commit cfbf0eb134efd1c5d9a589f6ae2139d7fad60581
Author: Duncan Coutts <duncan@well-typed.com>
Date:   Thu May 12 16:25:41 2011 +0100

    Make the GHCi linker handle partially stripped object files (#5004)

    When you use 'strip --strip-unneeded' on a ELF format .o or .a file, if
    the object file has no global/exported symbols then 'strip' ends up
    removing the symbol table entirely. Previously the GHCi linker assumed
    there would always be exactly one symbol table and exactly one string
    table. In fact, in ELF object files there is no such limitation, instead
    each section points to the other sections it needs, in particular
    relocation sections have a link to the symbol table section they use and
    symbol table sections have a link to the corresponding string table.
    So instead of assuming there will always be a global symbol and string
    table, all we have to do is validate and follow these links. Then, when
    we encounter an empty object file that has no symbols then we handle it
    correctly, because since it's empty we never process any relocations and
    so never have to follow any links to non-existant symbol tables.

    Also, in the case where an object is fully stripped, we can now detect
    this more reliably and emit a more helpful error message, e.g:

    libHSghc-7.1.20110509.a(DsMeta.o): relocation section #2 has no symbol table
    This object file has probably been fully striped. Such files cannot be linked.

Index: ghc-7.0.3/rts/Linker.c
===================================================================
--- ghc-7.0.3.orig/rts/Linker.c	2011-08-10 11:49:12.285949093 +0100
+++ ghc-7.0.3/rts/Linker.c	2011-08-10 11:48:59.658456201 +0100
@@ -2208,6 +2208,7 @@
             //  stgFree(oc->image);
             // #endif
             stgFree(oc->fileName);
+            stgFree(oc->archiveMemberName);
             stgFree(oc->symbols);
             stgFree(oc->sections);
             stgFree(oc);
@@ -3540,31 +3541,6 @@
  * Generic ELF functions
  */
 
-static char *
-findElfSection ( void* objImage, Elf_Word sh_type )
-{
-   char* ehdrC = (char*)objImage;
-   Elf_Ehdr* ehdr = (Elf_Ehdr*)ehdrC;
-   Elf_Shdr* shdr = (Elf_Shdr*)(ehdrC + ehdr->e_shoff);
-   char* sh_strtab = ehdrC + shdr[ehdr->e_shstrndx].sh_offset;
-   char* ptr = NULL;
-   int i;
-
-   for (i = 0; i < ehdr->e_shnum; i++) {
-      if (shdr[i].sh_type == sh_type
-          /* Ignore the section header's string table. */
-          && i != ehdr->e_shstrndx
-          /* Ignore string tables named .stabstr, as they contain
-             debugging info. */
-          && 0 != memcmp(".stabstr", sh_strtab + shdr[i].sh_name, 8)
-         ) {
-         ptr = ehdrC + shdr[i].sh_offset;
-         break;
-      }
-   }
-   return ptr;
-}
-
 static int
 ocVerifyImage_ELF ( ObjectCode* oc )
 {
@@ -3572,7 +3548,6 @@
    Elf_Sym*  stab;
    int i, j, nent, nstrtab, nsymtabs;
    char* sh_strtab;
-   char* strtab;
 
    char*     ehdrC = (char*)(oc->image);
    Elf_Ehdr* ehdr  = (Elf_Ehdr*)ehdrC;
@@ -3654,20 +3629,64 @@
                ehdrC + shdr[i].sh_offset,
                       ehdrC + shdr[i].sh_offset + shdr[i].sh_size - 1));
 
