- Fix the case when GDB leaks memory because value_struct_elt does not call

check_typedef.  (Doug Evans, BZ 15695, filed as RH BZ 1013453).

gdb-dlopen-stap-probe-inhibit.patch

@@ -0,0 +1,60 @@
+https://bugzilla.redhat.com/show_bug.cgi?id=911712
+
+--- ./gdb/objfiles.c-orig	2013-03-21 18:51:00.141957331 +0100
++++ ./gdb/objfiles.c	2013-03-21 19:20:25.615519748 +0100
+@@ -615,6 +615,7 @@ free_objfile (struct objfile *objfile)
+   obstack_free (&objfile->objfile_obstack, 0);
+ 
+   /* Rebuild section map next time we need it.  */
++  gdb_assert (!get_objfile_pspace_data (objfile->pspace)->inhibit_updates);
+   get_objfile_pspace_data (objfile->pspace)->objfiles_changed_p = 1;
+ 
+   xfree (objfile);
+@@ -1284,6 +1285,21 @@ bsearch_cmp (const void *key, const void
+   return 1;
+ }
+ 
++static void
++update_space_info_sections (struct objfile_pspace_info *pspace_info)
++{
++  if (pspace_info->objfiles_changed_p && !pspace_info->inhibit_updates)
++    {
++      update_section_map (current_program_space,
++			  &pspace_info->sections,
++			  &pspace_info->num_sections);
++
++      /* Don't need updates to section map until objfiles are added,
++         removed or relocated.  */
++      pspace_info->objfiles_changed_p = 0;
++    }
++}
++
+ /* Returns a section whose range includes PC or NULL if none found.   */
+ 
+ struct obj_section *
+@@ -1298,16 +1314,7 @@ find_pc_section (CORE_ADDR pc)
+     return s;
+ 
+   pspace_info = get_objfile_pspace_data (current_program_space);
+-  if (pspace_info->objfiles_changed_p && !pspace_info->inhibit_updates)
+-    {
+-      update_section_map (current_program_space,
+-			  &pspace_info->sections,
+-			  &pspace_info->num_sections);
+-
+-      /* Don't need updates to section map until objfiles are added,
+-         removed or relocated.  */
+-      pspace_info->objfiles_changed_p = 0;
+-    }
++  update_space_info_sections (pspace_info);
+ 
+   /* The C standard (ISO/IEC 9899:TC2) requires the BASE argument to
+      bsearch be non-NULL.  */
+@@ -1471,6 +1478,7 @@ objfiles_changed (void)
+ void
+ inhibit_section_map_updates (void)
+ {
++  update_space_info_sections (get_objfile_pspace_data (current_program_space));
+   get_objfile_pspace_data (current_program_space)->inhibit_updates = 1;
+ }
+ 

gdb-f18-open-emptyname.patch

@@ -0,0 +1,20 @@
+gdb ""
+SEGV
+From archer.patch probably from archer-tromey-dwz-multifile-rebase .
+
+--- gdb-7.5.1/gdb/exec.c-orig	2013-04-12 16:58:46.426785095 +0200
++++ gdb-7.5.1/gdb/exec.c	2013-04-12 17:10:18.566927513 +0200
+@@ -226,11 +226,11 @@ exec_file_attach (char *filename, int fr
+ 	     &scratch_pathname);
+ 	}
+ #endif
++      if (scratch_chan < 0)
++	perror_with_name (filename);
+ 
+       cleanups = make_cleanup (xfree, scratch_pathname);
+ 
+-      if (scratch_chan < 0)
+-	perror_with_name (filename);
+       exec_bfd = gdb_bfd_fopen (scratch_pathname, gnutarget,
+ 				write_files ? FOPEN_RUB : FOPEN_RB,
+ 				scratch_chan);

