os_dep.c

linux下建立JAVA虚拟机的源码KAFFE

/*
 * Copyright 1988, 1989 Hans-J. Boehm, Alan J. Demers
 * Copyright (c) 1991-1995 by Xerox Corporation.  All rights reserved.
 * Copyright (c) 1996-1999 by Silicon Graphics.  All rights reserved.
 * Copyright (c) 1999 by Hewlett-Packard Company.  All rights reserved.
 *
 * THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED
 * OR IMPLIED.  ANY USE IS AT YOUR OWN RISK.
 *
 * Permission is hereby granted to use or copy this program
 * for any purpose,  provided the above notices are retained on all copies.
 * Permission to modify the code and to distribute modified code is granted,
 * provided the above notices are retained, and a notice that the code was
 * modified is included with the above copyright notice.
 */

# include "private/gc_priv.h"

# if defined(LINUX) && !defined(POWERPC)
#   include <linux/version.h>
#   if (LINUX_VERSION_CODE <= 0x10400)
      /* Ugly hack to get struct sigcontext_struct definition.  Required      */
      /* for some early 1.3.X releases.  Will hopefully go away soon. */
      /* in some later Linux releases, asm/sigcontext.h may have to   */
      /* be included instead.                                         */
#     define __KERNEL__
#     include <asm/signal.h>
#     undef __KERNEL__
#   else
      /* Kernels prior to 2.1.1 defined struct sigcontext_struct instead of */
      /* struct sigcontext.  libc6 (glibc2) uses "struct sigcontext" in     */
      /* prototypes, so we have to include the top-level sigcontext.h to    */
      /* make sure the former gets defined to be the latter if appropriate. */
#     include <features.h>
#     if 2 <= __GLIBC__
#       if 2 == __GLIBC__ && 0 == __GLIBC_MINOR__
	  /* glibc 2.1 no longer has sigcontext.h.  But signal.h	*/
	  /* has the right declaration for glibc 2.1.			*/
#         include <sigcontext.h>
#       endif /* 0 == __GLIBC_MINOR__ */
#     else /* not 2 <= __GLIBC__ */
        /* libc5 doesn't have <sigcontext.h>: go directly with the kernel   */
        /* one.  Check LINUX_VERSION_CODE to see which we should reference. */
#       include <asm/sigcontext.h>
#     endif /* 2 <= __GLIBC__ */
#   endif
# endif
# if !defined(OS2) && !defined(PCR) && !defined(AMIGA) && !defined(MACOS) \
    && !defined(MSWINCE)
#   include <sys/types.h>
#   if !defined(MSWIN32) && !defined(SUNOS4)
#   	include <unistd.h>
#   endif
# endif

# include <stdio.h>
# if defined(MSWINCE)
#   define SIGSEGV 0 /* value is irrelevant */
# else
#   include <signal.h>
# endif

#if defined(LINUX) || defined(LINUX_STACKBOTTOM)
# include <ctype.h>
#endif

/* Blatantly OS dependent routines, except for those that are related 	*/
/* to dynamic loading.							*/

# if defined(HEURISTIC2) || defined(SEARCH_FOR_DATA_START)
#   define NEED_FIND_LIMIT
# endif

# if !defined(STACKBOTTOM) && defined(HEURISTIC2)
#   define NEED_FIND_LIMIT
# endif

# if (defined(SUNOS4) && defined(DYNAMIC_LOADING)) && !defined(PCR)
#   define NEED_FIND_LIMIT
# endif

# if (defined(SVR4) || defined(AUX) || defined(DGUX) \
      || (defined(LINUX) && defined(SPARC))) && !defined(PCR)
#   define NEED_FIND_LIMIT
# endif

#if defined(FREEBSD) && (defined(I386) || defined(powerpc) || defined(__powerpc__))
#  include <machine/trap.h>
#  if !defined(PCR)
#    define NEED_FIND_LIMIT
#  endif
#endif

