From: heltai Date: Tue, 10 Dec 2013 09:41:00 +0000 (+0000) Subject: Merged from trunk. X-Git-Url: https://gitweb.dealii.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=1fc50d6d9aaa50e56be43417b57930650de0460c;p=dealii-svn.git Merged from trunk. git-svn-id: https://svn.dealii.org/branches/branch_manifold_id@31967 0785d39b-7218-0410-832d-ea1e28bc413d --- diff --git a/deal.II/VERSION b/deal.II/VERSION index 091a617214..6cf8469d8c 100644 --- a/deal.II/VERSION +++ b/deal.II/VERSION @@ -1 +1 @@ -8.1.pre +8.2.pre diff --git a/deal.II/cmake/macros/macro_add_test.cmake b/deal.II/cmake/macros/macro_add_test.cmake index c4a2b90a1e..96c677b5c1 100644 --- a/deal.II/cmake/macros/macro_add_test.cmake +++ b/deal.II/cmake/macros/macro_add_test.cmake @@ -131,7 +131,7 @@ MACRO(DEAL_II_ADD_TEST _category _test_name _comparison_file) SET(_diff_target ${_test_name}.mpirun${_n_cpu}.${_build_lowercase}.diff) # diff target name SET(_test_full ${_category}/${_test_name}.mpirun=${_n_cpu}.${_build_lowercase}) # full test name SET(_test_directory ${CMAKE_CURRENT_BINARY_DIR}/${_target}/mpirun=${_n_cpu}) # directory to run the test in - SET(_run_command mpirun -np ${_n_cpu} ${CMAKE_CURRENT_BINARY_DIR}/${_target}/${_target}) # the command to issue + SET(_run_command ${MPIEXEC} ${MPIEXEC_NUMPROC_FLAG} ${_n_cpu} ${MPIEXEC_PREFLAGS} ${CMAKE_CURRENT_BINARY_DIR}/${_target}/${_target} ${MPIEXEC_POSTFLAGS}) # the command to issue ENDIF() diff --git a/deal.II/cmake/modules/FindTRILINOS.cmake b/deal.II/cmake/modules/FindTRILINOS.cmake index f36c514ed0..a19b46bd44 100644 --- a/deal.II/cmake/modules/FindTRILINOS.cmake +++ b/deal.II/cmake/modules/FindTRILINOS.cmake @@ -36,20 +36,18 @@ INCLUDE(FindPackageHandleStandardArgs) SET_IF_EMPTY(TRILINOS_DIR "$ENV{TRILINOS_DIR}") # -# Do not include TrilinosConfig.cmake directly, it is just too big o_O +# Include the trilinos package configuration: # -# Just search for the file: -# -FIND_FILE(TRILINOS_CONFIG - NAMES TrilinosConfig.cmake trilinos-config.cmake +FIND_PACKAGE(TRILINOS_CONFIG + CONFIG QUIET + NAMES Trilinos TRILINOS HINTS + ${TRILINOS_DIR}/lib/cmake/Trilinos ${TRILINOS_DIR} PATH_SUFFIXES lib64/cmake/Trilinos lib/cmake/Trilinos lib${LIB_SUFFIX}/cmake/Trilinos - include/trilinos - include/Trilinos NO_SYSTEM_ENVIRONMENT_PATH ) @@ -59,38 +57,6 @@ ENDIF() SET(TRILINOS_CONFIG_SAVED "${TRILINOS_CONFIG}" CACHE INTERNAL "" FORCE) -IF(NOT TRILINOS_CONFIG MATCHES "-NOTFOUND") - - SET(_filtered_trilinos_config - "${CMAKE_BINARY_DIR}${CMAKE_FILES_DIRECTORY}/TrilinosConfig.cmake" - ) - - IF(_new_trilinos_config) - GET_FILENAME_COMPONENT(_trilinos_path "${TRILINOS_CONFIG}" PATH) - FILE(WRITE ${_filtered_trilinos_config} "SET(_cmake_current_list_dir ${_trilinos_path})\n") - - # - # Only pick up every line that starts with "^SET("... - # - FILE(STRINGS "${TRILINOS_CONFIG}" _trilinos_config_filtered REGEX "^SET") - - FOREACH(_line ${_trilinos_config_filtered}) - STRING(REPLACE "CMAKE_CURRENT_LIST_DIR" "_cmake_current_list_dir" - _line "${_line}" - ) - FILE(APPEND ${_filtered_trilinos_config} "${_line}\n") - ENDFOREACH() - ENDIF() - - # - # ... and include only that: - # - INCLUDE(${_filtered_trilinos_config}) - - SET(TRILINOS_CONFIG_FOUND TRUE) -ENDIF() - - # # Look for the one include file that we'll query for further information: # @@ -106,18 +72,16 @@ FIND_FILE(EPETRA_CONFIG_H Epetra_config.h NO_CMAKE_SYSTEM_PATH NO_CMAKE_FIND_ROOT_PATH ) -IF(EPETRA_CONFIG_H MATCHES "-NOTFOUND") - SET(TRILINOS_CONFIG_FOUND FALSE) -ELSE() - SET(TRILINOS_INCLUDE_DIRS ${Trilinos_INCLUDE_DIRS}) - # - # *Boy* Sanitize the include paths given by TrilinosConfig.cmake... - # - STRING(REGEX REPLACE - "(lib64|lib)\\/cmake\\/Trilinos\\/\\.\\.\\/\\.\\.\\/\\.\\.\\/" "" - TRILINOS_INCLUDE_DIRS "${TRILINOS_INCLUDE_DIRS}" - ) -ENDIF() + + +# +# *Boy* Sanitize the include paths given by TrilinosConfig.cmake... +# +SET(TRILINOS_INCLUDE_DIRS ${Trilinos_INCLUDE_DIRS}) +STRING(REGEX REPLACE + "(lib64|lib)\\/cmake\\/Trilinos\\/\\.\\.\\/\\.\\.\\/\\.\\.\\/" "" + TRILINOS_INCLUDE_DIRS "${TRILINOS_INCLUDE_DIRS}" + ) # @@ -143,11 +107,9 @@ FOREACH(_library ${Trilinos_LIBRARIES}) MARK_AS_ADVANCED(TRILINOS_LIBRARY_${_library}) - IF(TRILINOS_LIBRARY_${_library} MATCHES "-NOTFOUND") - SET(TRILINOS_CONFIG_FOUND FALSE) - ELSE() - LIST(APPEND TRILINOS_LIBRARIES ${TRILINOS_LIBRARY_${_library}}) - ENDIF() + LIST(APPEND TRILINOS_LIBRARIES ${TRILINOS_LIBRARY_${_library}}) + LIST(APPEND TRILINOS_LIBRARY_VARIABLES TRILINOS_LIBRARY_${_library}) + ENDFOREACH() # @@ -161,11 +123,12 @@ LIST(APPEND TRILINOS_LIBRARIES FIND_PACKAGE_HANDLE_STANDARD_ARGS(TRILINOS DEFAULT_MSG TRILINOS_LIBRARIES # cosmetic: Gives nice output TRILINOS_CONFIG_FOUND + EPETRA_CONFIG_H + ${TRILINOS_LIBRARY_VARIABLES} ) MARK_AS_ADVANCED(TRILINOS_CONFIG EPETRA_CONFIG_H) - IF(TRILINOS_FOUND) # # Extract version numbers: @@ -249,6 +212,10 @@ IF(TRILINOS_FOUND) ELSE() + SET(TRILINOS_LIBRARIES) + SET(TRILINOS_INCLUDE_DIRS) + SET(TRILINOS_CONFIG_SAVED "" CACHE INTERNAL "" FORCE) + SET(TRILINOS_DIR "" CACHE PATH "An optional hint to a Trilinos installation" ) diff --git a/deal.II/cmake/scripts/run_testsuite.cmake b/deal.II/cmake/scripts/run_testsuite.cmake index f4605d6760..ffa6253e5b 100644 --- a/deal.II/cmake/scripts/run_testsuite.cmake +++ b/deal.II/cmake/scripts/run_testsuite.cmake @@ -379,7 +379,9 @@ IF(${_result} EQUAL 0) STRING(REGEX REPLACE "^${_svn_WC_ROOT}/" "" _branch ${_svn_WC_URL}) STRING(REGEX REPLACE "^branches/" "" _branch ${_branch}) + STRING(REGEX REPLACE "^releases/" "" _branch ${_branch}) STRING(REGEX REPLACE "/deal.II$" "" _branch ${_branch}) + STRING(REGEX REPLACE "/" "" _branch ${_branch}) SET(CTEST_BUILD_NAME "${CTEST_BUILD_NAME}-${_branch}") ENDIF() diff --git a/deal.II/cmake/setup_testsubproject.cmake b/deal.II/cmake/setup_testsubproject.cmake index b1bd6bf6ee..6ed0f5dfb4 100644 --- a/deal.II/cmake/setup_testsubproject.cmake +++ b/deal.II/cmake/setup_testsubproject.cmake @@ -33,6 +33,16 @@ # TEST_TIME_LIMIT # +# +# Silence warnings: +# +FOREACH(_var + MPIEXEC MPIEXEC_NUMPROC_FLAG MPIEXEC_POSTFLAGS MPIEXEC_PREFLAGS + ) + SET(${_var} ${${_var}}) +ENDFOREACH() + + # # Load all macros: # diff --git a/deal.II/cmake/setup_testsuite.cmake b/deal.II/cmake/setup_testsuite.cmake index 98580373d0..564a99acbf 100644 --- a/deal.II/cmake/setup_testsuite.cmake +++ b/deal.II/cmake/setup_testsuite.cmake @@ -91,7 +91,15 @@ SET(_options) LIST(APPEND _options -DDEAL_II_SOURCE_DIR=${CMAKE_SOURCE_DIR}) LIST(APPEND _options -DDEAL_II_BINARY_DIR=${CMAKE_BINARY_DIR}) FOREACH(_var - DIFF_DIR NUMDIFF_DIR TEST_DIFF TEST_PICKUP_REGEX TEST_TIME_LIMIT + DIFF_DIR + NUMDIFF_DIR + TEST_DIFF + TEST_PICKUP_REGEX + TEST_TIME_LIMIT + MPIEXEC + MPIEXEC_NUMPROC_FLAG + MPIEXEC_PREFLAGS + MPIEXEC_POSTFLAGS ) # always undefine: LIST(APPEND _options "-U${_var}") diff --git a/deal.II/doc/CMakeLists.txt b/deal.II/doc/CMakeLists.txt index c4ec069aff..b7efaf534c 100644 --- a/deal.II/doc/CMakeLists.txt +++ b/deal.II/doc/CMakeLists.txt @@ -21,31 +21,7 @@ IF(DEAL_II_COMPONENT_DOCUMENTATION) MESSAGE(STATUS "Setup documentation") - # - # Required packages for the document generation: - # - FIND_PACKAGE(Perl) - FIND_PACKAGE(Doxygen) - - # - # Do we have all necessary dependencies? - # - IF(NOT PERL_FOUND) - MESSAGE(FATAL_ERROR - "Could not find a perl installation which is required for building the documentation" - ) - ENDIF() - IF(NOT DOXYGEN_FOUND OR NOT DOXYGEN_DOT_FOUND) - MESSAGE(FATAL_ERROR - "Could not find doxygen and dot which is required for building the documentation" - ) - ENDIF() - - # - # Generate the source documentation via doxygen: - # ADD_SUBDIRECTORY(doxygen) - ADD_DEPENDENCIES(documentation doxygen) # # Install the static elements of the html documentation: diff --git a/deal.II/doc/authors.html b/deal.II/doc/authors.html index 7e7e9252b3..7e180f8a3d 100644 --- a/deal.II/doc/authors.html +++ b/deal.II/doc/authors.html @@ -115,7 +115,7 @@ interface, FE_Nedelec for arbitrary polynomial degrees.
  • Daniel Arndt: - FE_Q_DG0 element, ConstraintMatrix fixes, linear algebra. + FE_Q_DG0 element, ConstraintMatrix fixes, linear algebra, periodicity, bug fixes.
  • Juan Carlos Araujo Cabarcas: Work on ARPACK. @@ -134,10 +134,10 @@ Step-25 tutorial program.
  • Chih-Che Chueh: - Step-43 tutorial program. + Step-43 tutorial program. Bug fixes.
  • Denis Davydov: - FEValuesExtractors::Tensor and associated functionality, various other changes. + FEValuesExtractors::Tensor and associated functionality, bug fixes, various other changes.
  • Kevin Dugan: Fixes in VectorTools::interpolate. @@ -175,6 +175,9 @@ Lots of work on multigrid for adaptive meshes;   multigrid in the MeshWorker framework;   step-16. Various fixes. +
  • Armin Ghajar Jazi: + Small fixes. +
  • Xing Jin: step-24 tutorial program. @@ -194,7 +197,7 @@ framework. step-37 and step-48 tutorial programs.
  • Oleh Krehel: - Many documentation fixes. + Documentation fixes, small fixes.
  • Tobias Leicht: Lots of work on internal data structures: anisotropic refinement @@ -220,8 +223,11 @@
  • Andrew McBride: FEValues extractors for symmetric tensors. step-44. +
  • Craig Michoski: + Periodic boundaries +
  • Scott Miller: - Enhancements to FE_Nothing. + Enhancements to FE_Nothing, step-51, many smaller fixes.
  • Helmut Müller: Multiprocessor detection on Mac OS X. @@ -241,7 +247,7 @@ throughout the library.
  • Jean-Paul Pelteret: - step-44, linear algebra. + step-44, linear algebra, bug fixes.
  • Jonathan Pitt: Enhancements to FE_Nothing. Changes in a variety of places. @@ -252,6 +258,9 @@
  • Florian Prill: Gauss Lobatto quadrature, random bug fixes. +
  • Andreas Putz: + GridIn::read_vtk(). +
  • Daniel Castanon Quiroz: Torus domain and boundary. @@ -264,6 +273,9 @@   refinement functions;   MinRes linear solver. +
  • Mayank Sabharwal: + GridIn::read_vtk(). +
  • Abner Salgado-Gonzalez: step-35 tutorial program. Other fixes. diff --git a/deal.II/doc/developers/cmake-internals.html b/deal.II/doc/developers/cmake-internals.html index b2308511d2..e500d2c9ac 100644 --- a/deal.II/doc/developers/cmake-internals.html +++ b/deal.II/doc/developers/cmake-internals.html @@ -144,7 +144,7 @@ SET_TARGET_PROPERTIES(${DEAL_II_BASE_NAME}${DEAL_II_${build}_SUFFIX} This sets up all cached variables prior to the call to PROJECT(deal.II). For details see the comment at the top. Furthermore, some bookkeeping for compiler and linker flags - takes place, see the section + takes place, see the section about compile flags.
  • setup_deal_ii.cmake: This file is included immediately after the call to @@ -400,7 +400,7 @@ SET(FEATURE_DIR "" CACHE PATH "An optional hint to a FEATURE directory") module is used to determine whether an external dependency can be resolved or not. Depending on the current state of DEAL_II_WITH_<FEATURE> (see - here) the variables + here) the variables 
     FEATURE_LIBRARIES
 FEATURE_INCLUDE_DIRS
diff --git a/deal.II/doc/developers/navbar.html b/deal.II/doc/developers/navbar.html
index 4d92590038..676c05884a 100644
--- a/deal.II/doc/developers/navbar.html
+++ b/deal.II/doc/developers/navbar.html
@@ -43,9 +43,9 @@
   Online Resources
   
    
     Recent changes
    
+    CDash
    
     Regression tests
    
-    Build tests
    
-    Performance tests
    
+    Performance tests
    
   
    
 
   
    
diff --git a/deal.II/doc/developers/testsuite.html b/deal.II/doc/developers/testsuite.html
index 3c55dae252..7247194f88 100644
--- a/deal.II/doc/developers/testsuite.html
+++ b/deal.II/doc/developers/testsuite.html
@@ -738,7 +738,7 @@ CTEST_SOURCE_DIRECTORY
   - The source directory of deal.II (usually ending in "[...]/deal.II"
     (equivalent to https://svn.dealii.org/trunk/deal.II)
     Note: This is _not_ the test directory ending in "[...]/tests"
-  - If unspecified, "../deal.II" and "../../$ relative to the location
+  - If unspecified, "../deal.II" and "../../" relative to the location
     of this script is used. If this is not a source directory, an error
     thrown.
 
@@ -798,7 +798,7 @@ MAKEOPTS
 
       
    Running the build test suite is simple and we encourage deal.II
       users with configurations not found on the test suite page to
+      href="http://cdash.kyomu.43-1.org/index.php?project=deal.II">CDash page to
       participate. Assuming you checked out deal.II into the directory
       deal.II, running it is as simple as:
 
    diff --git a/deal.II/doc/developers/toc.html b/deal.II/doc/developers/toc.html
index ca6902a604..6ca72c7ed5 100644
--- a/deal.II/doc/developers/toc.html
+++ b/deal.II/doc/developers/toc.html
@@ -72,23 +72,26 @@ have the following resources available:
 
    Resources
    
 
 
      
-
    •  
      Results of regression tests: Every night,
-     the regression tests are run on some of our machines. The log of the
-     results can be seen here.
-     
      
-
-
    •  
      
-      Results of build tests: We also build the library each night on
-      a variety of computers, using different compilers. The results of
-      these builds is listed as well.
-     
      
-
-
    •  
        
+  
    • 
+    
      
+      CDash: The CDash instance is the central place where
+      all test results for regression and build tests are gathered.
+    
      
+  
    • 
+    
      
+      Regression tests:
+      An overview of results from a blessed regression tester that tests
+      every single svn revision is available at this place
+    
      
+
+  
    • 
+    
      
+      
       Results of benchmarks: We run a certain number of tests
       and track the performance over time.
-     
      
+    
      
 
    
 
 
    
diff --git a/deal.II/doc/doxygen/CMakeLists.txt b/deal.II/doc/doxygen/CMakeLists.txt
index 17fd2d2b1f..cdeeeebb78 100644
--- a/deal.II/doc/doxygen/CMakeLists.txt
+++ b/deal.II/doc/doxygen/CMakeLists.txt
@@ -14,6 +14,27 @@
 ##
 ## ---------------------------------------------------------------------
 
+#
+# Generate the source documentation via doxygen:
+#
+
+FIND_PACKAGE(Perl)
+FIND_PACKAGE(Doxygen)
+
+#
+# Do we have all necessary dependencies?
+#
+IF(NOT PERL_FOUND)
+  MESSAGE(FATAL_ERROR
+    "Could not find a perl installation which is required for building the documentation"
+    )
+ENDIF()
+IF(NOT DOXYGEN_FOUND)
+  MESSAGE(FATAL_ERROR
+    "Could not find doxygen which is required for building the documentation"
+    )
+ENDIF()
+
 #
 # A glob. I'm sorry
 #
@@ -25,7 +46,6 @@ file(GLOB DEAL_II_STEPS
 # Prepare the example steps for doxygen:
 #
 ADD_SUBDIRECTORY(tutorial) # uses ${DEAL_II_STEPS}
-ADD_SUBDIRECTORY(images)
 
 #
 # Prepare auxiliary files for doxygen:
@@ -35,24 +55,6 @@ CONFIGURE_FILE(
   ${CMAKE_CURRENT_BINARY_DIR}/options.dox
   )
 
-CONFIGURE_FILE(
-  ${CMAKE_CURRENT_SOURCE_DIR}/deal.css
-  ${CMAKE_CURRENT_BINARY_DIR}/deal.II/deal.css
-  COPYONLY
-  )
-
-CONFIGURE_FILE(
-  ${CMAKE_CURRENT_SOURCE_DIR}/stylesheet.css
-  ${CMAKE_CURRENT_BINARY_DIR}/stylesheet.css
-  COPYONLY
-  )
-
-CONFIGURE_FILE(
-  ${CMAKE_CURRENT_SOURCE_DIR}/images/logo200.png
-  ${CMAKE_CURRENT_BINARY_DIR}/logo200.png
-  COPYONLY
-  )
-
 CONFIGURE_FILE(
   ${CMAKE_CURRENT_SOURCE_DIR}/scripts/mod_header.pl.in
   ${CMAKE_CURRENT_BINARY_DIR}/scripts/mod_header.pl
@@ -129,7 +131,6 @@ ADD_CUSTOM_COMMAND(
   WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
   DEPENDS
     tutorial
-    images
     ${CMAKE_CURRENT_BINARY_DIR}/options.dox
     ${CMAKE_CURRENT_BINARY_DIR}/header.html
     ${CMAKE_CURRENT_BINARY_DIR}/footer.html
@@ -140,10 +141,10 @@ ADD_CUSTOM_COMMAND(
 ADD_CUSTOM_TARGET(doxygen ALL
   DEPENDS ${CMAKE_BINARY_DIR}/doxygen.log
   )
+ADD_DEPENDENCIES(documentation doxygen)
 
 INSTALL(FILES
   ${CMAKE_CURRENT_BINARY_DIR}/deal.tag
-  ${CMAKE_CURRENT_SOURCE_DIR}/stylesheet.css
   DESTINATION ${DEAL_II_DOCHTML_RELDIR}/doxygen
   COMPONENT documentation
   )
diff --git a/deal.II/doc/doxygen/deal.css b/deal.II/doc/doxygen/deal.css
deleted file mode 100644
index 716718c669..0000000000
--- a/deal.II/doc/doxygen/deal.css
+++ /dev/null
@@ -1,18 +0,0 @@
-//
-// Copyright (C) 1998 - 2013 by the deal.II authors
-//
-
-table.tutorial {
-    color: black;
-    border: 1px solid #aaa;
-    background-color: #f9f9f9;
-    padding: 5px;
-    font-size: 95%;
-}
-h3 {
-        font-size: 120%;
-        text-decoration:underline;
-}
-h4 {
-	font-size: 100%;
-}
diff --git a/deal.II/doc/doxygen/headers/multithreading.h b/deal.II/doc/doxygen/headers/multithreading.h
index 218b16803c..7ec8ea3d69 100644
--- a/deal.II/doc/doxygen/headers/multithreading.h
+++ b/deal.II/doc/doxygen/headers/multithreading.h
@@ -643,7 +643,12 @@
  * relatively independent jobs: for example, assembling local
  * contributions to the global linear system on each cell of a mesh;
  * evaluating an error estimator on each cell; or postprocessing on
- * each cell computed data for output fall into this class.
+ * each cell computed data for output fall into this class. These
+ * cases can be treated using a software design pattern we call
+ * "%WorkStream". In the following, we will walk through the rationale
+ * for this pattern and its implementation; more details as well as
+ * examples for the speedup that can be achieved with it are given in
+ * the @ref workstream_paper .
  *
  * Code like this could then be written like this:
  * @code
@@ -1011,7 +1016,7 @@
  *
  * As a final point: What if, for some reason, my assembler and copier
  * function do not match the above signature with three and one argument,
- * respectively? That's not a problem either. The WorkStream class offers two
+ * respectively? That's not a problem either. The WorkStream namespace offers two
  * versions of the WorkStream::run() function: one that takes an object and
  * the addresses of two member functions, and one that simply takes two
  * function objects that can be called with three and one argument,
@@ -1027,9 +1032,14 @@
      // ...is the same as:
      WorkStream::run (dof_handler.begin_active(),
                       dof_handler.end(),
-                      std_cxx1x::bind(&MyClass::assemble_on_one_cell, *this,
-                                  std_cxx1x::_1, std_cxx1x::_2, std_cxx1x::_3),
-                      std_cxx1x::bind(&MyClass::copy_local_to_global, *this, std_cxx1x::_1),
+                      std_cxx1x::bind(&MyClass::assemble_on_one_cell,
+                                      *this,
+                                      std_cxx1x::_1,
+                                      std_cxx1x::_2,
+                                      std_cxx1x::_3),
+                      std_cxx1x::bind(&MyClass::copy_local_to_global,
+                                      *this,
+                                      std_cxx1x::_1),
                       per_task_data);
  * @endcode
  * Note how std_cxx1x::bind produces a function object that takes three
@@ -1063,21 +1073,23 @@
      WorkStream::run (dof_handler.begin_active(),
                       dof_handler.end(),
                       std_cxx1x::bind(&MyClass::assemble_on_one_cell,
-                                  *this,
-                                  current_solution,
-                                  std_cxx1x::_1,
-                                  std_cxx1x::_2,
-                                  std_cxx1x::_3,
-                                  previous_time+time_step),
+                                      *this,
+                                      current_solution,
+                                      std_cxx1x::_1,
+                                      std_cxx1x::_2,
+                                      std_cxx1x::_3,
+                                      previous_time+time_step),
                       std_cxx1x::bind(&MyClass::copy_local_to_global,
-                                  *this, std_cxx1x::_1),
+                                      *this,
+                                      std_cxx1x::_1),
                       per_task_data);
  * @endcode
  * Here, we bind the object, the linearization point argument, and the
  * current time argument to the function before we hand it off to
  * WorkStream::run(). WorkStream::run() will then simply call the
  * function with the cell and scratch and per task objects which will be filled
- * in at the positions indicated by std_cxx1x::_1, std_cxx1x::_2 and std_cxx1x::_3.
+ * in at the positions indicated by std_cxx1x::_1, std_cxx1x::_2
+ * and std_cxx1x::_3.
  *
  * There are refinements to the WorkStream::run function shown above.
  * For example, one may realize that the basic idea above can only scale
diff --git a/deal.II/doc/doxygen/images/CMakeLists.txt b/deal.II/doc/doxygen/images/CMakeLists.txt
deleted file mode 100644
index 54a6589526..0000000000
--- a/deal.II/doc/doxygen/images/CMakeLists.txt
+++ /dev/null
@@ -1,32 +0,0 @@
-## ---------------------------------------------------------------------
-## $Id$
-##
-## Copyright (C) 2013 by the deal.II authors
-##
-## This file is part of the deal.II library.
-##
-## The deal.II library is free software; you can use it, redistribute
-## it, and/or modify it under the terms of the GNU Lesser General
-## Public License as published by the Free Software Foundation; either
-## version 2.1 of the License, or (at your option) any later version.
-## The full text of the license can be found in the file LICENSE at
-## the top level of the deal.II distribution.
-##
-## ---------------------------------------------------------------------
-ADD_CUSTOM_COMMAND(
-  OUTPUT
-    ${CMAKE_CURRENT_BINARY_DIR}/structure.png
-    ${CMAKE_CURRENT_BINARY_DIR}/structure.cmapx
-  COMMAND ${DOXYGEN_DOT_EXECUTABLE}
-    -Tpng -o ${CMAKE_CURRENT_BINARY_DIR}/structure.png
-    -Tcmapx -o ${CMAKE_CURRENT_BINARY_DIR}/structure.cmapx
-    ${CMAKE_CURRENT_SOURCE_DIR}/structure.dot
-  DEPENDS
-    ${CMAKE_CURRENT_SOURCE_DIR}/structure.dot
-  )
-
-ADD_CUSTOM_TARGET(images
-  DEPENDS
-    ${CMAKE_CURRENT_BINARY_DIR}/structure.png
-    ${CMAKE_CURRENT_BINARY_DIR}/structure.cmapx
-  )
diff --git a/deal.II/doc/doxygen/images/structure.dot b/deal.II/doc/doxygen/images/structure.dot
deleted file mode 100644
index 936be67955..0000000000
--- a/deal.II/doc/doxygen/images/structure.dot
+++ /dev/null
@@ -1,37 +0,0 @@
-digraph G
-{
-  graph[rankdir="TB",aspect=1.5,bgcolor="transparent"];
-  
-  edge [fontname="FreeSans",fontsize=10,labelfontname="FreeSans",labelfontsize=10];
-  node [fontname="FreeSans",fontsize=10,
-        shape=record,height=0.2,width=0.4,
-        color="black", fillcolor="white", style="filled"];
-
-  FE [label="FiniteElement",URL="../deal.II/group_fe.html"];
-  Tria [label="Triangulation",URL="../deal.II/group_grid.html"];
-  DoFHandler [label="DoFHandler",URL="../deal.II/group_dofs.html"];
-  Quadrature [label="Quadrature",URL="../deal.II/group_Quadrature.html"];
-  Mapping [label="Mapping",URL="../deal.II/group_mapping.html"];
-  FEValues [label="FEValues",URL="../deal.II/group_feaccess.html"];
-#  MeshWorker [label="MeshWorker",URL="../deal.II/group_MeshWorker.html"];
-#  Forms [label="(Bi-)Linear Forms",URL="../deal.II/group_Integrators.html"];
-  Linear [label="Discrete System",URL="../deal.II/group_LAC.html"];
-  LinearSolver [label="Solver",URL="../deal.II/group_Solvers.html"];
-  Output [label="Output",URL="../deal.II/group_output.html"];
-
-  Tria -> DoFHandler [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-#  Tria -> FEValues [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  FE -> DoFHandler [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  FE -> FEValues [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  Mapping -> FEValues [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  Quadrature -> FEValues [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  FEValues -> Linear [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-#  FEValues -> MeshWorker [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-#  FEValues -> Forms [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-#  Forms -> MeshWorker [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  DoFHandler -> Linear [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-#  DoFHandler -> MeshWorker [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-#  MeshWorker -> Linear [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  Linear -> LinearSolver [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-  LinearSolver -> Output [color="black",fontsize=10,style="solid",fontname="FreeSans"];
-}
diff --git a/deal.II/doc/doxygen/options.dox.in b/deal.II/doc/doxygen/options.dox.in
index 9f691ebb7d..d4a33e0e23 100644
--- a/deal.II/doc/doxygen/options.dox.in
+++ b/deal.II/doc/doxygen/options.dox.in
@@ -3,7 +3,7 @@
 
 PROJECT_NAME           = "The deal.II Library"
 PROJECT_NUMBER         = "Reference documentation for deal.II version @DEAL_II_PACKAGE_VERSION@"
-PROJECT_LOGO           = "logo200.png"
+PROJECT_LOGO           = @CMAKE_CURRENT_SOURCE_DIR@/images/logo200.png
 OUTPUT_DIRECTORY       = .
 
 # The doxygen documentation says this about the following flag:
@@ -103,7 +103,7 @@ MATHJAX_RELPATH        = http://cdn.mathjax.org/mathjax/latest
 MATHJAX_EXTENSIONS     = TeX/AMSmath TeX/AMSsymbols
 HTML_COLORSTYLE_HUE    = 220
 HTML_COLORSTYLE_SAT    = 200
-HTML_EXTRA_STYLESHEET  = stylesheet.css
+HTML_EXTRA_STYLESHEET  = @CMAKE_CURRENT_SOURCE_DIR@/stylesheet.css
 #---------------------------------------------------------------------------
 # configuration options related to the LaTeX output
 #---------------------------------------------------------------------------
diff --git a/deal.II/doc/doxygen/scripts/mod_header.pl.in b/deal.II/doc/doxygen/scripts/mod_header.pl.in
index 6007349ce1..2a2b91cf6e 100644
--- a/deal.II/doc/doxygen/scripts/mod_header.pl.in
+++ b/deal.II/doc/doxygen/scripts/mod_header.pl.in
@@ -6,7 +6,6 @@ $year += 1900;
 
 if (m'')
 {
-    print '', "\n";
     print '', "\n";
     print '', "\n";
     print '', "\n";
diff --git a/deal.II/doc/doxygen/tutorial/tutorial.h.in b/deal.II/doc/doxygen/tutorial/tutorial.h.in
index d7495be2a8..aafad4e83e 100644
--- a/deal.II/doc/doxygen/tutorial/tutorial.h.in
+++ b/deal.II/doc/doxygen/tutorial/tutorial.h.in
@@ -54,9 +54,8 @@
  * cmake ., build it via make and run it using
  * make run. The latter command also compiles the program if
  * that has not already been done. The CMakefiles.txt files in the
- * different directories are based on the small program Makefile template
- * described in this
- * section.
+ * different directories are based on the 
+ * small program Makefile template.
  *
  * 
  * 
    Connections between tutorial programs
    
@@ -859,7 +858,7 @@
  *   
  *
  * 
- * 
    Vector problems
    
+ * 
    %Vector problems
    
  * 
  *
  *   
diff --git a/deal.II/doc/news/8.0.0-vs-8.1.0.h b/deal.II/doc/news/8.0.0-vs-8.1.0.h
new file mode 100644
index 0000000000..4afef43f2f
--- /dev/null
+++ b/deal.II/doc/news/8.0.0-vs-8.1.0.h
@@ -0,0 +1,730 @@
+// ---------------------------------------------------------------------
+// $Id$
+//
+// Copyright (C) 2013 by the deal.II authors
+//
+// This file is part of the deal.II library.
+//
+// The deal.II library is free software; you can use it, redistribute
+// it, and/or modify it under the terms of the GNU Lesser General
+// Public License as published by the Free Software Foundation; either
+// version 2.1 of the License, or (at your option) any later version.
+// The full text of the license can be found in the file LICENSE at
+// the top level of the deal.II distribution.
+//
+// ---------------------------------------------------------------------
+
+/**
+ * @page changes_between_8_0_and_8_1 Changes between Version 8.0 and 8.1
+
+
    
+This is the list of changes made between the release of
+deal.II version 8.0.0 and that of 8.1.0.
+All entries are signed with the names of the author.
+
    
+
+
+
+
+
+
+
    Incompatibilities
    
+
+
    
+Following are a few modifications to the library that unfortunately
+are incompatible with previous versions of the library, but which we
+deem necessary for the future maintainability of the
+library. Unfortunately, some of these changes will require
+modifications to application programs. We apologize for the
+inconvenience this causes.
+
    
+
+
      
+  
    1. Changed: During the implementation of the 64-bit features for deal.II
+  8.0, many linear algebra classes obtained a local typedef
+  size_type indicating the integer type that is used to index
+  into them. In some instances, this was accidentally set to
+  types::global_dof_index (which may be a 64-bit data type)
+  even in cases where this is clearly not going to work, for example for
+  FullMatrix::size_type, since we will not be able to store full matrix
+  objects of sizes for which a 32-bit index type is not sufficient. In
+  these cases, the typedef was reverted to just unsigned int.
+  
      
+  (Wolfgang Bangerth, 2013/12/04)
+  
      2. 
+
+  
    2.  Removed: With the switch of the testsuite to CMake, the old report_features
+  and build test facilities are removed.
+  
      
+  (Matthias Maier, 2013/12/03)
+  
      3. 
+
+  
    3. 
+  Changed: The kinds of template arguments for the VectorTools::interpolate
+  function taking a Mapping as first argument has changed. This was done to
+  work around a bug in the Intel ICC compiler which led to linker errors. Since
+  the actual function argument list remains unchanged, the only way you will
+  notice this change is if you explicitly specify template arguments.
+  The only place one would typically do that is if you take the address of
+  a template function. Since this is not a common operation, the impact of this
+  change is probably limited.
+  
      
+  (Wolfgang Bangerth, 2013/11/27)
+  
      4. 
+
+  
    4. 
+  Changed: The ghost handling of the parallel::distributed::Vector class has
+  been reworked: The vector now carries a global state that stores whether
+  ghost elements have been updated or not. If a vector has ghost elements, it
+  does not allow calls to compress() any more. Instead, a compress operation
+  can now only be done when the ghost entries have been cleared before by
+  calling zero_out_ghosts() or operator=0. The state can be queried by the new
+  method has_ghost_elements(). This change avoids spurious entries to be
+  inserted with compress(), but requires some change in user codes. The
+  behavior of a ghosted vector is now very similar to ghosted PETSc and
+  Trilinos vectors. The only difference is that the same vector can
+  also be used as a non-ghosted vector which is designed for use in assembly
+  routines.
+  
      
+  (Martin Kronbichler, 2013/10/18)
+  
      5. 
+
+  
    5. 
+  Removed: GridTools::collect_periodic_face_pairs. This function is superseded
+  by GridTools::collect_periodic_faces which exports an
+  std::vector> instead.
+  
      
+  (Matthias Maier, 2013/09/30)
+  
      6. 
+
+  
    6. 
+  Removed: The member function face_to_equivalent_cell_index() in
+  FiniteElementData has been removed. It had been deprecated a while
+  back already. Please use FiniteElement::face_to_cell_index() instead.
+  
      
+  (Wolfgang Bangerth, 2013/08/09)
+  
      7. 
+
+  
    7. 
+  Changed: The typedefs DataOut::cell_iterator and
+  DataOut::active_cell_iterator were previously defined as
+  DoFHandler::(active)_cell_iterator, while they are now
+  Triangulation::(active)_cell_iterator. This is necessary to support DataOut
+  on multiple DoFHandler objects. This affects possible overloading of
+  DataOut::next_cell(cell_iterator). Use the typedef
+  DataOut::(active)_cell_iterator as argument type instead.
+  
      
+  (Martin Kronbichler, 2013/07/24)
+  
      8. 
+
    
+
+
+
+
+
+
    General
    
+
+
+
      
+  
    1.  Improved: The build system now supports usage of the library
+  out of the build directory without prior installation. This is done by
+  exporting an additional project configuration just for the build directory.
+  Furthermore, a bunch of convenience targets get now defined that just build
+  individual components (such as just the documentation or the libraries), and
+  if CMAKE_INSTALL_PREFIX is set, also install that specific component.
+  
      
+  (Matthias Maier, Luca Heltai, 2013/12/03)
+  
      2. 
+
+  
    2.  Fixed: Missing instantiations of SparseDirectMUMPS have been added.
+  
      
+  (Timo Heister, 2013/11/25)
+  
      3. 
+
+  
    3.  New: introduced "make test" that runs a minimal set of tests. We
+  encourage every user to run this, especially if they run in to problems.
+  The tests are automatically picked depending on the configuration and
+  will be shipped with every release.
+  
      
+  (Timo Heister, Matthias Maier, 2013/11/08)
+  
      4. 
+
+  
    4.  Changed: It is now possible to restore a parallel Triangulation
+  (and solutions) with a different number of processors it was saved with
+  using Triangulation::save() and Triangulation::load().
+  
      
+  (Timo Heister, 2013/11/02)
+  
      5. 
+
+  
    5.  Added support for Windows: It is now possible again to use gcc on Windows
+  in order to compile the library. We support gcc-4.8.1 on Cygwin64 and MinGW-w64.
+  
      
+  (Matthias Maier, 2013/11/01)
+  
      6. 
+
+  
    6.  Changed: step-9, step-13 and step-14 have been converted to use the
+  more modern WorkStream concept for assembling linear systems and computing
+  error indicators in parallel.
+  
      
+  (Bruno Turcksin, Wolfgang Bangerth, 2013/10/26)
+  
      7. 
+
+  
    7.  New: The testsuite is now ported to 
+  CMake and uses CTest as test driver.
+  
      
+  (Wolfgang Bangerth, Timo Heister, Matthias Maier, Bruno Turcksin, 2013/10/20)
+  
      8. 
+
+  
    8. 
+  Changed: multithreadinfo::n_default_threads is now deprecated. Use the
+  new n_threads() function instead, which works correctly with TBB.
+  
      
+  (Timo Heister, 2013/10/02)
+  
      9. 
+
+  
    9. 
+  Changed: if configured with TBB but the number of threads is set to 1,
+  do not bother to use TBB in workstream.
+  
      
+  (Timo Heister, 2013/10/02)
+  
      10. 
+
+  
    10. 
+  New: step-51 demonstrates the use of hybridized discontinuous Galerkin
+  methods in deal.II, using the face elements FE_FaceQ. The programs solves a
+  scalar convection-diffusion equation.
+  
      
+  (Martin Kronbichler and Scott Miller, 2013/10/01)
+  
      11. 
+
+  
    11. 
+  New: There is now an element FE_FaceP that can be combined with FE_DGP in
+  hybridized DG methods.
+  
      
+  (Martin Kronbichler, 2013/09/17)
+  
      12. 
+
+  
    12. 
+  Fixed: The DataOutBase::XDMFEntry class now has a proper serialization
+  function to allow for checkpointing.
+  
      
+  (Eric Heien, 2013/09/27)
+  
      13. 
+
+  
    13. 
+  New: DataOutBase::DataOutFilter provides a way to remove duplicate vertices
+  and values from a solution vector when generating output. Currently it only
+  supports HDF5/XDMF output.
+  
      
+  (Eric Heien, 2013/09/27)
+  
      14. 
+
+  
    14. 
+  Removed: DataOutBase::HDF5MemStream was removed and the functionality replaced
+  by DataOutBase::DataOutFilter. The user only manipulates these through
+  DataOutBase::write_hdf5_parallel so this change should be transparent.
+  
      
+  (Eric Heien, 2013/09/27)
+  
      15. 
+
+  
    15. 
+  New: Like the usual DoFHandler class, the hp::DoFHandler class now also
+  has a cache that makes operations such as cell-@>get_dof_indices(...)
+  faster. This should accelerate many parts of the library that deal with
+  hp finite elements.
+  
      
+  (Wolfgang Bangerth, 2013/09/10)
+  
      16. 
+
+  
    16. 
+  New: parallel::distributed::Triangulation now supports periodic boundaries.
+  DoFTools::make_periodicity_constraints and similar functions are now working
+  on parallel::distributed::Triangulation objects.
+  
      
+  (Tobin Isaac, Craig Michoski, Daniel Arndt, 2013/09/06)
+  
      17. 
+
+  
    17. 
+  New: It is now possible to compile and link deal.II against LLVM's libcxx. For
+  this, a few issues with C++ standard violations are resolved.
+  
      
+  (Matthias Maier, 2013/08/09)
+  
      18. 
+
    
+
+
+
+
+
+
    Specific improvements
    
+
+
      
+  
    1.  Fixed: Since the introduction of ThreadLocalStorage in version 8.0, the
+  way in which FEValues objects visit cells in a parallel assembly loop is no
+  longer deterministic. Therefore, the detection of CellSimilarity that can
+  speed up computations of certain geometric quantities (shape gradients) on
+  cells that are translations is disabled when the number of threads is
+  greater than one. This produces somewhat slower code (usually not more than
+  a few percent) but ensures exact reproducibility of results.
+  
      
+  (Martin Kronbichler, Wolfgang Bangerth, 2013/12/09)
+  
      2. 
+
+  
    2.  Fixed: Several functions in namespace GridTools were not instantiated
+  for parallel::distributed::Triangulation objects. This is now fixed.
+  
      
+  (Denis Davydov, Wolfgang Bangerth, 2013/12/01)
+  
      3. 
+
+  
    3.  Improved: The methods ConstraintMatrix::distribute_local_to_global
+  now use scratch data that is private to each thread instead of allocating
+  it for every cell anew. This gives better performance, in particular in
+  parallel, of these operations, while maintaining thread-safety (when
+  accessing non-overlapping rows, no race condition can exist).
+  
      
+  (Martin Kronbichler, 2013/12/03)
+  
      4. 
+
+  
    4.  Improved: When attempting operations such as FEValues::get_function_values()
+  or FEValues::shape_value(), the FEValues object needs to know that what these
+  functions return has been computed previously. What is computed is specified
+  by the update flags that are passed to the constructor of all FEValues, FEFaceValues
+  and FESubfaceValues objects. If a user attempts an operation for which the
+  corresponding flag was not specified, an exception is generated. This exception
+  did say previously what the cause was, but it was not overly explicit.
+  The exception now generates a message that says exactly what is going wrong.
+  
      
+  (Wolfgang Bangerth, 2013/12/01)
+  
      5. 
+
+  
    5.  Fixed: GridGenerator::truncated_cone() failed if half_length < 0.5*radius in 3d.
+  
      
+  (Timo Heister, 2013/11/25)
+  
      6. 
+
+  
    6.  Fixed: make_hanging_node_constraints failed with an exception in a
+  parallel::distributed computation if the element is
+  RaviartThomas (and probably others).
+  
      
+  (Timo Heister, 2013/11/23)
+  
      7. 
+
+  
    7.  Improved: CMake: Added a configuration check for incompatible ninja
+  + icc setup, fixed several setup and performance issues with the
+  testsuite.
+  
      
+  (Matthias Maier, 2013/11/20)
+  
      8. 
+
+  
    8.  Changed: when a dealii::Exception is thrown, defer the symbol lookup of the
+  stack trace to when it is needed. This improves performance if what() is never
+  called.
+  
      
+  (Timo Heister, 2013/11/17)
+  
      9. 
+
+  
    9.  Fixed: GridGenerator::parallelogram was not instantiated properly
+  when using intel compilers.
+  
      
+  (Timo Heister, 2013/11/17)
+  
      10. 
+
+  
    10. 
+  Fixed: MappingQ1::transform_real_to_unit_cell() could fail in some
+  cases with very elongated and twisted cells. This should now be fixed
+  with an algorithm that uses a better method of computing the Newton
+  convergence.
+  
      
+  (Wolfgang Bangerth, 2013/11/17)
+  
      11. 
+
+  
    11. 
+  Fixed: VectorTools::compute_no_normal_flux_constraints had a bug that
+  only appeared in rare cases at vertices of the domain if one adjacent
+  cell had two boundary indicators selected for no normal flux and another
+  had only one. This is now fixed.
+  
      
+  (Wolfgang Bangerth, 2013/11/17)
+  
      12. 
+
+  
    12.  Fixed: dealii::FETools::interpolation_difference was
+  not working for TrilinosWrappers::MPI::Vectors with ghost
+  entries. The TrilinosWrappers::VectorBase class has now a
+  get_mpi_communicator method similar to the PETSc vector
+  classes.
+  
      
+  (Martin Steigemann, Martin Kronbichler, 2013/11/17)
+  
      13. 
+
+  
    13.  Fixed: Bundled fparser is now compiled with FP_USE_THREAD_SAFE_EVAL in
+  case of enabled threading support so that it is thread safe.
+  
      
+  (Matthias Maier, reported by Francesco Cattoglio 2013/11/16)
+  
      14. 
+
+  
    14.  Fixed: The CellData class now has a default constructor that
+  sets the material and boundary indicators to zero. This fixes certain
+  internal errors with the Intel ICC compiler.
+  
      
+  (Wolfgang Bangerth, 2013/11/13)
+  
      15. 
+
+  
    15.  Cleanup: Removed obsolete files and files with unknown licensing
+  status from the source tree. Along the way, parameter_gui now uses
+  default icons from the desktop environment instead of bundled ones.
+  
      
+  (Matthias Maier, 2013/11/11)
+  
      16. 
+
+  
    16.  New: There is now a framework for coloring graphs, with functions
+  in namespace GraphColoring.
+  
      
+  (Bruno Turcksin, Martin Kronbichler, 2013/11/06)
+  
      17. 
+
+  
    17. 
+  Fixed: the DerivativeApproximation class was not working correctly when
+  used with parallel vectors.
+  (Timo Heister, 2013/10/28)
+  
      18. 
+
+  
    18. 
+  ~Subscriptor and ~GrowingVectorMemory no longer throw an exception (the
+  former if disable_abort_on_exception was called) to be compatible with the
+  C++11 standard which otherwise requires the program to immediately call
+  std::terminate. This was done with a new macro "AssertNothrow".
+  
      
+  (Wolfgang Bangerth, Matthias Maier, Bruno Turcksin 2013/10/22)
+  
      19. 
+
+  
    19. 
+  dealii::SolverControl::NoConvergence now inherits dealii::ExceptionBase and
+  is thrown via AssertThrow(false, ... ).
+  
      
+  (Matthias Maier, 2013/10/20)
+  
      20. 
+
+  
    20. 
+  New: parallel::distributed::BlockVector has now methods update_ghost_values,
+  compress, zero_out_ghosts, and has_ghost_elements that do the respective
+  operation on each block of parallel::distributed::Vector.
+  
      
+  (Martin Kronbichler, 2013/10/18)
+  
      21. 
+
+  
    21. 
+  Fixed: When deriving from DataOut to filter the cells where output is generated, there were two different bugs that result in segmentation faults or wrong cells written (example, step-18).
+  
      
+  (Timo Heister, 2013/10/16)
+  
      22. 
+
+  
    22. 
+  New: GridIn::read_vtk() reads 2d and 3d meshes in VTK format.
+  
      
+  (Mayank Sabharwal, Andreas Putz, 2013/10/07)
+  
      23. 
+
+  
    23. 
+  Fixed: ConstraintMatrix would not compress() the IndexSet in the constructor
+  leading to crashes that only happen in release mode. This is now fixed.
+  
      
+  (Timo Heister, 2013/09/27)
+  
      24. 
+
+  
    24. 
+  Fixed: PetscWrappers::MatrixBase::row_length() no longer worked after recent changes
+  to PETSc (around PETSc release 3.4). This is now fixed.
+  
      
+  (Wolfgang Bangerth, 2013/09/24)
+  
      25. 
+
+  
    25. 
+  New: Added write_visit_record that allows writing .visit files with multiple blocks
+  and multiple time steps.
+  
      
+  (Fahad Alrashed, 2013/09/21)
+  
      26. 
+
+  
    26. 
+  Changed: GridTools::have_same_coarse_mesh was only instantiated for
+  MGDoFHandler arguments in debug mode. This is now fixed.
+  
      
+  (Timo Heister, 2013/09/20)
+  
      27. 
+
+  
    27. 
+  Changed: GridTools::find_active_cell_around_point now throws the exception
+  GridTools::ExcPointNotFound
+  if the point is outside the Triangulation. This exception can be caught.
+  
      
+  (Timo Heister, 2013/09/18)
+  
      28. 
+
+  
    28. 
+  Changed: we now call MPI_Init_thread instead of MPI_Init.
+  
      
+  (Timo Heister, 2013/09/17)
+  
      29. 
+
+  
    29. 
+  Enhancement: It is now possible to use the build directory directly without
+  the need to install first. For this, a second copy of all necessary project
+  configuration files (deal.IIConfig.cmake, ..., Make.global_options) are
+  generated and deployed in the build directory. (This is fully compatible with
+  the old possibility to install into the build dir.)
+  
      
+  (Matthias Maier, 2013/09/15)
+  
      30. 
+
+  
    30. 
+  Fixed: DoFTools::extract_locally_*_dofs now instantiated for hp::DofHandler.
+  
      
+  (Jean-Paul Pelteret, 2013/09/11)
+  
      31. 
+
+  
    31. 
+  Changed: distributed::parallel:BlockVector::operator= now allows importing
+  of ghost values like all other vector types. Also added some new constructors
+  for BlockVector and Vector using IndexSets to mirror the other linear algebra
+  classes.
+  
      
+  (Timo Heister, 2013/09/04)
+  
      32. 
+
+  
    32. 
+  Fixed: VectorTools::compute_no_normal_flux_constraints had a bug that
+  only manifested on complex meshes. This is now fixed.
+  
      
+  (Chih-Che Chueh, Wolfgang Bangerth, 2013/09/04)
+  
      33. 
+
+  
    33. 
+  New: All vector classes now have functions extract_subvector_to()
+  that allow extracting not just a single value but a whole set.
+  
      
+  (Fahad Alrashed, 2013/09/02)
+  
      34. 
+
+  
    34. 
+  Fixed: common/Make.global_options now exports enable-threads
+  correctly, furthermore, lib-suffix, shared-lib-suffix
+  and static-lib-suffix are now exported as well for better legacy
+  support.
+  
      
+  (Matthias Maier, 2013/08/30)
+  
      35. 
+
+  
    35. 
+  New: The ParameterHandler class can now deal with including one parameter
+  file from another.
+  
      
+  (Wolfgang Bangerth, 2013/08/25)
+  
      36. 
+
+  
    36. 
+  New: The method VectorTools::compute_normal_flux_constraints can be used to
+  force a vector finite element function to be normal to the boundary.
+  
      
