188 then that process will be reparented to
189 the nearest still living ancestor subreaper.
190 Subsequently, calls to
192 in the orphaned process will now return the PID of the subreaper process,
193 and when the orphan terminates, it is the subreaper process that
196 signal and will be able to
198 on the process to discover its termination status.
200 The setting of the "child subreaper" attribute
201 is not inherited by children created by
205 The setting is preserved across
208 Establishing a subreaper process is useful in session management frameworks
209 where a hierarchical group of processes is managed by a subreaper process
210 that needs to be informed when one of the processes\(emfor example,
211 a double-forked daemon\(emterminates
212 (perhaps so that it can restart that process).
217 employ a subreaper process for similar reasons.
219 .BR PR_GET_CHILD_SUBREAPER " (since Linux 3.4)"
220 Return the "child subreaper" setting of the caller,
221 in the location pointed to by
222 .IR "(int\ *) arg2" .
224 .BR PR_SET_DUMPABLE " (since Linux 2.3.20)"
225 Set the state of the "dumpable" flag,
226 which determines whether core dumps are produced for the calling process
227 upon delivery of a signal whose default behavior is to produce a core dump.
229 In kernels up to and including 2.6.12,
232 .RB ( SUID_DUMP_DISABLE ,
233 process is not dumpable) or 1
234 .RB ( SUID_DUMP_USER ,
235 process is dumpable).
236 Between kernels 2.6.13 and 2.6.17,
237 .\" commit abf75a5033d4da7b8a7e92321d74021d1fcfb502
238 the value 2 was also permitted,
239 which caused any binary which normally would not be dumped
240 to be dumped readable by root only;
241 for security reasons, this feature has been removed.
242 .\" See http://marc.theaimsgroup.com/?l=linux-kernel&m=115270289030630&w=2
243 .\" Subject: Fix prctl privilege escalation (CVE-2006-2451)
244 .\" From: Marcel Holtmann <marcel () holtmann ! org>
245 .\" Date: 2006-07-12 11:12:00
246 (See also the description of
247 .I /proc/sys/fs/\:suid_dumpable
251 Normally, this flag is set to 1.
252 However, it is reset to the current value contained in the file
253 .IR /proc/sys/fs/\:suid_dumpable
254 (which by default has the value 0),
255 in the following circumstances:
256 .\" See kernel/cred.c::commit_creds() (Linux 3.18 sources)
259 The process's effective user or group ID is changed.
261 The process's filesystem user or group ID is changed (see
262 .BR credentials (7)).
266 a set-user-ID or set-group-ID program, resulting in a change
267 of either the effective user ID or the effective group ID.
271 a program that has file capabilities (see
272 .BR capabilities (7)),
273 .\" See kernel/cred.c::commit_creds()
274 but only if the permitted capabilities
275 gained exceed those already permitted for the process.
276 .\" Also certain namespace operations;
279 Processes that are not dumpable can not be attached via
286 If a process is not dumpable,
287 the ownership of files in the process's
289 directory is affected as described in
292 .BR PR_GET_DUMPABLE " (since Linux 2.3.20)"
293 Return (as the function result) the current state of the calling
294 process's dumpable flag.
295 .\" Since Linux 2.6.13, the dumpable flag can have the value 2,
296 .\" but in 2.6.13 PR_GET_DUMPABLE simply returns 1 if the dumpable
297 .\" flags has a nonzero value. This was fixed in 2.6.14.
299 .BR PR_SET_ENDIAN " (since Linux 2.6.18, PowerPC only)"
300 Set the endian-ness of the calling process to the value given
301 in \fIarg2\fP, which should be one of the following:
302 .\" Respectively 0, 1, 2
304 .BR PR_ENDIAN_LITTLE ,
306 .B PR_ENDIAN_PPC_LITTLE
307 (PowerPC pseudo little endian).
309 .BR PR_GET_ENDIAN " (since Linux 2.6.18, PowerPC only)"
310 Return the endian-ness of the calling process,
311 in the location pointed to by
312 .IR "(int\ *) arg2" .
314 .BR PR_SET_FP_MODE " (since Linux 4.0, only on MIPS)"
315 .\" commit 9791554b45a2acc28247f66a5fd5bbc212a6b8c8
316 On the MIPS architecture,
317 user-space code can be built using an ABI which permits linking
318 with code that has more restrictive floating-point (FP) requirements.
319 For example, user-space code may be built to target the O32 FPXX ABI
320 and linked with code built for either one of the more restrictive
322 When more restrictive code is linked in,
323 the overall requirement for the process is to use the more
324 restrictive floating-point mode.
326 Because the kernel has no means of knowing in advance
327 which mode the process should be executed in,
328 and because these restrictions can
329 change over the lifetime of the process, the
331 operation is provided to allow control of the floating-point mode
334 .\" https://dmz-portal.mips.com/wiki/MIPS_O32_ABI_-_FR0_and_FR1_Interlinking
336 .I (unsigned int) arg2
337 argument is a bit mask describing the floating-point mode used:
345 mode), the 32 floating-point registers are 32 bits wide,
346 and 64-bit registers are represented as a pair of registers
347 (even- and odd- numbered,
348 with the even-numbered register containing the lower 32 bits,
349 and the odd-numbered register containing the higher 32 bits).