gdb-rhbz1013453-value-struct-elt-memory-leak.patch

@@ -0,0 +1,136 @@
+https://sourceware.org/ml/gdb-patches/2013-07/msg00469.html
+
+Hi.
+This patch adds the missing calls to check_typedef
+and adds a testcase to show the issue.
+
+The PR is not closeable yet, but the remaining issues are more cleanups
+than actual bug fixes.
+
+Regression tested on amd64-linux.
+
+I will check this in in a few days if there are no objections.
+
+[The multiple calls to check_typedef (value_type (*argp)) are
+bothersome, but the code is simpler this way, and I expect
+resolving the rest of the issues in 15695 to potentially change this
+code significantly.]
+
+---
+
+commit cbb25189b69e501ddca64917d810b54bb1466c93
+Author: Doug Evans <dje@google.com>
+Date:   Thu Aug 1 23:59:47 2013 +0000
+
+    	PR symtab/15695
+    	* valops.c (value_struct_elt): Add missing call to check_typedef.
+    	(value_find_oload_method_list): Ditto.
+    
+    	testsuite/
+    	* gdb.base/func-ptr.exp: New file.
+    	* gdb.base/func-ptr.c: New file.
+
+2013-08-01  Doug Evans  <dje@google.com>
+ 
+	PR symtab/15695
+	* valops.c (value_struct_elt): Add missing call to check_typedef.
+	(value_find_oload_method_list): Ditto.
+
+2013-08-01  Doug Evans  <dje@google.com>
+
+	PR symtab/15695
+	* gdb.base/func-ptr.exp: New file.
+	* gdb.base/func-ptr.c: New file.
+
+Index: gdb-7.5.1/gdb/testsuite/gdb.base/func-ptr.c
+===================================================================
+--- /dev/null
++++ gdb-7.5.1/gdb/testsuite/gdb.base/func-ptr.c
+@@ -0,0 +1,30 @@
++/* This testcase is part of GDB, the GNU debugger.
++
++   Copyright 2013 Free Software Foundation, Inc.
++
++   This program is free software; you can redistribute it and/or modify
++   it under the terms of the GNU General Public License as published by
++   the Free Software Foundation; either version 3 of the License, or
++   (at your option) any later version.
++
++   This program is distributed in the hope that it will be useful,
++   but WITHOUT ANY WARRANTY; without even the implied warranty of
++   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
++   GNU General Public License for more details.
++
++   You should have received a copy of the GNU General Public License
++   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
++
++void
++bar ()
++{
++}
++
++typedef void foo (void);
++foo *pbar = bar;
++
++int
++main ()
++{
++  return 0;
++}
+Index: gdb-7.5.1/gdb/testsuite/gdb.base/func-ptr.exp
+===================================================================
+--- /dev/null
++++ gdb-7.5.1/gdb/testsuite/gdb.base/func-ptr.exp
+@@ -0,0 +1,30 @@
++# Copyright 2013 Free Software Foundation, Inc.
++
++# This program is free software; you can redistribute it and/or modify
++# it under the terms of the GNU General Public License as published by
++# the Free Software Foundation; either version 3 of the License, or
++# (at your option) any later version.
++#
++# This program is distributed in the hope that it will be useful,
++# but WITHOUT ANY WARRANTY; without even the implied warranty of
++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
++# GNU General Public License for more details.
++#
++# You should have received a copy of the GNU General Public License
++# along with this program.  If not, see <http://www.gnu.org/licenses/>.
++
++# This testcase exercises bug 15695.
++# Trying to print foo->bar if foo is a pointer to a typedef of a pointer
++# to a function will put gdb into an infinite loop.
++
++if { [prepare_for_testing func-ptr.exp "func-ptr" {func-ptr.c} {debug}] } {
++    return -1
++}
++
++if ![runto_main] {
++    fail "Can't run to main"
++    return 0
++}
++
++# This would put gdb into an infinite loop.
++gdb_test "print pbar->baz" "Attempt to extract .*"
+Index: gdb-7.5.1/gdb/valops.c
+===================================================================
+--- gdb-7.5.1.orig/gdb/valops.c
++++ gdb-7.5.1/gdb/valops.c
+@@ -2465,7 +2465,7 @@ value_struct_elt (struct value **argp, s
+     {
+       *argp = value_ind (*argp);
+       /* Don't coerce fn pointer to fn and then back again!  */
+-      if (TYPE_CODE (value_type (*argp)) != TYPE_CODE_FUNC)
++      if (TYPE_CODE (check_typedef (value_type (*argp))) != TYPE_CODE_FUNC)
+ 	*argp = coerce_array (*argp);
+       t = check_typedef (value_type (*argp));
+     }
+@@ -2629,7 +2629,7 @@ value_find_oload_method_list (struct val
+     {
+       *argp = value_ind (*argp);
+       /* Don't coerce fn pointer to fn and then back again!  */
+-      if (TYPE_CODE (value_type (*argp)) != TYPE_CODE_FUNC)
++      if (TYPE_CODE (check_typedef (value_type (*argp))) != TYPE_CODE_FUNC)
+ 	*argp = coerce_array (*argp);
+       t = check_typedef (value_type (*argp));
+     }

gdb-x86_64-i386-syscall-restart.patch

@@ -102,7 +102,7 @@ gdb/
  	}
      }
 +
-+  if (gdbarch_ptr_bit (gdbarch) == 32)
++  if (gdbarch_bfd_arch_info (gdbarch)->bits_per_word == 32)
 +    {
 +      /* Sign-extend %eax as during return from a syscall it is being checked
 +	 for -ERESTART* values -512 being above 0xfffffffffffffe00; tested by

gdb.spec

@@ -34,7 +34,7 @@ Version: 7.5.1
 
 # The release always contains a leading reserved number, start it at 1.
 # `upstream' is not a part of `name' to stay fully rpm dependencies compatible for the testing.
-Release: 35%{?dist}
+Release: 43%{?dist}
 