-      if (shdr[i].sh_type == SHT_REL) {
-          IF_DEBUG(linker,debugBelch("Rel  " ));
-      } else if (shdr[i].sh_type == SHT_RELA) {
-          IF_DEBUG(linker,debugBelch("RelA " ));
-      } else {
-          IF_DEBUG(linker,debugBelch("     "));
+#define SECTION_INDEX_VALID(ndx) (ndx > SHN_UNDEF && ndx < ehdr->e_shnum)
+
+      switch (shdr[i].sh_type) {
+
+        case SHT_REL:
+        case SHT_RELA:
+          IF_DEBUG(linker,debugBelch( shdr[i].sh_type == SHT_REL ? "Rel  " : "RelA "));
+
+          if (!SECTION_INDEX_VALID(shdr[i].sh_link)) {
+            if (shdr[i].sh_link == SHN_UNDEF)
+              errorBelch("\n%s: relocation section #%d has no symbol table\n"
+                         "This object file has probably been fully striped. "
+                         "Such files cannot be linked.\n",
+                         oc->archiveMemberName ? oc->archiveMemberName : oc->fileName, i);
+            else
+              errorBelch("\n%s: relocation section #%d has an invalid link field (%d)\n",
+                         oc->archiveMemberName ? oc->archiveMemberName : oc->fileName,
+                         i, shdr[i].sh_link);
+            return 0;
+          }
+          if (shdr[shdr[i].sh_link].sh_type != SHT_SYMTAB) {
+            errorBelch("\n%s: relocation section #%d does not link to a symbol table\n",
+                       oc->archiveMemberName ? oc->archiveMemberName : oc->fileName, i);
+            return 0;
+          }
+          if (!SECTION_INDEX_VALID(shdr[i].sh_info)) {
+            errorBelch("\n%s: relocation section #%d has an invalid info field (%d)\n",
+                       oc->archiveMemberName ? oc->archiveMemberName : oc->fileName,
+                       i, shdr[i].sh_info);
+            return 0;
+          }
+
+          break;
+        case SHT_SYMTAB:
+          IF_DEBUG(linker,debugBelch("Sym  "));
+
+          if (!SECTION_INDEX_VALID(shdr[i].sh_link)) {
+            errorBelch("\n%s: symbol table section #%d has an invalid link field (%d)\n",
+                       oc->archiveMemberName ? oc->archiveMemberName : oc->fileName,
+                       i, shdr[i].sh_link);
+            return 0;
+          }
+          if (shdr[shdr[i].sh_link].sh_type != SHT_STRTAB) {
+            errorBelch("\n%s: symbol table section #%d does not link to a string table\n",
+                       oc->archiveMemberName ? oc->archiveMemberName : oc->fileName, i);
+
+            return 0;
+          }
+          break;
+        case SHT_STRTAB: IF_DEBUG(linker,debugBelch("Str  ")); break;
+        default:         IF_DEBUG(linker,debugBelch("     ")); break;
       }
       if (sh_strtab) {
           IF_DEBUG(linker,debugBelch("sname=%s\n", sh_strtab + shdr[i].sh_name ));
       }
    }
 
-   IF_DEBUG(linker,debugBelch( "\nString tables" ));
-   strtab = NULL;
+   IF_DEBUG(linker,debugBelch( "\nString tables\n" ));
    nstrtab = 0;
    for (i = 0; i < ehdr->e_shnum; i++) {
       if (shdr[i].sh_type == SHT_STRTAB
@@ -3677,18 +3696,16 @@
              debugging info. */
           && 0 != memcmp(".stabstr", sh_strtab + shdr[i].sh_name, 8)
          ) {
-         IF_DEBUG(linker,debugBelch("   section %d is a normal string table", i ));
-         strtab = ehdrC + shdr[i].sh_offset;
+         IF_DEBUG(linker,debugBelch("   section %d is a normal string table\n", i ));
          nstrtab++;
       }
    }
-   if (nstrtab != 1) {
-      errorBelch("%s: no string tables, or too many", oc->fileName);
-      return 0;
+   if (nstrtab == 0) {
+      IF_DEBUG(linker,debugBelch("   no normal string tables (potentially, but not necessarily a problem)\n"));
    }
 
    nsymtabs = 0;
-   IF_DEBUG(linker,debugBelch( "\nSymbol tables" ));
+   IF_DEBUG(linker,debugBelch( "Symbol tables\n" ));
    for (i = 0; i < ehdr->e_shnum; i++) {
       if (shdr[i].sh_type != SHT_SYMTAB) continue;
       IF_DEBUG(linker,debugBelch( "section %d is a symbol table\n", i ));
@@ -3730,13 +3747,17 @@
          }
          IF_DEBUG(linker,debugBelch("  " ));
 
