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,
164
                         rtsBool mapped, char *archiveMemberName
165
#if (USE_MMAP == 0) && defined (darwin_HOST_OS)
166 167 168 169
                       , int misalignment
#endif
                       );

170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201
// 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;
}


202
#if defined(OBJFORMAT_ELF)
203 204 205
static int ocVerifyImage_ELF    ( ObjectCode* oc );
static int ocGetNames_ELF       ( ObjectCode* oc );
static int ocResolve_ELF        ( ObjectCode* oc );
206
static int ocRunInit_ELF        ( ObjectCode* oc );
207
#if NEED_SYMBOL_EXTRAS
208
static int ocAllocateSymbolExtras_ELF ( ObjectCode* oc );
209
#endif
210
#elif defined(OBJFORMAT_PEi386)
211 212 213
static int ocVerifyImage_PEi386 ( ObjectCode* oc );
static int ocGetNames_PEi386    ( ObjectCode* oc );
static int ocResolve_PEi386     ( ObjectCode* oc );
214
static int ocRunInit_PEi386     ( ObjectCode* oc );
215
static void *lookupSymbolInDLLs ( unsigned char *lbl );
216 217 218 219
/* See Note [mingw-w64 name decoration scheme] */
#ifndef x86_64_HOST_ARCH
 static void zapTrailingAtSign   ( unsigned char *sym );
#endif
220
static char *allocateImageAndTrampolines (
Austin Seipp's avatar
Austin Seipp committed
221
   pathchar* arch_name, char* member_name,
222
#if defined(x86_64_HOST_ARCH)
Austin Seipp's avatar
Austin Seipp committed
223
   FILE* f,
224 225 226 227 228 229 230 231 232
#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
233 234 235 236
#elif defined(OBJFORMAT_MACHO)
static int ocVerifyImage_MachO    ( ObjectCode* oc );
static int ocGetNames_MachO       ( ObjectCode* oc );
static int ocResolve_MachO        ( ObjectCode* oc );
237
static int ocRunInit_MachO        ( ObjectCode* oc );
238

239
#if (USE_MMAP == 0)
240
static int machoGetMisalignment( FILE * );
Ian Lynagh's avatar
Ian Lynagh committed
241
#endif
242
#if NEED_SYMBOL_EXTRAS
243 244
static int ocAllocateSymbolExtras_MachO ( ObjectCode* oc );
#endif
245
#ifdef powerpc_HOST_ARCH
246
static void machoInitSymbolsWithoutUnderscore( void );
247
#endif
248
#endif
249

250 251
static void freeProddableBlocks (ObjectCode *oc);

252
#if USE_MMAP
253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
/**
 * 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

280 281 282 283 284 285 286 287 288 289 290 291 292
/* 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
293
 *
294 295 296 297 298 299 300 301
 * 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
302 303 304 305 306 307 308 309 310 311 312 313
 *
 * 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);
314 315
 */
#define X86_64_ELF_NONPIC_HACK 1
316

317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342
/* 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)
 */
343
#if !defined(ALWAYS_PIC) && defined(x86_64_HOST_ARCH)
344 345 346 347 348 349 350 351 352

#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
353
static void *mmap_32bit_base = (void *)MMAP_32BIT_BASE_DEFAULT;
354 355 356 357 358 359 360
#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

361 362
/* -----------------------------------------------------------------------------
 * Insert symbols into hash tables, checking for duplicates.
363 364
 *
 * Returns: 0 on failure, nonzero on success
365
 */
366

367
static int ghciInsertSymbolTable(
368 369
   pathchar* obj_name,
   HashTable *table,
370
   const char* key,
371 372 373
   void *data,
   HsBool weak,
   ObjectCode *owner)
374
{
375 376 377 378 379 380 381 382
   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);
383 384 385 386 387 388 389
      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
390
   {
391 392 393 394
      /* override the weak definition with the non-weak one */
      pinfo->value = data;
      pinfo->owner = owner;
      pinfo->weak = HS_BOOL_FALSE;
395
      return 1;
Simon Marlow's avatar
Simon Marlow committed
396 397
   }
   debugBelch(
398
      "GHC runtime linker: fatal error: I found a duplicate definition for symbol\n"
399 400
      "   %s\n"
      "whilst processing object file\n"
401
      "   %" PATH_FMT "\n"
402 403 404 405
      "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"
406
      "     loaded twice.\n",
407 408
      (char*)key,
      obj_name
Simon Marlow's avatar
Simon Marlow committed
409
   );
410
   return 0;
411
}
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437

