3 ;;;; tags which are set during the build process and which end up in
4 ;;;; CL:*FEATURES* in the target SBCL, plus some comments about other
5 ;;;; CL:*FEATURES* tags which have special meaning to SBCL or which
6 ;;;; have a special conventional meaning
8 ;;;; Note that the recommended way to customize the features of a
9 ;;;; local build of SBCL is not to edit this file, but instead to
10 ;;;; tweak customize-target-features.lisp. (You must create this file
11 ;;;; first; it is not in the SBCL distribution, and is in fact
12 ;;;; explicitly excluded from the distribution in places like
13 ;;;; .cvsignore.) If you define a function in
14 ;;;; customize-target-features.lisp, it will be used to transform the
15 ;;;; target features list after it's read and before it's used. E.g.,
16 ;;;; you can use code like this:
18 ;;;; (flet ((enable (x) (pushnew x list))
19 ;;;; (disable (x) (setf list (remove x list))))
20 ;;;; #+nil (enable :sb-show)
21 ;;;; (enable :sb-after-xc-core)
22 ;;;; #+nil (disable :sb-doc)
24 ;;;; By thus editing a local file (one which is not in the source
25 ;;;; distribution, and which is in .cvsignore) your customizations
26 ;;;; will remain local even if you do things like "cvs update",
27 ;;;; will not show up if you try to submit a patch with "cvs diff",
28 ;;;; and might even stay out of the way if you use other non-CVS-based
29 ;;;; methods to upgrade the files or store your configuration.
31 ;;;; This software is part of the SBCL system. See the README file for
32 ;;;; more information.
34 ;;;; This software is derived from the CMU CL system, which was
35 ;;;; written at Carnegie Mellon University and released into the
36 ;;;; public domain. The software is in the public domain and is
37 ;;;; provided with absolutely no warranty. See the COPYING and CREDITS
38 ;;;; files for more information.
42 ;; features present in all builds
47 ;; FIXME: Isn't there a :x3jsomething feature which we should set too?
48 ;; No. CLHS says ":x3j13 [...] A conforming implementation might or
49 ;; might not contain such a feature." -- CSR, 2002-02-21
54 ;; Douglas Thomas Crosher's conservative generational GC (the only one
55 ;; we currently support for X86).
56 ;; :gencgc used to be here; CSR moved it into
57 ;; local-target-features.lisp-expr via make-config.sh, as alpha,
58 ;; sparc and ppc ports don't currently support it. -- CSR, 2002-02-21
61 ;; features present in this particular build
64 ;; Setting this enables the compilation of documentation strings
65 ;; from the system sources into the target Lisp executable.
66 ;; Traditional Common Lisp folk will want this option set.
67 ;; I (WHN) made it optional because I came to Common Lisp from
68 ;; C++ through Scheme, so I'm accustomed to asking
69 ;; Emacs about things that I'm curious about instead of asking
70 ;; the executable I'm running.
73 ;; Do regression and other tests when building the system. You might
74 ;; or might not want this if you're not a developer, depending on how
75 ;; paranoid you are. You probably do want it if you are a developer.
76 ;; This test does not affect the target system (in much the same way
77 ;; as :sb-after-xc-core, below).
80 ;; Make more debugging information available (for debugging SBCL
81 ;; itself). If you aren't hacking or troubleshooting SBCL itself,
82 ;; you probably don't want this set.
84 ;; At least two varieties of debugging information are enabled by this
86 ;; * SBCL is compiled with a higher level of OPTIMIZE DEBUG, so that
87 ;; the debugger can tell more about the state of the system.
88 ;; * Various code to print debugging messages, and similar debugging code,
89 ;; is compiled only when this feature is present.
91 ;; Note that the extra information recorded by the compiler at
92 ;; this higher level of OPTIMIZE DEBUG includes the source location
93 ;; forms. In order for the debugger to use this information, it has to
94 ;; re-READ the source file. In an ordinary installation of SBCL, this
95 ;; re-READing may not work very well, for either of two reasons:
96 ;; * The sources aren't present on the system in the same location that
97 ;; they were on the system where SBCL was compiled.