#if (defined(NETBSD) || defined(OPENBSD)) && defined(__ELF__) \
    && !defined(NEED_FIND_LIMIT)
   /* Used by GC_init_netbsd_elf() below.	*/
#  define NEED_FIND_LIMIT
#endif

#ifdef NEED_FIND_LIMIT
#   include <setjmp.h>
#endif

#ifdef AMIGA
# define GC_AMIGA_DEF
# include "AmigaOS.c"
# undef GC_AMIGA_DEF
#endif

#if defined(MSWIN32) || defined(MSWINCE)
# define WIN32_LEAN_AND_MEAN
# define NOSERVICE
# include <windows.h>
#endif

#ifdef MACOS
# include <Processes.h>
#endif

#ifdef IRIX5
# include <sys/uio.h>
# include <malloc.h>   /* for locking */
#endif
#if defined(USE_MMAP) || defined(USE_MUNMAP)
# ifndef USE_MMAP
    --> USE_MUNMAP requires USE_MMAP
# endif
# include <sys/types.h>
# include <sys/mman.h>
# include <sys/stat.h>
# include <errno.h>
#endif

#ifdef UNIX_LIKE
# include <fcntl.h>
# if defined(SUNOS5SIGS) && !defined(FREEBSD)
#  include <sys/siginfo.h>
# endif
  /* Define SETJMP and friends to be the version that restores	*/
  /* the signal mask.						*/
# define SETJMP(env) sigsetjmp(env, 1)
# define LONGJMP(env, val) siglongjmp(env, val)
# define JMP_BUF sigjmp_buf
#else
# define SETJMP(env) setjmp(env)
# define LONGJMP(env, val) longjmp(env, val)
# define JMP_BUF jmp_buf
#endif

#ifdef DARWIN
/* for get_etext and friends */
#include <mach-o/getsect.h>
#endif

#ifdef DJGPP
  /* Apparently necessary for djgpp 2.01.  May cause problems with	*/
  /* other versions.							*/
  typedef long unsigned int caddr_t;
#endif

#ifdef PCR
# include "il/PCR_IL.h"
# include "th/PCR_ThCtl.h"
# include "mm/PCR_MM.h"
#endif

#if !defined(NO_EXECUTE_PERMISSION)
# define OPT_PROT_EXEC PROT_EXEC
#else
# define OPT_PROT_EXEC 0
#endif

#if defined(LINUX) && \
    (defined(USE_PROC_FOR_LIBRARIES) || defined(IA64) || !defined(SMALL_CONFIG))

/* We need to parse /proc/self/maps, either to find dynamic libraries,	*/
/* and/or to find the register backing store base (IA64).  Do it once	*/
/* here.								*/

#define READ read

/* Repeatedly perform a read call until the buffer is filled or	*/
/* we encounter EOF.						*/
ssize_t GC_repeat_read(int fd, char *buf, size_t count)
{
    ssize_t num_read = 0;
    ssize_t result;
    
    while (num_read < count) {
	result = READ(fd, buf + num_read, count - num_read);
	if (result < 0) return result;
	if (result == 0) break;
	num_read += result;
    }
    return num_read;
}

/*
 * Apply fn to a buffer containing the contents of /proc/self/maps.
 * Return the result of fn or, if we failed, 0.
 * We currently do nothing to /proc/self/maps other than simply read
 * it.  This code could be simplified if we could determine its size
 * ahead of time.
 */

word GC_apply_to_maps(word (*fn)(char *))
{
    int f;
    int result;
    size_t maps_size = 4000;  /* Initial guess. 	*/
    static char init_buf[1];
    static char *maps_buf = init_buf;
    static size_t maps_buf_sz = 1;

