[PATCH 6/8] gdbserver build-id attribute generator

[Aleksandar Ristovski <ARistovski@qnx.com>]

Notable changes: removed predicated find_phdr_p and changed find_phdr
to look for a given p_type.

    	* doc/gdb.texinfo (Library List Format for SVR4 Targets): Add
    	'build-id' in description, example, new attribute in dtd.
    	* features/library-list-svr4.dtd (library-list-svr4): New
    	'build-id' attribute.
    	* linux-low.c (linux-maps.h, search.h): Include.
	(ElfXX_Ehdr, ElfXX_Phdr, ElfXX_Nhdr): New.
	(ELFXX_FLD, ELFXX_SIZEOF, ELFXX_ROUNDUP, BUILD_ID_INVALID): New.
    	(find_phdr): New.
    	(get_dynamic): Use find_pdhr to traverse program headers.
    	(struct mapping_entry): New structure.
    	(mapping_entry_s): New typedef, new vector type def.
    	(free_mapping_entry, compare_mapping_entry,
    	 compare_mapping_entry_range, compare_mapping_entry_inode): New.
    	(struct find_memory_region_callback_data): New.
    	(find_memory_region_callback): New fwd. declaration.
    	(read_build_id, find_memory_region_callback, get_hex_build_id): New.
    	(linux_qxfer_libraries_svr4): Add optional build-id attribute
    	to reply XML document.
---
 gdb/doc/gdb.texinfo                |   17 +-
 gdb/features/library-list-svr4.dtd |   13 +-
 gdb/gdbserver/linux-low.c          |  373 +++++++++++++++++++++++++++++++++---
 3 files changed, 366 insertions(+), 37 deletions(-)

diff --git a/gdb/doc/gdb.texinfo b/gdb/doc/gdb.texinfo
index bf7e25e..2920f63 100644
--- a/gdb/doc/gdb.texinfo
+++ b/gdb/doc/gdb.texinfo
@@ -40359,6 +40359,8 @@ memory address.  It is a displacement of absolute memory address against
 address the file was prelinked to during the library load.
 @item
 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
+@item
+@code{build-id}, hex encoded @code{NT_GNU_BUILD_ID} note, if it exists.
 @end itemize
 
 Additionally the single @code{main-lm} attribute specifies address of
@@ -40376,7 +40378,7 @@ looks like this:
   <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
            l_ld="0xe4eefc"/>
   <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
-           l_ld="0x152350"/>
+           l_ld="0x152350" build-id="9afccf7cc41e6293476223fe72480854"/>
 </library-list-svr>
 @end smallexample
 
@@ -40385,13 +40387,14 @@ The format of an SVR4 library list is described by this DTD:
 @smallexample
 <!-- library-list-svr4: Root element with versioning -->
 <!ELEMENT library-list-svr4  (library)*>
-<!ATTLIST library-list-svr4  version CDATA   #FIXED  "1.0">
-<!ATTLIST library-list-svr4  main-lm CDATA   #IMPLIED>
+<!ATTLIST library-list-svr4  version  CDATA   #FIXED  "1.0">
+<!ATTLIST library-list-svr4  main-lm  CDATA   #IMPLIED>
 <!ELEMENT library            EMPTY>
-<!ATTLIST library            name    CDATA   #REQUIRED>
-<!ATTLIST library            lm      CDATA   #REQUIRED>
-<!ATTLIST library            l_addr  CDATA   #REQUIRED>
-<!ATTLIST library            l_ld    CDATA   #REQUIRED>
+<!ATTLIST library            name     CDATA   #REQUIRED>
+<!ATTLIST library            lm       CDATA   #REQUIRED>
+<!ATTLIST library            l_addr   CDATA   #REQUIRED>
+<!ATTLIST library            l_ld     CDATA   #REQUIRED>
+<!ATTLIST library            build-id CDATA   #IMPLIED>
 @end smallexample
 