98 ;; * SBCL is using the standard readtable, without the added hackage
99 ;; which allows it to handle things like target features.
100 ;; If you want to be able to use the extra debugging information,
101 ;; therefore, be sure to keep the sources around, and run with the
102 ;; readtable configured so that the system sources can be read.
105 ;; Build SBCL with the old CMU CL low level debugger, "ldb". In the
106 ;; ideal world you would not need this unless you are messing with
107 ;; SBCL at a very low level (e.g., trying to diagnose GC problems, or
108 ;; trying to debug assembly code for a port to a new CPU). However,
109 ;; experience shows that sooner or later everyone lose()'s, in which
110 ;; case SB-LDB can at least provide an informative backtrace.
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 ;; Compile the C runtime with support for low-level debugging output
126 ;; through FSHOW and FSHOW_SIGNAL. If enabled, this feature allows
127 ;; users to turn on such debugging output using environment variables at
131 ;; Setting this makes SBCL more "fluid", i.e. more amenable to
132 ;; modification at runtime, by suppressing various INLINE declarations,
133 ;; compiler macro definitions, FREEZE-TYPE declarations; and by
134 ;; suppressing various burning-our-ships-behind-us actions after
135 ;; initialization is complete; and so forth. This tends to clobber the
136 ;; performance of the system, so unless you have some special need for
137 ;; this when hacking SBCL itself, you don't want this set.
140 ;; Enable code for collecting statistics on usage of various operations,
141 ;; useful for performance tuning of the SBCL system itself. This code
142 ;; is probably pretty stale (having not been tested since the fork from
143 ;; base CMU CL) but might nonetheless be a useful starting point for
144 ;; anyone who wants to collect such statistics in the future.
147 ;; Enable code for detecting concurrent accesses to the same hash-table
148 ;; in multiple threads. Note that this implementation is currently
149 ;; (2007-09-11) somewhat too eager: even though in the current implementation
150 ;; multiple readers are thread safe as long as there are no writers, this
151 ;; code will also trap multiple readers.
152 ; :sb-hash-table-debug
154 ;; Enabled automatically by make-config.sh for platforms which implement
155 ;; short vector SIMD intrinsics.
159 ;; Enabled automatically by make-config.sh for platforms which implement
160 ;; the %READ-CYCLE-COUNTER VOP. Can be disabled manually: affects TIME.
162 ;; FIXME: Should this be :SB-CYCLE-COUNTER instead? If so, then the same goes
163 ;; for :COMPARE-AND-SWAP-VOPS as well, and a bunch of others. Perhaps
164 ;; built-time convenience features like this should all live in eg. SB!INT
169 ;; Enabled automatically for platforms which implement complex arithmetic
170 ;; VOPs. Such platforms should implement real-complex, complex-real and
171 ;; complex-complex addition and subtractions (for complex-single-float
172 ;; and complex-double-float). They should also also implement complex-real
173 ;; and real-complex multiplication, complex-real division, and
174 ;; sb!vm::swap-complex, which swaps the real and imaginary parts.
175 ;; Finally, they should implement conjugate and complex-real, real-complex
176 ;; and complex-complex CL:= (complex-complex EQL would usually be a good
179 ; :complex-float-vops
181 ;; Enabled automatically for platforms which implement VOPs for EQL
182 ;; of single and double floats.
186 ;; Enabled automatically for platform that can implement inline constants.
188 ;; Such platform must implement 5 functions, in SB!VM:
189 ;; * canonicalize-inline-constant: converts a constant descriptor (list) into
190 ;; a canonical description, to be used as a key in an EQUAL hash table
191 ;; and to guide the generation of the constant itself.