+  (Martin Kronbichler, 2013/08/23)
+  
      37. 
+
+  
    37. 
+  Improved: MappingQ now uses the points of the Gauss-Lobatto quadrature
+  formula as support points instead of equispaced ones. This allows its use
+  for high polynomial orders and also gives better interpolation of circular
+  boundaries. Beware that mappings of order three and higher will behave
+  slightly differently now (usually better).
+  
      
+  (Martin Kronbichler, 2013/08/23)
+  
      38. 
+
+  
    38. 
+  Improved: Several .cc files in the deal.II directory have been split in
+  order to better utilize multiple processors when compiling in parallel and
+  reduce memory requirements of the compilation stage.
+  
      
+  (Martin Kronbichler, 2013/08/22)
+  
      39. 
+
+  
    39. 
+  Fixed: The ParameterHandler::declare_entry() did not check that the
+  default value of a parameter indeed satisfies the pattern given for this
+  parameter (despite a statement in the documentation that this checking
+  would happen). This is now fixed.
+  
      
+  (Wolfgang Bangerth, 2013/08/21)
+  
      40. 
+
+  
    40. 
+  New: Patterns::List and Patterns::Map now accept a string
+  different than the default comma that denotes the separator
+  between entries of the list or map.
+  
      
+  (Wolfgang Bangerth, 2013/08/21)
+  
      41. 
+
+  
    41. 
+  Fixed: Some operations in the MappingQ class are now done in higher
+  precision arithmetic to mitigate the ill-conditioning that appears
+  when using mappings of high order (say, order 6 or 8 or 10).
+  
      
+  (Juan Carlos Araujo Cabarcas, 2013/08/20)
+  
      42. 
+
+  
    42. 
+  Fixed: The SLEPcWrappers classes could not be compiled for 64-bit
+  indices. This is now fixed.
+  
      
+  (Denis Davydov, Wolfgang Bangerth, 2013/08/20)
+  
      43. 
+
+  
    43. 
+  Fixed: SolutionTransfer used to crash whenever one transfered in the hp
+  context between cells that use FE_Nothing and FE_Q. This is now fixed.
+  
      
+  (Krzyszof Bzowski, Wolfgang Bangerth, 2013/08/18)
+  
      44. 
+
+  
    44. 
+  Fixed: Under some circumstances (see http://code.google.com/p/dealii/issues/detail?id=82)
+  the DoFTools::make_periodicity_constraints() function could create cycles in
+  the ConstraintMatrix object. This is now fixed.
+  
      
+  (David Emerson, Wolfgang Bangerth, 2013/08/16)
+  
      45. 
+
+  
    45. 
+  New: There is now a function ConstraintMatrix::are_identity_constrained().
+  
      
+  (Wolfgang Bangerth, 2013/08/16)
+  
      46. 
+
+  
    46. 
+  New: TableHandler::write_text() now also supports output in
+  org-mode (http://orgmode.org/) format via a new entry in the
+  TableHandler::TextOutputFormat enumeration.
+  
      
+  (Oleh Krehel, 2013/08/15)
+  
      47. 
+
+  
    47. 
+  New: There are now global functions scalar_product
+  that compute the scalar product (double contraction) between
+  tensors of rank 2.
+  
      
+  (Scott Miller, 2013/08/14)
+  
      48. 
+
+  
    48. 
+  Fixed: Creating objects of type MappingQ was previously only possible
+  for low order polynomials. For orders higher than around 6, one ran
+  into assertions that tested for internal consistency. These assertions
+  have now been appropriately relaxes for the growth of round-off errors
+  with growing polynomial degrees.
+  
      
+  (Juan Carlos Araujo Cabarcas, Wolfgang Bangerth, 2013/08/14)
+  
      49. 
+
+  
    49. 
+  New: MappingQEulerian is now also instantiated for vector elements
+  of type TrilinosWrappers::Vector as well as the MPI and block
+  variants.
+  
      
+  (Armin Ghajar Jazi, 2013/08/14)
+  
      50. 
+
+  
    50. 
+  Fixed: The FiniteElement::face_to_cell_index() function had a bug
+  that made it work incorrectly for elements that have more than one
+  degree of freedom per line (in 2d) or per quad (in 3d). This is now
+  fixed for the most common cases, namely the FE_Q elements as well
+  as elements composed of FESystem elements. For all other cases, an
+  exception is generated reporting that this case is not implemented.
+  If you run into this, let us know.
+  
      
+  (Wolfgang Bangerth, 2013/08/10)
+  
      51. 
+
+  
    51. 
+  New: DataOutBase::VtkFlags now has a flag
+  DataOutBase::VtkFlags::print_date_and_time that can be used to suppress output
+  of date and time in output files. This is useful in test suites where a newer
+  run at a different time produces differences against previously stored files,
+  even though the actual data is exactly the same.
+  
      
+  (Oleh Krehel, 2013/08/06)
+  
      52. 
+
+  
    52. 
+  Fixed: The various block matrix classes are all derived from BlockMatrixBase
+  which had race conditions when the set() or add() functions were called from
+  different threads. This is now fixed.
+  
      
+  (Wolfgang Bangerth, 2013/08/05)
+  
      53. 
+
+  
    53. 
+  Fixed: various fixes with assignment and reinit of PETScWrappers::MPI::Vector.
+  
      
+  (Timo Heister, 2013/08/05)
+  
      54. 
+
+  
    54. Fixed: An assertion wrongly triggered in
+  DoFTools::make_hanging_node_constraints when used with a particular
+  combination of FESystem elements containing FE_Nothing. This is now fixed.
+  
      
+  (Denis Davydov, Wolfgang Bangerth, 2013/08/01)
+  
      55. 
+
+  
    55. 
+  New: Add has_ghost_elements() for PETScWrappers::MPI::BlockVector and
+  TrilinosWrappers::MPI::BlockVector.
+  
      
+  (Timo Heister, 2013/08/01)
+  
      56. 
+
+  
    56. 
+  SparsityTools::distribute_sparsity_pattern did not work correctly for
+  block systems, this has been fixed (function has a different signature).
+  
      
+  (Timo Heister, 2013/07/31)
+  
      57. 
+
+  
    57. Fixed: When typing make run in the step-32 directory,
+  the program was executed with mpirun -np 2 ./step-32. This
+  assumes that a program mpirun exists, but also does that
+  deal.II was in fact compiled with MPI support on. Neither was intended.
+  This is now fixed.
+  
      
+  (Wolfgang Bangerth, 2013/07/24)
+  
      58. 
+
+  
    58. New: The DataOut, DataOutFaces, and DataOutRotation classes now allow
+  the output of data vectors using different DoFHandler objects (based on the
+  same triangulation), by new functions add_data_vector. This is used in the
+  step-31 tutorial program which avoids creating a joint DoFHandler just for
+  output.
+  
      
+  (Martin Kronbichler, 2013/07/24)
+  
      59. 
+
+  
    59. Changed: GridGenerator used to be a class with only static members
+  but is now a namespace, like all other similar constructs in deal.II.
+  
      
+  (Wolfgang Bangerth, 2013/07/24)
+  
      60. 
+
+  
    60. Changed: In GridGenerator, several functions had erroneously been changed
+  to take an argument of type size_type rather than unsigned
+  int. GridGenerator::size_type was a typedef to
+  types::global_dof_index, which for most users was unsigned int
+  anyway, but could also be set to be a 64-bit integer type. In any case, the
+  change has been reverted and these functions take just a regular
+  unsigned int again.
+  
      
+  (Wolfgang Bangerth, 2013/07/24)
+  
      61. 
+
    
+
+
+*/
diff --git a/deal.II/doc/news/changes.h b/deal.II/doc/news/changes.h
index 93c0ecd3a3..1cdd7325ca 100644
--- a/deal.II/doc/news/changes.h
+++ b/deal.II/doc/news/changes.h
@@ -15,11 +15,11 @@
 // ---------------------------------------------------------------------
 
 /**
-// * @page changes_after_8_0 Changes after Version 8.0
+// * @page changes_after_8_1 Changes after Version 8.1
 
 
    
 This is the list of changes made after the release of
-deal.II version 8.0.0.
+deal.II version 8.1.0.
 All entries are signed with the names of the authors.
 
    
 
@@ -40,69 +40,6 @@ inconvenience this causes.
 
      
-  
    1.  Removed: With the switch of the testsuite to CMake, the old report_features
-  and build test facilities are removed.
-  
      
-  (Matthias Maier, 2013/12/03)
-  
      2. 
-
-  
    2. 
-  Changed: The kinds of template arguments for the VectorTools::interpolate
-  function taking a Mapping as first argument has changed. This was done to
-  work around a bug in the Intel ICC compiler which led to linker errors. Since
-  the actual function argument list remains unchanged, the only way you will
-  notice this change is if you explicitly specify template arguments.
-  The only place one would typically do that is if you take the address of
-  a template function. Since this is not a common operation, the impact of this
-  change is probably limited.
-  
      
-  (Wolfgang Bangerth, 2013/11/27)
-  
      3. 
-
-  
    3. 
-  Changed: The ghost handling of the parallel::distributed::Vector class has
-  been reworked: The vector now carries a global state that stores whether
-  ghost elements have been updated or not. If a vector has ghost elements, it
-  does not allow calls to compress() any more. Instead, a compress operation
-  can now only be done when the ghost entries have been cleared before by
-  calling zero_out_ghosts() or operator=0. The state can be queried by the new
-  method has_ghost_elements(). This change avoids spurious entries to be
-  inserted with compress(), but requires some change in user codes. The
-  behavior of a ghosted vector is now very similar to ghosted PETSc and
-  Trilinos vectors. The only difference is that the same vector can
-  also be used as a non-ghosted vector which is designed for use in assembly
-  routines.
-  
      
-  (Martin Kronbichler, 2013/10/18)
-  
      4. 
-
-  
    4. 
-  Removed: GridTools::collect_periodic_face_pairs. This function is superseded
-  by GridTools::collect_periodic_faces which exports an
-  std::vector> instead.
-  
      
-  (Matthias Maier, 2013/09/30)
-  
      5. 
-
-  
    5. 
-  Removed: The member function face_to_equivalent_cell_index() in
-  FiniteElementData has been removed. It had been deprecated a while
-  back already. Please use FiniteElement::face_to_cell_index() instead.
-  
      
-  (Wolfgang Bangerth, 2013/08/09)
-  
      6. 
-
-  
    6. 
-  Changed: The typedefs DataOut::cell_iterator and
-  DataOut::active_cell_iterator were previously defined as
-  DoFHandler::(active)_cell_iterator, while they are now
-  Triangulation::(active)_cell_iterator. This is necessary to support DataOut
-  on multiple DoFHandler objects. This affects possible overloading of
-  DataOut::next_cell(cell_iterator). Use the typedef
-  DataOut::(active)_cell_iterator as argument type instead.
-  
      
-  (Martin Kronbichler, 2013/07/24)
-  
      7. 
 
    
 
 
@@ -113,130 +50,6 @@ inconvenience this causes.
 
 
 
      
-  
    1.  Improved: The build system now supports usage of the library
-  out of the build directory without prior installation. This is done by
-  exporting an additional project configuration just for the build directory.
-  Furthermore, a bunch of convenience targets get now defined that just build
-  individual components (such as just the documentation or the libraries), and
-  if CMAKE_INSTALL_PREFIX is set, also install that specific component.
-  
      
-  (Matthias Maier, Luca Heltai, 2013/12/03)
-  
      2. 
-
-  
    2.  Fixed: Missing instantiations of SparseDirectMUMPS have been added.
-  
      
-  (Timo Heister, 2013/11/25)
-  
      3. 
-
-  
    3.  New: introduced "make test" that runs a minimal set of tests. We
-  encourage every user to run this, especially if they run in to problems.
-  The tests are automatically picked depending on the configuration and
-  will be shipped with every release.
-  
      
-  (Timo Heister, Matthias Maier, 2013/11/08)
-  
      4. 
-
-  
    4.  Changed: It is now possible to restore a parallel Triangulation
-  (and solutions) with a different number of processors it was saved with
-  using Triangulation::save() and Triangulation::load().
-  
      
-  (Timo Heister, 2013/11/02)
-  
      5. 
-
-  
    5.  Added support for Windows: It is now possible again to use gcc on Windows
-  in order to compile the library. We support gcc-4.8.1 on Cygwin64 and MinGW-w64.
-  
      
-  (Matthias Maier, 2013/11/01)
-  
      6. 
-
-  
    6.  Changed: step-9, step-13 and step-14 have been converted to use the
-  more modern WorkStream concept for assembling linear systems and computing
-  error indicators in parallel.
-  
      
-  (Bruno Turcksin, Wolfgang Bangerth, 2013/10/26)
-  
      7. 
-
-  
    7.  New: The testsuite is now ported to 
-  CMake and uses CTest as test driver.
-  
      
-  (Wolfgang Bangerth, Timo Heister, Matthias Maier, Bruno Turcksin, 2013/10/20)
-  
      8. 
-
-  
    8. 
-  Changed: multithreadinfo::n_default_threads is now deprecated. Use the
-  new n_threads() function instead, which works correctly with TBB.
-  
      
-  (Timo Heister, 2013/10/02)
-  
      9. 
-
-  
    9. 
-  Changed: if configured with TBB but the number of threads is set to 1,
-  do not bother to use TBB in workstream.
-  
      
-  (Timo Heister, 2013/10/02)
-  
      10. 
-
-  
    10. 
-  New: step-51 demonstrates the use of hybridized discontinuous Galerkin
-  methods in deal.II, using the face elements FE_FaceQ. The programs solves a
-  scalar convection-diffusion equation.
-  
      
-  (Martin Kronbichler and Scott Miller, 2013/10/01)
-  
      11. 
-
-  
    11. 
-  New: There is now an element FE_FaceP that can be combined with FE_DGP in
-  hybridized DG methods.
-  
      
-  (Martin Kronbichler, 2013/09/17)
-  
      12. 
-
-  
    12. 
-  Fixed: The DataOutBase::XDMFEntry class now has a proper serialization
-  function to allow for checkpointing.
-  
      
-  (Eric Heien, 2013/09/27)
-  
      13. 
-
-  
    13. 
-  New: DataOutBase::DataOutFilter provides a way to remove duplicate vertices
-  and values from a solution vector when generating output. Currently it only
-  supports HDF5/XDMF output.
-  
      
-  (Eric Heien, 2013/09/27)
-  
      14. 
-
-  
    14. 
-  Removed: DataOutBase::HDF5MemStream was removed and the functionality replaced
-  by DataOutBase::DataOutFilter. The user only manipulates these through
-  DataOutBase::write_hdf5_parallel so this change should be transparent.
-  
      
-  (Eric Heien, 2013/09/27)
-  
      15. 
-
-  
    15. 
-  New: Like the usual DoFHandler class, the hp::DoFHandler class now also
-  has a cache that makes operations such as cell-@>get_dof_indices(...)
-  faster. This should accelerate many parts of the library that deal with
-  hp finite elements.
-  
      
-  (Wolfgang Bangerth, 2013/09/10)
-  
      16. 
-
-  
    16. 
-  New: parallel::distributed::Triangulation now supports periodic boundaries.
-  DoFTools::make_periodicity_constraints and similar functions are now working
-  on parallel::distributed::Triangulation objects.
-  
      
-  (Tobin Isaac, Craig Michoski, Daniel Arndt, 2013/09/06)
-  
      17. 
-
-  
    17. 
-  New: It is now possible to compile and link deal.II against LLVM's libcxx. For
-  this, a few issues with C++ standard violations are resolved.
-  
      
-  (Matthias Maier, 2013/08/09)
-  
      18. 
 
    
 
 
@@ -246,460 +59,6 @@ inconvenience this causes.
 
    Specific improvements
      
-  
    1.  Fixed: Several functions in namespace GridTools were not instantiated
-  for parallel::distributed::Triangulation objects. This is now fixed.
-  
      
-  (Denis Davydov, Wolfgang Bangerth, 2013/12/01)
-  
      2. 
-
-  
    2.  Improved: The methods ConstraintMatrix::distribute_local_to_global
-  now use scratch data that is private to each thread instead of allocating
-  it for every cell anew. This gives better performance, in particular in
-  parallel, of these operations, while maintaining thread-safety (when
-  accessing non-overlapping rows, no race condition can exist).
-  
      
-  (Martin Kronbichler, 2013/12/03)
-  
      3. 
-
-  
    3.  Improved: When attempting operations such as FEValues::get_function_values()
-  or FEValues::shape_value(), the FEValues object needs to know that what these
-  functions return has been computed previously. What is computed is specified
-  by the update flags that are passed to the constructor of all FEValues, FEFaceValues
-  and FESubfaceValues objects. If a user attempts an operation for which the
-  corresponding flag was not specified, an exception is generated. This exception
-  did say previously what the cause was, but it was not overly explicit.
-  The exception now generates a message that says exactly what is going wrong.
-  
      
-  (Wolfgang Bangerth, 2013/12/01)
-  
      4. 
-
-  
    4.  Fixed: GridGenerator::truncated_cone() failed if half_length < 0.5*radius in 3d.
-  
      
-  (Timo Heister, 2013/11/25)
-  
      5. 
-
-  
    5.  Fixed: make_hanging_node_constraints failed with an exception in a
-  parallel::distributed computation if the element is
-  RaviartThomas (and probably others).
-  
      
-  (Timo Heister, 2013/11/23)
-  
      6. 
-
-  
    6.  Improved: CMake: Added a configuration check for incompatible ninja
-  + icc setup, fixed several setup and performance issues with the
-  testsuite.
-  
      
-  (Matthias Maier, 2013/11/20)
-  
      7. 
-
-  
    7.  Changed: when a dealii::Exception is thrown, defer the symbol lookup of the
-  stack trace to when it is needed. This improves performance if what() is never
-  called.
-  
      
-  (Timo Heister, 2013/11/17)
-  
      8. 
-
-  
    8.  Fixed: GridGenerator::parallelogram was not instantiated properly
-  when using intel compilers.
-  
      
-  (Timo Heister, 2013/11/17)
-  
      9. 
-
-  
    9. 
-  Fixed: MappingQ1::transform_real_to_unit_cell() could fail in some
-  cases with very elongated and twisted cells. This should now be fixed
-  with an algorithm that uses a better method of computing the Newton
-  convergence.
-  
      
-  (Wolfgang Bangerth, 2013/11/17)
-  
      10. 
-
-  
    10. 
-  Fixed: VectorTools::compute_no_normal_flux_constraints had a bug that
-  only appeared in rare cases at vertices of the domain if one adjacent
-  cell had two boundary indicators selected for no normal flux and another
-  had only one. This is now fixed.
-  
      
-  (Wolfgang Bangerth, 2013/11/17)
-  
      11. 
-
-  
    11.  Fixed: dealii::FETools::interpolation_difference was
-  not working for TrilinosWrappers::MPI::Vectors with ghost
-  entries. The TrilinosWrappers::VectorBase class has now a
-  get_mpi_communicator method similar to the PETSc vector
-  classes.
-  
      
-  (Martin Steigemann, Martin Kronbichler, 2013/11/17)
-  
      12. 
-
-  
    12.  Fixed: Bundled fparser is now compiled with FP_USE_THREAD_SAFE_EVAL in
-  case of enabled threading support so that it is thread safe.
-  
      
-  (Matthias Maier, reported by Francesco Cattoglio 2013/11/16)
-  
      13. 
-
-  
    13.  Fixed: The CellData class now has a default constructor that
-  sets the material and boundary indicators to zero. This fixes certain
-  internal errors with the Intel ICC compiler.
-  
      
-  (Wolfgang Bangerth, 2013/11/13)
-  
      14. 
-
-  
    14.  Cleanup: Removed obsolete files and files with unknown licensing
-  status from the source tree. Along the way, parameter_gui now uses
-  default icons from the desktop environment instead of bundled ones.
-  
      
-  (Matthias Maier, 2013/11/11)
-  
      15. 
-
-  
    15.  New: There is now a framework for coloring graphs, with functions
-  in namespace GraphColoring.
-  
      
-  (Bruno Turcksin, Martin Kronbichler, 2013/11/06)
-  
      16. 
-
-  
    16. 
-  Fixed: the DerivativeApproximation class was not working correctly when
-  used with parallel vectors.
-  (Timo Heister, 2013/10/28)
-  
      17. 
-
-  
    17. 
-  ~Subscriptor and ~GrowingVectorMemory no longer throw an exception (the
-  former if disable_abort_on_exception was called) to be compatible with the
-  C++11 standard which otherwise requires the program to immediately call
-  std::terminate. This was done with a new macro "AssertNothrow".
-  
      
-  (Wolfgang Bangerth, Matthias Maier, Bruno Turcksin 2013/10/22)
-  
      18. 
-
-  
    18. 
-  dealii::SolverControl::NoConvergence now inherits dealii::ExceptionBase and
-  is thrown via AssertThrow(false, ... ).
-  
      
-  (Matthias Maier, 2013/10/20)
-  
      19. 
-
-  
    19. 
-  New: parallel::distributed::BlockVector has now methods update_ghost_values,
-  compress, zero_out_ghosts, and has_ghost_elements that do the respective
-  operation on each block of parallel::distributed::Vector.
-  
      
-  (Martin Kronbichler, 2013/10/18)
-  
      20. 
-
-  
    20. 
-  Fixed: When deriving from DataOut to filter the cells where output is generated, there were two different bugs that result in segmentation faults or wrong cells written (example, step-18).
-  
      
-  (Timo Heister, 2013/10/16)
-  
      21. 
-
-  
    21. 
-  New: GridIn::read_vtk() reads 2d and 3d meshes in VTK format.
-  
      
-  (Mayank Sabharwal, Andreas Putz, 2013/10/07)
-  
      22. 
-
-  
    22. 
-  Fixed: ConstraintMatrix would not compress() the IndexSet in the constructor
-  leading to crashes that only happen in release mode. This is now fixed.
-  
      
-  (Timo Heister, 2013/09/27)
-  
      23. 
-
-  
    23. 
-  Fixed: PetscWrappers::MatrixBase::row_length() no longer worked after recent changes
-  to PETSc (around PETSc release 3.4). This is now fixed.
-  
      
-  (Wolfgang Bangerth, 2013/09/24)
-  
      24. 
-
-  
    24. 
-  New: Added write_visit_record that allows writing .visit files with multiple blocks
-  and multiple time steps.
-  
      
-  (Fahad Alrashed, 2013/09/21)
-  
      25. 
-
-  
    25. 
-  Changed: GridTools::have_same_coarse_mesh was only instantiated for
-  MGDoFHandler arguments in debug mode. This is now fixed.
-  
      
-  (Timo Heister, 2013/09/20)
-  
      26. 
-
-  
    26. 
-  Changed: GridTools::find_active_cell_around_point now throws the exception
-  GridTools::ExcPointNotFound
-  if the point is outside the Triangulation. This exception can be caught.
-  
      
-  (Timo Heister, 2013/09/18)
-  
      27. 
-
-  
    27. 
-  Changed: we now call MPI_Init_thread instead of MPI_Init.
-  
      
-  (Timo Heister, 2013/09/17)
-  
      28. 
-
-  
    28. 
-  Enhancement: It is now possible to use the build directory directly without
-  the need to install first. For this, a second copy of all necessary project
-  configuration files (deal.IIConfig.cmake, ..., Make.global_options) are
-  generated and deployed in the build directory. (This is fully compatible with
-  the old possibility to install into the build dir.)
-  
      
-  (Matthias Maier, 2013/09/15)
-  
      29. 
-
-  
    29. 
-  Fixed: DoFTools::extract_locally_*_dofs now instantiated for hp::DofHandler.
-  
      
-  (Jean-Paul Pelteret, 2013/09/11)
-  
      30. 
-
-  
    30. 
-  Changed: distributed::parallel:BlockVector::operator= now allows importing
-  of ghost values like all other vector types. Also added some new constructors
-  for BlockVector and Vector using IndexSets to mirror the other linear algebra
-  classes.
-  
      
-  (Timo Heister, 2013/09/04)
-  
      31. 
-
-  
    31. 
-  Fixed: VectorTools::compute_no_normal_flux_constraints had a bug that
-  only manifested on complex meshes. This is now fixed.
-  
      
-  (Chih-Che Chueh, Wolfgang Bangerth, 2013/09/04)
-  
      32. 
-
-  
    32. 
-  New: All vector classes now have functions extract_subvector_to()
-  that allow extracting not just a single value but a whole set.
-  
      
-  (Fahad Alrashed, 2013/09/02)
-  
      33. 
-
-  
    33. 
-  Fixed: common/Make.global_options now exports enable-threads
-  correctly, furthermore, lib-suffix, shared-lib-suffix
-  and static-lib-suffix are now exported as well for better legacy
-  support.
-  
      
-  (Matthias Maier, 2013/08/30)
-  
      34. 
-
-  
    34. 
-  New: The ParameterHandler class can now deal with including one parameter
-  file from another.
-  
      
-  (Wolfgang Bangerth, 2013/08/25)
-  
      35. 
-
-  
    35. 
-  New: The method VectorTools::compute_normal_flux_constraints can be used to
-  force a vector finite element function to be normal to the boundary.
-  
      
-  (Martin Kronbichler, 2013/08/23)
-  
      36. 
-
-  
    36. 
-  Improved: MappingQ now uses the points of the Gauss-Lobatto quadrature
-  formula as support points instead of equispaced ones. This allows its use
-  for high polynomial orders and also gives better interpolation of circular
-  boundaries. Beware that mappings of order three and higher will behave
-  slightly differently now (usually better).
-  
      
-  (Martin Kronbichler, 2013/08/23)
-  
      37. 
-
-  
    37. 
-  Improved: Several .cc files in the deal.II directory have been split in
-  order to better utilize multiple processors when compiling in parallel and
-  reduce memory requirements of the compilation stage.
-  
      
-  (Martin Kronbichler, 2013/08/22)
-  
      38. 
-
-  
    38. 
-  Fixed: The ParameterHandler::declare_entry() did not check that the
-  default value of a parameter indeed satisfies the pattern given for this
-  parameter (despite a statement in the documentation that this checking
-  would happen). This is now fixed.
-  
      
-  (Wolfgang Bangerth, 2013/08/21)
-  
      39. 
-
-  
    39. 
-  New: Patterns::List and Patterns::Map now accept a string
-  different than the default comma that denotes the separator
-  between entries of the list or map.
-  
      
-  (Wolfgang Bangerth, 2013/08/21)
-  
      40. 
-
-  
    40. 
-  Fixed: Some operations in the MappingQ class are now done in higher
-  precision arithmetic to mitigate the ill-conditioning that appears
-  when using mappings of high order (say, order 6 or 8 or 10).
-  
      
-  (Juan Carlos Araujo Cabarcas, 2013/08/20)
-  
      41. 
-
-  
    41. 
-  Fixed: The SLEPcWrappers classes could not be compiled for 64-bit
-  indices. This is now fixed.
-  
      
-  (Denis Davydov, Wolfgang Bangerth, 2013/08/20)
-  
      42. 
-
-  
    42. 
-  Fixed: SolutionTransfer used to crash whenever one transfered in the hp
-  context between cells that use FE_Nothing and FE_Q. This is now fixed.
-  
      
-  (Krzyszof Bzowski, Wolfgang Bangerth, 2013/08/18)
-  
      43. 
-
-  
    43. 
-  Fixed: Under some circumstances (see http://code.google.com/p/dealii/issues/detail?id=82)
-  the DoFTools::make_periodicity_constraints() function could create cycles in
-  the ConstraintMatrix object. This is now fixed.
-  
      
-  (David Emerson, Wolfgang Bangerth, 2013/08/16)
-  
      44. 
-
-  
    44. 
-  New: There is now a function ConstraintMatrix::are_identity_constrained().
-  
      
-  (Wolfgang Bangerth, 2013/08/16)
-  
      45. 
-
-  
    45. 
-  New: TableHandler::write_text() now also supports output in
-  org-mode (http://orgmode.org/) format via a new entry in the
-  TableHandler::TextOutputFormat enumeration.
-  
      
-  (Oleh Krehel, 2013/08/15)
-  
      46. 
-
-  
    46. 
-  New: There are now global functions scalar_product
-  that compute the scalar product (double contraction) between
-  tensors of rank 2.
-  
      
-  (Scott Miller, 2013/08/14)
-  
      47. 
-
-  
    47. 
-  Fixed: Creating objects of type MappingQ was previously only possible
-  for low order polynomials. For orders higher than around 6, one ran
-  into assertions that tested for internal consistency. These assertions
-  have now been appropriately relaxes for the growth of round-off errors
-  with growing polynomial degrees.
-  
      
-  (Juan Carlos Araujo Cabarcas, Wolfgang Bangerth, 2013/08/14)
-  
      48. 
-
-  
    48. 
-  New: MappingQEulerian is now also instantiated for vector elements
-  of type TrilinosWrappers::Vector as well as the MPI and block
-  variants.
-  
      
-  (Armin Ghajar Jazi, 2013/08/14)
-  
      49. 
-
-  
    49. 
-  Fixed: The FiniteElement::face_to_cell_index() function had a bug
-  that made it work incorrectly for elements that have more than one
-  degree of freedom per line (in 2d) or per quad (in 3d). This is now
-  fixed for the most common cases, namely the FE_Q elements as well
-  as elements composed of FESystem elements. For all other cases, an
-  exception is generated reporting that this case is not implemented.
-  If you run into this, let us know.
-  
      
-  (Wolfgang Bangerth, 2013/08/10)
-  
      50. 
-
-  
    50. 
-  New: DataOutBase::VtkFlags now has a flag
-  DataOutBase::VtkFlags::print_date_and_time that can be used to suppress output
-  of date and time in output files. This is useful in test suites where a newer
-  run at a different time produces differences against previously stored files,
-  even though the actual data is exactly the same.
-  
      
-  (Oleh Krehel, 2013/08/06)
-  
      51. 
-
-  
    51. 
-  Fixed: The various block matrix classes are all derived from BlockMatrixBase
-  which had race conditions when the set() or add() functions were called from
-  different threads. This is now fixed.
-  
      
-  (Wolfgang Bangerth, 2013/08/05)
-  
      52. 
-
-  
    52. 
-  Fixed: various fixes with assignment and reinit of PETScWrappers::MPI::Vector.
-  
      
-  (Timo Heister, 2013/08/05)
-  
      53. 
-
-  
    53. Fixed: An assertion wrongly triggered in
-  DoFTools::make_hanging_node_constraints when used with a particular
-  combination of FESystem elements containing FE_Nothing. This is now fixed.
-  
      
-  (Denis Davydov, Wolfgang Bangerth, 2013/08/01)
-  
      54. 
-
-  
    54. 
-  New: Add has_ghost_elements() for PETScWrappers::MPI::BlockVector and
-  TrilinosWrappers::MPI::BlockVector.
-  
      
-  (Timo Heister, 2013/08/01)
-  
      55. 
-
-  
    55. 
-  SparsityTools::distribute_sparsity_pattern did not work correctly for
-  block systems, this has been fixed (function has a different signature).
-  
      
-  (Timo Heister, 2013/07/31)
-  
      56. 
-
-  
    56. Fixed: When typing make run in the step-32 directory,
-  the program was executed with mpirun -np 2 ./step-32. This
-  assumes that a program mpirun exists, but also does that
-  deal.II was in fact compiled with MPI support on. Neither was intended.
-  This is now fixed.
-  
      
-  (Wolfgang Bangerth, 2013/07/24)
-  
      57. 
-
-  
    57. New: The DataOut, DataOutFaces, and DataOutRotation classes now allow
-  the output of data vectors using different DoFHandler objects (based on the
-  same triangulation), by new functions add_data_vector. This is used in the
-  step-31 tutorial program which avoids creating a joint DoFHandler just for
-  output.
-  
      
-  (Martin Kronbichler, 2013/07/24)
-  
      58. 
-
-  
    58. Changed: GridGenerator used to be a class with only static members
-  but is now a namespace, like all other similar constructs in deal.II.
-  
      
-  (Wolfgang Bangerth, 2013/07/24)
-  
      59. 
-
-  
    59. Changed: In GridGenerator, several functions had erroneously been changed
-  to take an argument of type size_type rather than unsigned
-  int. GridGenerator::size_type was a typedef to
-  types::global_dof_index, which for most users was unsigned int
-  anyway, but could also be set to be a 64-bit integer type. In any case, the
-  change has been reverted and these functions take just a regular
-  unsigned int again.
-  
      
-  (Wolfgang Bangerth, 2013/07/24)
-  
      60. 
 
    
 
 
diff --git a/deal.II/doc/news/news.html b/deal.II/doc/news/news.html
index b2eabc101b..a4be9e53c6 100644
--- a/deal.II/doc/news/news.html
+++ b/deal.II/doc/news/news.html
@@ -27,15 +27,30 @@
     mailing list for additional news.
     
    
 
+
     
    
       
    Recent changes
    
 	Changes in the library since the last major release are
-	here.
+	here.
       
    
 
+
     
    
+      
    
+          2013/12/24: Version 8.1 released
+      
    
+      
    
+          deal.II version 8.1 was released today. A full list of changes can be
+	  found here.
+	  
    
+	  deal.II can be downloaded at
+	  this site.
+      
    
+
       
    
           2013/11/01: step-42 solves an elasto-plastic contact problem 
       
    
diff --git a/deal.II/doc/publications/dealii.bib b/deal.II/doc/publications/dealii.bib
index 08d39692c9..96fb9bc2bc 100644
--- a/deal.II/doc/publications/dealii.bib
+++ b/deal.II/doc/publications/dealii.bib
@@ -12,70 +12,20 @@
 }
 
 
-%% The deal.II web page and online reference.
-%% Please cite specific versions only if you really need to
-
-@Manual{DealIIReference,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and T. Heister and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org}
-}
-
-@Manual{DealIIReference600,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org/6.0.0},
-  year = 2007
+@Misc{dealII80,
+          title =  {The {\tt deal.{I}{I}} Library, Version 8.0},
+          author = {W. Bangerth and T. Heister and L. Heltai and G. Kanschat
+                    and M. Kronbichler and M. Maier and B. Turcksin and T. D. Young},
+          note =   {\texttt{http://www.dealii.org}},
+          url =    {http://www.dealii.org},
+          year =   {2013}
+          }
 }
 
-@Manual{DealIIReference610,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org/6.1.0},
-  year = 2008
-}
 
-@Manual{DealIIReference621,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org/6.2.1},
-  year = 2009
-}
 
-@Manual{DealIIReference631,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org/6.3.1},
-  year = 2010
-}
 
-@Manual{DealIIReference700,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org/7.0.0},
-  year = 2011
-}
 
-@Manual{DealIIReference710,
-  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
-                  Technical Reference},
-  author = 	 {W. Bangerth and G. Kanschat},
-  note =         {\texttt{http://www.dealii.org}},
-  url =          {http://www.dealii.org/7.1.0},
-  year = 2011
-}
 
 %% Additional papers on deal.II and special aspects
 
@@ -146,3 +96,68 @@
   doi   =        {10.1016/j.compfluid.2012.04.012}
 }
 
+
+
+%% older references:
+
+@Manual{DealIIReference,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and T. Heister and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org}
+}
+
+@Manual{DealIIReference600,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org/6.0.0},
+  year = 2007
+}
+
+@Manual{DealIIReference610,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org/6.1.0},
+  year = 2008
+}
+
+@Manual{DealIIReference621,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org/6.2.1},
+  year = 2009
+}
+
+@Manual{DealIIReference631,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org/6.3.1},
+  year = 2010
+}
+
+@Manual{DealIIReference700,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org/7.0.0},
+  year = 2011
+}
+
+@Manual{DealIIReference710,
+  title = 	 {{\tt deal.{I}{I}} Differential Equations Analysis Library,
+                  Technical Reference},
+  author = 	 {W. Bangerth and G. Kanschat},
+  note =         {\texttt{http://www.dealii.org}},
+  url =          {http://www.dealii.org/7.1.0},
+  year = 2011
+}
diff --git a/deal.II/doc/publications/index.html b/deal.II/doc/publications/index.html
index 65522db8e5..d65bb6f0ae 100644
--- a/deal.II/doc/publications/index.html
+++ b/deal.II/doc/publications/index.html
@@ -106,27 +106,22 @@
 }
 
       
-      
  •  W. Bangerth,
-	T. Heister
-        and
-	G. Kanschat
-        
    
-        deal.II Differential Equations
-          Analysis Library, Technical Reference
+      
  • 
+	Wolfgang Bangerth, Timo Heister, Luca Heltai, Guido Kanschat, Martin Kronbichler, Matthias Maier, and Toby D. Young 
    
+        The deal.II Library, Version 8.0
         
         
    
-	To be found on the project webpage http://www.dealii.org
+	arxiv:1312.2266, 2013.
 	
    
 
    -@Manual{DealIIReference,
-  title =  {{\tt deal.{I}{I}} Differential Equations Analysis Library, Technical Reference},
-  author = {W. Bangerth and T. Heister and G. Kanschat and others},
-  note =   {\texttt{http://www.dealii.org}},
-  url =    {http://www.dealii.org}
+@Misc{dealII80,
+          title =  {The {\tt deal.{I}{I}} Library, Version 8.0},
+          author = {W. Bangerth and T. Heister and L. Heltai and G. Kanschat
+                    and M. Kronbichler and M. Maier and B. Turcksin and T. D. Young},
+          note =   {\texttt{http://www.dealii.org}},
+          url =    {http://www.dealii.org},
+          year =   {2013}
+          }
 }
 
    
       
    • 
@@ -354,11 +349,33 @@
 	  Preprint, MPI Magdeburg, 2013.
 	
 
+
+	
  •  A. Bonito, D. Devaud
+	  
    
+	  Adaptive Finite Element Methods for the Stokes Problems with Discontinuous Coefficients
+	  
    
+	  submitted, 2013.
+	
    • 
+
 	
  •  A. Bonito, R. DeVore, R. H. Nochetto
 	  
    
 	  Adaptive Finite Element Methods for Elliptic Problems with Discontinuous Coefficients
 	  
    
-	  arXiv:1301.3203, submitted, 2013.
+	  SIAM J. Numer. Anal.  51  (2013),  no. 6, 3106--3134.
+	
    • 
+
+	
  •  A. Bonito, R. Glowinski
+	  
    
+	  On the Nodal Set of the Eigenfunctions of the Laplace-Beltrami Operator for Bounded Surfaces in R3: A Computational Approach
+	  
    
+	  submitted, 2013.
+	
    • 
+
+	
  •  A. Bonito, J.-L. Guermond, S. Lee
+	  
    
+	  Two Modified Pressure-Correction Projection Methods for the Time-Dependent Stokes Equations
+	  
    
+	  submitted, 2013.
 	
    • 
 
 	
  •  A. Bonito, I. Kyza, R. H. Nochetto
@@ -384,6 +401,13 @@
 	  SIAM J. Numer. Anal., 51(1), 577--608, 2013.
 	
    • 
 
+	
  •  A. Bonito, J. E. Pasciak
+	  
    
+	  Numerical Approximation of Fractional Powers of Elliptic Operators
+	  
    
+	  submitted, 2013.
+	
    • 
+
 	
  •  J. Bosch, M. Stoll, P. Benner
 	  
    
 	  Fast solution of Cahn-Hilliard variational
diff --git a/deal.II/doc/readme.html b/deal.II/doc/readme.html
index f4c41181f3..28f2040c78 100644
--- a/deal.II/doc/readme.html
+++ b/deal.II/doc/readme.html
@@ -82,9 +82,12 @@
   
  • Mac OS X: GCC version 4.1 or later; Clang version 3.0 or later.
     Please see the deal.II Wiki for installation instructions.
    • 
-  
  • Windows: GCC version 4.8.* (Cygwin or MinGW, 64bit).
-    Please see the deal.II Wiki for installation instructions.
    • 
+  
  • Windows: Currently, Windows is not supported officially.
+    Please have a look at the
+    
+      FAQ and at the deal.II Wiki for some strategies to use deal.II on
+    Windows.
    • 
   
 
 
@@ -94,12 +97,6 @@
   please report it as a bug.
 
    
 
-
    
-  Please note that the Microsoft Visual C++ compiler, that is widely
-  used on Microsoft Windows, is not supported. If you need a native
-  Windows target, use the MinGW compiler instead.
-
    
-
 
 
 
    Additional software requirements
    
diff --git a/deal.II/doc/screen.css b/deal.II/doc/screen.css
index b13ec87ed6..0644b896be 100644
--- a/deal.II/doc/screen.css
+++ b/deal.II/doc/screen.css
@@ -1,6 +1,6 @@
-//
-// Copyright (C) 1998 - 2013 by the deal.II authors
-//
+/*
+ Copyright (C) 1998 - 2013 by the deal.II authors
+*/
 
 body {
     background-image: none;
diff --git a/deal.II/doc/users/cmakelists.html b/deal.II/doc/users/cmakelists.html
index e6a45736ff..b1f13d8d7a 100644
--- a/deal.II/doc/users/cmakelists.html
+++ b/deal.II/doc/users/cmakelists.html
@@ -61,7 +61,7 @@
   based on the DEAL_II_SETUP_TARGET macro. This method gives
   full control of what's happening and is easily extensible to more complex
   projects, as exemplified in the subsections here and later in the section
-  on advanced topics. Here is a full example
+  on advanced topics. Here is a full example
   (plain text
   version):
 
diff --git a/deal.II/doc/users/doxygen.html b/deal.II/doc/users/doxygen.html
index 3b226823ba..331aa020b6 100644
--- a/deal.II/doc/users/doxygen.html
+++ b/deal.II/doc/users/doxygen.html
@@ -32,6 +32,7 @@
       Doxygen, if the online documentation is available,
       here. For older revisions,
 	links are provided here by revision number:
+	8.1.0,
 	8.0.0,
 	7.3.0,
 	7.2.0,
diff --git a/deal.II/examples/step-26/step-26.cc b/deal.II/examples/step-26/step-26.cc
index d047fc7b5c..c7dd258c06 100644
--- a/deal.II/examples/step-26/step-26.cc
+++ b/deal.II/examples/step-26/step-26.cc
@@ -380,8 +380,8 @@ namespace Step26
   template
   void HeatEquation::run()
   {
-    const unsigned int initial_global_refinement = (dim == 2 ? 1 : 2);
-    const unsigned int n_adaptive_pre_refinement_steps = 1;
+    const unsigned int initial_global_refinement = 2;
+    const unsigned int n_adaptive_pre_refinement_steps = 4;
 
     GridGenerator::hyper_L (triangulation);
     triangulation.refine_global (initial_global_refinement);
@@ -458,7 +458,7 @@ start_time_iteration:
                                              system_rhs);
         }
 
-        constraints.condense (system_rhs);
+        constraints.condense (system_matrix, system_rhs);
 
         solve_time_step();
 
diff --git a/deal.II/examples/step-42/doc/intro.dox b/deal.II/examples/step-42/doc/intro.dox
index 27ce612c59..62168896d1 100644
--- a/deal.II/examples/step-42/doc/intro.dox
+++ b/deal.II/examples/step-42/doc/intro.dox
@@ -55,7 +55,7 @@ step-32.
 The classical formulation of the problem possesses the following form:
 @f{align*}
  \varepsilon(\mathbf u) &= A\sigma + \varepsilon^p & &\quad\text{in } \Omega,\\
-  -\textrm{\textrm{div}}\ \sigma &= \mathbf f & &\quad\text{in } \Omega,\\
+  -\textrm{div}\ \sigma &= \mathbf f & &\quad\text{in } \Omega,\\
   \varepsilon^p:(\tau - \sigma) &\geq 0\quad\forall\tau\text{ with
   }\mathcal{F}(\tau)\leq 0 & &\quad\text{in } \Omega,\\
   \mathbf u &= 0 & &\quad\text{on }\Gamma_D,\\
diff --git a/deal.II/include/deal.II/base/data_out_base.h b/deal.II/include/deal.II/base/data_out_base.h
index 5ea4eb017b..db413f05fd 100644
--- a/deal.II/include/deal.II/base/data_out_base.h
+++ b/deal.II/include/deal.II/base/data_out_base.h
@@ -3179,15 +3179,16 @@ public:
    * Read or write the data of this object for serialization
    */
   template 
-  void serialize(Archive &ar, const unsigned int version) {
+  void serialize(Archive &ar, const unsigned int version)
+  {
     ar &valid
-       &h5_sol_filename
-       &h5_mesh_filename
-       &entry_time
-       &num_nodes
-       &num_cells
-       &dimension
-       &attribute_dims;
+    &h5_sol_filename
+    &h5_mesh_filename
+    &entry_time
+    &num_nodes
+    &num_cells
+    &dimension
+    &attribute_dims;
   }
 
   /// Get the XDMF content associated with this entry.
diff --git a/deal.II/include/deal.II/base/exceptions.h b/deal.II/include/deal.II/base/exceptions.h
index 73506c7b0e..635c50d048 100644
--- a/deal.II/include/deal.II/base/exceptions.h
+++ b/deal.II/include/deal.II/base/exceptions.h
@@ -76,7 +76,7 @@ public:
   /**
    * Override the standard function that returns the description of the error.
    */
-  virtual const char* what() const throw();
+  virtual const char *what() const throw();
 
   /**
    * Get exception name.
@@ -266,7 +266,7 @@ namespace deal_II_exceptions
       // Fill the fields of the exception object
       e.set_fields (file, line, function, cond, exc_name);
 
-      switch(handling)
+      switch (handling)
         {
         case abort_on_exception:
           dealii::deal_II_exceptions::internals::abort(e);
@@ -297,15 +297,15 @@ namespace deal_II_exceptions
  */
 #ifdef DEBUG
 #define Assert(cond, exc)                                                   \
-{                                                                           \
-  if (!(cond))                                                              \
-    ::dealii::deal_II_exceptions::internals::                               \
-    issue_error(::dealii::deal_II_exceptions::internals::abort_on_exception,\
-                __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc); \
-}
+  {                                                                           \
+    if (!(cond))                                                              \
+      ::dealii::deal_II_exceptions::internals::                               \
+      issue_error(::dealii::deal_II_exceptions::internals::abort_on_exception,\
+                  __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc); \
+  }
 #else
 #define Assert(cond, exc)                                                   \
-{}
+  {}
 #endif
 
 
@@ -325,16 +325,16 @@ namespace deal_II_exceptions
  */
 #ifdef DEBUG
 #define AssertNothrow(cond, exc)                                            \
-{                                                                           \
-  if (!(cond))                                                              \
-    ::dealii::deal_II_exceptions::internals::                               \
-    issue_error(                                                            \
-      ::dealii::deal_II_exceptions::internals::abort_nothrow_on_exception,  \
-      __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc);           \
-}
+  {                                                                           \
+    if (!(cond))                                                              \
+      ::dealii::deal_II_exceptions::internals::                               \
+      issue_error(                                                            \
+          ::dealii::deal_II_exceptions::internals::abort_nothrow_on_exception,  \
+          __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc);           \
+  }
 #else
 #define AssertNothrow(cond, exc)                                            \
-{}
+  {}
 #endif
 
 
@@ -352,20 +352,20 @@ namespace deal_II_exceptions
  */
 #ifdef HAVE_BUILTIN_EXPECT
 #define AssertThrow(cond, exc)                                              \
-{                                                                           \
-  if (__builtin_expect(!(cond), false))                                     \
-    ::dealii::deal_II_exceptions::internals::                               \
-    issue_error(::dealii::deal_II_exceptions::internals::throw_on_exception,\
-                __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc); \
-}
+  {                                                                           \
+    if (__builtin_expect(!(cond), false))                                     \
+      ::dealii::deal_II_exceptions::internals::                               \
+      issue_error(::dealii::deal_II_exceptions::internals::throw_on_exception,\
+                  __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc); \
+  }
 #else /*ifdef HAVE_BUILTIN_EXPECT*/
 #define AssertThrow(cond, exc)                                              \
-{                                                                           \
-  if (!(cond))                                                              \
-    ::dealii::deal_II_exceptions::internals::                               \
-    issue_error(::dealii::deal_II_exceptions::internals::throw_on_exception,\
-                __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc); \
-}
+  {                                                                           \
+    if (!(cond))                                                              \
+      ::dealii::deal_II_exceptions::internals::                               \
+      issue_error(::dealii::deal_II_exceptions::internals::throw_on_exception,\
+                  __FILE__, __LINE__, __PRETTY_FUNCTION__, #cond, #exc, exc); \
+  }
 #endif /*ifdef HAVE_BUILTIN_EXPECT*/
 
 
diff --git a/deal.II/include/deal.II/base/geometry_info.h b/deal.II/include/deal.II/base/geometry_info.h
index 2b8b62331f..9859d77473 100644
--- a/deal.II/include/deal.II/base/geometry_info.h
+++ b/deal.II/include/deal.II/base/geometry_info.h
@@ -2172,11 +2172,11 @@ struct GeometryInfo
   static void
   alternating_form_at_vertices
 #ifndef DEAL_II_CONSTEXPR_BUG
-    (const Point (&vertices)[vertices_per_cell],
-     Tensor (&forms)[vertices_per_cell]);
+  (const Point (&vertices)[vertices_per_cell],
+   Tensor (&forms)[vertices_per_cell]);
 #else
-    (const Point *vertices,
-     Tensor *forms);
+  (const Point *vertices,
+   Tensor *forms);
 #endif
 
   /**
@@ -2372,7 +2372,8 @@ RefinementCase<2>::cut_axis (const unsigned int i)
   Assert (i < dim, ExcIndexRange(i, 0, dim));
 
   static const RefinementCase options[dim] = { RefinementPossibilities<2>::cut_x,
-					       RefinementPossibilities<2>::cut_y };
+                                               RefinementPossibilities<2>::cut_y
+                                             };
   return options[i];
 }
 
@@ -2387,8 +2388,9 @@ RefinementCase<3>::cut_axis (const unsigned int i)
   Assert (i < dim, ExcIndexRange(i, 0, dim));
 
   static const RefinementCase options[dim] = { RefinementPossibilities<3>::cut_x,
-					       RefinementPossibilities<3>::cut_y,
-					       RefinementPossibilities<3>::cut_z };
+                                               RefinementPossibilities<3>::cut_y,
+                                               RefinementPossibilities<3>::cut_z
+                                             };
   return options[i];
 }
 
@@ -2577,7 +2579,7 @@ GeometryInfo<2>::child_cell_from_point (const Point<2> &p)
 
   return (p[0] <= 0.5 ?
           (p[1] <= 0.5 ? 0 : 2) :
-            (p[1] <= 0.5 ? 1 : 3));
+          (p[1] <= 0.5 ? 1 : 3));
 }
 