    /* Read /proc/self/maps, growing maps_buf as necessary.	*/
        /* Note that we may not allocate conventionally, and	*/
        /* thus can't use stdio.				*/
	do {
	    if (maps_size >= maps_buf_sz) {
	      /* Grow only by powers of 2, since we leak "too small" buffers. */
	      while (maps_size >= maps_buf_sz) maps_buf_sz *= 2;
	      maps_buf = GC_scratch_alloc(maps_buf_sz);
	      if (maps_buf == 0) return 0;
	    }
	    f = open("/proc/self/maps", O_RDONLY);
	    if (-1 == f) return 0;
	    maps_size = 0;
	    do {
	        result = GC_repeat_read(f, maps_buf, maps_buf_sz-1);
	        if (result <= 0) return 0;
	        maps_size += result;
	    } while (result == maps_buf_sz-1);
	    close(f);
	} while (maps_size >= maps_buf_sz);
        maps_buf[maps_size] = '\0';
	
    /* Apply fn to result. */
	return fn(maps_buf);
}

#endif /* Need GC_apply_to_maps */

#if defined(LINUX) && (defined(USE_PROC_FOR_LIBRARIES) || defined(IA64))
//
//  GC_parse_map_entry parses an entry from /proc/self/maps so we can
//  locate all writable data segments that belong to shared libraries.
//  The format of one of these entries and the fields we care about
//  is as follows:
//  XXXXXXXX-XXXXXXXX r-xp 00000000 30:05 260537     name of mapping...\n
//  ^^^^^^^^ ^^^^^^^^ ^^^^          ^^
//  start    end      prot          maj_dev
//
//  Note that since about auguat 2003 kernels, the columns no longer have
//  fixed offsets on 64-bit kernels.  Hence we no longer rely on fixed offsets
//  anywhere, which is safer anyway.
//

/*
 * Assign various fields of the first line in buf_ptr to *start, *end,
 * *prot_buf and *maj_dev.  Only *prot_buf may be set for unwritable maps.
 */
char *GC_parse_map_entry(char *buf_ptr, word *start, word *end,
                                char *prot_buf, unsigned int *maj_dev)
{
    char *start_start, *end_start, *prot_start, *maj_dev_start;
    char *p;
    char *endp;

    if (buf_ptr == NULL || *buf_ptr == '\0') {
        return NULL;
    }

    p = buf_ptr;
    while (isspace(*p)) ++p;
    start_start = p;
    GC_ASSERT(isxdigit(*start_start));
    *start = strtoul(start_start, &endp, 16); p = endp;
    GC_ASSERT(*p=='-');

    ++p;
    end_start = p;
    GC_ASSERT(isxdigit(*end_start));
    *end = strtoul(end_start, &endp, 16); p = endp;
    GC_ASSERT(isspace(*p));

    while (isspace(*p)) ++p;
    prot_start = p;
    GC_ASSERT(*prot_start == 'r' || *prot_start == '-');
    memcpy(prot_buf, prot_start, 4);
    prot_buf[4] = '\0';
    if (prot_buf[1] == 'w') {/* we can skip the rest if it's not writable. */
	/* Skip past protection field to offset field */
          while (!isspace(*p)) ++p; while (isspace(*p)) ++p;
          GC_ASSERT(isxdigit(*p));
	/* Skip past offset field, which we ignore */
          while (!isspace(*p)) ++p; while (isspace(*p)) ++p;
	maj_dev_start = p;
        GC_ASSERT(isxdigit(*maj_dev_start));
        *maj_dev = strtoul(maj_dev_start, NULL, 16);
    }

    while (*p && *p++ != '\n');

    return p;
}

#endif /* Need to parse /proc/self/maps. */	

#if defined(SEARCH_FOR_DATA_START)
  /* The I386 case can be handled without a search.  The Alpha case	*/
  /* used to be handled differently as well, but the rules changed	*/
  /* for recent Linux versions.  This seems to be the easiest way to	*/
  /* cover all versions.						*/

