Linker.c 220 KB
Newer Older
1 2
/* -----------------------------------------------------------------------------
 *
Gabor Greif's avatar
typo  
Gabor Greif committed
3
 * (c) The GHC Team, 2000-2012
4 5 6 7 8
 *
 * RTS Object Linker
 *
 * ---------------------------------------------------------------------------*/

sof's avatar
sof committed
9
#if 0
10
#include "PosixSource.h"
sof's avatar
sof committed
11
#endif
12

13
/* Linux needs _GNU_SOURCE to get RTLD_DEFAULT from <dlfcn.h> and
14 15
   MREMAP_MAYMOVE from <sys/mman.h>.
 */
16 17
#if defined(__linux__)  || defined(__GLIBC__)
#define _GNU_SOURCE 1
18 19
#endif

20 21
#include "Rts.h"
#include "HsFFI.h"
Simon Marlow's avatar
Simon Marlow committed
22 23

#include "sm/Storage.h"
Simon Marlow's avatar
Simon Marlow committed
24
#include "Stats.h"
25
#include "Hash.h"
26
#include "LinkerInternals.h"
27
#include "RtsUtils.h"
28
#include "Trace.h"
Simon Marlow's avatar
Simon Marlow committed
29
#include "StgPrimFloat.h" // for __int_encodeFloat etc.
30
#include "Proftimer.h"
31
#include "GetEnv.h"
32
#include "Stable.h"
33
#include "RtsSymbols.h"
Simon Marlow's avatar
Simon Marlow committed
34 35 36 37

#if !defined(mingw32_HOST_OS)
#include "posix/Signals.h"
#endif
38

Simon Marlow's avatar
Simon Marlow committed
39 40 41
// get protos for is*()
#include <ctype.h>

42
#ifdef HAVE_SYS_TYPES_H
43
#include <sys/types.h>
44 45
#endif

Ian Lynagh's avatar
Ian Lynagh committed
46
#include <inttypes.h>
47 48
#include <stdlib.h>
#include <string.h>
49 50
#include <stdio.h>
#include <assert.h>
Simon Marlow's avatar
Simon Marlow committed
51
#include <libgen.h>
52

53
#ifdef HAVE_SYS_STAT_H
54
#include <sys/stat.h>
55
#endif
56

57
#if defined(HAVE_DLFCN_H)
58
#include <dlfcn.h>
59
#endif
60

61 62 63
#if (defined(powerpc_HOST_ARCH) && defined(linux_HOST_OS)) \
 || (!defined(powerpc_HOST_ARCH) && \
    (   defined(linux_HOST_OS)     || defined(freebsd_HOST_OS) || \
64 65
        defined(dragonfly_HOST_OS) || defined(netbsd_HOST_OS ) || \
        defined(openbsd_HOST_OS  ) || defined(darwin_HOST_OS ) || \
66 67
        defined(kfreebsdgnu_HOST_OS) || defined(gnu_HOST_OS  ) || \
        defined(solaris2_HOST_OS)))
68
/* Don't use mmap on powerpc/darwin as the mmap there doesn't support
69 70 71 72
 * reallocating but we need to allocate jump islands just after each
 * object images. Otherwise relative branches to jump islands can fail
 * due to 24-bits displacement overflow.
 */
73
#define USE_MMAP 1
74 75
#include <fcntl.h>
#include <sys/mman.h>
dons's avatar
dons committed
76 77 78 79 80

#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif

81 82 83 84
#else

#define USE_MMAP 0

85 86
#endif

87 88 89 90

/* PowerPC has relative branch instructions with only 24 bit displacements
 * and therefore needs jump islands contiguous with each object code module.
 */
91
#if (USE_MMAP && defined(powerpc_HOST_ARCH) && defined(linux_HOST_OS))
92 93 94 95 96
#define USE_CONTIGUOUS_MMAP 1
#else
#define USE_CONTIGUOUS_MMAP 0
#endif

pcapriotti's avatar
pcapriotti committed
97
#if defined(linux_HOST_OS) || defined(solaris2_HOST_OS) || defined(freebsd_HOST_OS) || defined(kfreebsdgnu_HOST_OS) || defined(dragonfly_HOST_OS) || defined(netbsd_HOST_OS) || defined(openbsd_HOST_OS) || defined(gnu_HOST_OS)
98
#  define OBJFORMAT_ELF
Ian Lynagh's avatar
Ian Lynagh committed
99 100
#  include <regex.h>    // regex is already used by dlopen() so this is OK
                        // to use here without requiring an additional lib
101
#elif defined (mingw32_HOST_OS)
102
#  define OBJFORMAT_PEi386
103
#  include <windows.h>
104
#  include <shfolder.h> /* SHGetFolderPathW */
sof's avatar
sof committed
105
#  include <math.h>
106
#elif defined(darwin_HOST_OS)
107
#  define OBJFORMAT_MACHO
108
#  include <regex.h>
Ian Lynagh's avatar
Ian Lynagh committed
109 110
#  include <mach/machine.h>
#  include <mach-o/fat.h>
111 112 113
#  include <mach-o/loader.h>
#  include <mach-o/nlist.h>
#  include <mach-o/reloc.h>
114 115 116
#if defined(powerpc_HOST_ARCH)
#  include <mach-o/ppc/reloc.h>
#endif
117 118 119
#if defined(x86_64_HOST_ARCH)
#  include <mach-o/x86_64/reloc.h>
#endif
120 121
#endif