 @node Memory Map Format
diff --git a/gdb/features/library-list-svr4.dtd b/gdb/features/library-list-svr4.dtd
index cae7fd8..fdd6ec0 100644
--- a/gdb/features/library-list-svr4.dtd
+++ b/gdb/features/library-list-svr4.dtd
@@ -6,11 +6,12 @@
 
 <!-- library-list-svr4: Root element with versioning -->
 <!ELEMENT library-list-svr4  (library)*>
-<!ATTLIST library-list-svr4  version CDATA   #FIXED  "1.0">
-<!ATTLIST library-list-svr4  main-lm CDATA   #IMPLIED>
+<!ATTLIST library-list-svr4  version  CDATA   #FIXED  "1.0">
+<!ATTLIST library-list-svr4  main-lm  CDATA   #IMPLIED>
 
 <!ELEMENT library            EMPTY>
-<!ATTLIST library            name    CDATA   #REQUIRED>
-<!ATTLIST library            lm      CDATA   #REQUIRED>
-<!ATTLIST library            l_addr  CDATA   #REQUIRED>
-<!ATTLIST library            l_ld    CDATA   #REQUIRED>
+<!ATTLIST library            name     CDATA   #REQUIRED>
+<!ATTLIST library            lm       CDATA   #REQUIRED>
+<!ATTLIST library            l_addr   CDATA   #REQUIRED>
+<!ATTLIST library            l_ld     CDATA   #REQUIRED>
+<!ATTLIST library            build-id CDATA   #IMPLIED>
diff --git a/gdb/gdbserver/linux-low.c b/gdb/gdbserver/linux-low.c
index 72c51e0..d819b21 100644
--- a/gdb/gdbserver/linux-low.c
+++ b/gdb/gdbserver/linux-low.c
@@ -20,6 +20,7 @@
 #include "linux-low.h"
 #include "linux-osdata.h"
 #include "agent.h"
+#include "linux-maps.h"
 
 #include "gdb_wait.h"
 #include <stdio.h>
@@ -43,6 +44,7 @@
 #include "gdb_stat.h"
 #include <sys/vfs.h>
 #include <sys/uio.h>
+#include <search.h>
 #ifndef ELFMAG0
 /* Don't include <linux/elf.h> here.  If it got included by gdb_proc_service.h
    then ELFMAG0 will have been defined.  If it didn't get included by
@@ -118,6 +120,33 @@ typedef struct
 } Elf64_auxv_t;
 #endif
 
+typedef union ElfXX_Ehdr
+{
+  Elf32_Ehdr _32;
+  Elf64_Ehdr _64;
+} ElfXX_Ehdr;
+
+typedef union ElfXX_Phdr
+{
+  Elf32_Phdr _32;
+  Elf64_Phdr _64;
+} ElfXX_Phdr;
+
+typedef union ElfXX_Nhdr
+{
+  Elf32_Nhdr _32;
+  Elf64_Nhdr _64;
+} ElfXX_Nhdr;
+
+#define ELFXX_FLD(hdr, fld) ((is_elf64) ? (hdr)._64.fld : (hdr)._32.fld)
+#define ELFXX_SIZEOF(hdr) ((is_elf64) ? sizeof ((hdr)._64) \
+				      : sizeof ((hdr)._32))
+#define ELFXX_ROUNDUP(what) ((is_elf64) ? (((what) + sizeof (Elf64_Word) - 1) \
+					   & ~(sizeof (Elf64_Word) - 1))      \
+					: (((what) + sizeof (Elf32_Word) - 1) \
+					   & ~(sizeof (Elf32_Word) - 1)))
+#define BUILD_ID_INVALID "?"
+
 /* ``all_threads'' is keyed by the LWP ID, which we use as the GDB protocol
    representation of the thread ID.
 
