1 ;;;; tags which are set during the build process and which end up in
2 ;;;; CL:*FEATURES* in the target SBCL, plus some comments about other
3 ;;;; CL:*FEATURES* tags which have special meaning to SBCL or which
4 ;;;; have a special conventional meaning
6 ;;;; Note that the recommended way to customize the features of a
7 ;;;; local build of SBCL is not to edit this file, but instead to
8 ;;;; tweak customize-target-features.lisp. If you define a function
9 ;;;; in customize-target-features.lisp, it will be used to transform
10 ;;;; the target features list after it's read and before it's used.
11 ;;;; E.g. you can use code like this:
13 ;;;; (flet ((enable (x) (pushnew x list))
14 ;;;; (disable (x) (setf list (remove x list))))
15 ;;;; #+nil (enable :sb-show)
16 ;;;; (enable :sb-after-xc-core)
17 ;;;; #+nil (disable :sb-doc)
19 ;;;; By thus editing a local file (one which is not in the source
20 ;;;; distribution, and which is in .cvsignore) your customizations
21 ;;;; will remain local even if you do things like "cvs update",
22 ;;;; will not show up if you try to submit a patch with "cvs diff",
23 ;;;; and might even stay out of the way if you use other non-CVS-based
24 ;;;; methods to upgrade the files or store your configuration.
26 ;;;; This software is part of the SBCL system. See the README file for
27 ;;;; more information.
29 ;;;; This software is derived from the CMU CL system, which was
30 ;;;; written at Carnegie Mellon University and released into the
31 ;;;; public domain. The software is in the public domain and is
32 ;;;; provided with absolutely no warranty. See the COPYING and CREDITS
33 ;;;; files for more information.
37 ;; features present in all builds
42 ;; FIXME: Isn't there a :x3jsomething feature which we should set too?
43 ;; No. CLHS says ":x3j13 [...] A conforming implementation might or
44 ;; might not contain such a feature." -- CSR, 2002-02-21
49 ;; Douglas Thomas Crosher's conservative generational GC (the only one
50 ;; we currently support for X86).
51 ;; :gencgc used to be here; CSR moved it into
52 ;; local-target-features.lisp-expr via make-config.sh, as alpha,
53 ;; sparc and ppc ports don't currently support it. -- CSR, 2002-02-21
55 ;; We're running under a UNIX. This is sort of redundant, and it was also
56 ;; sort of redundant under CMU CL, which we inherited it from: neither SBCL
57 ;; nor CMU CL supports anything but UNIX (and "technically not UNIX"es
58 ;; such as *BSD and Linux). But someday, maybe we might, and in that case
59 ;; we'd presumably remove this, so its presence conveys the information
60 ;; that the system isn't one which follows such a change.
64 ;; features present in this particular build
67 ;; Setting this enables the compilation of documentation strings
68 ;; from the system sources into the target Lisp executable.
69 ;; Traditional Common Lisp folk will want this option set.
70 ;; I (WHN) made it optional because I came to Common Lisp from
71 ;; C++ through Scheme, so I'm accustomed to asking
72 ;; Emacs about things that I'm curious about instead of asking
73 ;; the executable I'm running.
76 ;; Do regression and other tests when building the system. You
77 ;; might or might not want this if you're not a developer,
78 ;; depending on how paranoid you are. You probably do want it if
79 ;; you are a developer.
82 ;; Make more debugging information available (for debugging SBCL
83 ;; itself). If you aren't hacking or troubleshooting SBCL itself,
84 ;; you probably don't want this set.
86 ;; At least two varieties of debugging information are enabled by this
88 ;; * SBCL is compiled with a higher level of OPTIMIZE DEBUG, so that
89 ;; the debugger can tell more about the state of the system.
90 ;; * Various code to print debugging messages, and similar debugging code,
91 ;; is compiled only when this feature is present.