# ifdef LINUX
    /* Some Linux distributions arrange to define __data_start.  Some	*/
    /* define data_start as a weak symbol.  The latter is technically	*/
    /* broken, since the user program may define data_start, in which	*/
    /* case we lose.  Nonetheless, we try both, prefering __data_start.	*/
    /* We assume gcc-compatible pragmas.	*/
#   pragma weak __data_start
    extern int __data_start[];
#   pragma weak data_start
    extern int data_start[];
# endif /* LINUX */
  extern int _end[];

  ptr_t GC_data_start;

  void GC_init_linux_data_start()
  {
    extern ptr_t GC_find_limit();

#   ifdef LINUX
      /* Try the easy approaches first:	*/
      if ((ptr_t)__data_start != 0) {
	  GC_data_start = (ptr_t)(__data_start);
	  return;
      }
      if ((ptr_t)data_start != 0) {
	  GC_data_start = (ptr_t)(data_start);
	  return;
      }
#   endif /* LINUX */
    GC_data_start = GC_find_limit((ptr_t)(_end), FALSE);
  }
#endif

# ifdef ECOS

# ifndef ECOS_GC_MEMORY_SIZE
# define ECOS_GC_MEMORY_SIZE (448 * 1024)
# endif /* ECOS_GC_MEMORY_SIZE */

// setjmp() function, as described in ANSI para 7.6.1.1
#undef SETJMP
#define SETJMP( __env__ )  hal_setjmp( __env__ )

// FIXME: This is a simple way of allocating memory which is
// compatible with ECOS early releases.  Later releases use a more
// sophisticated means of allocating memory than this simple static
// allocator, but this method is at least bound to work.
static char memory[ECOS_GC_MEMORY_SIZE];
static char *brk = memory;

static void *tiny_sbrk(ptrdiff_t increment)
{
  void *p = brk;

  brk += increment;

  if (brk >  memory + sizeof memory)
    {
      brk -= increment;
      return NULL;
    }

  return p;
}
#define sbrk tiny_sbrk
# endif /* ECOS */

#if (defined(NETBSD) || defined(OPENBSD)) && defined(__ELF__)
  ptr_t GC_data_start;

  void GC_init_netbsd_elf()
  {
    extern ptr_t GC_find_limit();
    extern char **environ;
	/* This may need to be environ, without the underscore, for	*/
	/* some versions.						*/
    GC_data_start = GC_find_limit((ptr_t)&environ, FALSE);
  }
#endif

# ifdef OS2

# include <stddef.h>

# if !defined(__IBMC__) && !defined(__WATCOMC__) /* e.g. EMX */

struct exe_hdr {
    unsigned short      magic_number;
    unsigned short      padding[29];
    long                new_exe_offset;
};

#define E_MAGIC(x)      (x).magic_number
#define EMAGIC          0x5A4D  
#define E_LFANEW(x)     (x).new_exe_offset

struct e32_exe {
    unsigned char       magic_number[2]; 
    unsigned char       byte_order; 
    unsigned char       word_order; 
    unsigned long       exe_format_level;
    unsigned short      cpu;       
    unsigned short      os;
    unsigned long       padding1[13];
    unsigned long       object_table_offset;
    unsigned long       object_count;    
    unsigned long       padding2[31];
};

#define E32_MAGIC1(x)   (x).magic_number[0]
#define E32MAGIC1       'L'
#define E32_MAGIC2(x)   (x).magic_number[1]
#define E32MAGIC2       'X'
#define E32_BORDER(x)   (x).byte_order
#define E32LEBO         0
#define E32_WORDER(x)   (x).word_order
#define E32LEWO         0
#define E32_CPU(x)      (x).cpu
#define E32CPU286       1
#define E32_OBJTAB(x)   (x).object_table_offset
#define E32_OBJCNT(x)   (x).object_count

struct o32_obj {
    unsigned long       size;  
    unsigned long       base;
    unsigned long       flags;  
    unsigned long       pagemap;