xref: /freebsd/share/man/man9/stack.9 (revision 02e9120893770924227138ba49df1edb3896112a)
1.\"
2.\" Copyright (c) 2007-2009 Robert N. M. Watson
3.\" All rights reserved.
4.\"
5.\" Redistribution and use in source and binary forms, with or without
6.\" modification, are permitted provided that the following conditions
7.\" are met:
8.\" 1. Redistributions of source code must retain the above copyright
9.\"    notice(s), this list of conditions and the following disclaimer as
10.\"    the first lines of this file unmodified other than the possible
11.\"    addition of one or more copyright notices.
12.\" 2. Redistributions in binary form must reproduce the above copyright
13.\"    notice(s), this list of conditions and the following disclaimer in the
14.\"    documentation and/or other materials provided with the distribution.
15.\"
16.\" THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER(S) ``AS IS'' AND ANY
17.\" EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
18.\" WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
19.\" DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY
20.\" DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
21.\" (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
22.\" SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
23.\" CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
26.\" DAMAGE.
27.\"
28.Dd March 6, 2022
29.Dt STACK 9
30.Os
31.Sh NAME
32.Nm stack
33.Nd kernel thread stack tracing routines
34.Sh SYNOPSIS
35.In sys/param.h
36.In sys/stack.h
37.Pp
38In the kernel configuration file:
39.Cd "options DDB"
40.Cd "options STACK"
41.Pp
42.Ft struct stack *
43.Fn stack_create "int flags"
44.Ft void
45.Fn stack_destroy "struct stack *st"
46.Ft int
47.Fn stack_put "struct stack *st" "vm_offset_t pc"
48.Ft void
49.Fn stack_copy "const struct stack *src" "struct stack *dst"
50.Ft void
51.Fn stack_zero "struct stack *st"
52.Ft void
53.Fn stack_print "const struct stack *st"
54.Ft void
55.Fn stack_print_ddb "const struct stack *st"
56.Ft void
57.Fn stack_print_short "const struct stack *st"
58.Ft void
59.Fn stack_print_short_ddb "const struct stack *st"
60.Ft void
61.Fn stack_sbuf_print "struct sbuf *sb" "const struct stack *st"
62.Ft void
63.Fn stack_sbuf_print_ddb "struct sbuf *sb" "const struct stack *st"
64.Ft void
65.Fn stack_save "struct stack *st"
66.Ft int
67.Fn stack_save_td "struct stack *st" "struct thread *td"
68.Sh DESCRIPTION
69The
70.Nm
71KPI allows querying of kernel stack trace information and the automated
72generation of kernel stack trace strings for the purposes of debugging and
73tracing.
74To use the KPI, at least one of
75.Cd "options DDB"
76and
77.Cd "options STACK"
78must be compiled into the kernel.
79.Pp
80Each stack trace is described by a
81.Vt "struct stack" .
82It can be declared in the usual ways, including on the stack, and optionally
83initialized with
84.Fn stack_zero ,
85though this is not necessary before saving a trace.
86It can also be dynamically allocated with
87.Fn stack_create .
88The
89.Ar flags
90argument is passed to
91.Xr malloc 9 .
92This dynamic allocation must be freed with
93.Fn stack_destroy .
94.Pp
95A trace of the current thread's kernel call stack may be captured using
96.Fn stack_save .
97.Fn stack_save_td
98can be used to capture the kernel stack of a caller-specified thread.
99Callers of
100.Fn stack_save_td
101must own the thread lock of the specified thread,
102and the thread's stack must not be swapped out.
103.Fn stack_save_td
104can capture the kernel stack of a running thread, though note that this is
105not implemented on all platforms.
106If the thread is running, the caller must also hold the process lock for the
107target thread.
108.Pp
109.Fn stack_print
110and
111.Fn stack_print_short
112may be used to print a stack trace using the kernel
113.Xr printf 9 ,
114and may sleep as a result of acquiring
115.Xr sx 9
116locks in the kernel linker while looking up symbol names.
117In locking-sensitive environments, the unsynchronized
118.Fn stack_print_ddb
119and
120.Fn stack_print_short_ddb
121variants may be invoked.
122This function bypasses kernel linker locking, making it usable in
123.Xr ddb 4 ,
124but not in a live system where linker data structures may change.
125.Pp
126.Fn stack_sbuf_print
127may be used to construct a human-readable string, including conversion (where
128possible) from a simple kernel instruction pointer to a named symbol and
129offset.
130The argument
131.Ar sb
132must be an initialized
133.Dv struct sbuf
134as described in
135.Xr sbuf 9 .
136This function may sleep if an auto-extending
137.Dv struct sbuf
138is used, or due to kernel linker locking.
139In locking-sensitive environments, such as
140.Xr ddb 4 ,
141the unsynchronized
142.Fn stack_sbuf_print_ddb
143variant may be invoked to avoid kernel linker locking; it should be used with
144a fixed-length sbuf.
145.Pp
146The utility functions
147.Nm stack_zero ,
148.Nm stack_copy ,
149and
150.Nm stack_put
151may be used to manipulate stack data structures directly.
152.Sh RETURN VALUES
153.Fn stack_put
154returns 0 on success.
155Otherwise the
156.Dv struct stack
157does not contain space to record additional frames, and a non-zero value is
158returned.
159.Pp
160.Fn stack_save_td
161returns 0 when the stack capture was successful and a non-zero error number
162otherwise.
163In particular,
164.Er EBUSY
165is returned if the thread was running in user mode at the time that the
166capture was attempted, and
167.Er EOPNOTSUPP
168is returned if the operation is not implemented.
169.Sh SEE ALSO
170.Xr ddb 4 ,
171.Xr printf 9 ,
172.Xr sbuf 9 ,
173.Xr sx 9
174.Sh AUTHORS
175.An -nosplit
176The
177.Nm
178function suite was created by
179.An Antoine Brodin .
180.Nm
181was extended by
182.An Robert Watson
183for general-purpose use outside of
184.Xr ddb 4 .
185