From 0882710510059d9bf10d3e2324e0441029b50ce9 Mon Sep 17 00:00:00 2001
From: Mark Harmstone <mark@harmstone.com>
Date: Mon, 31 Oct 2022 00:15:54 +0000
Subject: ld: Add publics stream to PDB files

---
 ld/pdb.c                             | 347 ++++++++++++++++++++++++++++++++++-
 ld/pdb.h                             |  47 +++++
 ld/testsuite/ld-pe/pdb.exp           | 107 ++++++++++-
 ld/testsuite/ld-pe/pdb1-publics.d    |  41 +++++
 ld/testsuite/ld-pe/pdb1-sym-record.d |   7 +
 ld/testsuite/ld-pe/pdb1.s            |  18 +-
 6 files changed, 559 insertions(+), 8 deletions(-)
 create mode 100644 ld/testsuite/ld-pe/pdb1-publics.d
 create mode 100644 ld/testsuite/ld-pe/pdb1-sym-record.d

(limited to 'ld')

diff --git a/ld/pdb.c b/ld/pdb.c
index 1190dcf..8e151da 100644
--- a/ld/pdb.c
+++ b/ld/pdb.c
@@ -383,10 +383,31 @@ get_arch_number (bfd *abfd)
   return IMAGE_FILE_MACHINE_I386;
 }
 
+/* Return the index of a given output section.  */
+static uint16_t
+find_section_number (bfd *abfd, asection *sect)
+{
+  uint16_t i = 1;
+
+  for (asection *s = abfd->sections; s; s = s->next)
+    {
+      if (s == sect)
+	return i;
+
+      /* Empty sections aren't output.  */
+      if (s->size != 0)
+	i++;
+    }
+
+  return 0;
+}
+
 /* Stream 4 is the debug information (DBI) stream.  */
 static bool
 populate_dbi_stream (bfd *stream, bfd *abfd,
-		     uint16_t section_header_stream_num)
+		     uint16_t section_header_stream_num,
+		     uint16_t sym_rec_stream_num,
+		     uint16_t publics_stream_num)
 {
   struct pdb_dbi_stream_header h;
   struct optional_dbg_header opt;
@@ -396,9 +417,9 @@ populate_dbi_stream (bfd *stream, bfd *abfd,
   bfd_putl32 (1, &h.age);
   bfd_putl16 (0xffff, &h.global_stream_index);
   bfd_putl16 (0x8e1d, &h.build_number); // MSVC 14.29
-  bfd_putl16 (0xffff, &h.public_stream_index);
+  bfd_putl16 (publics_stream_num, &h.public_stream_index);
   bfd_putl16 (0, &h.pdb_dll_version);
-  bfd_putl16 (0xffff, &h.sym_record_stream);
+  bfd_putl16 (sym_rec_stream_num, &h.sym_record_stream);
   bfd_putl16 (0, &h.pdb_dll_rbld);
   bfd_putl32 (0, &h.mod_info_size);
   bfd_putl32 (0, &h.section_contribution_size);
@@ -433,6 +454,293 @@ populate_dbi_stream (bfd *stream, bfd *abfd,
   return true;
 }
 