353 (on supported hardware),
354 the 32 floating-point registers are 64 bits wide (so called
357 Note that modern MIPS implementations (MIPS R6 and newer) support
362 Applications that use the O32 FP32 ABI can operate only when this bit is
365 or they can be used with FRE enabled, see below).
366 Applications that use the O32 FP64 ABI
367 (and the O32 FP64A ABI, which exists to
368 provide the ability to operate with existing FP32 code; see below)
369 can operate only when this bit is
372 Applications that use the O32 FPXX ABI can operate with either
378 Enable emulation of 32-bit floating-point mode.
379 When this mode is enabled,
380 it emulates 32-bit floating-point operations
381 by raising a reserved-instruction exception
382 on every instruction that uses 32-bit formats and
383 the kernel then handles the instruction in software.
384 (The problem lies in the discrepancy of handling odd-numbered registers
385 which are the high 32 bits of 64-bit registers with even numbers in
387 mode and the lower 32-bit parts of odd-numbered 64-bit registers in
390 Enabling this bit is necessary when code with the O32 FP32 ABI should operate
391 with code with compatible the O32 FPXX or O32 FP64A ABIs (which require
393 FPU mode) or when it is executed on newer hardware (MIPS R6 onwards)
396 mode support when a binary with the FP32 ABI is used.
398 Note that this mode makes sense only when the FPU is in 64-bit mode
401 Note that the use of emulation inherently has a significant performance hit
402 and should be avoided if possible.
405 In the N32/N64 ABI, 64-bit floating-point mode is always used,
406 so FPU emulation is not required and the FPU always operates in
410 This option is mainly intended for use by the dynamic linker
420 .BR PR_GET_FP_MODE " (since Linux 4.0, only on MIPS)"
421 Get the current floating-point mode (see the description of
426 the call returns a bit mask which represents the current floating-point mode.
436 .BR PR_SET_FPEMU " (since Linux 2.4.18, 2.5.9, only on ia64)"
437 Set floating-point emulation control bits to \fIarg2\fP.
440 to silently emulate floating-point operation accesses, or
442 to not emulate floating-point operations and send
446 .BR PR_GET_FPEMU " (since Linux 2.4.18, 2.5.9, only on ia64)"
447 Return floating-point emulation control bits,
448 in the location pointed to by
449 .IR "(int\ *) arg2" .
451 .BR PR_SET_FPEXC " (since Linux 2.4.21, 2.5.32, only on PowerPC)"
452 Set floating-point exception mode to \fIarg2\fP.
453 Pass \fBPR_FP_EXC_SW_ENABLE\fP to use FPEXC for FP exception enables,
454 \fBPR_FP_EXC_DIV\fP for floating-point divide by zero,
455 \fBPR_FP_EXC_OVF\fP for floating-point overflow,
456 \fBPR_FP_EXC_UND\fP for floating-point underflow,
457 \fBPR_FP_EXC_RES\fP for floating-point inexact result,
458 \fBPR_FP_EXC_INV\fP for floating-point invalid operation,
459 \fBPR_FP_EXC_DISABLED\fP for FP exceptions disabled,
460 \fBPR_FP_EXC_NONRECOV\fP for async nonrecoverable exception mode,
461 \fBPR_FP_EXC_ASYNC\fP for async recoverable exception mode,
462 \fBPR_FP_EXC_PRECISE\fP for precise exception mode.
464 .BR PR_GET_FPEXC " (since Linux 2.4.21, 2.5.32, only on PowerPC)"
465 Return floating-point exception mode,
466 in the location pointed to by
467 .IR "(int\ *) arg2" .
469 .BR PR_SET_KEEPCAPS " (since Linux 2.2.18)"
470 Set the state of the calling thread's "keep capabilities" flag.
471 The effect if this flag is described in
472 .BR capabilities (7).
474 must be either 0 (clear the flag)
476 The "keep capabilities" value will be reset to 0 on subsequent calls to
479 .BR PR_GET_KEEPCAPS " (since Linux 2.2.18)"
480 Return (as the function result) the current state of the calling thread's
481 "keep capabilities" flag.
484 for a description of this flag.
486 .BR PR_MCE_KILL " (since Linux 2.6.32)"
487 Set the machine check memory corruption kill policy for the calling thread.
491 .BR PR_MCE_KILL_CLEAR ,
492 clear the thread memory corruption kill policy and use the system-wide default.
493 (The system-wide default is defined by
494 .IR /proc/sys/vm/memory_failure_early_kill ;
500 .BR PR_MCE_KILL_SET ,
501 use a thread-specific memory corruption kill policy.
504 defines whether the policy is
506 .RB ( PR_MCE_KILL_EARLY ),
508 .RB ( PR_MCE_KILL_LATE ),
509 or the system-wide default
510 .RB ( PR_MCE_KILL_DEFAULT ).