@@ -5432,15 +5461,43 @@ get_phdr_phnum_from_proc_auxv (const int pid, const int is_elf64,
   return 0;
 }
 
+/* Linearly traverse pheaders given in PHDR until supplied
+   predicate function returns 1.  If supplied predicate function
+   did return 1, stop traversal and return that PHDR.  */
+
+static const void *
+find_phdr (const int is_elf64, const void *const phdr_begin,
+	   const void *const phdr_end, const ULONGEST p_type)
+{
+#define PHDR_NEXT(hdrp) ((const void *) ((const gdb_byte *) (hdrp) + \
+			 ELFXX_SIZEOF (*hdrp)))
+
+  const ElfXX_Phdr *phdr = phdr_begin;
+
+  if (phdr == NULL)
+    return NULL;
+
+  while (PHDR_NEXT (phdr) <= phdr_end)
+    {
+      if (ELFXX_FLD (*phdr, p_type) == p_type)
+	return phdr;
+      phdr = PHDR_NEXT (phdr);
+    }
+
+  return NULL;
+#undef PHDR_NEXT
+}
+
 /* Return &_DYNAMIC (via PT_DYNAMIC) in the inferior, or 0 if not present.  */
 
 static CORE_ADDR
 get_dynamic (const int pid, const int is_elf64)
 {
   CORE_ADDR phdr_memaddr, relocation;
-  int num_phdr, i;
+  int num_phdr;
   unsigned char *phdr_buf;
   const int phdr_size = is_elf64 ? sizeof (Elf64_Phdr) : sizeof (Elf32_Phdr);
+  const void *phdr;
 
   if (get_phdr_phnum_from_proc_auxv (pid, is_elf64, &phdr_memaddr, &num_phdr))
     return 0;
@@ -5454,21 +5511,23 @@ get_dynamic (const int pid, const int is_elf64)
   /* Compute relocation: it is expected to be 0 for "regular" executables,
      non-zero for PIE ones.  */
   relocation = -1;
-  for (i = 0; relocation == -1 && i < num_phdr; i++)
-    if (is_elf64)
-      {
-	Elf64_Phdr *const p = (Elf64_Phdr *) (phdr_buf + i * phdr_size);
+  phdr = find_phdr (is_elf64, phdr_buf, phdr_buf + num_phdr * phdr_size,
+		    PT_PHDR);
+  if (phdr != NULL)
+    {
+      if (is_elf64)
+	{
+	  const Elf64_Phdr *const p = phdr;
 
-	if (p->p_type == PT_PHDR)
 	  relocation = phdr_memaddr - p->p_vaddr;
-      }
-    else
-      {
-	Elf32_Phdr *const p = (Elf32_Phdr *) (phdr_buf + i * phdr_size);
+	}
+      else
+	{
+	  const Elf32_Phdr *const p = phdr;
 
-	if (p->p_type == PT_PHDR)
 	  relocation = phdr_memaddr - p->p_vaddr;
-      }
+	}
+    }
 
   if (relocation == -1)
     {
@@ -5485,21 +5544,22 @@ get_dynamic (const int pid, const int is_elf64)
       return 0;
     }
 
-  for (i = 0; i < num_phdr; i++)
+  phdr = find_phdr (is_elf64, phdr_buf, phdr_buf + num_phdr * phdr_size,
+		    PT_DYNAMIC);
+
+  if (phdr != NULL)
     {
       if (is_elf64)
 	{
-	  Elf64_Phdr *const p = (Elf64_Phdr *) (phdr_buf + i * phdr_size);
+	  const Elf64_Phdr *const p = phdr;
 
-	  if (p->p_type == PT_DYNAMIC)
-	    return p->p_vaddr + relocation;
+	  return p->p_vaddr + relocation;
 	}
       else
 	{
-	  Elf32_Phdr *const p = (Elf32_Phdr *) (phdr_buf + i * phdr_size);
+	  const Elf32_Phdr *const p = phdr;
 
-	  if (p->p_type == PT_DYNAMIC)
-	    return p->p_vaddr + relocation;
+	  return p->p_vaddr + relocation;
 	}
     }
 
@@ -5641,6 +5701,253 @@ struct link_map_offsets
     int l_prev_offset;
   };
 