+/* Used as parameter to qsort, to sort publics by hash.  */
+static int
+public_compare_hash (const void *s1, const void *s2)
+{
+  const struct public *p1 = *(const struct public **) s1;
+  const struct public *p2 = *(const struct public **) s2;
+
+  if (p1->hash < p2->hash)
+    return -1;
+  if (p1->hash > p2->hash)
+    return 1;
+
+  return 0;
+}
+
+/* Used as parameter to qsort, to sort publics by address.  */
+static int
+public_compare_addr (const void *s1, const void *s2)
+{
+  const struct public *p1 = *(const struct public **) s1;
+  const struct public *p2 = *(const struct public **) s2;
+
+  if (p1->section < p2->section)
+    return -1;
+  if (p1->section > p2->section)
+    return 1;
+
+  if (p1->address < p2->address)
+    return -1;
+  if (p1->address > p2->address)
+    return 1;
+
+  return 0;
+}
+
+/* The publics stream is a hash map of S_PUB32 records, which are stored
+   in the symbol record stream.  Each S_PUB32 entry represents a symbol
+   from the point of view of the linker: a section index, an offset within
+   the section, and a mangled name.  Compare with S_GDATA32 and S_GPROC32,
+   which are the same thing but generated by the compiler.  */
+static bool
+populate_publics_stream (bfd *stream, bfd *abfd, bfd *sym_rec_stream)
+{
+  struct publics_header header;
+  struct globals_hash_header hash_header;
+  const unsigned int num_buckets = 4096;
+  unsigned int num_entries = 0, filled_buckets = 0;
+  unsigned int buckets_size, sym_hash_size;
+  char int_buf[sizeof (uint32_t)];
+  struct public *publics_head = NULL, *publics_tail = NULL;
+  struct public **buckets;
+  struct public **sorted = NULL;
+  bool ret = false;
+
+  buckets = xmalloc (sizeof (struct public *) * num_buckets);
+  memset (buckets, 0, sizeof (struct public *) * num_buckets);
+
+  /* Loop through the global symbols in our input files, and write S_PUB32
+     records in the symbol record stream for those that make it into the
+     final image.  */
+  for (bfd *in = coff_data (abfd)->link_info->input_bfds; in;
+       in = in->link.next)
+    {
+      for (unsigned int i = 0; i < in->symcount; i++)
+	{
+	  struct bfd_symbol *sym = in->outsymbols[i];
+
+	  if (sym->flags & BSF_GLOBAL)
+	    {
+	      struct pubsym ps;
+	      uint16_t record_length;
+	      const char *name = sym->name;
+	      size_t name_len = strlen (name);
+	      struct public *p = xmalloc (sizeof (struct public));
+	      unsigned int padding = 0;
+	      uint16_t section;
+	      uint32_t flags = 0;
+
+	      section =
+		find_section_number (abfd, sym->section->output_section);
+
+	      if (section == 0)
+		continue;
+
+	      p->next = NULL;
+	      p->offset = bfd_tell (sym_rec_stream);
+	      p->hash = calc_hash (name, name_len) % num_buckets;
+	      p->section = section;
+	      p->address = sym->section->output_offset + sym->value;
+
+	      record_length = sizeof (struct pubsym) + name_len + 1;
+
+	      if (record_length % 4)
+		padding = 4 - (record_length % 4);
+
+	      /* Assume that all global symbols in executable sections
+		 are functions.  */
+	      if (sym->section->flags & SEC_CODE)
+		flags = PUBSYM_FUNCTION;
+
+	      bfd_putl16 (record_length + padding - sizeof (uint16_t),
+			  &ps.record_length);
+	      bfd_putl16 (S_PUB32, &ps.record_type);
+	      bfd_putl32 (flags, &ps.flags);
+	      bfd_putl32 (p->address, &ps.offset);
+	      bfd_putl16 (p->section, &ps.section);
+
+	      if (bfd_bwrite (&ps, sizeof (struct pubsym), sym_rec_stream) !=
+		  sizeof (struct pubsym))
+		goto end;
+
+	      if (bfd_bwrite (name, name_len + 1, sym_rec_stream) !=
+		  name_len + 1)
+		goto end;
+
+	      for (unsigned int j = 0; j < padding; j++)
+		{
+		  uint8_t b = 0;
+
+		  if (bfd_bwrite (&b, sizeof (uint8_t), sym_rec_stream) !=
+		      sizeof (uint8_t))
+		    goto end;
+		}
+
+	      if (!publics_head)
+		publics_head = p;
+	      else
+		publics_tail->next = p;
+
+	      publics_tail = p;
+	      num_entries++;
+	    }
+	}
+    }
+
+
+  if (num_entries > 0)
+    {
+      /* Create an array of pointers, sorted by hash value.  */
+
