[SCM] Library for computing Fast Fourier Transforms branch, upstream, updated. upstream/3.3.2-1-g6a18b11

Julian Taylor jtaylor.debian at googlemail.com
Sun Dec 2 17:12:04 UTC 2012


The following commit has been merged in the upstream branch:
commit 6a18b11870ba535799c29d69c430d5e6ba3adc94
Author: Julian Taylor <jtaylor.debian at googlemail.com>
Date:   Mon Nov 26 19:46:15 2012 +0100

    Imported Upstream version 3.3.3

diff --git a/ChangeLog b/ChangeLog
index b697fc1..770d2ee 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,132 @@
+Sat Nov 24 22:37:54 EST 2012  stevenj at fftw.org
+  * fixed deadlock bug caused by bogosity flag getting out of synch between processes; thanks to Michael Pippig for the bug report
+
+    M ./kernel/ifftw.h +1
+    M ./kernel/planner.c -3 +6
+    M ./mpi/api.c +12
+
+Wed Nov 21 18:34:29 EST 2012  athena at fftw.org
+  * Updated NEWS
+
+    M ./NEWS -2 +7
+
+Wed Nov 21 18:33:15 EST 2012  athena at fftw.org
+  * use 2x2 AVX transposition instead of individual stores.
+  
+  This seems to improve single-precision AVX on Sandy Bridge machines.
+  
+
+    M ./simd-support/simd-avx.h -2 +14
+
+Tue Nov 20 12:18:00 EST 2012  stevenj at fftw.org
+  * revert part of Taylor patch to acx_mpi.m4: do not link -lmpi if mpicc works without libraries, as -lmpi may be some completely different MPI implementation
+
+    M ./m4/acx_mpi.m4 -3 +3
+
+Tue Nov 20 11:44:57 EST 2012  stevenj at fftw.org
+  * fix deadlock bug (thanks to Michael Pippig for the bug report and patch, and to Graham Dennis for the bug report) in which some processes called MPI_Alltoall and some called MPI_Alltoallv
+
+    M ./mpi/transpose-alltoall.c -3 +2
+
+Mon Oct 29 15:20:01 EDT 2012  athena at fftw.org
+  * fix texinfo quirk
+
+    M ./doc/tutorial.texi -2 +2
+
+Mon Oct 29 09:16:43 EDT 2012  athena at fftw.org
+  * clarify that padding only applies to in-place transforms
+
+    M ./doc/tutorial.texi -5 +10
+
+Sun Oct 28 18:42:48 EDT 2012  athena at fftw.org
+  * make the index-computation logic less paranoid
+  
+  The problem is that for each K and for each expression of the form P[I
+  + STRIDE * K] in a loop, most compilers will try to lift an induction
+  variable PK := &P[I + STRIDE * K].  In large codelets we have many
+  such values of K.  For example, a codelet of size 32 with 4 input
+  pointers will generate O(128) induction variables, which will likely
+  overflow the register set, which is likely worse than doing the index
+  computation in the first place.
+  
+  In the past we (wisely and correctly) assumed that compilers will do
+  the wrong thing, and consequently we disabled the induction-variable
+  "optimization" altogether by setting STRIDE ^= ZERO, where ZERO is a
+  value guaranteed to be 0.  Since the compiler does not know that
+  ZERO=0, it cannot perform its "optimization" and it is forced to
+  behave sensibly.
+  
+  With this patch, FFTW is a little bit less paranoid.  FFTW now
+  disables the induction-variable optimization" only when we estimate
+  that the codelet uses more than ESTIMATED_AVAILABLE_INDEX_REGISTERS
+  induction variables.
+  
+  Currently we set ESTIMATED_AVAILABLE_INDEX_REGISTERS=16.  16 registers ought
+  to be enough for anybody (or so the amd64 and ARM ISA's seem to imply).
+  
+
+    M ./genfft/gen_hc2c.ml -1 +1
+    M ./genfft/gen_hc2cdft.ml -1 +1
+    M ./genfft/gen_hc2cdft_c.ml -1 +1
+    M ./genfft/gen_hc2hc.ml -1 +1
+    M ./genfft/gen_notw.ml -2 +2
+    M ./genfft/gen_notw_c.ml -2 +2
+    M ./genfft/gen_r2cb.ml -3 +3
+    M ./genfft/gen_r2cf.ml -3 +3
+    M ./genfft/gen_r2r.ml -2 +2
+    M ./genfft/gen_twiddle.ml -1 +1
+    M ./genfft/gen_twiddle_c.ml -1 +1
+    M ./genfft/gen_twidsq.ml -2 +2
+    M ./genfft/gen_twidsq_c.ml -2 +2
+    M ./genfft/genutil.ml -1 +2
+    M ./kernel/ifftw.h -3 +20
+
+Sun Oct 28 18:33:24 EDT 2012  athena at fftw.org
+  * silence warnings
+
+    M ./kernel/buffered.c +1
+    M ./rdft/rank0.c +1
+
+Sat Oct 27 09:58:49 EDT 2012  athena at fftw.org
+  * bump version to 3.3.3
+
+    M ./NEWS +7
+    M ./configure.ac -1 +1
+
+Sat Oct 27 09:55:15 EDT 2012  athena at fftw.org
+  * evaluate plans for >1ms when using gettimeofday()
+  
+  The previous limit 10ms was too paranoid, and it made life difficult
+  on machines without an "official" cycle counter, such as ARM.
+
+    M ./kernel/timer.c -1 +1
+
+Sat Oct 27 09:46:04 EDT 2012  athena at fftw.org
+  * use 4-way NEON SIMD instead of 2-way
+  
+  Kai-Uwe Bloem tried to warn me a year ago that 128-bit NEON was better
+  than 64-bit NEON even on machines with a 64-bit pipe, but I foolishly
+  did not listen.  Now that 128-bit NEON pipes are starting to appear on
+  the market it is definitely time to switch.
+  
+
+    M ./simd-support/simd-neon.h -55 +100
+
+Wed Sep 26 14:21:12 EDT 2012  athena at fftw.org
+  * Note that fftw-3.3 includes MPI support
+
+    M ./doc/intro.texi -5 +4
+
+Wed Jul 18 11:25:40 EDT 2012  athena at fftw.org
+  * remove obsolete unused function
+
+    M ./dft/bluestein.c -14
+
+Fri Jun 29 15:57:14 EDT 2012  stevenj at fftw.org
+  * whoops, call omp_get_max_threads; thanks to Hanno Rein for the bug report
+
+    M ./doc/threads.texi -1 +1
+
 Sat Apr 28 10:55:09 EDT 2012  athena at fftw.org
   * Fix libfftw3/libfftw3_threads chicken-egg problem
   