511 Early kill means that the thread receives a
513 signal as soon as hardware memory corruption is detected inside
515 In late kill mode, the process is killed only when it accesses a corrupted page.
518 for more information on the
521 The policy is inherited by children.
524 arguments must be zero for future compatibility.
526 .BR PR_MCE_KILL_GET " (since Linux 2.6.32)"
527 Return the current per-process machine check kill policy.
530 arguments must be zero.
532 .BR PR_SET_MM " (since Linux 3.3)"
533 .\" commit 028ee4be34a09a6d48bdf30ab991ae933a7bc036
534 Modify certain kernel memory map descriptor fields
535 of the calling process.
536 Usually these fields are set by the kernel and dynamic loader (see
538 for more information) and a regular application should not use this feature.
539 However, there are cases, such as self-modifying programs,
540 where a program might find it useful to change its own memory map.
542 The calling process must have the
547 is one of the options below, while
549 provides a new value for the option.
554 arguments must be zero if unused.
557 .\" commit 52b3694157e3aa6df871e283115652ec6f2d31e0
558 this feature is available only if the kernel is built with the
559 .BR CONFIG_CHECKPOINT_RESTORE
563 .BR PR_SET_MM_START_CODE
564 Set the address above which the program text can run.
565 The corresponding memory area must be readable and executable,
566 but not writable or shareable (see
570 for more information).
572 .BR PR_SET_MM_END_CODE
573 Set the address below which the program text can run.
574 The corresponding memory area must be readable and executable,
575 but not writable or shareable.
577 .BR PR_SET_MM_START_DATA
578 Set the address above which initialized and
579 uninitialized (bss) data are placed.
580 The corresponding memory area must be readable and writable,
581 but not executable or shareable.
583 .B PR_SET_MM_END_DATA
584 Set the address below which initialized and
585 uninitialized (bss) data are placed.
586 The corresponding memory area must be readable and writable,
587 but not executable or shareable.
589 .BR PR_SET_MM_START_STACK
590 Set the start address of the stack.
591 The corresponding memory area must be readable and writable.
593 .BR PR_SET_MM_START_BRK
594 Set the address above which the program heap can be expanded with
597 The address must be greater than the ending address of
598 the current program data segment.
599 In addition, the combined size of the resulting heap and
600 the size of the data segment can't exceed the
609 The requirements for the address are the same as for the
610 .BR PR_SET_MM_START_BRK
613 The following options are available since Linux 3.5.
614 .\" commit fe8c7f5cbf91124987106faa3bdf0c8b955c4cf7
616 .BR PR_SET_MM_ARG_START
617 Set the address above which the program command line is placed.
619 .BR PR_SET_MM_ARG_END
620 Set the address below which the program command line is placed.
622 .BR PR_SET_MM_ENV_START
623 Set the address above which the program environment is placed.
625 .BR PR_SET_MM_ENV_END
626 Set the address below which the program environment is placed.
628 The address passed with
629 .BR PR_SET_MM_ARG_START ,
630 .BR PR_SET_MM_ARG_END ,
631 .BR PR_SET_MM_ENV_START ,
633 .BR PR_SET_MM_ENV_END
634 should belong to a process stack area.
635 Thus, the corresponding memory area must be readable, writable, and
636 (depending on the kernel configuration) have the
642 Set a new auxiliary vector.
645 argument should provide the address of the vector.
648 is the size of the vector.
650 .BR PR_SET_MM_EXE_FILE
651 .\" commit b32dfe377102ce668775f8b6b1461f7ad428f8b6
654 symbolic link with a new one pointing to a new executable file
655 identified by the file descriptor provided in
658 The file descriptor should be obtained with a regular
662 To change the symbolic link, one needs to unmap all existing
663 executable memory areas, including those created by the kernel itself
664 (for example the kernel usually creates at least one executable
665 memory area for the ELF
669 In Linux 4.9 and earlier, the
670 .\" commit 3fb4afd9a504c2386b8435028d43283216bf588e
671 .BR PR_SET_MM_EXE_FILE
672 operation can be performed only once in a process's lifetime;
673 attempting to perform the operation a second time results in the error
675 This restriction was enforced for security reasons that were subsequently
677 and the restriction was removed in Linux 4.10 because some
678 user-space applications needed to perform this operation more than once.
680 The following options are available since Linux 3.18.
681 .\" commit f606b77f1a9e362451aca8f81d8f36a3a112139e
684 Provides one-shot access to all the addresses by passing in a
685 .I struct prctl_mm_map
686 (as defined in \fI<linux/prctl.h>\fP).
689 argument should provide the size of the struct.
691 This feature is available only if the kernel is built with the
692 .BR CONFIG_CHECKPOINT_RESTORE
695 .BR PR_SET_MM_MAP_SIZE
696 Returns the size of the
697 .I struct prctl_mm_map
699 This allows user space to find a compatible struct.
702 argument should be a pointer to an unsigned int.