192 ;; * inline-constant-value: given a canonical constant descriptor, computes
194 ;; 1. A label that will be used to emit the constant (usually a
196 ;; 2. A value that will be returned to code generators referring to
197 ;; the constant (on x86oids, an EA object)
198 ;; * sort-inline-constants: Receives a vector of unique constants;
199 ;; the car of each entry is the constant descriptor, and the cdr the
200 ;; corresponding label. Destructively returns a vector of constants
201 ;; sorted in emission order. It could actually perform arbitrary
202 ;; modifications to the vector, e.g. to fuse constants of different
204 ;; * emit-constant-segment-header: receives the vector of sorted constants
205 ;; and a flag (true iff speed > space). Expected to emit padding
206 ;; of some sort between the ELSEWHERE segment and the constants, or some
208 ;; * emit-inline-constant: receives a constant descriptor and its associated
209 ;; label. Emits the constant.
211 ;; Implementing this features lets VOP generators use sb!c:register-inline-constant
212 ;; to get handles (as returned by sb!vm:inline-constant-value) from constant
217 ;; Peter Van Eynde's increase-bulletproofness code for CMU CL
219 ;; Some of the code which was #+high-security before the fork has now
220 ;; been either made unconditional, deleted, or rewritten into
221 ;; unrecognizability, but some remains. What remains is not maintained
222 ;; or tested in current SBCL, but I haven't gone out of my way to
226 ; :high-security-support
228 ;; low-level thread primitives support
230 ;; As of SBCL 1.0.33.26, threads are part of the default build on
231 ;; x86oid Linux. Other platforms that support them include
232 ;; x86oid Darwin, FreeBSD, and Solaris.
237 ;; While on linux we are able to use futexes for our locking
238 ;; primitive, on other platforms we don't have this luxury.
242 ;; On some operating systems the FS segment register (used for SBCL's
243 ;; thread local storage) is not reliably preserved in signal
244 ;; handlers, so we need to restore its value from the pthread thread
246 ; :restore-fs-segment-register-from-tls
248 ;; On some x86oid operating systems (darwin) SIGTRAP is not reliably
249 ;; delivered for the INT3 instruction, so we use the UD2 instruction
250 ;; which generates SIGILL instead.
253 ;; Support for detection of unportable code (when applied to the
254 ;; COMMON-LISP package, or SBCL-internal pacakges) or bad-neighbourly
255 ;; code (when applied to user-level packages), relating to material
256 ;; alteration to packages or to bindings in symbols in packages.
259 ;; Support for the entirety of the 21-bit character space defined by
260 ;; the Unicode consortium, rather than the classical 8-bit ISO-8859-1
264 ;; Support for a full evaluator that can execute all the CL special
265 ;; forms, as opposed to the traditional SBCL evaluator which called
266 ;; COMPILE for everything complicated.
269 ;; Record source location information for variables, classes, conditions,
270 ;; packages, etc. Gives much better information on M-. in Slime, but
271 ;; increases core size by about 100kB.
274 ;; Record xref data for SBCL internals. This can be rather useful for
275 ;; people who want to develop on SBCL itself because it'll make M-?
276 ;; (slime-edit-uses) work which lists call/expansion/etc. sites.
277 ;; It'll increase the core size by major 5-6mB, though.
278 ; :sb-xref-for-internals
280 ;; We support package local nicknames. No :sb-prefix here as we vainly
281 ;; believe our API is worth copying to other implementations as well.
282 ;; This doesn't affect the build at all, merely declares how things are.
283 :package-local-nicknames
285 ;; This affects the definition of a lot of things in bignum.lisp. It
286 ;; doesn't seem to be documented anywhere what systems it might apply
287 ;; to. It doesn't seem to be needed for X86 systems anyway.
290 ;; This is set in classic CMU CL, and presumably there it means
291 ;; that the floating point arithmetic implementation
292 ;; conforms to IEEE's standard. Here it definitely means that the
293 ;; floating point arithmetic implementation conforms to IEEE's standard.
294 ;; I (WHN 19990702) haven't tried to verify
295 ;; that it does conform, but it should at least mostly conform (because
296 ;; the underlying x86 hardware tries).