 
@@ -2594,10 +2596,10 @@ GeometryInfo<3>::child_cell_from_point (const Point<3> &p)
   return (p[0] <= 0.5 ?
           (p[1] <= 0.5 ?
            (p[2] <= 0.5 ? 0 : 4) :
-             (p[2] <= 0.5 ? 2 : 6)) :
-            (p[1] <= 0.5 ?
-             (p[2] <= 0.5 ? 1 : 5) :
-             (p[2] <= 0.5 ? 3 : 7)));
+           (p[2] <= 0.5 ? 2 : 6)) :
+          (p[1] <= 0.5 ?
+           (p[2] <= 0.5 ? 1 : 5) :
+           (p[2] <= 0.5 ? 3 : 7)));
 }
 
 
diff --git a/deal.II/include/deal.II/base/graph_coloring.h b/deal.II/include/deal.II/base/graph_coloring.h
index d8cc56d27a..f3c58a24b3 100644
--- a/deal.II/include/deal.II/base/graph_coloring.h
+++ b/deal.II/include/deal.II/base/graph_coloring.h
@@ -236,15 +236,15 @@ namespace GraphColoring
       // intersection of the conflict indices.
       for (unsigned int i=0; i::iterator degrees_it;
diff --git a/deal.II/include/deal.II/base/logstream.h b/deal.II/include/deal.II/base/logstream.h
index fe03109f66..2c9519b6d2 100644
--- a/deal.II/include/deal.II/base/logstream.h
+++ b/deal.II/include/deal.II/base/logstream.h
@@ -551,7 +551,7 @@ private:
    */
   Threads::ThreadLocalStorage > outstreams;
 
-  template  friend LogStream &operator << (LogStream & log, const T &t);
+  template  friend LogStream &operator << (LogStream &log, const T &t);
 };
 
 
@@ -567,7 +567,7 @@ private:
  */
 template 
 inline
-LogStream & operator<< (LogStream &log, const T &t)
+LogStream &operator<< (LogStream &log, const T &t)
 {
   // print to the internal stringstream
   log.get_stream() << t;
diff --git a/deal.II/include/deal.II/base/parallel.h b/deal.II/include/deal.II/base/parallel.h
index 819d60f296..51c762f1c3 100644
--- a/deal.II/include/deal.II/base/parallel.h
+++ b/deal.II/include/deal.II/base/parallel.h
@@ -110,7 +110,7 @@ namespace parallel
        * three elements.
        */
       template 
+                typename I3, typename I4>
       static
       void
       apply (const F &f,
@@ -173,8 +173,8 @@ namespace parallel
    * module.
    */
   template 
+            typename OutputIterator,
+            typename Predicate>
   void transform (const InputIterator &begin_in,
                   const InputIterator &end_in,
                   OutputIterator       out,
@@ -239,9 +239,9 @@ namespace parallel
    * module.
    */
   template 
+            typename InputIterator2,
+            typename OutputIterator,
+            typename Predicate>
   void transform (const InputIterator1 &begin_in1,
                   const InputIterator1 &end_in1,
                   InputIterator2        in2,
@@ -307,10 +307,10 @@ namespace parallel
    * module.
    */
   template 
+            typename InputIterator2,
+            typename InputIterator3,
+            typename OutputIterator,
+            typename Predicate>
   void transform (const InputIterator1 &begin_in1,
                   const InputIterator1 &end_in1,
                   InputIterator2        in2,
@@ -563,7 +563,7 @@ namespace parallel
      * for each subrange.
      */
     template 
+              typename Function>
     struct ReductionOnSubranges
     {
       /**
diff --git a/deal.II/include/deal.II/base/synchronous_iterator.h b/deal.II/include/deal.II/base/synchronous_iterator.h
index 4bc42d379e..14b40192ba 100644
--- a/deal.II/include/deal.II/base/synchronous_iterator.h
+++ b/deal.II/include/deal.II/base/synchronous_iterator.h
@@ -166,7 +166,7 @@ void advance (std_cxx1x::tuple &t,
  * Advance a tuple of iterators by $n$.
  */
 template 
+          typename I3, typename I4>
 inline
 void advance (std_cxx1x::tuple &t,
               const unsigned int              n)
@@ -206,7 +206,7 @@ void advance_by_one (std_cxx1x::tuple &t)
  * Advance a tuple of iterators by 1.
  */
 template 
+          typename I3, typename I4>
 inline
 void advance_by_one (std_cxx1x::tuple &t)
 {
diff --git a/deal.II/include/deal.II/base/thread_management.h b/deal.II/include/deal.II/base/thread_management.h
index ab3cde9387..337e0933f9 100644
--- a/deal.II/include/deal.II/base/thread_management.h
+++ b/deal.II/include/deal.II/base/thread_management.h
@@ -1034,7 +1034,7 @@ namespace Threads
      * for each number of arguments.
      */
     template ::value>
+              int length = std_cxx1x::tuple_size::value>
     struct fun_ptr_helper;
 
 
@@ -1477,7 +1477,7 @@ namespace Threads
        * returned by the function called on
        * the thread.
        */
-        std_cxx1x::shared_ptr > ret_val;
+      std_cxx1x::shared_ptr > ret_val;
 
       /**
        * Start the thread and
@@ -1570,8 +1570,8 @@ namespace Threads
      * Copy constructor.
      */
     Thread (const Thread &t)
-    :
-    thread_descriptor (t.thread_descriptor)
+      :
+      thread_descriptor (t.thread_descriptor)
     {}
 
     /**
@@ -2330,14 +2330,14 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   internal::fun_encapsulator,3>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   internal::fun_encapsulator,3>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3))
@@ -2355,14 +2355,14 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   internal::fun_encapsulator,3>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   internal::fun_encapsulator,3>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3))
@@ -2380,14 +2380,14 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   internal::fun_encapsulator,3>
   spawn (const C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3) const) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   internal::fun_encapsulator,3>
   spawn (const C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3) const)
@@ -2409,14 +2409,14 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   internal::fun_encapsulator,4>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   internal::fun_encapsulator,4>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4))
@@ -2434,14 +2434,14 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   internal::fun_encapsulator,4>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   internal::fun_encapsulator,4>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4))
@@ -2459,14 +2459,14 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   internal::fun_encapsulator,4>
   spawn (const C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4) const) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   internal::fun_encapsulator,4>
   spawn (const C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4) const)
@@ -2487,16 +2487,16 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   internal::fun_encapsulator,5>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   internal::fun_encapsulator,5>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5))
@@ -2514,16 +2514,16 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   internal::fun_encapsulator,5>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   internal::fun_encapsulator,5>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5))
@@ -2541,16 +2541,16 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   internal::fun_encapsulator,5>
   spawn (const C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5) const) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   internal::fun_encapsulator,5>
   spawn (const C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5) const)
@@ -2571,16 +2571,16 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   internal::fun_encapsulator,6>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   internal::fun_encapsulator,6>
   spawn (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6))
@@ -2598,16 +2598,16 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   internal::fun_encapsulator,6>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6)) DEAL_II_DEPRECATED;
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   internal::fun_encapsulator,6>
   spawn (C &c, RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6))
@@ -2625,8 +2625,8 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   internal::fun_encapsulator,6>
@@ -2634,8 +2634,8 @@ namespace Threads
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   internal::fun_encapsulator,6>
@@ -2657,9 +2657,9 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   internal::fun_encapsulator,7>
@@ -2667,9 +2667,9 @@ namespace Threads
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   internal::fun_encapsulator,7>
@@ -2688,9 +2688,9 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   internal::fun_encapsulator,7>
@@ -2698,9 +2698,9 @@ namespace Threads
 
 
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   internal::fun_encapsulator,7>
@@ -2719,9 +2719,9 @@ namespace Threads
    * @deprecated Use new_thread() instead.
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   internal::fun_encapsulator
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3),
@@ -3215,7 +3215,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3),
@@ -3241,7 +3241,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3) const,
@@ -3270,7 +3270,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4),
@@ -3298,7 +3298,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4),
@@ -3326,7 +3326,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4) const,
@@ -3356,8 +3356,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5),
@@ -3387,8 +3387,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5),
@@ -3418,8 +3418,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5) const,
@@ -3451,8 +3451,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6),
@@ -3484,8 +3484,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6),
@@ -3517,8 +3517,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6) const,
@@ -3552,9 +3552,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6,Arg7),
@@ -3588,9 +3588,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6,Arg7),
@@ -3624,9 +3624,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6,Arg7) const,
@@ -3662,9 +3662,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -3701,9 +3701,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -3740,9 +3740,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -3781,9 +3781,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   Thread
   new_thread (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -3822,9 +3822,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -3863,9 +3863,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   Thread
   new_thread (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -4341,8 +4341,8 @@ namespace Threads
      * Copy constructor.
      */
     Task (const Task &t)
-    :
-    task_descriptor (t.task_descriptor)
+      :
+      task_descriptor (t.task_descriptor)
     {}
 
 
@@ -4649,7 +4649,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3),
@@ -4675,7 +4675,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3),
@@ -4701,7 +4701,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3) const,
@@ -4730,7 +4730,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4),
@@ -4758,7 +4758,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4),
@@ -4786,7 +4786,7 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3, typename Arg4>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4) const,
@@ -4816,8 +4816,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5),
@@ -4847,8 +4847,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5),
@@ -4878,8 +4878,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5) const,
@@ -4911,8 +4911,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6),
@@ -4944,8 +4944,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6),
@@ -4977,8 +4977,8 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6) const,
@@ -5012,9 +5012,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6,Arg7),
@@ -5048,9 +5048,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6,Arg7),
@@ -5084,9 +5084,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,Arg6,Arg7) const,
@@ -5122,9 +5122,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -5161,9 +5161,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -5200,9 +5200,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -5241,9 +5241,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   Task
   new_task (RT (*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -5282,9 +5282,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
@@ -5323,9 +5323,9 @@ namespace Threads
    * @ingroup threads
    */
   template 
+            typename Arg1, typename Arg2, typename Arg3,
+            typename Arg4, typename Arg5, typename Arg6,
+            typename Arg7, typename Arg8, typename Arg9>
   inline
   Task
   new_task (RT (C::*fun_ptr)(Arg1,Arg2,Arg3,Arg4,Arg5,
diff --git a/deal.II/include/deal.II/base/work_stream.h b/deal.II/include/deal.II/base/work_stream.h
index 4236b80911..363c28bbc7 100644
--- a/deal.II/include/deal.II/base/work_stream.h
+++ b/deal.II/include/deal.II/base/work_stream.h
@@ -162,8 +162,8 @@ namespace WorkStream
        * A class that creates a sequence of items from a range of iterators.
        */
       template 
+                typename ScratchData,
+                typename CopyData>
       class IteratorRangeToItemStream : public tbb::filter
       {
       public:
@@ -481,8 +481,8 @@ namespace WorkStream
        * run in parallel.
        */
       template 
+                typename ScratchData,
+                typename CopyData>
       class Worker : public tbb::filter
       {
       public:
@@ -635,8 +635,8 @@ namespace WorkStream
        * they are created.
        */
       template 
+                typename ScratchData,
+                typename CopyData>
       class Copier : public tbb::filter
       {
       public:
@@ -722,8 +722,8 @@ namespace WorkStream
        * A class that creates a sequence of items from a range of iterators.
        */
       template 
+                typename ScratchData,
+                typename CopyData>
       class IteratorRangeToItemStream : public tbb::filter
       {
       public:
@@ -1009,8 +1009,8 @@ namespace WorkStream
        * run in parallel.
        */
       template 
+                typename ScratchData,
+                typename CopyData>
       class WorkerAndCopier : public tbb::filter
       {
       public:
@@ -1198,10 +1198,10 @@ namespace WorkStream
    * are generated.
    */
   template 
+            typename Copier,
+            typename Iterator,
+            typename ScratchData,
+            typename CopyData>
   void
   run (const Iterator                          &begin,
        const typename identity::type &end,
@@ -1331,10 +1331,10 @@ namespace WorkStream
    * are generated.
    */
   template 
+            typename Copier,
+            typename Iterator,
+            typename ScratchData,
+            typename CopyData>
   void
   run (const std::vector > &colored_iterators,
        Worker                                   worker,
@@ -1443,9 +1443,9 @@ namespace WorkStream
    * are generated.
    */
   template 
+            typename Iterator,
+            typename ScratchData,
+            typename CopyData>
   void
   run (const Iterator                          &begin,
        const typename identity::type &end,
diff --git a/deal.II/include/deal.II/distributed/tria.h b/deal.II/include/deal.II/distributed/tria.h
index 1bfe32cd5a..f3285eb009 100644
--- a/deal.II/include/deal.II/distributed/tria.h
+++ b/deal.II/include/deal.II/distributed/tria.h
@@ -159,7 +159,10 @@ namespace internal
 }
 
 //forward declaration of the data type for periodic face pairs
-namespace GridTools {template  struct PeriodicFacePair;}
+namespace GridTools
+{
+  template  struct PeriodicFacePair;
+}
 
 namespace parallel
 {
@@ -679,7 +682,7 @@ namespace parallel
       register_data_attach (const std::size_t size,
                             const std_cxx1x::function & pack_callback);
+                                                            void *)> &pack_callback);
 
       /**
        * The given function is called for
@@ -691,7 +694,7 @@ namespace parallel
       notify_ready_to_unpack (const unsigned int offset,
                               const std_cxx1x::function & unpack_callback);
+                                                              const void *)> &unpack_callback);
 
       /**
        * Returns a permutation vector for the order the coarse
@@ -723,7 +726,7 @@ namespace parallel
        */
       void
       add_periodicity
-      (const std::vector >&);
+      (const std::vector > &);
 
 
 
@@ -934,8 +937,8 @@ namespace parallel
        */
       void
       fill_vertices_with_ghost_neighbors
-        (std::map >
-         &vertices_with_ghost_neighbors);
+      (std::map >
+       &vertices_with_ghost_neighbors);
 
       template  friend class dealii::internal::DoFHandler::Policy::ParallelDistributed;
     };
@@ -1040,8 +1043,8 @@ namespace parallel
        */
       void
       fill_vertices_with_ghost_neighbors
-        (std::map >
-         &vertices_with_ghost_neighbors);
+      (std::map >
+       &vertices_with_ghost_neighbors);
 
     };
   }
diff --git a/deal.II/include/deal.II/dofs/dof_accessor.templates.h b/deal.II/include/deal.II/dofs/dof_accessor.templates.h
index 6ccc484d85..e3dc97f94b 100644
--- a/deal.II/include/deal.II/dofs/dof_accessor.templates.h
+++ b/deal.II/include/deal.II/dofs/dof_accessor.templates.h
@@ -514,7 +514,7 @@ namespace internal
       {
         return dof_handler.faces->lines.
                get_dof_index (dof_handler,
-                       obj_index,
+                              obj_index,
                               fe_index,
                               local_index,
                               obj_level);
@@ -814,7 +814,7 @@ namespace internal
                           const dealii::internal::int2type<1> &)
       {
         return dof_handler.levels[obj_level]->fe_index_is_active(obj_index,
-               fe_index);
+                                                                 fe_index);
       }
 
 
@@ -826,7 +826,7 @@ namespace internal
                            const unsigned int obj_index,
                            const dealii::internal::int2type<1> &)
       {
-	// on a cell, the number of active elements is one
+        // on a cell, the number of active elements is one
         return 1;
       }
 
@@ -841,7 +841,7 @@ namespace internal
                            const unsigned int n,
                            const dealii::internal::int2type<1> &)
       {
-	Assert (n==0, ExcMessage("On cells, there can only be one active FE index"));
+        Assert (n==0, ExcMessage("On cells, there can only be one active FE index"));
         return dof_handler.levels[obj_level]->active_fe_index (obj_index);
       }
 
@@ -902,7 +902,7 @@ namespace internal
                           const dealii::internal::int2type<2> &)
       {
         return dof_handler.levels[obj_level]->fe_index_is_active(obj_index,
-               fe_index);
+                                                                 fe_index);
       }
 
 
@@ -914,7 +914,7 @@ namespace internal
                            const unsigned int obj_index,
                            const dealii::internal::int2type<2> &)
       {
-	// on a cell, the number of active elements is one
+        // on a cell, the number of active elements is one
         return 1;
       }
 
@@ -929,7 +929,7 @@ namespace internal
                            const unsigned int n,
                            const dealii::internal::int2type<2> &)
       {
-	Assert (n==0, ExcMessage("On cells, there can only be one active FE index"));
+        Assert (n==0, ExcMessage("On cells, there can only be one active FE index"));
         return dof_handler.levels[obj_level]->active_fe_index (obj_index);
       }
 
@@ -1007,7 +1007,7 @@ namespace internal
                           const dealii::internal::int2type<3> &)
       {
         return dof_handler.levels[obj_level]->fe_index_is_active(obj_index,
-               fe_index);
+                                                                 fe_index);
       }
 
 
@@ -1050,7 +1050,7 @@ namespace internal
                            const unsigned int obj_index,
                            const dealii::internal::int2type<3> &)
       {
-	// on a cell, the number of active elements is one
+        // on a cell, the number of active elements is one
         return 1;
       }
 
@@ -1065,7 +1065,7 @@ namespace internal
                            const unsigned int n,
                            const dealii::internal::int2type<3> &)
       {
-	Assert (n==0, ExcMessage("On cells, there can only be one active FE index"));
+        Assert (n==0, ExcMessage("On cells, there can only be one active FE index"));
         return dof_handler.levels[obj_level]->active_fe_index (obj_index);
       }
 
@@ -2571,7 +2571,7 @@ namespace internal
         Assert (static_cast(accessor.present_index)
                 <
                 accessor.dof_handler->levels[accessor.present_level]
-                                             ->cell_cache_offsets.size(),
+                ->cell_cache_offsets.size(),
                 ExcInternalError());
         Assert (accessor.dof_handler->levels[accessor.present_level]
                 ->cell_cache_offsets[accessor.present_index]
@@ -2582,12 +2582,12 @@ namespace internal
 
         std::vector dof_indices (dofs_per_cell);
         static_cast,lda> &>
-          (accessor).get_dof_indices (dof_indices, accessor.active_fe_index());
+        (accessor).get_dof_indices (dof_indices, accessor.active_fe_index());
 
         types::global_dof_index *next_dof_index
-        = &accessor.dof_handler->levels[accessor.present_level]
-           ->cell_dof_indices_cache[accessor.dof_handler->levels[accessor.present_level]
-                                                                 ->cell_cache_offsets[accessor.present_index]];
+          = &accessor.dof_handler->levels[accessor.present_level]
+            ->cell_dof_indices_cache[accessor.dof_handler->levels[accessor.present_level]
+                                     ->cell_cache_offsets[accessor.present_index]];
         for (unsigned int i=0; ilevels[accessor.level()]
-	  ->active_fe_index(accessor.present_index);
+               ->active_fe_index(accessor.present_index);
       }
 
 
@@ -2821,7 +2821,7 @@ namespace internal
                 ExcMessage ("DoFHandler not initialized"));
 
         accessor.dof_handler->levels[accessor.level()]
-	  ->set_active_fe_index (accessor.present_index, i);
+        ->set_active_fe_index (accessor.present_index, i);
       }
 
 
@@ -3033,7 +3033,7 @@ namespace internal
 
 
       template 
+                class OutputMatrix, typename OutputVector>
       static
       void
       distribute_local_to_global (const DoFCellAccessor, lda> &accessor,
@@ -3078,7 +3078,7 @@ namespace internal
 
 
       template 
+                class OutputMatrix, typename OutputVector>
       static
       void
       distribute_local_to_global (const DoFCellAccessor, lda> &accessor,
@@ -3287,8 +3287,8 @@ DoFCellAccessor::get_dof_indices (std::vector &
   AssertDimension (dof_indices.size(), this->get_fe().dofs_per_cell);
 
   const types::global_dof_index *cache
-  = this->dof_handler->levels[this->present_level]
-    ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
+    = this->dof_handler->levels[this->present_level]
+      ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
   for (unsigned int i=0; iget_fe().dofs_per_cell; ++i, ++cache)
     dof_indices[i] = *cache;
 }
@@ -3357,8 +3357,8 @@ DoFCellAccessor::get_dof_values (const InputVector &values,
           typename DoFCellAccessor::ExcVectorDoesNotMatch());
 
   const types::global_dof_index *cache
-  = this->dof_handler->levels[this->present_level]
-    ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
+    = this->dof_handler->levels[this->present_level]
+      ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
 
   values.extract_subvector_to (cache,
                                cache + this->get_fe().dofs_per_cell,
@@ -3385,12 +3385,12 @@ DoFCellAccessor::get_dof_values (const ConstraintMatrix &constraints,
           == this->get_fe().dofs_per_cell,
           typename DoFCellAccessor::ExcVectorDoesNotMatch());
   Assert (values.size() == this->get_dof_handler().n_dofs(),
-      typename DoFCellAccessor::ExcVectorDoesNotMatch());
+          typename DoFCellAccessor::ExcVectorDoesNotMatch());
 
 
   const types::global_dof_index *cache
-  = this->dof_handler->levels[this->present_level]
-    ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
+    = this->dof_handler->levels[this->present_level]
+      ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
 
   constraints.get_dof_values(values, *cache, local_values_begin,
                              local_values_end);
@@ -3414,12 +3414,12 @@ DoFCellAccessor::set_dof_values (const Vector &local_values,
           == this->get_fe().dofs_per_cell,
           typename DoFCellAccessor::ExcVectorDoesNotMatch());
   Assert (values.size() == this->get_dof_handler().n_dofs(),
-      typename DoFCellAccessor::ExcVectorDoesNotMatch());
+          typename DoFCellAccessor::ExcVectorDoesNotMatch());
 
 
   const types::global_dof_index *cache
-  = this->dof_handler->levels[this->present_level]
-    ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
+    = this->dof_handler->levels[this->present_level]
+      ->get_cell_cache_start (this->present_index, this->get_fe().dofs_per_cell);
 
   for (unsigned int i=0; iget_fe().dofs_per_cell; ++i, ++cache)
     values(*cache) = local_values(i);
diff --git a/deal.II/include/deal.II/dofs/dof_handler.h b/deal.II/include/deal.II/dofs/dof_handler.h
index c56d20ed5e..5feae64959 100644
--- a/deal.II/include/deal.II/dofs/dof_handler.h
+++ b/deal.II/include/deal.II/dofs/dof_handler.h
@@ -1262,8 +1262,8 @@ namespace internal
    *
    * Defined in dof_handler.cc.
    */
-    template
-    std::string policy_to_string(const dealii::internal::DoFHandler::Policy::PolicyBase & policy);
+  template
+  std::string policy_to_string(const dealii::internal::DoFHandler::Policy::PolicyBase &policy);
 
 }
 
@@ -1327,11 +1327,11 @@ void DoFHandler::load (Archive &ar,
                            "the one that was associated with the DoFHandler previously stored."));
   AssertThrow (policy_name == internal::policy_to_string(*policy),
                ExcMessage (std::string ("The policy currently associated with this DoFHandler (")
-			   + internal::policy_to_string(*policy)
-			   +std::string(") does not match the one that was associated with the "
-				       "DoFHandler previously stored (")
-			   + policy_name
-			   + ")."));
+                           + internal::policy_to_string(*policy)
+                           +std::string(") does not match the one that was associated with the "
+                                        "DoFHandler previously stored (")
+                           + policy_name
+                           + ")."));
 }
 
 
diff --git a/deal.II/include/deal.II/dofs/dof_levels.h b/deal.II/include/deal.II/dofs/dof_levels.h
index c657388fe9..6982bc7f7b 100644
--- a/deal.II/include/deal.II/dofs/dof_levels.h
+++ b/deal.II/include/deal.II/dofs/dof_levels.h
@@ -101,7 +101,7 @@ namespace internal
        */
       const types::global_dof_index *
       get_cell_cache_start (const unsigned int obj_index,
-                               const unsigned int dofs_per_cell) const;
+                            const unsigned int dofs_per_cell) const;
 
       /**
        * Determine an estimate for the
@@ -125,7 +125,7 @@ namespace internal
     inline
     const types::global_dof_index *
     DoFLevel::get_cell_cache_start (const unsigned int obj_index,
-                                             const unsigned int dofs_per_cell) const
+                                         const unsigned int dofs_per_cell) const
     {
       Assert (obj_index*dofs_per_cell+dofs_per_cell
               <=
diff --git a/deal.II/include/deal.II/dofs/dof_tools.h b/deal.II/include/deal.II/dofs/dof_tools.h
index f1a02d0fe9..d00ab54b23 100644
--- a/deal.II/include/deal.II/dofs/dof_tools.h
+++ b/deal.II/include/deal.II/dofs/dof_tools.h
@@ -53,7 +53,10 @@ class ConstraintMatrix;
 template  class InterGridMap;
 template  class Mapping;
 
-namespace GridTools {template  struct PeriodicFacePair;}
+namespace GridTools
+{
+  template  struct PeriodicFacePair;
+}
 
 //TODO: map_support_points_to_dofs should generate a multimap, rather than just a map, since several dofs may be located at the same support point
 
@@ -1024,8 +1027,8 @@ namespace DoFTools
   template
   void
   make_periodicity_constraints
-  (const std::vector > 
-     &periodic_faces,
+  (const std::vector >
+   &periodic_faces,
    dealii::ConstraintMatrix &constraint_matrix,
    const ComponentMask      &component_mask = ComponentMask());
 
diff --git a/deal.II/include/deal.II/fe/fe_raviart_thomas.h b/deal.II/include/deal.II/fe/fe_raviart_thomas.h
index 0831bf5c2c..8f46f9f577 100644
--- a/deal.II/include/deal.II/fe/fe_raviart_thomas.h
+++ b/deal.II/include/deal.II/fe/fe_raviart_thomas.h
@@ -43,7 +43,7 @@ template  class MappingQ;
  *
  * We follow the usual definition of the degree of RT elements, which
  * denotes the polynomial degree of the largest complete polynomial
- * subspace contained in the RT space. Then, approxiamtion order of
+ * subspace contained in the RT space. Then, approximation order of
  * the function itself is degree+1, as with usual polynomial
  * spaces. The numbering so chosen implies the sequence
  * @f[
diff --git a/deal.II/include/deal.II/fe/fe_tools.h b/deal.II/include/deal.II/fe/fe_tools.h
index 1cc390f751..362edb59d4 100644
--- a/deal.II/include/deal.II/fe/fe_tools.h
+++ b/deal.II/include/deal.II/fe/fe_tools.h
@@ -552,9 +552,9 @@ namespace FETools
    * @p distribute function of your hanging node constraints object.
    */
   template  class DH1,
-           template  class DH2,
-           class InVector, class OutVector>
+            template  class DH1,
+            template  class DH2,
+            class InVector, class OutVector>
   void
   interpolate (const DH1 &dof1,
                const InVector          &u1,
@@ -578,9 +578,9 @@ namespace FETools
    * DoF values at the discontinuities.
    */
   template  class DH1,
-           template  class DH2,
-           class InVector, class OutVector>
+            template  class DH1,
+            template  class DH2,
+            class InVector, class OutVector>
   void interpolate (const DH1  &dof1,
                     const InVector           &u1,
                     const DH2  &dof2,
@@ -611,8 +611,8 @@ namespace FETools
    * type @p hp::DoFHandler.
    */
   template  class DH,
-           class InVector, class OutVector, int spacedim>
+            template  class DH,
+            class InVector, class OutVector, int spacedim>
   void back_interpolate (const DH            &dof1,
                          const InVector           &u1,
                          const FiniteElement &fe2,
diff --git a/deal.II/include/deal.II/fe/fe_values.h b/deal.II/include/deal.II/fe/fe_values.h
index 153c1b5721..0802018b05 100644
--- a/deal.II/include/deal.II/fe/fe_values.h
+++ b/deal.II/include/deal.II/fe/fe_values.h
@@ -2220,13 +2220,13 @@ public:
    */
   DeclException1 (ExcAccessToUninitializedField,
                   char *,
-		  << ("You are requesting information from an FEValues/FEFaceValues/FESubfaceValues "
-		      "object for which this kind of information has not been computed. What "
-		      "information these objects compute is determined by the update_* flags you "
-		      "pass to the constructor. Here, the operation you are attempting requires "
-		      "the <")
-		  << arg1
-		  << "> flag to be set, but it was apparently not specified upon construction.");
+                  << ("You are requesting information from an FEValues/FEFaceValues/FESubfaceValues "
+                      "object for which this kind of information has not been computed. What "
+                      "information these objects compute is determined by the update_* flags you "
+                      "pass to the constructor. Here, the operation you are attempting requires "
+                      "the <")
+                  << arg1
+                  << "> flag to be set, but it was apparently not specified upon construction.");
   /**
    * @todo Document this
    *
@@ -3896,7 +3896,7 @@ const std::vector > &
 FEValuesBase::get_quadrature_points () const
 {
   Assert (this->update_flags & update_quadrature_points,
-	  ExcAccessToUninitializedField("update_quadrature_points"));
+          ExcAccessToUninitializedField("update_quadrature_points"));
   return this->quadrature_points;
 }
 
@@ -3908,7 +3908,7 @@ const std::vector &
 FEValuesBase::get_JxW_values () const
 {
   Assert (this->update_flags & update_JxW_values,
-	  ExcAccessToUninitializedField("update_JxW_values"));
+          ExcAccessToUninitializedField("update_JxW_values"));
   return this->JxW_values;
 }
 
@@ -3920,7 +3920,7 @@ const std::vector > &
 FEValuesBase::get_jacobians () const
 {
   Assert (this->update_flags & update_jacobians,
-	  ExcAccessToUninitializedField("update_jacobians"));
+          ExcAccessToUninitializedField("update_jacobians"));
   return this->jacobians;
 }
 
@@ -3932,7 +3932,7 @@ const std::vector > &
 FEValuesBase::get_jacobian_grads () const
 {
   Assert (this->update_flags & update_jacobian_grads,
-	  ExcAccessToUninitializedField("update_jacobians_grads"));
+          ExcAccessToUninitializedField("update_jacobians_grads"));
   return this->jacobian_grads;
 }
 
@@ -3944,7 +3944,7 @@ const std::vector > &
 FEValuesBase::get_inverse_jacobians () const
 {
   Assert (this->update_flags & update_inverse_jacobians,
-	  ExcAccessToUninitializedField("update_inverse_jacobians"));
+          ExcAccessToUninitializedField("update_inverse_jacobians"));
   return this->inverse_jacobians;
 }
 
@@ -3956,7 +3956,7 @@ const Point &
 FEValuesBase::quadrature_point (const unsigned int i) const
 {
   Assert (this->update_flags & update_quadrature_points,
-	  ExcAccessToUninitializedField("update_quadrature_points"));
+          ExcAccessToUninitializedField("update_quadrature_points"));
   Assert (iquadrature_points.size(), ExcIndexRange(i, 0, this->quadrature_points.size()));
 
   return this->quadrature_points[i];
@@ -3971,7 +3971,7 @@ double
 FEValuesBase::JxW (const unsigned int i) const
 {
   Assert (this->update_flags & update_JxW_values,
-	  ExcAccessToUninitializedField("update_JxW_values"));
+          ExcAccessToUninitializedField("update_JxW_values"));
   Assert (iJxW_values.size(), ExcIndexRange(i, 0, this->JxW_values.size()));
 
   return this->JxW_values[i];
@@ -3985,7 +3985,7 @@ const DerivativeForm<1,dim,spacedim> &
 FEValuesBase::jacobian (const unsigned int i) const
 {
   Assert (this->update_flags & update_jacobians,
-	  ExcAccessToUninitializedField("update_jacobians"));
+          ExcAccessToUninitializedField("update_jacobians"));
   Assert (ijacobians.size(), ExcIndexRange(i, 0, this->jacobians.size()));
 
   return this->jacobians[i];
@@ -3999,7 +3999,7 @@ const DerivativeForm<2,dim,spacedim> &
 FEValuesBase::jacobian_grad (const unsigned int i) const
 {
   Assert (this->update_flags & update_jacobian_grads,
-	  ExcAccessToUninitializedField("update_jacobians_grads"));
+          ExcAccessToUninitializedField("update_jacobians_grads"));
   Assert (ijacobian_grads.size(), ExcIndexRange(i, 0, this->jacobian_grads.size()));
 
   return this->jacobian_grads[i];
@@ -4013,7 +4013,7 @@ const DerivativeForm<1,spacedim,dim> &
 FEValuesBase::inverse_jacobian (const unsigned int i) const
 {
   Assert (this->update_flags & update_inverse_jacobians,
-	  ExcAccessToUninitializedField("update_inverse_jacobians"));
+          ExcAccessToUninitializedField("update_inverse_jacobians"));
   Assert (iinverse_jacobians.size(), ExcIndexRange(i, 0, this->inverse_jacobians.size()));
 
   return this->inverse_jacobians[i];
diff --git a/deal.II/include/deal.II/grid/grid_generator.h b/deal.II/include/deal.II/grid/grid_generator.h
index 7307374abe..0ad8634481 100644
--- a/deal.II/include/deal.II/grid/grid_generator.h
+++ b/deal.II/include/deal.II/grid/grid_generator.h
@@ -606,12 +606,12 @@ namespace GridGenerator
    */
   template
   void hyper_cube_with_cylindrical_hole (
-      Triangulation &triangulation,
-      const double        inner_radius = .25,
-      const double        outer_radius = .5,
-      const double        L = .5,
-      const unsigned int  repetition = 1,
-      const bool          colorize = false);
+    Triangulation &triangulation,
+    const double        inner_radius = .25,
+    const double        outer_radius = .5,
+    const double        L = .5,
+    const unsigned int  repetition = 1,
+    const bool          colorize = false);
 
   /**
    * Produce a ring of cells in 3D that is cut open, twisted and glued
diff --git a/deal.II/include/deal.II/grid/grid_tools.h b/deal.II/include/deal.II/grid/grid_tools.h
index 29cefdfe52..b65081d4a5 100644
--- a/deal.II/include/deal.II/grid/grid_tools.h
+++ b/deal.II/include/deal.II/grid/grid_tools.h
@@ -1066,7 +1066,9 @@ namespace GridTools
 
   /**
    * This function will collect periodic face pairs on the
-   * coarsest mesh level of the given @p container (a Triangulation or DoFHandler).
+   * coarsest mesh level of the given @p container (a Triangulation or
+   * DoFHandler) and add them to the vector @p matched_pairs leaving the
+   * original contents intact.
    *
    * Define a 'first' boundary as all boundary faces having boundary_id
    * @p b_id1 and a 'second' boundary consisting of all faces belonging
@@ -1093,15 +1095,21 @@ namespace GridTools
    * parallel::distributed::Triangulation::add_periodicity to enforce
    * periodicity algebraically.
    *
+   * @note Because elements will be added to @p matched_pairs (and existing
+   * entries will be preserved), it is possible to call this function several
+   * times with different boundary ids to generate a vector with all periodic
+   * pairs.
+   *
    * @author Daniel Arndt, Matthias Maier, 2013
    */
   template
-  std::vector >
+  void
   collect_periodic_faces
   (const CONTAINER &container,
    const types::boundary_id b_id1,
    const types::boundary_id b_id2,
    const int                direction,
+   std::vector > &matched_pairs,
    const dealii::Tensor<1,CONTAINER::space_dimension> &offset = dealii::Tensor<1,CONTAINER::space_dimension>());
 
 
@@ -1116,7 +1124,8 @@ namespace GridTools
    * face with local face index 2*dimension+1 and boundary
    * indicator @p b_id.
    *
-   * This function will collect periodic face pairs on the coarsest mesh level.
+   * This function will collect periodic face pairs on the coarsest mesh level
+   * and add them to @p matched_pairs leaving the original contents intact.
    *
    * @note This version of collect_periodic_face_pairs  will not work on
    * meshes with cells not in @ref GlossFaceOrientation
@@ -1125,11 +1134,12 @@ namespace GridTools
    * @author Daniel Arndt, Matthias Maier, 2013
    */
   template
-  std::vector >
+  void
   collect_periodic_faces
   (const CONTAINER                 &container,
    const types::boundary_id b_id,
    const int                direction,
+   std::vector > &matched_pairs,
    const dealii::Tensor<1,CONTAINER::space_dimension> &offset = dealii::Tensor<1,CONTAINER::space_dimension>());
 
 
diff --git a/deal.II/include/deal.II/grid/tria_objects.h b/deal.II/include/deal.II/grid/tria_objects.h
index 27e5af16d7..db5c01d685 100644
--- a/deal.II/include/deal.II/grid/tria_objects.h
+++ b/deal.II/include/deal.II/grid/tria_objects.h
@@ -175,7 +175,7 @@ namespace internal
         void serialize(Archive &ar,
                        const unsigned int version);
       };
-      
+
       /**
        * Store boundary and material data. For
        * example, in one dimension, this field
diff --git a/deal.II/include/deal.II/hp/dof_faces.h b/deal.II/include/deal.II/hp/dof_faces.h
index 3aa6f9c4d5..7812e9c530 100644
--- a/deal.II/include/deal.II/hp/dof_faces.h
+++ b/deal.II/include/deal.II/hp/dof_faces.h
@@ -405,8 +405,8 @@ namespace internal
             return *(pointer + 1 + local_index);
           else
             pointer += static_cast(
-              dof_handler.get_fe()[*pointer]
-              .template n_dofs_per_object() + 1);
+                         dof_handler.get_fe()[*pointer]
+                         .template n_dofs_per_object() + 1);
         }
     }
 
@@ -634,8 +634,8 @@ namespace internal
             return true;
           else
             pointer += static_cast(
-              dof_handler.get_fe()[*pointer]
-              .template n_dofs_per_object()+1);
+                         dof_handler.get_fe()[*pointer]
+                         .template n_dofs_per_object()+1);
         }
     }
 
diff --git a/deal.II/include/deal.II/hp/dof_handler.h b/deal.II/include/deal.II/hp/dof_handler.h
index 4ee8bab130..8b0c2b1ace 100644
--- a/deal.II/include/deal.II/hp/dof_handler.h
+++ b/deal.II/include/deal.II/hp/dof_handler.h
@@ -790,7 +790,7 @@ namespace hp
      * levels[] tree of
      * the Triangulation objects.
      */
-    std::vector levels;
+    std::vector levels;
 
     /**
      * Space to store the DoF
diff --git a/deal.II/include/deal.II/hp/dof_level.h b/deal.II/include/deal.II/hp/dof_level.h
index 0f9e43fd85..a582379c97 100644
--- a/deal.II/include/deal.II/hp/dof_level.h
+++ b/deal.II/include/deal.II/hp/dof_level.h
@@ -250,7 +250,7 @@ namespace internal
        */
       void
       set_active_fe_index (const unsigned int obj_index,
-			   const unsigned int fe_index);
+                           const unsigned int fe_index);
 
       /**
        * Return a pointer to the beginning of the DoF indices cache
@@ -265,7 +265,7 @@ namespace internal
        */
       const types::global_dof_index *
       get_cell_cache_start (const unsigned int obj_index,
-                               const unsigned int dofs_per_cell) const;
+                            const unsigned int dofs_per_cell) const;
 
       /**
        * Determine an estimate for the
@@ -330,7 +330,7 @@ namespace internal
                           "information is available"));
 
       Assert (fe_index == active_fe_indices[obj_index],
-	      ExcMessage ("FE index does not match that of the present cell"));
+              ExcMessage ("FE index does not match that of the present cell"));
 
       // see if the dof_indices array has been compressed for this
       // particular cell
@@ -363,7 +363,7 @@ namespace internal
       Assert ((signed_active_fe_index_type)active_fe_indices[obj_index]>=0,
               ExcMessage ("This function can no longer be called after compressing the dof_indices array"));
       Assert (fe_index == active_fe_indices[obj_index],
-	      ExcMessage ("FE index does not match that of the present cell"));
+              ExcMessage ("FE index does not match that of the present cell"));
       dof_indices[dof_offsets[obj_index]+local_index] = global_index;
     }
 
@@ -399,7 +399,7 @@ namespace internal
     void
     DoFLevel::
     set_active_fe_index (const unsigned int obj_index,
-			 const unsigned int fe_index)
+                         const unsigned int fe_index)
     {
       Assert (obj_index < active_fe_indices.size(),
               ExcIndexRange (obj_index, 0, active_fe_indices.size()));
@@ -412,7 +412,7 @@ namespace internal
     inline
     const types::global_dof_index *
     DoFLevel::get_cell_cache_start (const unsigned int obj_index,
-                                        const unsigned int dofs_per_cell) const
+                                    const unsigned int dofs_per_cell) const
     {
       Assert (obj_index < cell_cache_offsets.size(),
               ExcInternalError());
diff --git a/deal.II/include/deal.II/integrators/divergence.h b/deal.II/include/deal.II/integrators/divergence.h
index 836936425b..79438a8fba 100644
--- a/deal.II/include/deal.II/integrators/divergence.h
+++ b/deal.II/include/deal.II/integrators/divergence.h
@@ -187,18 +187,18 @@ namespace LocalIntegrators
     }
 
 
-/**
- * Cell matrix for gradient. The derivative is on the trial function.
- * \f[
- * \int_Z \nabla u \cdot \mathbf v\,dx
- * \f]
- *
- * This is the strong gradient and the trial space should be at least
- * in H1. The test functions can be discontinuous.
- *
- * @author Guido Kanschat
- * @date 2011
- */
+    /**
+     * Cell matrix for gradient. The derivative is on the trial function.
+     * \f[
+     * \int_Z \nabla u \cdot \mathbf v\,dx
+     * \f]
+     *
+     * This is the strong gradient and the trial space should be at least
+     * in H1. The test functions can be discontinuous.
+     *
+     * @author Guido Kanschat
+     * @date 2011
+     */
     template 
     void gradient_matrix(
       FullMatrix &M,
diff --git a/deal.II/include/deal.II/integrators/laplace.h b/deal.II/include/deal.II/integrators/laplace.h
index 61a58afbb8..156c274ef0 100644
--- a/deal.II/include/deal.II/integrators/laplace.h
+++ b/deal.II/include/deal.II/integrators/laplace.h
@@ -397,35 +397,47 @@ namespace LocalIntegrators
         {
           const double dx = fe1.JxW(k);
           const Point &n = fe1.normal_vector(k);
-	  for (unsigned int i=0; i
+            typename MATRIX2, typename INVERSE>
   void solve(
     const MATRIX1 &A,
     const MATRIX2 &B,
@@ -228,7 +228,7 @@ ArpackSolver::ArpackSolver (SolverControl &control,
 
 
 template 
+          typename MATRIX2, typename INVERSE>
 inline
 void ArpackSolver::solve (
   const MATRIX1 &system_matrix,
diff --git a/deal.II/include/deal.II/lac/block_list.h b/deal.II/include/deal.II/lac/block_list.h
index 97bfc00e20..782ccb4742 100644
--- a/deal.II/include/deal.II/lac/block_list.h
+++ b/deal.II/include/deal.II/lac/block_list.h
@@ -465,7 +465,7 @@ BlockList::cell_generates_vertex_patch(
       if (same_level_only && cell->neighbor(4)->neighbor(2)->level() != cell->level()) break;
       if (cell->neighbor(4)->neighbor(0)->at_boundary(2)) break;
       if (same_level_only && cell->neighbor(4)->neighbor(0)->neighbor(2)->level() != cell->level()) break;
-      // No break here
+    // No break here
     case 2:
       if (cell->at_boundary(2)) break;
       if (same_level_only && cell->neighbor(2)->level() != cell->level()) break;
@@ -578,7 +578,7 @@ BlockList::initialize_vertex_patches_mg(
                     indices[fe.face_to_cell_index(i,5)] = numbers::invalid_dof_index;
                 }
               add(k, indices);
-              // no break here
+            // no break here
             case 1:
               cell->get_mg_dof_indices(indices);
               for (unsigned int i=0; i
+            class VectorType>
   void vmult_block_nonblock (BlockVectorType          &dst,
                              const VectorType &src) const;
 
@@ -1106,7 +1106,7 @@ protected:
    * compiler.
    */
   template 
+            class VectorType>
   void vmult_nonblock_block (VectorType    &dst,
                              const BlockVectorType &src) const;
 
@@ -1174,7 +1174,7 @@ protected:
    * compiler.
    */
   template 
+            class VectorType>
   void Tvmult_block_nonblock (BlockVectorType  &dst,
                               const VectorType &src) const;
 
@@ -1197,7 +1197,7 @@ protected:
    * compiler.
    */
   template 
+            class VectorType>
   void Tvmult_nonblock_block (VectorType    &dst,
                               const BlockVectorType &src) const;
 
@@ -1301,7 +1301,7 @@ private:
      * beginning of their use, so there is no point actually copying
      * anything.
      */
-    TemporaryData & operator = (const TemporaryData &)
+    TemporaryData &operator = (const TemporaryData &)
     {
       return *this;
     }
@@ -2537,7 +2537,7 @@ vmult_block_block (BlockVectorType       &dst,
 
 template 
 template 
+          class VectorType>
 void
 BlockMatrixBase::
 vmult_nonblock_block (VectorType    &dst,
@@ -2557,7 +2557,7 @@ vmult_nonblock_block (VectorType    &dst,
 
 template 
 template 
+          class VectorType>
 void
 BlockMatrixBase::
 vmult_block_nonblock (BlockVectorType  &dst,
@@ -2638,7 +2638,7 @@ Tvmult_block_block (BlockVectorType       &dst,
 
 template 
 template 
+          class VectorType>
 void
 BlockMatrixBase::
 Tvmult_block_nonblock (BlockVectorType  &dst,
@@ -2659,7 +2659,7 @@ Tvmult_block_nonblock (BlockVectorType  &dst,
 
 template 
 template 
+          class VectorType>
 void
 BlockMatrixBase::
 Tvmult_nonblock_block (VectorType    &dst,
diff --git a/deal.II/include/deal.II/lac/block_sparse_matrix.h b/deal.II/include/deal.II/lac/block_sparse_matrix.h
index 75d3c5811e..35932009e9 100644
--- a/deal.II/include/deal.II/lac/block_sparse_matrix.h
+++ b/deal.II/include/deal.II/lac/block_sparse_matrix.h
@@ -278,7 +278,7 @@ public:
    * only one block column.
    */
   template 
+            typename nonblock_number>
   void vmult (BlockVector          &dst,
               const Vector &src) const;
 
@@ -290,7 +290,7 @@ public:
    * only one block row.
    */
   template 
+            typename nonblock_number>
   void vmult (Vector    &dst,
               const BlockVector &src) const;
 
@@ -325,7 +325,7 @@ public:
    * only one block row.
    */
   template 
+            typename nonblock_number>
   void Tvmult (BlockVector  &dst,
                const Vector &src) const;
 
@@ -337,7 +337,7 @@ public:
    * only one block column.
    */
   template 
+            typename nonblock_number>
   void Tvmult (Vector    &dst,
                const BlockVector &src) const;
 
@@ -501,7 +501,7 @@ BlockSparseMatrix::vmult (BlockVector       &dst,
 
 template 
 template 
+          typename nonblock_number>
 inline
 void
 BlockSparseMatrix::vmult (BlockVector     &dst,
@@ -514,7 +514,7 @@ BlockSparseMatrix::vmult (BlockVector     &dst,
 
 template 
 template 
+          typename nonblock_number>
 inline
 void
 BlockSparseMatrix::vmult (Vector         &dst,
@@ -551,7 +551,7 @@ BlockSparseMatrix::Tvmult (BlockVector       &dst,
 
 template 
 template 
+          typename nonblock_number>
 inline
 void
 BlockSparseMatrix::Tvmult (BlockVector     &dst,
@@ -564,7 +564,7 @@ BlockSparseMatrix::Tvmult (BlockVector     &dst,
 
 template 
 template 
+          typename nonblock_number>
 inline
 void
 BlockSparseMatrix::Tvmult (Vector         &dst,
diff --git a/deal.II/include/deal.II/lac/block_vector_base.h b/deal.II/include/deal.II/lac/block_vector_base.h
index 8991123408..a924af5d66 100644
--- a/deal.II/include/deal.II/lac/block_vector_base.h
+++ b/deal.II/include/deal.II/lac/block_vector_base.h
@@ -2553,7 +2553,7 @@ template 
 template 
 inline
 void BlockVectorBase::extract_subvector_to (const std::vector &indices,
-                                           std::vector &values) const
+                                                        std::vector &values) const
 {
   for (size_type i = 0; i < indices.size(); ++i)
     values[i] = operator()(indices[i]);
@@ -2568,10 +2568,12 @@ void BlockVectorBase::extract_subvector_to (ForwardIterator
                                                         const ForwardIterator    indices_end,
                                                         OutputIterator           values_begin) const
 {
-  while (indices_begin != indices_end) {
-    *values_begin = operator()(*indices_begin);
-    indices_begin++; values_begin++;
-  }
+  while (indices_begin != indices_end)
+    {
+      *values_begin = operator()(*indices_begin);
+      indices_begin++;
+      values_begin++;
+    }
 }
 
 #endif // DOXYGEN
diff --git a/deal.II/include/deal.II/lac/chunk_sparse_matrix.templates.h b/deal.II/include/deal.II/lac/chunk_sparse_matrix.templates.h
index f9d396419d..62f8409c8f 100644
--- a/deal.II/include/deal.II/lac/chunk_sparse_matrix.templates.h
+++ b/deal.II/include/deal.II/lac/chunk_sparse_matrix.templates.h
@@ -61,8 +61,8 @@ namespace internal
      * fragment.
      */
     template 
+              typename SrcIterator,
+              typename DstIterator>
     inline
     void
     chunk_vmult_add (const size_type      chunk_size,
@@ -92,8 +92,8 @@ namespace internal
      * the residual.
      */
     template 
+              typename SrcIterator,
+              typename DstIterator>
     inline
     void
     chunk_vmult_subtract (const size_type chunk_size,
@@ -123,8 +123,8 @@ namespace internal
      * chunk_size to the destination vector fragment.
      */
     template 
+              typename SrcIterator,
+              typename DstIterator>
     inline
     void
     chunk_Tvmult_add (const size_type      chunk_size,
@@ -150,9 +150,9 @@ namespace internal
      * individual chunk.
      */
     template 
+              typename MatrixIterator,
+              typename SrcIterator1,
+              typename SrcIterator2>
     inline
     result_type
     chunk_matrix_scalar_product (const size_type      chunk_size,
@@ -190,8 +190,8 @@ namespace internal
      * task scheduler.
      */
     template 
+              typename InVector,
+              typename OutVector>
     void vmult_add_on_subrange (const ChunkSparsityPattern &cols,
                                 const unsigned int  begin_row,
                                 const unsigned int  end_row,
diff --git a/deal.II/include/deal.II/lac/constraint_matrix.h b/deal.II/include/deal.II/lac/constraint_matrix.h
index e6cecd0ceb..33bf9bf3bb 100644
--- a/deal.II/include/deal.II/lac/constraint_matrix.h
+++ b/deal.II/include/deal.II/lac/constraint_matrix.h
@@ -215,7 +215,7 @@ public:
    * IndexSet, whereas otherwise it contains the global problem size and the
    * local range.
    */
-  const IndexSet & get_local_lines() const;
+  const IndexSet &get_local_lines() const;
 
   /**
    * This function copies the content of @p constraints_in with DoFs that are
@@ -436,7 +436,7 @@ public:
    * index2=index1.
    */
   bool are_identity_constrained (const size_type index1,
-                                     const size_type index2) const;
+                                 const size_type index2) const;
 
   /**
    * Return the maximum number of other dofs that one dof is constrained
@@ -857,7 +857,7 @@ public:
    * prevent data races.
    */
   template 
+            class VectorType>
   void
   distribute_local_to_global (ForwardIteratorVec local_vector_begin,
                               ForwardIteratorVec local_vector_end,
@@ -1035,7 +1035,7 @@ public:
    * set (i.e., distribute to be called).
    */
   template 
+            class VectorType>
   void
   get_dof_values (const VectorType  &global_vector,
                   ForwardIteratorInd local_indices_begin,
@@ -1290,76 +1290,6 @@ private:
    */
   bool sorted;
 