704 This feature is available only if the kernel is built with the
705 .BR CONFIG_CHECKPOINT_RESTORE
709 .BR PR_MPX_ENABLE_MANAGEMENT ", " PR_MPX_DISABLE_MANAGEMENT " (since Linux 3.19) "
710 .\" commit fe3d197f84319d3bce379a9c0dc17b1f48ad358c
711 .\" See also http://lwn.net/Articles/582712/
712 .\" See also https://gcc.gnu.org/wiki/Intel%20MPX%20support%20in%20the%20GCC%20compiler
713 Enable or disable kernel management of Memory Protection eXtensions (MPX)
721 .\" commit e9d1b4f3c60997fe197bf0243cb4a41a44387a88
722 arguments must be zero.
724 MPX is a hardware-assisted mechanism for performing bounds checking on
726 It consists of a set of registers storing bounds information
727 and a set of special instruction prefixes that tell the CPU on which
728 instructions it should do bounds enforcement.
729 There is a limited number of these registers and
730 when there are more pointers than registers,
731 their contents must be "spilled" into a set of tables.
732 These tables are called "bounds tables" and the MPX
735 whether the kernel manages their allocation and freeing.
737 When management is enabled, the kernel will take over allocation
738 and freeing of the bounds tables.
739 It does this by trapping the #BR exceptions that result
740 at first use of missing bounds tables and
741 instead of delivering the exception to user space,
742 it allocates the table and populates the bounds directory
743 with the location of the new table.
744 For freeing, the kernel checks to see if bounds tables are
745 present for memory which is not allocated, and frees them if so.
747 Before enabling MPX management using
748 .BR PR_MPX_ENABLE_MANAGEMENT ,
749 the application must first have allocated a user-space buffer for
750 the bounds directory and placed the location of that directory in the
754 These calls fail if the CPU or kernel does not support MPX.
755 Kernel support for MPX is enabled via the
756 .BR CONFIG_X86_INTEL_MPX
757 configuration option.
758 You can check whether the CPU supports MPX by looking for the 'mpx'
759 CPUID bit, like with the following command:
763 cat /proc/cpuinfo | grep ' mpx '
767 A thread may not switch in or out of long (64-bit) mode while MPX is
770 All threads in a process are affected by these calls.
774 inherits the state of MPX management.
777 MPX management is reset to a state as if
778 .BR PR_MPX_DISABLE_MANAGEMENT
781 For further information on Intel MPX, see the kernel source file
782 .IR Documentation/x86/intel_mpx.txt .
784 .BR PR_SET_NAME " (since Linux 2.6.9)"
785 Set the name of the calling thread,
786 using the value in the location pointed to by
787 .IR "(char\ *) arg2" .
788 The name can be up to 16 bytes long,
789 .\" TASK_COMM_LEN in include/linux/sched.h
790 including the terminating null byte.
791 (If the length of the string, including the terminating null byte,
792 exceeds 16 bytes, the string is silently truncated.)
793 This is the same attribute that can be set via
794 .BR pthread_setname_np (3)
796 .BR pthread_getname_np (3).
797 The attribute is likewise accessible via
798 .IR /proc/self/task/[tid]/comm ,
801 is the name of the calling thread.
803 .BR PR_GET_NAME " (since Linux 2.6.11)"
804 Return the name of the calling thread,
805 in the buffer pointed to by
806 .IR "(char\ *) arg2" .
807 The buffer should allow space for up to 16 bytes;
808 the returned string will be null-terminated.
810 .BR PR_SET_NO_NEW_PRIVS " (since Linux 3.5)"
811 Set the calling thread's
813 attribute to the value in
819 promises not to grant privileges to do anything
820 that could not have been done without the
823 rendering the set-user-ID and set-group-ID mode bits,
824 and file capabilities non-functional).
827 attribute cannot be unset.
828 The setting of this attribute is inherited by children created by
836 the value of a thread's
838 attribute can be viewed via the
841 .IR /proc/[pid]/status
844 For more information, see the kernel source file
845 .IR Documentation/userspace\-api/no_new_privs.rst
846 .\" commit 40fde647ccb0ae8c11d256d271e24d385eed595b
848 .IR Documentation/prctl/no_new_privs.txt
853 .BR PR_GET_NO_NEW_PRIVS " (since Linux 3.5)"
854 Return (as the function result) the value of the
856 attribute for the calling thread.
857 A value of 0 indicates the regular
860 A value of 1 indicates
862 will operate in the privilege-restricting mode described above.
864 .BR PR_SET_PDEATHSIG " (since Linux 2.1.57)"
865 Set the parent-death signal
866 of the calling process to \fIarg2\fP (either a signal value
867 in the range 1..maxsig, or 0 to clear).
868 This is the signal that the calling process will get when its
872 .\" https://bugzilla.kernel.org/show_bug.cgi?id=43300
873 the "parent" in this case is considered to be the
875 that created this process.
876 In other words, the signal will be sent when that thread terminates
878 .BR pthread_exit (3)),
879 rather than after all of the threads in the parent process terminate.
881 The parent-death signal is sent upon subsequent termination of the parent
882 thread and also upon termination of each subreaper process
883 (see the description of
884 .B PR_SET_CHILD_SUBREAPER
885 above) to which the caller is subsequently reparented.
886 If the parent thread and all ancestor subreapers have already terminated
889 operation, then no parent-death signal is sent to the caller.