+      sorted = xmalloc (sizeof (struct public *) * num_entries);
+
+      struct public *p = publics_head;
+      for (unsigned int i = 0; i < num_entries; i++)
+	{
+	  sorted[i] = p;
+	  p = p->next;
+	}
+
+      qsort (sorted, num_entries, sizeof (struct public *),
+	     public_compare_hash);
+
+      /* Populate the buckets.  */
+
+      for (unsigned int i = 0; i < num_entries; i++)
+	{
+	  if (!buckets[sorted[i]->hash])
+	    {
+	      buckets[sorted[i]->hash] = sorted[i];
+	      filled_buckets++;
+	    }
+
+	  sorted[i]->index = i;
+	}
+    }
+
+  buckets_size = num_buckets / 8;
+  buckets_size += sizeof (uint32_t);
+  buckets_size += filled_buckets * sizeof (uint32_t);
+
+  sym_hash_size = sizeof (hash_header);
+  sym_hash_size += num_entries * sizeof (struct hash_record);
+  sym_hash_size += buckets_size;
+
+  /* Output the publics header.  */
+
+  bfd_putl32 (sym_hash_size, &header.sym_hash_size);
+  bfd_putl32 (num_entries * sizeof (uint32_t), &header.addr_map_size);
+  bfd_putl32 (0, &header.num_thunks);
+  bfd_putl32 (0, &header.thunks_size);
+  bfd_putl32 (0, &header.thunk_table);
+  bfd_putl32 (0, &header.thunk_table_offset);
+  bfd_putl32 (0, &header.num_sects);
+
+  if (bfd_bwrite (&header, sizeof (header), stream) != sizeof (header))
+    goto end;
+
+  /* Output the global hash header.  */
+
+  bfd_putl32 (GLOBALS_HASH_SIGNATURE, &hash_header.signature);
+  bfd_putl32 (GLOBALS_HASH_VERSION_70, &hash_header.version);
+  bfd_putl32 (num_entries * sizeof (struct hash_record),
+	      &hash_header.entries_size);
+  bfd_putl32 (buckets_size, &hash_header.buckets_size);
+
+  if (bfd_bwrite (&hash_header, sizeof (hash_header), stream) !=
+      sizeof (hash_header))
+    goto end;
+
+  /* Write the entries in hash order.  */
+
+  for (unsigned int i = 0; i < num_entries; i++)
+    {
+      struct hash_record hr;
+
+      bfd_putl32 (sorted[i]->offset + 1, &hr.offset);
+      bfd_putl32 (1, &hr.reference);
+
+      if (bfd_bwrite (&hr, sizeof (hr), stream) != sizeof (hr))
+	goto end;
+    }
+
+  /* Write the bitmap for filled and unfilled buckets.  */
+
+  for (unsigned int i = 0; i < num_buckets; i += 8)
+    {
+      uint8_t v = 0;
+
+      for (unsigned int j = 0; j < 8; j++)
+	{
+	  if (buckets[i + j])
+	    v |= 1 << j;
+	}
+
+      if (bfd_bwrite (&v, sizeof (v), stream) != sizeof (v))
+	goto end;
+    }
+
+  /* Add a 4-byte gap.  */
+
+  bfd_putl32 (0, int_buf);
+
+  if (bfd_bwrite (int_buf, sizeof (uint32_t), stream) != sizeof (uint32_t))
+    goto end;
+
+  /* Write the bucket offsets.  */
+
+  for (unsigned int i = 0; i < num_buckets; i++)
+    {
+      if (buckets[i])
+	{
+	  /* 0xc is size of internal hash_record structure in
+	     Microsoft's parser.  */
+	  bfd_putl32 (buckets[i]->index * 0xc, int_buf);
+
+	  if (bfd_bwrite (int_buf, sizeof (uint32_t), stream) !=
+	      sizeof (uint32_t))
+	    goto end;
+	}
+    }
+
+  /* Write the address map: offsets into the symbol record stream of
+     S_PUB32 records, ordered by address.  */
+
+  if (num_entries > 0)
+    {
+      qsort (sorted, num_entries, sizeof (struct public *),
+	     public_compare_addr);
+
+      for (unsigned int i = 0; i < num_entries; i++)
+	{
+	  bfd_putl32 (sorted[i]->offset, int_buf);
+
+	  if (bfd_bwrite (int_buf, sizeof (uint32_t), stream) !=
+	      sizeof (uint32_t))
+	    goto end;
+	}
+    }
+
+  ret = true;
+
+end:
+  free (buckets);
+
+  while (publics_head)
+    {
+      struct public *p = publics_head->next;
+
+      free (publics_head);
+      publics_head = p;
+    }
+
+  free (sorted);
+
+  return ret;
+}
+
 /* The section header stream contains a copy of the section headers
    from the PE file, in the same format.  */
 static bool
@@ -494,8 +802,9 @@ create_pdb_file (bfd *abfd, const char *pdb_name, const unsigned char *guid)
 {
   bfd *pdb;
   bool ret = false;
-  bfd *info_stream, *dbi_stream, *names_stream;
-  uint16_t section_header_stream_num;
+  bfd *info_stream, *dbi_stream, *names_stream, *sym_rec_stream,
+    *publics_stream;
+  uint16_t section_header_stream_num, sym_rec_stream_num, publics_stream_num;
 