diff --git a/Makefile.in b/Makefile.in
index 294b305..34bc0a1 100644
--- a/Makefile.in
+++ b/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -17,6 +17,23 @@
 
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -124,6 +141,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DATA = $(pkgconfig_DATA)
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
@@ -447,7 +469,6 @@ fftw.pc: $(top_builddir)/config.status $(srcdir)/fftw.pc.in
 	cd $(top_builddir) && $(SHELL) ./config.status $@
 install-libLTLIBRARIES: $(lib_LTLIBRARIES)
 	@$(NORMAL_INSTALL)
-	test -z "$(libdir)" || $(MKDIR_P) "$(DESTDIR)$(libdir)"
 	@list='$(lib_LTLIBRARIES)'; test -n "$(libdir)" || list=; \
 	list2=; for p in $$list; do \
 	  if test -f $$p; then \
@@ -455,6 +476,8 @@ install-libLTLIBRARIES: $(lib_LTLIBRARIES)
 	  else :; fi; \
 	done; \
 	test -z "$$list2" || { \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(libdir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(libdir)" || exit 1; \
 	  echo " $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL) $(INSTALL_STRIP_FLAG) $$list2 '$(DESTDIR)$(libdir)'"; \
 	  $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL) $(INSTALL_STRIP_FLAG) $$list2 "$(DESTDIR)$(libdir)"; \
 	}
@@ -495,8 +518,11 @@ distclean-libtool:
 	-rm -f libtool config.lt
 install-pkgconfigDATA: $(pkgconfig_DATA)
 	@$(NORMAL_INSTALL)
-	test -z "$(pkgconfigdir)" || $(MKDIR_P) "$(DESTDIR)$(pkgconfigdir)"
 	@list='$(pkgconfig_DATA)'; test -n "$(pkgconfigdir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(pkgconfigdir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(pkgconfigdir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
@@ -681,13 +707,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
@@ -773,7 +796,7 @@ distcheck: dist
 	*.zip*) \
 	  unzip $(distdir).zip ;;\
 	esac
-	chmod -R a-w $(distdir); chmod a+w $(distdir)
+	chmod -R a-w $(distdir); chmod u+w $(distdir)
 	mkdir $(distdir)/_build
 	mkdir $(distdir)/_inst
 	chmod a-w $(distdir)
diff --git a/NEWS b/NEWS
index 6d78c1a..0f7a3fc 100644
--- a/NEWS
+++ b/NEWS
@@ -1,3 +1,15 @@
+FFTW 3.3.3
+
+* Fix deadlock bug in MPI transforms (thanks to Michael Pippig for the
+  bug report and patch, and to Graham Dennis for the bug report).
+
+* Use 128-bit ARM NEON instructions instead of 64-bits.  This change
+  appears to speed up even ARM processors with a 64-bit NEON pipe.
+
+* Speed improvements for single-precision AVX.
+
+* Speed up planner on machines without "official" cycle counters, such as ARM.
+
 FFTW 3.3.2
 
 * Removed an archaic stack-alignment hack that was failing with
diff --git a/aclocal.m4 b/aclocal.m4
index 40fbae5..7d4f87b 100644
--- a/aclocal.m4
+++ b/aclocal.m4
@@ -1,4 +1,4 @@
-# generated automatically by aclocal 1.11.3 -*- Autoconf -*-
+# generated automatically by aclocal 1.11.6 -*- Autoconf -*-
 
 # Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
 # 2005, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation,
@@ -14,8 +14,8 @@
 
 m4_ifndef([AC_AUTOCONF_VERSION],
   [m4_copy([m4_PACKAGE_VERSION], [AC_AUTOCONF_VERSION])])dnl
-m4_if(m4_defn([AC_AUTOCONF_VERSION]), [2.68],,
-[m4_warning([this file was generated for autoconf 2.68.
+m4_if(m4_defn([AC_AUTOCONF_VERSION]), [2.69],,
+[m4_warning([this file was generated for autoconf 2.69.
 You have another version of autoconf.  It may work, but is not guaranteed to.
 If you have problems, you may need to regenerate the build system entirely.
 To do so, use the procedure documented by the package, typically `autoreconf'.])])
@@ -38,7 +38,7 @@ AC_DEFUN([AM_AUTOMAKE_VERSION],
 [am__api_version='1.11'
 dnl Some users find AM_AUTOMAKE_VERSION and mistake it for a way to
 dnl require some minimum version.  Point them to the right macro.
-m4_if([$1], [1.11.3], [],
+m4_if([$1], [1.11.6], [],
       [AC_FATAL([Do not call $0, use AM_INIT_AUTOMAKE([$1]).])])dnl
 ])
 
@@ -54,7 +54,7 @@ m4_define([_AM_AUTOCONF_VERSION], [])
 # Call AM_AUTOMAKE_VERSION and AM_AUTOMAKE_VERSION so they can be traced.
 # This function is AC_REQUIREd by AM_INIT_AUTOMAKE.
 AC_DEFUN([AM_SET_CURRENT_AUTOMAKE_VERSION],
-[AM_AUTOMAKE_VERSION([1.11.3])dnl
+[AM_AUTOMAKE_VERSION([1.11.6])dnl
 m4_ifndef([AC_AUTOCONF_VERSION],
   [m4_copy([m4_PACKAGE_VERSION], [AC_AUTOCONF_VERSION])])dnl
 _AM_AUTOCONF_VERSION(m4_defn([AC_AUTOCONF_VERSION]))])
diff --git a/api/Makefile.in b/api/Makefile.in
index 14b9e3e..eb62b24 100644
--- a/api/Makefile.in
+++ b/api/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -17,6 +17,23 @@
 
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -97,6 +114,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libapi_la_SOURCES)
 DIST_SOURCES = $(libapi_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 am__vpath_adj_setup = srcdirstrip=`echo "$(srcdir)" | sed 's|.|.|g'`;
 am__vpath_adj = case $$p in \
     $(srcdir)/*) f=`echo "$$p" | sed "s|^$$srcdirstrip/||"`;; \
@@ -455,8 +477,11 @@ clean-libtool:
 	-rm -rf .libs _libs
 install-includeHEADERS: $(include_HEADERS)
 	@$(NORMAL_INSTALL)
-	test -z "$(includedir)" || $(MKDIR_P) "$(DESTDIR)$(includedir)"
 	@list='$(include_HEADERS)'; test -n "$(includedir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(includedir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(includedir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
@@ -473,8 +498,11 @@ uninstall-includeHEADERS:
 	dir='$(DESTDIR)$(includedir)'; $(am__uninstall_files_from_dir)
 install-nodist_includeHEADERS: $(nodist_include_HEADERS)
 	@$(NORMAL_INSTALL)
-	test -z "$(includedir)" || $(MKDIR_P) "$(DESTDIR)$(includedir)"
 	@list='$(nodist_include_HEADERS)'; test -n "$(includedir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(includedir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(includedir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
diff --git a/compile b/compile
index b1f4749..862a14e 100755
--- a/compile
+++ b/compile
@@ -1,7 +1,7 @@
 #! /bin/sh
 # Wrapper for compilers which do not understand '-c -o'.
 
-scriptversion=2012-01-04.17; # UTC
+scriptversion=2012-03-05.13; # UTC
 
 # Copyright (C) 1999, 2000, 2003, 2004, 2005, 2009, 2010, 2012 Free
 # Software Foundation, Inc.
@@ -79,6 +79,48 @@ func_file_conv ()
   esac
 }
 