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);
}
438 439 440
/* -----------------------------------------------------------------------------
 * initialize the object linker
 */
441 442 443 444


static int linker_init_done = 0 ;

445
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
446
static void *dl_prog_handle;
447 448 449 450 451
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
452 453
#elif defined(OBJFORMAT_PEi386)
void addDLLHandle(pathchar* dll_name, HINSTANCE instance);
454
#endif
455

456 457 458 459
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
460
    // be able to unload object files when they contain CAFs.
461 462 463
    initLinker_(1);
}

464
void
465
initLinker_ (int retain_cafs)
466
{
467
    RtsSymbolVal *sym;
Simon Marlow's avatar
Simon Marlow committed
468
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
469
    int compileResult;
Simon Marlow's avatar
Simon Marlow committed
470
#endif
471

472 473
    IF_DEBUG(linker, debugBelch("initLinker: start\n"));

474
    /* Make initLinker idempotent, so we can call it
Gabor Greif's avatar
typo  
Gabor Greif committed
475
       before every relevant operation; that means we
476
       don't need to initialise the linker separately */
Ian Lynagh's avatar
Ian Lynagh committed
477 478 479
    if (linker_init_done == 1) {
        IF_DEBUG(linker, debugBelch("initLinker: idempotent return\n"));
        return;
480 481
    } else {
        linker_init_done = 1;
482 483
    }

484 485 486
    objects = NULL;
    unloaded_objects = NULL;

487 488
#if defined(THREADED_RTS)
    initMutex(&linker_mutex);
489
    initMutex(&linker_unloaded_mutex);
490
#if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
491
    initMutex(&dl_mutex);
492
#endif
493
#endif
494 495 496 497
    symhash = allocStrHashTable();

    /* populate the symbol table with stuff from the RTS */
    for (sym = rtsSyms; sym->lbl != NULL; sym++) {
498 499 500 501
        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
502
        IF_DEBUG(linker, debugBelch("initLinker: inserting rts symbol %s, %p\n", sym->lbl, sym->addr));
503
    }
504
#   if defined(OBJFORMAT_MACHO) && defined(powerpc_HOST_ARCH)
505 506
    machoInitSymbolsWithoutUnderscore();
#   endif
507 508 509 510 511
    /* 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.
    */
512 513 514 515
    if (! ghciInsertSymbolTable(WSTR("(GHCi special symbols)"),
                                symhash, "__dso_handle", (void *)0x12345687, HS_BOOL_FALSE, NULL)) {
        barf("ghciInsertSymbolTable failed");
    }
516

Gabor Greif's avatar
Gabor Greif committed
517
    // Redirect newCAF to newRetainedCAF if retain_cafs is true.
518 519
    if (! ghciInsertSymbolTable(WSTR("(GHCi built-in symbols)"), symhash,
                                MAYBE_LEADING_UNDERSCORE_STR("newCAF"),
520
                                retain_cafs ? newRetainedCAF : newGCdCAF,
521 522 523
                                HS_BOOL_FALSE, NULL)) {
        barf("ghciInsertSymbolTable failed");
    }
524

525
#   if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
526
#   if defined(RTLD_DEFAULT)
527 528
    dl_prog_handle = RTLD_DEFAULT;
#   else
529
    dl_prog_handle = dlopen(NULL, RTLD_LAZY);
530
#   endif /* RTLD_DEFAULT */
531 532

    compileResult = regcomp(&re_invalid,
533
           "(([^ \t()])+\\.so([^ \t:()])*):([ \t])*(invalid ELF header|file too short)",
534
           REG_EXTENDED);
Ian Lynagh's avatar
Ian Lynagh committed
535 536 537
    if (compileResult != 0) {
        barf("Compiling re_invalid failed");
    }
538
    compileResult = regcomp(&re_realso,
539
           "(GROUP|INPUT) *\\( *([^ )]+)",
540
           REG_EXTENDED);
Ian Lynagh's avatar
Ian Lynagh committed
541 542 543
    if (compileResult != 0) {
        barf("Compiling re_realso failed");
    }
544
#   endif
545