299 ;; CMU CL had, and we inherited, code to support 80-bit LONG-FLOAT on the x86
300 ;; architecture. Nothing has been done to actively destroy the long float
301 ;; support, but it hasn't been thoroughly maintained, and needs at least
302 ;; some maintenance before it will work. (E.g. the LONG-FLOAT-only parts of
303 ;; genesis are still implemented in terms of unportable CMU CL functions
304 ;; which are not longer available at genesis time in SBCL.) A deeper
305 ;; problem is SBCL's bootstrap process implicitly assumes that the
306 ;; cross-compilation host will be able to make the same distinctions
307 ;; between floating point types that it does. This assumption is
308 ;; fundamentally sleazy, even though in practice it's unlikely to break down
309 ;; w.r.t. distinguishing SINGLE-FLOAT from DOUBLE-FLOAT; it's much more
310 ;; likely to break down w.r.t. distinguishing DOUBLE-FLOAT from LONG-FLOAT.
311 ;; Still it's likely to be quite doable to get LONG-FLOAT support working
312 ;; again, if anyone's sufficiently motivated.
315 ;; Some platforms don't use a 32-bit off_t by default, and thus can't
316 ;; handle files larger than 2GB. This feature will control whether
317 ;; we'll try to use platform-specific compilation options to enable a
318 ;; 64-bit off_t. The intent is for this feature to be automatically
319 ;; enabled by make-config.sh on platforms where it's needed and known
320 ;; to work, you shouldn't be enabling it manually. You might however
321 ;; want to disable it, if you need to pass file descriptors to
322 ;; foreign code that uses a 32-bit off_t.
325 ;; Enabled automatically on platforms that have VOPs to compute the
326 ;; high half of a full word-by-word multiplication. When disabled,
327 ;; SB-KERNEL:%MULTIPLY-HIGH is implemented in terms of
328 ;; SB-BIGNUM:%MULTIPLY.
329 ; :multiply-high-vops
331 ;; SBCL has optional support for zlib-based compressed core files. Enable
332 ;; this feature to compile it in. Obviously, doing so adds a dependency
334 ; :sb-core-compression
336 ;; On certain thread-enabled platforms, synchronization between threads
337 ;; for the purpose of stopping and starting the world around GC can be
338 ;; performed using safepoints instead of signals. Enable this feature
339 ;; to compile with safepoints and to use them for GC.
340 ;; (Replaces use of SIG_STOP_FOR_GC.)
343 ;; When compiling with safepoints, the INTERRUPT-THREAD mechanism can
344 ;; also use safepoints to roll the target thread to a point at which it
345 ;; can be interrupted safely, instead of using a signal for this
346 ;; purpose. Enable this feature in addition to :SB-SAFEPOINT to enable
348 ;; (Replaces use of SIGPIPE, except to wake up syscalls.)
351 ;; When compiling with safepoints and thruptions, the TIMER facility
352 ;; can replace its use of setitimer with a background thread.
353 ;; (Replaces use of SIGALRM.)
356 ;; This platform implements VOPs for %ash/right, variable-width shift right
360 ;; miscellaneous notes on other things which could have special significance
361 ;; in the *FEATURES* list
364 ;; Any target feature which affects binary compatibility of fasl files
365 ;; needs to be recorded in *FEATURES-POTENTIALLY-AFFECTING-FASL-FORMAT*
368 ;; notes on the :NIL and :IGNORE features:
370 ;; #+NIL is used to comment out forms. Occasionally #+IGNORE is used
371 ;; for this too. So don't use :NIL or :IGNORE as the names of features..
373 ;; notes on :SB-XC and :SB-XC-HOST features (which aren't controlled by this
374 ;; file, but are instead temporarily pushed onto *FEATURES* or
375 ;; *TARGET-FEATURES* during some phases of cross-compilation):
377 ;; :SB-XC-HOST stands for "cross-compilation host" and is in *FEATURES*
378 ;; during the first phase of cross-compilation bootstrapping, when the
379 ;; host Lisp is being used to compile the cross-compiler.