   pdb = bfd_openw (pdb_name, "pdb");
   if (!pdb)
@@ -554,6 +863,24 @@ create_pdb_file (bfd *abfd, const char *pdb_name, const unsigned char *guid)
       goto end;
     }
 
+  sym_rec_stream = add_stream (pdb, NULL, &sym_rec_stream_num);
+
+  if (!sym_rec_stream)
+    {
+      einfo (_("%P: warning: cannot create symbol record stream "
+	       "in PDB file: %E\n"));
+      goto end;
+    }
+
+  publics_stream = add_stream (pdb, NULL, &publics_stream_num);
+
+  if (!publics_stream)
+    {
+      einfo (_("%P: warning: cannot create publics stream "
+	       "in PDB file: %E\n"));
+      goto end;
+    }
+
   if (!create_section_header_stream (pdb, abfd, &section_header_stream_num))
     {
       einfo (_("%P: warning: cannot create section header stream "
@@ -561,13 +888,21 @@ create_pdb_file (bfd *abfd, const char *pdb_name, const unsigned char *guid)
       goto end;
     }
 
-  if (!populate_dbi_stream (dbi_stream, abfd, section_header_stream_num))
+  if (!populate_dbi_stream (dbi_stream, abfd, section_header_stream_num,
+			    sym_rec_stream_num, publics_stream_num))
     {
       einfo (_("%P: warning: cannot populate DBI stream "
 	       "in PDB file: %E\n"));
       goto end;
     }
 
+  if (!populate_publics_stream (publics_stream, abfd, sym_rec_stream))
+    {
+      einfo (_("%P: warning: cannot populate publics stream "
+	       "in PDB file: %E\n"));
+      goto end;
+    }
+
   if (!populate_info_stream (pdb, info_stream, guid))
     {
       einfo (_("%P: warning: cannot populate info stream "
diff --git a/ld/pdb.h b/ld/pdb.h
index e5f53b4..1a80101 100644
--- a/ld/pdb.h
+++ b/ld/pdb.h
@@ -28,6 +28,8 @@
 #include "bfd.h"
 #include <stdbool.h>
 
+#define S_PUB32				0x110e
+
 /* PDBStream70 in pdb1.h */
 struct pdb_stream_70
 {
@@ -91,6 +93,51 @@ struct pdb_dbi_stream_header
 
 #define DBI_STREAM_VERSION_70		19990903
 
+/* PSGSIHDR in gsi.h */
+struct publics_header
+{
+  uint32_t sym_hash_size;
+  uint32_t addr_map_size;
+  uint32_t num_thunks;
+  uint32_t thunks_size;
+  uint32_t thunk_table;
+  uint32_t thunk_table_offset;
+  uint32_t num_sects;
+};
+
+/* GSIHashHdr in gsi.h */
+struct globals_hash_header
+{
+  uint32_t signature;
+  uint32_t version;
+  uint32_t entries_size;
+  uint32_t buckets_size;
+};
+
+/* HRFile in gsi.h */
+struct hash_record
+{
+  uint32_t offset;
+  uint32_t reference;
+};
+
+#define GLOBALS_HASH_SIGNATURE		0xffffffff
+#define GLOBALS_HASH_VERSION_70		0xf12f091a
+
+/* PUBSYM32 in cvinfo.h */
+struct pubsym
+{
+  uint16_t record_length;
+  uint16_t record_type;
+  uint32_t flags;
+  uint32_t offset;
+  uint16_t section;
+  /* followed by null-terminated string */
+} ATTRIBUTE_PACKED;
+
+/* see bitset CV_PUBSYMFLAGS in cvinfo.h */
+#define PUBSYM_FUNCTION			0x2
+
 struct optional_dbg_header
 {
   uint16_t fpo_stream;
diff --git a/ld/testsuite/ld-pe/pdb.exp b/ld/testsuite/ld-pe/pdb.exp
index cee0721..ee314c4 100644
--- a/ld/testsuite/ld-pe/pdb.exp
+++ b/ld/testsuite/ld-pe/pdb.exp
@@ -395,12 +395,111 @@ proc check_section_stream { img pdb } {
     return 1
 }
 