891 The parent-death signal is process-directed (see
893 and, if the child installs a handler using the
900 argument of the handler contains the PID of the terminating parent process.
902 The parent-death signal setting is cleared for the child of a
905 (since Linux 2.4.36 / 2.6.23)
906 .\" commit d2d56c5f51028cb9f3d800882eb6f4cbd3f9099f
907 cleared when executing a set-user-ID or set-group-ID binary,
908 or a binary that has associated capabilities (see
909 .BR capabilities (7));
910 otherwise, this value is preserved across
913 .BR PR_GET_PDEATHSIG " (since Linux 2.3.15)"
914 Return the current value of the parent process death signal,
915 in the location pointed to by
916 .IR "(int\ *) arg2" .
918 .BR PR_SET_PTRACER " (since Linux 3.4)"
919 .\" commit 2d514487faf188938a4ee4fb3464eeecfbdcf8eb
920 .\" commit bf06189e4d14641c0148bea16e9dd24943862215
921 This is meaningful only when the Yama LSM is enabled and in mode 1
922 ("restricted ptrace", visible via
923 .IR /proc/sys/kernel/yama/ptrace_scope ).
924 When a "ptracer process ID" is passed in \fIarg2\fP,
925 the caller is declaring that the ptracer process can
927 the calling process as if it were a direct process ancestor.
930 operation replaces the previous "ptracer process ID".
935 set to 0 clears the caller's "ptracer process ID".
939 .BR PR_SET_PTRACER_ANY ,
940 the ptrace restrictions introduced by Yama are effectively disabled for the
943 For further information, see the kernel source file
944 .IR Documentation/admin\-guide/LSM/Yama.rst
945 .\" commit 90bb766440f2147486a2acc3e793d7b8348b0c22
947 .IR Documentation/security/Yama.txt
950 .BR PR_SET_SECCOMP " (since Linux 2.6.23)"
951 .\" See http://thread.gmane.org/gmane.linux.kernel/542632
952 .\" [PATCH 0 of 2] seccomp updates
953 .\" andrea@cpushare.com
954 Set the secure computing (seccomp) mode for the calling thread, to limit
955 the available system calls.
958 system call provides a superset of the functionality of
961 The seccomp mode is selected via
963 (The seccomp constants are defined in
964 .IR <linux/seccomp.h> .)
969 .BR SECCOMP_MODE_STRICT ,
970 the only system calls that the thread is permitted to make are
978 Other system calls result in the delivery of a
981 Strict secure computing mode is useful for number-crunching applications
982 that may need to execute untrusted byte code,
983 perhaps obtained by reading from a pipe or socket.
984 This operation is available only
985 if the kernel is configured with
992 .BR SECCOMP_MODE_FILTER " (since Linux 3.5),"
993 the system calls allowed are defined by a pointer
994 to a Berkeley Packet Filter passed in
996 This argument is a pointer to
997 .IR "struct sock_fprog" ;
998 it can be designed to filter
999 arbitrary system calls and system call arguments.
1000 This mode is available only if the kernel is configured with
1001 .B CONFIG_SECCOMP_FILTER
1005 .BR SECCOMP_MODE_FILTER
1008 then the seccomp mode is inherited by children created by
1012 is permitted, then the seccomp mode is preserved across
1014 If the filters permit
1016 calls, then additional filters can be added;
1017 they are run in order until the first non-allow result is seen.
1019 For further information, see the kernel source file
1020 .IR Documentation/userspace\-api/seccomp_filter.rst
1021 .\" commit c061f33f35be0ccc80f4b8e0aea5dfd2ed7e01a3
1023 .IR Documentation/prctl/seccomp_filter.txt
1026 .BR PR_GET_SECCOMP " (since Linux 2.6.23)"
1027 Return (as the function result)
1028 the secure computing mode of the calling thread.
1029 If the caller is not in secure computing mode, this operation returns 0;
1030 if the caller is in strict secure computing mode, then the
1034 signal to be sent to the process.
1035 If the caller is in filter mode, and this system call is allowed by the
1036 seccomp filters, it returns 2; otherwise, the process is killed with a
1039 This operation is available only
1040 if the kernel is configured with
1044 Since Linux 3.8, the
1047 .IR /proc/[pid]/status
1048 file provides a method of obtaining the same information,
1049 without the risk that the process is killed; see
1052 .BR PR_SET_SECUREBITS " (since Linux 2.6.26)"
1053 Set the "securebits" flags of the calling thread to the value supplied in
1056 .BR capabilities (7).
1058 .BR PR_GET_SECUREBITS " (since Linux 2.6.26)"
1059 Return (as the function result)
1060 the "securebits" flags of the calling thread.
1062 .BR capabilities (7).
1064 .BR PR_GET_SPECULATION_CTRL " (since Linux 4.17)"
1065 Returns the state of the speculation misfeature specified in
1067 Currently, the only permitted value for this argument is
1068 .BR PR_SPEC_STORE_BYPASS
1069 (otherwise the call fails with the error
1072 The return value uses bits 0-3 with the following meaning:
1076 Mitigation can be controlled per thread by
1077 .B PR_SET_SPECULATION_CTRL
1080 The speculation feature is enabled, mitigation is disabled.
1083 The speculation feature is disabled, mitigation is enabled
1085 .BR PR_SPEC_FORCE_DISABLE
1088 but cannot be undone.