-  /**
-   * Scratch data that is used during calls to distribute_local_to_global and
-   * add_entries_local_to_global. In order to avoid frequent memory
-   * allocation, we keep the data alive from one call to the next.
-   */
-  struct ScratchData
-  {
-    /**
-     * Constructor, does nothing.
-     */
-    ScratchData () :
-      in_use (false)
-    {}
-
-    /**
-     * Copy constructor, does nothing
-     */
-    ScratchData (const ScratchData &) :
-      in_use (false)
-    {}
-
-    /**
-     * Stores whether the data is currently in use.
-     */
-    bool in_use;
-
-    /**
-     * Temporary array for column indices
-     */
-    std::vector columns;
-
-    /**
-     * Temporary array for column values
-     */
-    std::vector    values;
-
-    /**
-     * Temporary array for block start indices
-     */
-    std::vector block_starts;
-
-    /**
-     * Temporary array for vector indices
-     */
-    std::vector vector_indices;
-
-    /**
-     * Data array for reorder row/column indices. Use a shared ptr to
-     * global_rows to avoid defining in the .h file
-     */
-    std_cxx1x::shared_ptr global_rows;
-
-    /**
-     * Data array for reorder row/column indices. Use a shared ptr to
-     * global_rows to avoid defining in the .h file
-     */
-    std_cxx1x::shared_ptr global_columns;
-  };
-
-  /**
-   * Here comes the actual data structure for the scratch data. It is made
-   * mutable since it is modified in a const function. Since only one thread
-   * can access it at a time, no conflicting access can occur. For this to be
-   * valid, we need to make sure that no call within
-   * distribute_local_to_global is made that by itself can spawn
-   * tasks. Otherwise, we might end up in a situation where several threads
-   * fight for the data.
-   */
-  mutable Threads::ThreadLocalStorage scratch_data;
-
   /**
    * Internal function to calculate the index of line @p line in the vector
    * lines_cache using local_lines.
@@ -1472,8 +1402,7 @@ ConstraintMatrix::ConstraintMatrix (const IndexSet &local_constraints)
   :
   lines (),
   local_lines (local_constraints),
-  sorted (false),
-  scratch_data (ScratchData())
+  sorted (false)
 {
   // make sure the IndexSet is compressed. Otherwise this can lead to crashes
   // that are hard to find (only happen in release mode).
@@ -1490,8 +1419,7 @@ ConstraintMatrix::ConstraintMatrix (const ConstraintMatrix &constraint_matrix)
   lines (constraint_matrix.lines),
   lines_cache (constraint_matrix.lines_cache),
   local_lines (constraint_matrix.local_lines),
-  sorted (constraint_matrix.sorted),
-  scratch_data (ScratchData())
+  sorted (constraint_matrix.sorted)
 {}
 
 
@@ -1705,7 +1633,7 @@ void ConstraintMatrix::distribute_local_to_global (
 
 
 template 
+          class VectorType>
 inline
 void ConstraintMatrix::distribute_local_to_global (
   ForwardIteratorVec local_vector_begin,
@@ -1748,7 +1676,7 @@ ConstraintMatrix::distribute_local_to_global (
 
 
 template 
+          class VectorType>
 inline
 void ConstraintMatrix::get_dof_values (const VectorType  &global_vector,
                                        ForwardIteratorInd local_indices_begin,
diff --git a/deal.II/include/deal.II/lac/constraint_matrix.templates.h b/deal.II/include/deal.II/lac/constraint_matrix.templates.h
index d2e4dce533..36c3304668 100644
--- a/deal.II/include/deal.II/lac/constraint_matrix.templates.h
+++ b/deal.II/include/deal.II/lac/constraint_matrix.templates.h
@@ -1309,15 +1309,14 @@ namespace internals
         {
           AssertDimension(data.size(), individual_size.size()*row_length);
           // no space left in this row, need to double row_length and
-          // rearrange the data items
+          // rearrange the data items. Move all items to the right except the
+          // first one, starting at the back. Since individual_size contains
+          // at least one element when we get here, subtracting 1 works fine.
           data.resize(2*data.size());
-          for (size_type i=individual_size.size(); i>0; )
-            {
-              --i;
-              std::memmove(&data[i*row_length*2], &data[i*row_length],
-                           individual_size[i]*
-                           sizeof(std::pair));
-            }
+          for (size_type i=individual_size.size()-1; i>0; --i)
+            std::memmove(&data[i*row_length*2], &data[i*row_length],
+                         individual_size[i]*
+                         sizeof(std::pair));
           row_length *= 2;
         }
       data[index*row_length+my_length] = pair;
@@ -1633,6 +1632,142 @@ namespace internals
       }
   }
 
+
+
+  /**
+   * Scratch data that is used during calls to distribute_local_to_global and
+   * add_entries_local_to_global. In order to avoid frequent memory
+   * allocation, we keep the data alive from one call to the next in a static
+   * variable. Since we want to allow for different number types in matrices,
+   * this is a template.
+   *
+   * Since each thread gets its private version of scratch data out of the
+   * ThreadLocalStorage, no conflicting access can occur. For this to be
+   * valid, we need to make sure that no call within
+   * distribute_local_to_global is made that by itself can spawn
+   * tasks. Otherwise, we might end up in a situation where several threads
+   * fight for the data.
+   *
+   * Access to the scratch data is only through the accessor class which
+   * handles the access as well as marking the data as used.
+   */
+  template 
+  class ConstraintMatrixData
+  {
+  public:
+    struct ScratchData
+    {
+      /**
+       * Constructor, does nothing.
+       */
+      ScratchData ()
+        :
+        in_use (false)
+      {}
+
+      /**
+       * Copy constructor, does nothing
+       */
+      ScratchData (const ScratchData &)
+        :
+        in_use (false)
+      {}
+
+      /**
+       * Stores whether the data is currently in use.
+       */
+      bool in_use;
+
+      /**
+       * Temporary array for column indices
+       */
+      std::vector columns;
+
+      /**
+       * Temporary array for column values
+       */
+      std::vector    values;
+
+      /**
+       * Temporary array for block start indices
+       */
+      std::vector block_starts;
+
+      /**
+       * Temporary array for vector indices
+       */
+      std::vector vector_indices;
+
+      /**
+       * Data array for reorder row/column indices. Use a shared ptr to
+       * global_rows to avoid defining in the .h file
+       */
+      GlobalRowsFromLocal global_rows;
+
+      /**
+       * Data array for reorder row/column indices. Use a shared ptr to
+       * global_rows to avoid defining in the .h file
+       */
+      GlobalRowsFromLocal global_columns;
+    };
+
+    /**
+     * Accessor class to guard access to scratch_data
+     */
+    class ScratchDataAccessor
+    {
+    public:
+      /**
+       * Constructor. Grabs a scratch data object on the current thread and
+       * mark it as used
+       */
+      ScratchDataAccessor()
+        :
+        my_scratch_data(&ConstraintMatrixData::scratch_data.get())
+      {
+        Assert(my_scratch_data->in_use == false,
+               ExcMessage("Access to thread-local scratch data tried, but it is already "
+                          "in use"));
+        my_scratch_data->in_use = true;
+      }
+
+      /**
+       * Destructor. Mark scratch data as available again.
+       */
+      ~ScratchDataAccessor()
+      {
+        my_scratch_data->in_use = false;
+      }
+
+      /**
+       * Dereferencing operator.
+       */
+      ScratchData &operator* ()
+      {
+        return *my_scratch_data;
+      }
+
+      /**
+       * Dereferencing operator.
+       */
+      ScratchData *operator-> ()
+      {
+        return my_scratch_data;
+      }
+
+    private:
+      ScratchData *my_scratch_data;
+    };
+
+  private:
+    /**
+     * The actual data object that contains a scratch data for each thread.
+     */
+    static Threads::ThreadLocalStorage scratch_data;
+  };
+
+
+
   // function for block matrices: Find out where in the list of local dofs
   // (sorted according to global ids) the individual blocks start. Transform
   // the global indices to block-local indices in order to be able to use
@@ -2373,17 +2508,10 @@ ConstraintMatrix::distribute_local_to_global (
 
   const size_type n_local_dofs = local_dof_indices.size();
 
-  ScratchData &my_scratch_data = scratch_data.get();
-  Assert(my_scratch_data.in_use == false,
-         ExcMessage("Access to thread-local scratch data tried, but it is already "
-                    "in use"));
-  // TODO: might want to have a scoped variable for in_use here and in the
-  // methods below
-  my_scratch_data.in_use = true;
-
-  if (my_scratch_data.global_rows.get() == 0)
-    my_scratch_data.global_rows.reset(new internals::GlobalRowsFromLocal());
-  internals::GlobalRowsFromLocal &global_rows = *my_scratch_data.global_rows;
+  typename internals::ConstraintMatrixData::ScratchDataAccessor
+  scratch_data;
+
+  internals::GlobalRowsFromLocal &global_rows = scratch_data->global_rows;
   global_rows.reinit(n_local_dofs);
   make_sorted_row_list (local_dof_indices, global_rows);
 
@@ -2395,18 +2523,14 @@ ConstraintMatrix::distribute_local_to_global (
   // an array in any case since we cannot know about the actual data type in
   // the ConstraintMatrix class (unless we do cast). This involves a little
   // bit of logic to determine the type of the matrix value.
-  std::vector & cols = my_scratch_data.columns;
-  std::vector    & vals = my_scratch_data.values;
-  std::vector      values_non_double;
+  std::vector &cols = scratch_data->columns;
+  std::vector     &vals = scratch_data->values;
   SparseMatrix *sparse_matrix
     = dynamic_cast *>(&global_matrix);
   if (use_dealii_matrix == false)
     {
       cols.resize (n_actual_dofs);
-      if (types_are_equal::value == false)
-        values_non_double.resize(n_actual_dofs);
-      else
-        vals.resize (n_actual_dofs);
+      vals.resize (n_actual_dofs);
     }
   else
     Assert (sparse_matrix != 0, ExcInternalError());
@@ -2423,14 +2547,13 @@ ConstraintMatrix::distribute_local_to_global (
           size_type *col_ptr = &cols[0];
           // cast is uncritical here and only used to avoid compiler
           // warnings. We never access a non-double array
-          number *val_ptr = types_are_equal::value ?
-            reinterpret_cast(&vals[0]) : &values_non_double[0];
+          number *val_ptr = &vals[0];
           internals::resolve_matrix_row (global_rows, global_rows, i, 0,
                                          n_actual_dofs,
                                          local_matrix, col_ptr, val_ptr);
           const size_type n_values = col_ptr - &cols[0];
           if (n_values > 0)
-            global_matrix.add(row, n_values, &cols[0], val_ptr-n_values, false,
+            global_matrix.add(row, n_values, &cols[0], &vals[0], false,
                               true);
         }
       else
@@ -2457,7 +2580,6 @@ ConstraintMatrix::distribute_local_to_global (
   internals::set_matrix_diagonals (global_rows, local_dof_indices,
                                    local_matrix, *this,
                                    global_matrix, global_vector, use_inhomogeneities_for_rhs);
-  my_scratch_data.in_use = false;
 }
 
 
@@ -2479,19 +2601,11 @@ ConstraintMatrix::distribute_local_to_global (
   const size_type n_local_row_dofs = row_indices.size();
   const size_type n_local_col_dofs = col_indices.size();
 
-  ScratchData &my_scratch_data = scratch_data.get();
-  Assert(my_scratch_data.in_use == false,
-         ExcMessage("Access to thread-local scratch data tried, but it is already "
-                    "in use"));
-  my_scratch_data.in_use = true;
-
-  if (my_scratch_data.global_rows.get() == 0)
-    my_scratch_data.global_rows.reset(new internals::GlobalRowsFromLocal());
-  if (my_scratch_data.global_columns.get() == 0)
-    my_scratch_data.global_columns.reset(new internals::GlobalRowsFromLocal());
-  internals::GlobalRowsFromLocal &global_rows = *my_scratch_data.global_rows;
+  typename internals::ConstraintMatrixData::ScratchDataAccessor
+  scratch_data;
+  internals::GlobalRowsFromLocal &global_rows = scratch_data->global_rows;
   global_rows.reinit(n_local_row_dofs);
-  internals::GlobalRowsFromLocal &global_cols = *my_scratch_data.global_columns;
+  internals::GlobalRowsFromLocal &global_cols = scratch_data->global_columns;
   global_cols.reinit(n_local_col_dofs);
   make_sorted_row_list (row_indices, global_rows);
   make_sorted_row_list (col_indices, global_cols);
@@ -2501,14 +2615,10 @@ ConstraintMatrix::distribute_local_to_global (
 
   // create arrays for the column data (indices and values) that will then be
   // written into the matrix. Shortcut for deal.II sparse matrix
-  std::vector & cols = my_scratch_data.columns;
-  std::vector    & vals = my_scratch_data.values;
-  std::vector      values_non_double;
+  std::vector &cols = scratch_data->columns;
+  std::vector     &vals = scratch_data->values;
   cols.resize(n_actual_col_dofs);
-  if (types_are_equal::value == true)
-    vals.resize(n_actual_col_dofs);
-  else
-    values_non_double.resize(n_actual_col_dofs);
+  vals.resize(n_actual_col_dofs);
 
   // now do the actual job.
   for (size_type i=0; i::value ?
-        reinterpret_cast(&vals[0]) : &values_non_double[0];
+      number    *val_ptr = &vals[0];
       internals::resolve_matrix_row (global_rows, global_cols, i, 0,
                                      n_actual_col_dofs,
                                      local_matrix, col_ptr, val_ptr);
       const size_type n_values = col_ptr - &cols[0];
       if (n_values > 0)
-        global_matrix.add(row, n_values, &cols[0], val_ptr-n_values,
-                          false, true);
+        global_matrix.add(row, n_values, &cols[0], &vals[0], false, true);
     }
-
-  my_scratch_data.in_use = false;
 }
 
 
@@ -2563,22 +2669,17 @@ distribute_local_to_global (const FullMatrix     &local_matrix,
     }
   Assert (sorted == true, ExcMatrixNotClosed());
 
-  ScratchData &my_scratch_data = scratch_data.get();
-  Assert(my_scratch_data.in_use == false,
-         ExcMessage("Access to thread-local scratch data tried, but it is already "
-                    "in use"));
-  my_scratch_data.in_use = true;
+  typename internals::ConstraintMatrixData::ScratchDataAccessor
+  scratch_data;
 
   const size_type n_local_dofs = local_dof_indices.size();
-  if (my_scratch_data.global_rows.get() == 0)
-    my_scratch_data.global_rows.reset(new internals::GlobalRowsFromLocal());
-  internals::GlobalRowsFromLocal &global_rows = *my_scratch_data.global_rows;
+  internals::GlobalRowsFromLocal &global_rows = scratch_data->global_rows;
   global_rows.reinit(n_local_dofs);
 
   make_sorted_row_list (local_dof_indices, global_rows);
   const size_type n_actual_dofs = global_rows.size();
 
-  std::vector &global_indices = my_scratch_data.vector_indices;
+  std::vector &global_indices = scratch_data->vector_indices;
   if (use_vectors == true)
     {
       global_indices.resize(n_actual_dofs);
@@ -2588,20 +2689,16 @@ distribute_local_to_global (const FullMatrix     &local_matrix,
 
   // additional construct that also takes care of block indices.
   const size_type num_blocks   = global_matrix.n_block_rows();
-  std::vector &block_starts = my_scratch_data.block_starts;
+  std::vector &block_starts = scratch_data->block_starts;
   block_starts.resize(num_blocks+1);
   internals::make_block_starts (global_matrix, global_rows, block_starts);
 
-  std::vector & cols = my_scratch_data.columns;
-  std::vector    & vals = my_scratch_data.values;
-  std::vector      values_non_double;
+  std::vector &cols = scratch_data->columns;
+  std::vector     &vals = scratch_data->values;
   if (use_dealii_matrix == false)
     {
       cols.resize (n_actual_dofs);
-      if (types_are_equal::value == true)
-        vals.resize(n_actual_dofs);
-      else
-        values_non_double.resize(n_actual_dofs);
+      vals.resize (n_actual_dofs);
     }
 
   // the basic difference to the non-block variant from now onwards is that we
@@ -2621,16 +2718,14 @@ distribute_local_to_global (const FullMatrix     &local_matrix,
               if (use_dealii_matrix == false)
                 {
                   size_type *col_ptr = &cols[0];
-                  number *val_ptr = types_are_equal::value ?
-                    reinterpret_cast(&vals[0]) : &values_non_double[0];
+                  number *val_ptr = &vals[0];
                   internals::resolve_matrix_row (global_rows, global_rows, i,
                                                  start_block, end_block,
                                                  local_matrix, col_ptr, val_ptr);
                   const size_type n_values = col_ptr - &cols[0];
                   if (n_values > 0)
                     global_matrix.block(block, block_col).add(row, n_values,
-                                                              &cols[0],
-                                                              val_ptr-n_values,
+                                                              &cols[0], &vals[0],
                                                               false, true);
                 }
               else
@@ -2661,8 +2756,6 @@ distribute_local_to_global (const FullMatrix     &local_matrix,
   internals::set_matrix_diagonals (global_rows, local_dof_indices,
                                    local_matrix, *this,
                                    global_matrix, global_vector, use_inhomogeneities_for_rhs);
-
-  my_scratch_data.in_use = false;
 }
 
 
@@ -2686,11 +2779,7 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
       AssertDimension (dof_mask.n_cols(), n_local_dofs);
     }
 
-  ScratchData &my_scratch_data = scratch_data.get();
-  Assert(my_scratch_data.in_use == false,
-         ExcMessage("Access to thread-local scratch data tried, but it is already "
-                    "in use"));
-  my_scratch_data.in_use = true;
+  internals::ConstraintMatrixData::ScratchDataAccessor scratch_data;
 
   // if the dof mask is not active, all we have to do is to add some indices
   // in a matrix format. To do this, we first create an array of all the
@@ -2698,7 +2787,7 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
   // plus some indices that come from constraints.
   if (dof_mask_is_active == false)
     {
-      std::vector & actual_dof_indices = my_scratch_data.columns;
+      std::vector &actual_dof_indices = scratch_data->columns;
       actual_dof_indices.resize(n_local_dofs);
       make_sorted_row_list (local_dof_indices, actual_dof_indices);
       const size_type n_actual_dofs = actual_dof_indices.size();
@@ -2727,7 +2816,6 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
               sparsity_pattern.add (local_dof_indices[i], local_dof_indices[i]);
           }
 
-      my_scratch_data.in_use = false;
       return;
     }
 
@@ -2735,16 +2823,14 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
   // complicated case: we need to filter out some indices. then the function
   // gets similar to the function for distributing matrix entries, see there
   // for additional comments.
-  if (my_scratch_data.global_rows.get() == 0)
-    my_scratch_data.global_rows.reset(new internals::GlobalRowsFromLocal());
-  internals::GlobalRowsFromLocal &global_rows = *my_scratch_data.global_rows;
+  internals::GlobalRowsFromLocal &global_rows = scratch_data->global_rows;
   global_rows.reinit(n_local_dofs);
   make_sorted_row_list (local_dof_indices, global_rows);
   const size_type n_actual_dofs = global_rows.size();
 
   // create arrays for the column indices that will then be written into the
   // sparsity pattern.
-  std::vector & cols = my_scratch_data.columns;
+  std::vector &cols = scratch_data->columns;
   cols.resize(n_actual_dofs);
 
   for (size_type i=0; i &local_dof_indices,
   internals::set_sparsity_diagonals (global_rows, local_dof_indices,
                                      dof_mask, keep_constrained_entries,
                                      sparsity_pattern);
-  my_scratch_data.in_use = false;
 }
 
 
@@ -2846,11 +2931,7 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
   const size_type n_local_dofs = local_dof_indices.size();
   const size_type num_blocks = sparsity_pattern.n_block_rows();
 
-  ScratchData &my_scratch_data = scratch_data.get();
-  Assert(my_scratch_data.in_use == false,
-         ExcMessage("Access to thread-local scratch data tried, but it is already "
-                    "in use"));
-  my_scratch_data.in_use = true;
+  internals::ConstraintMatrixData::ScratchDataAccessor scratch_data;
 
   bool dof_mask_is_active = false;
   if (dof_mask.n_rows() == n_local_dofs)
@@ -2861,13 +2942,13 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
 
   if (dof_mask_is_active == false)
     {
-      std::vector & actual_dof_indices = my_scratch_data.columns;
+      std::vector &actual_dof_indices = scratch_data->columns;
       actual_dof_indices.resize(n_local_dofs);
       make_sorted_row_list (local_dof_indices, actual_dof_indices);
       const size_type n_actual_dofs = actual_dof_indices.size();
 
       // additional construct that also takes care of block indices.
-      std::vector &block_starts = my_scratch_data.block_starts;
+      std::vector &block_starts = scratch_data->block_starts;
       block_starts.resize(num_blocks+1);
       internals::make_block_starts (sparsity_pattern, actual_dof_indices,
                                     block_starts);
@@ -2908,25 +2989,22 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
               sparsity_pattern.add (local_dof_indices[i], local_dof_indices[i]);
           }
 
-      my_scratch_data.in_use = false;
       return;
     }
 
   // difficult case with dof_mask, similar to the distribute_local_to_global
   // function for block matrices
-  if (my_scratch_data.global_rows.get() == 0)
-    my_scratch_data.global_rows.reset(new internals::GlobalRowsFromLocal());
-  internals::GlobalRowsFromLocal &global_rows = *my_scratch_data.global_rows;
+  internals::GlobalRowsFromLocal &global_rows = scratch_data->global_rows;
   global_rows.reinit(n_local_dofs);
   make_sorted_row_list (local_dof_indices, global_rows);
   const size_type n_actual_dofs = global_rows.size();
 
   // additional construct that also takes care of block indices.
-  std::vector & block_starts = my_scratch_data.block_starts;
+  std::vector &block_starts = scratch_data->block_starts;
   block_starts.resize(num_blocks+1);
   internals::make_block_starts(sparsity_pattern, global_rows, block_starts);
 
-  std::vector &cols = my_scratch_data.columns;
+  std::vector &cols = scratch_data->columns;
   cols.resize(n_actual_dofs);
 
   // the basic difference to the non-block variant from now onwards is that we
@@ -2956,7 +3034,6 @@ add_entries_local_to_global (const std::vector &local_dof_indices,
   internals::set_sparsity_diagonals (global_rows, local_dof_indices,
                                      dof_mask, keep_constrained_entries,
                                      sparsity_pattern);
-  my_scratch_data.in_use = false;
 }
 
 
diff --git a/deal.II/include/deal.II/lac/full_matrix.h b/deal.II/include/deal.II/lac/full_matrix.h
index ad8a7fc09a..4be38349f0 100644
--- a/deal.II/include/deal.II/lac/full_matrix.h
+++ b/deal.II/include/deal.II/lac/full_matrix.h
@@ -61,17 +61,17 @@ template  class LAPACKFullMatrix;
  *
  * @author Guido Kanschat, Franz-Theo Suttmeier, Wolfgang Bangerth, 1993-2004
  */
-template
+template 
 class FullMatrix : public Table<2,number>
 {
 public:
   /**
-   * Declare type of container size.
+   * A type of used to index into this container. Because we can not
+   * expect to store matrices bigger than what can be indexed by a regular
+   * unsigned integer, unsigned int is completely sufficient
+   * as an index type.
    */
-  // A FullMatrix will never require to use unsigned long long int instead of
-  // unsigned but because a ConstraintMatrix can be a SparseMatrix or a
-  // FullMatrix, we need have the same interface.
-  typedef types::global_dof_index size_type;
+  typedef unsigned int size_type;
 
   /**
    * Type of matrix entries. In analogy to
@@ -324,7 +324,7 @@ public:
   /**
    * Variable assignment operator.
    */
-  template
+  template 
   FullMatrix &
   operator = (const FullMatrix &);
 
@@ -446,10 +446,10 @@ public:
    *      rows and columns in the current object. In other words,
    *      the current object is not resized for this operation.
    */
-  template 
+  template 
   void extract_submatrix_from (const MatrixType &matrix,
-                               const std::vector &row_index_set,
-                               const std::vector &column_index_set);
+                               const std::vector &row_index_set,
+                               const std::vector &column_index_set);
 
   /**
    * Copy the elements of the current matrix object into a specified
@@ -463,10 +463,10 @@ public:
    *      rows and columns in the current object. In other words,
    *      the current object is not resized for this operation.
    */
-  template 
+  template 
   void
-  scatter_matrix_to (const std::vector &row_index_set,
-                     const std::vector &column_index_set,
+  scatter_matrix_to (const std::vector &row_index_set,
+                     const std::vector &column_index_set,
                      MatrixType &matrix) const;
 
   /**
@@ -487,7 +487,7 @@ public:
    * determined either by the size
    * of this or src.
    */
-  template
+  template 
   void fill (const FullMatrix &src,
              const size_type dst_offset_i = 0,
              const size_type dst_offset_j = 0,
@@ -499,7 +499,7 @@ public:
    * Make function of base class
    * available.
    */
-  template
+  template 
   void fill (const number2 *);
 
   /**
@@ -520,7 +520,7 @@ public:
    * possible to duplicate rows or
    * columns by this method.
    */
-  template
+  template 
   void fill_permutation (const FullMatrix       &src,
                          const std::vector &p_rows,
                          const std::vector &p_cols);
@@ -612,7 +612,7 @@ public:
    * complex-valued, but not mixed, for
    * this function to make sense.
    */
-  template
+  template 
   number2 matrix_norm_square (const Vector &v) const;
 
   /**
@@ -630,7 +630,7 @@ public:
    * complex-valued, but not mixed, for
    * this function to make sense.
    */
-  template
+  template 
   number2 matrix_scalar_product (const Vector &u,
                                  const Vector &v) const;
 
@@ -825,7 +825,7 @@ public:
    * convertible to the data type
    * of this matrix.
    */
-  template
+  template 
   void add (const number               a,
             const FullMatrix &A);
 
@@ -842,7 +842,7 @@ public:
    * convertible to the data type
    * of this matrix.
    */
-  template
+  template 
   void add (const number               a,
             const FullMatrix &A,
             const number               b,
@@ -861,7 +861,7 @@ public:
    * is convertible to the data
    * type of this matrix.
    */
-  template
+  template 
   void add (const number               a,
             const FullMatrix &A,
             const number               b,
@@ -886,7 +886,7 @@ public:
    * size of this or src
    * and the given offsets.
    */
-  template
+  template 
   void add (const FullMatrix &src,
             const number factor,
             const size_type dst_offset_i = 0,
@@ -901,7 +901,7 @@ public:
    *
    * A += s BT
    */
-  template
+  template 
   void Tadd (const number               s,
              const FullMatrix &B);
 
@@ -927,7 +927,7 @@ public:
    * of this or
    * src.
    */
-  template
+  template 
   void Tadd (const FullMatrix &src,
              const number               factor,
              const size_type dst_offset_i = 0,
@@ -959,10 +959,10 @@ public:
    * and are indeed ignored in the
    * implementation.
    */
-  template 
+  template 
   void add (const size_type     row,
-            const size_type     n_cols,
-            const size_type    *col_indices,
+            const unsigned int  n_cols,
+            const index_type   *col_indices,
             const number2      *values,
             const bool          elide_zero_values = true,
             const bool          col_indices_are_sorted = false);
@@ -1030,7 +1030,7 @@ public:
    * Assignment *this =
    * a*A.
    */
-  template
+  template 
   void equ (const number               a,
             const FullMatrix &A);
 
@@ -1038,7 +1038,7 @@ public:
    * Assignment *this = a*A +
    * b*B.
    */
-  template
+  template 
   void equ (const number               a,
             const FullMatrix &A,
             const number               b,
@@ -1048,7 +1048,7 @@ public:
    * Assignment *this = a*A +
    * b*B + c*C.
    */
-  template
+  template 
   void equ (const number               a,
             const FullMatrix &A,
             const number               b,
@@ -1176,7 +1176,7 @@ public:
    * usually results in considerable
    * performance gains.
    */
-  template
+  template 
   void mmult (FullMatrix       &C,
               const FullMatrix &B,
               const bool                 adding=false) const;
@@ -1208,7 +1208,7 @@ public:
    * BLAS usually results in considerable
    * performance gains.
    */
-  template
+  template 
   void Tmmult (FullMatrix       &C,
                const FullMatrix &B,
                const bool                 adding=false) const;
@@ -1240,7 +1240,7 @@ public:
    * usually results in considerable
    * performance gains.
    */
-  template
+  template 
   void mTmult (FullMatrix       &C,
                const FullMatrix &B,
                const bool                 adding=false) const;
@@ -1273,7 +1273,7 @@ public:
    * BLAS usually results in considerable
    * performance gains.
    */
-  template
+  template 
   void TmTmult (FullMatrix       &C,
                 const FullMatrix &B,
                 const bool                 adding=false) const;
@@ -1322,7 +1322,7 @@ public:
    * Source and destination must
    * not be the same vector.
    */
-  template
+  template 
   void vmult (Vector       &w,
               const Vector &v,
               const bool             adding=false) const;
@@ -1334,7 +1334,7 @@ public:
    * Source and destination must
    * not be the same vector.
    */
-  template
+  template 
   void vmult_add (Vector       &w,
                   const Vector &v) const;
 
@@ -1357,7 +1357,7 @@ public:
    * Source and destination must
    * not be the same vector.
    */
-  template
+  template 
   void Tvmult (Vector       &w,
                const Vector &v,
                const bool             adding=false) const;
@@ -1370,7 +1370,7 @@ public:
    * Source and destination must
    * not be the same vector.
    */
-  template
+  template 
   void Tvmult_add (Vector       &w,
                    const Vector &v) const;
 
@@ -1398,7 +1398,7 @@ public:
    * dst must not be the same
    * vector.
    */
-  template
+  template 
   number residual (Vector       &dst,
                    const Vector &x,
                    const Vector &b) const;
@@ -1420,7 +1420,7 @@ public:
    * same object for @p dst and @p
    * src.
    */
-  template
+  template 
   void forward (Vector       &dst,
                 const Vector &src) const;
 
@@ -1434,7 +1434,7 @@ public:
    * same object for @p dst and @p
    * src.
    */
-  template
+  template 
   void backward (Vector       &dst,
                  const Vector &src) const;
 
@@ -1572,12 +1572,12 @@ FullMatrix::copy_transposed (const MATRIX &M)
 
 
 template 
-template 
+template 
 inline
 void
 FullMatrix::extract_submatrix_from (const MatrixType &matrix,
-                                            const std::vector &row_index_set,
-                                            const std::vector &column_index_set)
+                                            const std::vector &row_index_set,
+                                            const std::vector &column_index_set)
 {
   AssertDimension(row_index_set.size(), this->n_rows());
   AssertDimension(column_index_set.size(), this->n_cols());
@@ -1593,11 +1593,11 @@ FullMatrix::extract_submatrix_from (const MatrixType &matrix,
 
 
 template 
-template 
+template 
 inline
 void
-FullMatrix::scatter_matrix_to (const std::vector &row_index_set,
-                                       const std::vector &column_index_set,
+FullMatrix::scatter_matrix_to (const std::vector &row_index_set,
+                                       const std::vector &column_index_set,
                                        MatrixType &matrix) const
 {
   AssertDimension(row_index_set.size(), this->n_rows());
@@ -1627,7 +1627,7 @@ FullMatrix::set (const size_type i,
 
 
 template 
-template
+template 
 void
 FullMatrix::vmult_add (Vector       &w,
                                const Vector &v) const
@@ -1637,7 +1637,7 @@ FullMatrix::vmult_add (Vector       &w,
 
 
 template 
-template
+template 
 void
 FullMatrix::Tvmult_add (Vector       &w,
                                 const Vector &v) const
@@ -1833,13 +1833,13 @@ FullMatrix::add (const size_type r, const size_type c, const number v)
 
 
 template 
-template 
+template 
 inline
 void
-FullMatrix::add (const size_type  row,
-                         const size_type  n_cols,
-                         const size_type *col_indices,
-                         const number2   *values,
+FullMatrix::add (const size_type    row,
+                         const unsigned int n_cols,
+                         const index_type  *col_indices,
+                         const number2     *values,
                          const bool,
                          const bool)
 {
diff --git a/deal.II/include/deal.II/lac/lapack_templates.h b/deal.II/include/deal.II/lac/lapack_templates.h
index ddebfbdf03..676fa71406 100644
--- a/deal.II/include/deal.II/lac/lapack_templates.h
+++ b/deal.II/include/deal.II/lac/lapack_templates.h
@@ -720,7 +720,7 @@ syev (const char *, const char *, const int *, number1 *, const int *, number2 *
 inline void
 syev (const char *jobz, const char *uplo, const int *n, double *A, const int *lda, double *w, double *work, const int *lwork, int *info)
 {
-  dsyev_ (char*jobz,char*uplo,int*n,double*A,int*lda,double*w,double*work,int*lwork,int*info);
+  dsyev_ (char *jobz,char *uplo,int *n,double*A,int *lda,double*w,double*work,int *lwork,int *info);
 }
 #else
 inline void
@@ -735,7 +735,7 @@ syev (const char *, const char *, const int *, double *, const int *, double *,
 inline void
 syev (const char *jobz, const char *uplo, const int *n, float *A, const int *lda, float *w, float *work, const int *lwork, int *info)
 {
-  ssyev_ (char*jobz,char*uplo,int*n,double*A,int*lda,double*w,double*work,int*lwork,int*info);
+  ssyev_ (char *jobz,char *uplo,int *n,double*A,int *lda,double*w,double*work,int *lwork,int *info);
 }
 #else
 inline void
diff --git a/deal.II/include/deal.II/lac/matrix_out.h b/deal.II/include/deal.II/lac/matrix_out.h
index 55c8452217..dbb94464f7 100644
--- a/deal.II/include/deal.II/lac/matrix_out.h
+++ b/deal.II/include/deal.II/lac/matrix_out.h
@@ -134,7 +134,7 @@ public:
      *
      * Default value is one.
      */
-    size_type block_size;
+    unsigned int block_size;
 
     /**
      * If true, plot
@@ -149,9 +149,9 @@ public:
      * structure to their default
      * values.
      */
-    Options (const bool      show_absolute_values = false,
-             const size_type block_size           = 1,
-             const bool      discontinuous        = false);
+    Options (const bool         show_absolute_values = false,
+             const unsigned int block_size           = 1,
+             const bool         discontinuous        = false);
   };
 
   /**
@@ -365,16 +365,18 @@ MatrixOut::get_gridpoint_value (const Matrix   &matrix,
         return std::fabs(get_element (matrix, i, j));
       else
         return get_element (matrix, i, j);
-    };
+    }
 
   // if blocksize greater than one,
   // then compute average of elements
   double average = 0;
   size_type n_elements = 0;
   for (size_type row=i*options.block_size;
-       row < std::min(matrix.m(), (i+1)*options.block_size); ++row)
+       row < std::min(size_type(matrix.m()),
+                      size_type((i+1)*options.block_size)); ++row)
     for (size_type col=j*options.block_size;
-         col < std::min(matrix.m(), (j+1)*options.block_size); ++col, ++n_elements)
+         col < std::min(size_type(matrix.m()),
+                        size_type((j+1)*options.block_size)); ++col, ++n_elements)
       if (options.show_absolute_values == true)
         average += std::fabs(get_element (matrix, row, col));
       else
diff --git a/deal.II/include/deal.II/lac/parallel_block_vector.h b/deal.II/include/deal.II/lac/parallel_block_vector.h
index 7e3f1bb190..7ffd04094b 100644
--- a/deal.II/include/deal.II/lac/parallel_block_vector.h
+++ b/deal.II/include/deal.II/lac/parallel_block_vector.h
@@ -423,8 +423,8 @@ namespace parallel
     template 
     inline
     BlockVector::BlockVector (const std::vector &local_ranges,
-        const std::vector &ghost_indices,
-        const MPI_Comm  communicator)
+                                      const std::vector &ghost_indices,
+                                      const MPI_Comm  communicator)
     {
       std::vector sizes(local_ranges.size());
       for (unsigned int i=0; i
     inline
     BlockVector::BlockVector (const std::vector &local_ranges,
-        const MPI_Comm  communicator)
+                                      const MPI_Comm  communicator)
     {
       std::vector sizes(local_ranges.size());
       for (unsigned int i=0; in_blocks() == 0 || this->n_blocks() == v.n_blocks(),
-                    ExcDimensionMismatch(this->n_blocks(), v.n_blocks()));
+              ExcDimensionMismatch(this->n_blocks(), v.n_blocks()));
 
       if (this->n_blocks() != v.n_blocks())
         reinit(v.n_blocks(), true);
diff --git a/deal.II/include/deal.II/lac/parallel_vector.h b/deal.II/include/deal.II/lac/parallel_vector.h
index fbc865798f..db837d4983 100644
--- a/deal.II/include/deal.II/lac/parallel_vector.h
+++ b/deal.II/include/deal.II/lac/parallel_vector.h
@@ -1750,7 +1750,8 @@ namespace parallel
       while (indices_begin != indices_end)
         {
           *values_begin = operator()(*indices_begin);
-          indices_begin++; values_begin++;
+          indices_begin++;
+          values_begin++;
         }
     }
 
diff --git a/deal.II/include/deal.II/lac/petsc_matrix_base.h b/deal.II/include/deal.II/lac/petsc_matrix_base.h
index 8d0ed3179e..870647f24c 100644
--- a/deal.II/include/deal.II/lac/petsc_matrix_base.h
+++ b/deal.II/include/deal.II/lac/petsc_matrix_base.h
@@ -84,10 +84,10 @@ namespace PETScWrappers
                   const size_type   row,
                   const size_type   index);
 
-	/**
-	 * Copy constructor.
-	 */
-	Accessor (const Accessor &a);
+        /**
+         * Copy constructor.
+         */
+        Accessor (const Accessor &a);
 
         /**
          * Row number of the element
diff --git a/deal.II/include/deal.II/lac/petsc_parallel_block_vector.h b/deal.II/include/deal.II/lac/petsc_parallel_block_vector.h
index f55e61d5cc..1928d507df 100644
--- a/deal.II/include/deal.II/lac/petsc_parallel_block_vector.h
+++ b/deal.II/include/deal.II/lac/petsc_parallel_block_vector.h
@@ -448,7 +448,7 @@ namespace PETScWrappers
       // we only allow assignment to vectors with the same number of blocks
       // or to an empty BlockVector
       Assert (n_blocks() == 0 || n_blocks() == v.n_blocks(),
-          ExcDimensionMismatch(n_blocks(), v.n_blocks()));
+              ExcDimensionMismatch(n_blocks(), v.n_blocks()));
 
       if (this->n_blocks() != v.n_blocks())
         reinit(v.n_blocks());
diff --git a/deal.II/include/deal.II/lac/petsc_vector_base.h b/deal.II/include/deal.II/lac/petsc_vector_base.h
index 709e1e6185..6de405f3b5 100644
--- a/deal.II/include/deal.II/lac/petsc_vector_base.h
+++ b/deal.II/include/deal.II/lac/petsc_vector_base.h
@@ -1264,97 +1264,99 @@ namespace PETScWrappers
     // if we are dealing
     // with a parallel vector
     if (ghosted )
-    {
-
-      int ierr;
-
-      // there is the possibility
-      // that the vector has
-      // ghost elements. in that
-      // case, we first need to
-      // figure out which
-      // elements we own locally,
-      // then get a pointer to
-      // the elements that are
-      // stored here (both the
-      // ones we own as well as
-      // the ghost elements). in
-      // this array, the locally
-      // owned elements come
-      // first followed by the
-      // ghost elements whose
-      // position we can get from
-      // an index set
-      PetscInt begin, end, i;
-      ierr = VecGetOwnershipRange (vector, &begin, &end);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
+      {
+
+        int ierr;
+
+        // there is the possibility
+        // that the vector has
+        // ghost elements. in that
+        // case, we first need to
+        // figure out which
+        // elements we own locally,
+        // then get a pointer to
+        // the elements that are
+        // stored here (both the
+        // ones we own as well as
+        // the ghost elements). in
+        // this array, the locally
+        // owned elements come
+        // first followed by the
+        // ghost elements whose
+        // position we can get from
+        // an index set
+        PetscInt begin, end, i;
+        ierr = VecGetOwnershipRange (vector, &begin, &end);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
+
+        Vec locally_stored_elements = PETSC_NULL;
+        ierr = VecGhostGetLocalForm(vector, &locally_stored_elements);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
+
+        PetscInt lsize;
+        ierr = VecGetSize(locally_stored_elements, &lsize);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
+
+        PetscScalar *ptr;
+        ierr = VecGetArray(locally_stored_elements, &ptr);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
+
+        for (i = 0; i < n_idx; i++)
+          {
+            const unsigned int index = *(indices_begin+i);
+            if ( index>=static_cast(begin)
+                 && index(end) )
+              {
+                //local entry
+                *(values_begin+i) = *(ptr+index-begin);
+              }
+            else
+              {
+                //ghost entry
+                const unsigned int ghostidx
+                  = ghost_indices.index_within_set(index);
+
+                Assert(ghostidx+end-begin<(unsigned int)lsize, ExcInternalError());
+                *(values_begin+i) = *(ptr+ghostidx+end-begin);
+              }
+          }
+
+        ierr = VecRestoreArray(locally_stored_elements, &ptr);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
+
+        ierr = VecGhostRestoreLocalForm(vector, &locally_stored_elements);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
 
-      Vec locally_stored_elements = PETSC_NULL;
-      ierr = VecGhostGetLocalForm(vector, &locally_stored_elements);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
-
-      PetscInt lsize;
-      ierr = VecGetSize(locally_stored_elements, &lsize);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
-
-      PetscScalar *ptr;
-      ierr = VecGetArray(locally_stored_elements, &ptr);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
-
-      for (i = 0; i < n_idx; i++) {
-        const unsigned int index = *(indices_begin+i);
-        if ( index>=static_cast(begin)
-            && index(end) )
-        {
-          //local entry
-          *(values_begin+i) = *(ptr+index-begin);
-        }
-        else
-        {
-          //ghost entry
-          const unsigned int ghostidx
-          = ghost_indices.index_within_set(index);
-
-          Assert(ghostidx+end-begin<(unsigned int)lsize, ExcInternalError());
-          *(values_begin+i) = *(ptr+ghostidx+end-begin);
-        }
       }
-
-      ierr = VecRestoreArray(locally_stored_elements, &ptr);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
-
-      ierr = VecGhostRestoreLocalForm(vector, &locally_stored_elements);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
-
-    }
     // if the vector is local or the
     // caller, then simply access the
     // element we are interested in
     else
-    {
-      int ierr;
+      {
+        int ierr;
 
-      PetscInt begin, end;
-      ierr = VecGetOwnershipRange (vector, &begin, &end);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
+        PetscInt begin, end;
+        ierr = VecGetOwnershipRange (vector, &begin, &end);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
 
-      PetscScalar *ptr;
-      ierr = VecGetArray(vector, &ptr);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
+        PetscScalar *ptr;
+        ierr = VecGetArray(vector, &ptr);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
 
-      for (PetscInt i = 0; i < n_idx; i++) {
-        const unsigned int index = *(indices_begin+i);
+        for (PetscInt i = 0; i < n_idx; i++)
+          {
+            const unsigned int index = *(indices_begin+i);
 
-        Assert(index>=static_cast(begin)
-            && index(end), ExcInternalError());
+            Assert(index>=static_cast(begin)
+                   && index(end), ExcInternalError());
 
-        *(values_begin+i) = *(ptr+index-begin);
-      }
+            *(values_begin+i) = *(ptr+index-begin);
+          }
 
-      ierr = VecRestoreArray(vector, &ptr);
-      AssertThrow (ierr == 0, ExcPETScError(ierr));
+        ierr = VecRestoreArray(vector, &ptr);
+        AssertThrow (ierr == 0, ExcPETScError(ierr));
 
-    }
+      }
   }
 
 #endif // DOXYGEN
diff --git a/deal.II/include/deal.II/lac/precondition_selector.h b/deal.II/include/deal.II/lac/precondition_selector.h
index 21fe9100bc..5ff4284df7 100644
--- a/deal.II/include/deal.II/lac/precondition_selector.h
+++ b/deal.II/include/deal.II/lac/precondition_selector.h
@@ -94,7 +94,7 @@ template  class SparseMatrix;
  * @author Ralf Hartmann, 1999
  */
 template ,
-         class VECTOR = dealii::Vector >
+          class VECTOR = dealii::Vector >
 class PreconditionSelector : public Subscriptor
 {
 public:
diff --git a/deal.II/include/deal.II/lac/slepc_solver.h b/deal.II/include/deal.II/lac/slepc_solver.h
index eefb46b6a9..ad8f01b418 100644
--- a/deal.II/include/deal.II/lac/slepc_solver.h
+++ b/deal.II/include/deal.II/lac/slepc_solver.h
@@ -300,7 +300,7 @@ namespace SLEPcWrappers
      */
     void
     solve (const unsigned int n_eigenpairs,
-	   unsigned int *n_converged);
+           unsigned int *n_converged);
 
     /**
      * Access the real parts of solutions for a solved eigenvector
diff --git a/deal.II/include/deal.II/lac/sparse_direct.h b/deal.II/include/deal.II/lac/sparse_direct.h
index c9df62991f..4075ba85dd 100644
--- a/deal.II/include/deal.II/lac/sparse_direct.h
+++ b/deal.II/include/deal.II/lac/sparse_direct.h
@@ -93,7 +93,7 @@ public:
    * preconditioners.
    */
   class AdditionalData
-{};
+  {};
 
 
   /**
diff --git a/deal.II/include/deal.II/lac/sparse_matrix.h b/deal.II/include/deal.II/lac/sparse_matrix.h
index 6f9519ae27..4efc719c17 100644
--- a/deal.II/include/deal.II/lac/sparse_matrix.h
+++ b/deal.II/include/deal.II/lac/sparse_matrix.h
@@ -1497,8 +1497,8 @@ public:
    */
   template 
   void print (STREAM &out,
-	      const bool across = false,
-	      const bool diagonal_first = true) const;
+              const bool across = false,
+              const bool diagonal_first = true) const;
 
   /**
    * Print the matrix in the usual format, i.e. as a matrix and not as a list
@@ -2477,8 +2477,8 @@ template 
 template 
 inline
 void SparseMatrix::print (STREAM &out,
-				  const bool across,
-				  const bool diagonal_first) const
+                                  const bool across,
+                                  const bool diagonal_first) const
 {
   Assert (cols != 0, ExcNotInitialized());
   Assert (val != 0, ExcNotInitialized());
diff --git a/deal.II/include/deal.II/lac/sparse_matrix.templates.h b/deal.II/include/deal.II/lac/sparse_matrix.templates.h
index e007579833..01caf34c11 100644
--- a/deal.II/include/deal.II/lac/sparse_matrix.templates.h
+++ b/deal.II/include/deal.II/lac/sparse_matrix.templates.h
@@ -437,8 +437,8 @@ namespace internal
      * task scheduler.
      */
     template 
+              typename InVector,
+              typename OutVector>
     void vmult_on_subrange (const size_type    begin_row,
                             const size_type    end_row,
                             const number      *values,
@@ -823,7 +823,7 @@ namespace internal
      * scheduler.
      */
     template 
+              typename InVector>
     number matrix_norm_sqr_on_subrange (const size_type    begin_row,
                                         const size_type    end_row,
                                         const number      *values,
@@ -886,7 +886,7 @@ namespace internal
      * scheduler.
      */
     template 
+              typename InVector>
     number matrix_scalar_product_on_subrange (const size_type    begin_row,
                                               const size_type    end_row,
                                               const number      *values,
@@ -1269,8 +1269,8 @@ namespace internal
      * scheduler.
      */
     template 
+              typename InVector,
+              typename OutVector>
     number residual_sqr_on_subrange (const size_type    begin_row,
                                      const size_type    end_row,
                                      const number      *values,
diff --git a/deal.II/include/deal.II/lac/sparsity_tools.h b/deal.II/include/deal.II/lac/sparsity_tools.h
index 860eedfae7..89b4a2b21c 100644
--- a/deal.II/include/deal.II/lac/sparsity_tools.h
+++ b/deal.II/include/deal.II/lac/sparsity_tools.h
@@ -256,7 +256,7 @@ namespace SparsityTools
   /**
    * similar to the function above, but includes support for
    * BlockCompressedSimpleSparsityPattern.
-   * @p owned_set_per_cpu is typically DoFHandler::locally_owned_dofs_per_processor 
+   * @p owned_set_per_cpu is typically DoFHandler::locally_owned_dofs_per_processor
    * and @p myrange are locally_relevant_dofs.
    */
   template 
diff --git a/deal.II/include/deal.II/lac/trilinos_parallel_block_vector.h b/deal.II/include/deal.II/lac/trilinos_parallel_block_vector.h
index d684216bdc..7ec7a0e217 100644
--- a/deal.II/include/deal.II/lac/trilinos_parallel_block_vector.h
+++ b/deal.II/include/deal.II/lac/trilinos_parallel_block_vector.h
@@ -133,8 +133,8 @@ namespace TrilinosWrappers
        * be ignored.
        */
       BlockVector (const std::vector ¶llel_partitioning,
-          const std::vector &ghost_values,
-          const MPI_Comm              &communicator);
+                   const std::vector &ghost_values,
+                   const MPI_Comm              &communicator);
 
 
       /**
@@ -448,8 +448,8 @@ namespace TrilinosWrappers
 
     inline
     BlockVector::BlockVector (const std::vector ¶llel_partitioning,
-        const std::vector &ghost_values,
-        const MPI_Comm              &communicator)
+                              const std::vector &ghost_values,
+                              const MPI_Comm              &communicator)
     {
       reinit(parallel_partitioning, ghost_values, communicator);
     }
diff --git a/deal.II/include/deal.II/lac/trilinos_precondition.h b/deal.II/include/deal.II/lac/trilinos_precondition.h
index 3936118d50..0f295a99e1 100644
--- a/deal.II/include/deal.II/lac/trilinos_precondition.h
+++ b/deal.II/include/deal.II/lac/trilinos_precondition.h
@@ -128,7 +128,7 @@ namespace TrilinosWrappers
      * Apply the transpose preconditioner.
      */
     virtual void Tvmult (VectorBase       &dst,
-                 const VectorBase &src) const;
+                         const VectorBase &src) const;
 
     /**
      * Apply the preconditioner on
@@ -1532,15 +1532,15 @@ namespace TrilinosWrappers
                  const VectorBase &src) const;
 
     /**
-     * Apply the preconditioner on deal.II data structures 
-     * instead of the ones provided in the Trilinos wrapper class, 
+     * Apply the preconditioner on deal.II data structures
+     * instead of the ones provided in the Trilinos wrapper class,
      * i.e., dst = src.
      */
     void vmult (dealii::Vector       &dst,
                 const dealii::Vector &src) const;
 
     /**
-     * Apply the transpose preconditioner on deal.II data structures 
+     * Apply the transpose preconditioner on deal.II data structures
      * instead of the ones provided in the Trilinos wrapper class,
      * i.e. dst = src.
      */
@@ -1548,7 +1548,7 @@ namespace TrilinosWrappers
                  const dealii::Vector &src) const;
 
     /**
-     * Apply the preconditioner on deal.II parallel data structures 
+     * Apply the preconditioner on deal.II parallel data structures
      * instead of the ones provided in the Trilinos wrapper class,
      * i.e., dst = src.
      */
@@ -1556,7 +1556,7 @@ namespace TrilinosWrappers
                 const dealii::parallel::distributed::Vector &src) const;
 