122 123 124 125
#if defined(x86_64_HOST_ARCH) && defined(darwin_HOST_OS)
#define ALWAYS_PIC
#endif

126 127 128 129
#if defined(dragonfly_HOST_OS)
#include <sys/tls.h>
#endif

130 131 132 133 134 135 136
typedef struct _RtsSymbolInfo {
    void *value;
    const ObjectCode *owner;
    HsBool weak;
} RtsSymbolInfo;

/* Hash table mapping symbol names to RtsSymbolInfo */
137
static /*Str*/HashTable *symhash;
138

139
/* List of currently loaded objects */
Ian Lynagh's avatar
Ian Lynagh committed
140
ObjectCode *objects = NULL;     /* initially empty */
141

142 143 144 145
/* List of objects that have been unloaded via unloadObj(), but are waiting
   to be actually freed via checkUnload() */
ObjectCode *unloaded_objects = NULL; /* initially empty */

146
#ifdef THREADED_RTS
147
/* This protects all the Linker's global state except unloaded_objects */
148
Mutex linker_mutex;
149 150 151 152
/*
 * This protects unloaded_objects.  We have a separate mutex for this, because
 * the GC needs to access unloaded_objects in checkUnload, while the linker only
 * needs to access unloaded_objects in unloadObj(), so this allows most linker
153
 * operations proceed concurrently with the GC.
154 155
 */
Mutex linker_unloaded_mutex;
156 157
#endif

158 159 160
/* Type of the initializer */
typedef void (*init_t) (int argc, char **argv, char **env);

161
static HsInt isAlreadyLoaded( pathchar *path );
162
static HsInt loadOc( ObjectCode* oc );
163
static ObjectCode* mkOc( pathchar *path, char *image, int imageSize,
Erik de Castro Lopo's avatar
Erik de Castro Lopo committed
164 165
                         rtsBool mapped, char *archiveMemberName,
                         int misalignment
166 167
                       );

168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199
// Use wchar_t for pathnames on Windows (#5697)
#if defined(mingw32_HOST_OS)
#define pathcmp wcscmp
#define pathlen wcslen
#define pathopen _wfopen
#define pathstat _wstat
#define struct_stat struct _stat
#define open wopen
#define WSTR(s) L##s
#else
#define pathcmp strcmp
#define pathlen strlen
#define pathopen fopen
#define pathstat stat
#define struct_stat struct stat
#define WSTR(s) s
#endif

static pathchar* pathdup(pathchar *path)
{
    pathchar *ret;
#if defined(mingw32_HOST_OS)
    ret = wcsdup(path);
#else
    /* sigh, strdup() isn't a POSIX function, so do it the long way */
    ret = stgMallocBytes( strlen(path)+1, "loadObj" );
    strcpy(ret, path);
#endif
    return ret;
}


200
#if defined(OBJFORMAT_ELF)
201 202 203
static int ocVerifyImage_ELF    ( ObjectCode* oc );
static int ocGetNames_ELF       ( ObjectCode* oc );
static int ocResolve_ELF        ( ObjectCode* oc );
204
static int ocRunInit_ELF        ( ObjectCode* oc );
205
#if NEED_SYMBOL_EXTRAS
206
static int ocAllocateSymbolExtras_ELF ( ObjectCode* oc );
207
#endif
208
#elif defined(OBJFORMAT_PEi386)
209 210 211
static int ocVerifyImage_PEi386 ( ObjectCode* oc );
static int ocGetNames_PEi386    ( ObjectCode* oc );
static int ocResolve_PEi386     ( ObjectCode* oc );
212
static int ocRunInit_PEi386     ( ObjectCode* oc );
213
static void *lookupSymbolInDLLs ( unsigned char *lbl );
214 215 216 217
/* See Note [mingw-w64 name decoration scheme] */
#ifndef x86_64_HOST_ARCH
 static void zapTrailingAtSign   ( unsigned char *sym );
#endif
218
static char *allocateImageAndTrampolines (
Austin Seipp's avatar
Austin Seipp committed
219
   pathchar* arch_name, char* member_name,
220
#if defined(x86_64_HOST_ARCH)
Austin Seipp's avatar
Austin Seipp committed
221
   FILE* f,
222 223 224 225 226 227 228 229 230
#endif
   int size );
#if defined(x86_64_HOST_ARCH)
static int ocAllocateSymbolExtras_PEi386 ( ObjectCode* oc );
static size_t makeSymbolExtra_PEi386( ObjectCode* oc, size_t, char* symbol );
#define PEi386_IMAGE_OFFSET 4
#else
#define PEi386_IMAGE_OFFSET 0
#endif
231 232 233 234
#elif defined(OBJFORMAT_MACHO)
static int ocVerifyImage_MachO    ( ObjectCode* oc );
static int ocGetNames_MachO       ( ObjectCode* oc );
static int ocResolve_MachO        ( ObjectCode* oc );
235
static int ocRunInit_MachO        ( ObjectCode* oc );
236

237
#if (USE_MMAP == 0)
238
static int machoGetMisalignment( FILE * );
Ian Lynagh's avatar
Ian Lynagh committed
239
#endif
240
#if NEED_SYMBOL_EXTRAS
241 242
static int ocAllocateSymbolExtras_MachO ( ObjectCode* oc );
#endif
243
#ifdef powerpc_HOST_ARCH
244
static void machoInitSymbolsWithoutUnderscore( void );
245
#endif
246
#endif
247