+proc get_publics_stream_index { pdb } {
+    global ar
+
+    set exec_output [run_host_cmd "$ar" "x --output tmpdir $pdb 0003"]
+
+    if ![string match "" $exec_output] {
+	return -1
+    }
+
+    set fi [open tmpdir/0003]
+    fconfigure $fi -translation binary
+
+    # skip fields
+    seek $fi 16
+
+    # read substream sizes
+
+    set data [read $fi 2]
+    binary scan $data s index
+
+    close $fi
+
+    return $index
+}
+
+proc get_sym_record_stream_index { pdb } {
+    global ar
+
+    set exec_output [run_host_cmd "$ar" "x --output tmpdir $pdb 0003"]
+
+    if ![string match "" $exec_output] {
+	return -1
+    }
+
+    set fi [open tmpdir/0003]
+    fconfigure $fi -translation binary
+
+    # skip fields
+    seek $fi 20
+
+    # read substream sizes
+
+    set data [read $fi 2]
+    binary scan $data s index
+
+    close $fi
+
+    return $index
+}
+
+proc check_publics_stream { pdb } {
+    global ar
+    global objdump
+    global srcdir
+    global subdir
+
+    set publics_index [get_publics_stream_index $pdb]
+
+    if { $publics_index == -1 } {
+	return 0
+    }
+
+    set index_str [format "%04x" $publics_index]
+
+    set exec_output [run_host_cmd "$ar" "x --output tmpdir $pdb $index_str"]
+
+    if ![string match "" $exec_output] {
+	return 0
+    }
+
+    set exp [file_contents "$srcdir/$subdir/pdb1-publics.d"]
+    set got [run_host_cmd "$objdump" "-s --target=binary tmpdir/$index_str"]
+    if ![string match $exp $got] {
+	return 0
+    }
+
+    set sym_record_index [get_sym_record_stream_index $pdb]
+
+    if { $sym_record_index == -1 } {
+	return 0
+    }
+
+    set index_str [format "%04x" $sym_record_index]
+
+    set exec_output [run_host_cmd "$ar" "x --output tmpdir $pdb $index_str"]
+
+    if ![string match "" $exec_output] {
+	return 0
+    }
+
+    set exp [file_contents "$srcdir/$subdir/pdb1-sym-record.d"]
+    set got [run_host_cmd "$objdump" "-s --target=binary tmpdir/$index_str"]
+    if ![string match $exp $got] {
+	return 0
+    }
+
+    return 1
+}
+
 if ![ld_assemble $as $srcdir/$subdir/pdb1.s tmpdir/pdb1.o] {
     unsupported "Build pdb1.o"
     return
 }
 