     /**
-     * Apply the transpose preconditioner on deal.II parallel data structures 
+     * Apply the transpose preconditioner on deal.II parallel data structures
      * instead of the ones provided in the Trilinos wrapper class,
      * i.e., dst = src.
      */
diff --git a/deal.II/include/deal.II/lac/trilinos_vector_base.h b/deal.II/include/deal.II/lac/trilinos_vector_base.h
index d5e2217ad9..848a8ad556 100644
--- a/deal.II/include/deal.II/lac/trilinos_vector_base.h
+++ b/deal.II/include/deal.II/lac/trilinos_vector_base.h
@@ -1361,10 +1361,12 @@ namespace TrilinosWrappers
                                          const ForwardIterator    indices_end,
                                          OutputIterator           values_begin) const
   {
-    while (indices_begin != indices_end) {
-      *values_begin = operator()(*indices_begin);
-      indices_begin++; values_begin++;
-    }
+    while (indices_begin != indices_end)
+      {
+        *values_begin = operator()(*indices_begin);
+        indices_begin++;
+        values_begin++;
+      }
   }
 
 
@@ -2215,7 +2217,7 @@ namespace TrilinosWrappers
   }
 
 
-  
+
 #endif // DOXYGEN
 
 }
diff --git a/deal.II/include/deal.II/lac/vector.h b/deal.II/include/deal.II/lac/vector.h
index fb67e97c1c..8dae825247 100644
--- a/deal.II/include/deal.II/lac/vector.h
+++ b/deal.II/include/deal.II/lac/vector.h
@@ -1366,10 +1366,12 @@ void Vector::extract_subvector_to (ForwardIterator          indices_begi
                                            const ForwardIterator    indices_end,
                                            OutputIterator           values_begin) const
 {
-  while (indices_begin != indices_end) {
-    *values_begin = operator()(*indices_begin);
-    indices_begin++; values_begin++;
-  }
+  while (indices_begin != indices_end)
+    {
+      *values_begin = operator()(*indices_begin);
+      indices_begin++;
+      values_begin++;
+    }
 }
 
 
diff --git a/deal.II/include/deal.II/lac/vector.templates.h b/deal.II/include/deal.II/lac/vector.templates.h
index 2d729e578e..91529fd539 100644
--- a/deal.II/include/deal.II/lac/vector.templates.h
+++ b/deal.II/include/deal.II/lac/vector.templates.h
@@ -580,7 +580,7 @@ namespace internal
     // The code returns the result as the last argument in order to make
     // spawning tasks simpler and use automatic template deduction.
     template 
+              typename ResultType, typename size_type>
     void accumulate (const Operation   &op,
                      const Number      *X,
                      const Number2     *Y,
@@ -635,18 +635,18 @@ namespace internal
                   r2 = op(X, Y, power);
                   for (size_type j=1; j<8; ++j)
                     r2 += op(X, Y, power);
-                  // no break
+                // no break
                 case 2:
                   r1 = op(X, Y, power);
                   for (size_type j=1; j<8; ++j)
                     r1 += op(X, Y, power);
                   r1 += r2;
-                  // no break
+                // no break
                 case 1:
                   r2 = op(X, Y, power);
                   for (size_type j=1; j<8; ++j)
                     r2 += op(X, Y, power);
-                  // no break
+                // no break
                 default:
                   for (size_type j=0; jlocal_range().second-
                                         vector_partitioner->local_range().first)
-                                        + vector_partitioner->ghost_indices().n_elements();
+                                       + vector_partitioner->ghost_indices().n_elements();
       for (std::size_t i=0; i 0)
                   {
@@ -1970,7 +1970,7 @@ not_connect:
         {
           out << "Entries row " << row << ": ";
           const unsigned int *glob_indices = begin_indices(row),
-            *end_row = end_indices(row);
+                              *end_row = end_indices(row);
           unsigned int index = 0;
           const std::pair
           *con_it = begin_indicators(row),
diff --git a/deal.II/include/deal.II/matrix_free/fe_evaluation.h b/deal.II/include/deal.II/matrix_free/fe_evaluation.h
index 6e75bc5331..425175f5af 100644
--- a/deal.II/include/deal.II/matrix_free/fe_evaluation.h
+++ b/deal.II/include/deal.II/matrix_free/fe_evaluation.h
@@ -83,7 +83,7 @@ namespace internal
  * @author Katharina Kormann and Martin Kronbichler, 2010, 2011
  */
 template 
+          int n_components_, typename Number>
 class FEEvaluationBase
 {
 public:
@@ -745,7 +745,7 @@ protected:
  * @author Katharina Kormann and Martin Kronbichler, 2010, 2011
  */
 template 
+          int n_components_, typename Number>
 class FEEvaluationAccess :
   public FEEvaluationBase
 {
@@ -1070,7 +1070,7 @@ protected:
  * @author Katharina Kormann and Martin Kronbichler, 2010, 2011
  */
 template 
+          int n_components_ = 1, typename Number = double >
 class FEEvaluationGeneral :
   public FEEvaluationAccess::value,
@@ -1219,7 +1219,7 @@ protected:
  * @author Katharina Kormann and Martin Kronbichler, 2010, 2011
  */
 template 
+          int n_components_ = 1, typename Number = double >
 class FEEvaluation :
   public FEEvaluationGeneral
 {
@@ -1427,7 +1427,7 @@ protected:
 /*----------------------- FEEvaluationBase ----------------------------------*/
 
 template 
+          int n_components_, typename Number>
 inline
 FEEvaluationBase
 ::FEEvaluationBase (const MatrixFree &data_in,
@@ -1478,7 +1478,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 void
 FEEvaluationBase
@@ -1553,7 +1553,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 unsigned int
 FEEvaluationBase
@@ -1566,7 +1566,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 internal::MatrixFreeFunctions::CellType
 FEEvaluationBase
@@ -1664,7 +1664,7 @@ namespace internal
                       VectorType         &vec,
                       Number             &res) const
     {
-      res = vector_access (const_cast(vec), index);
+      res = vector_access (const_cast(vec), index);
     }
 
     void pre_constraints (const Number &,
@@ -1679,7 +1679,7 @@ namespace internal
                              VectorType        &vec,
                              Number            &res) const
     {
-      res += weight * vector_access (const_cast(vec), index);
+      res += weight * vector_access (const_cast(vec), index);
     }
 
     void post_constraints (const Number &sum,
@@ -1803,7 +1803,7 @@ namespace internal
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2124,7 +2124,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2149,7 +2149,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2178,7 +2178,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2207,7 +2207,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2227,7 +2227,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2249,7 +2249,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2271,7 +2271,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2295,7 +2295,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2322,7 +2322,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2349,7 +2349,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2373,7 +2373,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2401,7 +2401,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2429,7 +2429,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 template
 inline
 void
@@ -2549,7 +2549,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components, typename Number>
 inline
 const VectorizedArray *
 FEEvaluationBase::
@@ -2561,7 +2561,7 @@ begin_dof_values () const
 
 
 template 
+          int n_components, typename Number>
 inline
 VectorizedArray *
 FEEvaluationBase::
@@ -2576,7 +2576,7 @@ begin_dof_values ()
 
 
 template 
+          int n_components, typename Number>
 inline
 const VectorizedArray *
 FEEvaluationBase::
@@ -2590,7 +2590,7 @@ begin_values () const
 
 
 template 
+          int n_components, typename Number>
 inline
 VectorizedArray *
 FEEvaluationBase::
@@ -2605,7 +2605,7 @@ begin_values ()
 
 
 template 
+          int n_components, typename Number>
 inline
 const VectorizedArray *
 FEEvaluationBase::
@@ -2619,7 +2619,7 @@ begin_gradients () const
 
 
 template 
+          int n_components, typename Number>
 inline
 VectorizedArray *
 FEEvaluationBase::
@@ -2634,7 +2634,7 @@ begin_gradients ()
 
 
 template 
+          int n_components, typename Number>
 inline
 const VectorizedArray *
 FEEvaluationBase::
@@ -2647,7 +2647,7 @@ begin_hessians () const
 
 
 template 
+          int n_components, typename Number>
 inline
 VectorizedArray *
 FEEvaluationBase::
@@ -2659,7 +2659,7 @@ begin_hessians ()
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,VectorizedArray >
 FEEvaluationBase
@@ -2675,7 +2675,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,VectorizedArray >
 FEEvaluationBase
@@ -2693,7 +2693,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,Tensor<1,dim,VectorizedArray > >
 FEEvaluationBase
@@ -2782,7 +2782,7 @@ namespace internal
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,Tensor<2,dim,VectorizedArray > >
 FEEvaluationBase
@@ -2908,7 +2908,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,Tensor<1,dim,VectorizedArray > >
 FEEvaluationBase
@@ -2987,7 +2987,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,VectorizedArray >
 FEEvaluationBase
@@ -3011,7 +3011,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 void
 FEEvaluationBase
@@ -3029,7 +3029,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 void
 FEEvaluationBase
@@ -3058,7 +3058,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 void
 FEEvaluationBase
@@ -3101,7 +3101,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 Tensor<1,n_components_,VectorizedArray >
 FEEvaluationBase
@@ -3127,7 +3127,7 @@ FEEvaluationBase
 
 
 template 
+          int n_components_, typename Number>
 inline
 FEEvaluationAccess
 ::FEEvaluationAccess (const MatrixFree &data_in,
@@ -3537,7 +3537,7 @@ FEEvaluationAccess
 }
 
 template 
+          typename Number>
 inline
 void
 FEEvaluationAccess
@@ -3685,7 +3685,7 @@ FEEvaluationAccess
 /*----------------------- FEEvaluationGeneral -------------------------------*/
 
 template 
+          typename Number>
 inline
 FEEvaluationGeneral
 ::FEEvaluationGeneral (const MatrixFree &data_in,
@@ -3735,7 +3735,7 @@ FEEvaluationGeneral
             {
               proposed_quad_comp = no;
               break;
-           }
+            }
       if (proposed_dof_comp  != numbers::invalid_unsigned_int &&
           proposed_quad_comp != numbers::invalid_unsigned_int)
         {
@@ -3805,7 +3805,7 @@ namespace internal
   // like the functions below and corresponds to a usual matrix-matrix
   // product
   template 
+            int direction, bool dof_to_quad, bool add>
   inline
   void
   apply_tensor_product (const Number *shape_data,
@@ -3880,7 +3880,7 @@ namespace internal
   // fe_degree+1 degrees of freedom per direction and not n_q_points_1d for
   // those directions lower than the one currently applied
   template 
+            bool dof_to_quad, bool add>
   inline
   void
   apply_tensor_product_face (const Number *shape_data,
@@ -3969,7 +3969,7 @@ namespace internal
   // ones and in addition use the symmetry in entries to reduce the number of
   // read operations.
   template 
+            int direction, bool dof_to_quad, bool add>
   inline
   void
   apply_tensor_product_values (const Number *shape_values,
@@ -4172,7 +4172,7 @@ namespace internal
   // zeros and ones and in addition use the symmetry in entries to
   // reduce the number of read operations.
   template 
+            int direction, bool dof_to_quad, bool add>
   inline
   void
   apply_tensor_product_gradients (const Number *shape_gradients,
@@ -4316,7 +4316,7 @@ namespace internal
   // form assuming the symmetries of unit cell shape hessians for
   // finite elements in FEEvaluation
   template 
+            int direction, bool dof_to_quad, bool add>
   inline
   void
   apply_tensor_product_hessians (const Number *shape_hessians,
@@ -4688,7 +4688,7 @@ namespace internal
   //        [ 2.82  -1.528  2.673  0    -13.51  ]
   //        [-1      0.518 -0.75   2.482 10     ]
   template 
+            int direction, bool dof_to_quad, bool add>
   inline
   void
   apply_tensor_product_gradients_gl (const Number *shape_gradients,
@@ -5185,7 +5185,7 @@ namespace internal
 
 
 template 
+          typename Number>
 inline
 void
 FEEvaluationGeneral
@@ -5199,7 +5199,7 @@ FEEvaluationGeneral
 
 
 template 
+          typename Number>
 inline
 void
 FEEvaluationGeneral
@@ -5212,7 +5212,7 @@ FEEvaluationGeneral
 
 
 template 
+          typename Number>
 inline
 Point >
 FEEvaluationGeneral
@@ -5254,7 +5254,7 @@ FEEvaluationGeneral
 
 
 template 
+          typename Number>
 template 
 inline
 void
@@ -5270,7 +5270,7 @@ FEEvaluationGeneral
 
 
 template 
+          typename Number>
 template 
 inline
 void
@@ -5286,7 +5286,7 @@ FEEvaluationGeneral
 
 
 template 
+          typename Number>
 template 
 inline
 void
@@ -5304,7 +5304,7 @@ FEEvaluationGeneral
 
 
 template 
+          typename Number>
 inline
 FEEvaluation
 ::FEEvaluation (const MatrixFree &data_in,
@@ -5408,7 +5408,7 @@ FEEvaluation
 
 
 template 
+          typename Number>
 inline
 void
 FEEvaluation
@@ -5422,7 +5422,7 @@ FEEvaluation
 
 
 template 
+          typename Number>
 inline
 void
 FEEvaluation
@@ -5434,7 +5434,7 @@ FEEvaluation
 
 
 template 
+          typename Number>
 template 
 inline
 void
@@ -5447,18 +5447,18 @@ FEEvaluation
   // for all other orders, we choose even-odd
   if (fe_degree > 1 || n_q_points_1d > 3)
     internal::apply_tensor_product_evenodd, direction, dof_to_quad, add, 0>
-           (shape_val_evenodd, in, out);
+             VectorizedArray, direction, dof_to_quad, add, 0>
+             (shape_val_evenodd, in, out);
   else
     internal::apply_tensor_product_values, direction, dof_to_quad, add>
-           (this->data.shape_values.begin(), in, out);
+             VectorizedArray, direction, dof_to_quad, add>
+             (this->data.shape_values.begin(), in, out);
 }
 
 
 
 template 
+          typename Number>
 template 
 inline
 void
@@ -5468,12 +5468,12 @@ FEEvaluation
 {
   if (fe_degree > 1 || n_q_points_1d > 3)
     internal::apply_tensor_product_evenodd, direction, dof_to_quad, add, 1>
-           (shape_gra_evenodd, in, out);
+             VectorizedArray, direction, dof_to_quad, add, 1>
+             (shape_gra_evenodd, in, out);
   else
     internal::apply_tensor_product_gradients, direction, dof_to_quad, add>
-           (this->data.shape_gradients.begin(), in, out);
+             VectorizedArray, direction, dof_to_quad, add>
+             (this->data.shape_gradients.begin(), in, out);
 }
 
 
@@ -5482,7 +5482,7 @@ FEEvaluation
 // the same symmetry relations hold. However, it is not possible to omit some
 // values that are zero for the values
 template 
+          typename Number>
 template 
 inline
 void
@@ -5492,12 +5492,12 @@ FEEvaluation
 {
   if (fe_degree > 1 || n_q_points_1d > 3)
     internal::apply_tensor_product_evenodd, direction, dof_to_quad, add, 2>
-           (shape_hes_evenodd, in, out);
+             VectorizedArray, direction, dof_to_quad, add, 2>
+             (shape_hes_evenodd, in, out);
   else
     internal::apply_tensor_product_hessians, direction, dof_to_quad, add>
-           (this->data.shape_hessians.begin(), in, out);
+             VectorizedArray, direction, dof_to_quad, add>
+             (this->data.shape_hessians.begin(), in, out);
 }
 
 
diff --git a/deal.II/include/deal.II/matrix_free/matrix_free.templates.h b/deal.II/include/deal.II/matrix_free/matrix_free.templates.h
index 193df65ef7..359deb1d8f 100644
--- a/deal.II/include/deal.II/matrix_free/matrix_free.templates.h
+++ b/deal.II/include/deal.II/matrix_free/matrix_free.templates.h
@@ -595,7 +595,7 @@ void MatrixFree::initialize_indices
         for (types::global_dof_index i=start_index; iis_constrained(i)==true)
             dof_info[no].constrained_dofs.
-              push_back(static_cast(i-start_index));
+            push_back(static_cast(i-start_index));
       }
     }
 
diff --git a/deal.II/include/deal.II/meshworker/loop.h b/deal.II/include/deal.II/meshworker/loop.h
index 73ed8f5bd7..ecbf0c53d5 100644
--- a/deal.II/include/deal.II/meshworker/loop.h
+++ b/deal.II/include/deal.II/meshworker/loop.h
@@ -114,7 +114,7 @@ namespace MeshWorker
     const std_cxx1x::function &boundary_worker,
     const std_cxx1x::function& face_worker,
+                                    typename INFOBOX::CellInfo &)> &face_worker,
     const bool cells_first,
     const bool unique_faces_only)
   {
@@ -260,7 +260,7 @@ namespace MeshWorker
             const std_cxx1x::function &boundary_worker,
             const std_cxx1x::function& face_worker,
+                                            typename INFOBOX::CellInfo &)> &face_worker,
             ASSEMBLER &assembler,
             bool cells_first = true,
             bool unique_faces_only = true)
diff --git a/deal.II/include/deal.II/meshworker/simple.h b/deal.II/include/deal.II/meshworker/simple.h
index e43257df15..9f279b3c8b 100644
--- a/deal.II/include/deal.II/meshworker/simple.h
+++ b/deal.II/include/deal.II/meshworker/simple.h
@@ -915,7 +915,7 @@ namespace MeshWorker
           for (unsigned int j=0; j= threshold)
-		if (!mg_constrained_dofs->at_refinement_edge(level, i2[k]))
+                if (!mg_constrained_dofs->at_refinement_edge(level, i2[k]))
                   G.add(i1[j], i2[k], M(k,j));
         }
     }
@@ -961,40 +961,40 @@ namespace MeshWorker
       AssertDimension(M.m(), i1.size());
       AssertDimension(M.n(), i2.size());
       Assert(mg_constrained_dofs != 0, ExcInternalError());
-      
+
       for (unsigned int j=0; j= threshold)
-					     // Enter values into matrix only if j corresponds to a
-					     // degree of freedom on the refinemenent edge, k does
-					     // not, and both are not on the boundary. This is part
-					     // the difference between the complete matrix with no
-					     // boundary condition at the refinement edge and and
-					     // the matrix assembled above by assemble().
-	    
-					     // Thus the logic is: enter the row if it is
-					     // constrained by hanging node constraints (actually,
-					     // the whole refinement edge), but not if it is
-					     // constrained by a boundary constraint.
-	    if (mg_constrained_dofs->at_refinement_edge(level, i1[j]) &&
-		!mg_constrained_dofs->at_refinement_edge(level, i2[k]))
-	      {
-		if (mg_constrained_dofs->set_boundary_values())
-		  {
-		    if ((!mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
-			 !mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]))
-			||
-			(mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
-			 mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]) &&
-			 i1[j] == i2[k]))
-		      G.add(i1[j], i2[k], M(j,k));
-		  }
-		else
-		  G.add(i1[j], i2[k], M(j,k));
-	      }
+        for (unsigned int k=0; k= threshold)
+            // Enter values into matrix only if j corresponds to a
+            // degree of freedom on the refinemenent edge, k does
+            // not, and both are not on the boundary. This is part
+            // the difference between the complete matrix with no
+            // boundary condition at the refinement edge and and
+            // the matrix assembled above by assemble().
+
+            // Thus the logic is: enter the row if it is
+            // constrained by hanging node constraints (actually,
+            // the whole refinement edge), but not if it is
+            // constrained by a boundary constraint.
+            if (mg_constrained_dofs->at_refinement_edge(level, i1[j]) &&
+                !mg_constrained_dofs->at_refinement_edge(level, i2[k]))
+              {
+                if (mg_constrained_dofs->set_boundary_values())
+                  {
+                    if ((!mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
+                         !mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]))
+                        ||
+                        (mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
+                         mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]) &&
+                         i1[j] == i2[k]))
+                      G.add(i1[j], i2[k], M(j,k));
+                  }
+                else
+                  G.add(i1[j], i2[k], M(j,k));
+              }
     }
-    
-    
+
+
     template 
     inline void
     MGMatrixSimple::assemble_out(
@@ -1007,28 +1007,28 @@ namespace MeshWorker
       AssertDimension(M.n(), i1.size());
       AssertDimension(M.m(), i2.size());
       Assert(mg_constrained_dofs != 0, ExcInternalError());
-      
+
       for (unsigned int j=0; j= threshold)
-	    if (mg_constrained_dofs->at_refinement_edge(level, i1[j]) &&
-		!mg_constrained_dofs->at_refinement_edge(level, i2[k]))
-	      {
-		if (mg_constrained_dofs->set_boundary_values())
-		  {
-		    if ((!mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
-			 !mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]))
-			||
-			(mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
-			 mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]) &&
-			 i1[j] == i2[k]))
-		      G.add(i1[j], i2[k], M(k,j));
-		  }
-		else
-		  G.add(i1[j], i2[k], M(k,j));
-	      }
+        for (unsigned int k=0; k= threshold)
+            if (mg_constrained_dofs->at_refinement_edge(level, i1[j]) &&
+                !mg_constrained_dofs->at_refinement_edge(level, i2[k]))
+              {
+                if (mg_constrained_dofs->set_boundary_values())
+                  {
+                    if ((!mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
+                         !mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]))
+                        ||
+                        (mg_constrained_dofs->at_refinement_edge_boundary(level, i1[j]) &&
+                         mg_constrained_dofs->at_refinement_edge_boundary(level, i2[k]) &&
+                         i1[j] == i2[k]))
+                      G.add(i1[j], i2[k], M(k,j));
+                  }
+                else
+                  G.add(i1[j], i2[k], M(k,j));
+              }
     }
-  
+
 
     template 
     template 
@@ -1182,7 +1182,7 @@ namespace MeshWorker
       ResidualSimple::initialize(c);
     }
 
-    
+
     template 
     template 
     inline void
diff --git a/deal.II/include/deal.II/multigrid/sparse_matrix_collection.h b/deal.II/include/deal.II/multigrid/sparse_matrix_collection.h
index 4859e1ef7f..5e128a2f39 100644
--- a/deal.II/include/deal.II/multigrid/sparse_matrix_collection.h
+++ b/deal.II/include/deal.II/multigrid/sparse_matrix_collection.h
@@ -30,32 +30,32 @@ DEAL_II_NAMESPACE_OPEN
 
 namespace mg
 {
-/**
- * Handler and storage for all five SparseMatrix object involved in
- * using multigrid with local refinement.
- *
- * @author Baerbel Janssen, Guido Kanschat
- * @date 2013
- */
+  /**
+   * Handler and storage for all five SparseMatrix object involved in
+   * using multigrid with local refinement.
+   *
+   * @author Baerbel Janssen, Guido Kanschat
+   * @date 2013
+   */
   template 
   class SparseMatrixCollection : public Subscriptor
   {
-    public:
-      void resize(const unsigned int minlevel, const unsigned  int maxlevel);
-      
-      template 
-      void reinit(const DH& dof_handler);
-
-      void set_zero();
-      
-      MGLevelObject sparsity;
-      MGLevelObject sparsity_edge;
-      
-      MGLevelObject > matrix;
-      MGLevelObject > matrix_down;
-      MGLevelObject > matrix_up;
-      MGLevelObject > matrix_in;
-      MGLevelObject > matrix_out;
+  public:
+    void resize(const unsigned int minlevel, const unsigned  int maxlevel);
+
+    template 
+    void reinit(const DH &dof_handler);
+
+    void set_zero();
+
+    MGLevelObject sparsity;
+    MGLevelObject sparsity_edge;
+
+    MGLevelObject > matrix;
+    MGLevelObject > matrix_down;
+    MGLevelObject > matrix_up;
+    MGLevelObject > matrix_in;
+    MGLevelObject > matrix_out;
   };
 
 
@@ -81,28 +81,28 @@ namespace mg
   template 
   template 
   void
-  SparseMatrixCollection::reinit(const DH& dof_handler)
+  SparseMatrixCollection::reinit(const DH &dof_handler)
   {
     AssertIndexRange(sparsity.max_level(), dof_handler.get_tria().n_levels());
-    
+
     for (unsigned int level=sparsity.min_level();
-	 level<=sparsity.max_level();++level)
+         level<=sparsity.max_level(); ++level)
       {
-	CompressedSparsityPattern c_sparsity(dof_handler.n_dofs(level));      
-	MGTools::make_flux_sparsity_pattern(dof_handler, c_sparsity, level);
-	sparsity[level].copy_from(c_sparsity);
-	matrix[level].reinit(sparsity[level]);
-	matrix_in[level].reinit(sparsity[level]);
-	matrix_out[level].reinit(sparsity[level]);
-	if (level>0)
-	  {
-	    CompressedSparsityPattern ci_sparsity;
-	    ci_sparsity.reinit(dof_handler.n_dofs(level-1), dof_handler.n_dofs(level));
-	    MGTools::make_flux_sparsity_pattern_edge(dof_handler, ci_sparsity, level);
-	    sparsity_edge[level].copy_from(ci_sparsity);
-	    matrix_up[level].reinit(sparsity_edge[level]);
-	    matrix_down[level].reinit(sparsity_edge[level]);
-	  }
+        CompressedSparsityPattern c_sparsity(dof_handler.n_dofs(level));
+        MGTools::make_flux_sparsity_pattern(dof_handler, c_sparsity, level);
+        sparsity[level].copy_from(c_sparsity);
+        matrix[level].reinit(sparsity[level]);
+        matrix_in[level].reinit(sparsity[level]);
+        matrix_out[level].reinit(sparsity[level]);
+        if (level>0)
+          {
+            CompressedSparsityPattern ci_sparsity;
+            ci_sparsity.reinit(dof_handler.n_dofs(level-1), dof_handler.n_dofs(level));
+            MGTools::make_flux_sparsity_pattern_edge(dof_handler, ci_sparsity, level);
+            sparsity_edge[level].copy_from(ci_sparsity);
+            matrix_up[level].reinit(sparsity_edge[level]);
+            matrix_down[level].reinit(sparsity_edge[level]);
+          }
       }
   }
 
@@ -116,7 +116,7 @@ namespace mg
     matrix_up = 0.;
     matrix_down = 0.;
   }
-  
+
 }
 
 DEAL_II_NAMESPACE_CLOSE
diff --git a/deal.II/include/deal.II/numerics/derivative_approximation.h b/deal.II/include/deal.II/numerics/derivative_approximation.h
index 97e7cfc4b7..fb7062f195 100644
--- a/deal.II/include/deal.II/numerics/derivative_approximation.h
+++ b/deal.II/include/deal.II/numerics/derivative_approximation.h
@@ -552,7 +552,7 @@ private:
    * we are to work on.
    */
   template  class DH, class InputVector, int spacedim>
+            template  class DH, class InputVector, int spacedim>
   static void
   approximate_derivative (const Mapping    &mapping,
                           const DH         &dof,
@@ -566,10 +566,10 @@ private:
    * on the cell.
    */
   template  class DH, class InputVector, int spacedim>
+            template  class DH, class InputVector, int spacedim>
   static void
   approximate (SynchronousIterators::active_cell_iterator>,
-                Vector::iterator> > const &cell,
+               Vector::iterator> > const &cell,
                const Mapping    &mapping,
                const DH         &dof,
                const InputVector     &solution,
@@ -580,7 +580,7 @@ private:
    * derivative tensor.
    */
   template  class DH, class InputVector, int spacedim>
+            template  class DH, class InputVector, int spacedim>
   static void
   approximate_cell (const Mapping                            &mapping,
                     const DH                                 &dof,
diff --git a/deal.II/include/deal.II/numerics/fe_field_function.h b/deal.II/include/deal.II/numerics/fe_field_function.h
index b4df6928f1..af75bf5e74 100644
--- a/deal.II/include/deal.II/numerics/fe_field_function.h
+++ b/deal.II/include/deal.II/numerics/fe_field_function.h
@@ -168,8 +168,8 @@ namespace Functions
    * @author Luca Heltai, 2006, Markus Buerg, 2012, Wolfgang Bangerth, 2013
    */
   template ,
-           typename VECTOR=Vector >
+            typename DH=DoFHandler,
+            typename VECTOR=Vector >
   class FEFieldFunction :  public Function
   {
   public:
diff --git a/deal.II/include/deal.II/numerics/vector_tools.h b/deal.II/include/deal.II/numerics/vector_tools.h
index 28c5fa3f10..9d14cdcc10 100644
--- a/deal.II/include/deal.II/numerics/vector_tools.h
+++ b/deal.II/include/deal.II/numerics/vector_tools.h
@@ -539,8 +539,8 @@ namespace VectorTools
    * constraints object.
    */
   template  class DH,
-           class VECTOR>
+            template  class DH,
+            class VECTOR>
   void
   interpolate_to_different_mesh (const DH &dof1,
                                  const VECTOR            &u1,
@@ -573,8 +573,8 @@ namespace VectorTools
    * at hanging nodes.
    */
   template  class DH,
-           class VECTOR>
+            template  class DH,
+            class VECTOR>
   void
   interpolate_to_different_mesh (const DH &dof1,
                                  const VECTOR            &u1,
@@ -597,8 +597,8 @@ namespace VectorTools
    * destination DoFHandler.
    */
   template  class DH,
-           class VECTOR>
+            template  class DH,
+            class VECTOR>
   void
   interpolate_to_different_mesh (const InterGridMap > &intergridmap,
                                  const VECTOR                           &u1,
@@ -686,8 +686,8 @@ namespace VectorTools
                 VECTOR                   &vec,
                 const bool                enforce_zero_boundary = false,
                 const hp::QCollection  &q_boundary = hp::QCollection(dim > 1 ?
-                                                        QGauss(2) :
-                                                        Quadrature(0)),
+                                                            QGauss(2) :
+                                                            Quadrature(0)),
                 const bool                project_to_boundary_first = false);
 
   /**
@@ -703,8 +703,8 @@ namespace VectorTools
                 VECTOR                   &vec,
                 const bool                enforce_zero_boundary = false,
                 const hp::QCollection  &q_boundary = hp::QCollection(dim > 1 ?
-                                                        QGauss(2) :
-                                                        Quadrature(0)),
+                                                            QGauss(2) :
+                                                            Quadrature(0)),
                 const bool                project_to_boundary_first = false);
 
   /**
diff --git a/deal.II/include/deal.II/numerics/vector_tools.templates.h b/deal.II/include/deal.II/numerics/vector_tools.templates.h
index 5462558aff..a5d886252e 100644
--- a/deal.II/include/deal.II/numerics/vector_tools.templates.h
+++ b/deal.II/include/deal.II/numerics/vector_tools.templates.h
@@ -89,7 +89,7 @@ namespace VectorTools
     const bool                  fe_is_system = (n_components != 1);
 
     typename DH::active_cell_iterator cell = dof.begin_active(),
-						    endc = dof.end();
+                                                    endc = dof.end();
 
     // For FESystems many of the
     // unit_support_points will appear
@@ -200,7 +200,7 @@ namespace VectorTools
     hp::MappingCollection mapping_collection (mapping);
 
     hp::FEValues fe_values (mapping_collection,
-                                                      fe, support_quadrature, update_quadrature_points);
+                                          fe, support_quadrature, update_quadrature_points);
 
     for (; cell!=endc; ++cell)
       if (cell->is_locally_owned())
@@ -392,8 +392,8 @@ namespace VectorTools
 
 
   template  class DH,
-           class Vector>
+            template  class DH,
+            class Vector>
   void
   interpolate_to_different_mesh (const DH &dof1,
                                  const Vector            &u1,
@@ -416,8 +416,8 @@ namespace VectorTools
 
 
   template  class DH,
-           class Vector>
+            template  class DH,
+            class Vector>
   void
   interpolate_to_different_mesh (const DH &dof1,
                                  const Vector            &u1,
@@ -438,8 +438,8 @@ namespace VectorTools
 
 
   template  class DH,
-           class Vector>
+            template  class DH,
+            class Vector>
   void
   interpolate_to_different_mesh (const InterGridMap > &intergridmap,
                                  const Vector                           &u1,
@@ -507,12 +507,12 @@ namespace VectorTools
               template  class M_or_MC,
               template  class Q_or_QC>
     void project_compute_b_v (const M_or_MC   &mapping,
-                  const DH &dof,
-                  const Function &function,
-                  const bool                enforce_zero_boundary,
-                  const Q_or_QC  &q_boundary,
-                  const bool                project_to_boundary_first,
-                  std::map &boundary_values)
+                              const DH &dof,
+                              const Function &function,
+                              const bool                enforce_zero_boundary,
+                              const Q_or_QC  &q_boundary,
+                              const bool                project_to_boundary_first,
+                              std::map &boundary_values)
     {
       if (enforce_zero_boundary == true)
         // no need to project boundary
@@ -574,14 +574,14 @@ namespace VectorTools
               template  class M_or_MC,
               template  class Q_or_QC>
     void do_project (const M_or_MC   &mapping,
-                  const DH &dof,
-                  const ConstraintMatrix   &constraints,
-                  const Q_or_QC       &quadrature,
-                  const Function &function,
-                  Vector                   &vec_result,
-                  const bool                enforce_zero_boundary,
-                  const Q_or_QC  &q_boundary,
-                  const bool                project_to_boundary_first)
+                     const DH &dof,
+                     const ConstraintMatrix   &constraints,
+                     const Q_or_QC       &quadrature,
+                     const Function &function,
+                     Vector                   &vec_result,
+                     const bool                enforce_zero_boundary,
+                     const Q_or_QC  &q_boundary,
+                     const bool                project_to_boundary_first)
     {
       Assert (dof.get_fe().n_components() == function.n_components,
               ExcDimensionMismatch(dof.get_fe().n_components(),
@@ -673,9 +673,9 @@ namespace VectorTools
                 const bool                project_to_boundary_first)
   {
     do_project (mapping, dof, constraints, quadrature,
-        function, vec_result,
-        enforce_zero_boundary, q_boundary,
-        project_to_boundary_first);
+                function, vec_result,
+                enforce_zero_boundary, q_boundary,
+                project_to_boundary_first);
   }
 
 
@@ -707,9 +707,9 @@ namespace VectorTools
                 const bool                project_to_boundary_first)
   {
     do_project (mapping, dof, constraints, quadrature,
-        function, vec_result,
-        enforce_zero_boundary, q_boundary,
-        project_to_boundary_first);
+                function, vec_result,
+                enforce_zero_boundary, q_boundary,
+                project_to_boundary_first);
   }
 
 
@@ -1484,11 +1484,11 @@ namespace VectorTools
               template  class M_or_MC>
     static inline
     void do_interpolate_boundary_values (const M_or_MC &,
-                                      const DH                 &dof,
-                                      const typename FunctionMap::type &function_map,
-                                      std::map &boundary_values,
-                                      const ComponentMask       &component_mask,
-                                      const dealii::internal::int2type<1>)
+                                         const DH                 &dof,
+                                         const typename FunctionMap::type &function_map,
+                                         std::map &boundary_values,
+                                         const ComponentMask       &component_mask,
+                                         const dealii::internal::int2type<1>)
     {
       const unsigned int dim = DH::dimension;
       const unsigned int spacedim=DH::space_dimension;
@@ -1564,11 +1564,11 @@ namespace VectorTools
     static inline
     void
     do_interpolate_boundary_values (const M_or_MC &mapping,
-                                 const DH                 &dof,
-                                 const typename FunctionMap::type &function_map,
-                                 std::map &boundary_values,
-                                 const ComponentMask       &component_mask,
-                                 const dealii::internal::int2type)
+                                    const DH                 &dof,
+                                    const typename FunctionMap::type &function_map,
+                                    std::map &boundary_values,
+                                    const ComponentMask       &component_mask,
+                                    const dealii::internal::int2type)
     {
       const unsigned int dim = DH::dimension;
       const unsigned int spacedim=DH::space_dimension;
@@ -1747,17 +1747,17 @@ namespace VectorTools
                                    :
                                    (dim == 2 ?
                                     (i<2*fe.dofs_per_vertex ? i : i+2*fe.dofs_per_vertex)
+                                    :
+                                    (dim == 3 ?
+                                     (i<4*fe.dofs_per_vertex ?
+                                      i
                                       :
-                                      (dim == 3 ?
-                                       (i<4*fe.dofs_per_vertex ?
-                                        i
-                                        :
-                                        (i<4*fe.dofs_per_vertex+4*fe.dofs_per_line ?
-                                         i+4*fe.dofs_per_vertex
-                                         :
-                                         i+4*fe.dofs_per_vertex+8*fe.dofs_per_line))
+                                      (i<4*fe.dofs_per_vertex+4*fe.dofs_per_line ?
+                                       i+4*fe.dofs_per_vertex
                                        :
-                                       numbers::invalid_unsigned_int)));
+                                       i+4*fe.dofs_per_vertex+8*fe.dofs_per_line))
+                                     :
+                                     numbers::invalid_unsigned_int)));
                               Assert (cell_i < fe.dofs_per_cell, ExcInternalError());
 
                               // make sure that if this is not a primitive
@@ -1783,7 +1783,7 @@ namespace VectorTools
                     }
                   else
                     // fe has only one component, so save some computations
-                  {
+                    {
                       // get only the one component that this function has
                       dof_values_scalar.resize (fe.dofs_per_face);
                       function_map.find(boundary_component)->second
@@ -1811,8 +1811,8 @@ namespace VectorTools
                                const ComponentMask       &component_mask_)
   {
     do_interpolate_boundary_values (mapping, dof, function_map, boundary_values,
-                                 component_mask_,
-                                 dealii::internal::int2type());
+                                    component_mask_,
+                                    dealii::internal::int2type());
   }
 
 
@@ -1842,8 +1842,8 @@ namespace VectorTools
                                const ComponentMask       &component_mask_)
   {
     do_interpolate_boundary_values (mapping, dof, function_map, boundary_values,
-                                 component_mask_,
-                                 dealii::internal::int2type());
+                                    component_mask_,
+                                    dealii::internal::int2type());
   }
 
 
@@ -1972,11 +1972,11 @@ namespace VectorTools
               template  class Q_or_QC>
     void
     do_project_boundary_values (const M_or_MC   &mapping,
-                             const DH &dof,
-                             const typename FunctionMap::type &boundary_functions,
-                             const Q_or_QC        &q,
-                             std::map  &boundary_values,
-                             std::vector       component_mapping)
+                                const DH &dof,
+                                const typename FunctionMap::type &boundary_functions,
+                                const Q_or_QC        &q,
+                                std::map  &boundary_values,
+                                std::vector       component_mapping)
     {
       // in 1d, projection onto the 0d end points == interpolation
       if (dim == 1)
@@ -1987,11 +1987,11 @@ namespace VectorTools
           return;
         }
 
-  //TODO:[?] In project_boundary_values, no condensation of sparsity
-  //    structures, matrices and right hand sides or distribution of
-  //    solution vectors is performed. This is ok for dim<3 because then
-  //    there are no constrained nodes on the boundary, but is not
-  //    acceptable for higher dimensions. Fix this.
+      //TODO:[?] In project_boundary_values, no condensation of sparsity
+      //    structures, matrices and right hand sides or distribution of
+      //    solution vectors is performed. This is ok for dim<3 because then
+      //    there are no constrained nodes on the boundary, but is not
+      //    acceptable for higher dimensions. Fix this.
 
       if (component_mapping.size() == 0)
         {
@@ -2047,8 +2047,8 @@ namespace VectorTools
       // but it needs to be implemented
       if (dim>=3)
         {
-  #ifdef DEBUG
-  // Assert that there are no hanging nodes at the boundary
+#ifdef DEBUG
+          // Assert that there are no hanging nodes at the boundary
           int level = -1;
           for (typename DH::active_cell_iterator cell = dof.begin_active();
                cell != dof.end(); ++cell)
@@ -2064,7 +2064,7 @@ namespace VectorTools
                       }
                   }
               }
-  #endif
+#endif
         }
       sparsity.compress();
 
@@ -2185,9 +2185,9 @@ namespace VectorTools
                                 std::vector component_mapping)
   {
     do_project_boundary_values (mapping, dof,
-        boundary_functions,
-        q, boundary_values,
-        component_mapping);
+                                boundary_functions,
+                                q, boundary_values,
+                                component_mapping);
   }
 
 
@@ -2200,9 +2200,9 @@ namespace VectorTools
                                 std::vector component_mapping)
   {
     project_boundary_values (hp::StaticMappingQ1::mapping_collection, dof,
-        boundary_function,
-        q, boundary_values,
-        component_mapping);
+                             boundary_function,
+                             q, boundary_values,
+                             component_mapping);
   }
 
 
@@ -2299,11 +2299,11 @@ namespace VectorTools
 
 
     template 
-    std::ostream & operator << (std::ostream &out,
-				const VectorDoFTuple &vdt)
+    std::ostream &operator << (std::ostream &out,
+                               const VectorDoFTuple &vdt)
     {
       for (unsigned int d=0; d normal_vector
                       = (cell->face(face_no)->get_boundary()
                          .normal_vector (fe_values.quadrature_point(i)));
-		    if (normal_vector * fe_values.normal_vector(i) < 0)
-		      normal_vector *= -1;
+                    if (normal_vector * fe_values.normal_vector(i) < 0)
+                      normal_vector *= -1;
                     Assert (std::fabs(normal_vector.norm() - 1) < 1e-14,
                             ExcInternalError());
                     for (unsigned int d=0; dcenter() << std::endl
-			      << "   normal=" << normal_vector << std::endl;
+                    std::cout << "Adding normal vector:" << std::endl
+                              << "   dofs=" << vector_dofs << std::endl
+                              << "   cell=" << cell << " at " << cell->center() << std::endl
+                              << "   normal=" << normal_vector << std::endl;
 #endif
                   }
             }
@@ -4319,14 +4319,14 @@ namespace VectorTools
           same_dof_range[1] = dof_to_normals_map.end();
 
 #ifdef DEBUG_NO_NORMAL_FLUX
-	std::cout << "For dof indices <" << p->first << ">, found the following normals"
-		  << std::endl;
+        std::cout << "For dof indices <" << p->first << ">, found the following normals"
+                  << std::endl;
         for (typename DoFToNormalsMap::const_iterator
              q = same_dof_range[0];
              q != same_dof_range[1]; ++q)
-	  std::cout << "   " << q->second.first
-		    << " from cell " << q->second.second
-		    << std::endl;
+          std::cout << "   " << q->second.first
+                    << " from cell " << q->second.second
+                    << std::endl;
 #endif
 
 
@@ -4366,13 +4366,13 @@ namespace VectorTools
         Assert (cell_to_normals_map.size() >= 1, ExcInternalError());
 
 #ifdef DEBUG_NO_NORMAL_FLUX
-	std::cout << "   cell_to_normals_map:" << std::endl;
+        std::cout << "   cell_to_normals_map:" << std::endl;
         for (typename CellToNormalsMap::const_iterator
              x = cell_to_normals_map.begin();
              x != cell_to_normals_map.end(); ++x)
-	  std::cout << "      " << x->first << " -> ("
-		    << x->second.first << ',' << x->second.second << ')'
-		    << std::endl;
+          std::cout << "      " << x->first << " -> ("
+                    << x->second.first << ',' << x->second.second << ')'
+                    << std::endl;
 #endif
 
         // count the maximum number of contributions from each cell
@@ -4391,17 +4391,17 @@ namespace VectorTools
 
         switch (max_n_contributions_per_cell)
           {
-            // first deal with the case that a number of cells all have
-            // registered that they have a normal vector defined at the
-            // location of a given vector dof, and that each of them have
-            // encountered this vector dof exactly once while looping over all
-            // their faces. as stated in the documentation, this is the case
-            // where we want to simply average over all normal vectors
-	    //
-	    // the typical case is in 2d where multiple cells meet at one
-	    // vertex sitting on the boundary. same in 3d for a vertex that
-	    // is associated with only one of the boundary indicators passed
-	    // to this function
+          // first deal with the case that a number of cells all have
+          // registered that they have a normal vector defined at the
+          // location of a given vector dof, and that each of them have
+          // encountered this vector dof exactly once while looping over all
+          // their faces. as stated in the documentation, this is the case
+          // where we want to simply average over all normal vectors
+          //
+          // the typical case is in 2d where multiple cells meet at one
+          // vertex sitting on the boundary. same in 3d for a vertex that
+          // is associated with only one of the boundary indicators passed
+          // to this function
           case 1:
           {
 
@@ -4512,21 +4512,21 @@ namespace VectorTools
             // if this is dim-1 for all cells (if a single cell has
             // contributed dim, or if all adjacent cells have contributed 1
             // normal vector, this is already handled above).
-	    //
-	    // we only implement the case that all cells contribute
-	    // dim-1 because we assume that we are following an edge
-	    // of the domain (think: we are looking at a vertex
-	    // located on one of the edges of a refined cube where the
-	    // boundary indicators of the two adjacent faces of the
-	    // cube are both listed in the set of boundary indicators
-	    // passed to this function). in that case, all cells along
-	    // that edge of the domain are assumed to have contributed
-	    // dim-1 normal vectors. however, there are cases where
-	    // this assumption is not justified (see the lengthy
-	    // explanation in test no_flux_12.cc) and in those cases
-	    // we simply ignore the cell that contributes only
-	    // once. this is also discussed at length in the
-	    // documentation of this function.
+            //
+            // we only implement the case that all cells contribute
+            // dim-1 because we assume that we are following an edge
+            // of the domain (think: we are looking at a vertex
+            // located on one of the edges of a refined cube where the
+            // boundary indicators of the two adjacent faces of the
+            // cube are both listed in the set of boundary indicators
+            // passed to this function). in that case, all cells along
+            // that edge of the domain are assumed to have contributed
+            // dim-1 normal vectors. however, there are cases where
+            // this assumption is not justified (see the lengthy
+            // explanation in test no_flux_12.cc) and in those cases
+            // we simply ignore the cell that contributes only
+            // once. this is also discussed at length in the
+            // documentation of this function.
             //
             // for each contributing cell compute the tangential vector that
             // remains unconstrained
@@ -4537,21 +4537,21 @@ namespace VectorTools
                  ++contribution)
               {
 #ifdef DEBUG_NO_NORMAL_FLUX
-		std::cout << "   Treating edge case with dim-1 contributions." << std::endl
-			  << "   Looking at cell " << contribution->first
-			  << " which has contributed these normal vectors:"
-			  << std::endl;
-		for (typename std::list >::const_iterator
-                       t = contribution->second.begin();
-		     t != contribution->second.end();
-		     ++t)
-		  std::cout << "      " << *t << std::endl;
+                std::cout << "   Treating edge case with dim-1 contributions." << std::endl
+                          << "   Looking at cell " << contribution->first
+                          << " which has contributed these normal vectors:"
+                          << std::endl;
+                for (typename std::list >::const_iterator
+                     t = contribution->second.begin();
+                     t != contribution->second.end();
+                     ++t)
+                  std::cout << "      " << *t << std::endl;
 #endif
 
-		// as mentioned above, simply ignore cells that only
-		// contribute once
+                // as mentioned above, simply ignore cells that only
+                // contribute once
                 if (contribution->second.size() < dim-1)
-		  continue;
+                  continue;
 
                 Tensor<1,dim> normals[dim-1];
                 {
@@ -4653,7 +4653,7 @@ namespace VectorTools
 
   template  class DH, int spacedim>
   void
-  compute_normal_flux_constraints (const DH&dof_handler,
+  compute_normal_flux_constraints (const DH &dof_handler,
                                    const unsigned int     first_vector_component,
                                    const std::set &boundary_ids,
                                    ConstraintMatrix      &constraints,
@@ -4698,7 +4698,7 @@ namespace VectorTools
                       std::max(n_scalar_indices,
                                fe.face_system_to_component_index(i).second+1);
                     cell_vector_dofs[fe.face_system_to_component_index(i).second]
-                      [fe.face_system_to_component_index(i).first-first_vector_component]
+                    [fe.face_system_to_component_index(i).first-first_vector_component]
                       = face_dofs[i];
                   }
 
@@ -4713,8 +4713,8 @@ namespace VectorTools
     // can find constrained ones
     unsigned int n_total_constraints_found = 0;
     for (typename std::set,
-           PointComparator >::const_iterator it=vector_dofs.begin();
-           it!=vector_dofs.end(); ++it)
+         PointComparator >::const_iterator it=vector_dofs.begin();
+         it!=vector_dofs.end(); ++it)
       {
         unsigned int n_constraints = 0;
         bool is_constrained[dim];
@@ -4753,7 +4753,7 @@ namespace VectorTools
                   normal[d] = 1.;
                 }
             AssertIndexRange(constrained_index, dim);
-            const std::vector >* constrained
+            const std::vector > *constrained
               = no_normal_flux_constraints.get_constraint_entries((*it)[constrained_index]);
             // find components to which this index is constrained to
             Assert(constrained != 0, ExcInternalError());
@@ -4997,7 +4997,7 @@ namespace VectorTools
               double sum = 0;
               for (unsigned int k=0; k            &p,
       Point q1, q2, q3, q4;
       Vector
       v1(this->n_components), v2(this->n_components),
-         v3(this->n_components), v4(this->n_components);
+      v3(this->n_components), v4(this->n_components);
       const double h_inv_12=1./(12*h);
       for (unsigned int i=0; i > &patch
 
 template <>
 void DataOutBase::write_svg<1,1> (const std::vector > &patches,
-                             const std::vector &data_names,
-                             const std::vector > &vector_data_ranges,
-                             const SvgFlags &flags,
-                             std::ostream &out)
+                                  const std::vector &data_names,
+                                  const std::vector > &vector_data_ranges,
+                                  const SvgFlags &flags,
+                                  std::ostream &out)
 {
   AssertThrow (false, ExcNotImplemented());
 }
@@ -6533,7 +6533,7 @@ void DataOutInterface::write_vtu_in_parallel (const char *filename
   MPI_Info_create(&info);
   MPI_File fh;
   err = MPI_File_open(comm, const_cast(filename),
-                MPI_MODE_CREATE | MPI_MODE_WRONLY, info, &fh);
+                      MPI_MODE_CREATE | MPI_MODE_WRONLY, info, &fh);
   AssertThrow(err==0, ExcMessage("Unable to open file with MPI_File_open!"));
 
 
diff --git a/deal.II/source/base/exceptions.cc b/deal.II/source/base/exceptions.cc
index e98a4d89b9..80e04826cc 100644
--- a/deal.II/source/base/exceptions.cc
+++ b/deal.II/source/base/exceptions.cc
@@ -118,7 +118,7 @@ void ExceptionBase::set_fields (const char *f,
 #endif
 }
 