+# func_cl_dashL linkdir
+# Make cl look for libraries in LINKDIR
+func_cl_dashL ()
+{
+  func_file_conv "$1"
+  if test -z "$lib_path"; then
+    lib_path=$file
+  else
+    lib_path="$lib_path;$file"
+  fi
+  linker_opts="$linker_opts -LIBPATH:$file"
+}
+
+# func_cl_dashl library
+# Do a library search-path lookup for cl
+func_cl_dashl ()
+{
+  lib=$1
+  found=no
+  save_IFS=$IFS
+  IFS=';'
+  for dir in $lib_path $LIB
+  do
+    IFS=$save_IFS
+    if $shared && test -f "$dir/$lib.dll.lib"; then
+      found=yes
+      lib=$dir/$lib.dll.lib
+      break
+    fi
+    if test -f "$dir/$lib.lib"; then
+      found=yes
+      lib=$dir/$lib.lib
+      break
+    fi
+  done
+  IFS=$save_IFS
+
+  if test "$found" != yes; then
+    lib=$lib.lib
+  fi
+}
+
 # func_cl_wrapper cl arg...
 # Adjust compile command to suit cl
 func_cl_wrapper ()
@@ -109,43 +151,34 @@ func_cl_wrapper ()
 	      ;;
 	  esac
 	  ;;
+	-I)
+	  eat=1
+	  func_file_conv "$2" mingw
+	  set x "$@" -I"$file"
+	  shift
+	  ;;
 	-I*)
 	  func_file_conv "${1#-I}" mingw
 	  set x "$@" -I"$file"
 	  shift
 	  ;;
+	-l)
+	  eat=1
+	  func_cl_dashl "$2"
+	  set x "$@" "$lib"
+	  shift
+	  ;;
 	-l*)
-	  lib=${1#-l}
-	  found=no
-	  save_IFS=$IFS
-	  IFS=';'
-	  for dir in $lib_path $LIB
-	  do
-	    IFS=$save_IFS
-	    if $shared && test -f "$dir/$lib.dll.lib"; then
-	      found=yes
-	      set x "$@" "$dir/$lib.dll.lib"
-	      break
-	    fi
-	    if test -f "$dir/$lib.lib"; then
-	      found=yes
-	      set x "$@" "$dir/$lib.lib"
-	      break
-	    fi
-	  done
-	  IFS=$save_IFS
-
-	  test "$found" != yes && set x "$@" "$lib.lib"
+	  func_cl_dashl "${1#-l}"
+	  set x "$@" "$lib"
 	  shift
 	  ;;
+	-L)
+	  eat=1
+	  func_cl_dashL "$2"
+	  ;;
 	-L*)
-	  func_file_conv "${1#-L}"
-	  if test -z "$lib_path"; then
-	    lib_path=$file
-	  else
-	    lib_path="$lib_path;$file"
-	  fi
-	  linker_opts="$linker_opts -LIBPATH:$file"
+	  func_cl_dashL "${1#-L}"
 	  ;;
 	-static)
 	  shared=false
diff --git a/config.sub b/config.sub
index c894da4..6205f84 100755
--- a/config.sub
+++ b/config.sub
@@ -4,7 +4,7 @@
 #   2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010,
 #   2011, 2012 Free Software Foundation, Inc.
 
-timestamp='2012-02-10'
+timestamp='2012-04-18'
 
 # This file is (in principle) common to ALL GNU software.
 # The presence of a machine in this file suggests that SOME GNU software
@@ -225,6 +225,12 @@ case $os in
 	-isc*)
 		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
 		;;
+	-lynx*178)
+		os=-lynxos178
+		;;
+	-lynx*5)
+		os=-lynxos5
+		;;
 	-lynx*)
 		os=-lynxos
 		;;
@@ -1537,6 +1543,9 @@ case $basic_machine in
 	c4x-* | tic4x-*)
 		os=-coff
 		;;
+	hexagon-*)
+		os=-elf
+		;;
 	tic54x-*)
 		os=-coff
 		;;
diff --git a/configure b/configure
index 4d5c607..431d3ec 100755
--- a/configure
+++ b/configure
@@ -1,13 +1,11 @@
 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.68 for fftw 3.3.2.
+# Generated by GNU Autoconf 2.69 for fftw 3.3.3.
 #
 # Report bugs to <fftw at fftw.org>.
 #
 #
-# Copyright (C) 1992, 1993, 1994, 1995, 1996, 1998, 1999, 2000, 2001,
-# 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 Free Software
-# Foundation, Inc.
+# Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
 #
 #
 # This configure script is free software; the Free Software Foundation
@@ -136,6 +134,31 @@ export LANGUAGE
 # CDPATH.
 (unset CDPATH) >/dev/null 2>&1 && unset CDPATH
 
+# Use a proper internal environment variable to ensure we don't fall
+  # into an infinite loop, continuously re-executing ourselves.
+  if test x"${_as_can_reexec}" != xno && test "x$CONFIG_SHELL" != x; then
+    _as_can_reexec=no; export _as_can_reexec;
+    # We cannot yet assume a decent shell, so we have to provide a
+# neutralization value for shells without unset; and this also
+# works around shells that cannot unset nonexistent variables.
+# Preserve -v and -x to the replacement shell.
+BASH_ENV=/dev/null
+ENV=/dev/null
+(unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV
+case $- in # ((((
+  *v*x* | *x*v* ) as_opts=-vx ;;
+  *v* ) as_opts=-v ;;
+  *x* ) as_opts=-x ;;
+  * ) as_opts= ;;
+esac
+exec $CONFIG_SHELL $as_opts "$as_myself" ${1+"$@"}
+# Admittedly, this is quite paranoid, since all the known shells bail
+# out after a failed `exec'.
+$as_echo "$0: could not re-execute with $CONFIG_SHELL" >&2
+as_fn_exit 255
+  fi
+  # We don't want this to propagate to other subprocesses.
+          { _as_can_reexec=; unset _as_can_reexec;}
 if test "x$CONFIG_SHELL" = x; then
   as_bourne_compatible="if test -n \"\${ZSH_VERSION+set}\" && (emulate sh) >/dev/null 2>&1; then :
   emulate sh
@@ -169,7 +192,8 @@ if ( set x; as_fn_ret_success y && test x = \"\$1\" ); then :
 else
   exitcode=1; echo positional parameters were not saved.
 fi
-test x\$exitcode = x0 || exit 1"
+test x\$exitcode = x0 || exit 1
+test -x / || exit 1"
   as_suggested="  as_lineno_1=";as_suggested=$as_suggested$LINENO;as_suggested=$as_suggested" as_lineno_1a=\$LINENO
   as_lineno_2=";as_suggested=$as_suggested$LINENO;as_suggested=$as_suggested" as_lineno_2a=\$LINENO
   eval 'test \"x\$as_lineno_1'\$as_run'\" != \"x\$as_lineno_2'\$as_run'\" &&
@@ -222,21 +246,25 @@ IFS=$as_save_IFS
 