+
+/* Structure for holding a mapping.  Only mapping
+   containing l_ld can have hex_build_id set.  */
+
+struct mapping_entry
+{
+  /* Fields are populated from linux_find_memory_region parameters.  */
+
+  ULONGEST vaddr;
+  ULONGEST size;
+  ULONGEST offset;
+  ULONGEST inode;
+
+  /* Hex encoded string allocated using xmalloc, and
+     needs to be freed.  It can be NULL.  */
+
+  char *hex_build_id;
+};
+
+typedef struct mapping_entry mapping_entry_s;
+
+DEF_VEC_O(mapping_entry_s);
+
+static void
+free_mapping_entry (VEC (mapping_entry_s) *lst)
+{
+  int ix;
+  mapping_entry_s *p;
+
+  for (ix = 0; VEC_iterate (mapping_entry_s, lst, ix, p); ++ix)
+    xfree (p->hex_build_id);
+
+  VEC_free (mapping_entry_s, lst);
+}
+
+/* Used for finding a mapping containing the given
+   l_ld passed in K.  */
+
+static int
+compare_mapping_entry_range (const void *const k, const void *const b)
+{
+  const ULONGEST key = *(CORE_ADDR*) k;
+  const mapping_entry_s *const p = b;
+
+  if (key < p->vaddr)
+    return -1;
+
+  if (key < p->vaddr + p->size)
+    return 0;
+
+  return 1;
+}
+
+struct find_memory_region_callback_data
+{
+  unsigned is_elf64;
+
+  /* Return.  Ordered list of all object mappings sorted in
+     ascending order by VADDR.  Must be freed with free_mapping_entry.  */
+  VEC (mapping_entry_s) *list;
+};
+
+static linux_find_memory_region_ftype find_memory_region_callback;
+
+/* Read build-id from PT_NOTE.  */
+
+static void
+read_build_id (struct find_memory_region_callback_data *const p,
+	       mapping_entry_s *const bil, const CORE_ADDR load_addr,
+	       const CORE_ADDR l_addr)
+{
+  const int is_elf64 = p->is_elf64;
+  ElfXX_Ehdr ehdr;
+
+  if (linux_read_memory (load_addr, (unsigned char *) &ehdr,
+			 ELFXX_SIZEOF (ehdr)) == 0
+      && ELFXX_FLD (ehdr, e_ident[EI_MAG0]) == ELFMAG0
+      && ELFXX_FLD (ehdr, e_ident[EI_MAG1]) == ELFMAG1
+      && ELFXX_FLD (ehdr, e_ident[EI_MAG2]) == ELFMAG2
+      && ELFXX_FLD (ehdr, e_ident[EI_MAG3]) == ELFMAG3)
+    {
+      const ElfXX_Phdr *phdr;
+      void *phdr_buf;
+      const unsigned e_phentsize = ELFXX_FLD (ehdr, e_phentsize);
+
+      gdb_assert (ELFXX_FLD (ehdr, e_phnum) < 100);  /* Basic sanity check.  */
+      gdb_assert (e_phentsize == ELFXX_SIZEOF (*phdr));
+      phdr_buf = alloca (ELFXX_FLD (ehdr, e_phnum) * e_phentsize);
+
+      if (linux_read_memory (load_addr + ELFXX_FLD (ehdr, e_phoff), phdr_buf,
+			     ELFXX_FLD (ehdr, e_phnum) * e_phentsize) != 0)
+	{
+	  warning ("Could not read program header.");
+	  return;
+	}
+
+      phdr = phdr_buf;
+
+      for (;;)
+	{
+	  gdb_byte *pt_note;
+	  const gdb_byte *pt_end;
+	  const ElfXX_Nhdr *nhdr;
+
+	  phdr = find_phdr (p->is_elf64, phdr, (gdb_byte *) phdr_buf
+			    + ELFXX_FLD (ehdr, e_phnum) * e_phentsize,
+			    PT_NOTE);
+	  if (phdr == NULL)
+	    break;
+	  pt_note = xmalloc (ELFXX_FLD (*phdr, p_memsz));
+	  pt_end = (gdb_byte*) pt_note + ELFXX_FLD (*phdr, p_memsz);
+
+	  if (linux_read_memory (ELFXX_FLD (*phdr, p_vaddr) + l_addr, pt_note,
+				 ELFXX_FLD (*phdr, p_memsz)) != 0)
+	    {
+	      xfree (pt_note);
+	      warning ("Could not read note.");
+	      break;
+	    }
+
+	  nhdr = (void *) pt_note;
+	  while ((void *) nhdr < (void *) pt_end)
+	    {
+	      const size_t namesz
+		= ELFXX_ROUNDUP (ELFXX_FLD (*nhdr, n_namesz));