93 ;; Note that the extra information recorded by the compiler at
94 ;; this higher level of OPTIMIZE DEBUG includes the source location
95 ;; forms. In order for the debugger to use this information, it has to
96 ;; re-READ the source file. In an ordinary installation of SBCL, this
97 ;; re-READing may not work very well, for either of two reasons:
98 ;; * The sources aren't present on the system in the same location that
99 ;; they were on the system where SBCL was compiled.
100 ;; * SBCL is using the standard readtable, without the added hackage
101 ;; which allows it to handle things like target features.
102 ;; If you want to be able to use the extra debugging information,
103 ;; therefore, be sure to keep the sources around, and run with the
104 ;; readtable configured so that the system sources can be read.
107 ;; Build SBCL with the old CMU CL low level debugger, "ldb". If
108 ;; are aren't messing with CMU CL at a very low level (e.g.
109 ;; trying to diagnose GC problems, or trying to debug assembly
110 ;; code for a port to a new CPU) you shouldn't need this.
113 ;; This isn't really a target Lisp feature at all, but controls
114 ;; whether the build process produces an after-xc.core file. This
115 ;; can be useful for shortening the edit/compile/debug cycle when
116 ;; you modify SBCL's own source code, as in slam.sh. Otherwise
117 ;; you don't need it.
120 ;; Enable extra debugging output in the assem.lisp assembler/scheduler
121 ;; code. (This is the feature which was called :DEBUG in the
122 ;; original CMU CL code.)
125 ;; Setting this makes SBCL more "fluid", i.e. more amenable to
126 ;; modification at runtime, by suppressing various INLINE declarations,
127 ;; compiler macro definitions, FREEZE-TYPE declarations; and by
128 ;; suppressing various burning-our-ships-behind-us actions after
129 ;; initialization is complete; and so forth. This tends to clobber the
130 ;; performance of the system, so unless you have some special need for
131 ;; this when hacking SBCL itself, you don't want this set.
134 ;; Enable code for collecting statistics on usage of various operations,
135 ;; useful for performance tuning of the SBCL system itself. This code
136 ;; is probably pretty stale (having not been tested since the fork from
137 ;; base CMU CL) but might nonetheless be a useful starting point for
138 ;; anyone who wants to collect such statistics in the future.
141 ;; Peter Van Eynde's increase-bulletproofness code for CMU CL
143 ;; Some of the code which was #+high-security before the fork has now
144 ;; been either made unconditional, deleted, or rewritten into
145 ;; unrecognizability, but some remains. What remains is not maintained
146 ;; or tested in current SBCL, but I haven't gone out of my way to
150 ; :high-security-support
152 ;; multiprocessing support
154 ;; This is not maintained or tested in current SBCL. I haven't gone out
155 ;; of my way to break it, but since it's derived from an old version of
156 ;; CMU CL where multiprocessing was pretty shaky, it's likely to be very
158 ;; :MP enables multiprocessing
159 ;; :MP-I486 is used, only within the multiprocessing code, to control
160 ;; what seems to control processor-version-specific code. It's
161 ;; probably for 486 or later, i.e. could be set as long as
162 ;; you know you're not running on a 386, but it doesn't seem
163 ;; to be documented anywhere, so that's just a guess.
167 ;; This affects the definition of a lot of things in bignum.lisp. It
168 ;; doesn't seem to be documented anywhere what systems it might apply
169 ;; to. It doesn't seem to be needed for X86 systems anyway.
172 ;; This is probably true for some processor types, but not X86. It
173 ;; affects a lot of floating point code.
174 ; :negative-zero-is-not-zero
176 ;; This is set in classic CMU CL, and presumably there it means
177 ;; that the floating point arithmetic implementation
178 ;; conforms to IEEE's standard. Here it definitely means that the
179 ;; floating point arithmetic implementation conforms to IEEE's standard.
180 ;; I (WHN 19990702) haven't tried to verify
181 ;; that it does conform, but it should at least mostly conform (because
182 ;; the underlying x86 hardware tries).