 
       if test "x$CONFIG_SHELL" != x; then :
-  # We cannot yet assume a decent shell, so we have to provide a
-	# neutralization value for shells without unset; and this also
-	# works around shells that cannot unset nonexistent variables.
-	# Preserve -v and -x to the replacement shell.
-	BASH_ENV=/dev/null
-	ENV=/dev/null
-	(unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV
-	export CONFIG_SHELL
-	case $- in # ((((
-	  *v*x* | *x*v* ) as_opts=-vx ;;
-	  *v* ) as_opts=-v ;;
-	  *x* ) as_opts=-x ;;
-	  * ) as_opts= ;;
-	esac
-	exec "$CONFIG_SHELL" $as_opts "$as_myself" ${1+"$@"}
+  export CONFIG_SHELL
+             # We cannot yet assume a decent shell, so we have to provide a
+# neutralization value for shells without unset; and this also
+# works around shells that cannot unset nonexistent variables.
+# Preserve -v and -x to the replacement shell.
+BASH_ENV=/dev/null
+ENV=/dev/null
+(unset BASH_ENV) >/dev/null 2>&1 && unset BASH_ENV ENV
+case $- in # ((((
+  *v*x* | *x*v* ) as_opts=-vx ;;
+  *v* ) as_opts=-v ;;
+  *x* ) as_opts=-x ;;
+  * ) as_opts= ;;
+esac
+exec $CONFIG_SHELL $as_opts "$as_myself" ${1+"$@"}
+# Admittedly, this is quite paranoid, since all the known shells bail
+# out after a failed `exec'.
+$as_echo "$0: could not re-execute with $CONFIG_SHELL" >&2
+exit 255
 fi
 
     if test x$as_have_required = xno; then :
@@ -339,6 +367,14 @@ $as_echo X"$as_dir" |
 
 
 } # as_fn_mkdir_p
+
+# as_fn_executable_p FILE
+# -----------------------
+# Test if FILE is an executable regular file.
+as_fn_executable_p ()
+{
+  test -f "$1" && test -x "$1"
+} # as_fn_executable_p
 # as_fn_append VAR VALUE
 # ----------------------
 # Append the text in VALUE to the end of the definition contained in VAR. Take
@@ -460,6 +496,10 @@ as_cr_alnum=$as_cr_Letters$as_cr_digits
   chmod +x "$as_me.lineno" ||
     { $as_echo "$as_me: error: cannot create $as_me.lineno; rerun with a POSIX shell" >&2; as_fn_exit 1; }
 
+  # If we had to re-execute with $CONFIG_SHELL, we're ensured to have
+  # already done that, so ensure we don't try to do so again and fall
+  # in an infinite loop.  This has already happened in practice.
+  _as_can_reexec=no; export _as_can_reexec
   # Don't try to exec as it changes $[0], causing all sort of problems
   # (the dirname of $[0] is not the place where we might find the
   # original and so on.  Autoconf is especially sensitive to this).
@@ -494,16 +534,16 @@ if (echo >conf$$.file) 2>/dev/null; then
     # ... but there are two gotchas:
     # 1) On MSYS, both `ln -s file dir' and `ln file dir' fail.
     # 2) DJGPP < 2.04 has no symlinks; `ln -s' creates a wrapper executable.
-    # In both cases, we have to default to `cp -p'.
+    # In both cases, we have to default to `cp -pR'.
     ln -s conf$$.file conf$$.dir 2>/dev/null && test ! -f conf$$.exe ||
-      as_ln_s='cp -p'
+      as_ln_s='cp -pR'
   elif ln conf$$.file conf$$ 2>/dev/null; then
     as_ln_s=ln
   else
-    as_ln_s='cp -p'
+    as_ln_s='cp -pR'
   fi
 else
-  as_ln_s='cp -p'
+  as_ln_s='cp -pR'
 fi
 rm -f conf$$ conf$$.exe conf$$.dir/conf$$.file conf$$.file
 rmdir conf$$.dir 2>/dev/null
@@ -515,28 +555,8 @@ else
   as_mkdir_p=false
 fi
 
