Re: [PATCH 1/4] python support for fetching separate debug files: build_id

[Doug Evans <dje at google dot com>]

Eli Zaretskii writes:
 > > From: Doug Evans <dje@google.com>
 > > Date: Thu, 20 Nov 2014 13:20:13 -0800
 > > 
 > > This patch just provides a Python API call to fetch the build id from
 > > objfiles.  In my use-case separate debug files are looked up via the
 > > build id.
 > 
 > OK for the documentation parts.
 > 
 > Thanks.

Thanks.

Here is what I committed.
[no real change from original]

commit 7c50a93137df660f7b2d9d68c0db748a9cb7868f
Author: Doug Evans <dje@google.com>
Date:   Thu Dec 4 11:32:24 2014 -0800

    New python attribute gdb.Objfile.build_id.
    
    gdb/ChangeLog:
    
    	* NEWS: Mention gdb.Objfile.build_id.
    	* build-id.c (build_id_bfd_get): Make non-static.
    	* build-id.h (build_id_bfd_get): Add declaration.
    	* python/py-objfile.c: #include "build-id.h", "elf-bfd.h".
    	(OBJFPY_REQUIRE_VALID): New macro.
    	(objfpy_get_build_id): New function.
    	(objfile_getset): Add "build_id".
    	* utils.c (make_hex_string): New function.
    	* utils.h (make_hex_string): Add declaration.
    
    gdb/doc/ChangeLog:
    
    	* python.texi (Objfiles In Python): Document Objfile.build_id.
    
    gdb/testsuite/ChangeLog:
    
    	* lib/gdb.exp (get_build_id): New function.
    	(build_id_debug_filename_get): Rewrite to use it.
    	* gdb.python/py-objfile.exp: Add test for objfile.build_id.

diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 352d107..e6da0b5 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,15 @@
+2014-12-04  Doug Evans  <dje@google.com>
+
+	* NEWS: Mention gdb.Objfile.build_id.
+	* build-id.c (build_id_bfd_get): Make non-static.
+	* build-id.h (build_id_bfd_get): Add declaration.
+	* python/py-objfile.c: #include "build-id.h", "elf-bfd.h".
+	(OBJFPY_REQUIRE_VALID): New macro.
+	(objfpy_get_build_id): New function.
+	(objfile_getset): Add "build_id".
+	* utils.c (make_hex_string): New function.
+	* utils.h (make_hex_string): Add declaration.
+
 2014-12-04  Jan Kratochvil  <jan.kratochvil@redhat.com>
 
 	* block.c (block_lookup_symbol_primary): New function.
diff --git a/gdb/NEWS b/gdb/NEWS
index 7262502..16ed91e 100644
--- a/gdb/NEWS
+++ b/gdb/NEWS
@@ -21,6 +21,8 @@
   ** New attribute 'producer' for gdb.Symtab objects.
   ** gdb.Objfile objects have a new attribute "progspace",
      which is the gdb.Progspace object of the containing program space.
+  ** gdb.Objfile objects have a new attribute "build_id",
+     which is the build ID generated when the file was built.
   ** A new event "gdb.clear_objfiles" has been added, triggered when
      selecting a new file to debug.
   ** You can now add attributes to gdb.Objfile and gdb.Progspace objects.
diff --git a/gdb/build-id.c b/gdb/build-id.c
index 0f553ce..5c1f415 100644
--- a/gdb/build-id.c
+++ b/gdb/build-id.c
@@ -27,9 +27,9 @@
 #include "objfiles.h"
 #include "filenames.h"
 
-/* Locate NT_GNU_BUILD_ID from ABFD and return its content.  */
+/* See build-id.h.  */
 
-static const struct elf_build_id *
+const struct elf_build_id *
 build_id_bfd_get (bfd *abfd)
 {
   if (!bfd_check_format (abfd, bfd_object)
diff --git a/gdb/build-id.h b/gdb/build-id.h
index ddd2645..548ea5e 100644
--- a/gdb/build-id.h
+++ b/gdb/build-id.h
@@ -20,6 +20,10 @@
 #ifndef BUILD_ID_H
 #define BUILD_ID_H
 
