RE: [PATCH v3 3/3] btrace, vdso: add vdso target sections

["Metzger, Markus T" <markus.t.metzger@intel.com>]

> -----Original Message-----
> From: Pedro Alves [mailto:palves@redhat.com]
> Sent: Thursday, May 29, 2014 7:04 PM


> Hmm, wait, nothing is removing the vdso symbols target sections, like ever,
> not even when the process dies, isn't it?  I guess that's it.

Yes, that seems to be it.

Target sections are typically owned by an so_list object and are removed
when the so_list object is removed.

Symfile target sections are owned by an objfile object and are removed
in a free_objfile observer in symfile.c.  That observer only removes target
sections for OBJF_USERLOADED objfiles, though.

The vdso target sections are owned by the vdso BFD and are never removed.

When I change symfile_add_from_memory to create target sections like this:

	add_target_sections_of_objfile (objf);

and further drop the restriction to only remove OBJF_USERLOADED objfiles in
symfile_free_objfile, the test passes.

diff --git a/gdb/symfile-mem.c b/gdb/symfile-mem.c
index b29421e..ef48f7d 100644
--- a/gdb/symfile-mem.c
+++ b/gdb/symfile-mem.c
@@ -92,7 +92,6 @@ symbol_file_add_from_memory (struct bfd *templ, CORE_ADDR addr,
   struct section_addr_info *sai;
   unsigned int i;
   struct cleanup *cleanup;
-  struct target_section *sections, *sections_end, *tsec;
 
   if (bfd_get_flavour (templ) != bfd_target_elf_flavour)
     error (_("add-symbol-file-from-memory not supported for this target"));
@@ -132,22 +131,7 @@ symbol_file_add_from_memory (struct bfd *templ, CORE_ADDR addr,
                                   from_tty ? SYMFILE_VERBOSE : 0,
                                    sai, OBJF_SHARED, NULL);
 
-  sections = NULL;
-  sections_end = NULL;
-
-  if (build_section_table (nbfd, &sections, &sections_end) == 0)
-    {
-      make_cleanup (xfree, sections);
-
-      /* Adjust the target section addresses by the load address.  */
-      for (tsec = sections; tsec != sections_end; ++tsec)
-       {
-         tsec->addr += loadbase;
-         tsec->endaddr += loadbase;
-       }
-
-      add_target_sections (&nbfd, sections, sections_end);
-    }
+  add_target_sections_of_objfile (objf);
 
   /* This might change our ideas about frames already looked at.  */
   reinit_frame_cache ();
diff --git a/gdb/symfile.c b/gdb/symfile.c
index 64a83c6..caa0722 100644
--- a/gdb/symfile.c
+++ b/gdb/symfile.c
@@ -3893,7 +3893,7 @@ static void
 symfile_free_objfile (struct objfile *objfile)
 {
   /* Remove the target sections of user-added objfiles.  */
-  if (objfile != 0 && objfile->flags & OBJF_USERLOADED)
+  if (objfile != NULL)
     remove_target_sections ((void *) objfile);
 }

Alternatively, I could add another OBJF_ flag and another free_objfile observer
to also remove the symfile-from-memory target sections.

Regards,
Markus.

Intel GmbH
Dornacher Strasse 1
85622 Feldkirchen/Muenchen, Deutschland
Sitz der Gesellschaft: Feldkirchen bei Muenchen
Geschaeftsfuehrer: Christian Lamprechter, Hannes Schwaderer, Douglas Lusk
Registergericht: Muenchen HRB 47456
Ust.-IdNr./VAT Registration No.: DE129385895
Citibank Frankfurt a.M. (BLZ 502 109 00) 600119052