-if test -x / >/dev/null 2>&1; then
-  as_test_x='test -x'
-else
-  if ls -dL / >/dev/null 2>&1; then
-    as_ls_L_option=L
-  else
-    as_ls_L_option=
-  fi
-  as_test_x='
-    eval sh -c '\''
-      if test -d "$1"; then
-	test -d "$1/.";
-      else
-	case $1 in #(
-	-*)set "./$1";;
-	esac;
-	case `ls -ld'$as_ls_L_option' "$1" 2>/dev/null` in #((
-	???[sx]*):;;*)false;;esac;fi
-    '\'' sh
-  '
-fi
-as_executable_p=$as_test_x
+as_test_x='test -x'
+as_executable_p=as_fn_executable_p
 
 # Sed expression to map a string onto a valid CPP name.
 as_tr_cpp="eval sed 'y%*$as_cr_letters%P$as_cr_LETTERS%;s%[^_$as_cr_alnum]%_%g'"
@@ -570,8 +590,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='fftw'
 PACKAGE_TARNAME='fftw'
-PACKAGE_VERSION='3.3.2'
-PACKAGE_STRING='fftw 3.3.2'
+PACKAGE_VERSION='3.3.3'
+PACKAGE_STRING='fftw 3.3.3'
 PACKAGE_BUGREPORT='fftw at fftw.org'
 PACKAGE_URL=''
 
@@ -1287,8 +1307,6 @@ target=$target_alias
 if test "x$host_alias" != x; then
   if test "x$build_alias" = x; then
     cross_compiling=maybe
-    $as_echo "$as_me: WARNING: if you wanted to set the --build type, don't use --host.
-    If a cross compiler is detected then cross compile mode will be used" >&2
   elif test "x$build_alias" != "x$host_alias"; then
     cross_compiling=yes
   fi
@@ -1374,7 +1392,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures fftw 3.3.2 to adapt to many kinds of systems.
+\`configure' configures fftw 3.3.3 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1444,7 +1462,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of fftw 3.3.2:";;
+     short | recursive ) echo "Configuration of fftw 3.3.3:";;
    esac
   cat <<\_ACEOF
 
@@ -1586,10 +1604,10 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-fftw configure 3.3.2
-generated by GNU Autoconf 2.68
+fftw configure 3.3.3
+generated by GNU Autoconf 2.69
 
-Copyright (C) 2010 Free Software Foundation, Inc.
+Copyright (C) 2012 Free Software Foundation, Inc.
 This configure script is free software; the Free Software Foundation
 gives unlimited permission to copy, distribute and modify it.
 _ACEOF
@@ -1665,7 +1683,7 @@ $as_echo "$ac_try_echo"; } >&5
 	 test ! -s conftest.err
        } && test -s conftest$ac_exeext && {
 	 test "$cross_compiling" = yes ||
-	 $as_test_x conftest$ac_exeext
+	 test -x conftest$ac_exeext
        }; then :
   ac_retval=0
 else
@@ -1878,7 +1896,8 @@ int
 main ()
 {
 static int test_array [1 - 2 * !(($2) >= 0)];
-test_array [0] = 0
+test_array [0] = 0;
+return test_array [0];
 
   ;
   return 0;
@@ -1894,7 +1913,8 @@ int
 main ()
 {
 static int test_array [1 - 2 * !(($2) <= $ac_mid)];
-test_array [0] = 0
+test_array [0] = 0;
+return test_array [0];
 
   ;
   return 0;
@@ -1920,7 +1940,8 @@ int
 main ()
 {
 static int test_array [1 - 2 * !(($2) < 0)];
-test_array [0] = 0
+test_array [0] = 0;
+return test_array [0];
 
   ;
   return 0;
@@ -1936,7 +1957,8 @@ int
 main ()
 {
 static int test_array [1 - 2 * !(($2) >= $ac_mid)];
-test_array [0] = 0
+test_array [0] = 0;
+return test_array [0];
 
   ;
   return 0;
@@ -1970,7 +1992,8 @@ int
 main ()
 {
 static int test_array [1 - 2 * !(($2) <= $ac_mid)];
-test_array [0] = 0
+test_array [0] = 0;
+return test_array [0];
 
   ;
   return 0;
@@ -2295,7 +2318,7 @@ $as_echo "$ac_try_echo"; } >&5
 	 test ! -s conftest.err
        } && test -s conftest$ac_exeext && {
 	 test "$cross_compiling" = yes ||
-	 $as_test_x conftest$ac_exeext
+	 test -x conftest$ac_exeext
        }; then :
   ac_retval=0
 else
@@ -2317,8 +2340,8 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by fftw $as_me 3.3.2, which was
-generated by GNU Autoconf 2.68.  Invocation command line was
+It was created by fftw $as_me 3.3.3, which was
+generated by GNU Autoconf 2.69.  Invocation command line was
 