-const char* ExceptionBase::what() const throw()
+const char *ExceptionBase::what() const throw()
 {
   // If no error c_string was generated so far, do it now:
   if (what_str == "")
diff --git a/deal.II/source/base/function_parser.cc b/deal.II/source/base/function_parser.cc
index c26b0594e5..285ff50a7a 100644
--- a/deal.II/source/base/function_parser.cc
+++ b/deal.II/source/base/function_parser.cc
@@ -24,14 +24,14 @@
 namespace fparser
 {
   class FunctionParser: public ::FunctionParser
-{};
+  {};
 }
 #else
 
 namespace fparser
 {
   class FunctionParser
-{};
+  {};
 }
 
 #endif
diff --git a/deal.II/source/base/geometry_info.cc b/deal.II/source/base/geometry_info.cc
index 7bb041574e..4ca5e47b44 100644
--- a/deal.II/source/base/geometry_info.cc
+++ b/deal.II/source/base/geometry_info.cc
@@ -1843,11 +1843,11 @@ void
 GeometryInfo::
 alternating_form_at_vertices
 #ifndef DEAL_II_CONSTEXPR_BUG
-    (const Point (&vertices)[vertices_per_cell],
-     Tensor (&forms)[vertices_per_cell])
+(const Point (&vertices)[vertices_per_cell],
+ Tensor (&forms)[vertices_per_cell])
 #else
-    (const Point *vertices,
-     Tensor *forms)
+(const Point *vertices,
+ Tensor *forms)
 #endif
 {
   // for each of the vertices,
diff --git a/deal.II/source/base/logstream.cc b/deal.II/source/base/logstream.cc
index 447ff34110..75b401e569 100644
--- a/deal.II/source/base/logstream.cc
+++ b/deal.II/source/base/logstream.cc
@@ -159,26 +159,32 @@ LogStream::operator<< (std::ostream& (*p) (std::ostream &))
   {
     // Implement a minimalistic stream buffer that only stores the fact
     // whether overflow or sync was called
-    public:
-      QueryStreambuf()
-        : flushed_(false), newline_written_(false)
-      {
-      }
-      bool flushed() { return flushed_; }
-      bool newline_written() { return newline_written_; }
-    private:
-      int_type overflow(int_type ch)
-        {
-          newline_written_ = true;
-          return ch;
-        }
-      int sync()
-        {
-          flushed_ = true;
-          return 0;
-        }
-      bool flushed_;
-      bool newline_written_;
+  public:
+    QueryStreambuf()
+      : flushed_(false), newline_written_(false)
+    {
+    }
+    bool flushed()
+    {
+      return flushed_;
+    }
+    bool newline_written()
+    {
+      return newline_written_;
+    }
+  private:
+    int_type overflow(int_type ch)
+    {
+      newline_written_ = true;
+      return ch;
+    }
+    int sync()
+    {
+      flushed_ = true;
+      return 0;
+    }
+    bool flushed_;
+    bool newline_written_;
   } query_streambuf;
 
   {
diff --git a/deal.II/source/base/mpi.cc b/deal.II/source/base/mpi.cc
index 4b93f9be56..97a7c89b66 100644
--- a/deal.II/source/base/mpi.cc
+++ b/deal.II/source/base/mpi.cc
@@ -190,7 +190,7 @@ namespace Utilities
       void max_reduce ( const void *in_lhs_,
                         void *inout_rhs_,
                         int *len,
-                        MPI_Datatype * )
+                        MPI_Datatype *)
       {
         const MinMaxAvg *in_lhs = static_cast(in_lhs_);
         MinMaxAvg *inout_rhs = static_cast(inout_rhs_);
@@ -443,7 +443,7 @@ namespace Utilities
       // when running PETSc, because we initialize MPI ourselves before calling
       // PetscInitialize
 #ifdef DEAL_II_WITH_MPI
-     int mpi_err = 0;
+      int mpi_err = 0;
 
       int MPI_has_been_started = 0;
       MPI_Initialized(&MPI_has_been_started);
diff --git a/deal.II/source/base/multithread_info.cc b/deal.II/source/base/multithread_info.cc
index a3a92163de..815af8bd45 100644
--- a/deal.II/source/base/multithread_info.cc
+++ b/deal.II/source/base/multithread_info.cc
@@ -105,7 +105,7 @@ unsigned int MultithreadInfo::get_n_cpus()
 void MultithreadInfo::set_thread_limit(const unsigned int max_threads)
 {
   Assert(n_max_threads==numbers::invalid_unsigned_int,
-      ExcMessage("Calling set_thread_limit() more than once is not supported!"));
+         ExcMessage("Calling set_thread_limit() more than once is not supported!"));
 
   unsigned int max_threads_env = numbers::invalid_unsigned_int;
   char *penv;
diff --git a/deal.II/source/base/path_search.cc b/deal.II/source/base/path_search.cc
index eaa59d4ff0..9a1b0808b6 100644
--- a/deal.II/source/base/path_search.cc
+++ b/deal.II/source/base/path_search.cc
@@ -131,19 +131,19 @@ PathSearch::find (const std::string &filename,
       // the whole filename specified, including (possibly)
       // the suffix
       {
-	real_name = *path + filename;
-	if (debug > 1)
-	  deallog << "PathSearch[" << cls << "] trying "
-		  << real_name << std::endl;
-	FILE *fp = fopen(real_name.c_str(), open_mode);
-	if (fp != 0)
-	  {
-	    if (debug > 0)
-	      deallog << "PathSearch[" << cls << "] opened "
-		      << real_name << std::endl;
-	    fclose(fp);
-	    return real_name;
-	  }
+        real_name = *path + filename;
+        if (debug > 1)
+          deallog << "PathSearch[" << cls << "] trying "
+                  << real_name << std::endl;
+        FILE *fp = fopen(real_name.c_str(), open_mode);
+        if (fp != 0)
+          {
+            if (debug > 0)
+              deallog << "PathSearch[" << cls << "] opened "
+                      << real_name << std::endl;
+            fclose(fp);
+            return real_name;
+          }
       }
 
       // try again with the suffix appended, unless there is
@@ -153,13 +153,13 @@ PathSearch::find (const std::string &filename,
           real_name = *path + filename + suffix;
           if (debug > 1)
             deallog << "PathSearch[" << cls << "] trying "
-            << real_name << std::endl;
+                    << real_name << std::endl;
           FILE *fp = fopen(real_name.c_str(), open_mode);
           if (fp != 0)
             {
               if (debug > 0)
                 deallog << "PathSearch[" << cls << "] opened "
-                << real_name << std::endl;
+                        << real_name << std::endl;
               fclose(fp);
               return real_name;
             }
diff --git a/deal.II/source/base/polynomial.cc b/deal.II/source/base/polynomial.cc
index cd0b61f7e9..19ecd4654f 100644
--- a/deal.II/source/base/polynomial.cc
+++ b/deal.II/source/base/polynomial.cc
@@ -149,9 +149,9 @@ namespace Polynomials
             }
             break;
 
-            // manually implement size 1 (values only), size 2 (value + first
-            // derivative), and size 3 (up to second derivative) since they
-            // might be called often. then, we can unroll the loop.
+          // manually implement size 1 (values only), size 2 (value + first
+          // derivative), and size 3 (up to second derivative) since they
+          // might be called often. then, we can unroll the loop.
           case 1:
             values[0] = 1;
             for (unsigned int i=0; i::Quadrature (const Quadrature<0> &)
 
 template 
 Quadrature::Quadrature (const Quadrature &q)
-  :
-  Subscriptor(),
-  quadrature_points (Utilities::fixed_power(q.size())),
-  weights (Utilities::fixed_power(q.size()))
+:
+Subscriptor(),
+            quadrature_points (Utilities::fixed_power(q.size())),
+            weights (Utilities::fixed_power(q.size()))
 {
   Assert (dim <= 3, ExcNotImplemented());
 
diff --git a/deal.II/source/base/table_handler.cc b/deal.II/source/base/table_handler.cc
index 831434c98d..78c9a8089f 100644
--- a/deal.II/source/base/table_handler.cc
+++ b/deal.II/source/base/table_handler.cc
@@ -391,7 +391,7 @@ void TableHandler::write_text(std::ostream &out,
       out << std::flush;
       return;
     }
-    
+
     case simple_table_with_separate_column_description:
     {
       // write the captions
@@ -417,7 +417,7 @@ void TableHandler::write_text(std::ostream &out,
       out << std::flush;
       return;
     }
-    
+
     case table_with_separate_column_description:
     {
       // writing the captions for table_with_separate_column_description
@@ -430,7 +430,7 @@ void TableHandler::write_text(std::ostream &out,
         }
       break;
     }
-    
+
     case table_with_headers:
     {
       // This format output supercolumn headers and aligns them centered
@@ -507,12 +507,12 @@ void TableHandler::write_text(std::ostream &out,
       out << '\n';
       break;
     }
-    
+
     default:
       Assert (false, ExcInternalError());
     }
 
-  
+
   // finally output the data itself for
   // table_with_headers or table_with_separate_column_description:
   for (unsigned int i=0; i::connectivity_destroy) (p4est_connectivity_t *connectivity)
@@ -325,7 +325,7 @@ namespace internal
       = p4est_connectivity_save;
 
     int (&functions<2>::connectivity_is_valid) (types<2>::connectivity
-                                                 *connectivity)
+                                                *connectivity)
       = p4est_connectivity_is_valid;
 
     types<2>::connectivity *
@@ -624,7 +624,7 @@ namespace internal
       = p8est_connectivity_save;
 
     int (&functions<3>::connectivity_is_valid) (types<3>::connectivity
-                                                   *connectivity)
+                                                *connectivity)
       = p8est_connectivity_is_valid;
 
     types<3>::connectivity *
@@ -771,7 +771,7 @@ namespace
                                std::vector &vertex_touch_count,
                                std::vector::active_cell_iterator,unsigned int> > >
-                               & vertex_to_cell)
+                               &vertex_to_cell)
   {
     vertex_touch_count.resize (triangulation.n_vertices());
     vertex_to_cell.resize (triangulation.n_vertices());
@@ -795,7 +795,7 @@ namespace
                              std::vector &edge_touch_count,
                              std::vector::active_cell_iterator,unsigned int> > >
-                             & edge_to_cell)
+                             &edge_to_cell)
   {
     Assert (triangulation.n_levels() == 1, ExcInternalError());
 
@@ -825,7 +825,7 @@ namespace
                             const std::vector &vertex_touch_count,
                             const std::vector::active_cell_iterator,unsigned int> > >
-                            & vertex_to_cell,
+                            &vertex_to_cell,
                             const std::vector &coarse_cell_to_p4est_tree_permutation,
                             const bool set_vertex_info,
                             typename internal::p4est::types::connectivity *connectivity)
@@ -2082,10 +2082,11 @@ namespace parallel
     Triangulation::
     load(const char *filename)
     {
-      if (parallel_ghost != 0) {
-        dealii::internal::p4est::functions::ghost_destroy (parallel_ghost);
-        parallel_ghost = 0;
-      }
+      if (parallel_ghost != 0)
+        {
+          dealii::internal::p4est::functions::ghost_destroy (parallel_ghost);
+          parallel_ghost = 0;
+        }
       dealii::internal::p4est::functions::destroy (parallel_forest);
       parallel_forest = 0;
       dealii::internal::p4est::functions::connectivity_destroy (connectivity);
@@ -2116,9 +2117,9 @@ namespace parallel
           // this command changes the distribution for some reason, so we
           // will leave it in here.
           dealii::internal::p4est::functions::
-                partition (parallel_forest,
-                           /* prepare coarsening */ 1,
-                           /* weight_callback */ NULL);
+          partition (parallel_forest,
+                     /* prepare coarsening */ 1,
+                     /* weight_callback */ NULL);
 
 
         }
@@ -2347,7 +2348,7 @@ namespace parallel
       num_ett = std::accumulate (edge_touch_count.begin(),
                                  edge_touch_count.end(),
                                  0);
-      
+
       // now create a connectivity object with the right sizes for all arrays
       const bool set_vertex_info
 #ifdef DEBUG
@@ -2470,7 +2471,7 @@ namespace parallel
       // function. Note that the smoothing flag is used in the normal
       // refinement process.
       typename Triangulation::MeshSmoothing
-        save_smooth = this->smooth_grid;
+      save_smooth = this->smooth_grid;
 
       // We will refine manually to match the p4est further down, which
       // obeys a level difference of 2 at each vertex (see the balance call
@@ -2528,13 +2529,13 @@ namespace parallel
           parallel_ghost = 0;
         }
       parallel_ghost = dealii::internal::p4est::functions::ghost_new (parallel_forest,
-                   (dim == 2
-                    ?
-                    typename dealii::internal::p4est::types::
-                    balance_type(P4EST_CONNECT_CORNER)
-                    :
-                    typename dealii::internal::p4est::types::
-                    balance_type(P8EST_CONNECT_CORNER)));
+                       (dim == 2
+                        ?
+                        typename dealii::internal::p4est::types::
+                        balance_type(P4EST_CONNECT_CORNER)
+                        :
+                        typename dealii::internal::p4est::types::
+                        balance_type(P8EST_CONNECT_CORNER)));
 
       Assert (parallel_ghost, ExcInternalError());
 
@@ -2767,7 +2768,7 @@ namespace parallel
                       Assert(cell->neighbor(f)->active(), ExcInternalError());
                       Assert(cell->neighbor(f)->subdomain_id() != numbers::artificial_subdomain_id, ExcInternalError());
                       Assert(cell->neighbor(f)->level_subdomain_id() == numbers::artificial_subdomain_id
-                           || cell->neighbor(f)->level_subdomain_id() == cell->neighbor(f)->subdomain_id(), ExcInternalError());
+                             || cell->neighbor(f)->level_subdomain_id() == cell->neighbor(f)->subdomain_id(), ExcInternalError());
                       cell->neighbor(f)->set_level_subdomain_id(cell->neighbor(f)->subdomain_id());
                     }
                 }
@@ -3049,7 +3050,7 @@ namespace parallel
     register_data_attach (const std::size_t size,
                           const std_cxx1x::function & pack_callback)
+                                                         void *)> &pack_callback)
     {
       Assert(size>0, ExcMessage("register_data_attach(), size==0"));
       Assert(attached_data_pack_callbacks.size()==n_attached_datas,
@@ -3072,7 +3073,7 @@ namespace parallel
     notify_ready_to_unpack (const unsigned int offset,
                             const std_cxx1x::function & unpack_callback)
+                                                            const void *)> &unpack_callback)
     {
       Assert (offset < attached_data_size, ExcMessage ("invalid offset in notify_ready_to_unpack()"));
       Assert (n_attached_datas > 0, ExcMessage ("notify_ready_to_unpack() called too often"));
@@ -3150,7 +3151,7 @@ namespace parallel
       struct find_ghosts
       {
         typename dealii::parallel::distributed::Triangulation *triangulation;
-        sc_array_t * subids;
+        sc_array_t *subids;
         std::map >
         *vertices_with_ghost_neighbors;
       };
@@ -3163,12 +3164,12 @@ namespace parallel
       template 
       void
       find_ghosts_corner
-      (typename dealii::internal::p4est::iter::corner_info * info,
+      (typename dealii::internal::p4est::iter::corner_info *info,
        void *user_data)
       {
         int i, j;
         int nsides = info->sides.elem_count;
-        typename dealii::internal::p4est::iter::corner_side * sides =
+        typename dealii::internal::p4est::iter::corner_side *sides =
           (typename dealii::internal::p4est::iter::corner_side *)
           (info->sides.array);
         struct find_ghosts *fg = static_cast *>(user_data);
@@ -3180,35 +3181,41 @@ namespace parallel
         *vertices_with_ghost_neighbors = fg->vertices_with_ghost_neighbors;
 
         subids->elem_count = 0;
-        for (i = 0; i < nsides; i++) {
-          if (sides[i].is_ghost) {
-            typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].quad));
-            Assert (cell->is_ghost(), ExcMessage ("ghost quad did not find ghost cell"));
-            dealii::types::subdomain_id *subid =
-              static_cast(sc_array_push (subids));
-            *subid = cell->subdomain_id();
+        for (i = 0; i < nsides; i++)
+          {
+            if (sides[i].is_ghost)
+              {
+                typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].quad));
+                Assert (cell->is_ghost(), ExcMessage ("ghost quad did not find ghost cell"));
+                dealii::types::subdomain_id *subid =
+                  static_cast(sc_array_push (subids));
+                *subid = cell->subdomain_id();
+              }
           }
-        }
 
-        if (!subids->elem_count) {
-          return;
-        }
+        if (!subids->elem_count)
+          {
+            return;
+          }
 
         nsubs = (int) subids->elem_count;
         subdomain_ids = (dealii::types::subdomain_id *) (subids->array);
 
-        for (i = 0; i < nsides; i++) {
-          if (!sides[i].is_ghost) {
-            typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].quad));
+        for (i = 0; i < nsides; i++)
+          {
+            if (!sides[i].is_ghost)
+              {
+                typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].quad));
 
-            Assert (!cell->is_ghost(), ExcMessage ("local quad found ghost cell"));
+                Assert (!cell->is_ghost(), ExcMessage ("local quad found ghost cell"));
 
-            for (j = 0; j < nsubs; j++) {
-              (*vertices_with_ghost_neighbors)[cell->vertex_index(sides[i].corner)]
-              .insert (subdomain_ids[j]);
-            }
+                for (j = 0; j < nsubs; j++)
+                  {
+                    (*vertices_with_ghost_neighbors)[cell->vertex_index(sides[i].corner)]
+                    .insert (subdomain_ids[j]);
+                  }
+              }
           }
-        }
 
         subids->elem_count = 0;
       }
@@ -3219,12 +3226,12 @@ namespace parallel
       template 
       void
       find_ghosts_edge
-      (typename dealii::internal::p4est::iter::edge_info * info,
+      (typename dealii::internal::p4est::iter::edge_info *info,
        void *user_data)
       {
         int i, j, k;
         int nsides = info->sides.elem_count;
-        typename dealii::internal::p4est::iter::edge_side * sides =
+        typename dealii::internal::p4est::iter::edge_side *sides =
           (typename dealii::internal::p4est::iter::edge_side *)
           (info->sides.array);
         struct find_ghosts *fg = static_cast *>(user_data);
@@ -3236,40 +3243,50 @@ namespace parallel
         *vertices_with_ghost_neighbors = fg->vertices_with_ghost_neighbors;
 
         subids->elem_count = 0;
-        for (i = 0; i < nsides; i++) {
-          if (sides[i].is_hanging) {
-            for (j = 0; j < 2; j++) {
-              if (sides[i].is.hanging.is_ghost[j]) {
-                typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
-                dealii::types::subdomain_id *subid =
-                  static_cast(sc_array_push (subids));
-                *subid = cell->subdomain_id();
+        for (i = 0; i < nsides; i++)
+          {
+            if (sides[i].is_hanging)
+              {
+                for (j = 0; j < 2; j++)
+                  {
+                    if (sides[i].is.hanging.is_ghost[j])
+                      {
+                        typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
+                        dealii::types::subdomain_id *subid =
+                          static_cast(sc_array_push (subids));
+                        *subid = cell->subdomain_id();
+                      }
+                  }
               }
-            }
           }
-        }
 
-        if (!subids->elem_count) {
-          return;
-        }
+        if (!subids->elem_count)
+          {
+            return;
+          }
 
         nsubs = (int) subids->elem_count;
         subdomain_ids = (dealii::types::subdomain_id *) (subids->array);
 
-        for (i = 0; i < nsides; i++) {
-          if (sides[i].is_hanging) {
-            for (j = 0; j < 2; j++) {
-              if (!sides[i].is.hanging.is_ghost[j]) {
-                typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
+        for (i = 0; i < nsides; i++)
+          {
+            if (sides[i].is_hanging)
+              {
+                for (j = 0; j < 2; j++)
+                  {
+                    if (!sides[i].is.hanging.is_ghost[j])
+                      {
+                        typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
 
-                for (k = 0; k < nsubs; k++) {
-                  (*vertices_with_ghost_neighbors)[cell->vertex_index(p8est_edge_corners[sides[i].edge][1^j])]
-                  .insert (subdomain_ids[k]);
-                }
+                        for (k = 0; k < nsubs; k++)
+                          {
+                            (*vertices_with_ghost_neighbors)[cell->vertex_index(p8est_edge_corners[sides[i].edge][1^j])]
+                            .insert (subdomain_ids[k]);
+                          }
+                      }
+                  }
               }
-            }
           }
-        }
 
         subids->elem_count = 0;
       }
@@ -3280,12 +3297,12 @@ namespace parallel
       template 
       void
       find_ghosts_face
-      (typename dealii::internal::p4est::iter::face_info * info,
+      (typename dealii::internal::p4est::iter::face_info *info,
        void *user_data)
       {
         int i, j, k;
         int nsides = info->sides.elem_count;
-        typename dealii::internal::p4est::iter::face_side * sides =
+        typename dealii::internal::p4est::iter::face_side *sides =
           (typename dealii::internal::p4est::iter::face_side *)
           (info->sides.array);
         struct find_ghosts *fg = static_cast *>(user_data);
@@ -3298,46 +3315,58 @@ namespace parallel
         int limit = (dim == 2) ? 2 : 4;
 
         subids->elem_count = 0;
-        for (i = 0; i < nsides; i++) {
-          if (sides[i].is_hanging) {
-            for (j = 0; j < limit; j++) {
-              if (sides[i].is.hanging.is_ghost[j]) {
-                typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
-                dealii::types::subdomain_id *subid =
-                  static_cast(sc_array_push (subids));
-                *subid = cell->subdomain_id();
+        for (i = 0; i < nsides; i++)
+          {
+            if (sides[i].is_hanging)
+              {
+                for (j = 0; j < limit; j++)
+                  {
+                    if (sides[i].is.hanging.is_ghost[j])
+                      {
+                        typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
+                        dealii::types::subdomain_id *subid =
+                          static_cast(sc_array_push (subids));
+                        *subid = cell->subdomain_id();
+                      }
+                  }
               }
-            }
           }
-        }
 
-        if (!subids->elem_count) {
-          return;
-        }
+        if (!subids->elem_count)
+          {
+            return;
+          }
 
         nsubs = (int) subids->elem_count;
         subdomain_ids = (dealii::types::subdomain_id *) (subids->array);
 
-        for (i = 0; i < nsides; i++) {
-          if (sides[i].is_hanging) {
-            for (j = 0; j < limit; j++) {
-              if (!sides[i].is.hanging.is_ghost[j]) {
-                typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
+        for (i = 0; i < nsides; i++)
+          {
+            if (sides[i].is_hanging)
+              {
+                for (j = 0; j < limit; j++)
+                  {
+                    if (!sides[i].is.hanging.is_ghost[j])
+                      {
+                        typename dealii::parallel::distributed::Triangulation::cell_iterator cell = cell_from_quad (triangulation, sides[i].treeid, *(sides[i].is.hanging.quad[j]));
 
-                for (k = 0; k < nsubs; k++) {
-                  if (dim == 2) {
-                    (*vertices_with_ghost_neighbors)[cell->vertex_index(p4est_face_corners[sides[i].face][(limit - 1)^j])]
-                    .insert (subdomain_ids[k]);
-                  }
-                  else {
-                    (*vertices_with_ghost_neighbors)[cell->vertex_index(p8est_face_corners[sides[i].face][(limit - 1)^j])]
-                    .insert (subdomain_ids[k]);
+                        for (k = 0; k < nsubs; k++)
+                          {
+                            if (dim == 2)
+                              {
+                                (*vertices_with_ghost_neighbors)[cell->vertex_index(p4est_face_corners[sides[i].face][(limit - 1)^j])]
+                                .insert (subdomain_ids[k]);
+                              }
+                            else
+                              {
+                                (*vertices_with_ghost_neighbors)[cell->vertex_index(p8est_face_corners[sides[i].face][(limit - 1)^j])]
+                                .insert (subdomain_ids[k]);
+                              }
+                          }
+                      }
                   }
-                }
               }
-            }
           }
-        }
 
         subids->elem_count = 0;
       }
@@ -3347,8 +3376,8 @@ namespace parallel
     void
     Triangulation<1,1>::
     fill_vertices_with_ghost_neighbors
-      (std::map >
-       &vertices_with_ghost_neighbors)
+    (std::map >
+     &vertices_with_ghost_neighbors)
     {
       Assert (false, ExcNotImplemented());
     }
@@ -3357,8 +3386,8 @@ namespace parallel
     void
     Triangulation<1,2>::
     fill_vertices_with_ghost_neighbors
-      (std::map >
-       &vertices_with_ghost_neighbors)
+    (std::map >
+     &vertices_with_ghost_neighbors)
     {
       Assert (false, ExcNotImplemented());
     }
@@ -3367,8 +3396,8 @@ namespace parallel
     void
     Triangulation<1,3>::
     fill_vertices_with_ghost_neighbors
-      (std::map >
-       &vertices_with_ghost_neighbors)
+    (std::map >
+     &vertices_with_ghost_neighbors)
     {
       Assert (false, ExcNotImplemented());
     }
@@ -3381,8 +3410,8 @@ namespace parallel
     void
     Triangulation<2,2>::
     fill_vertices_with_ghost_neighbors
-      (std::map >
-       &vertices_with_ghost_neighbors)
+    (std::map >
+     &vertices_with_ghost_neighbors)
     {
       struct find_ghosts<2,2> fg;
 
@@ -3404,8 +3433,8 @@ namespace parallel
     void
     Triangulation<2,3>::
     fill_vertices_with_ghost_neighbors
-      (std::map >
-       &vertices_with_ghost_neighbors)
+    (std::map >
+     &vertices_with_ghost_neighbors)
     {
       struct find_ghosts<2,3> fg;
 
@@ -3427,8 +3456,8 @@ namespace parallel
     void
     Triangulation<3,3>::
     fill_vertices_with_ghost_neighbors
-      (std::map >
-       &vertices_with_ghost_neighbors)
+    (std::map >
+     &vertices_with_ghost_neighbors)
     {
       struct find_ghosts<3,3> fg;
 
@@ -3452,43 +3481,43 @@ namespace parallel
     template
     void
     Triangulation::add_periodicity
-      (const std::vector >&
-        periodicity_vector)
+    (const std::vector > &
+     periodicity_vector)
     {
-      #if DEAL_II_P4EST_VERSION_GTE(0,3,4,1)
+#if DEAL_II_P4EST_VERSION_GTE(0,3,4,1)
       Assert (triangulation_has_content == true,
               ExcMessage ("The triangulation is empty!"));
       Assert (this->n_levels() == 1,
               ExcMessage ("The triangulation is refined!"));
-      
+
       typedef std::vector >
-        FaceVector;
+      FaceVector;
       typename FaceVector::const_iterator it, periodic_end;
       it = periodicity_vector.begin();
       periodic_end = periodicity_vector.end();
-      
+
       for (; itcell[0];
-        const cell_iterator second_cell = it->cell[1];
-        const unsigned int face_left = it->face_idx[0];
-        const unsigned int face_right = it->face_idx[1];
-        
-        //respective cells of the matching faces in p4est
-        const unsigned int tree_left
-          = coarse_cell_to_p4est_tree_permutation[std::distance(this->begin(),
-                                                                first_cell)];
-        const unsigned int tree_right
-          = coarse_cell_to_p4est_tree_permutation[std::distance(this->begin(),
-                                                                second_cell)];
-
-        //TODO Add support for non default orientation.
-        Assert(it->orientation == 1,
-               ExcMessage("Found a face match with non standard orientation. "
-                          "This function is only suitable for meshes with "
-                          "cells in default orientation"));
-
-        dealii::internal::p4est::functions::
+        {
+          const cell_iterator first_cell = it->cell[0];
+          const cell_iterator second_cell = it->cell[1];
+          const unsigned int face_left = it->face_idx[0];
+          const unsigned int face_right = it->face_idx[1];
+
+          //respective cells of the matching faces in p4est
+          const unsigned int tree_left
+            = coarse_cell_to_p4est_tree_permutation[std::distance(this->begin(),
+                                                                  first_cell)];
+          const unsigned int tree_right
+            = coarse_cell_to_p4est_tree_permutation[std::distance(this->begin(),
+                                                                  second_cell)];
+
+          //TODO Add support for non default orientation.
+          Assert(it->orientation == 1,
+                 ExcMessage("Found a face match with non standard orientation. "
+                            "This function is only suitable for meshes with "
+                            "cells in default orientation"));
+
+          dealii::internal::p4est::functions::
           connectivity_join_faces (connectivity,
                                    tree_left,
                                    tree_right,
@@ -3496,50 +3525,50 @@ namespace parallel
                                    face_right,
                                    /* orientation */ 0);
 
-        /* The orientation parameter above describes the difference between
-         * the cell on the left and the cell on the right would number of the
-         * corners of the face.  In the periodic domains most users will want,
-         * this orientation will be 0, i.e. the two cells would number the
-         * corners the same way.  More exotic periodicity, like moebius strips
-         * or converting an unstructured quad/hex mesh into a periodic domain,
-         * are not supported right now, and undefined behavior will occur if
-         * users try to make them periodic.  This may be addressed at a later
-         * date.
-         */
-      }
-      
-      
+          /* The orientation parameter above describes the difference between
+           * the cell on the left and the cell on the right would number of the
+           * corners of the face.  In the periodic domains most users will want,
+           * this orientation will be 0, i.e. the two cells would number the
+           * corners the same way.  More exotic periodicity, like moebius strips
+           * or converting an unstructured quad/hex mesh into a periodic domain,
+           * are not supported right now, and undefined behavior will occur if
+           * users try to make them periodic.  This may be addressed at a later
+           * date.
+           */
+        }
+
+
       Assert(dealii::internal::p4est::functions::connectivity_is_valid
-               (connectivity) == 1, ExcInternalError());
-      
+             (connectivity) == 1, ExcInternalError());
+
       // now create a forest out of the connectivity data structure
       dealii::internal::p4est::functions::destroy (parallel_forest);
       parallel_forest
         = dealii::internal::p4est::functions::
-            new_forest (mpi_communicator,
-                        connectivity,
-                       /* minimum initial number of quadrants per tree */ 0,
-                       /* minimum level of upfront refinement */ 0,
-                       /* use uniform upfront refinement */ 1,
-                       /* user_data_size = */ 0,
-                       /* user_data_constructor = */ NULL,
-                       /* user_pointer */ this);
+          new_forest (mpi_communicator,
+                      connectivity,
+                      /* minimum initial number of quadrants per tree */ 0,
+                      /* minimum level of upfront refinement */ 0,
+                      /* use uniform upfront refinement */ 1,
+                      /* user_data_size = */ 0,
+                      /* user_data_constructor = */ NULL,
+                      /* user_pointer */ this);
 
 
       try
-      {
-        copy_local_forest_to_triangulation ();
-      }
+        {
+          copy_local_forest_to_triangulation ();
+        }
       catch (const typename Triangulation::DistortedCellList &)
-      {
-        // the underlying triangulation should not be checking for distorted
-        // cells
-        AssertThrow (false, ExcInternalError());
-      }
-      
-      #else
+        {
+          // the underlying triangulation should not be checking for distorted
+          // cells
+          AssertThrow (false, ExcInternalError());
+        }
+
+#else
       Assert(false, ExcMessage ("Need p4est version >= 0.3.4.1!"));
-      #endif
+#endif
     }
 
 
@@ -3707,12 +3736,13 @@ namespace parallel
       const std::vector perm = triangulation->get_p4est_tree_to_coarse_cell_permutation ();
       types::global_dof_index dealii_index = perm[treeidx];
 
-      for (i = 0; i < l; i++) {
-        typename dealii::Triangulation::cell_iterator cell (triangulation, i, dealii_index);
-        child_id = internal::p4est::functions::quadrant_ancestor_id (&quad, i + 1);
-        Assert (cell->has_children (), ExcMessage ("p4est quadrant does not correspond to a cell!"));
-        dealii_index = cell->child_index(child_id);
-      }
+      for (i = 0; i < l; i++)
+        {
+          typename dealii::Triangulation::cell_iterator cell (triangulation, i, dealii_index);
+          child_id = internal::p4est::functions::quadrant_ancestor_id (&quad, i + 1);
+          Assert (cell->has_children (), ExcMessage ("p4est quadrant does not correspond to a cell!"));
+          dealii_index = cell->child_index(child_id);
+        }
 
       typename dealii::Triangulation::cell_iterator out_cell (triangulation, l, dealii_index);
 
diff --git a/deal.II/source/dofs/dof_handler.cc b/deal.II/source/dofs/dof_handler.cc
index 119a9ee2c5..ce78e1d5d2 100644
--- a/deal.II/source/dofs/dof_handler.cc
+++ b/deal.II/source/dofs/dof_handler.cc
@@ -66,18 +66,17 @@ namespace internal
 namespace internal
 {
   template
-  std::string policy_to_string(const dealii::internal::DoFHandler::Policy::PolicyBase & policy)
+  std::string policy_to_string(const dealii::internal::DoFHandler::Policy::PolicyBase &policy)
   {
     std::string policy_name;
     if (dynamic_cast*>(&policy))
-        policy_name = "Policy::Sequential<";
+      policy_name = "Policy::Sequential<";
+    else if (dynamic_cast*>(&policy))
+      policy_name = "Policy::ParallelDistributed<";
     else
-      if (dynamic_cast*>(&policy))
-          policy_name = "Policy::ParallelDistributed<";
-      else
-        AssertThrow(false, ExcNotImplemented());
+      AssertThrow(false, ExcNotImplemented());
     policy_name += Utilities::int_to_string(dim)+
-            ","+Utilities::int_to_string(spacedim)+">";
+                   ","+Utilities::int_to_string(spacedim)+">";
     return policy_name;
   }
 
@@ -168,10 +167,10 @@ namespace internal
                           16*dof_handler.selected_fe->dofs_per_quad;
             break;
 
-            // the following numbers are not based on actual counting but by
-            // extrapolating the number sequences from the previous ones (for
-            // example, for dofs_per_vertex, the sequence above is 19, 21, 28,
-            // 30, 37, and is continued as follows):
+          // the following numbers are not based on actual counting but by
+          // extrapolating the number sequences from the previous ones (for
+          // example, for dofs_per_vertex, the sequence above is 19, 21, 28,
+          // 30, 37, and is continued as follows):
           case 9:
             max_couplings=39*dof_handler.selected_fe->dofs_per_vertex +
                           59*dof_handler.selected_fe->dofs_per_line +
diff --git a/deal.II/source/dofs/dof_handler_policy.cc b/deal.II/source/dofs/dof_handler_policy.cc
index 3db4b2104a..3e71bbd9c1 100644
--- a/deal.II/source/dofs/dof_handler_policy.cc
+++ b/deal.II/source/dofs/dof_handler_policy.cc
@@ -1216,7 +1216,7 @@ namespace internal
               //have us as a neighbor. In this case we and the children are active.
               if (dealii_cell->active())
                 {
-                  for (unsigned int f=0;f::faces_per_cell;++f)
+                  for (unsigned int f=0; f::faces_per_cell; ++f)
                     {
                       if (dealii_cell->at_boundary(f))
                         continue;
diff --git a/deal.II/source/dofs/dof_tools.cc b/deal.II/source/dofs/dof_tools.cc
index 51b5f2ec55..15beb53ea5 100644
--- a/deal.II/source/dofs/dof_tools.cc
+++ b/deal.II/source/dofs/dof_tools.cc
@@ -615,19 +615,19 @@ namespace DoFTools
                            :
                            (dim == 2 ?
                             (i<2*fe.dofs_per_vertex ? i : i+2*fe.dofs_per_vertex)
+                            :
+                            (dim == 3 ?
+                             (i<4*fe.dofs_per_vertex ?
+                              i
                               :
-                              (dim == 3 ?
-                               (i<4*fe.dofs_per_vertex ?
-                                i
-                                :
-                                (i<4*fe.dofs_per_vertex+4*fe.dofs_per_line ?
-                                 i+4*fe.dofs_per_vertex
-                                 :
-                                 i+4*fe.dofs_per_vertex+8*fe.dofs_per_line))
+                              (i<4*fe.dofs_per_vertex+4*fe.dofs_per_line ?
+                               i+4*fe.dofs_per_vertex
                                :
-                               numbers::invalid_unsigned_int)));
+                               i+4*fe.dofs_per_vertex+8*fe.dofs_per_line))
+                             :
+                             numbers::invalid_unsigned_int)));
                       if (fe.is_primitive (cell_index))
-                      {
+                        {
                           if (component_mask[fe.face_system_to_component_index(i).first]
                               == true)
                             selected_dofs.add_index (face_dof_indices[i]);
diff --git a/deal.II/source/dofs/dof_tools_constraints.cc b/deal.II/source/dofs/dof_tools_constraints.cc
index 4ef7047725..890f2ac036 100644
--- a/deal.II/source/dofs/dof_tools_constraints.cc
+++ b/deal.II/source/dofs/dof_tools_constraints.cc
@@ -844,7 +844,7 @@ namespace DoFTools
                       continue;
                     for (unsigned int dof=0; dof!=fe.dofs_per_line; ++dof)
                       dofs_on_children.push_back(
-                          this_face->child(child)->dof_index(dof, fe_index));
+                        this_face->child(child)->dof_index(dof, fe_index));
                   }
                 // note: can get fewer DoFs when we have artificial cells
                 Assert(dofs_on_children.size() <= n_dofs_on_children, ExcInternalError());
@@ -1058,8 +1058,8 @@ namespace DoFTools
                     if (cell->neighbor_child_on_subface (face, child)->is_artificial())
                       continue;
                     for (unsigned int dof=0; dof!=fe.dofs_per_quad; ++dof)
-                        dofs_on_children.push_back(
-                          this_face->child(child)->dof_index(dof, fe_index));
+                      dofs_on_children.push_back(
+                        this_face->child(child)->dof_index(dof, fe_index));
                   }
 
                 // note: can get fewer DoFs when we have artificial cells:
@@ -1580,13 +1580,13 @@ namespace DoFTools
                         // there, we have FESystem(FE_Q(1),FE_DGQ(0)) vs.
                         // FESystem(FE_Q(1),FE_DGQ(1)). neither of them
                         // dominates the other.
-			//
-			// a final possibility is that we have something like
-			// FESystem(FE_Q(1),FE_Q(1)) vs
-			// FESystem(FE_Q(1),FE_Nothing()), see
-			// hp/fe_nothing_18/19.
-			//
-			// in any case, the point is that it doesn't
+                        //
+                        // a final possibility is that we have something like
+                        // FESystem(FE_Q(1),FE_Q(1)) vs
+                        // FESystem(FE_Q(1),FE_Nothing()), see
+                        // hp/fe_nothing_18/19.
+                        //
+                        // in any case, the point is that it doesn't
                         // matter. there is nothing to do here.
                         break;
                       }
@@ -1730,28 +1730,30 @@ namespace DoFTools
           face_1->get_dof_indices(dofs_1, face_1_index);
           face_2->get_dof_indices(dofs_2, face_2_index);
 
-          for (unsigned int i=0; i < dofs_per_face; i++) {
-            if (dofs_1[i] == numbers::invalid_dof_index ||
-                dofs_2[i] == numbers::invalid_dof_index) {
-              /* If either of these faces have no indices, stop.  This is so
-               * that there is no attempt to match artificial cells of
-               * parallel distributed triangulations.
-               *
-               * While it seems like we ought to be able to avoid even calling
-               * set_periodicity_constraints for artificial faces, this
-               * situation can arise when a face that is being made periodic
-               * is only partially touched by the local subdomain.
-               * make_periodicity_constraints will be called recursively even
-               * for the section of the face that is not touched by the local
-               * subdomain.
-               *
-               * Until there is a better way to determine if the cells that
-               * neighbor a face are artificial, we simply test to see if the
-               * face does not have a valid dof initialization.
-               */
-              return;
+          for (unsigned int i=0; i < dofs_per_face; i++)
+            {
+              if (dofs_1[i] == numbers::invalid_dof_index ||
+                  dofs_2[i] == numbers::invalid_dof_index)
+                {
+                  /* If either of these faces have no indices, stop.  This is so
+                   * that there is no attempt to match artificial cells of
+                   * parallel distributed triangulations.
+                   *
+                   * While it seems like we ought to be able to avoid even calling
+                   * set_periodicity_constraints for artificial faces, this
+                   * situation can arise when a face that is being made periodic
+                   * is only partially touched by the local subdomain.
+                   * make_periodicity_constraints will be called recursively even
+                   * for the section of the face that is not touched by the local
+                   * subdomain.
+                   *
+                   * Until there is a better way to determine if the cells that
+                   * neighbor a face are artificial, we simply test to see if the
+                   * face does not have a valid dof initialization.
+                   */
+                  return;
+                }
             }
-          }
 
           // Well, this is a hack:
           //
@@ -1833,7 +1835,7 @@ namespace DoFTools
                       // orientation:
                       const unsigned int j =
                         cell_to_rotated_face_index[fe.face_to_cell_index(identity_constraint_target,
-                            0, /* It doesn't really matter, just assume
+                                                                         0, /* It doesn't really matter, just assume
                            * we're on the first face...
                            */
                                                                          face_orientation, face_flip, face_rotation)];
@@ -2030,16 +2032,15 @@ namespace DoFTools
             ExcMessage ("The boundary indicators b_id1 and b_id2 must be"
                         "different to denote different boundaries."));
 
-    typedef std::vector > FaceVector;
+    std::vector > matched_faces;
 
     // Collect matching periodic cells on the coarsest level:
-    FaceVector matched_faces =
-      GridTools::collect_periodic_faces(dof_handler, b_id1, b_id2,
-                                        direction, offset);
+    GridTools::collect_periodic_faces(dof_handler, b_id1, b_id2, direction,
+                                      matched_faces, offset);
 
     make_periodicity_constraints
-      (matched_faces, constraint_matrix, component_mask);
+    (matched_faces, constraint_matrix, component_mask);
   }
 
 
@@ -2081,16 +2082,15 @@ namespace DoFTools
     Assert(dim == space_dim,
            ExcNotImplemented());
 
-    typedef std::vector > FaceVector;
+    std::vector > matched_faces;
 
     // Collect matching periodic cells on the coarsest level:
-    const FaceVector matched_faces =
-      GridTools::collect_periodic_faces(dof_handler, b_id,
-                                        direction, offset);
+    GridTools::collect_periodic_faces(dof_handler, b_id, direction,
+                                      matched_faces, offset);
 
     make_periodicity_constraints
-      (matched_faces, constraint_matrix, component_mask);
+    (matched_faces, constraint_matrix, component_mask);
   }
 
 
@@ -2099,7 +2099,7 @@ namespace DoFTools
   void
   make_periodicity_constraints
   (const std::vector >
-    &periodic_faces,
+   &periodic_faces,
    dealii::ConstraintMatrix &constraint_matrix,
    const ComponentMask      &component_mask)
   {
@@ -2112,26 +2112,26 @@ namespace DoFTools
 
     // And apply the low level make_periodicity_constraints function to
     // every matching pair:
-    for(; it!=end_periodic; ++it)
-    {
-      typedef typename DH::face_iterator FaceIterator;
-      const FaceIterator face_1 = it->cell[0]->face(it->face_idx[0]);
-      const FaceIterator face_2 = it->cell[1]->face(it->face_idx[1]);
+    for (; it!=end_periodic; ++it)
+      {
+        typedef typename DH::face_iterator FaceIterator;
+        const FaceIterator face_1 = it->cell[0]->face(it->face_idx[0]);
+        const FaceIterator face_2 = it->cell[1]->face(it->face_idx[1]);
 
-      Assert(face_1->at_boundary() && face_2->at_boundary(),
-             ExcInternalError());
+        Assert(face_1->at_boundary() && face_2->at_boundary(),
+               ExcInternalError());
 
-      Assert (face_1 != face_2,
-              ExcInternalError());
+        Assert (face_1 != face_2,
+                ExcInternalError());
 
-      make_periodicity_constraints(face_1,
-                                   face_2,
-                                   constraint_matrix,
-                                   component_mask,
-                                   it->orientation[0],
-                                   it->orientation[1],
-                                   it->orientation[2]);
-    }
+        make_periodicity_constraints(face_1,
+                                     face_2,
+                                     constraint_matrix,
+                                     component_mask,
+                                     it->orientation[0],
+                                     it->orientation[1],
+                                     it->orientation[2]);
+      }
   }
 
 
@@ -2161,14 +2161,14 @@ namespace DoFTools
        */
       template 
       void compute_intergrid_weights_3 (
-          const typename dealii::DoFHandler::active_cell_iterator &cell,
-          const Assembler::Scratch                                              &,
-          Assembler::CopyData                                     ©_data,
-	  const unsigned int                                                     coarse_component,
-          const FiniteElement                                     &coarse_fe,
-          const InterGridMap >                 &coarse_to_fine_grid_map,
-          const std::vector >                            ¶meter_dofs,
-	  const std::vector                            &weight_mapping)
+        const typename dealii::DoFHandler::active_cell_iterator &cell,
+        const Assembler::Scratch &,
+        Assembler::CopyData                                     ©_data,
+        const unsigned int                                                     coarse_component,
+        const FiniteElement                                     &coarse_fe,
+        const InterGridMap >                 &coarse_to_fine_grid_map,
+        const std::vector >                            ¶meter_dofs,
+        const std::vector                            &weight_mapping)
       {
         // for each cell on the parameter grid: find out which degrees of
         // freedom on the fine grid correspond in which way to the degrees
@@ -2225,8 +2225,8 @@ namespace DoFTools
         // parameter grid cell
         cell->get_dof_indices (copy_data.parameter_dof_indices);
 
-	// reset the output array to a pristine state
-	copy_data.global_parameter_representation.clear ();
+        // reset the output array to a pristine state
+        copy_data.global_parameter_representation.clear ();
 
         // loop over all dofs on this cell and check whether they are
         // interesting for us
@@ -2240,7 +2240,7 @@ namespace DoFTools
                 = coarse_fe.system_to_component_index(local_dof).second;
 
               copy_data.global_parameter_representation.push_back(
-                  dealii::Vector (n_fine_dofs));
+                dealii::Vector (n_fine_dofs));
 
               // distribute the representation of
               // @p{local_parameter_dof} on the parameter grid cell
@@ -2260,9 +2260,9 @@ namespace DoFTools
        */
       template 
       void copy_intergrid_weights_3(const Assembler::CopyData                ©_data,
-				    const unsigned int                                      coarse_component,
+                                    const unsigned int                                      coarse_component,
                                     const FiniteElement                      &coarse_fe,
-				    const std::vector             &weight_mapping,
+                                    const std::vector             &weight_mapping,
                                     std::vector > &weights)
       {
         unsigned int pos = 0;
@@ -2270,7 +2270,7 @@ namespace DoFTools
           if (coarse_fe.system_to_component_index(local_dof).first
               ==
               coarse_component)
-          {
+            {
               // now that we've got the global representation of each
               // parameter dof, we've only got to clobber the non-zero
               // entries in that vector and store the result
@@ -2295,7 +2295,7 @@ namespace DoFTools
               // several threads operating on different intergrid
               // weights, have only one mutex for all of them
               for (types::global_dof_index i=0; i copy_data;
 
         WorkStream::run(coarse_grid.begin_active(),
-			coarse_grid.end(),
-			std_cxx1x::bind(&compute_intergrid_weights_3,
-					std_cxx1x::_1,
-					std_cxx1x::_2,
-					std_cxx1x::_3,
-					coarse_component,
-					std_cxx1x::cref(coarse_grid.get_fe()),
-					std_cxx1x::cref(coarse_to_fine_grid_map),
-					std_cxx1x::cref(parameter_dofs),
-					std_cxx1x::cref(weight_mapping)),
-			std_cxx1x::bind(©_intergrid_weights_3,
-					std_cxx1x::_1,
-					coarse_component,
-					std_cxx1x::cref(coarse_grid.get_fe()),
-					std_cxx1x::cref(weight_mapping),
-					std_cxx1x::ref(weights)),
-			scratch,
-			copy_data);
+                        coarse_grid.end(),
+                        std_cxx1x::bind(&compute_intergrid_weights_3,
+                                        std_cxx1x::_1,
+                                        std_cxx1x::_2,
+                                        std_cxx1x::_3,
+                                        coarse_component,
+                                        std_cxx1x::cref(coarse_grid.get_fe()),
+                                        std_cxx1x::cref(coarse_to_fine_grid_map),
+                                        std_cxx1x::cref(parameter_dofs),
+                                        std_cxx1x::cref(weight_mapping)),
+                        std_cxx1x::bind(©_intergrid_weights_3,
+                                        std_cxx1x::_1,
+                                        coarse_component,
+                                        std_cxx1x::cref(coarse_grid.get_fe()),
+                                        std_cxx1x::cref(weight_mapping),
+                                        std_cxx1x::ref(weights)),
+                        scratch,
+                        copy_data);
       }
 
 
@@ -2862,57 +2862,57 @@ namespace DoFTools
       if (!cell->is_artificial()
           &&
           cell->at_boundary ())
-      {
-        const FiniteElement &fe = cell->get_fe();
+        {
+          const FiniteElement &fe = cell->get_fe();
 
-        // get global indices of dofs on the cell
-        cell_dofs.resize (fe.dofs_per_cell);
-        cell->get_dof_indices (cell_dofs);
+          // get global indices of dofs on the cell
+          cell_dofs.resize (fe.dofs_per_cell);
+          cell->get_dof_indices (cell_dofs);
 
-        for (unsigned int face_no = 0; face_no < GeometryInfo::faces_per_cell;
-             ++face_no)
-          {
-            const typename DH::face_iterator face = cell->face(face_no);
-
-            // if face is on the boundary and satisfies the correct
-            // boundary id property
-            if (face->at_boundary ()
-                &&
-                ((boundary_indicator == numbers::invalid_boundary_id)
-                 ||
-                 (face->boundary_indicator() == boundary_indicator)))
-              {
-                // get indices and physical location on this face
-                face_dofs.resize (fe.dofs_per_face);
-                face->get_dof_indices (face_dofs, cell->active_fe_index());
+          for (unsigned int face_no = 0; face_no < GeometryInfo::faces_per_cell;
+               ++face_no)
+            {
+              const typename DH::face_iterator face = cell->face(face_no);
+
+              // if face is on the boundary and satisfies the correct
+              // boundary id property
+              if (face->at_boundary ()
+                  &&
+                  ((boundary_indicator == numbers::invalid_boundary_id)
+                   ||
+                   (face->boundary_indicator() == boundary_indicator)))
+                {
+                  // get indices and physical location on this face
+                  face_dofs.resize (fe.dofs_per_face);
+                  face->get_dof_indices (face_dofs, cell->active_fe_index());
 