-if ![ld_link $ld "tmpdir/pdb1.exe" "--pdb=tmpdir/pdb1.pdb tmpdir/pdb1.o"] {
+if ![ld_link $ld "tmpdir/pdb1.exe" "--pdb=tmpdir/pdb1.pdb --gc-sections -e foo tmpdir/pdb1.o"] {
     fail "Could not create a PE image with a PDB file"
     return
 }
@@ -441,3 +540,9 @@ if [check_section_stream tmpdir/pdb1.exe tmpdir/pdb1.pdb] {
 } else {
     fail "Invalid section stream"
 }
+
+if [check_publics_stream tmpdir/pdb1.pdb] {
+    pass "Valid publics stream"
+} else {
+    fail "Invalid publics stream"
+}
diff --git a/ld/testsuite/ld-pe/pdb1-publics.d b/ld/testsuite/ld-pe/pdb1-publics.d
new file mode 100644
index 0000000..f7df2d9
--- /dev/null
+++ b/ld/testsuite/ld-pe/pdb1-publics.d
@@ -0,0 +1,41 @@
+
+*:     file format binary
+
+Contents of section .data:
+ 0000 2c020000 08000000 00000000 00000000  ,...............
+ 0010 00000000 00000000 00000000 ffffffff  ................
+ 0020 1a092ff1 10000000 0c020000 15000000  ../.............
+ 0030 01000000 01000000 01000000 00000000  ................
+ 0040 00000000 00000000 00000000 00000000  ................
+ 0050 00000000 00000000 00000000 00000000  ................
+ 0060 00000000 00000000 00000000 00000000  ................
+ 0070 00000000 00000000 00000000 00000000  ................
+ 0080 00000000 00000000 00000000 00000000  ................
+ 0090 00000000 00000000 00000000 00000000  ................
+ 00a0 00000000 00000000 00000000 00000000  ................
+ 00b0 00000000 00000000 00000000 00000000  ................
+ 00c0 00000000 00000000 00000000 00000000  ................
+ 00d0 00000000 00000000 00000000 00000001  ................
+ 00e0 00000000 00000000 00000000 00000000  ................
+ 00f0 00000000 00000000 00000000 00000000  ................
+ 0100 00000000 00000000 00000000 00000000  ................
+ 0110 00000000 00000000 00000000 00000000  ................
+ 0120 00000000 00000000 00000000 00000000  ................
+ 0130 00000000 00000000 00000000 00000000  ................
+ 0140 00000000 00000000 00000000 00000000  ................
+ 0150 00000000 00000000 00000000 00000000  ................
+ 0160 00000000 00000000 00000000 00000000  ................
+ 0170 00000000 00000000 00000000 00000000  ................
+ 0180 00000000 00000000 00000000 00000000  ................
+ 0190 00000000 00000000 00000000 01000000  ................
+ 01a0 00000000 00000000 00000000 00000000  ................
+ 01b0 00000000 00000000 00000000 00000000  ................
+ 01c0 00000000 00000000 00000000 00000000  ................
+ 01d0 00000000 00000000 00000000 00000000  ................
+ 01e0 00000000 00000000 00000000 00000000  ................
+ 01f0 00000000 00000000 00000000 00000000  ................
+ 0200 00000000 00000000 00000000 00000000  ................
+ 0210 00000000 00000000 00000000 00000000  ................
+ 0220 00000000 00000000 00000000 00000000  ................
+ 0230 00000000 00000000 00000000 00000000  ................
+ 0240 00000000 0c000000 00000000 14000000  ................
\ No newline at end of file
diff --git a/ld/testsuite/ld-pe/pdb1-sym-record.d b/ld/testsuite/ld-pe/pdb1-sym-record.d
new file mode 100644
index 0000000..2078a5e
--- /dev/null
+++ b/ld/testsuite/ld-pe/pdb1-sym-record.d
@@ -0,0 +1,7 @@
+
+*:     file format binary
+
+Contents of section .data:
+ 0000 12000e11 02000000 08000000 0100666f  ..............fo
+ 0010 6f000000 12000e11 00000000 04000000  o...............
+ 0020 02006261 72000000                    ..bar...        
\ No newline at end of file
diff --git a/ld/testsuite/ld-pe/pdb1.s b/ld/testsuite/ld-pe/pdb1.s
index 30a8cfc..846814b 100644
--- a/ld/testsuite/ld-pe/pdb1.s
+++ b/ld/testsuite/ld-pe/pdb1.s
@@ -1,5 +1,21 @@
 .text
 
+	.long 0x12345678
+	.long 0x9abcdef0
+
 .global foo
-foo:
+foo:	# section 0001, offset 00000008
+	.secrel32 bar
+
+.data
+	.long 0x12345678
+
+.global bar
+bar:	# section 0002, offset 00000004
+	.long 0x9abcdef0
+
+.section "gcsect"
+
+.global baz
+baz:	# unreferenced, will be GC'd out
 	.long 0x12345678
-- 
cgit v1.1