248 249
static void freeProddableBlocks (ObjectCode *oc);

250
#if USE_MMAP
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277
/**
 * An allocated page being filled by the allocator
 */
struct m32_alloc_t {
   void * base_addr;             // Page address
   unsigned int current_size;    // Number of bytes already reserved
};

#define M32_MAX_PAGES 32

/**
 * Allocator
 *
 * Currently an allocator is just a set of pages being filled. The maximum
 * number of pages can be configured with M32_MAX_PAGES.
 */
typedef struct m32_allocator_t {
   struct m32_alloc_t pages[M32_MAX_PAGES];
} * m32_allocator;

// We use a global memory allocator
static struct m32_allocator_t allocator;

struct m32_allocator_t;
static void m32_allocator_init(struct m32_allocator_t *m32);
#endif

278 279 280 281 282 283 284 285 286 287 288 289 290
/* on x86_64 we have a problem with relocating symbol references in
 * code that was compiled without -fPIC.  By default, the small memory
 * model is used, which assumes that symbol references can fit in a
 * 32-bit slot.  The system dynamic linker makes this work for
 * references to shared libraries by either (a) allocating a jump
 * table slot for code references, or (b) moving the symbol at load
 * time (and copying its contents, if necessary) for data references.
 *
 * We unfortunately can't tell whether symbol references are to code
 * or data.  So for now we assume they are code (the vast majority
 * are), and allocate jump-table slots.  Unfortunately this will
 * SILENTLY generate crashing code for data references.  This hack is
 * enabled by X86_64_ELF_NONPIC_HACK.
Ian Lynagh's avatar
Ian Lynagh committed
291
 *
292 293 294 295 296 297 298 299
 * One workaround is to use shared Haskell libraries.  This is
 * coming.  Another workaround is to keep the static libraries but
 * compile them with -fPIC, because that will generate PIC references
 * to data which can be relocated.  The PIC code is still too green to
 * do this systematically, though.
 *
 * See bug #781
 * See thread http://www.haskell.org/pipermail/cvs-ghc/2007-September/038458.html
300 301 302 303 304 305 306 307 308 309 310 311
 *
 * Naming Scheme for Symbol Macros
 *
 * SymI_*: symbol is internal to the RTS. It resides in an object
 *         file/library that is statically.
 * SymE_*: symbol is external to the RTS library. It might be linked
 *         dynamically.
 *
 * Sym*_HasProto  : the symbol prototype is imported in an include file
 *                  or defined explicitly
 * Sym*_NeedsProto: the symbol is undefined and we add a dummy
 *                  default proto extern void sym(void);
312 313
 */
#define X86_64_ELF_NONPIC_HACK 1
314

315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
/* Link objects into the lower 2Gb on x86_64.  GHC assumes the
 * small memory model on this architecture (see gcc docs,
 * -mcmodel=small).
 *
 * MAP_32BIT not available on OpenBSD/amd64
 */
#if defined(x86_64_HOST_ARCH) && defined(MAP_32BIT)
#define TRY_MAP_32BIT MAP_32BIT
#else
#define TRY_MAP_32BIT 0
#endif

/*
 * Due to the small memory model (see above), on x86_64 we have to map
 * all our non-PIC object files into the low 2Gb of the address space
 * (why 2Gb and not 4Gb?  Because all addresses must be reachable
 * using a 32-bit signed PC-relative offset). On Linux we can do this
 * using the MAP_32BIT flag to mmap(), however on other OSs
 * (e.g. *BSD, see #2063, and also on Linux inside Xen, see #2512), we
 * can't do this.  So on these systems, we have to pick a base address
 * in the low 2Gb of the address space and try to allocate memory from
 * there.
 *
 * We pick a default address based on the OS, but also make this
 * configurable via an RTS flag (+RTS -xm)
 */
341
#if !defined(ALWAYS_PIC) && defined(x86_64_HOST_ARCH)
342 343 344 345 346 347 348 349 350

#if defined(MAP_32BIT)
// Try to use MAP_32BIT
#define MMAP_32BIT_BASE_DEFAULT 0
#else
// A guess: 1Gb.
#define MMAP_32BIT_BASE_DEFAULT 0x40000000
#endif

Ian Lynagh's avatar
Ian Lynagh committed
351
static void *mmap_32bit_base = (void *)MMAP_32BIT_BASE_DEFAULT;
352 353 354 355 356 357 358
#endif

/* MAP_ANONYMOUS is MAP_ANON on some systems, e.g. OpenBSD */
#if !defined(MAP_ANONYMOUS) && defined(MAP_ANON)
#define MAP_ANONYMOUS MAP_ANON
#endif

359 360
/* -----------------------------------------------------------------------------
 * Insert symbols into hash tables, checking for duplicates.
361 362
 *
 * Returns: 0 on failure, nonzero on success
363
 */
364