1092 then the CPU is not affected by the speculation misfeature.
1096 is set, then per-thread control of the mitigation is available.
1099 for the speculation misfeature will fail.
1106 arguments must be specified as 0; otherwise the call fails with the error
1109 .BR PR_SET_SPECULATION_CTRL " (since Linux 4.17)"
1110 .\" commit b617cfc858161140d69cc0b5cc211996b557a1c7
1111 .\" commit 356e4bfff2c5489e016fdb925adbf12a1e3950ee
1112 Sets the state of the speculation misfeature specified in
1114 Currently, the only permitted value for this argument is
1115 .B PR_SPEC_STORE_BYPASS
1116 (otherwise the call fails with the error
1118 This setting is a per-thread attribute.
1121 argument is used to hand in the control value,
1122 which is one of the following:
1126 The speculation feature is enabled, mitigation is disabled.
1129 The speculation feature is disabled, mitigation is enabled
1131 .BR PR_SPEC_FORCE_DISABLE
1134 but cannot be undone.
1137 prctl(..., PR_SPEC_ENABLE)
1138 will fail with the error
1144 will result in the call failing with the error
1151 arguments must be specified as 0; otherwise the call fails with the error
1154 The speculation feature can also be controlled by the
1155 .B spec_store_bypass_disable
1157 This parameter may enforce a read-only policy which will result in the
1159 call failing with the error
1161 For further details, see the kernel source file
1162 .IR Documentation/admin-guide/kernel-parameters.txt .
1164 .BR PR_SET_THP_DISABLE " (since Linux 3.15)"
1165 .\" commit a0715cc22601e8830ace98366c0c2bd8da52af52
1166 Set the state of the "THP disable" flag for the calling thread.
1169 has a nonzero value, the flag is set, otherwise it is cleared.
1170 Setting this flag provides a method
1171 for disabling transparent huge pages
1172 for jobs where the code cannot be modified, and using a malloc hook with
1174 is not an option (i.e., statically allocated data).
1175 The setting of the "THP disable" flag is inherited by a child created via
1177 and is preserved across
1181 .BR PR_TASK_PERF_EVENTS_DISABLE " (since Linux 2.6.31)"
1182 Disable all performance counters attached to the calling process,
1183 regardless of whether the counters were created by
1184 this process or another process.
1185 Performance counters created by the calling process for other
1186 processes are unaffected.
1187 For more information on performance counters, see the Linux kernel source file
1188 .IR tools/perf/design.txt .
1191 .BR PR_TASK_PERF_COUNTERS_DISABLE ;
1192 .\" commit 1d1c7ddbfab358445a542715551301b7fc363e28
1193 renamed (retaining the same numerical value)
1197 .BR PR_TASK_PERF_EVENTS_ENABLE " (since Linux 2.6.31)"
1199 .BR PR_TASK_PERF_EVENTS_DISABLE ;
1200 enable performance counters attached to the calling process.
1203 .BR PR_TASK_PERF_COUNTERS_ENABLE ;
1204 .\" commit 1d1c7ddbfab358445a542715551301b7fc363e28
1206 .\" commit cdd6c482c9ff9c55475ee7392ec8f672eddb7be6
1210 .BR PR_GET_THP_DISABLE " (since Linux 3.15)"
1211 Return (via the function result) the current setting of the "THP disable"
1212 flag for the calling thread:
1213 either 1, if the flag is set, or 0, if it is not.
1215 .BR PR_GET_TID_ADDRESS " (since Linux 3.5)"
1216 .\" commit 300f786b2683f8bb1ec0afb6e1851183a479c86d
1220 .BR set_tid_address (2)
1223 .B CLONE_CHILD_CLEARTID
1224 flag, in the location pointed to by
1225 .IR "(int\ **)\ arg2" .
1226 This feature is available only if the kernel is built with the
1227 .BR CONFIG_CHECKPOINT_RESTORE
1231 system call does not have a compat implementation for
1232 the AMD64 x32 and MIPS n32 ABIs,
1233 and the kernel writes out a pointer using the kernel's pointer size,
1234 this operation expects a user-space buffer of 8 (not 4) bytes on these ABIs.
1236 .BR PR_SET_TIMERSLACK " (since Linux 2.6.28)"
1237 .\" See https://lwn.net/Articles/369549/
1238 .\" commit 6976675d94042fbd446231d1bd8b7de71a980ada
1239 Each thread has two associated timer slack values:
1240 a "default" value, and a "current" value.
1241 This operation sets the "current" timer slack value for the calling thread.
1242 If the nanosecond value supplied in
1244 is greater than zero, then the "current" value is set to this value.
1247 is less than or equal to zero,
1248 .\" It seems that it's not possible to set the timer slack to zero;
1249 .\" The minimum value is 1? Seems a little strange.