546
#if !defined(ALWAYS_PIC) && defined(x86_64_HOST_ARCH)
547 548 549 550 551
    if (RtsFlags.MiscFlags.linkerMemBase != 0) {
        // User-override for mmap_32bit_base
        mmap_32bit_base = (void*)RtsFlags.MiscFlags.linkerMemBase;
    }
#endif
552 553 554 555 556 557 558

#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.
     */
559 560
    addDLL(WSTR("msvcrt"));
    addDLL(WSTR("kernel32"));
561
    addDLLHandle(WSTR("*.exe"), GetModuleHandle(NULL));
562
#endif
563

564
#if USE_MMAP
565
    m32_allocator_init(&allocator);
Tamar Christina's avatar
Tamar Christina committed
566
#endif
567

568 569
    IF_DEBUG(linker, debugBelch("initLinker: done\n"));
    return;
570 571
}

572 573 574 575 576 577 578 579 580 581 582
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
583 584 585
   if (linker_init_done == 1) {
       freeHashTable(symhash, free);
   }
586 587 588
#ifdef THREADED_RTS
   closeMutex(&linker_mutex);
#endif
589 590
}

591
/* -----------------------------------------------------------------------------
592 593 594
 *                  Loading DLL or .so dynamic libraries
 * -----------------------------------------------------------------------------
 *
595 596 597 598
 * 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
599 600
 * dl-handle.  Returns NULL if success, otherwise ptr to an err msg.
 *
601
 * In the PEi386 case, open the DLLs and put handles to them in a
602
 * linked list.  When looking for a symbol, try all handles in the
603 604 605 606 607
 * 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.
608
 *
609
 */
610 611 612 613 614 615

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

typedef
   struct _OpenedDLL {
616
      pathchar*          name;
617 618
      struct _OpenedDLL* next;
      HINSTANCE instance;
619
   }
620 621 622 623
   OpenedDLL;

/* A list thereof. */
static OpenedDLL* opened_dlls = NULL;
624 625 626 627 628 629 630 631 632 633 634 635

/* 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;

636 637 638 639 640 641 642 643 644 645
/* 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;
}

646 647
#endif

648
#  if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
649

650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671
/* 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;

672
static const char *
673 674
internal_dlopen(const char *dll_name)
{
675
   OpenedSO* o_so;
676
   void *hdl;
677 678
   const char *errmsg;
   char *errmsg_copy;
679

680 681
   // omitted: RTLD_NOW
   // see http://www.haskell.org/pipermail/cvs-ghc/2007-September/038570.html
682 683 684 685 686 687 688 689 690 691
   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);
692
   hdl = dlopen(dll_name, RTLD_LAZY|RTLD_LOCAL); /* see Note [RTLD_LOCAL] */
dons's avatar
dons committed
693

694
   errmsg = NULL;
695 696 697 698
   if (hdl == NULL) {
      /* dlopen failed; return a ptr to the error msg. */
      errmsg = dlerror();
      if (errmsg == NULL) errmsg = "addDLL: unknown error";
699 700 701
      errmsg_copy = stgMallocBytes(strlen(errmsg)+1, "addDLL");
      strcpy(errmsg_copy, errmsg);
      errmsg = errmsg_copy;
702 703 704 705 706
   } else {
      o_so = stgMallocBytes(sizeof(OpenedSO), "addDLL");
      o_so->handle = hdl;
      o_so->next   = openedSOs;
      openedSOs    = o_so;
707
   }
708

709 710 711 712 713
   RELEASE_LOCK(&dl_mutex);
   //--------------- End critical section -------------------

   return errmsg;
}
714

715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732
/*
  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.
*/

733
static void *
734
internal_dlsym(const char *symbol) {
735 736 737 738 739 740
    OpenedSO* o_so;
    void *v;

    // We acquire dl_mutex as concurrent dl* calls may alter dlerror
    ACQUIRE_LOCK(&dl_mutex);
    dlerror();
741 742 743 744 745 746 747
    // look in program first
    v = dlsym(dl_prog_handle, symbol);
    if (dlerror() == NULL) {
        RELEASE_LOCK(&dl_mutex);
        return v;
    }