365
static int ghciInsertSymbolTable(
366 367
   pathchar* obj_name,
   HashTable *table,
368
   const char* key,
369 370 371
   void *data,
   HsBool weak,
   ObjectCode *owner)
372
{
373 374 375 376 377 378 379 380
   RtsSymbolInfo *pinfo = lookupStrHashTable(table, key);
   if (!pinfo) /* new entry */
   {
      pinfo = stgMallocBytes(sizeof (*pinfo), "ghciInsertToSymbolTable");
      pinfo->value = data;
      pinfo->owner = owner;
      pinfo->weak = weak;
      insertStrHashTable(table, key, pinfo);
381 382 383 384 385 386 387
      return 1;
   }
   else if ((!pinfo->weak || pinfo->value) && weak)
   {
     return 1; /* duplicate weak symbol, throw it away */
   }
   else if (pinfo->weak) /* weak symbol is in the table */
Simon Marlow's avatar
Simon Marlow committed
388
   {
389 390 391 392
      /* override the weak definition with the non-weak one */
      pinfo->value = data;
      pinfo->owner = owner;
      pinfo->weak = HS_BOOL_FALSE;
393
      return 1;
Simon Marlow's avatar
Simon Marlow committed
394 395
   }
   debugBelch(
396
      "GHC runtime linker: fatal error: I found a duplicate definition for symbol\n"
397 398
      "   %s\n"
      "whilst processing object file\n"
399
      "   %" PATH_FMT "\n"
400 401 402 403
      "This could be caused by:\n"
      "   * Loading two different object files which export the same symbol\n"
      "   * Specifying the same object file twice on the GHCi command line\n"
      "   * An incorrect `package.conf' entry, causing some object to be\n"
404
      "     loaded twice.\n",
405 406
      (char*)key,
      obj_name
Simon Marlow's avatar
Simon Marlow committed
407
   );
408
   return 0;
409
}
410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435

static HsBool ghciLookupSymbolTable(HashTable *table,
    const char *key, void **result)
{
    RtsSymbolInfo *pinfo = lookupStrHashTable(table, key);
    if (!pinfo) {
        *result = NULL;
        return HS_BOOL_FALSE;
    }
    if (pinfo->weak)
        IF_DEBUG(linker, debugBelch("lookup: promoting %s\n", key));
    /* Once it's looked up, it can no longer be overridden */
    pinfo->weak = HS_BOOL_FALSE;

    *result = pinfo->value;
    return HS_BOOL_TRUE;
}

static void ghciRemoveSymbolTable(HashTable *table, const char *key,
    ObjectCode *owner)
{
    RtsSymbolInfo *pinfo = lookupStrHashTable(table, key);
    if (!pinfo || owner != pinfo->owner) return;
    removeStrHashTable(table, key, NULL);
    stgFree(pinfo);
}
436 437 438
/* -----------------------------------------------------------------------------
 * initialize the object linker
 */
439 440 441 442


static int linker_init_done = 0 ;

443
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
444
static void *dl_prog_handle;
445 446 447 448 449
static regex_t re_invalid;
static regex_t re_realso;
#ifdef THREADED_RTS
static Mutex dl_mutex; // mutex to protect dlopen/dlerror critical section
#endif
450 451
#elif defined(OBJFORMAT_PEi386)
void addDLLHandle(pathchar* dll_name, HINSTANCE instance);
452
#endif
453

454 455 456 457
void initLinker (void)
{
    // default to retaining CAFs for backwards compatibility.  Most
    // users will want initLinker_(0): otherwise unloadObj() will not
Gabor Greif's avatar
Gabor Greif committed
458
    // be able to unload object files when they contain CAFs.
459 460 461
    initLinker_(1);
}

462
void
463
initLinker_ (int retain_cafs)
464
{
465
    RtsSymbolVal *sym;
Simon Marlow's avatar
Simon Marlow committed
466
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
467
    int compileResult;
Simon Marlow's avatar
Simon Marlow committed
468
#endif
469

470 471
    IF_DEBUG(linker, debugBelch("initLinker: start\n"));

472
    /* Make initLinker idempotent, so we can call it
Gabor Greif's avatar
typo  
Gabor Greif committed
473
       before every relevant operation; that means we
474
       don't need to initialise the linker separately */
Ian Lynagh's avatar
Ian Lynagh committed
475 476 477
    if (linker_init_done == 1) {
        IF_DEBUG(linker, debugBelch("initLinker: idempotent return\n"));
        return;
478 479
    } else {
        linker_init_done = 1;
480 481
    }

482 483 484
    objects = NULL;
    unloaded_objects = NULL;

485 486
#if defined(THREADED_RTS)
    initMutex(&linker_mutex);
487
    initMutex(&linker_unloaded_mutex);
488
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
489
    initMutex(&dl_mutex);
490
#endif
491
#endif
492 493 494 495
    symhash = allocStrHashTable();

    /* populate the symbol table with stuff from the RTS */
    for (sym = rtsSyms; sym->lbl != NULL; sym++) {
496 497 498 499
        if (! ghciInsertSymbolTable(WSTR("(GHCi built-in symbols)"),
                                    symhash, sym->lbl, sym->addr, HS_BOOL_FALSE, NULL)) {
            barf("ghciInsertSymbolTable failed");
        }
Ian Lynagh's avatar
Ian Lynagh committed
500
        IF_DEBUG(linker, debugBelch("initLinker: inserting rts symbol %s, %p\n", sym->lbl, sym->addr));
501
    }
502
#   if defined(OBJFORMAT_MACHO) && defined(powerpc_HOST_ARCH)
503 504
    machoInitSymbolsWithoutUnderscore();
#   endif
505 506 507 508 509
    /* GCC defines a special symbol __dso_handle which is resolved to NULL if
       referenced from a statically linked module. We need to mimic this, but
       we cannot use NULL because we use it to mean nonexistent symbols. So we
       use an arbitrary (hopefully unique) address here.
    */
510 511 512 513
    if (! ghciInsertSymbolTable(WSTR("(GHCi special symbols)"),
                                symhash, "__dso_handle", (void *)0x12345687, HS_BOOL_FALSE, NULL)) {
        barf("ghciInsertSymbolTable failed");
    }
514