-                // enter those dofs into the list that match the component
-                // signature.
-                for (unsigned int i=0; i::iterator it_index_on_cell
-                      = std::find (cell_dofs.begin(), cell_dofs.end(), face_dofs[i]);
-                    Assert (it_index_on_cell != cell_dofs.end(), ExcInvalidIterator());
-                    const unsigned int index_on_cell = std::distance(cell_dofs.begin(),
-                                                                     it_index_on_cell);
-                    const ComponentMask &nonzero_component_array
-                      = cell->get_fe().get_nonzero_components (index_on_cell);
-                    bool nonzero = false;
-                    for (unsigned int c=0; c::iterator it_index_on_cell
+                        = std::find (cell_dofs.begin(), cell_dofs.end(), face_dofs[i]);
+                      Assert (it_index_on_cell != cell_dofs.end(), ExcInvalidIterator());
+                      const unsigned int index_on_cell = std::distance(cell_dofs.begin(),
+                                                                       it_index_on_cell);
+                      const ComponentMask &nonzero_component_array
+                        = cell->get_fe().get_nonzero_components (index_on_cell);
+                      bool nonzero = false;
+                      for (unsigned int c=0; c::FiniteElement (
       system_to_component_table.resize(this->dofs_per_cell);
       face_system_to_component_table.resize(this->dofs_per_face);
       for (unsigned int j=0 ; jdofs_per_cell ; ++j)
-	system_to_component_table[j] = std::pair(0,j);
+        system_to_component_table[j] = std::pair(0,j);
       for (unsigned int j=0 ; jdofs_per_face ; ++j)
-	face_system_to_component_table[j] = std::pair(0,j);
+        face_system_to_component_table[j] = std::pair(0,j);
     }
-  
+
   for (unsigned int j=0 ; jdofs_per_cell ; ++j)
     system_to_base_table[j] = std::make_pair(std::make_pair(0U,0U),j);
   for (unsigned int j=0 ; jdofs_per_face ; ++j)
     face_system_to_base_table[j] = std::make_pair(std::make_pair(0U,0U),j);
-  
+
   // Fill with default value; may be changed by constructor of derived class.
   base_to_block_indices.reinit(1,1);
 
diff --git a/deal.II/source/fe/fe_dgq.cc b/deal.II/source/fe/fe_dgq.cc
index 07139d90a8..7b23f08e94 100644
--- a/deal.II/source/fe/fe_dgq.cc
+++ b/deal.II/source/fe/fe_dgq.cc
@@ -312,8 +312,8 @@ FE_DGQ::rotate_indices (std::vector &numbers,
     {
       switch (direction)
         {
-          // Rotate xy-plane
-          // counter-clockwise
+        // Rotate xy-plane
+        // counter-clockwise
         case 'z':
           for (unsigned int iz=0; iz<((dim>2) ? n:1); ++iz)
             for (unsigned int j=0; j::rotate_indices (std::vector &numbers,
                   numbers[l++] = k;
                 }
           break;
-          // Rotate xy-plane
-          // clockwise
+        // Rotate xy-plane
+        // clockwise
         case 'Z':
           for (unsigned int iz=0; iz<((dim>2) ? n:1); ++iz)
             for (unsigned int iy=0; iy::rotate_indices (std::vector &numbers,
                   numbers[k] = l++;
                 }
           break;
-          // Rotate yz-plane
-          // counter-clockwise
+        // Rotate yz-plane
+        // counter-clockwise
         case 'x':
           Assert (dim>2, ExcDimensionMismatch (dim,3));
           for (unsigned int iz=0; iz::rotate_indices (std::vector &numbers,
                   numbers[l++] = k;
                 }
           break;
-          // Rotate yz-plane
-          // clockwise
+        // Rotate yz-plane
+        // clockwise
         case 'X':
           Assert (dim>2, ExcDimensionMismatch (dim,3));
           for (unsigned int iz=0; iz::FE_FaceQ (const unsigned int degree)
         for (unsigned int iy=0; iy <= ((codim>1) ? this->degree : 0) ; ++iy)
           for (unsigned int ix=0; ix<=this->degree; ++ix)
             {
-	      Point p;
-	      
+              Point p;
+
               p(0) = points[ix][0];
               if (codim>1)
                 p(1) = points[iy][0];
@@ -164,7 +164,7 @@ get_subface_interpolation_matrix (const FiniteElement &x_source_fe
               (typename FiniteElement::
                ExcInterpolationNotImplemented ()));
 
-      // generate a quadrature with the unit face support points. 
+      // generate a quadrature with the unit face support points.
       const Quadrature face_quadrature (source_fe->get_unit_face_support_points ());
 
       // Rule of thumb for FP accuracy, that can be expected for a given
@@ -291,8 +291,8 @@ FE_FaceP::FE_FaceP (const unsigned int degree)
   :
   FE_PolyFace, dim, spacedim>
   (PolynomialSpace(Polynomials::Legendre::generate_complete_basis(degree)),
-  FiniteElementData(get_dpo_vector(degree), 1, degree, FiniteElementData::L2),
-  std::vector(1,true))
+   FiniteElementData(get_dpo_vector(degree), 1, degree, FiniteElementData::L2),
+   std::vector(1,true))
 {}
 
 
@@ -437,7 +437,7 @@ get_subface_interpolation_matrix (const FiniteElement &x_source_fe
       const double eps = 2e-13*(this->degree+1)*(dim-1);
 
       FullMatrix mass (face_quadrature.size(), source_fe->dofs_per_face);
- 
+
       for (unsigned int k = 0; k < face_quadrature.size(); ++k)
         {
           const Point p =
@@ -462,9 +462,9 @@ get_subface_interpolation_matrix (const FiniteElement &x_source_fe
           for (unsigned int k = 0; k < face_quadrature.size(); ++k)
             {
               const Point p = numbers::invalid_unsigned_int ?
-                face_quadrature.point(k) :
-                GeometryInfo::child_to_cell_coordinates (face_quadrature.point(k),
-                                                                subface);
+                                     face_quadrature.point(k) :
+                                     GeometryInfo::child_to_cell_coordinates (face_quadrature.point(k),
+                                         subface);
               v_in(k) = this->poly_space.compute_value(i, p);
             }
           const double result = H.least_squares(v_out, v_in);
diff --git a/deal.II/source/fe/fe_q_base.cc b/deal.II/source/fe/fe_q_base.cc
index 40096c09cf..10db7b4839 100644
--- a/deal.II/source/fe/fe_q_base.cc
+++ b/deal.II/source/fe/fe_q_base.cc
@@ -606,10 +606,10 @@ get_subface_interpolation_matrix (const FiniteElement &x_source_fe
 // be done for the face orientation flag in 3D.
       const Quadrature subface_quadrature
         = subface == numbers::invalid_unsigned_int
-        ?
-        QProjector::project_to_face (quad_face_support, 0)
-        :
-        QProjector::project_to_subface (quad_face_support, 0, subface);
+          ?
+          QProjector::project_to_face (quad_face_support, 0)
+          :
+          QProjector::project_to_subface (quad_face_support, 0, subface);
       for (unsigned int i=0; idofs_per_face; ++i)
         {
           const Point &p = subface_quadrature.point (i);
@@ -995,35 +995,35 @@ face_to_cell_index (const unsigned int face_index,
       // face orientation, face flips, etc
       unsigned int adjusted_dof_index_on_line;
       switch (dim)
-      {
-      case 1:
-        Assert (false, ExcInternalError());
-
-      case 2:
-        // in 2d, only face_flip has a meaning. if it is set, consider
-        // dofs in reverse order
-        if (face_flip == false)
+        {
+        case 1:
+          Assert (false, ExcInternalError());
+
+        case 2:
+          // in 2d, only face_flip has a meaning. if it is set, consider
+          // dofs in reverse order
+          if (face_flip == false)
+            adjusted_dof_index_on_line = dof_index_on_line;
+          else
+            adjusted_dof_index_on_line = this->dofs_per_line - dof_index_on_line - 1;
+          break;
+
+        case 3:
+          // in 3d, things are difficult. someone will have to think
+          // about how this code here should look like, by drawing a bunch
+          // of pictures of how all the faces can look like with the various
+          // flips and rotations.
+          //
+          // that said, the Q2 case is easy enough to implement, as is the case
+          // where everything is in standard orientation
+          Assert ((this->dofs_per_line <= 1) ||
+                  ((face_orientation == true) &&
+                   (face_flip == false) &&
+                   (face_rotation == false)),
+                  ExcNotImplemented());
           adjusted_dof_index_on_line = dof_index_on_line;
-        else
-          adjusted_dof_index_on_line = this->dofs_per_line - dof_index_on_line - 1;
-        break;
-
-      case 3:
-        // in 3d, things are difficult. someone will have to think
-        // about how this code here should look like, by drawing a bunch
-        // of pictures of how all the faces can look like with the various
-        // flips and rotations.
-        //
-        // that said, the Q2 case is easy enough to implement, as is the case
-        // where everything is in standard orientation
-        Assert ((this->dofs_per_line <= 1) ||
-                ((face_orientation == true) &&
-                 (face_flip == false) &&
-                 (face_rotation == false)),
-                ExcNotImplemented());
-        adjusted_dof_index_on_line = dof_index_on_line;
-        break;
-      }
+          break;
+        }
 
       return (this->first_line_index
               + GeometryInfo::face_to_cell_lines(face, face_line,
diff --git a/deal.II/source/fe/fe_system.cc b/deal.II/source/fe/fe_system.cc
index 2b5e504a5e..bf70ec76a6 100644
--- a/deal.II/source/fe/fe_system.cc
+++ b/deal.II/source/fe/fe_system.cc
@@ -48,7 +48,7 @@ namespace
 template 
 FESystem::InternalData::InternalData(const unsigned int n_base_elements,
                                                    const bool         compute_hessians)
-:
+  :
   compute_hessians (compute_hessians),
   base_fe_datas(n_base_elements),
   base_fe_values_datas(n_base_elements)
@@ -765,11 +765,11 @@ face_to_cell_index (const unsigned int face_dof_index,
 
   const unsigned int
   base_face_to_cell_index
-  = this->base_element(face_base_index.first.first).face_to_cell_index (face_base_index.second,
-                                                                        face,
-                                                                        face_orientation,
-                                                                        face_flip,
-                                                                        face_rotation);
+    = this->base_element(face_base_index.first.first).face_to_cell_index (face_base_index.second,
+        face,
+        face_orientation,
+        face_flip,
+        face_rotation);
 
   // it would be nice if we had a base_to_system_index function, but
   // all that exists is a component_to_system_index function. we can't do
diff --git a/deal.II/source/fe/fe_tools.cc b/deal.II/source/fe/fe_tools.cc
index 4ba84c5a43..b72d5a6671 100644
--- a/deal.II/source/fe/fe_tools.cc
+++ b/deal.II/source/fe/fe_tools.cc
@@ -390,7 +390,7 @@ namespace FETools
           start_indices[i] = k;
           k += block_data[i];
         }
-    
+
     for (unsigned int i=0; i
diff --git a/deal.II/source/fe/fe_tools_interpolate.cc b/deal.II/source/fe/fe_tools_interpolate.cc
index a1983255a1..eee88ee74b 100644
--- a/deal.II/source/fe/fe_tools_interpolate.cc
+++ b/deal.II/source/fe/fe_tools_interpolate.cc
@@ -52,9 +52,9 @@ DEAL_II_NAMESPACE_OPEN
 namespace FETools
 {
   template  class DH1,
-           template  class DH2,
-           class InVector, class OutVector>
+            template  class DH1,
+            template  class DH2,
+            class InVector, class OutVector>
   void
   interpolate(const DH1 &dof1,
               const InVector           &u1,
@@ -69,9 +69,9 @@ namespace FETools
 
 
   template  class DH1,
-           template  class DH2,
-           class InVector, class OutVector>
+            template  class DH1,
+            template  class DH2,
+            class InVector, class OutVector>
   void
   interpolate (const DH1 &dof1,
                const InVector           &u1,
@@ -314,8 +314,8 @@ namespace FETools
 
 
   template  class DH,
-           class InVector, class OutVector, int spacedim>
+            template  class DH,
+            class InVector, class OutVector, int spacedim>
   void
   back_interpolate(const DH            &dof1,
                    const InVector           &u1,
diff --git a/deal.II/source/fe/fe_values.cc b/deal.II/source/fe/fe_values.cc
index 23534addf4..639d937fd9 100644
--- a/deal.II/source/fe/fe_values.cc
+++ b/deal.II/source/fe/fe_values.cc
@@ -15,6 +15,7 @@
 // ---------------------------------------------------------------------
 
 #include 
+#include 
 #include 
 #include 
 #include 
@@ -3256,15 +3257,33 @@ void
 FEValuesBase::check_cell_similarity
 (const typename Triangulation::cell_iterator &cell)
 {
-  // case that there has not been any cell
-  // before
+  // Unfortunately, the detection of simple geometries with CellSimilarity is
+  // sensitive to the first cell detected. When doing this with multiple
+  // threads, each thread will get its own scratch data object with an
+  // FEValues object in the implementation framework from late 2013, which is
+  // initialized to the first cell the thread sees. As this number might
+  // different between different runs (after all, the tasks are scheduled
+  // dynamically onto threads), this slight deviation leads to difference in
+  // roundoff errors that propagate through the program. Therefore, we need to
+  // disable CellSimilarity in case there is more than one thread in the
+  // problem. This will likely not affect many MPI test cases as there
+  // multithreading is disabled on default, but in many other situations
+  // because we rarely explicitly set the number of threads.
+  //
+  // TODO: Is it reasonable to introduce a flag "unsafe" in the constructor of
+  // FEValues to re-enable this feature?
+  if (multithread_info.n_threads() > 1)
+    {
+      cell_similarity = CellSimilarity::none;
+      return;
+    }
+
+  // case that there has not been any cell before
   if (this->present_cell.get() == 0)
     cell_similarity = CellSimilarity::none;
   else
-    // in MappingQ, data can have been
-    // modified during the last call. Then,
-    // we can't use that data on the new
-    // cell.
+    // in MappingQ, data can have been modified during the last call. Then, we
+    // can't use that data on the new cell.
     if (cell_similarity == CellSimilarity::invalid_next_cell)
       cell_similarity = CellSimilarity::none;
     else
@@ -3282,8 +3301,7 @@ FEValuesBase::check_cell_similarity
           != cell->direction_flag() )
         cell_similarity =  CellSimilarity::inverted_translation;
     }
-  // TODO: here, one could implement other
-  // checks for similarity, e.g. for
+  // TODO: here, one could implement other checks for similarity, e.g. for
   // children of a parallelogram.
 }
 
@@ -3393,7 +3411,7 @@ namespace
         (typeid(*present_cell.get()) == typeid(Type)))
       {
         // call destructor of the old object
-        static_cast(present_cell.get())->~Type();
+        static_cast(present_cell.get())->~Type();
 
         // then construct a new object in-place
         new(const_cast(static_cast(present_cell.get()))) Type(new_cell);
diff --git a/deal.II/source/fe/mapping_q.cc b/deal.II/source/fe/mapping_q.cc
index c653133dfd..6b46cd43b7 100644
--- a/deal.II/source/fe/mapping_q.cc
+++ b/deal.II/source/fe/mapping_q.cc
@@ -134,7 +134,7 @@ MappingQ::MappingQ (const unsigned int p,
   // Qp mapping of cells at the boundary.
   const QGaussLobatto<1> points(degree+1);
   tensor_pols = new TensorProductPolynomials
-    (Polynomials::generate_complete_Lagrange_basis(points.get_points()));
+  (Polynomials::generate_complete_Lagrange_basis(points.get_points()));
   Assert (n_shape_functions==tensor_pols->n(),
           ExcInternalError());
   Assert(n_inner+n_outer==n_shape_functions, ExcInternalError());
@@ -468,8 +468,8 @@ MappingQ::set_laplace_on_quad_vector(Table<2,double> &loqvs) const
   double const *loqv_ptr=0;
   switch (degree)
     {
-      // for degree==1, we shouldn't have to compute any support points, since
-      // all of them are on the vertices
+    // for degree==1, we shouldn't have to compute any support points, since
+    // all of them are on the vertices
 
     case 2:
     {
@@ -606,14 +606,14 @@ MappingQ::compute_laplace_vector(Table<2,double> &lvs) const
   for (unsigned int point=0; point::compute_laplace_vector(Table<2,double> &lvs) const
   for (unsigned int point=0; point S_1(n_inner);
   S_1.invert(S);
diff --git a/deal.II/source/fe/mapping_q1.cc b/deal.II/source/fe/mapping_q1.cc
index 69537d32b5..4696547bd2 100644
--- a/deal.II/source/fe/mapping_q1.cc
+++ b/deal.II/source/fe/mapping_q1.cc
@@ -1683,7 +1683,7 @@ transform_real_to_unit_cell_internal
   // early out if we already have our point
   if (f.square() < 1e-24 * cell->diameter() * cell->diameter())
     return p_unit;
-  
+
   // we need to compare the position of the computed p(x) against the given
   // point 'p'. We will terminate the iteration and return 'x' if they are
   // less than eps apart. The question is how to choose eps -- or, put maybe
@@ -1727,7 +1727,7 @@ transform_real_to_unit_cell_internal
 #ifdef DEBUG_TRANSFORM_REAL_TO_UNIT_CELL
       std::cout << "Newton iteration " << newton_iteration << std::endl;
 #endif
-      
+
       // f'(x)
       Tensor<2,spacedim> df;
       for (unsigned int k=0; k 0.05)
             step_length /= 2;
           else
-	    AssertThrow (false,
-			 (typename Mapping::ExcTransformationFailed()));
+            AssertThrow (false,
+                         (typename Mapping::ExcTransformationFailed()));
         }
       while (true);
 
       ++newton_iteration;
       if (newton_iteration > newton_iteration_limit)
-	AssertThrow (false,
-		     (typename Mapping::ExcTransformationFailed()));
+        AssertThrow (false,
+                     (typename Mapping::ExcTransformationFailed()));
       last_f_weighted_norm = (df_inverse * f).norm();
     }
   while (last_f_weighted_norm > eps);
-    
+
   return p_unit;
 }
 
diff --git a/deal.II/source/grid/grid_generator.cc b/deal.II/source/grid/grid_generator.cc
index 158ef79423..0fc9d742a3 100644
--- a/deal.II/source/grid/grid_generator.cc
+++ b/deal.II/source/grid/grid_generator.cc
@@ -731,12 +731,12 @@ namespace GridGenerator
 
     switch (dim)
       {
-        // A line (1d parallelepiped)
+      // A line (1d parallelepiped)
       case 1:
         vertices[1] = corners[0];
         break;
 
-        // A parallelogram (2d parallelepiped)
+      // A parallelogram (2d parallelepiped)
       case 2:
         // assign corners to vertices:
         vertices[1] = corners[0];
@@ -746,7 +746,7 @@ namespace GridGenerator
         vertices[3] = vertices[1] + vertices[2];
         break;
 
-        // A parallelepiped (3d parallelepiped)
+      // A parallelepiped (3d parallelepiped)
       case 3:
         // assign corners to vertices:
         vertices[1] = corners[0];
@@ -2500,8 +2500,8 @@ namespace GridGenerator
     // Determine number of cells and vertices
     const unsigned int
     n_cells = static_cast(std::ceil (half_length /
-                                                    std::max (radius_0,
-                                                              radius_1)));
+                                                   std::max (radius_0,
+                                                             radius_1)));
     const unsigned int n_vertices = 4 * (n_cells + 1);
     std::vector > vertices_tmp(n_vertices);
 
diff --git a/deal.II/source/grid/grid_in.cc b/deal.II/source/grid/grid_in.cc
index e61de813d6..c41f034ca3 100644
--- a/deal.II/source/grid/grid_in.cc
+++ b/deal.II/source/grid/grid_in.cc
@@ -63,7 +63,7 @@ void GridIn::read_vtk(std::istream &in)
   text[2] = "ASCII";
   text[3] = "DATASET UNSTRUCTURED_GRID";
 
-  for(unsigned int i = 0;i < 4;i++)//Checking for the match between initial strings/text in the file.
+  for (unsigned int i = 0; i < 4; i++) //Checking for the match between initial strings/text in the file.
     {
       getline(in,line);
 
@@ -95,15 +95,15 @@ void GridIn::read_vtk(std::istream &in)
       in>>no_vertices;// taking the no. of vertices
       in.ignore(256, '\n');//ignoring the number beside the total no. of points.
 
-      for(unsigned int count = 0; count < no_vertices; count++)//loop to read three values till the no . vertices is satisfied
+      for (unsigned int count = 0; count < no_vertices; count++) //loop to read three values till the no . vertices is satisfied
         {
-	  // VTK format always specifies vertex coordinates with 3 components
-	  Point<3> x;
+          // VTK format always specifies vertex coordinates with 3 components
+          Point<3> x;
           in >> x(0) >> x(1) >> x(2);
 
           vertices.push_back(Point());
-	  for (unsigned int d=0; d::read_vtk(std::istream &in)
 
       if (dim == 3)
         {
-          for(unsigned int count = 0; count < total_cells; count++)
+          for (unsigned int count = 0; count < total_cells; count++)
             {
               in>>type;
 
-              if(type == 8)
+              if (type == 8)
                 {
 
                   cells.push_back(CellData());
 
-                  for(j = 0; j < type; j++)//loop to feed data
+                  for (j = 0; j < type; j++) //loop to feed data
                     in >> cells.back().vertices[j];
 
                   cells.back().material_id = 0;
 
-                  for(j = 0; j < type; j++)//loop to feed the data of the vertices to the cell
+                  for (j = 0; j < type; j++) //loop to feed the data of the vertices to the cell
                     {
                       cells.back().vertices[j] = vertex_indices[cells.back().vertices[j]];
                     }
@@ -171,12 +171,12 @@ void GridIn::read_vtk(std::istream &in)
                   subcelldata.boundary_quads.push_back(CellData<2>());
                   int no_quads = 0;
 
-                  for(j = 0; j < type; j++)//loop to feed the data to the boundary
+                  for (j = 0; j < type; j++) //loop to feed the data to the boundary
                     {
                       in >> subcelldata.boundary_quads.back().vertices[j];
                     }
                   subcelldata.boundary_quads.back().material_id = 0;
-                  for(unsigned int j = 0; j < type; j++)
+                  for (unsigned int j = 0; j < type; j++)
                     {
                       subcelldata.boundary_quads.back().vertices[j] = vertex_indices[subcelldata.boundary_quads.back().vertices[j]];
                     }
@@ -192,20 +192,20 @@ void GridIn::read_vtk(std::istream &in)
 
       else if (dim == 2)
         {
-          for(unsigned int count = 0; count < total_cells; count++)
+          for (unsigned int count = 0; count < total_cells; count++)
             {
               in>>type;
 
-              if(type == 4)
+              if (type == 4)
                 {
                   cells.push_back(CellData());
 
-                  for(j = 0; j < type; j++)//loop to feed data
+                  for (j = 0; j < type; j++) //loop to feed data
                     in >> cells.back().vertices[j];
 
                   cells.back().material_id = 0;
 
-                  for(j = 0; j < type; j++)//loop to feed the data of the vertices to the cell
+                  for (j = 0; j < type; j++) //loop to feed the data of the vertices to the cell
                     {
                       cells.back().vertices[j] = vertex_indices[cells.back().vertices[j]];
                     }
@@ -213,19 +213,19 @@ void GridIn::read_vtk(std::istream &in)
                   no_cells++;
                 }
 
-              else if(type == 2)
+              else if (type == 2)
                 {
                   //If this is encountered, the pointer comes out of the loop
                   //and starts processing boundaries.
                   subcelldata.boundary_lines.push_back(CellData<1>());
                   int no_lines = 0;
 
-                  for(j = 0; j < type; j++)//loop to feed the data to the boundary
+                  for (j = 0; j < type; j++) //loop to feed the data to the boundary
                     {
                       in >> subcelldata.boundary_lines.back().vertices[j];
                     }
                   subcelldata.boundary_lines.back().material_id = 0;
-                  for(unsigned int j = 0; j < type; j++)
+                  for (unsigned int j = 0; j < type; j++)
                     {
                       subcelldata.boundary_lines.back().vertices[j] = vertex_indices[subcelldata.boundary_lines.back().vertices[j]];
                     }
@@ -250,10 +250,10 @@ void GridIn::read_vtk(std::istream &in)
         {
           in.ignore(256, '\n');
 
-          while(!in.eof())
+          while (!in.eof())
             {
               in>>keyword_3;
-              if(std::strcmp(keyword_3,"12") && std::strcmp(keyword_3,"9"))
+              if (std::strcmp(keyword_3,"12") && std::strcmp(keyword_3,"9"))
                 {
                   break;
                 }
@@ -274,11 +274,11 @@ void GridIn::read_vtk(std::istream &in)
 
           in.ignore(256, '\n');
 
-          for(unsigned int i = 0;i < 2;i++)
+          for (unsigned int i = 0; i < 2; i++)
             {
               getline(in,linenew);
 
-              if(linenew.compare(textnew[i]) == 0)
+              if (linenew.compare(textnew[i]) == 0)
                 {}
               else
                 AssertThrow (false,
@@ -286,7 +286,7 @@ void GridIn::read_vtk(std::istream &in)
                                          textnew[i] + "> section"));
             }
 
-          for(unsigned int i = 0; i < no_cells; i++)//assigning IDs to cells.
+          for (unsigned int i = 0; i < no_cells; i++) //assigning IDs to cells.
             {
               int id;
               in>>id;
@@ -295,7 +295,7 @@ void GridIn::read_vtk(std::istream &in)
 
           if (dim == 3)
             {
-              for(unsigned int i = 0; i < no_quads; i++)//assigning IDs to bounds.
+              for (unsigned int i = 0; i < no_quads; i++) //assigning IDs to bounds.
                 {
                   int id;
                   in>>id;
@@ -304,7 +304,7 @@ void GridIn::read_vtk(std::istream &in)
             }
           else if (dim == 2)
             {
-              for(unsigned int i = 0; i < no_lines; i++)//assigning IDs to bounds.
+              for (unsigned int i = 0; i < no_lines; i++) //assigning IDs to bounds.
                 {
                   int id;
                   in>>id;
@@ -323,9 +323,9 @@ void GridIn::read_vtk(std::istream &in)
                                         cells,
                                         subcelldata);
 
-      if(dim == spacedim)
+      if (dim == spacedim)
         GridReordering::invert_all_cells_of_negative_grid(vertices,
-                                                                         cells);
+            cells);
 
       GridReordering::reorder_cells(cells);
       tria->create_triangulation_compatibility(vertices,
@@ -2642,8 +2642,8 @@ void GridIn::read (std::istream &in,
       return;
 
     case vtk:
-         read_vtk (in);
-         return;
+      read_vtk (in);
+      return;
 
     case unv:
       read_unv (in);
diff --git a/deal.II/source/grid/grid_out.cc b/deal.II/source/grid/grid_out.cc
index c110dcfdc8..edf857458e 100644
--- a/deal.II/source/grid/grid_out.cc
+++ b/deal.II/source/grid/grid_out.cc
@@ -687,7 +687,7 @@ void GridOut::write_dx (const Triangulation &tria,
 
   for (unsigned int i=0; i::invert_all_cells_of_negative_grid(const std::vector
+void
+GridReordering<1,3>::reorder_cells (std::vector > &)
+{
+  // there should not be much to do
+  // in 1d...
+}
+
+
+template<>
+void
+GridReordering<1,3>::invert_all_cells_of_negative_grid(const std::vector > &,
+                                                       std::vector > &)
+{
+  // nothing to be done in 1d
+}
+
+
 namespace internal
 {
   namespace GridReordering2d
diff --git a/deal.II/source/grid/grid_tools.cc b/deal.II/source/grid/grid_tools.cc
index f62d935d63..12be439467 100644
--- a/deal.II/source/grid/grid_tools.cc
+++ b/deal.II/source/grid/grid_tools.cc
@@ -982,7 +982,7 @@ next_cell:
       }
 
     AssertThrow (best_cell.first.state() == IteratorState::valid,
-            ExcPointNotFound(p));
+                 ExcPointNotFound(p));
 
     return best_cell;
   }
@@ -1105,7 +1105,7 @@ next_cell:
       }
 
     AssertThrow (best_cell.first.state() == IteratorState::valid,
-            ExcPointNotFound(p));
+                 ExcPointNotFound(p));
 
     return best_cell;
   }
@@ -1136,7 +1136,7 @@ next_cell:
     // pointer/index clashes when saving/restoring them. The following approach
     // works, but this map can get quite big. Not sure about more efficient solutions.
     std::map< std::pair, unsigned int >
-      indexmap;
+    indexmap;
     unsigned int index = 0;
     for (typename Triangulation::active_cell_iterator
          cell = triangulation.begin_active();
@@ -1171,7 +1171,7 @@ next_cell:
               (cell->neighbor(f)->has_children() == false))
             {
               unsigned int other_index = indexmap.find(
-                  std::pair(cell->neighbor(f)->level(),cell->neighbor(f)->index()))->second;
+                                           std::pair(cell->neighbor(f)->level(),cell->neighbor(f)->index()))->second;
               cell_connectivity.add (index, other_index);
               cell_connectivity.add (other_index, index);
             }
@@ -2364,12 +2364,13 @@ next_cell:
    * Internally used in collect_periodic_faces
    */
   template
-  std::vector >
-  match_periodic_face_pairs 
-    (std::set > &pairs1,
-     std::set::type, unsigned int> > &pairs2,
-     const int direction,
-     const dealii::Tensor<1,CellIterator::AccessorType::space_dimension> &offset)
+  void
+  match_periodic_face_pairs
+  (std::set > &pairs1,
+   std::set::type, unsigned int> > &pairs2,
+   const int direction,
+   std::vector > &matched_pairs,
+   const dealii::Tensor<1,CellIterator::AccessorType::space_dimension> &offset)
   {
     static const int space_dim = CellIterator::AccessorType::space_dimension;
     Assert (0<=direction && direction > matched_faces;
+    unsigned int n_matches = 0;
 
     // Match with a complexity of O(n^2). This could be improved...
     std::bitset<3> orientation;
     typedef typename std::set
-       >::const_iterator PairIterator;
+     >::const_iterator PairIterator;
     for (PairIterator it1 = pairs1.begin(); it1 != pairs1.end(); ++it1)
       {
         for (PairIterator it2 = pairs2.begin(); it2 != pairs2.end(); ++it2)
@@ -2401,30 +2402,31 @@ next_cell:
                 // remove the matched cell in pairs2 to speed up the
                 // matching:
                 const PeriodicFacePair matched_face
-                  = {{cell1, cell2},{face_idx1, face_idx2}, orientation};
-                matched_faces.push_back(matched_face);
+                = {{cell1, cell2},{face_idx1, face_idx2}, orientation};
+                matched_pairs.push_back(matched_face);
                 pairs2.erase(it2);
+                ++n_matches;
                 break;
               }
           }
       }
 
-    AssertThrow (matched_faces.size() == pairs1.size() && pairs2.size() == 0,
+    //Assure that all faces are matched
+    AssertThrow (n_matches == pairs1.size() && pairs2.size() == 0,
                  ExcMessage ("Unmatched faces on periodic boundaries"));
-
-    return matched_faces;
   }
 
 
 
   template
-  std::vector >
+  void
   collect_periodic_faces
-    (const CONTAINER                 &container,
-     const types::boundary_id b_id1,
-     const types::boundary_id b_id2,
-     const int                direction,
-     const dealii::Tensor<1,CONTAINER::space_dimension> &offset)
+  (const CONTAINER          &container,
+   const types::boundary_id b_id1,
+   const types::boundary_id b_id2,
+   const int                direction,
+   std::vector > &matched_pairs,
+   const dealii::Tensor<1,CONTAINER::space_dimension> &offset)
   {
     static const int dim = CONTAINER::dimension;
     static const int space_dim = CONTAINER::space_dimension;
@@ -2444,18 +2446,18 @@ next_cell:
           {
             const typename CONTAINER::face_iterator face = cell->face(i);
             if (face->at_boundary() && face->boundary_indicator() == b_id1)
-            {
-              const std::pair pair1
-                = std::make_pair(cell, i);
-              pairs1.insert(pair1);
-            } 
+              {
+                const std::pair pair1
+                  = std::make_pair(cell, i);
+                pairs1.insert(pair1);
+              }
 
             if (face->at_boundary() && face->boundary_indicator() == b_id2)
-            {
-              const std::pair pair2
-                = std::make_pair(cell, i);
-              pairs2.insert(pair2);
-            }
+              {
+                const std::pair pair2
+                  = std::make_pair(cell, i);
+                pairs2.insert(pair2);
+              }
           }
       }
 
@@ -2463,17 +2465,19 @@ next_cell:
             ExcMessage ("Unmatched faces on periodic boundaries"));
 
     // and call match_periodic_face_pairs that does the actual matching:
-    return match_periodic_face_pairs(pairs1, pairs2, direction, offset);
+    match_periodic_face_pairs(pairs1, pairs2, direction, matched_pairs, offset);
   }
 
 
 
   template
-  std::vector >
-  collect_periodic_faces (const CONTAINER                 &container,
-                          const types::boundary_id b_id,
-                          const int                direction,
-                          const dealii::Tensor<1,CONTAINER::space_dimension> &offset)
+  void
+  collect_periodic_faces
+  (const CONTAINER          &container,
+   const types::boundary_id b_id,
+   const int                direction,
+   std::vector > &matched_pairs,
+   const dealii::Tensor<1,CONTAINER::space_dimension> &offset)
   {
     static const int dim = CONTAINER::dimension;
     static const int space_dim = CONTAINER::space_dimension;
@@ -2491,48 +2495,47 @@ next_cell:
 
     for (typename CONTAINER::cell_iterator cell = container.begin(0);
          cell != container.end(0); ++cell)
-         {
-           const typename CONTAINER::face_iterator face_1 = cell->face(2*direction);
-           const typename CONTAINER::face_iterator face_2 = cell->face(2*direction+1);
-
-           if (face_1->at_boundary() && face_1->boundary_indicator() == b_id)
-           {
-             const std::pair pair1
-               = std::make_pair(cell, 2*direction);
-             pairs1.insert(pair1);
-           }
-
-           if (face_2->at_boundary() && face_2->boundary_indicator() == b_id)
-           {
-             const std::pair pair2
-               = std::make_pair(cell, 2*direction+1);
-             pairs2.insert(pair2);
-           }
-         }
+      {
+        const typename CONTAINER::face_iterator face_1 = cell->face(2*direction);
+        const typename CONTAINER::face_iterator face_2 = cell->face(2*direction+1);
+
+        if (face_1->at_boundary() && face_1->boundary_indicator() == b_id)
+          {
+            const std::pair pair1
+              = std::make_pair(cell, 2*direction);
+            pairs1.insert(pair1);
+          }
+
+        if (face_2->at_boundary() && face_2->boundary_indicator() == b_id)
+          {
+            const std::pair pair2
+              = std::make_pair(cell, 2*direction+1);
+            pairs2.insert(pair2);
+          }
+      }
 
     Assert (pairs1.size() == pairs2.size(),
             ExcMessage ("Unmatched faces on periodic boundaries"));
 
-    // and call match_periodic_face_pairs that does the actual matching:
 
-    typedef std::vector >
-      FaceVector;
+#ifdef DEBUG
+    const unsigned int size_old = matched_pairs.size();
+#endif
 
-    FaceVector matching = match_periodic_face_pairs(pairs1, pairs2,
-                                                    direction, offset);
+    // and call match_periodic_face_pairs that does the actual matching:
+    match_periodic_face_pairs(pairs1, pairs2, direction, matched_pairs, offset);
 
 #ifdef DEBUG
-    for (typename FaceVector::iterator pairing = matching.begin();
-         pairing != matching.end(); ++pairing)
-    {
-      Assert(pairing->orientation == 1,
-             ExcMessage("Found a face match with non standard orientation. "
-                        "This function is only suitable for meshes with cells "
-                        "in default orientation"));
-    }
+    //check for standard orientation
+    const unsigned int size_new = matched_pairs.size();
+    for (unsigned int i = size_old; i < size_new; ++i)
+      {
+        Assert(matched_pairs[i].orientation == 1,
+               ExcMessage("Found a face match with non standard orientation. "
+                          "This function is only suitable for meshes with cells "
+                          "in default orientation"));
+      }
 #endif
-
-    return matching;
   }
 
 
diff --git a/deal.II/source/grid/grid_tools.inst.in b/deal.II/source/grid/grid_tools.inst.in
index 6e827d7111..8089c7dad6 100644
--- a/deal.II/source/grid/grid_tools.inst.in
+++ b/deal.II/source/grid/grid_tools.inst.in
@@ -311,21 +311,19 @@ for (X : TRIANGULATION_AND_DOFHANDLERS; deal_II_dimension : DIMENSIONS ; deal_II
     #if deal_II_dimension >= 2
 
       template
-      std::vector >
-      collect_periodic_faces
-                                 (const X &,
-                                  const types::boundary_id,
-                                  const types::boundary_id,
-                                  const int,
-                                  const Tensor<1,X::space_dimension> &);
+      void collect_periodic_faces (const X &,
+                                      const types::boundary_id,
+                                      const types::boundary_id,
+                                      const int,
+                                      std::vector > &,
+                                      const Tensor<1,X::space_dimension> &);
 
       template
-      std::vector >
-      collect_periodic_faces
-                                 (const X &,
-                                  const types::boundary_id,
-                                  const int,
-                                  const Tensor<1,X::space_dimension> &);
+      void collect_periodic_faces (const X &,
+                                      const types::boundary_id,
+                                      const int,
+                                      std::vector > &,
+                                      const Tensor<1,X::space_dimension> &);
 
     #endif
 
@@ -341,20 +339,22 @@ for (deal_II_dimension : DIMENSIONS ; deal_II_space_dimension : SPACE_DIMENSIONS
      namespace GridTools \{
 
       template
-      std::vector::cell_iterator> >
+      void
       collect_periodic_faces >
                                  (const parallel::distributed::Triangulation &,
                                   const types::boundary_id,
                                   const types::boundary_id,
                                   const int,
+                                  std::vector::cell_iterator> > &,
                                   const Tensor<1,parallel::distributed::Triangulation::space_dimension> &);
 
       template
-      std::vector::cell_iterator> >
+      void
       collect_periodic_faces >
                                  (const parallel::distributed::Triangulation &,
                                   const types::boundary_id,
                                   const int,
+                                  std::vector::cell_iterator> > &,
                                   const Tensor<1,parallel::distributed::Triangulation::space_dimension> &);
 
      \}
diff --git a/deal.II/source/grid/tria_boundary.cc b/deal.II/source/grid/tria_boundary.cc
index 56031cde91..d28ab418b9 100644
--- a/deal.II/source/grid/tria_boundary.cc
+++ b/deal.II/source/grid/tria_boundary.cc
@@ -262,7 +262,7 @@ get_line_support_points (const unsigned int n_intermediate_points) const
       if (points[n_intermediate_points].get() == 0)
         {
           std_cxx1x::shared_ptr >
-            quadrature (new QGaussLobatto<1>(n_intermediate_points+2));
+          quadrature (new QGaussLobatto<1>(n_intermediate_points+2));
           points[n_intermediate_points] = quadrature;
         }
     }
diff --git a/deal.II/source/grid/tria_objects.cc b/deal.II/source/grid/tria_objects.cc
index ff6b22a296..13eac3c181 100644
--- a/deal.II/source/grid/tria_objects.cc
+++ b/deal.II/source/grid/tria_objects.cc
@@ -192,14 +192,19 @@ namespace internal
                                        RefinementCase::no_refinement);
             }
 
+          // first reserve, then resize. Otherwise the std library can decide to allocate
+          // more entries.
+          boundary_or_material_id.reserve (new_size);
           boundary_or_material_id.resize (new_size);
 
+          user_data.reserve (new_size);
+          user_data.resize (new_size);
+
           manifold_id.reserve (new_size);
           manifold_id.insert (manifold_id.end(),
 			      new_size-manifold_id.size(),
 			      numbers::flat_manifold_id);
 
-          user_data.resize (new_size);
         }
 
       if (n_unused_singles==0)
@@ -271,10 +276,10 @@ namespace internal
                            4*new_size-children.size(),
                            -1);
 
-	  // for the following two fields, we know exactly how many elements
-	  // we need, so first reserve then resize (resize itself, at least
-	  // with some compiler libraries, appears to round up the size it
-	  // actually reserves)
+          // for the following two fields, we know exactly how many elements
+          // we need, so first reserve then resize (resize itself, at least
+          // with some compiler libraries, appears to round up the size it
+          // actually reserves)
           boundary_or_material_id.reserve (new_size);
           boundary_or_material_id.resize (new_size);
 
diff --git a/deal.II/source/hp/dof_handler.cc b/deal.II/source/hp/dof_handler.cc
index fa6d95cadd..85a37b7aac 100644
--- a/deal.II/source/hp/dof_handler.cc
+++ b/deal.II/source/hp/dof_handler.cc
@@ -543,7 +543,7 @@ namespace internal
             active_fe_backup(dof_handler.levels.size ());
             for (unsigned int level = 0; levelactive_fe_indices,
-			 active_fe_backup[level]);
+                         active_fe_backup[level]);
 
             // delete all levels and set them up
             // newly, since vectors are
@@ -870,7 +870,7 @@ namespace internal
             // non-invalid value later on
             dof_handler.faces->lines.dof_offsets
               = std::vector (dof_handler.tria->n_raw_lines(),
-                                             (unsigned int)(-1));
+                                           (unsigned int)(-1));
             dof_handler.faces->lines.dofs
               = std::vector (n_line_slots,
                                                       DoFHandler::invalid_dof_index);
@@ -1250,8 +1250,8 @@ namespace internal
               {
                 dof_handler.faces->quads.dof_offsets
                   = std::vector
-                                                         (dof_handler.tria->n_raw_quads(),
-                                                         (unsigned int)(-1));
+                    (dof_handler.tria->n_raw_quads(),
+                     (unsigned int)(-1));
                 dof_handler.faces->quads.dofs
                   = std::vector (n_quad_slots,
                                                           DoFHandler::invalid_dof_index);
@@ -1992,7 +1992,7 @@ namespace hp
             dofs_on_face.resize (dofs_per_face);
 
             cell->face(f)->get_dof_indices (dofs_on_face,
-	                                    cell->active_fe_index());
+                                            cell->active_fe_index());
             for (unsigned int i=0; i(),
                     ExcInternalError());
@@ -88,12 +88,12 @@ namespace internal
             // find the next cell active on this level
             unsigned int next_cell = cell+1;
             while ((next_cell(),
                     ExcInternalError());
@@ -172,12 +172,12 @@ namespace internal
             // find the next cell active on this level
             unsigned int next_cell = cell+1;
             while ((next_cell &indices) const
 {
   const unsigned int indices_size = indices.size();
-  const std::vector > * line_ptr;
+  const std::vector > *line_ptr;
   for (unsigned int i=0; i that can be passed to the
+// constructor of scratch_data (it won't allow one to be constructed in place).
+namespace internals
+{
+#define SCRATCH_INITIALIZER(Number,Name)                                \
+  ConstraintMatrixData::ScratchData scratch_data_initializer_##Name; \
+  template<> Threads::ThreadLocalStorage::ScratchData> \
+  ConstraintMatrixData::scratch_data(scratch_data_initializer_##Name)
+
+  SCRATCH_INITIALIZER(double,double);
+  SCRATCH_INITIALIZER(float,float);
+  SCRATCH_INITIALIZER(long double,ldouble);
+  SCRATCH_INITIALIZER(std::complex,cdouble);
+  SCRATCH_INITIALIZER(std::complex,cfloat);
+  SCRATCH_INITIALIZER(std::complex,cldouble);
+#undef SCRATCH_INITIALIZER
+}
+
+
 DEAL_II_NAMESPACE_CLOSE
diff --git a/deal.II/source/lac/full_matrix.inst.in b/deal.II/source/lac/full_matrix.inst.in
index f902a72697..7484bbd483 100644
--- a/deal.II/source/lac/full_matrix.inst.in
+++ b/deal.II/source/lac/full_matrix.inst.in
@@ -51,7 +51,7 @@ for (S1, S2 : REAL_SCALARS)
 
     template
       void FullMatrix::fill (
-	const FullMatrix&, types::global_dof_index, types::global_dof_index, types::global_dof_index, types::global_dof_index);
+	const FullMatrix&, size_type, size_type, size_type, size_type);
     template
       void FullMatrix::add (const S1, const FullMatrix&);
     template
@@ -63,12 +63,12 @@ for (S1, S2 : REAL_SCALARS)
 				    const S1, const FullMatrix&);
     template
       void FullMatrix::add (
-	const FullMatrix&, S1, types::global_dof_index, types::global_dof_index, types::global_dof_index, types::global_dof_index);
+	const FullMatrix&, S1, size_type, size_type, size_type, size_type);
     template
       void FullMatrix::Tadd (const S1, const FullMatrix&);
     template
       void FullMatrix::Tadd (
-	const FullMatrix&, S1, types::global_dof_index, types::global_dof_index, types::global_dof_index, types::global_dof_index);
+	const FullMatrix&, S1, size_type, size_type, size_type, size_type);
     template
       void FullMatrix::equ (const S1, const FullMatrix&);
     template
@@ -159,7 +159,7 @@ for (S1, S2 : COMPLEX_SCALARS)
   {
     template
       void FullMatrix::fill (
-	const FullMatrix&, types::global_dof_index, types::global_dof_index, types::global_dof_index, types::global_dof_index);
+	const FullMatrix&, size_type, size_type, size_type, size_type);
     template
       void FullMatrix::add (const S1, const FullMatrix&);
     template
@@ -171,13 +171,13 @@ for (S1, S2 : COMPLEX_SCALARS)
 				    const S1, const FullMatrix&);
     template
       void FullMatrix::add (
-	const FullMatrix&, S1, types::global_dof_index, types::global_dof_index, types::global_dof_index, types::global_dof_index);
+	const FullMatrix&, S1, size_type, size_type, size_type, size_type);
     template
       void FullMatrix::Tadd (const S1, const FullMatrix&);
     template
       void FullMatrix::Tadd (
-	const FullMatrix&, S1, types::global_dof_index, types::global_dof_index,
-  types::global_dof_index, types::global_dof_index);
+	const FullMatrix&, S1, size_type, size_type,
+  size_type, size_type);
     template
       void FullMatrix::equ (const S1, const FullMatrix&);
     template
diff --git a/deal.II/source/lac/matrix_out.cc b/deal.II/source/lac/matrix_out.cc
index 634d5dd7f9..abfe4f1116 100644
--- a/deal.II/source/lac/matrix_out.cc
+++ b/deal.II/source/lac/matrix_out.cc
@@ -19,9 +19,9 @@
 DEAL_II_NAMESPACE_OPEN
 
 
-MatrixOut::Options::Options (const bool      show_absolute_values,
-                             const size_type block_size,
-                             const bool      discontinuous)
+MatrixOut::Options::Options (const bool         show_absolute_values,
+                             const unsigned int block_size,
+                             const bool         discontinuous)
   :
   show_absolute_values (show_absolute_values),
   block_size (block_size),
diff --git a/deal.II/source/lac/slepc_solver.cc b/deal.II/source/lac/slepc_solver.cc
index 3fd24cd777..56cab3038d 100644
--- a/deal.II/source/lac/slepc_solver.cc
+++ b/deal.II/source/lac/slepc_solver.cc
@@ -159,12 +159,12 @@ namespace SLEPcWrappers
     // transformation and target the wanted eigenvalues
     if (transformation)
       {
-	// set transformation type if any
-	transformation->set_context (solver_data->eps);
-	
-	// set target eigenvalues to solve for
-	ierr = EPSSetTarget (solver_data->eps, target_eigenvalue);
-	AssertThrow (ierr == 0, ExcSLEPcError(ierr));
+        // set transformation type if any
+        transformation->set_context (solver_data->eps);
+
+        // set target eigenvalues to solve for
+        ierr = EPSSetTarget (solver_data->eps, target_eigenvalue);
+        AssertThrow (ierr == 0, ExcSLEPcError(ierr));
       }
 
     // set which portion of the eigenspectrum to solve for
@@ -195,7 +195,7 @@ namespace SLEPcWrappers
 
     // get number of converged eigenstates
     ierr = EPSGetConverged (solver_data->eps,
-                            reinterpret_cast(n_converged));
+                            reinterpret_cast(n_converged));
     AssertThrow (ierr == 0, ExcSLEPcError(ierr));
 
     PetscInt n_iterations   = 0;
@@ -211,10 +211,10 @@ namespace SLEPcWrappers
       // get the residual norm of the most extreme eigenvalue if and
       // only if at least one eigenvector has converged.
       if ((*n_converged)>0)
-	{
-	  ierr = EPSComputeResidualNorm (solver_data->eps, 0, &residual_norm);
-	  AssertThrow (ierr == 0, ExcSLEPcError(ierr));
-	}
+        {
+          ierr = EPSComputeResidualNorm (solver_data->eps, 0, &residual_norm);
+          AssertThrow (ierr == 0, ExcSLEPcError(ierr));
+        }
 
       // check the solver state
       const SolverControl::State state
diff --git a/deal.II/source/lac/solver_control.cc b/deal.II/source/lac/solver_control.cc
index 6eab7e4d4f..41ad2ca9eb 100644
--- a/deal.II/source/lac/solver_control.cc
+++ b/deal.II/source/lac/solver_control.cc
@@ -312,7 +312,7 @@ IterationNumberControl::IterationNumberControl(const unsigned int n,
                                                const bool m_log_history,
                                                const bool m_log_result)
   :
-  SolverControl (n, tolerance, m_log_history, m_log_result){}
+  SolverControl (n, tolerance, m_log_history, m_log_result) {}
 
 
 IterationNumberControl::~IterationNumberControl()
diff --git a/deal.II/source/lac/sparse_direct.cc b/deal.II/source/lac/sparse_direct.cc
index 4333b64fa1..77cdb3bd1f 100644
--- a/deal.II/source/lac/sparse_direct.cc
+++ b/deal.II/source/lac/sparse_direct.cc
@@ -514,7 +514,7 @@ SparseDirectMUMPS::~SparseDirectMUMPS ()
 template 
 void SparseDirectMUMPS::initialize_matrix (const Matrix &matrix)
 {
-    Assert(matrix.n() == matrix.m(), ExcMessage("Matrix needs to be square."));
+  Assert(matrix.n() == matrix.m(), ExcMessage("Matrix needs to be square."));
 
 
   // Check we haven't been here before:
diff --git a/deal.II/source/lac/sparsity_tools.cc b/deal.II/source/lac/sparsity_tools.cc
index 066aa09a9b..57a04430c0 100644
--- a/deal.II/source/lac/sparsity_tools.cc
+++ b/deal.II/source/lac/sparsity_tools.cc
@@ -506,130 +506,130 @@ namespace SparsityTools
 
   }
 
-template 
-void distribute_sparsity_pattern(CSP_t &csp,
+  template 
+  void distribute_sparsity_pattern(CSP_t &csp,
                                    const std::vector &owned_set_per_cpu,
                                    const MPI_Comm &mpi_comm,
                                    const IndexSet &myrange)
-{
-  size_type myid = Utilities::MPI::this_mpi_process(mpi_comm);
+  {
+    size_type myid = Utilities::MPI::this_mpi_process(mpi_comm);
 
-  typedef std::map > map_vec_t;
-  map_vec_t send_data;
+    typedef std::map > map_vec_t;
+    map_vec_t send_data;
 
-  {
-    unsigned int dest_cpu=0;
+    {
+      unsigned int dest_cpu=0;
 
-    size_type n_local_rel_rows = myrange.n_elements();
-    for (size_type row_idx=0; row_idx &dst = send_data[dest_cpu];
+          //save entries
+          std::vector &dst = send_data[dest_cpu];
 
-        dst.push_back(rlen); // number of entries
-        dst.push_back(row); // row index
-        for (size_type c=0; c send_to;
-    send_to.reserve(send_data.size());
-    for (map_vec_t::iterator it=send_data.begin(); it!=send_data.end(); ++it)
-      send_to.push_back(it->first);
-
-    num_receive =
-      Utilities::MPI::
-      compute_point_to_point_communication_pattern(mpi_comm, send_to).size();
-  }
+    unsigned int num_receive=0;
+    {
+      std::vector send_to;
+      send_to.reserve(send_data.size());
+      for (map_vec_t::iterator it=send_data.begin(); it!=send_data.end(); ++it)
+        send_to.push_back(it->first);
 
-  std::vector requests(send_data.size());
+      num_receive =
+        Utilities::MPI::
+        compute_point_to_point_communication_pattern(mpi_comm, send_to).size();
+    }
 
+    std::vector requests(send_data.size());
 
