#ifndef _SBCL_RUNTIME_H_
#define _SBCL_RUNTIME_H_
-#define QSHOW 0 /* Enable low-level debugging output? */
-#if QSHOW
-#define FSHOW(args) fprintf args
-#define SHOW(string) FSHOW((stderr, "/%s\n", string))
+#if defined(LISP_FEATURE_WIN32) && defined(LISP_FEATURE_SB_THREAD)
+# include "pthreads_win32.h"
+#else
+# include <signal.h>
+# ifdef LISP_FEATURE_SB_THREAD
+# include <pthread.h>
+# endif
+#endif
+
+#include <stdint.h>
+
+#if defined(LISP_FEATURE_SB_THREAD)
+#define thread_self() pthread_self()
+#define thread_kill pthread_kill
+#define thread_sigmask pthread_sigmask
+#define thread_mutex_lock(l) pthread_mutex_lock(l)
+#define thread_mutex_unlock(l) pthread_mutex_unlock(l)
#else
-#define FSHOW(args)
-#define SHOW(string)
+#define thread_self() 0
+#define thread_kill kill_safely
+#define thread_sigmask sigprocmask
+#define thread_mutex_lock(l) 0
+#define thread_mutex_unlock(l) 0
#endif
+#if defined(LISP_FEATURE_WIN32) && defined(LISP_FEATURE_SB_THREAD)
+void os_preinit();
+#endif
+
+#if defined(LISP_FEATURE_SB_SAFEPOINT)
+
+typedef enum {
+ GC_NONE=0,
+ GC_FLIGHT,
+ GC_MESSAGE,
+ GC_INVOKED,
+ GC_QUIET,
+ GC_SETTLED,
+ GC_COLLECT,
+ GC_NPHASES
+} gc_phase_t;
+
+void map_gc_page();
+void unmap_gc_page();
+int check_pending_interrupts();
+void gc_state_lock();
+void gc_state_wait(gc_phase_t);
+void gc_state_unlock();
+
+#endif
+
+/*
+ * The next few defines serve as configuration -- edit them inline if
+ * you are a developer and want to affect FSHOW behaviour.
+ */
+
+/* Block blockable interrupts for each SHOW, if not 0.
+ * (On Windows, this setting has no effect.)
+ *
+ * In principle, this is a "configuration option", but I am not aware of
+ * any reason why or when it would be advantageous to disable it. */
+#define QSHOW_SIGNAL_SAFE 1
+
/* Enable extra-verbose low-level debugging output for signals? (You
* probably don't want this unless you're trying to debug very early
* cold boot on a new machine, or one where you've just messed up
* causes output from signal handlers, and the i/o libraries aren't
* necessarily reentrant. But it can still be very convenient for
* figuring out what's going on when you have a signal handling
- * problem.. */
-#define QSHOW_SIGNALS 0
+ * problem.
+ *
+ * Possible values are:
+ * 0 -- Never show signal-related output. There is absolutely no
+ * run-time overhead from FSHOW_SIGNAL in this case.
+ *
+ * 1 -- (recommended)
+ * Show signal-related output only if selected at run-time
+ * (otherwise almost no run-time overhead).
+ *
+ * 2 -- Unconditionally show signal-related output.
+ * Very significant overhead.
+ *
+ * For reasons of tradition, we default to 0 on POSIX and 1 on Windows
+ * through :SB-QSHOW.
+ *
+ * With option 1, set up environment variable SBCL_DYNDEBUG to include
+ * "fshow" or "fshow_signal" before starting SBCL to enable output.
+ *
+ * There is no particular advantage to option 2 except that you do not
+ * need to set environment variables in this case.