Gabor Greif's avatar
Gabor Greif committed
515
    // Redirect newCAF to newRetainedCAF if retain_cafs is true.
516 517
    if (! ghciInsertSymbolTable(WSTR("(GHCi built-in symbols)"), symhash,
                                MAYBE_LEADING_UNDERSCORE_STR("newCAF"),
518
                                retain_cafs ? newRetainedCAF : newGCdCAF,
519 520 521
                                HS_BOOL_FALSE, NULL)) {
        barf("ghciInsertSymbolTable failed");
    }
522

523
#   if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
524
#   if defined(RTLD_DEFAULT)
525 526
    dl_prog_handle = RTLD_DEFAULT;
#   else
527
    dl_prog_handle = dlopen(NULL, RTLD_LAZY);
528
#   endif /* RTLD_DEFAULT */
529 530

    compileResult = regcomp(&re_invalid,
531
           "(([^ \t()])+\\.so([^ \t:()])*):([ \t])*(invalid ELF header|file too short)",
532
           REG_EXTENDED);
Ian Lynagh's avatar
Ian Lynagh committed
533 534 535
    if (compileResult != 0) {
        barf("Compiling re_invalid failed");
    }
536
    compileResult = regcomp(&re_realso,
537
           "(GROUP|INPUT) *\\( *([^ )]+)",
538
           REG_EXTENDED);
Ian Lynagh's avatar
Ian Lynagh committed
539 540 541
    if (compileResult != 0) {
        barf("Compiling re_realso failed");
    }
542
#   endif
543

544
#if !defined(ALWAYS_PIC) && defined(x86_64_HOST_ARCH)
545 546 547 548 549
    if (RtsFlags.MiscFlags.linkerMemBase != 0) {
        // User-override for mmap_32bit_base
        mmap_32bit_base = (void*)RtsFlags.MiscFlags.linkerMemBase;
    }
#endif
550 551 552 553 554 555 556

#if defined(mingw32_HOST_OS)
    /*
     * These two libraries cause problems when added to the static link,
     * but are necessary for resolving symbols in GHCi, hence we load
     * them manually here.
     */
557 558
    addDLL(WSTR("msvcrt"));
    addDLL(WSTR("kernel32"));
559
    addDLLHandle(WSTR("*.exe"), GetModuleHandle(NULL));
560
#endif
561

562
#if USE_MMAP
563
    m32_allocator_init(&allocator);
Tamar Christina's avatar
Tamar Christina committed
564
#endif
565

566 567
    IF_DEBUG(linker, debugBelch("initLinker: done\n"));
    return;
568 569
}

570 571 572 573 574 575 576 577 578 579 580
void
exitLinker( void ) {
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
   if (linker_init_done == 1) {
      regfree(&re_invalid);
      regfree(&re_realso);
#ifdef THREADED_RTS
      closeMutex(&dl_mutex);
#endif
   }
#endif
581 582 583
   if (linker_init_done == 1) {
       freeHashTable(symhash, free);
   }
584 585 586
#ifdef THREADED_RTS
   closeMutex(&linker_mutex);
#endif
587 588
}

589
/* -----------------------------------------------------------------------------
590 591 592
 *                  Loading DLL or .so dynamic libraries
 * -----------------------------------------------------------------------------
 *
593 594 595 596
 * Add a DLL from which symbols may be found.  In the ELF case, just
 * do RTLD_GLOBAL-style add, so no further messing around needs to
 * happen in order that symbols in the loaded .so are findable --
 * lookupSymbol() will subsequently see them by dlsym on the program's
597 598
 * dl-handle.  Returns NULL if success, otherwise ptr to an err msg.
 *
599
 * In the PEi386 case, open the DLLs and put handles to them in a
600
 * linked list.  When looking for a symbol, try all handles in the
601 602 603 604 605
 * list.  This means that we need to load even DLLs that are guaranteed
 * to be in the ghc.exe image already, just so we can get a handle
 * to give to loadSymbol, so that we can find the symbols.  For such
 * libraries, the LoadLibrary call should be a no-op except for returning
 * the handle.
606
 *
607
 */
608 609 610 611 612 613

#if defined(OBJFORMAT_PEi386)
/* A record for storing handles into DLLs. */

typedef
   struct _OpenedDLL {
614
      pathchar*          name;
615 616
      struct _OpenedDLL* next;
      HINSTANCE instance;
617
   }
618 619 620 621
   OpenedDLL;

/* A list thereof. */
static OpenedDLL* opened_dlls = NULL;
622 623 624 625 626 627 628 629 630 631 632 633

/* A record for storing indirectly linked functions from DLLs. */
typedef
   struct _IndirectAddr {
      void*                 addr;
      struct _IndirectAddr* next;
   }
   IndirectAddr;

/* A list thereof. */
static IndirectAddr* indirects = NULL;

634 635 636 637 638 639 640 641 642 643
/* Adds a DLL instance to the list of DLLs in which to search for symbols. */
void addDLLHandle(pathchar* dll_name, HINSTANCE instance) {
   OpenedDLL* o_dll;
   o_dll = stgMallocBytes( sizeof(OpenedDLL), "addDLLHandle" );
   o_dll->name     = dll_name ? pathdup(dll_name) : NULL;
   o_dll->instance = instance;
   o_dll->next     = opened_dlls;
   opened_dlls     = o_dll;
}