-  // send data
-  {
-    unsigned int idx=0;
-    for (map_vec_t::iterator it=send_data.begin(); it!=send_data.end(); ++it, ++idx)
-      MPI_Isend(&(it->second[0]),
-                it->second.size(),
-                DEAL_II_DOF_INDEX_MPI_TYPE,
-                it->first,
-                124,
-                mpi_comm,
-                &requests[idx]);
-  }
+
+    // send data
+    {
+      unsigned int idx=0;
+      for (map_vec_t::iterator it=send_data.begin(); it!=send_data.end(); ++it, ++idx)
+        MPI_Isend(&(it->second[0]),
+                  it->second.size(),
+                  DEAL_II_DOF_INDEX_MPI_TYPE,
+                  it->first,
+                  124,
+                  mpi_comm,
+                  &requests[idx]);
+    }
 
 //TODO: In the following, we read individual bytes and then reinterpret them
 //    as size_type objects. this is error prone. use properly typed reads that
 //    match the write above
-  {
-    //receive
-    std::vector recv_buf;
-    for (unsigned int index=0; index recv_buf;
+      for (unsigned int index=0; index::const_iterator ptr = recv_buf.begin();
+          std::vector::const_iterator end = recv_buf.end();
+          while (ptr+1::const_iterator ptr = recv_buf.begin();
-        std::vector::const_iterator end = recv_buf.end();
-        while (ptr+1
 (BlockCompressedSimpleSparsityPattern &csp,
-                                   const std::vector &owned_set_per_cpu,
-                                   const MPI_Comm &mpi_comm,
-                                   const IndexSet &myrange);
+ const std::vector &owned_set_per_cpu,
+ const MPI_Comm &mpi_comm,
+ const IndexSet &myrange);
 
 #endif
 
diff --git a/deal.II/source/lac/trilinos_block_vector.cc b/deal.II/source/lac/trilinos_block_vector.cc
index 7344ec1199..a0f645a3fe 100644
--- a/deal.II/source/lac/trilinos_block_vector.cc
+++ b/deal.II/source/lac/trilinos_block_vector.cc
@@ -62,7 +62,7 @@ namespace TrilinosWrappers
       // we only allow assignment to vectors with the same number of blocks
       // or to an empty BlockVector
       Assert (n_blocks() == 0 || n_blocks() == v.n_blocks(),
-                    ExcDimensionMismatch(n_blocks(), v.n_blocks()));
+              ExcDimensionMismatch(n_blocks(), v.n_blocks()));
 
       if (this->n_blocks() != v.n_blocks())
         reinit(v.n_blocks());
diff --git a/deal.II/source/lac/trilinos_precondition.cc b/deal.II/source/lac/trilinos_precondition.cc
index a014bdd2f3..d939607b0c 100644
--- a/deal.II/source/lac/trilinos_precondition.cc
+++ b/deal.II/source/lac/trilinos_precondition.cc
@@ -750,7 +750,7 @@ namespace TrilinosWrappers
 
   /* -------------------------- PreconditionAMG -------------------------- */
 
-  void 
+  void
   PreconditionIdentity::vmult(VectorBase       &dst,
                               const VectorBase &src) const
   {
@@ -764,7 +764,7 @@ namespace TrilinosWrappers
     dst = src;
   }
 
-  void 
+  void
   PreconditionIdentity::vmult(dealii::Vector       &dst,
                               const dealii::Vector &src) const
   {
diff --git a/deal.II/source/lac/trilinos_solver.cc b/deal.II/source/lac/trilinos_solver.cc
index 6eb7ae8567..7d206765fd 100644
--- a/deal.II/source/lac/trilinos_solver.cc
+++ b/deal.II/source/lac/trilinos_solver.cc
@@ -208,15 +208,15 @@ namespace TrilinosWrappers
         Assert (false, ExcNotImplemented());
       }
 
-    // Introduce the preconditioner, 
+    // Introduce the preconditioner,
     // if the identity preconditioner is used,
     // the precondioner is set to none, ...
     if (preconditioner.preconditioner.use_count()!=0)
-    {
-      ierr = solver.SetPrecOperator (const_cast
-                                     (preconditioner.preconditioner.get()));
-      AssertThrow (ierr == 0, ExcTrilinosError(ierr));
-    }
+      {
+        ierr = solver.SetPrecOperator (const_cast
+                                       (preconditioner.preconditioner.get()));
+        AssertThrow (ierr == 0, ExcTrilinosError(ierr));
+      }
     else
       solver.SetAztecOption(AZ_precond,AZ_none);
 
diff --git a/deal.II/source/multigrid/mg_tools.cc b/deal.II/source/multigrid/mg_tools.cc
index 5d74887ef2..a9da2451be 100644
--- a/deal.II/source/multigrid/mg_tools.cc
+++ b/deal.II/source/multigrid/mg_tools.cc
@@ -1336,17 +1336,17 @@ namespace MGTools
                                      :
                                      (dim == 2 ?
                                       (i<2*fe.dofs_per_vertex ? i : i+2*fe.dofs_per_vertex)
+                                      :
+                                      (dim == 3 ?
+                                       (i<4*fe.dofs_per_vertex ?
+                                        i
                                         :
-                                        (dim == 3 ?
-                                         (i<4*fe.dofs_per_vertex ?
-                                          i
-                                          :
-                                          (i<4*fe.dofs_per_vertex+4*fe.dofs_per_line ?
-                                           i+4*fe.dofs_per_vertex
-                                           :
-                                           i+4*fe.dofs_per_vertex+8*fe.dofs_per_line))
+                                        (i<4*fe.dofs_per_vertex+4*fe.dofs_per_line ?
+                                         i+4*fe.dofs_per_vertex
                                          :
-                                         numbers::invalid_unsigned_int)));
+                                         i+4*fe.dofs_per_vertex+8*fe.dofs_per_line))
+                                       :
+                                       numbers::invalid_unsigned_int)));
                                 Assert (cell_i < fe.dofs_per_cell, ExcInternalError());
 
                                 // make sure
@@ -1394,7 +1394,7 @@ namespace MGTools
                      const typename FunctionMap::type &function_map,
                      std::vector &boundary_indices,
                      const ComponentMask &component_mask)
-{
+  {
     Assert (boundary_indices.size() == dof.get_tria().n_global_levels(),
             ExcDimensionMismatch (boundary_indices.size(),
                                   dof.get_tria().n_global_levels()));
diff --git a/deal.II/source/numerics/data_out.cc b/deal.II/source/numerics/data_out.cc
index d242eaa816..546d89a217 100644
--- a/deal.II/source/numerics/data_out.cc
+++ b/deal.II/source/numerics/data_out.cc
@@ -294,7 +294,7 @@ build_one_patch (const std::pair *cell_and_index,
     }
 
   const unsigned int patch_idx =
-      (*data.cell_to_patch_index_map)[cell_and_index->first->level()][cell_and_index->first->index()];
+    (*data.cell_to_patch_index_map)[cell_and_index->first->level()][cell_and_index->first->index()];
   // did we mess up the indices?
   Assert(patch_idx < patches.size(), ExcInternalError());
 
@@ -373,13 +373,13 @@ void DataOut::build_patches (const Mappingtriangulation->end();
-        cell = next_locally_owned_cell(cell))
+         cell = next_locally_owned_cell(cell))
       {
         // move forward until active_cell points at the cell (cell) we are looking
         // at to compute the current active_index
         while (active_cell!=this->triangulation->end()
-            && cell->active()
-            && active_cell_iterator(cell) != active_cell)
+               && cell->active()
+               && active_cell_iterator(cell) != active_cell)
           {
             ++active_cell;
             ++active_index;
@@ -392,7 +392,7 @@ void DataOut::build_patches (const Mappinglevel()].size(),
                 ExcInternalError());
         Assert (active_index < this->triangulation->n_active_cells(),
-                        ExcInternalError());
+                ExcInternalError());
         cell_to_patch_index_map[cell->level()][cell->index()] = all_cells.size();
 
         all_cells.push_back (std::make_pair(cell, active_index));
@@ -451,18 +451,18 @@ void DataOut::build_patches (const Mapping::build_one_patch,
                                      this, std_cxx1x::_1, std_cxx1x::_2, std_cxx1x::_3,
                                      curved_cell_region,std_cxx1x::ref(this->patches)),
-		     // no copy-local-to-global function needed here
-		     std_cxx1x::function &)>(),
+                     // no copy-local-to-global function needed here
+                     std_cxx1x::function &)>(),
                      thread_data,
                      sample_patch,
-		     // experimenting shows that we can make things run a bit
-		     // faster if we increase the number of cells we work on
-		     // per item (i.e., WorkStream's chunk_size argument,
-		     // about 10% improvement) and the items in flight at any
-		     // given time (another 5% on the testcase discussed in
-		     // @ref workstream_paper, on 32 cores) and if
-		     8*multithread_info.n_threads(),
-		     64);
+                     // experimenting shows that we can make things run a bit
+                     // faster if we increase the number of cells we work on
+                     // per item (i.e., WorkStream's chunk_size argument,
+                     // about 10% improvement) and the items in flight at any
+                     // given time (another 5% on the testcase discussed in
+                     // @ref workstream_paper, on 32 cores) and if
+                     8*multithread_info.n_threads(),
+                     64);
 }
 
 
@@ -483,7 +483,7 @@ DataOut::next_cell (const typename DataOut::cell_iterator &cell)
   // convert the iterator to an active_iterator and advance this to the next
   // active cell
   typename Triangulation::
-    active_cell_iterator active_cell = cell;
+  active_cell_iterator active_cell = cell;
   ++active_cell;
   return active_cell;
 }
@@ -495,7 +495,7 @@ typename DataOut::cell_iterator
 DataOut::first_locally_owned_cell ()
 {
   typename DataOut::cell_iterator
-    cell = first_cell();
+  cell = first_cell();
 
   // skip cells if the current one has no children (is active) and is a ghost
   // or artificial cell
diff --git a/deal.II/source/numerics/data_out_dof_data.cc b/deal.II/source/numerics/data_out_dof_data.cc
index 37ff85b983..db456bc6eb 100644
--- a/deal.II/source/numerics/data_out_dof_data.cc
+++ b/deal.II/source/numerics/data_out_dof_data.cc
@@ -68,7 +68,7 @@ namespace internal
       if (use_face_values == false)
         {
           dealii::hp::QCollection
-            quadrature(QIterated(QTrapez<1>(), n_subdivisions));
+          quadrature(QIterated(QTrapez<1>(), n_subdivisions));
           n_q_points = quadrature[0].size();
           x_fe_values.resize(this->finite_elements.size());
           for (unsigned int i=0; ifinite_elements.size(); ++i)
@@ -93,7 +93,7 @@ namespace internal
       else
         {
           dealii::hp::QCollection
-            quadrature(QIterated(QTrapez<1>(), n_subdivisions));
+          quadrature(QIterated(QTrapez<1>(), n_subdivisions));
           n_q_points = quadrature[0].size();
           x_fe_face_values.resize(this->finite_elements.size());
           for (unsigned int i=0; ifinite_elements.size(); ++i)
@@ -157,7 +157,7 @@ namespace internal
         {
           Assert(data.x_fe_face_values.empty() == true, ExcInternalError());
           dealii::hp::QCollection
-            quadrature(QIterated(QTrapez<1>(), n_subdivisions));
+          quadrature(QIterated(QTrapez<1>(), n_subdivisions));
           x_fe_values.resize(this->finite_elements.size());
           for (unsigned int i=0; ifinite_elements.size(); ++i)
             {
@@ -181,7 +181,7 @@ namespace internal
       else
         {
           dealii::hp::QCollection
-            quadrature(QIterated(QTrapez<1>(), n_subdivisions));
+          quadrature(QIterated(QTrapez<1>(), n_subdivisions));
           x_fe_face_values.resize(this->finite_elements.size());
           for (unsigned int i=0; ifinite_elements.size(); ++i)
             {
@@ -631,7 +631,7 @@ namespace internal
 
 
 template 
+          int patch_dim, int patch_space_dim>
 DataOut_DoFData::DataOut_DoFData ()
   :
   triangulation(0,typeid(*this).name()),
@@ -677,7 +677,7 @@ attach_triangulation (const Triangulation &tr
 
 
 template 
+          int patch_dim, int patch_space_dim>
 template 
 void
 DataOut_DoFData::
@@ -715,7 +715,7 @@ add_data_vector (const VECTOR                             &vec,
 
 
 template 
+          int patch_dim, int patch_space_dim>
 template 
 void
 DataOut_DoFData::
@@ -783,7 +783,7 @@ add_data_vector (const VECTOR                             &vec,
 
 
 template 
+          int patch_dim, int patch_space_dim>
 template 
 void
 DataOut_DoFData::
@@ -810,7 +810,7 @@ add_data_vector (const VECTOR                           &vec,
 
 
 template 
+          int patch_dim, int patch_space_dim>
 template 
 void
 DataOut_DoFData::
@@ -833,7 +833,7 @@ add_data_vector (const DH                               &dof_handler,
 
 
 template 
+          int patch_dim, int patch_space_dim>
 template 
 void
 DataOut_DoFData::
@@ -866,7 +866,7 @@ add_data_vector (const DH                       &dof_handler,
 
 
 template 
+          int patch_dim, int patch_space_dim>
 template 
 void
 DataOut_DoFData::
@@ -906,7 +906,7 @@ add_data_vector (const DH                       &dof_handler,
 
 
 template 
+          int patch_dim, int patch_space_dim>
 void DataOut_DoFData::clear_data_vectors ()
 {
   dof_data.erase (dof_data.begin(), dof_data.end());
@@ -920,7 +920,7 @@ void DataOut_DoFData::clear_data_vectors ()
 
 
 template 
+          int patch_dim, int patch_space_dim>
 void
 DataOut_DoFData::
 clear_input_data_references ()
@@ -938,7 +938,7 @@ clear_input_data_references ()
 
 
 template 
+          int patch_dim, int patch_space_dim>
 void
 DataOut_DoFData::clear ()
 {
@@ -956,7 +956,7 @@ DataOut_DoFData::clear ()
 
 
 template 
+          int patch_dim, int patch_space_dim>
 std::vector
 DataOut_DoFData::
 get_dataset_names () const
@@ -984,7 +984,7 @@ get_dataset_names () const
 
 
 template 
+          int patch_dim, int patch_space_dim>
 std::vector >
 DataOut_DoFData::get_vector_data_ranges () const
 {
@@ -1078,7 +1078,7 @@ DataOut_DoFData::get_vector_data_ranges () const
 
 
 template 
+          int patch_dim, int patch_space_dim>
 const std::vector< dealii::DataOutBase::Patch > &
 DataOut_DoFData::get_patches () const
 {
@@ -1088,14 +1088,14 @@ DataOut_DoFData::get_patches () const
 
 
 template 
+          int patch_dim, int patch_space_dim>
 std::vector > >
 DataOut_DoFData::get_finite_elements() const
 {
   const unsigned int dhdim = DH::dimension;
   const unsigned int dhspacedim = DH::space_dimension;
   std::vector > >
-    finite_elements(this->dof_data.size());
+  finite_elements(this->dof_data.size());
   for (unsigned int i=0; idof_data.size(); ++i)
     {
       Assert (dof_data[i]->dof_handler != 0, ExcNoDoFHandlerSelected ());
@@ -1129,7 +1129,7 @@ DataOut_DoFData::get_finite_elements() const
 
 
 template 
+          int patch_dim, int patch_space_dim>
 std::size_t
 DataOut_DoFData::memory_consumption () const
 {
diff --git a/deal.II/source/numerics/data_out_rotation.cc b/deal.II/source/numerics/data_out_rotation.cc
index 5082a83008..82464c5557 100644
--- a/deal.II/source/numerics/data_out_rotation.cc
+++ b/deal.II/source/numerics/data_out_rotation.cc
@@ -187,7 +187,7 @@ build_one_patch (const cell_iterator *cell,
           // first fill dof_data
           for (unsigned int dataset=0; datasetdof_data.size(); ++dataset)
             {
-              const FEValuesBase & fe_patch_values
+              const FEValuesBase &fe_patch_values
                 = data.get_present_fe_values(dataset);
               const unsigned int n_components
                 = fe_patch_values.get_fe().n_components();
diff --git a/deal.II/source/numerics/derivative_approximation.cc b/deal.II/source/numerics/derivative_approximation.cc
index 79884b4258..dd2a9ca41d 100644
--- a/deal.II/source/numerics/derivative_approximation.cc
+++ b/deal.II/source/numerics/derivative_approximation.cc
@@ -631,7 +631,7 @@ approximate_derivative_tensor (const DH
 
 
 template  class DH, class InputVector, int spacedim>
+          template  class DH, class InputVector, int spacedim>
 void
 DerivativeApproximation::
 approximate_derivative (const Mapping    &mapping,
@@ -648,33 +648,33 @@ approximate_derivative (const Mapping    &mapping,
 
   // Only act on the locally owned cells
   typedef FilteredIterator::active_cell_iterator> CellFilter;
-  
+
   typedef std_cxx1x::tuple::iterator>
-    Iterators;
+  Iterators;
   SynchronousIterators begin(Iterators (CellFilter(IteratorFilters::LocallyOwnedCell(),
-            dof_handler.begin_active()),derivative_norm.begin())),
-    end(Iterators (CellFilter(IteratorFilters::LocallyOwnedCell(),dof_handler.end()),
-          derivative_norm.end()));
-  
-  // There is no need for a copier because there is no conflict between threads 
+                                                              dof_handler.begin_active()),derivative_norm.begin())),
+                                                                                       end(Iterators (CellFilter(IteratorFilters::LocallyOwnedCell(),dof_handler.end()),
+                                                                                           derivative_norm.end()));
+
+  // There is no need for a copier because there is no conflict between threads
   // to write in derivative_norm. Scratch and CopyData are also useless.
   WorkStream::run(begin,end,
-      static_cast const&,
-        internal::Assembler::Scratch const&,internal::Assembler::CopyData &)> >
-      (std_cxx1x::bind(&DerivativeApproximation::template approximate,
-                       std_cxx1x::_1,
-                       std_cxx1x::cref(mapping),
-                       std_cxx1x::cref(dof_handler),
-                       std_cxx1x::cref(solution),component)),
-    std_cxx1x::function (),
-    internal::Assembler::Scratch (),internal::Assembler::CopyData ());
+                  static_cast const &,
+                                                        internal::Assembler::Scratch const &,internal::Assembler::CopyData &)> >
+                  (std_cxx1x::bind(&DerivativeApproximation::template approximate,
+                                   std_cxx1x::_1,
+                                   std_cxx1x::cref(mapping),
+                                   std_cxx1x::cref(dof_handler),
+                                   std_cxx1x::cref(solution),component)),
+                  std_cxx1x::function (),
+                  internal::Assembler::Scratch (),internal::Assembler::CopyData ());
 }
 
 
 
 template  class DH, class InputVector, int spacedim>
+          template  class DH, class InputVector, int spacedim>
 void
 DerivativeApproximation::approximate (SynchronousIterators::active_cell_iterator>,Vector::iterator> > const &cell,
                                       const Mapping                  &mapping,
@@ -686,7 +686,7 @@ DerivativeApproximation::approximate (SynchronousIterators
-    (mapping,dof_handler,solution,component,std_cxx1x::get<0>(cell.iterators),derivative);
+  (mapping,dof_handler,solution,component,std_cxx1x::get<0>(cell.iterators),derivative);
   // evaluate the norm and fill the vector
   //*derivative_norm_on_this_cell
   *std_cxx1x::get<1>(cell.iterators) = DerivativeDescription::derivative_norm (derivative);
@@ -695,7 +695,7 @@ DerivativeApproximation::approximate (SynchronousIterators class DH, class InputVector, int spacedim>
+          template  class DH, class InputVector, int spacedim>
 void
 DerivativeApproximation::
 approximate_cell (const Mapping                   &mapping,
diff --git a/deal.II/source/numerics/matrix_tools.cc b/deal.II/source/numerics/matrix_tools.cc
index 0423f1f794..bdc0e62cd3 100644
--- a/deal.II/source/numerics/matrix_tools.cc
+++ b/deal.II/source/numerics/matrix_tools.cc
@@ -139,7 +139,7 @@ namespace MatrixCreator
     namespace AssemblerData
     {
       template 
+                int spacedim>
       struct Scratch
       {
         Scratch (const ::dealii::hp::FECollection &fe,
@@ -214,8 +214,8 @@ namespace MatrixCreator
 
 
     template 
+              int spacedim,
+              typename CellIterator>
     void mass_assembler (const CellIterator &cell,
                          MatrixCreator::internal::AssemblerData::Scratch &data,
                          MatrixCreator::internal::AssemblerData::CopyData              ©_data)
@@ -395,8 +395,8 @@ namespace MatrixCreator
 
 
     template 
+              int spacedim,
+              typename CellIterator>
     void laplace_assembler (const CellIterator &cell,
                             MatrixCreator::internal::AssemblerData::Scratch &data,
                             MatrixCreator::internal::AssemblerData::CopyData              ©_data)
@@ -576,7 +576,7 @@ namespace MatrixCreator
 
 
     template 
+              typename VectorType>
     void copy_local_to_global (const AssemblerData::CopyData &data,
                                MatrixType *matrix,
                                VectorType *right_hand_side)
@@ -629,12 +629,12 @@ namespace MatrixCreator
 
       template 
       CopyData::CopyData(CopyData const &data) :
-       dofs_per_cell(data.dofs_per_cell),
-       dofs(data.dofs),
-       dof_is_on_face(data.dof_is_on_face),
-       cell(data.cell),
-       cell_matrix(data.cell_matrix),
-       cell_vector(data.cell_vector)
+        dofs_per_cell(data.dofs_per_cell),
+        dofs(data.dofs),
+        dof_is_on_face(data.dof_is_on_face),
+        cell(data.cell),
+        cell_matrix(data.cell_matrix),
+        cell_vector(data.cell_vector)
       {}
     }
   }
@@ -878,7 +878,7 @@ namespace MatrixCreator
                                    typename FunctionMap::type const &boundary_functions,
                                    Function const *const coefficient,
                                    std::vector const &component_mapping)
-                                   
+
     {
       // All assertions for this function
       // are in the calling function
@@ -889,7 +889,7 @@ namespace MatrixCreator
       const bool         fe_is_primitive = fe.is_primitive();
 
       const unsigned int dofs_per_face = fe.dofs_per_face;
-      
+
       copy_data.cell = cell;
       copy_data.dofs_per_cell = fe.dofs_per_cell;
 
@@ -910,7 +910,7 @@ namespace MatrixCreator
 
       std::vector          rhs_values_scalar (fe_values.n_quadrature_points);
       std::vector > rhs_values_system (fe_values.n_quadrature_points,
-          Vector(n_function_components));
+                                                      Vector(n_function_components));
 
       copy_data.dofs.resize(copy_data.dofs_per_cell);
       cell->get_dof_indices (copy_data.dofs);
@@ -931,7 +931,7 @@ namespace MatrixCreator
             boundary_functions.end())
           {
             copy_data.cell_matrix.push_back(FullMatrix (copy_data.dofs_per_cell,
-                  copy_data.dofs_per_cell));
+                                                                copy_data.dofs_per_cell));
             copy_data.cell_vector.push_back(Vector (copy_data.dofs_per_cell));
             fe_values.reinit (cell, face);
 
@@ -1007,8 +1007,8 @@ namespace MatrixCreator
                                 }
                             }
                           copy_data.cell_vector.back()(i) += fe_values.shape_value(i,point)
-                                            * rhs_values_system[point](component_mapping[fe.system_to_component_index(i).first])
-                                            * weight;
+                                                             * rhs_values_system[point](component_mapping[fe.system_to_component_index(i).first])
+                                                             * weight;
                         }
                       else
                         {
@@ -1021,9 +1021,9 @@ namespace MatrixCreator
                                    * normal_adjustment[point][comp]
                                    * weight * coefficient_vector_values[point](comp);
                               copy_data.cell_vector.back()(i) += fe_values.shape_value_component(i,point,comp) *
-                                                rhs_values_system[point](component_mapping[comp])
-                                                * normal_adjustment[point][comp]
-                                                * weight;
+                                                                 rhs_values_system[point](component_mapping[comp])
+                                                                 * normal_adjustment[point][comp]
+                                                                 * weight;
                             }
                         }
                   }
@@ -1066,7 +1066,7 @@ namespace MatrixCreator
                                                               copy_data.dofs[i])
                                                     !=
                                                     dofs_on_face_vector.end());
-          }   
+          }
     }
 
     template 
@@ -1105,34 +1105,34 @@ namespace MatrixCreator
       // searches.
       unsigned int pos(0);
       for (unsigned int face=0; face::faces_per_cell; ++face)
-      {
-        // check if this face is on that part of
-        // the boundary we are interested in
-        if (boundary_functions.find(copy_data.cell->face(face)->boundary_indicator()) !=
-            boundary_functions.end())
-          {
-            for (unsigned int i=0; iface(face)->boundary_indicator()) !=
+              boundary_functions.end())
+            {
+              for (unsigned int i=0; i
     void
     create_boundary_mass_matrix_1<2,3> (DoFHandler<2,3>::active_cell_iterator const &cell,
-                                        MatrixCreator::internal::AssemblerBoundary::Scratch const 
-                                        &scratch,                                                 
+                                        MatrixCreator::internal::AssemblerBoundary::Scratch const
+                                        &scratch,
                                         MatrixCreator::internal::AssemblerBoundary::CopyData > ©_data,                                               
-                                        Mapping<2,3> const &mapping,                              
-                                        FiniteElement<2,3> const &fe,                             
-                                        Quadrature<1> const &q,                                   
-                                        FunctionMap<3>::type const &boundary_functions,  
-                                        Function<3> const *const coefficient,                     
-                                        std::vector const &component_mapping)       
+                                        3> > ©_data,
+                                        Mapping<2,3> const &mapping,
+                                        FiniteElement<2,3> const &fe,
+                                        Quadrature<1> const &q,
+                                        FunctionMap<3>::type const &boundary_functions,
+                                        Function<3> const *const coefficient,
+                                        std::vector const &component_mapping)
     {
       Assert(false,ExcNotImplemented());
     }
@@ -1159,16 +1159,16 @@ namespace MatrixCreator
     template <>
     void
     create_boundary_mass_matrix_1<1,3> (DoFHandler<1,3>::active_cell_iterator const &cell,
-                                        MatrixCreator::internal::AssemblerBoundary::Scratch const 
+                                        MatrixCreator::internal::AssemblerBoundary::Scratch const
                                         &scratch,
                                         MatrixCreator::internal::AssemblerBoundary::CopyData > ©_data,                             
+                                        3> > ©_data,
                                         Mapping<1,3> const &mapping,
                                         FiniteElement<1,3> const &fe,
                                         Quadrature<0> const &q,
                                         FunctionMap<3>::type const &boundary_functions,
                                         Function<3> const *const coefficient,
-                                        std::vector const &component_mapping)                
+                                        std::vector const &component_mapping)
     {
       Assert(false,ExcNotImplemented());
     }
@@ -1225,24 +1225,24 @@ namespace MatrixCreator
     MatrixCreator::internal::AssemblerBoundary::CopyData > copy_data;
 
     WorkStream::run(dof.begin_active(),dof.end(),
-        static_cast::active_cell_iterator
-          const &,MatrixCreator::internal::AssemblerBoundary::Scratch const &,
-          MatrixCreator::internal::AssemblerBoundary::CopyData > &)> > 
-        (std_cxx1x::bind(&create_boundary_mass_matrix_1,std_cxx1x::_1,std_cxx1x::_2,
-                         std_cxx1x::_3,
-                         std_cxx1x::cref(mapping),std_cxx1x::cref(fe),std_cxx1x::cref(q),
-                         std_cxx1x::cref(boundary_functions),coefficient,
-                         std_cxx1x::cref(component_mapping))), 
-        static_cast > const &)> > (std_cxx1x::bind(
-              ©_boundary_mass_matrix_1,
-	      std_cxx1x::_1,
-	      std_cxx1x::cref(boundary_functions),
-	      std_cxx1x::cref(dof_to_boundary_mapping),
-	      std_cxx1x::ref(matrix),
-	      std_cxx1x::ref(rhs_vector))),
-		    scratch,
-		    copy_data);
+                    static_cast::active_cell_iterator
+                                                          const &,MatrixCreator::internal::AssemblerBoundary::Scratch const &,
+                                                          MatrixCreator::internal::AssemblerBoundary::CopyData > &)> >
+                    (std_cxx1x::bind(&create_boundary_mass_matrix_1,std_cxx1x::_1,std_cxx1x::_2,
+                                     std_cxx1x::_3,
+                                     std_cxx1x::cref(mapping),std_cxx1x::cref(fe),std_cxx1x::cref(q),
+                                     std_cxx1x::cref(boundary_functions),coefficient,
+                                     std_cxx1x::cref(component_mapping))),
+                    static_cast > const &)> > (std_cxx1x::bind(
+                                                                ©_boundary_mass_matrix_1,
+                                                                std_cxx1x::_1,
+                                                                std_cxx1x::cref(boundary_functions),
+                                                                std_cxx1x::cref(dof_to_boundary_mapping),
+                                                                std_cxx1x::ref(matrix),
+                                                                std_cxx1x::ref(rhs_vector))),
+                    scratch,
+                    copy_data);
   }
 
 
@@ -1252,7 +1252,7 @@ namespace MatrixCreator
 
     template 
     void
-    create_hp_boundary_mass_matrix_1 (typename hp::DoFHandler::active_cell_iterator const 
+    create_hp_boundary_mass_matrix_1 (typename hp::DoFHandler::active_cell_iterator const
                                       &cell,
                                       MatrixCreator::internal::AssemblerBoundary::Scratch const &scratch,
                                       MatrixCreator::internal::AssemblerBoundary
@@ -1314,7 +1314,7 @@ namespace MatrixCreator
             const FEFaceValues &fe_values = x_fe_values.get_present_fe_values ();
 
             copy_data.cell_matrix.push_back(FullMatrix (copy_data.dofs_per_cell,
-                  copy_data.dofs_per_cell));
+                                                                copy_data.dofs_per_cell));
             copy_data.cell_vector.push_back(Vector (copy_data.dofs_per_cell));
 
             if (fe_is_system)
@@ -1349,8 +1349,8 @@ namespace MatrixCreator
                                     }
 
                                 copy_data.cell_vector.back()(i) += v *
-                                                  rhs_values_system[point](
-                                                    component_mapping[fe.system_to_component_index(i).first]) * weight;
+                                                                   rhs_values_system[point](
+                                                                     component_mapping[fe.system_to_component_index(i).first]) * weight;
                               }
                           }
                       }
@@ -1376,8 +1376,8 @@ namespace MatrixCreator
                                       copy_data.cell_matrix.back()(i,j) +=
                                         (u * v * weight * coefficient_vector_values[point](component_i));
                                     }
-                                copy_data.cell_vector.back()(i) += v * 
-                                  rhs_values_system[point](component_mapping[component_i]) * weight;
+                                copy_data.cell_vector.back()(i) += v *
+                                                                   rhs_values_system[point](component_mapping[component_i]) * weight;
                               }
                           }
                       }
@@ -1397,9 +1397,9 @@ namespace MatrixCreator
                                 copy_data.cell_matrix.back()(i,j) += (u * v * weight);
                               }
                           copy_data.cell_vector.back()(i) += v *
-                                            rhs_values_system[point](
-                                              fe.system_to_component_index(i).first) *
-                                            weight;
+                                                             rhs_values_system[point](
+                                                               fe.system_to_component_index(i).first) *
+                                                             weight;
                         }
                     }
               }
@@ -1424,8 +1424,8 @@ namespace MatrixCreator
                             for (unsigned int j=0; j::faces_per_cell; ++face)
-      {
-        // check if this face is on that part of
-        // the boundary we are interested in
-        if (boundary_functions.find(copy_data.cell->face(face)->boundary_indicator()) !=
-            boundary_functions.end())
-          {
+        {
+          // check if this face is on that part of
+          // the boundary we are interested in
+          if (boundary_functions.find(copy_data.cell->face(face)->boundary_indicator()) !=
+              boundary_functions.end())
+            {
 #ifdef DEBUG
-            // in debug mode: compute an element in the matrix which is
-            // guaranteed to belong to a boundary dof. We do this to check that the
-            // entries in the cell matrix are guaranteed to be zero if the
-            // respective dof is not on the boundary. Since because of
-            // round-off, the actual value of the matrix entry may be
-            // only close to zero, we assert that it is small relative to an element
-            // which is guaranteed to be nonzero. (absolute smallness does not
-            // suffice since the size of the domain scales in here)
-            //
-            // for this purpose we seek the diagonal of the matrix, where there
-            // must be an element belonging to the boundary. we take the maximum
-            // diagonal entry.
-            types::global_dof_index max_element = static_cast(0);
-            for (std::vector::const_iterator i=dof_to_boundary_mapping.begin();
-                 i!=dof_to_boundary_mapping.end(); ++i)
-              if ((*i != hp::DoFHandler::invalid_dof_index) &&
-                  (*i > max_element))
-                max_element = *i;
-            Assert (max_element  == matrix.n()-1, ExcInternalError());
-
-            double max_diag_entry = 0;
-            for (unsigned int i=0; i max_diag_entry)
-                max_diag_entry = std::fabs(copy_data.cell_matrix[pos](i,i));
+              // in debug mode: compute an element in the matrix which is
+              // guaranteed to belong to a boundary dof. We do this to check that the
+              // entries in the cell matrix are guaranteed to be zero if the
+              // respective dof is not on the boundary. Since because of
+              // round-off, the actual value of the matrix entry may be
+              // only close to zero, we assert that it is small relative to an element
+              // which is guaranteed to be nonzero. (absolute smallness does not
+              // suffice since the size of the domain scales in here)
+              //
+              // for this purpose we seek the diagonal of the matrix, where there
+              // must be an element belonging to the boundary. we take the maximum
+              // diagonal entry.
+              types::global_dof_index max_element = static_cast(0);
+              for (std::vector::const_iterator i=dof_to_boundary_mapping.begin();
+                   i!=dof_to_boundary_mapping.end(); ++i)
+                if ((*i != hp::DoFHandler::invalid_dof_index) &&
+                    (*i > max_element))
+                  max_element = *i;
+              Assert (max_element  == matrix.n()-1, ExcInternalError());
+
+              double max_diag_entry = 0;
+              for (unsigned int i=0; i max_diag_entry)
+                  max_diag_entry = std::fabs(copy_data.cell_matrix[pos](i,i));
 #endif
 
-            for (unsigned int i=0; i > copy_data;
 
     WorkStream::run(dof.begin_active(),dof.end(),
-        static_cast::active_cell_iterator
-          const &,MatrixCreator::internal::AssemblerBoundary::Scratch const &,
-          MatrixCreator::internal::AssemblerBoundary::CopyData > &)> > 
-        (std_cxx1x::bind( &create_hp_boundary_mass_matrix_1,std_cxx1x::_1,std_cxx1x::_2,
-                         std_cxx1x::_3,
-                         std_cxx1x::cref(mapping),std_cxx1x::cref(fe_collection),std_cxx1x::cref(q),
-                         std_cxx1x::cref(boundary_functions),coefficient,
-                         std_cxx1x::cref(component_mapping))), 
-		    static_cast > const &)> > (
-		       std_cxx1x::bind( ©_hp_boundary_mass_matrix_1,
-					std_cxx1x::_1,
-					std_cxx1x::cref(boundary_functions),
-					std_cxx1x::cref(dof_to_boundary_mapping),
-					std_cxx1x::ref(matrix),
-					std_cxx1x::ref(rhs_vector))),
-		    scratch,
-		    copy_data);
+                    static_cast::active_cell_iterator
+                                                          const &,MatrixCreator::internal::AssemblerBoundary::Scratch const &,
+                                                          MatrixCreator::internal::AssemblerBoundary::CopyData > &)> >
+                    (std_cxx1x::bind( &create_hp_boundary_mass_matrix_1,std_cxx1x::_1,std_cxx1x::_2,
+                                      std_cxx1x::_3,
+                                      std_cxx1x::cref(mapping),std_cxx1x::cref(fe_collection),std_cxx1x::cref(q),
+                                      std_cxx1x::cref(boundary_functions),coefficient,
+                                      std_cxx1x::cref(component_mapping))),
+                    static_cast > const &)> > (
+                      std_cxx1x::bind( ©_hp_boundary_mass_matrix_1,
+                                       std_cxx1x::_1,
+                                       std_cxx1x::cref(boundary_functions),
+                                       std_cxx1x::cref(dof_to_boundary_mapping),
+                                       std_cxx1x::ref(matrix),
+                                       std_cxx1x::ref(rhs_vector))),
+                    scratch,
+                    copy_data);
   }
 
 
@@ -1712,10 +1712,10 @@ namespace MatrixCreator
                      static_cast::active_cell_iterator>(dof.end()),
                      &MatrixCreator::internal::laplace_assembler::active_cell_iterator>,
                      std_cxx1x::bind (&MatrixCreator::internal::
-				      copy_local_to_global, Vector >,
-                                      std_cxx1x::_1, 
-				      &matrix, 
-				      (Vector *)NULL),
+                                      copy_local_to_global, Vector >,
+                                      std_cxx1x::_1,
+                                      &matrix,
+                                      (Vector *)NULL),
                      assembler_data,
                      copy_data);
   }
@@ -1769,10 +1769,10 @@ namespace MatrixCreator
                      static_cast::active_cell_iterator>(dof.end()),
                      &MatrixCreator::internal::laplace_assembler::active_cell_iterator>,
                      std_cxx1x::bind (&MatrixCreator::internal::
-				      copy_local_to_global, Vector >,
-                                      std_cxx1x::_1, 
-				      &matrix, 
-				      &rhs_vector),
+                                      copy_local_to_global, Vector >,
+                                      std_cxx1x::_1,
+                                      &matrix,
+                                      &rhs_vector),
                      assembler_data,
                      copy_data);
   }
@@ -1824,10 +1824,10 @@ namespace MatrixCreator
                      static_cast::active_cell_iterator>(dof.end()),
                      &MatrixCreator::internal::laplace_assembler::active_cell_iterator>,
                      std_cxx1x::bind (&MatrixCreator::internal::
-				      copy_local_to_global, Vector >,
-                                      std_cxx1x::_1, 
-				      &matrix, 
-				      (Vector *)0),
+                                      copy_local_to_global, Vector >,
+                                      std_cxx1x::_1,
+                                      &matrix,
+                                      (Vector *)0),
                      assembler_data,
                      copy_data);
   }
@@ -1878,10 +1878,10 @@ namespace MatrixCreator
                      static_cast::active_cell_iterator>(dof.end()),
                      &MatrixCreator::internal::laplace_assembler::active_cell_iterator>,
                      std_cxx1x::bind (&MatrixCreator::internal::
-				      copy_local_to_global, Vector >,
-                                      std_cxx1x::_1, 
-				      &matrix, 
-				      &rhs_vector),
+                                      copy_local_to_global, Vector >,
+                                      std_cxx1x::_1,
+                                      &matrix,
+                                      &rhs_vector),
                      assembler_data,
                      copy_data);
   }
diff --git a/deal.II/source/numerics/solution_transfer.cc b/deal.II/source/numerics/solution_transfer.cc
index beb4b0155d..5e0ac217ed 100644
--- a/deal.II/source/numerics/solution_transfer.cc
+++ b/deal.II/source/numerics/solution_transfer.cc
@@ -563,41 +563,41 @@ interpolate (const std::vector &all_in,
                           local_values.reinit (cell->has_children() ?
                                                cell->child(0)->get_fe().dofs_per_cell
                                                : cell->get_fe().dofs_per_cell, true);
-			  // do the interpolation. we get into trouble if the
-			  // interpolation_hp(new,old) matrix hasn't been computed.
-			  // this can happen if the respective elements don't support
-			  // the corresponding interpolation; if that's the case, then
-			  // the computation of the matrix simply sets the matrix
-			  // back to size zero. so if we get here and that is
-			  // the wrong size, then this may be because the elements
-			  // haven't implemented the correct function yet
-			  //
-			  // there is one wrinkle. we would like to only error out if
-			  // the size of the matrix is 0 times 0 but at least one
-			  // of the elements has more than one dof per cell. the
-			  // problem is that if you reinit a matrix to 4x0, it automatically
-			  // sets its size to 0x0. so we can only execute the following
-			  // test if *both* elements have dofs_per_cell>0, not if *at
-			  // least one* have.
-			  Assert (! ((interpolation_hp(new_fe_index,old_index).m() == 0)
-				     &&
-				     (interpolation_hp(new_fe_index,old_index).n() == 0)
-				     &&
-				     ((dof_handler->get_fe()[new_fe_index].dofs_per_cell > 0)
-				      &&
-				      (dof_handler->get_fe()[old_index].dofs_per_cell > 0))),
-				  ExcMessage ("The interpolation between two different "
-					      "elements you are trying to use here has "
-					      "not been implemented for this pair of "
-					      "elements!"));
+                          // do the interpolation. we get into trouble if the
+                          // interpolation_hp(new,old) matrix hasn't been computed.
+                          // this can happen if the respective elements don't support
+                          // the corresponding interpolation; if that's the case, then
+                          // the computation of the matrix simply sets the matrix
+                          // back to size zero. so if we get here and that is
+                          // the wrong size, then this may be because the elements
+                          // haven't implemented the correct function yet
+                          //
+                          // there is one wrinkle. we would like to only error out if
+                          // the size of the matrix is 0 times 0 but at least one
+                          // of the elements has more than one dof per cell. the
+                          // problem is that if you reinit a matrix to 4x0, it automatically
+                          // sets its size to 0x0. so we can only execute the following
+                          // test if *both* elements have dofs_per_cell>0, not if *at
+                          // least one* have.
+                          Assert (! ((interpolation_hp(new_fe_index,old_index).m() == 0)
+                                     &&
+                                     (interpolation_hp(new_fe_index,old_index).n() == 0)
+                                     &&
+                                     ((dof_handler->get_fe()[new_fe_index].dofs_per_cell > 0)
+                                      &&
+                                      (dof_handler->get_fe()[old_index].dofs_per_cell > 0))),
+                                  ExcMessage ("The interpolation between two different "
+                                              "elements you are trying to use here has "
+                                              "not been implemented for this pair of "
+                                              "elements!"));
 
                           // simple case where all children have the
                           // same FE index: just interpolate to their FE
                           // first and then use the standard routines
-			  if (tmp.size() > 0)
-			    interpolation_hp(new_fe_index,old_index).vmult (local_values, tmp);
-			  else
-			    local_values = 0;
+                          if (tmp.size() > 0)
+                            interpolation_hp(new_fe_index,old_index).vmult (local_values, tmp);
+                          else
+                            local_values = 0;
                         }
 
                       if (cell->has_children() == false)
@@ -615,21 +615,21 @@ interpolate (const std::vector &all_in,
                                                      interpolation_hp(c_index,old_index).n());
                                     local_values.reinit(cell->child(child)->get_fe().dofs_per_cell, true);
 
-				    // do the interpolation. same problem as above
-				    Assert (! ((interpolation_hp(c_index,old_index).m() == 0)
-					       &&
-					       (interpolation_hp(c_index,old_index).n() == 0)
-					       &&
-					       ((dof_handler->get_fe()[c_index].dofs_per_cell > 0)
-						&&
-						(dof_handler->get_fe()[old_index].dofs_per_cell > 0))),
-					    ExcMessage ("The interpolation between two different "
-							"elements you are trying to use here has "
-							"not been implemented for this pair of "
-							"elements!"));
-
-				    if (tmp.size() > 0)
-				      interpolation_hp(c_index,old_index).vmult (local_values, tmp);
+                                    // do the interpolation. same problem as above
+                                    Assert (! ((interpolation_hp(c_index,old_index).m() == 0)
+                                               &&
+                                               (interpolation_hp(c_index,old_index).n() == 0)
+                                               &&
+                                               ((dof_handler->get_fe()[c_index].dofs_per_cell > 0)
+                                                &&
+                                                (dof_handler->get_fe()[old_index].dofs_per_cell > 0))),
+                                            ExcMessage ("The interpolation between two different "
+                                                        "elements you are trying to use here has "
+                                                        "not been implemented for this pair of "
+                                                        "elements!"));
+
+                                    if (tmp.size() > 0)
+                                      interpolation_hp(c_index,old_index).vmult (local_values, tmp);
                                     else
                                       local_values = 0;
                                   }
diff --git a/deal.II/tests/quick_tests/CMakeLists.txt b/deal.II/tests/quick_tests/CMakeLists.txt
index a01e681928..232036aae0 100644
--- a/deal.II/tests/quick_tests/CMakeLists.txt
+++ b/deal.II/tests/quick_tests/CMakeLists.txt
@@ -23,31 +23,40 @@ ENABLE_TESTING()
 LIST(GET DEAL_II_BUILD_TYPES 0 _mybuild)
 MESSAGE(STATUS "Setup quick_tests in ${_mybuild} mode")
 
+SET(ALL_TESTS) # clean variable
+
 # define a macro to set up a quick test:
 MACRO(make_quicktest test_basename build_name mpi_run)
   STRING(TOLOWER ${build_name} _build_lowercase)
   SET(_target ${test_basename}.${_build_lowercase})
+  LIST(APPEND ALL_TESTS "${_target}")
   ADD_EXECUTABLE(${_target} EXCLUDE_FROM_ALL ${test_basename}.cc)
   DEAL_II_INSOURCE_SETUP_TARGET(${_target} ${build_name})
 
-  IF(NOT ${mpi_run} EQUAL "")
-  ADD_CUSTOM_COMMAND(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/${_target}-OK
-    DEPENDS ${_target}
-    COMMAND mpirun -n ${mpi_run} ./${_target} > ${CMAKE_CURRENT_BINARY_DIR}/${_target}-OK 2>&1 || (rm ${_target}-OK  && exit 1)
-    WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
-    )
+  IF("${mpi_run}" STREQUAL "")
+    SET(_command ./${_target})
   ELSE()
-  ADD_CUSTOM_COMMAND(OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/${_target}-OK
+    SET(_command mpirun -n ${mpi_run} ./${_target})
+  ENDIF()
+  ADD_CUSTOM_TARGET(${_target}.run
     DEPENDS ${_target}
-    COMMAND ./${_target} > ${CMAKE_CURRENT_BINARY_DIR}/${_target}-OK 2>&1 || (rm ${_target}-OK  && exit 1)
+    COMMAND
+      ${_command} > ${_target}-OK 2>&1
+      ||(echo "${_target}: RUN failed. Output:"
+         && cat ${_target}-OK
+         && rm ${_target}-OK
+         && exit 1)
+    COMMAND echo "${_target}: PASSED."
     WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
     )
-  ENDIF()
 
-  ADD_CUSTOM_TARGET(${_target}.run
-    DEPENDS ${_target}
-    DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/${_target}-OK
-    COMMAND echo "${_target}: PASSED.")
+  # this is a hack to make sure the -OK file is deleted
+  # even if compilation fails.
+  ADD_CUSTOM_TARGET(kill-${_target}-OK
+        COMMAND rm -f ${_target}-OK 
+    WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
+    )
+  ADD_DEPENDENCIES(${_target} kill-${_target}-OK)
 
   ADD_TEST(NAME ${_target}
     COMMAND ${CMAKE_COMMAND} -DTRGT=${_target}.run -DTEST=${_target}
@@ -59,7 +68,7 @@ MACRO(make_quicktest test_basename build_name mpi_run)
 ENDMACRO()
 
 
-# simple assembly/solver test. This makes sure we can compile and link correctly
+# Simple assembly/solver test. This makes sure we can compile and link correctly
 # in debug and release.
 FOREACH(_build ${DEAL_II_BUILD_TYPES})
   make_quicktest("step" ${_build} "")
@@ -96,7 +105,7 @@ ENDIF()
 
 # A custom test target:
 ADD_CUSTOM_TARGET(test
-  COMMAND ${CMAKE_CTEST_COMMAND} --force-new-ctest-process
+  COMMAND ${CMAKE_COMMAND} -D ALL_TESTS="${ALL_TESTS}" -P ${CMAKE_CURRENT_SOURCE_DIR}/run.cmake
   WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
   )
 ADD_DEPENDENCIES(test build_library)
diff --git a/deal.II/tests/quick_tests/run.cmake b/deal.II/tests/quick_tests/run.cmake
new file mode 100644
index 0000000000..4cf38f1f2a
--- /dev/null
+++ b/deal.II/tests/quick_tests/run.cmake
@@ -0,0 +1,64 @@
+## ---------------------------------------------------------------------
+## $Id$
+##
+## Copyright (C) 2013 by the deal.II authors
+##
+## This file is part of the deal.II library.
+##
+## The deal.II library is free software; you can use it, redistribute
+## it, and/or modify it under the terms of the GNU Lesser General
+## Public License as published by the Free Software Foundation; either
+## version 2.1 of the License, or (at your option) any later version.
+## The full text of the license can be found in the file LICENSE at
+## the top level of the deal.II distribution.
+##
+## ---------------------------------------------------------------------
+
+# This file is run when "make test" is executed by the user and is 
+# responsible for running the tests and printing some helpful
+# error messages.
+
+SEPARATE_ARGUMENTS(ALL_TESTS)
+
+EXECUTE_PROCESS(COMMAND ${CMAKE_CTEST_COMMAND} --force-new-ctest-process --output-on-failure -O quicktests.log RESULT_VARIABLE res_var)
+
+if(NOT "${res_var}" STREQUAL "0")
+  MESSAGE( "
+
+*******************************     WARNING     *******************************
+
+Some of the tests failed!
+
+Please scroll up or check the file tests/quick_tests/quicktests.log for the
+error messages. If you are unable to fix the problems, see the FAQ or write
+to the mailing list linked at http://www.dealii.org\n"
+    )
+
+  FOREACH(test ${ALL_TESTS})  
+    IF (${test} MATCHES "^affinity" AND NOT EXISTS ${test}-OK)
+      MESSAGE("
+The affinity test can fail when you are linking in a library like BLAS
+which uses OpenMP. Even without calling any BLAS functions, OpenMP messes
+with the thread affinity which causes TBB to run single-threaded only. You
+can fix this by exporting OMP_NUM_THREADS=1.\n"
+        )
+    ENDIF()
+
+    IF (${test} MATCHES "^step-petsc" AND NOT EXISTS ${test}-OK)
+      MESSAGE("
+Additional information about PETSc issues is available
+at:\nhttp://www.dealii.org/developer/external-libs/petsc.html\n"
+        )
+    ENDIF()
+
+    IF (${test} MATCHES "^p4est" AND NOT EXISTS ${test}-OK)
+      MESSAGE("
+The p4est test can fail if you are running an OpenMPI version before 1.5.
+This is a known problem and the only work around is to update to a more
+recent version or use a different MPI library like MPICH.\n"
+        )
+    ENDIF()
+
+  ENDFOREACH()
+
+ENDIF()
diff --git a/deal.II/tests/quick_tests/tbb.cc b/deal.II/tests/quick_tests/tbb.cc
index 654da479bf..0e8110caf9 100644
--- a/deal.II/tests/quick_tests/tbb.cc
+++ b/deal.II/tests/quick_tests/tbb.cc
@@ -76,7 +76,7 @@ void test2()
 				  std_cxx1x::_1),
 		  scratch_data(), copy_data());
   std::cout << "result: " << result << std::endl;
-  
+
   if (result != (3*3+5*5+1*1))
     exit(2);
 }
@@ -84,7 +84,7 @@ void test2()
 int main ()
 {
   std::cout << "TBB will use " << tbb::task_scheduler_init::default_num_threads() << " threads." << std::endl;
-  
+
   test1();
   test2();
 }