xref: /freebsd/share/man/man7/simd.7 (revision 7ef62cebc2f965b0f640263e179276928885e33d) 
1.\" Copyright (c) 2023 The FreeBSD Foundation
2.
3.\" This documentation was written by Robert Clausecker <fuz@FreeBSD.org>
4.\" under sponsorship from the FreeBSD Foundation.
5.
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in the
13.\"    documentation and/or other materials provided with the distribution.
14.
15.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ''AS IS'' AND
16.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
17.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
18.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
19.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
20.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
21.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
22.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
23.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
24.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
25.\" SUCH DAMAGE
26.
27.Dd August 5, 2023
28.Dt SIMD 7
29.Os
30.Sh NAME
31.Nm simd
32.Nd SIMD enhancements
33.
34.Sh DESCRIPTION
35On some architectures, the
36.Fx
37.Em libc
38provides enhanced implementations of commonly used functions, replacing
39the architecture-independent implementations used otherwise.
40Depending on architecture and function, an enhanced
41implementation of a function may either always be used or the
42.Em libc
43detects at runtime which SIMD instruction set extensions are
44supported and picks the most suitable implementation automatically.
45On
46.Cm amd64 ,
47the environment variable
48.Ev ARCHLEVEL
49can be used to override this mechanism.
50.Pp
51Enhanced functions are present in the following architectures:
52.Bl -column FUNCTION__ aarch64_ arm_ amd64_ i386_ ppc64_ -offset indent
53.It Em FUNCTION  Ta Em AARCH64 Ta Em ARM Ta Em AMD64  Ta Em I386 Ta Em PPC64
54.It    bcmp      Ta            Ta        Ta    S      Ta    S
55.It    bcopy     Ta            Ta    S   Ta    S      Ta    S    Ta    SV
56.It    bzero     Ta            Ta    S   Ta    S      Ta    S
57.It    div       Ta            Ta        Ta    S      Ta    S
58.It    index     Ta    S       Ta        Ta    S1
59.It    ldiv      Ta            Ta        Ta    S      Ta    S
60.It    lldiv     Ta            Ta        Ta    S
61.It    memcmp    Ta            Ta    S   Ta    S      Ta    S
62.It    memcpy    Ta    S       Ta    S   Ta    S      Ta    S    Ta    SV
63.It    memmove   Ta    S       Ta    S   Ta    S      Ta    S    Ta    SV
64.It    memset    Ta            Ta    S   Ta    S      Ta    S
65.It    rindex    Ta    S
66.It    stpcpy    Ta            Ta        Ta    S
67.It    strcat    Ta            Ta        Ta    S      Ta    S
68.It    strchr    Ta    S       Ta        Ta    S1     Ta    S
69.It    strchrnul Ta            Ta        Ta    S1
70.It    strcmp    Ta            Ta    S   Ta    S      Ta    S
71.It    strcpy    Ta            Ta        Ta    S      Ta    S    Ta    S2
72.It    strlen    Ta            Ta    S   Ta    S1
73.It    strncmp   Ta            Ta    S   Ta           Ta    S
74.It    strncpy   Ta            Ta        Ta           Ta         Ta    S2
75.It    strrchr   Ta    S       Ta        Ta           Ta    S
76.It    swab      Ta            Ta        Ta           Ta    S
77.It    wcschr    Ta            Ta        Ta           Ta    S
78.It    wcscmp    Ta            Ta        Ta           Ta    S
79.It    wcslen    Ta            Ta        Ta           Ta    S
80.It    wmemchr   Ta            Ta        Ta           Ta    S
81.El
82.Pp
83.Sy S Ns :\ scalar (non-SIMD),
84.Sy 1 Ns :\ amd64 baseline,
85.Sy 2 Ns :\ x86-64-v2
86or PowerPC\ 2.05,
87.Sy 3 Ns :\ x86-64-v3,
88.Sy 4 Ns :\ x86-64-v4,
89.Sy V Ns :\ PowerPC\ VSX.
90.