644 645
#endif

646
#  if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
647

648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669
/* Suppose in ghci we load a temporary SO for a module containing
       f = 1
   and then modify the module, recompile, and load another temporary
   SO with
       f = 2
   Then as we don't unload the first SO, dlsym will find the
       f = 1
   symbol whereas we want the
       f = 2
   symbol. We therefore need to keep our own SO handle list, and
   try SOs in the right order. */

typedef
   struct _OpenedSO {
      struct _OpenedSO* next;
      void *handle;
   }
   OpenedSO;

/* A list thereof. */
static OpenedSO* openedSOs = NULL;

670
static const char *
671 672
internal_dlopen(const char *dll_name)
{
673
   OpenedSO* o_so;
674
   void *hdl;
675 676
   const char *errmsg;
   char *errmsg_copy;
677

678 679
   // omitted: RTLD_NOW
   // see http://www.haskell.org/pipermail/cvs-ghc/2007-September/038570.html
680 681 682 683 684 685 686 687 688 689
   IF_DEBUG(linker,
      debugBelch("internal_dlopen: dll_name = '%s'\n", dll_name));

   //-------------- Begin critical section ------------------
   // This critical section is necessary because dlerror() is not
   // required to be reentrant (see POSIX -- IEEE Std 1003.1-2008)
   // Also, the error message returned must be copied to preserve it
   // (see POSIX also)

   ACQUIRE_LOCK(&dl_mutex);
690
   hdl = dlopen(dll_name, RTLD_LAZY|RTLD_LOCAL); /* see Note [RTLD_LOCAL] */
dons's avatar
dons committed
691

692
   errmsg = NULL;
693 694 695 696
   if (hdl == NULL) {
      /* dlopen failed; return a ptr to the error msg. */
      errmsg = dlerror();
      if (errmsg == NULL) errmsg = "addDLL: unknown error";
697 698 699
      errmsg_copy = stgMallocBytes(strlen(errmsg)+1, "addDLL");
      strcpy(errmsg_copy, errmsg);
      errmsg = errmsg_copy;
700 701 702 703 704
   } else {
      o_so = stgMallocBytes(sizeof(OpenedSO), "addDLL");
      o_so->handle = hdl;
      o_so->next   = openedSOs;
      openedSOs    = o_so;
705
   }
706

707 708 709 710 711
   RELEASE_LOCK(&dl_mutex);
   //--------------- End critical section -------------------

   return errmsg;
}
712

713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730
/*
  Note [RTLD_LOCAL]

  In GHCi we want to be able to override previous .so's with newly
  loaded .so's when we recompile something.  This further implies that
  when we look up a symbol in internal_dlsym() we have to iterate
  through the loaded libraries (in order from most recently loaded to
  oldest) looking up the symbol in each one until we find it.

  However, this can cause problems for some symbols that are copied
  by the linker into the executable image at runtime - see #8935 for a
  lengthy discussion.  To solve that problem we need to look up
  symbols in the main executable *first*, before attempting to look
  them up in the loaded .so's.  But in order to make that work, we
  have to always call dlopen with RTLD_LOCAL, so that the loaded
  libraries don't populate the global symbol table.
*/

731
static void *
732
internal_dlsym(const char *symbol) {
733 734 735 736 737 738
    OpenedSO* o_so;
    void *v;

    // We acquire dl_mutex as concurrent dl* calls may alter dlerror
    ACQUIRE_LOCK(&dl_mutex);
    dlerror();
739 740 741 742 743 744 745
    // look in program first
    v = dlsym(dl_prog_handle, symbol);
    if (dlerror() == NULL) {
        RELEASE_LOCK(&dl_mutex);
        return v;
    }

746 747 748 749 750 751 752 753 754 755
    for (o_so = openedSOs; o_so != NULL; o_so = o_so->next) {
        v = dlsym(o_so->handle, symbol);
        if (dlerror() == NULL) {
            RELEASE_LOCK(&dl_mutex);
            return v;
        }
    }
    RELEASE_LOCK(&dl_mutex);
    return v;
}
756 757 758
#  endif

const char *
759
addDLL( pathchar *dll_name )
760 761 762 763 764 765
{
#  if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
   /* ------------------- ELF DLL loader ------------------- */

#define NMATCH 5
   regmatch_t match[NMATCH];
766
   const char *errmsg;
767 768 769 770 771 772 773 774 775 776
   FILE* fp;
   size_t match_length;
#define MAXLINE 1000
   char line[MAXLINE];
   int result;

   IF_DEBUG(linker, debugBelch("addDLL: dll_name = '%s'\n", dll_name));
   errmsg = internal_dlopen(dll_name);

   if (errmsg == NULL) {
777 778
      return NULL;
   }
779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800

   // GHC Trac ticket #2615
   // On some systems (e.g., Gentoo Linux) dynamic files (e.g. libc.so)
   // contain linker scripts rather than ELF-format object code. This
   // code handles the situation by recognizing the real object code
   // file name given in the linker script.
   //
   // If an "invalid ELF header" error occurs, it is assumed that the
   // .so file contains a linker script instead of ELF object code.
   // In this case, the code looks for the GROUP ( ... ) linker
   // directive. If one is found, the first file name inside the
   // parentheses is treated as the name of a dynamic library and the
   // code attempts to dlopen that file. If this is also unsuccessful,
   // an error message is returned.