+	      const size_t descsz
+		= ELFXX_ROUNDUP (ELFXX_FLD (*nhdr, n_descsz));
+	      const size_t note_sz = ELFXX_SIZEOF (*nhdr) + namesz + descsz;
+
+	      if (((gdb_byte *) nhdr + note_sz) > pt_end || note_sz == 0
+		  || descsz == 0)
+		{
+		  warning ("Malformed PT_NOTE\n");
+		  break;
+		}
+	      if (ELFXX_FLD (*nhdr, n_type) == NT_GNU_BUILD_ID
+		  && ELFXX_FLD (*nhdr, n_namesz) == 4)
+		{
+		  const char gnu[4] = "GNU\0";
+		  const char *const pname
+		    = (char *) nhdr + ELFXX_SIZEOF (*nhdr);
+
+		  if (memcmp (pname, gnu, 4) == 0)
+		    {
+		      const size_t n_descsz = ELFXX_FLD (*nhdr, n_descsz);
+
+		      bil->hex_build_id = xmalloc (n_descsz * 2 + 1);
+		      bin2hex ((gdb_byte*) pname + namesz, bil->hex_build_id,
+					   n_descsz);
+		      xfree (pt_note);
+		      return;
+		    }
+		}
+	      nhdr = (void*) ((gdb_byte *) nhdr + note_sz);
+	    }
+	  xfree (pt_note);
+	}
+    }
+}
+
+/* Add mapping_entry.  */
+
+static int
+find_memory_region_callback (ULONGEST vaddr, ULONGEST size, ULONGEST offset,
+			     ULONGEST inode, int read, int write, int exec,
+			     int modified, const char *filename, void *data)
+{
+  if (inode != 0)
+    {
+      struct find_memory_region_callback_data *const p = data;
+      mapping_entry_s bil;
+
+      bil.vaddr = vaddr;
+      bil.size = size;
+      bil.offset = offset;
+      bil.inode = inode;
+      bil.hex_build_id = NULL;
+
+      VEC_safe_push (mapping_entry_s, p->list, &bil);
+    }
+
+  /* Continue the traversal.  */
+  return 0;
+}
+
+/* Linear reverse find starting from RBEGIN towards REND looking for
+   the lowest vaddr mapping of the same inode and zero offset.  */
+
+static mapping_entry_s *
+lrfind_mapping_entry (mapping_entry_s *const rbegin,
+		      const mapping_entry_s *const rend)
+{
+  mapping_entry_s *p;
+
+  for (p = rbegin - 1; p >= rend; --p)
+    if (p->offset == 0 && p->inode == rbegin->inode)
+      return p;
+
+  return NULL;
+}
+
+/* Get build-id for the given L_LD.  DATA must point to
+   already filled list of mapping_entry elements.
+
+   Return build_id as stored in the list element corresponding
+   to L_LD.
+
+   NULL may be returned if build-id could not be fetched.
+
+   Returned string must not be freed explicitly.  */
+
+static const char *
+get_hex_build_id (const CORE_ADDR l_addr, const CORE_ADDR l_ld,
+		  struct find_memory_region_callback_data *const data)
+{
+  mapping_entry_s *bil;
+
+  if (VEC_address (mapping_entry_s, data->list) == NULL)
+    return NULL;
+
+  bil = bsearch (&l_ld, VEC_address (mapping_entry_s, data->list),
+		 VEC_length (mapping_entry_s, data->list),
+		 sizeof (mapping_entry_s), compare_mapping_entry_range);
+
+  if (bil == NULL)
+    return NULL;
+
+  if (bil->hex_build_id == NULL)
+    {
+      mapping_entry_s *bil_min;
+
+      bil_min = lrfind_mapping_entry (bil, VEC_address (mapping_entry_s,
+							data->list));
+      if (bil_min != NULL)
+	read_build_id (data, bil, bil_min->vaddr, l_addr);
+      else
+	{
+	  /* Do not try to find hex_build_id again.  */
+	  bil->hex_build_id = xstrdup (BUILD_ID_INVALID);
+	  warning ("Could not determine load address; "
+		   "build-id can not be used.");
+	}
+    }
+
+  return bil->hex_build_id;
+}
+
 /* Construct qXfer:libraries-svr4:read reply.  */
 