-         IF_DEBUG(linker,debugBelch("name=%s\n", strtab + stab[j].st_name ));
+         IF_DEBUG(linker,debugBelch("name=%s\n",
+                        ehdrC + shdr[shdr[i].sh_link].sh_offset
+                              + stab[j].st_name ));
       }
    }
 
    if (nsymtabs == 0) {
-      errorBelch("%s: didn't find any symbol tables", oc->fileName);
-      return 0;
+     // Not having a symbol table is not in principle a problem.
+     // When an object file has no symbols then the 'strip' program
+     // typically will remove the symbol table entirely.
+     IF_DEBUG(linker,debugBelch("   no symbol tables (potentially, but not necessarily a problem)\n"));
    }
 
    return 1;
@@ -3783,16 +3804,11 @@
 
    char*     ehdrC    = (char*)(oc->image);
    Elf_Ehdr* ehdr     = (Elf_Ehdr*)ehdrC;
-   char*     strtab   = findElfSection ( ehdrC, SHT_STRTAB );
+   char*     strtab;
    Elf_Shdr* shdr     = (Elf_Shdr*) (ehdrC + ehdr->e_shoff);
 
    ASSERT(symhash != NULL);
 
-   if (!strtab) {
-      errorBelch("%s: no strtab", oc->fileName);
-      return 0;
-   }
-
    k = 0;
    for (i = 0; i < ehdr->e_shnum; i++) {
       /* Figure out what kind of section it is.  Logic derived from
@@ -3825,12 +3841,16 @@
 
       /* copy stuff into this module's object symbol table */
       stab = (Elf_Sym*) (ehdrC + shdr[i].sh_offset);
+      strtab = ehdrC + shdr[shdr[i].sh_link].sh_offset;
       nent = shdr[i].sh_size / sizeof(Elf_Sym);
 
       oc->n_symbols = nent;
       oc->symbols = stgMallocBytes(oc->n_symbols * sizeof(char*),
                                    "ocGetNames_ELF(oc->symbols)");
 
+      //TODO: we ignore local symbols anyway right? So we can use the
+      //      shdr[i].sh_info to get the index of the first non-local symbol
+      // ie we should use j = shdr[i].sh_info
       for (j = 0; j < nent; j++) {
 
          char  isLocal = FALSE; /* avoids uninit-var warning */
@@ -3928,21 +3948,24 @@
    relocations appear to be of this form. */
 static int
 do_Elf_Rel_relocations ( ObjectCode* oc, char* ehdrC,
-                         Elf_Shdr* shdr, int shnum,
-                         Elf_Sym*  stab, char* strtab )
+                         Elf_Shdr* shdr, int shnum )
 {
    int j;
    char *symbol;
    Elf_Word* targ;
    Elf_Rel*  rtab = (Elf_Rel*) (ehdrC + shdr[shnum].sh_offset);
+   Elf_Sym*  stab;
+   char*     strtab;
    int         nent = shdr[shnum].sh_size / sizeof(Elf_Rel);
    int target_shndx = shdr[shnum].sh_info;
    int symtab_shndx = shdr[shnum].sh_link;
+   int strtab_shndx = shdr[symtab_shndx].sh_link;
 
    stab  = (Elf_Sym*) (ehdrC + shdr[ symtab_shndx ].sh_offset);
+   strtab= (char*)    (ehdrC + shdr[ strtab_shndx ].sh_offset);
    targ  = (Elf_Word*)(ehdrC + shdr[ target_shndx ].sh_offset);
-   IF_DEBUG(linker,debugBelch( "relocations for section %d using symtab %d\n",
-                          target_shndx, symtab_shndx ));
+   IF_DEBUG(linker,debugBelch( "relocations for section %d using symtab %d and strtab %d\n",
+                          target_shndx, symtab_shndx, strtab_shndx ));
 