1250 the "current" timer slack is reset to the
1251 thread's "default" timer slack value.
1253 The "current" timer slack is used by the kernel to group timer expirations
1254 for the calling thread that are close to one another;
1255 as a consequence, timer expirations for the thread may be
1256 up to the specified number of nanoseconds late (but will never expire early).
1257 Grouping timer expirations can help reduce system power consumption
1258 by minimizing CPU wake-ups.
1260 The timer expirations affected by timer slack are those set by
1266 .BR epoll_pwait (2),
1267 .BR clock_nanosleep (2),
1271 (and thus the library functions implemented via futexes, including
1272 .\" List obtained by grepping for futex usage in glibc source
1273 .BR pthread_cond_timedwait (3),
1274 .BR pthread_mutex_timedlock (3),
1275 .BR pthread_rwlock_timedrdlock (3),
1276 .BR pthread_rwlock_timedwrlock (3),
1278 .BR sem_timedwait (3)).
1280 Timer slack is not applied to threads that are scheduled under
1281 a real-time scheduling policy (see
1282 .BR sched_setscheduler (2)).
1284 When a new thread is created,
1285 the two timer slack values are made the same as the "current" value
1286 of the creating thread.
1287 Thereafter, a thread can adjust its "current" timer slack value via
1288 .BR PR_SET_TIMERSLACK .
1289 The "default" value can't be changed.
1290 The timer slack values of
1292 (PID 1), the ancestor of all processes,
1293 are 50,000 nanoseconds (50 microseconds).
1294 The timer slack values are preserved across
1297 Since Linux 4.6, the "current" timer slack value of any process
1298 can be examined and changed via the file
1299 .IR /proc/[pid]/timerslack_ns .
1303 .BR PR_GET_TIMERSLACK " (since Linux 2.6.28)"
1304 Return (as the function result)
1305 the "current" timer slack value of the calling thread.
1307 .BR PR_SET_TIMING " (since Linux 2.6.0)"
1308 .\" Precisely: Linux 2.6.0-test4
1309 Set whether to use (normal, traditional) statistical process timing or
1310 accurate timestamp-based process timing, by passing
1311 .B PR_TIMING_STATISTICAL
1314 .B PR_TIMING_TIMESTAMP
1317 .B PR_TIMING_TIMESTAMP
1318 is not currently implemented
1319 (attempting to set this mode will yield the error
1321 .\" PR_TIMING_TIMESTAMP doesn't do anything in 2.6.26-rc8,
1322 .\" and looking at the patch history, it appears
1323 .\" that it never did anything.
1325 .BR PR_GET_TIMING " (since Linux 2.6.0)"
1326 .\" Precisely: Linux 2.6.0-test4
1327 Return (as the function result) which process timing method is currently
1330 .BR PR_SET_TSC " (since Linux 2.6.26, x86 only)"
1331 Set the state of the flag determining whether the timestamp counter
1332 can be read by the process.
1337 to allow it to be read, or
1341 when the process tries to read the timestamp counter.
1343 .BR PR_GET_TSC " (since Linux 2.6.26, x86 only)"
1344 Return the state of the flag determining whether the timestamp counter
1346 in the location pointed to by
1347 .IR "(int\ *) arg2" .
1350 (Only on: ia64, since Linux 2.3.48; parisc, since Linux 2.6.15;
1351 PowerPC, since Linux 2.6.18; Alpha, since Linux 2.6.22;
1352 .\" sh: 94ea5e449ae834af058ef005d16a8ad44fcf13d6
1353 .\" tile: 2f9ac29eec71a696cb0dcc5fb82c0f8d4dac28c9
1354 sh, since Linux 2.6.34; tile, since Linux 3.12)
1355 Set unaligned access control bits to \fIarg2\fP.
1357 \fBPR_UNALIGN_NOPRINT\fP to silently fix up unaligned user accesses,
1358 or \fBPR_UNALIGN_SIGBUS\fP to generate
1360 on unaligned user access.
1361 Alpha also supports an additional flag with the value
1362 of 4 and no corresponding named constant,
1363 which instructs kernel to not fix up
1364 unaligned accesses (it is analogous to providing the
1370 system call on Tru64).
1375 for information on versions and architectures)
1376 Return unaligned access control bits, in the location pointed to by
1377 .IR "(unsigned int\ *) arg2" .
1380 .BR PR_GET_DUMPABLE ,
1381 .BR PR_GET_KEEPCAPS ,
1382 .BR PR_GET_NO_NEW_PRIVS ,
1383 .BR PR_GET_THP_DISABLE ,
1384 .BR PR_CAPBSET_READ ,
1386 .BR PR_GET_TIMERSLACK ,
1387 .BR PR_GET_SECUREBITS ,
1388 .BR PR_MCE_KILL_GET ,
1389 .BR PR_CAP_AMBIENT + PR_CAP_AMBIENT_IS_SET ,
1392 return the nonnegative values described above.
1395 values return 0 on success.
1396 On error, \-1 is returned, and
1398 is set appropriately.