381 ;; :SB-XC stands for "cross compiler", and is in *FEATURES* during the second
382 ;; phase of cross-compilation bootstrapping, when the cross-compiler is
383 ;; being used to create the first target Lisp.
385 ;; notes on the :SB-ASSEMBLING feature (which isn't controlled by
388 ;; This is a flag for whether we're in the assembler. It's
389 ;; temporarily pushed onto the *FEATURES* list in the setup for
390 ;; the ASSEMBLE-FILE function. It would be a bad idea
391 ;; to use it as a name for a permanent feature.
393 ;; notes on local features (which are set automatically by the
394 ;; configuration script, and should not be set here unless you
395 ;; really, really know what you're doing):
397 ;; machine architecture features:
399 ;; any Intel 386 or better, or compatibles like the AMD K6 or K7
401 ;; any x86-64 CPU running in 64-bit mode
403 ;; DEC/Compaq Alpha CPU
405 ;; any Sun UltraSPARC (possibly also non-Ultras -- currently untested)
411 ;; any MIPS CPU (in little-endian mode with :little-endian)
413 ;; an ARM CPU (details yet to be determined)
415 ;; (CMU CL also had a :pentium feature, which affected the definition
416 ;; of some floating point vops. It was present but not enabled or
417 ;; documented in the CMU CL code that SBCL is derived from, and has
418 ;; now been moved to the backend-subfeatures mechanism.)
420 ;; properties derived from the machine architecture
421 ;; :control-stack-grows-downward-not-upward
422 ;; On the X86, the Lisp control stack grows downward. On the
423 ;; other supported CPU architectures as of sbcl-0.7.1.40, the
424 ;; system stack grows upward.
425 ;; Note that there are other stack-related differences between the
426 ;; X86 port and the other ports. E.g. on the X86, the Lisp control
427 ;; stack coincides with the C stack, meaning that on the X86 there's
428 ;; stuff on the control stack that the Lisp-level debugger doesn't
429 ;; understand very well. As of sbcl-0.7.1.40 things like that are
430 ;; just parameterized by #!+X86, but it'd probably be better to
431 ;; use new flags like :CONTROL-STACK-CONTAINS-C-STACK.
433 ;; :stack-allocatable-closures
434 ;; The compiler can allocate dynamic-extent closures on stack.
437 ;; Alien callbacks have been implemented for this platform.
439 ;; :compare-and-swap-vops
440 ;; The backend implements compare-and-swap VOPs.
442 ;; :memory-barrier-vops
443 ;; Memory barriers (for multi-threaded synchronization) have been
444 ;; implemented for this platform.
446 ;; operating system features:
447 ;; :unix = We're intended to run under some Unix-like OS. (This is not
448 ;; exclusive with the features which indicate which particular
449 ;; Unix-like OS we're intended to run under.)
450 ;; :linux = We're intended to run under some version of Linux.
451 ;; :bsd = We're intended to run under some version of BSD Unix. (This
452 ;; is not exclusive with the features which indicate which
453 ;; particular version of BSD we're intended to run under.)
454 ;; :freebsd = We're intended to run under FreeBSD.
455 ;; :openbsd = We're intended to run under OpenBSD.
456 ;; :netbsd = We're intended to run under NetBSD.
457 ;; :dragonfly = We're intended to run under DragonFly BSD.
458 ;; :darwin = We're intended to run under Darwin (including MacOS X).
459 ;; :sunos = We're intended to run under Solaris user environment
460 ;; with the SunOS kernel.
461 ;; :hpux = We're intended to run under HP-UX 11.11 or later
462 ;; :osf1 = We're intended to run under Tru64 (aka Digital Unix
464 ;; :win32 = We're intended to under some version of Microsoft Windows.
465 ;; (No others are supported by SBCL as of 1.0.8, but :hpux or :irix
466 ;; support could be ported from CMU CL if anyone is sufficiently
467 ;; motivated to do so.)