+ */
+#ifdef LISP_FEATURE_SB_QSHOW
+# define QSHOW_SIGNALS 1
+#else
+# define QSHOW_SIGNALS 0
+#endif
-/* FIXME: There seems to be no reason that LowtagOf can't be defined
- * as a (possibly inline) function instead of a macro. It would also
- * be reasonable to rename the constants in ALL CAPS. */
+/* Enable low-level debugging output, if not zero. Defaults to enabled
+ * if QSHOW_SIGNALS, disabled otherwise. Change it to 1 or 2 if you want
+ * low-level debugging output but not the whole signal mess. */
+#define QSHOW QSHOW_SIGNALS
-#define lowtag_Bits 3
-#define lowtag_Mask ((1<<lowtag_Bits)-1)
-#define LowtagOf(obj) ((obj)&lowtag_Mask)
-#define type_Bits 8
-#define type_Mask ((1<<type_Bits)-1)
+/*
+ * Configuration options end here -- the following defines do not
+ * generally need customization.
+ */
+
+#define odxprint(topic, fmt, ...) \
+ do \
+ if (dyndebug_config.dyndebug_##topic) \
+ odxprint_fun(fmt "\n", ##__VA_ARGS__); \
+ while (0)
+
+void odxprint_fun(const char *fmt, ...);
+void fshow_fun(void *ignored, const char *fmt, ...);
-/* FIXME: There seems to be no reason that TypeOf, HeaderValue,
- * Pointerp, PTR, CONS, SYMBOL, and FDEFN can't be defined
- * as (possibly inline) functions instead of macros. */
+/* Flags defined in a structure to avoid code duplication between
+ * declaration and definition. */
+extern struct dyndebug_config {
+ int dyndebug_fshow;
+ int dyndebug_fshow_signal;
+ int dyndebug_gencgc_verbose;
+ int dyndebug_safepoints;
+ int dyndebug_seh;
+ int dyndebug_misc;
+ int dyndebug_pagefaults;
+ int dyndebug_backtrace_when_lost;
+ int dyndebug_sleep_when_lost;
+ int dyndebug_io;
+ int dyndebug_runtime_link;
+} dyndebug_config;
+
+#ifdef LISP_FEATURE_GENCGC
+extern int gencgc_verbose;
+#endif
-#define TypeOf(obj) ((obj)&type_Mask)
-#define HeaderValue(obj) ((unsigned long) ((obj)>>type_Bits))
+void dyndebug_init(void);
-#define Pointerp(obj) ((obj) & 0x01)
-#define PTR(obj) ((unsigned long)((obj)&~lowtag_Mask))
+#if QSHOW_SIGNAL_SAFE == 1 && !defined(LISP_FEATURE_WIN32)
-#define CONS(obj) ((struct cons *)((obj)-type_ListPointer))
-#define SYMBOL(obj) ((struct symbol *)((obj)-type_OtherPointer))
-#define FDEFN(obj) ((struct fdefn *)((obj)-type_OtherPointer))
+extern sigset_t blockable_sigset;
+
+#define QSHOW_BLOCK \
+ sigset_t oldset; \
+ thread_sigmask(SIG_BLOCK, &blockable_sigset, &oldset)
+#define QSHOW_UNBLOCK thread_sigmask(SIG_SETMASK,&oldset,0)
+#else
+#define QSHOW_BLOCK
+#define QSHOW_UNBLOCK
+#endif
+
+/* The following macros duplicate the expansion of odxprint, because the
+ * extra level of parentheses around `args' prevents us from
+ * implementing FSHOW in terms of odxprint directly. (They also differ
+ * in a newline.)
+ */
+
+#if QSHOW
+# define FSHOW(args) \
+ do if (dyndebug_config.dyndebug_fshow) fshow_fun args; while (0)
+# define SHOW(string) FSHOW((stderr, "/%s\n", string))
+#else
+# define FSHOW(args)
+# define SHOW(string)
+#endif
+
+#if QSHOW_SIGNALS
+# define FSHOW_SIGNAL(args) \
+ do if (dyndebug_config.dyndebug_fshow_signal) fshow_fun args; while (0)
+#else
+# define FSHOW_SIGNAL(args)
+#endif
/* KLUDGE: These are in theory machine-dependent and OS-dependent, but
* in practice the "foo int" definitions work for all the machines
* that SBCL runs on as of 0.6.7. If we port to the Alpha or some
* other non-32-bit machine we'll probably need real machine-dependent
* and OS-dependent definitions again. */
-#if ((defined alpha) && !(defined __linux__))
-#error No u32,s32 definitions for this platform. Write some.