   // see if the error message is due to an invalid ELF header
   IF_DEBUG(linker, debugBelch("errmsg = '%s'\n", errmsg));
   result = regexec(&re_invalid, errmsg, (size_t) NMATCH, match, 0);
   IF_DEBUG(linker, debugBelch("result = %i\n", result));
   if (result == 0) {
      // success -- try to read the named file as a linker script
      match_length = (size_t) stg_min((match[1].rm_eo - match[1].rm_so),
Ian Lynagh's avatar
Ian Lynagh committed
801
                                 MAXLINE-1);
802 803 804 805
      strncpy(line, (errmsg+(match[1].rm_so)),match_length);
      line[match_length] = '\0'; // make sure string is null-terminated
      IF_DEBUG(linker, debugBelch ("file name = '%s'\n", line));
      if ((fp = fopen(line, "r")) == NULL) {
Ian Lynagh's avatar
Ian Lynagh committed
806
         return errmsg; // return original error if open fails
807
      }
808
      // try to find a GROUP or INPUT ( ... ) command
809
      while (fgets(line, MAXLINE, fp) != NULL) {
Ian Lynagh's avatar
Ian Lynagh committed
810 811
         IF_DEBUG(linker, debugBelch("input line = %s", line));
         if (regexec(&re_realso, line, (size_t) NMATCH, match, 0) == 0) {
812 813
            // success -- try to dlopen the first named file
            IF_DEBUG(linker, debugBelch("match%s\n",""));
814
            line[match[2].rm_eo] = '\0';
815
            stgFree((void*)errmsg); // Free old message before creating new one
816
            errmsg = internal_dlopen(line+match[2].rm_so);
Ian Lynagh's avatar
Ian Lynagh committed
817 818
            break;
         }
819 820 821
         // if control reaches here, no GROUP or INPUT ( ... ) directive
         // was found and the original error message is returned to the
         // caller
822 823 824 825
      }
      fclose(fp);
   }
   return errmsg;
826

827
#  elif defined(OBJFORMAT_PEi386)
828
   /* ------------------- Win32 DLL loader ------------------- */
829

830
   pathchar*      buf;
831
   OpenedDLL* o_dll;
832
   HINSTANCE  instance;
833

834
   /* debugBelch("\naddDLL; dll_name = `%s'\n", dll_name); */
835 836 837

   /* See if we've already got it, and ignore if so. */
   for (o_dll = opened_dlls; o_dll != NULL; o_dll = o_dll->next) {
838
      if (0 == pathcmp(o_dll->name, dll_name))
839
         return NULL;
840 841
   }

842 843 844 845
   /* The file name has no suffix (yet) so that we can try
      both foo.dll and foo.drv

      The documentation for LoadLibrary says:
Ian Lynagh's avatar
Ian Lynagh committed
846 847 848 849 850
        If no file name extension is specified in the lpFileName
        parameter, the default library extension .dll is
        appended. However, the file name string can include a trailing
        point character (.) to indicate that the module name has no
        extension. */
851

852 853 854
   size_t bufsize = pathlen(dll_name) + 10;
   buf = stgMallocBytes(bufsize * sizeof(wchar_t), "addDLL");
   snwprintf(buf, bufsize, L"%s.DLL", dll_name);
855
   instance = LoadLibraryW(buf);
856
   if (instance == NULL) {
857 858
       if (GetLastError() != ERROR_MOD_NOT_FOUND) goto error;
       // KAA: allow loading of drivers (like winspool.drv)
859
       snwprintf(buf, bufsize, L"%s.DRV", dll_name);
860
       instance = LoadLibraryW(buf);
861 862 863
       if (instance == NULL) {
           if (GetLastError() != ERROR_MOD_NOT_FOUND) goto error;
           // #1883: allow loading of unix-style libfoo.dll DLLs
864
           snwprintf(buf, bufsize, L"lib%s.DLL", dll_name);
865
           instance = LoadLibraryW(buf);
866 867 868 869
           if (instance == NULL) {
               goto error;
           }
       }
870
   }
sof's avatar
sof committed
871
   stgFree(buf);
872

873
   addDLLHandle(dll_name, instance);
874 875

   return NULL;
876 877 878

error:
   stgFree(buf);
879
   sysErrorBelch("%" PATH_FMT, dll_name);
Simon Marlow's avatar
Simon Marlow committed
880

881 882
   /* LoadLibrary failed; return a ptr to the error msg. */
   return "addDLL: could not load DLL";
Simon Marlow's avatar
Simon Marlow committed
883 884 885 886

#  else
   barf("addDLL: not implemented on this platform");
#  endif
887 888
}

889 890
/* -----------------------------------------------------------------------------
 * insert a symbol in the hash table
891 892
 *
 * Returns: 0 on failure, nozero on success
893
 */
894
HsInt insertSymbol(pathchar* obj_name, char* key, void* data)
895
{
896
    return ghciInsertSymbolTable(obj_name, symhash, key, data, HS_BOOL_FALSE, NULL);
897 898
}

899 900
/* -----------------------------------------------------------------------------
 * lookup a symbol in the hash table
901
 */
902
static void* lookupSymbol_ (char *lbl)
903
{
904
    void *val;
905
    IF_DEBUG(linker, debugBelch("lookupSymbol: looking up %s\n", lbl));
906