748 749 750 751 752 753 754 755 756 757
    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;
}
758 759 760
#  endif

const char *
761
addDLL( pathchar *dll_name )
762 763 764 765 766 767
{
#  if defined(OBJFORMAT_ELF) || defined(OBJFORMAT_MACHO)
   /* ------------------- ELF DLL loader ------------------- */

#define NMATCH 5
   regmatch_t match[NMATCH];
768
   const char *errmsg;
769 770 771 772 773 774 775 776 777 778
   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) {
779 780
      return NULL;
   }
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802

   // 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
803
                                 MAXLINE-1);
804 805 806 807
      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
808
         return errmsg; // return original error if open fails
809
      }
810
      // try to find a GROUP or INPUT ( ... ) command
811
      while (fgets(line, MAXLINE, fp) != NULL) {
Ian Lynagh's avatar
Ian Lynagh committed
812 813
         IF_DEBUG(linker, debugBelch("input line = %s", line));
         if (regexec(&re_realso, line, (size_t) NMATCH, match, 0) == 0) {
814 815
            // success -- try to dlopen the first named file
            IF_DEBUG(linker, debugBelch("match%s\n",""));
816
            line[match[2].rm_eo] = '\0';
817
            stgFree((void*)errmsg); // Free old message before creating new one
818
            errmsg = internal_dlopen(line+match[2].rm_so);
Ian Lynagh's avatar
Ian Lynagh committed
819 820
            break;
         }
821 822 823
         // if control reaches here, no GROUP or INPUT ( ... ) directive
         // was found and the original error message is returned to the
         // caller
824 825 826 827
      }
      fclose(fp);
   }
   return errmsg;
828

829
#  elif defined(OBJFORMAT_PEi386)
830
   /* ------------------- Win32 DLL loader ------------------- */
831

832
   pathchar*      buf;
833
   OpenedDLL* o_dll;
834
   HINSTANCE  instance;
835

836
   /* debugBelch("\naddDLL; dll_name = `%s'\n", dll_name); */
837 838 839

   /* See if we've already got it, and ignore if so. */
   for (o_dll = opened_dlls; o_dll != NULL; o_dll = o_dll->next) {
840
      if (0 == pathcmp(o_dll->name, dll_name))
841
         return NULL;
842 843
   }

844 845 846 847
   /* 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
848 849 850 851 852
        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. */
853

854 855 856
   size_t bufsize = pathlen(dll_name) + 10;
   buf = stgMallocBytes(bufsize * sizeof(wchar_t), "addDLL");
   snwprintf(buf, bufsize, L"%s.DLL", dll_name);
857
   instance = LoadLibraryW(buf);
858
   if (instance == NULL) {
859 860
       if (GetLastError() != ERROR_MOD_NOT_FOUND) goto error;
       // KAA: allow loading of drivers (like winspool.drv)
861
       snwprintf(buf, bufsize, L"%s.DRV", dll_name);
862
       instance = LoadLibraryW(buf);
863 864 865
       if (instance == NULL) {
           if (GetLastError() != ERROR_MOD_NOT_FOUND) goto error;
           // #1883: allow loading of unix-style libfoo.dll DLLs
866
           snwprintf(buf, bufsize, L"lib%s.DLL", dll_name);
867
           instance = LoadLibraryW(buf);
868 869 870 871
           if (instance == NULL) {
               goto error;
           }
       }
872
   }
sof's avatar
sof committed
873
   stgFree(buf);
874

875
   addDLLHandle(dll_name, instance);
876 877

   return NULL;
878 879 880

error:
   stgFree(buf);
881
   sysErrorBelch("%" PATH_FMT, dll_name);
Simon Marlow's avatar
Simon Marlow committed
882

883 884
   /* LoadLibrary failed; return a ptr to the error msg. */
   return "addDLL: could not load DLL";
Simon Marlow's avatar
Simon Marlow committed
885 886 887 888

#  else
   barf("addDLL: not implemented on this platform");
#  endif
889 890
}

891 892
/* -----------------------------------------------------------------------------
 * insert a symbol in the hash table
893 894
 *
 * Returns: 0 on failure, nozero on success
895
 */
896
HsInt insertSymbol(pathchar* obj_name, char* key, void* data)
897
{
898
    return ghciInsertSymbolTable(obj_name, symhash, key, data, HS_BOOL_FALSE, NULL);
899 900
}