91.Sh ENVIRONMENT
92.Bl -tag
93.It Ev ARCHLEVEL
94On
95.Em amd64 ,
96controls the level of SIMD enhancements used.
97If this variable is set to an architecture level from the list below
98and that architecture level is supported by the processor, SIMD
99enhancements up to
100.Ev ARCHLEVEL
101are used.
102If
103.Ev ARCHLEVEL
104is unset, not recognised, or not supported by the processor, the highest
105level of SIMD enhancements supported by the processor is used.
106.Pp
107A suffix beginning with
108.Sq ":"
109or
110.Sq "+"
111in
112.Ev ARCHLEVEL
113is ignored and may be used for future extensions.
114The architecture level can be prefixed with a
115.Sq "!"
116character to force use of the requested architecture level, even if the
117processor does not advertise that it is supported.
118This usually causes applications to crash and should only be used for
119testing purposes or if architecture level detection yields incorrect
120results.
121.Pp
122The architecture levels follow the AMD64 SysV ABI supplement:
123.Bl -tag -width x86-64-v2
124.It Cm scalar
125scalar enhancements only (no SIMD)
126.It Cm baseline
127cmov, cx8, x87 FPU, fxsr, MMX, osfxsr, SSE, SSE2
128.It Cm x86-64-v2
129cx16, lahf/sahf, popcnt, SSE3, SSSE3, SSE4.1, SSE4.2
130.It Cm x86-64-v3
131AVX, AVX2, BMI1, BMI2, F16C, FMA, lzcnt, movbe, osxsave
132.It Cm x86-64-v4
133AVX-512F/BW/CD/DQ/VL
134.El
135.El
136.
137.Sh DIAGNOSTICS
138.Bl -diag
139.It "Illegal Instruction"
140Printed by
141.Xr sh 1
142if a command is terminated through delivery of a
143.Dv SIGILL
144signal, see
145.Xr signal 3 .
146.Pp
147Use of an unsupported architecture level was forced by setting
148.Ev ARCHLEVEL
149to a string beginning with a
150.Sq "!"
151character, causing a process to crash due to use of an unsupported
152instruction.
153Unset
154.Ev ARCHLEVEL ,
155remove the
156.Sq "!"
157prefix or select a supported architecture level.
158.Pp
159Message may also appear for unrelated reasons.
160.El
161.
162.Sh SEE ALSO
163.Xr string 3 ,
164.Xr arch 7
165.Rs
166.%A H. J. Lu
167.%A Michael Matz
168.%A Milind Girkar
169.%A Jan Hubi\[u010D]ka \" \(vc
170.%A Andreas Jaeger
171.%A Mark Mitchell
172.%B System V Application Binary Interface
173.%D May 23, 2023
174.%T AMD64 Architecture Processor Supplement
175.%O Version 1.0
176.Re
177.
178.Sh HISTORY
179Architecture-specific enhanced
180.Em libc
181functions were added starting
182with
183.Fx 2.0
184for
185.Cm i386 ,
186.Fx 6.0
187for
188.Cm arm ,
189.Fx 6.1
190for
191.Cm amd64 ,
192.Fx 11.0
193for
194.Cm aarch64 ,
195and
196.Fx 12.0
197for
198.Cm powerpc64 .
199SIMD-enhanced functions were first added with
200.Fx 13.0
201for
202.Cm powerpc64
203and with
204.Fx 14.0
205for
206.Cm amd64 .
207.Pp
208A
209.Nm
210manual page appeared in
211.Fx 14.0 .
212.
213.Sh AUTHOR
214.An Robert Clausecker Aq Mt fuz@FreeBSD.org
215.
216.Sh CAVEATS
217Other parts of
218.Fx
219such as cryptographic routines in the kernel or in
220OpenSSL may also use SIMD enhancements.
221These enhancements are not subject to the
222.Ev ARCHLEVEL
223variable and may have their own configuration
224mechanism.
225.
226.Sh BUGS
227Use of SIMD enhancements cannot be configured on powerpc64.
228