907
    ASSERT(symhash != NULL);
908

909
    if (!ghciLookupSymbolTable(symhash, lbl, &val)) {
Ian Lynagh's avatar
Ian Lynagh committed
910
        IF_DEBUG(linker, debugBelch("lookupSymbol: symbol not found\n"));
911
#       if defined(OBJFORMAT_ELF)
912
        return internal_dlsym(lbl);
913
#       elif defined(OBJFORMAT_MACHO)
914

915
        /* HACK: On OS X, all symbols are prefixed with an underscore.
916
                 However, dlsym wants us to omit the leading underscore from the
917
                 symbol name -- the dlsym routine puts it back on before searching
Austin Seipp's avatar
Austin Seipp committed
918
                 for the symbol. For now, we simply strip it off here (and ONLY
919 920
                 here).
        */
921
        IF_DEBUG(linker, debugBelch("lookupSymbol: looking up %s with dlsym\n", lbl));
Austin Seipp's avatar
Austin Seipp committed
922
        ASSERT(lbl[0] == '_');
923
        return internal_dlsym(lbl + 1);
924
#       elif defined(OBJFORMAT_PEi386)
925
        void* sym;
926

927 928 929 930
/* See Note [mingw-w64 name decoration scheme] */
#ifndef x86_64_HOST_ARCH
         zapTrailingAtSign ( (unsigned char*)lbl );
#endif
Simon Marlow's avatar
Simon Marlow committed
931
        sym = lookupSymbolInDLLs((unsigned char*)lbl);
932
        return sym; // might be NULL if not found
933

ken's avatar
ken committed
934 935 936
#       else
        ASSERT(2+2 == 5);
        return NULL;
937
#       endif
938
    } else {
Ian Lynagh's avatar
Ian Lynagh committed
939 940
        IF_DEBUG(linker, debugBelch("lookupSymbol: value of %s is %p\n", lbl, val));
        return val;
941 942 943
    }
}

944 945 946 947 948 949 950 951
void* lookupSymbol( char *lbl )
{
    ACQUIRE_LOCK(&linker_mutex);
    char *r = lookupSymbol_(lbl);
    RELEASE_LOCK(&linker_mutex);
    return r;
}

952
/* -----------------------------------------------------------------------------
Gabor Greif's avatar
Gabor Greif committed
953
   Create a StablePtr for a foreign export.  This is normally called by
954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982
   a C function with __attribute__((constructor)), which is generated
   by GHC and linked into the module.

   If the object code is being loaded dynamically, then we remember
   which StablePtrs were allocated by the constructors and free them
   again in unloadObj().
   -------------------------------------------------------------------------- */

static ObjectCode *loading_obj = NULL;

StgStablePtr foreignExportStablePtr (StgPtr p)
{
    ForeignExportStablePtr *fe_sptr;
    StgStablePtr *sptr;

    sptr = getStablePtr(p);

    if (loading_obj != NULL) {
        fe_sptr = stgMallocBytes(sizeof(ForeignExportStablePtr),
                                 "foreignExportStablePtr");
        fe_sptr->stable_ptr = sptr;
        fe_sptr->next = loading_obj->stable_ptrs;
        loading_obj->stable_ptrs = fe_sptr;
    }

    return sptr;
}


983 984 985 986 987 988 989 990 991 992 993 994 995 996
/* -----------------------------------------------------------------------------
 * Debugging aid: look in GHCi's object symbol tables for symbols
 * within DELTA bytes of the specified address, and show their names.
 */
#ifdef DEBUG
void ghci_enquire ( char* addr );

void ghci_enquire ( char* addr )
{
   int   i;
   char* sym;
   char* a;
   const int DELTA = 64;
   ObjectCode* oc;
997

998 999 1000 1001 1002
   for (oc = objects; oc; oc = oc->next) {
      for (i = 0; i < oc->n_symbols; i++) {
         sym = oc->symbols[i];
         if (sym == NULL) continue;
         a = NULL;
1003
         if (a == NULL) {
1004
            ghciLookupSymbolTable(symhash, sym, (void **)&a);
Ian Lynagh's avatar
Ian Lynagh committed
1005
         }
1006
         if (a == NULL) {
Ian Lynagh's avatar
Ian Lynagh committed
1007
             // debugBelch("ghci_enquire: can't find %s\n", sym);
1008
         }
1009
         else if (addr-DELTA <= a && a <= addr+DELTA) {
1010
            debugBelch("%p + %3d  ==  `%s'\n", addr, (int)(a - addr), sym);
1011 1012 1013 1014 1015 1016
         }
      }
   }
}
#endif

1017
#if USE_MMAP
1018
#define ROUND_UP(x,size) ((x + size - 1) & ~(size - 1))
1019 1020 1021 1022 1023 1024
#define ROUND_DOWN(x,size) (x & ~(size - 1))

static StgWord getPageSize(void)
{
    static StgWord pagesize = 0;
    if (pagesize == 0) {
Simon Marlow's avatar
Simon Marlow committed
1025
        pagesize = sysconf(_SC_PAGESIZE);
1026
    }
Simon Marlow's avatar
Simon Marlow committed
1027
    return pagesize;
1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040
}

static StgWord roundUpToPage (StgWord size)
{
    return ROUND_UP(size, getPageSize());
}

#ifdef OBJFORMAT_ELF
static StgWord roundDownToPage (StgWord size)
{
    return ROUND_DOWN(size, getPageSize());
}
#endif
1041

1042 1043 1044
//
// Returns NULL on failure.
//
1045
static void * mmapForLinker (size_t bytes, nat flags, int fd, int offset)
1046 1047