+/* even on alpha, int happens to be 4 bytes. long is longer. */
+/* FIXME: these names really shouldn't reflect their length and this
+ is not quite right for some of the FFI stuff */
+#if defined(LISP_FEATURE_WIN32)&&defined(LISP_FEATURE_X86_64)
+typedef unsigned long long u64;
+typedef signed long long s64;
#else
-/* int happens to be 4 bytes on linux/alpha. long is longer. */
+typedef unsigned long u64;
+typedef signed long s64;
+#endif
typedef unsigned int u32;
typedef signed int s32;
+
+/* this is an integral type the same length as a machine pointer */
+typedef uintptr_t pointer_sized_uint_t;
+
+#ifdef _WIN64
+#define AMD64_SYSV_ABI __attribute__((sysv_abi))
+#else
+#define AMD64_SYSV_ABI
+#endif
+
+#include <sys/types.h>
+
+#if defined(LISP_FEATURE_SB_THREAD)
+typedef pthread_t os_thread_t;
+#else
+typedef pid_t os_thread_t;
+#endif
+
+typedef uintptr_t uword_t;
+typedef intptr_t sword_t;
+
+/* FIXME: we do things this way because of the alpha32 port. once
+ alpha64 has arrived, all this nastiness can go away */
+#if 64 == N_WORD_BITS
+#define LOW_WORD(c) ((pointer_sized_uint_t)c)
+#define OBJ_FMTX "lx"
+typedef uintptr_t lispobj;
+#else
+#define OBJ_FMTX "x"
#define LOW_WORD(c) ((long)(c) & 0xFFFFFFFFL)
+/* fake it on alpha32 */
+typedef unsigned int lispobj;
#endif
-typedef u32 lispobj;
+static inline int
+lowtag_of(lispobj obj)
+{
+ return obj & LOWTAG_MASK;
+}
-/* FIXME: There seems to be no reason that make_fixnum and fixnum_value
- * can't be implemented as (possibly inline) functions. */
-#define make_fixnum(n) ((lispobj)((n)<<2))
-#define fixnum_value(n) (((long)n)>>2)
+static inline int
+widetag_of(lispobj obj)
+{
+ return obj & WIDETAG_MASK;
+}
-/* Too bad ANSI C doesn't define "bool" as C++ does.. */
-typedef int boolean;
+static inline uword_t
+HeaderValue(lispobj obj)
+{
+ return obj >> N_WIDETAG_BITS;
+}
+
+static inline struct cons *
+CONS(lispobj obj)
+{
+ return (struct cons *)(obj - LIST_POINTER_LOWTAG);
+}
-/* FIXME: There seems to be no reason that SymbolValue, SetSymbolValue,
- * and SymbolFunction can't be defined as (possibly inline) functions
- * instead of macros. */
+static inline struct symbol *
+SYMBOL(lispobj obj)
+{
+ return (struct symbol *)(obj - OTHER_POINTER_LOWTAG);
+}
-#define SymbolValue(sym) \
- (((struct symbol *)((sym)-type_OtherPointer))->value)
-#define SetSymbolValue(sym,val) \
- (((struct symbol *)((sym)-type_OtherPointer))->value = (val))
+static inline struct fdefn *
+FDEFN(lispobj obj)
+{
+ return (struct fdefn *)(obj - OTHER_POINTER_LOWTAG);
+}
-/* This only works for static symbols. */
-/* FIXME: should be called StaticSymbolFunction, right? */
-#define SymbolFunction(sym) \
- (((struct fdefn *)(SymbolValue(sym)-type_OtherPointer))->function)
+/* Is the Lisp object obj something with pointer nature (as opposed to
+ * e.g. a fixnum or character or unbound marker)? */
+static inline int
+is_lisp_pointer(lispobj obj)
+{
+#if N_WORD_BITS == 64
+ return (obj & 3) == 3;
+#else
+ return obj & 1;
+#endif
+}
+
+#include "fixnump.h"
+
+/* Is the Lisp object obj something with immediate nature (e.g. a
+ * fixnum or character or unbound marker)? */
+static inline int
+is_lisp_immediate(lispobj obj)
+{
+ return (fixnump(obj)
+ || (widetag_of(obj) == CHARACTER_WIDETAG)