901 902
/* -----------------------------------------------------------------------------
 * lookup a symbol in the hash table
903
 */
904
static void* lookupSymbol_ (char *lbl)
905
{
906
    void *val;
907
    IF_DEBUG(linker, debugBelch("lookupSymbol: looking up %s\n", lbl));
908

909
    ASSERT(symhash != NULL);
910

911
    if (!ghciLookupSymbolTable(symhash, lbl, &val)) {
Ian Lynagh's avatar
Ian Lynagh committed
912
        IF_DEBUG(linker, debugBelch("lookupSymbol: symbol not found\n"));
913
#       if defined(OBJFORMAT_ELF)
914
        return internal_dlsym(lbl);
915
#       elif defined(OBJFORMAT_MACHO)
916

917
        /* HACK: On OS X, all symbols are prefixed with an underscore.
918
                 However, dlsym wants us to omit the leading underscore from the
919
                 symbol name -- the dlsym routine puts it back on before searching
Austin Seipp's avatar
Austin Seipp committed
920
                 for the symbol. For now, we simply strip it off here (and ONLY
921 922
                 here).
        */
923
        IF_DEBUG(linker, debugBelch("lookupSymbol: looking up %s with dlsym\n", lbl));
Austin Seipp's avatar
Austin Seipp committed
924
        ASSERT(lbl[0] == '_');
925
        return internal_dlsym(lbl + 1);
926
#       elif defined(OBJFORMAT_PEi386)
927
        void* sym;
928

929 930 931 932
/* See Note [mingw-w64 name decoration scheme] */
#ifndef x86_64_HOST_ARCH
         zapTrailingAtSign ( (unsigned char*)lbl );
#endif
Simon Marlow's avatar
Simon Marlow committed
933
        sym = lookupSymbolInDLLs((unsigned char*)lbl);
934
        return sym; // might be NULL if not found
935

ken's avatar
ken committed
936 937 938
#       else
        ASSERT(2+2 == 5);
        return NULL;
939
#       endif
940
    } else {
Ian Lynagh's avatar
Ian Lynagh committed
941 942
        IF_DEBUG(linker, debugBelch("lookupSymbol: value of %s is %p\n", lbl, val));
        return val;
943 944 945
    }
}

946 947 948 949 950 951 952 953
void* lookupSymbol( char *lbl )
{
    ACQUIRE_LOCK(&linker_mutex);
    char *r = lookupSymbol_(lbl);
    RELEASE_LOCK(&linker_mutex);
    return r;
}

954
/* -----------------------------------------------------------------------------
Gabor Greif's avatar
Gabor Greif committed
955
   Create a StablePtr for a foreign export.  This is normally called by
956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984
   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;
}


985 986 987 988 989 990 991 992 993 994 995 996 997 998
/* -----------------------------------------------------------------------------
 * 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;
999

1000 1001 1002 1003 1004
   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;
1005
         if (a == NULL) {
1006
            ghciLookupSymbolTable(symhash, sym, (void **)&a);
Ian Lynagh's avatar
Ian Lynagh committed
1007
         }
1008
         if (a == NULL) {
Ian Lynagh's avatar
Ian Lynagh committed
1009
             // debugBelch("ghci_enquire: can't find %s\n", sym);
1010
         }
1011
         else if (addr-DELTA <= a && a <= addr+DELTA) {
1012
            debugBelch("%p + %3d  ==  `%s'\n", addr, (int)(a - addr), sym);
1013 1014 1015 1016 1017 1018
         }
      }
   }
}
#endif

1019
#if USE_MMAP
1020
#define ROUND_UP(x,size) ((x + size - 1) & ~(size - 1))
1021 1022 1023 1024 1025 1026
#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
1027
        pagesize = sysconf(_SC_PAGESIZE);
1028
    }
Simon Marlow's avatar
Simon Marlow committed
1029
    return pagesize;
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042
}

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

#ifdef OBJFORMAT_ELF
static StgWord roundDownToPage (StgWord size)
{
    return ROUND_DOWN(size, getPageSize());
}
#endif
1043

1044 1045 1046
//
// Returns NULL on failure.
//
1047
static void * mmapForLinker (size_t bytes, nat fl