   $ $0 $@
 
@@ -2739,7 +2762,7 @@ case $as_dir/ in #((
     # by default.
     for ac_prog in ginstall scoinst install; do
       for ac_exec_ext in '' $ac_executable_extensions; do
-	if { test -f "$as_dir/$ac_prog$ac_exec_ext" && $as_test_x "$as_dir/$ac_prog$ac_exec_ext"; }; then
+	if as_fn_executable_p "$as_dir/$ac_prog$ac_exec_ext"; then
 	  if test $ac_prog = install &&
 	    grep dspmsg "$as_dir/$ac_prog$ac_exec_ext" >/dev/null 2>&1; then
 	    # AIX install.  It has an incompatible calling convention.
@@ -2908,7 +2931,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_STRIP="${ac_tool_prefix}strip"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -2948,7 +2971,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_STRIP="strip"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -2999,7 +3022,7 @@ do
   test -z "$as_dir" && as_dir=.
     for ac_prog in mkdir gmkdir; do
 	 for ac_exec_ext in '' $ac_executable_extensions; do
-	   { test -f "$as_dir/$ac_prog$ac_exec_ext" && $as_test_x "$as_dir/$ac_prog$ac_exec_ext"; } || continue
+	   as_fn_executable_p "$as_dir/$ac_prog$ac_exec_ext" || continue
 	   case `"$as_dir/$ac_prog$ac_exec_ext" --version 2>&1` in #(
 	     'mkdir (GNU coreutils) '* | \
 	     'mkdir (coreutils) '* | \
@@ -3052,7 +3075,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_AWK="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -3138,7 +3161,7 @@ fi
 
 # Define the identity of the package.
  PACKAGE='fftw'
- VERSION='3.3.2'
+ VERSION='3.3.3'
 
 
 cat >>confdefs.h <<_ACEOF
@@ -3698,7 +3721,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_CC="${ac_tool_prefix}gcc"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -3738,7 +3761,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_CC="gcc"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -3791,7 +3814,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_CC="${ac_tool_prefix}cc"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -3832,7 +3855,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     if test "$as_dir/$ac_word$ac_exec_ext" = "/usr/ucb/cc"; then
        ac_prog_rejected=yes
        continue
@@ -3890,7 +3913,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_CC="$ac_tool_prefix$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -3934,7 +3957,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_CC="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -4380,8 +4403,7 @@ cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 #include <stdarg.h>
 #include <stdio.h>
-#include <sys/types.h>
-#include <sys/stat.h>
+struct stat;
 /* Most of the following tests are stolen from RCS 5.7's src/conf.sh.  */
 struct buf { int x; };
 FILE * (*rcsopen) (struct buf *, struct stat *, int);
@@ -4969,7 +4991,7 @@ main ()
   return 0;
 }
 _ACEOF
-for ac_arg in '' -std=gnu99 -std=c99 -c99 -AC99 -xc99=all -qlanglvl=extc99
+for ac_arg in '' -std=gnu99 -std=c99 -c99 -AC99 -D_STDC_C99= -qlanglvl=extc99
 do
   CC="$ac_save_CC $ac_arg"
   if ac_fn_c_try_compile "$LINENO"; then :
@@ -5009,8 +5031,7 @@ cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 #include <stdarg.h>
 #include <stdio.h>
-#include <sys/types.h>
-#include <sys/stat.h>
+struct stat;
 /* Most of the following tests are stolen from RCS 5.7's src/conf.sh.  */
 struct buf { int x; };
 FILE * (*rcsopen) (struct buf *, struct stat *, int);
@@ -5176,7 +5197,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_AS="${ac_tool_prefix}as"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -5216,7 +5237,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_AS="as"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -5268,7 +5289,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_DLLTOOL="${ac_tool_prefix}dlltool"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -5308,7 +5329,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_DLLTOOL="dlltool"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -5360,7 +5381,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_OBJDUMP="${ac_tool_prefix}objdump"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -5400,7 +5421,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_OBJDUMP="objdump"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -5574,7 +5595,7 @@ do
     for ac_prog in sed gsed; do
     for ac_exec_ext in '' $ac_executable_extensions; do
       ac_path_SED="$as_dir/$ac_prog$ac_exec_ext"
-      { test -f "$ac_path_SED" && $as_test_x "$ac_path_SED"; } || continue
+      as_fn_executable_p "$ac_path_SED" || continue
 # Check for GNU ac_path_SED and select it if it is found.
   # Check for GNU $ac_path_SED
 case `"$ac_path_SED" --version 2>&1` in
@@ -5650,7 +5671,7 @@ do
     for ac_prog in grep ggrep; do
     for ac_exec_ext in '' $ac_executable_extensions; do
       ac_path_GREP="$as_dir/$ac_prog$ac_exec_ext"
-      { test -f "$ac_path_GREP" && $as_test_x "$ac_path_GREP"; } || continue
+      as_fn_executable_p "$ac_path_GREP" || continue
 # Check for GNU ac_path_GREP and select it if it is found.
   # Check for GNU $ac_path_GREP
 case `"$ac_path_GREP" --version 2>&1` in
@@ -5716,7 +5737,7 @@ do
     for ac_prog in egrep; do
     for ac_exec_ext in '' $ac_executable_extensions; do
       ac_path_EGREP="$as_dir/$ac_prog$ac_exec_ext"
-      { test -f "$ac_path_EGREP" && $as_test_x "$ac_path_EGREP"; } || continue
+      as_fn_executable_p "$ac_path_EGREP" || continue
 # Check for GNU ac_path_EGREP and select it if it is found.
   # Check for GNU $ac_path_EGREP
 case `"$ac_path_EGREP" --version 2>&1` in
@@ -5783,7 +5804,7 @@ do
     for ac_prog in fgrep; do
     for ac_exec_ext in '' $ac_executable_extensions; do
       ac_path_FGREP="$as_dir/$ac_prog$ac_exec_ext"
-      { test -f "$ac_path_FGREP" && $as_test_x "$ac_path_FGREP"; } || continue
+      as_fn_executable_p "$ac_path_FGREP" || continue
 # Check for GNU ac_path_FGREP and select it if it is found.
   # Check for GNU $ac_path_FGREP
 case `"$ac_path_FGREP" --version 2>&1` in
@@ -6039,7 +6060,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_DUMPBIN="$ac_tool_prefix$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -6083,7 +6104,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_DUMPBIN="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -6496,7 +6517,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_OBJDUMP="${ac_tool_prefix}objdump"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -6536,7 +6557,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_OBJDUMP="objdump"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -6659,10 +6680,6 @@ freebsd* | dragonfly*)
   fi
   ;;
 