+/* Locate NT_GNU_BUILD_ID from ABFD and return its content.  */
+
+extern const struct elf_build_id *build_id_bfd_get (bfd *abfd);
+
 /* Return true if ABFD has NT_GNU_BUILD_ID matching the CHECK value.
    Otherwise, issue a warning and return false.  */
 
diff --git a/gdb/doc/ChangeLog b/gdb/doc/ChangeLog
index 08b49f5..7f36a31 100644
--- a/gdb/doc/ChangeLog
+++ b/gdb/doc/ChangeLog
@@ -1,3 +1,7 @@
+2014-12-04  Doug Evans  <dje@google.com>
+
+	* python.texi (Objfiles In Python): Document Objfile.build_id.
+
 2014-12-02  Nick Bull  <nicholaspbull@gmail.com>
 
 	* python.texi (Events In Python): Document new events
diff --git a/gdb/doc/python.texi b/gdb/doc/python.texi
index 63db2b2..a951410 100644
--- a/gdb/doc/python.texi
+++ b/gdb/doc/python.texi
@@ -3495,6 +3495,17 @@ class.
 The file name of the objfile as a string.
 @end defvar
 
+@defvar Objfile.build_id
+The build ID of the objfile as a string.
+If the objfile does not have a build ID then the value is @code{None}.
+
+This is supported only on some operating systems, notably those which use
+the ELF format for binary files and the @sc{gnu} Binutils.  For more details
+about this feature, see the description of the @option{--build-id}
+command-line option in @ref{Options, , Command Line Options, ld.info,
+The GNU Linker}.
+@end defvar
+
 @defvar Objfile.progspace
 The containing program space of the objfile as a @code{gdb.Progspace}
 object.  @xref{Progspaces In Python}.
diff --git a/gdb/python/py-objfile.c b/gdb/python/py-objfile.c
index c99de87..05a7c21 100644
--- a/gdb/python/py-objfile.c
+++ b/gdb/python/py-objfile.c
@@ -22,6 +22,8 @@
 #include "charset.h"
 #include "objfiles.h"
 #include "language.h"
+#include "build-id.h"
+#include "elf-bfd.h"
 
 typedef struct
 {
@@ -51,9 +53,21 @@ static PyTypeObject objfile_object_type
 
 static const struct objfile_data *objfpy_objfile_data_key;
 
+/* Require that OBJF be a valid objfile.  */
+#define OBJFPY_REQUIRE_VALID(obj)				\
+  do {								\
+    if (!(obj)->objfile)					\
+      {								\
+	PyErr_SetString (PyExc_RuntimeError,			\
+			 _("Objfile no longer exists."));	\
+	return NULL;						\
+      }								\
+  } while (0)
+
 
 
 /* An Objfile method which returns the objfile's file name, or None.  */
+
 static PyObject *
 objfpy_get_filename (PyObject *self, void *closure)
 {
@@ -66,6 +80,38 @@ objfpy_get_filename (PyObject *self, void *closure)
   Py_RETURN_NONE;
 }
 
+/* An Objfile method which returns the objfile's build id, or None.  */
+
+static PyObject *
+objfpy_get_build_id (PyObject *self, void *closure)
+{
+  objfile_object *obj = (objfile_object *) self;
+  struct objfile *objfile = obj->objfile;
+  const struct elf_build_id *build_id = NULL;
+  volatile struct gdb_exception except;
+
+  OBJFPY_REQUIRE_VALID (obj);
+
+  TRY_CATCH (except, RETURN_MASK_ALL)
+    {
+      build_id = build_id_bfd_get (objfile->obfd);
+    }
+  GDB_PY_HANDLE_EXCEPTION (except);
+
+  if (build_id != NULL)
+    {
+      char *hex_form = make_hex_string (build_id->data, build_id->size);
+      PyObject *result;
+
+      result = PyString_Decode (hex_form, strlen (hex_form),
+				host_charset (), NULL);
+      xfree (hex_form);
+      return result;
+    }
+
+  Py_RETURN_NONE;
+}
+
 /* An Objfile method which returns the objfile's progspace, or None.  */
 
 static PyObject *
@@ -364,6 +410,8 @@ static PyGetSetDef objfile_getset[] =
     "The __dict__ for this objfile.", &objfile_object_type },
   { "filename", objfpy_get_filename, NULL,
     "The objfile's filename, or None.", NULL },
+  { "build_id", objfpy_get_build_id, NULL,
+    "The objfile's build id, or None.", NULL },
   { "progspace", objfpy_get_progspace, NULL,
     "The objfile's progspace, or None.", NULL },
   { "pretty_printers", objfpy_get_printers, objfpy_set_printers,
diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index f1584e0..b19a5c0 100644
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,3 +1,9 @@
+2014-12-04  Doug Evans  <dje@google.com>
+
+	* lib/gdb.exp (get_build_id): New function.
+	(build_id_debug_filename_get): Rewrite to use it.
+	* gdb.python/py-objfile.exp: Add test for objfile.build_id.
+
 2014-12-04  Maciej W. Rozycki  <macro@codesourcery.com>
 