 License: GPLv3+ and GPLv3+ with exceptions and GPLv2+ and GPLv2+ with exceptions and GPL+ and LGPLv2+ and BSD and Public Domain
 Group: Development/Debuggers
@@ -511,6 +511,7 @@ Patch548: gdb-test-expr-cumulative-archer.patch
 Patch579: gdb-7.2.50-sparc-add-workaround-to-broken-debug-files.patch
 
 # Fix dlopen of libpthread.so, patched glibc required (Gary Benson, BZ 669432).
+# Fix crash regression from the dlopen of libpthread.so fix (BZ 911712).
 #=push
 Patch618: gdb-dlopen-stap-probe-1of7.patch
 Patch717: gdb-dlopen-stap-probe-2of7.patch
@@ -522,6 +523,7 @@ Patch722: gdb-dlopen-stap-probe-7of7.patch
 Patch619: gdb-dlopen-stap-probe-test.patch
 Patch723: gdb-dlopen-stap-probe-test2.patch
 Patch822: gdb-dlopen-stap-probe-mapfailed.patch
+Patch826: gdb-dlopen-stap-probe-inhibit.patch
 
 # Work around PR libc/13097 "linux-vdso.so.1" warning message.
 #=push
@@ -607,6 +609,14 @@ Patch819: gdb-implicitpointer-offset.patch
 Patch820: gdb-rhbz890900-ppc-dis-1of2.patch
 Patch821: gdb-rhbz890900-ppc-dis-2of2.patch
 
+# Fix crash on 'gdb ""' (BZ 951251)..
+Patch830: gdb-f18-open-emptyname.patch
+
+# Fix the case when GDB leaks memory because value_struct_elt
+# does not call check_typedef.  (Doug Evans, BZ 15695, filed as
+# RH BZ 1013453).
+Patch844: gdb-rhbz1013453-value-struct-elt-memory-leak.patch
+
 %if 0%{!?rhel:1} || 0%{?rhel} > 6
 # RL_STATE_FEDORA_GDB would not be found for:
 # Patch642: gdb-readline62-ask-more-rh.patch
@@ -909,6 +919,7 @@ find -name "*.info*"|xargs rm -f
 %patch722 -p1
 %patch723 -p1
 %patch822 -p1
+%patch826 -p1
 %patch619 -p1
 %patch627 -p1
 %patch634 -p1
@@ -948,6 +959,8 @@ find -name "*.info*"|xargs rm -f
 %patch819 -p1
 %patch820 -p1
 %patch821 -p1
+%patch830 -p1
+%patch844 -p1
 
 %patch393 -p1
 %if 0%{!?el5:1} || 0%{?scl:1}
@@ -970,9 +983,9 @@ find -name "*.orig" | xargs rm -f
 # Change the version that gets printed at GDB startup, so it is RH specific.
 cat > gdb/version.in << _FOO
 %if 0%{!?rhel:1}
-Fedora (%{version}-%{release})
+Fedora %{version}-%{release}
 %else # !0%{!?rhel:1} 
-Red Hat Enterprise Linux (%{version}-%{release})
+Red Hat Enterprise Linux %{version}-%{release}
 %endif # !0%{!?rhel:1} 
 _FOO
 
@@ -1153,8 +1166,7 @@ $(: fmtutil: format directory '/builddir/.texmf-var/web2c' does not exist. ) \
 %else
      %{?_smp_mflags} \
 %endif
-     -C gdb/doc {gdb,annotate}{.info,/index.html} MAKEHTMLFLAGS=--no-split MAKEINFOFLAGS=--no-split
-     # -C gdb/doc {gdb,annotate}{.info,/index.html,.pdf} MAKEHTMLFLAGS=--no-split MAKEINFOFLAGS=--no-split
+     -C gdb/doc {gdb,annotate}{.info,/index.html,.pdf} MAKEHTMLFLAGS=--no-split MAKEINFOFLAGS=--no-split
 
 grep '#define HAVE_ZLIB_H 1' gdb/config.h
 
@@ -1284,8 +1296,8 @@ for pyo in "" "-O";do
 done
 %endif # 0%{?_enable_debug_packages:1} && 0%{!?_without_python:1}
 