 static int
@@ -5653,6 +5960,7 @@ linux_qxfer_libraries_svr4 (const char *annex, unsigned char *readbuf,
   struct process_info_private *const priv = current_process ()->private;
   char filename[PATH_MAX];
   int pid, is_elf64;
+  struct find_memory_region_callback_data data;
 
   static const struct link_map_offsets lmo_32bit_offsets =
     {
@@ -5688,6 +5996,13 @@ linux_qxfer_libraries_svr4 (const char *annex, unsigned char *readbuf,
   is_elf64 = elf_64_file_p (filename, &machine);
   lmo = is_elf64 ? &lmo_64bit_offsets : &lmo_32bit_offsets;
 
+  data.is_elf64 = is_elf64;
+  data.list = NULL;
+  VEC_reserve (mapping_entry_s, data.list, 16);
+  if (linux_find_memory_regions_full (pid, find_memory_region_callback, &data,
+      NULL) < 0)
+    warning ("Finding memory regions failed");
+
   if (priv->r_debug == 0)
     priv->r_debug = get_r_debug (pid, is_elf64);
 
@@ -5762,6 +6077,7 @@ linux_qxfer_libraries_svr4 (const char *annex, unsigned char *readbuf,
 	      /* 6x the size for xml_escape_text below.  */
 	      size_t len = 6 * strlen ((char *) libname);
 	      char *name;
+	      const char *hex_enc_build_id = NULL;
 
 	      if (!header_done)
 		{
@@ -5770,21 +6086,29 @@ linux_qxfer_libraries_svr4 (const char *annex, unsigned char *readbuf,
 		  header_done = 1;
 		}
 
-	      while (allocated < p - document + len + 200)
+	      name = xml_escape_text ((char *) libname);
+	      hex_enc_build_id = get_hex_build_id (l_addr, l_ld, &data);
+
+	      while (allocated < (p - document + len + 200
+				  + (hex_enc_build_id != NULL
+				     ? strlen (hex_enc_build_id) : 0)))
 		{
 		  /* Expand to guarantee sufficient storage.  */
-		  uintptr_t document_len = p - document;
+		  const ptrdiff_t document_len = p - document;
 
-		  document = xrealloc (document, 2 * allocated);
 		  allocated *= 2;
+		  document = xrealloc (document, allocated);
 		  p = document + document_len;
 		}
 
-	      name = xml_escape_text ((char *) libname);
 	      p += sprintf (p, "<library name=\"%s\" lm=\"0x%lx\" "
-			       "l_addr=\"0x%lx\" l_ld=\"0x%lx\"/>",
+			       "l_addr=\"0x%lx\" l_ld=\"0x%lx\"",
 			    name, (unsigned long) lm_addr,
 			    (unsigned long) l_addr, (unsigned long) l_ld);
+	      if (hex_enc_build_id != NULL
+		  && strcmp (hex_enc_build_id, BUILD_ID_INVALID) != 0)
+		p += sprintf (p, " build-id=\"%s\"", hex_enc_build_id);
+	      p += sprintf(p, "/>");
 	      free (name);
 	    }
 	  else if (lm_prev == 0)
@@ -5819,6 +6143,7 @@ linux_qxfer_libraries_svr4 (const char *annex, unsigned char *readbuf,
 
   memcpy (readbuf, document + offset, len);
   xfree (document);
+  free_mapping_entry (data.list);
 
   return len;
 }
-- 
1.7.10.4