+#if N_WORD_BITS == 64
+ || (widetag_of(obj) == SINGLE_FLOAT_WIDETAG)
+#endif
+ || (widetag_of(obj) == UNBOUND_MARKER_WIDETAG));
+}
+
+/* Convert from a lispobj with type bits to a native (ordinary
+ * C/assembly) pointer to the beginning of the object. */
+static inline lispobj *
+native_pointer(lispobj obj)
+{
+ return (lispobj *) ((pointer_sized_uint_t) (obj & ~LOWTAG_MASK));
+}
+
+/* inverse operation: create a suitably tagged lispobj from a native
+ * pointer or integer.*/
+static inline lispobj
+make_lispobj(void *o, int low_tag)
+{
+ return LOW_WORD(o) | low_tag;
+}
+
+#define MAKE_FIXNUM(n) (n << N_FIXNUM_TAG_BITS)
+static inline lispobj
+make_fixnum(sword_t n)
+{
+ return MAKE_FIXNUM(n);
+}
+
+static inline sword_t
+fixnum_value(lispobj n)
+{
+ return n >> N_FIXNUM_TAG_BITS;
+}
+
+#if defined(LISP_FEATURE_WIN32)
+/* KLUDGE: Avoid double definition of boolean by rpcndr.h included via
+ * shlobj.h.
+ *
+ * FIXME: We should probably arrange to use the rpcndr.h boolean on Windows,
+ * or get rid of our own boolean type. If the boolean type is only used in
+ * the runtime, and never passed to Lisp, then it doesn't matter which one
+ * we use.
+ */
+#define boolean rpcndr_boolean
+#include <shlobj.h>
+#undef boolean
+#endif
+typedef int boolean;
+
+static inline boolean
+other_immediate_lowtag_p(lispobj header)
+{
+ /* These lowtags are spaced 4 apart throughout the lowtag space. */
+ return (lowtag_of(header) & 3) == OTHER_IMMEDIATE_0_LOWTAG;
+}
/* KLUDGE: As far as I can tell there's no ANSI C way of saying
* "this function never returns". This is the way that you do it
- * in GCC later than version 2.7 or so. If you are using some
- * compiler that doesn't understand this, you could could just
- * change it to "typedef void never_returns" and nothing would
- * break, you might just get a few more bytes of compiled code or
- * a few more compiler warnings. -- WHN 2000-10-21 */
-typedef volatile void never_returns;
+ * in GCC later than version 2.5 or so. */
+#if defined(__GNUC__)
+#if __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 5)
+#define never_returns __attribute__ ((noreturn))
+#else
+#define never_returns
+#endif
+#else
+#define never_returns
+#endif
+
+extern void *successful_malloc (size_t size);
+extern char *copied_string (char *string);
+
+#if defined(LISP_FEATURE_SB_THREAD) && !defined(LISP_FEATURE_SB_SAFEPOINT)
+# define THREADS_USING_GCSIGNAL 1
+#endif
+
+/* Now that SPARC has precise GENCGC, several places that used to be
+ * #ifdef PCC need adjustment. Clearly, "PPC or SPARC" is as unhelpful
+ * a test as its reverse, "x86 or x86-64". However, the feature
+ * commonly used to differentiate between those two worlds is
+ * C_STACK_IS_CONTROL_STACK, and clearly (or at least in my humble
+ * opinion), at some point we'd like to have precise GC on x86 while
+ * still sharing the C stack, so stack usage ought not imply GC
+ * conservativeness. So let's have a helper feature that makes the code
+ * a bit more future-proof, even if it is itself currently defined in
+ * the naive way: */
+#if defined(LISP_FEATURE_GENCGC) && !defined(LISP_FEATURE_C_STACK_IS_CONTROL_STACK)
+# define GENCGC_IS_PRECISE 1
+#endif
+
+void *os_dlsym_default(char *name);
#endif /* _SBCL_RUNTIME_H_ */
projects / sbcl.git / blobdiff