-%if 0%{?rhel:1} && 0%{?rhel} <= 6
 %if 0%{!?_without_python:1}
+%if 0%{?rhel:1} && 0%{?rhel} <= 6
 # Temporarily now:
 for LIB in lib lib64;do
   LIBPATH="$RPM_BUILD_ROOT%{_datadir}/gdb/auto-load%{_root_prefix}/$LIB"
@@ -1307,8 +1319,15 @@ for i in `find $RPM_BUILD_ROOT%{_datadir}/gdb/python -name "*.py"` \
   # Files come from gdb-archer.patch and can be also further patched.
   touch -r $RPM_BUILD_DIR/%{gdb_src}/gdb/ChangeLog $i
 done
+%else # 0%{!?rhel:1} || 0%{?rhel} > 6
+# BZ 999645: /usr/share/gdb/auto-load/ needs filesystem symlinks
+mkdir -p $RPM_BUILD_ROOT%{_datadir}/gdb/auto-load
+for i in $(echo bin lib $(basename %{_libdir}) sbin|tr ' ' '\n'|sort -u);do
+  ln -s $(echo %{_root_prefix}|sed 's#^/*##')/$i \
+        $RPM_BUILD_ROOT%{_datadir}/gdb/auto-load/$i
+done
+%endif # 0%{!?rhel:1} || 0%{?rhel} > 6
 %endif # 0%{!?_without_python:1}
-%endif # 0%{?rhel:1} && 0%{?rhel} <= 6
 
 # gdb-add-index cannot be run even for SCL package on RHEL<=6.
 %if 0%{?rhel:1} && 0%{?rhel} <= 6
@@ -1410,13 +1429,24 @@ rm -rf $RPM_BUILD_ROOT
 %endif # %{have_inproctrace}
 %endif
 
+%pre
+for i in $(echo bin lib $(basename %{_libdir}) sbin|tr ' ' '\n'|sort -u);do
+  src="%{_datadir}/gdb/auto-load/$i"
+  dst="%{_datadir}/gdb/auto-load/%{_root_prefix}/$i"
+  if test -d $src -a ! -L $src;then
+    if ! rmdir 2>/dev/null $src;then
+      mv -n $src/* $dst/
+      rmdir $src
+    fi
+  fi
+done
+
 # It would break RHEL-5 by leaving excessive files for the doc subpackage.
 %endif # !noarch
 %if 0%{!?el5:1} || "%{_target_cpu}" == "noarch"
 
 %files doc
-%doc %{gdb_build}/gdb/doc/{gdb,annotate}.html
-#doc %{gdb_build}/gdb/doc/{gdb,annotate}.{html,pdf}
+%doc %{gdb_build}/gdb/doc/{gdb,annotate}.{html,pdf}
 %defattr(-,root,root)
 %{_infodir}/annotate.info*
 %{_infodir}/gdb.info*
@@ -1446,6 +1476,33 @@ fi
 %endif # 0%{!?el5:1} || "%{_target_cpu}" == "noarch"
 
 %changelog
+* Mon Sep 30 2013 Sergio Durigan Junior <sergiodj@redhat.com> - 7.5.1-43.fc18
+- Fix the case when GDB leaks memory because value_struct_elt does not call
+  check_typedef.  (Doug Evans, BZ 15695, filed as RH BZ 1013453).
+
+* Mon Sep  9 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-42.fc18
+- Fix the version string to be GNU standards compliant (BZ 1004949).
+
+* Fri Aug 30 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-41.fc18
+- New %%pre to fix failed upgrade of the previous commit (BZ 999645).
+- Fix false warnings of new %%pre during future upgrades (BZ 999645).
+
+* Wed Aug 28 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-40.fc18
+- Fix /usr/share/gdb/auto-load/ need of filesystem symlinks (BZ 999645).
+  It needs: yum remove gdb-heap; yum reinstall gdb; yum install gdb-heap
+
+* Sat Apr 27 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-39.fc18
+- Fix false "Unknown error 512" on x32 (H.J. Lu, BZ 956883).
+
+* Fri Apr 12 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-38.fc18
+- Fix crash on 'gdb ""' (BZ 951251)..
+
+* Thu Mar 21 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-37.fc18
+- Fix crash regression from the dlopen of libpthread.so fix (BZ 911712).
+
+* Fri Feb  8 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-36.fc18
+- Re-enable PDF in gdb-doc after texinfo RH BZ 876710 has been fixed.
+
 * Sun Jan 13 2013 Jan Kratochvil <jan.kratochvil@redhat.com> - 7.5.1-35.fc18
 - [testsuite] Fix gdb-orphanripper.c lockup on F-17 (/dev/pts/* glibc chown).