185 ;; CMU CL had, and we inherited, code to support 80-bit LONG-FLOAT on the x86
186 ;; architecture. Nothing has been done to actively destroy the long float
187 ;; support, but it hasn't been thoroughly maintained, and needs at least
188 ;; some maintenance before it will work. (E.g. the LONG-FLOAT-only parts of
189 ;; genesis are still implemented in terms of unportable CMU CL functions
190 ;; which are not longer available at genesis time in SBCL.) A deeper
191 ;; problem is SBCL's bootstrap process implicitly assumes that the
192 ;; cross-compilation host will be able to make the same distinctions
193 ;; between floating point types that it does. This assumption is
194 ;; fundamentally sleazy, even though in practice it's unlikely to break down
195 ;; w.r.t. distinguishing SINGLE-FLOAT from DOUBLE-FLOAT; it's much more
196 ;; likely to break down w.r.t. distinguishing DOUBLE-FLOAT from LONG-FLOAT.
197 ;; Still it's likely to be quite doable to get LONG-FLOAT support working
198 ;; again, if anyone's sufficiently motivated.
202 ;; miscellaneous notes on other things which could have special significance
203 ;; in the *FEATURES* list
206 ;; notes on the :NIL and :IGNORE features:
208 ;; #+NIL is used to comment out forms. Occasionally #+IGNORE is used
209 ;; for this too. So don't use :NIL or :IGNORE as the names of features..
211 ;; notes on :SB-XC and :SB-XC-HOST features (which aren't controlled by this
212 ;; file, but are instead temporarily pushed onto *FEATURES* or
213 ;; *TARGET-FEATURES* during some phases of cross-compilation):
215 ;; :SB-XC-HOST stands for "cross-compilation host" and is in *FEATURES*
216 ;; during the first phase of cross-compilation bootstrapping, when the
217 ;; host Lisp is being used to compile the cross-compiler.
219 ;; :SB-XC stands for "cross compiler", and is in *FEATURES* during the second
220 ;; phase of cross-compilation bootstrapping, when the cross-compiler is
221 ;; being used to create the first target Lisp.
223 ;; notes on the :SB-ASSEMBLING feature (which isn't controlled by
226 ;; This is a flag for whether we're in the assembler. It's
227 ;; temporarily pushed onto the *FEATURES* list in the setup for
228 ;; the ASSEMBLE-FILE function. It would be a bad idea
229 ;; to use it as a name for a permanent feature.
231 ;; notes on local features (which are set automatically by the
232 ;; configuration script, and should not be set here unless you
233 ;; really, really know what you're doing):
235 ;; machine architecture features:
237 ;; any Intel 386 or better, or compatibles like the AMD K6 or K7
239 ;; DEC/Compaq Alpha CPU
240 ;; (No other CPUs are supported by SBCL as of 0.6.12.15, but SPARC or
241 ;; PowerPC support could be ported from CMU CL if anyone is
242 ;; sufficiently motivated to do so, or if you're *really* motivated,
243 ;; you could write a port from scratch for a new CPU architecture.)
244 ;; (CMU CL also had a :pentium feature, which affected the definition
245 ;; of some floating point vops. It was present but not enabled or
246 ;; documented in the CMU CL code that SBCL is derived from, and is
247 ;; present but stale in SBCL as of 0.6.12.)
249 ;; operating system features:
250 ;; :linux = We're intended to run under some version of Linux.
251 ;; :bsd = We're intended to run under some version of BSD Unix. (This
252 ;; is not exclusive with the features which indicate which
253 ;; particular version of BSD we're intended to run under.)
254 ;; :freebsd = We're intended to run under FreeBSD.
255 ;; :openbsd = We're intended to run under FreeBSD.
256 ;; (No others are supported by SBCL as of 0.6.7, but :hpux or
257 ;; :solaris support could be ported from CMU CL if anyone is
258 ;; sufficiently motivated to do so, and it'd even be possible,
259 ;; though harder, to port the system to Microsoft Windows.)