1408 .BR SECCOMP_MODE_FILTER ,
1409 but the process does not have the
1411 capability or has not set the
1413 attribute (see the discussion of
1414 .BR PR_SET_NO_NEW_PRIVS
1424 .BR PR_SET_MM_EXE_FILE ,
1425 the file is not executable.
1433 .BR PR_SET_MM_EXE_FILE ,
1434 and the file descriptor passed in
1444 .BR PR_SET_MM_EXE_FILE ,
1445 and this the second attempt to change the
1447 symbolic link, which is prohibited.
1451 is an invalid address.
1456 .BR PR_SET_SECCOMP ,
1459 .BR SECCOMP_MODE_FILTER ,
1460 the system was built with
1461 .BR CONFIG_SECCOMP_FILTER ,
1464 is an invalid address.
1481 arguments were not specified as zero.
1485 is not valid value for this
1493 .BR PR_GET_SECCOMP ,
1494 and the kernel was not configured with
1495 .BR CONFIG_SECCOMP .
1500 .BR PR_SET_SECCOMP ,
1503 .BR SECCOMP_MODE_FILTER ,
1504 and the kernel was not configured with
1505 .BR CONFIG_SECCOMP_FILTER .
1511 and one of the following is true
1522 (the limit on the size of the user address space for this architecture);
1526 .BR PR_SET_MM_START_CODE ,
1527 .BR PR_SET_MM_END_CODE ,
1528 .BR PR_SET_MM_START_DATA ,
1529 .BR PR_SET_MM_END_DATA ,
1531 .BR PR_SET_MM_START_STACK ,
1532 and the permissions of the corresponding memory area are not as required;
1536 .BR PR_SET_MM_START_BRK
1541 is less than or equal to the end of the data segment
1542 or specifies a value that would cause the
1544 resource limit to be exceeded.
1554 .BR PR_SET_PTRACER_ANY ,
1555 or the PID of an existing process.
1563 is not a valid signal number.
1572 .B SUID_DUMP_DISABLE
1574 .BR SUID_DUMP_USER .
1583 .BR PR_TIMING_STATISTICAL .
1588 .BR PR_SET_NO_NEW_PRIVS
1602 .BR PR_GET_NO_NEW_PRIVS
1614 .BR PR_SET_THP_DISABLE
1625 .BR PR_GET_THP_DISABLE
1638 and an unused argument
1643 .BR PR_CAP_AMBIENT_CLEAR_ALL ,
1647 has an invalid value;
1651 .BR PR_CAP_AMBIENT_LOWER ,
1652 .BR PR_CAP_AMBIENT_RAISE ,
1654 .BR PR_CAP_AMBIENT_IS_SET
1657 does not specify a valid capability.
1662 .BR PR_SET_SPECULATION_CTRL
1663 the kernel or CPU does not support the requested speculation misfeature.
1668 .BR PR_MPX_ENABLE_MANAGEMENT
1670 .BR PR_MPX_DISABLE_MANAGEMENT
1671 and the kernel or the CPU does not support MPX management.
1672 Check that the kernel and processor have MPX support.
1677 .BR PR_SET_SPECULATION_CTRL
1678 implies that the control of the selected speculation misfeature is not possible.
1680 .BR PR_GET_SPECULATION_CTRL
1681 for the bit fields to determine which option is available.
1689 has an invalid or unsupported value.
1694 .BR PR_SET_SECUREBITS ,
1695 and the caller does not have the
1698 or tried to unset a "locked" flag,
1699 or tried to set a flag whose corresponding locked flag was set
1701 .BR capabilities (7)).
1706 .BR PR_SET_SPECULATION_CTRL
1707 wherein the speculation was disabled with
1708 .B PR_SPEC_FORCE_DISABLE
1709 and caller tried to enable it again.
1714 .BR PR_SET_KEEPCAPS ,
1716 .B SECBIT_KEEP_CAPS_LOCKED
1719 .BR capabilities (7)).
1724 .BR PR_CAPBSET_DROP ,
1725 and the caller does not have the
1733 and the caller does not have the
1744 .BR PR_CAP_AMBIENT_RAISE ,
1745 but either the capability specified in
1747 is not present in the process's permitted and inheritable capability sets,
1749 .B PR_CAP_AMBIENT_LOWER
1750 securebit has been set.
1755 .BR PR_SET_SPECULATION_CTRL
1759 .BR PR_SPEC_ENABLE ,
1760 .BR PR_SPEC_DISABLE ,
1762 .BR PR_SPEC_FORCE_DISABLE .
1767 .BR PR_GET_SPECULATION_CTRL
1769 .BR PR_SET_SPECULATION_CTRL
1770 and unused arguments to
1776 system call was introduced in Linux 2.1.57.
1777 .\" The library interface was added in glibc 2.0.6
1779 This call is Linux-specific.
1782 system call (also introduced in Linux 2.1.44
1783 as irix_prctl on the MIPS architecture),
1788 .BI "ptrdiff_t prctl(int " option ", int " arg2 ", int " arg3 );
1792 and options to get the maximum number of processes per user,
1793 get the maximum number of processors the calling process can use,
1794 find out whether a specified process is currently blocked,
1795 get or set the maximum stack size, and so on.