 	* gdb.cp/expand-psymtabs-cxx.exp: Accept any address of
diff --git a/gdb/testsuite/gdb.python/py-objfile.exp b/gdb/testsuite/gdb.python/py-objfile.exp
index 7bf41ed..74384ed 100644
--- a/gdb/testsuite/gdb.python/py-objfile.exp
+++ b/gdb/testsuite/gdb.python/py-objfile.exp
@@ -39,6 +39,16 @@ gdb_py_test_silent_cmd "python objfile = sym\[0\].symtab.objfile" \
 
 gdb_test "python print (objfile.filename)" ".*py-objfile.*" \
   "Get objfile file name"
+
+set binfile_build_id [get_build_id $binfile]
+if [string compare $binfile_build_id ""] {
+    verbose -log "binfile_build_id = $binfile_build_id"
+    gdb_test "python print (objfile.build_id)" "$binfile_build_id" \
+    "Get objfile build id"
+} else {
+    unsupported "build-id is not supported by the compiler"
+}
+
 gdb_test "python print (objfile.progspace)" "<gdb\.Progspace object at .*>" \
   "Get objfile program space"
 gdb_test "python print (objfile.is_valid())" "True" \
diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp
index 2c79bc1..a29b661 100644
--- a/gdb/testsuite/lib/gdb.exp
+++ b/gdb/testsuite/lib/gdb.exp
@@ -4330,14 +4330,14 @@ gdb_caching_proc gdb_has_argv0 {
 # foo.debug --> foo's debug info
 # foo --> like foo, but with a new .gnu_debuglink section pointing to foo.debug.
 
-# Return the build-id hex string (usually 160 bits as 40 hex characters)
-# converted to the form: .build-id/ab/cdef1234...89.debug
-# Return "" if no build-id found.
-proc build_id_debug_filename_get { exec } {
-    set tmp [standard_output_file "${exec}-tmp"]
+# Fetch the build id from the file.
+# Returns "" if there is none.
+
+proc get_build_id { filename } {
+    set tmp [standard_output_file "${filename}-tmp"]
     set objcopy_program [gdb_find_objcopy]
 
-    set result [catch "exec $objcopy_program -j .note.gnu.build-id -O binary $exec $tmp" output]
+    set result [catch "exec $objcopy_program -j .note.gnu.build-id -O binary $filename $tmp" output]
     verbose "result is $result"
     verbose "output is $output"
     if {$result == 1} {
@@ -4355,6 +4355,17 @@ proc build_id_debug_filename_get { exec } {
     }
     # Convert it to hex.
     binary scan $data H* data
+    return $data
+}
+
+# Return the build-id hex string (usually 160 bits as 40 hex characters)
+# converted to the form: .build-id/ab/cdef1234...89.debug
+# Return "" if no build-id found.
+proc build_id_debug_filename_get { filename } {
+    set data [get_build_id $filename]
+    if { $data == "" } {
+	return ""
+    }
     regsub {^..} $data {\0/} data
     return ".build-id/${data}.debug"
 }
diff --git a/gdb/utils.c b/gdb/utils.c
index 1f5f4f4..1ab183c 100644
--- a/gdb/utils.c
+++ b/gdb/utils.c
@@ -1112,6 +1112,23 @@ gdb_print_host_address (const void *addr, struct ui_file *stream)
 {
   fprintf_filtered (stream, "%s", host_address_to_string (addr));
 }
+
+/* See utils.h.  */
+
+char *
+make_hex_string (const gdb_byte *data, size_t length)
+{
+  char *result = xmalloc (length * 2 + 1);
+  char *p;
+  size_t i;
+
+  p = result;
+  for (i = 0; i < length; ++i)
+    p += sprintf (p, "%02x", data[i]);
+  *p = '\0';
+  return result;
+}
+
 
 
 /* A cleanup function that calls regfree.  */
diff --git a/gdb/utils.h b/gdb/utils.h
index 1568011..660b548 100644
--- a/gdb/utils.h
+++ b/gdb/utils.h
@@ -369,4 +369,9 @@ extern void warn_cant_dump_core (const char *reason);
 
 extern void dump_core (void);
 
+/* Return the hex string form of LENGTH bytes of DATA.
+   Space for the result is malloc'd, caller must free.  */
+
+extern char *make_hex_string (const gdb_byte *data, size_t length);
+
 #endif /* UTILS_H */