    /* Skip sections that we're not interested in. */
    {
@@ -4028,18 +4051,21 @@
    sparc-solaris relocations appear to be of this form. */
 static int
 do_Elf_Rela_relocations ( ObjectCode* oc, char* ehdrC,
-                          Elf_Shdr* shdr, int shnum,
-                          Elf_Sym*  stab, char* strtab )
+                          Elf_Shdr* shdr, int shnum )
 {
    int j;
    char *symbol = NULL;
    Elf_Addr targ;
    Elf_Rela* rtab = (Elf_Rela*) (ehdrC + shdr[shnum].sh_offset);
+   Elf_Sym*  stab;
+   char*     strtab;
    int         nent = shdr[shnum].sh_size / sizeof(Elf_Rela);
    int target_shndx = shdr[shnum].sh_info;
    int symtab_shndx = shdr[shnum].sh_link;
+   int strtab_shndx = shdr[symtab_shndx].sh_link;
 
    stab  = (Elf_Sym*) (ehdrC + shdr[ symtab_shndx ].sh_offset);
+   strtab= (char*)    (ehdrC + shdr[ strtab_shndx ].sh_offset);
    targ  = (Elf_Addr) (ehdrC + shdr[ target_shndx ].sh_offset);
    IF_DEBUG(linker,debugBelch( "relocations for section %d using symtab %d\n",
                           target_shndx, symtab_shndx ));
@@ -4308,35 +4334,20 @@
 static int
 ocResolve_ELF ( ObjectCode* oc )
 {
-   char *strtab;
    int   shnum, ok;
-   Elf_Sym*  stab  = NULL;
    char*     ehdrC = (char*)(oc->image);
    Elf_Ehdr* ehdr  = (Elf_Ehdr*) ehdrC;
    Elf_Shdr* shdr  = (Elf_Shdr*) (ehdrC + ehdr->e_shoff);
 
-   /* first find "the" symbol table */
-   stab = (Elf_Sym*) findElfSection ( ehdrC, SHT_SYMTAB );
-
-   /* also go find the string table */
-   strtab = findElfSection ( ehdrC, SHT_STRTAB );
-
-   if (stab == NULL || strtab == NULL) {
-      errorBelch("%s: can't find string or symbol table", oc->fileName);
-      return 0;
-   }
-
    /* Process the relocation sections. */
    for (shnum = 0; shnum < ehdr->e_shnum; shnum++) {
       if (shdr[shnum].sh_type == SHT_REL) {
-         ok = do_Elf_Rel_relocations ( oc, ehdrC, shdr,
-                                       shnum, stab, strtab );
+         ok = do_Elf_Rel_relocations ( oc, ehdrC, shdr, shnum );
          if (!ok) return ok;
       }
       else
       if (shdr[shnum].sh_type == SHT_RELA) {
-         ok = do_Elf_Rela_relocations ( oc, ehdrC, shdr,
-                                        shnum, stab, strtab );
+         ok = do_Elf_Rela_relocations ( oc, ehdrC, shdr, shnum );
          if (!ok) return ok;
       }
    }
@@ -4369,8 +4380,12 @@
 
   if( i == ehdr->e_shnum )
   {
-    errorBelch( "This ELF file contains no symtab" );
-    return 0;
+    // Not having a symbol table is not in principle a problem.
+    // When an object file has no symbols then the 'strip' program
+    // typically will remove the symbol table entirely.
+    IF_DEBUG(linker, debugBelch( "The ELF file %s contains no symtab\n",
+             oc->archiveMemberName ? oc->archiveMemberName : oc->fileName ));
+    return 1;
   }
 
   if( shdr[i].sh_entsize != sizeof( Elf_Sym ) )