-gnu*)
-  lt_cv_deplibs_check_method=pass_all
-  ;;
-
 haiku*)
   lt_cv_deplibs_check_method=pass_all
   ;;
@@ -6701,7 +6718,7 @@ irix5* | irix6* | nonstopux*)
   ;;
 
 # This must be glibc/ELF.
-linux* | k*bsd*-gnu | kopensolaris*-gnu)
+linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
   lt_cv_deplibs_check_method=pass_all
   ;;
 
@@ -6839,7 +6856,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_DLLTOOL="${ac_tool_prefix}dlltool"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -6879,7 +6896,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_DLLTOOL="dlltool"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -6979,7 +6996,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_AR="$ac_tool_prefix$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7023,7 +7040,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_AR="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7148,7 +7165,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_STRIP="${ac_tool_prefix}strip"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7188,7 +7205,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_STRIP="strip"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7247,7 +7264,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_RANLIB="${ac_tool_prefix}ranlib"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7287,7 +7304,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_RANLIB="ranlib"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7936,7 +7953,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_MANIFEST_TOOL="${ac_tool_prefix}mt"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -7976,7 +7993,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_MANIFEST_TOOL="mt"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8056,7 +8073,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_DSYMUTIL="${ac_tool_prefix}dsymutil"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8096,7 +8113,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_DSYMUTIL="dsymutil"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8148,7 +8165,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_NMEDIT="${ac_tool_prefix}nmedit"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8188,7 +8205,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_NMEDIT="nmedit"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8240,7 +8257,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_LIPO="${ac_tool_prefix}lipo"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8280,7 +8297,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_LIPO="lipo"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8332,7 +8349,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_OTOOL="${ac_tool_prefix}otool"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8372,7 +8389,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_OTOOL="otool"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8424,7 +8441,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_OTOOL64="${ac_tool_prefix}otool64"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -8464,7 +8481,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_OTOOL64="otool64"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -9592,7 +9609,7 @@ lt_prog_compiler_static=
       lt_prog_compiler_static='-non_shared'
       ;;
 
-    linux* | k*bsd*-gnu | kopensolaris*-gnu)
+    linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
       case $cc_basename in
       # old Intel for x86_64 which still supported -KPIC.
       ecc*)
@@ -11762,17 +11779,6 @@ freebsd* | dragonfly*)
   esac
   ;;
 
-gnu*)
-  version_type=linux # correct to gnu/linux during the next big refactor
-  need_lib_prefix=no
-  need_version=no
-  library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}${major} ${libname}${shared_ext}'
-  soname_spec='${libname}${release}${shared_ext}$major'
-  shlibpath_var=LD_LIBRARY_PATH
-  shlibpath_overrides_runpath=no
-  hardcode_into_libs=yes
-  ;;
-
 haiku*)
   version_type=linux # correct to gnu/linux during the next big refactor
   need_lib_prefix=no
@@ -11889,7 +11895,7 @@ linux*oldld* | linux*aout* | linux*coff*)
   ;;
 
 # This must be glibc/ELF.
-linux* | k*bsd*-gnu | kopensolaris*-gnu)
+linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
   version_type=linux # correct to gnu/linux during the next big refactor
   need_lib_prefix=no
   need_version=no
@@ -12952,7 +12958,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_OCAMLBUILD="ocamlbuild"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -13010,7 +13016,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_MPICC="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -13043,7 +13049,7 @@ test -n "$MPICC" || MPICC="$CC"
 if test x = x"$MPILIBS"; then
 	ac_fn_c_check_func "$LINENO" "MPI_Init" "ac_cv_func_MPI_Init"
 if test "x$ac_cv_func_MPI_Init" = xyes; then :
-  MPILIBS=""
+  MPILIBS=" "
 fi
 
 fi
@@ -13188,7 +13194,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_MPIRUN="mpirun"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -13453,9 +13459,13 @@ fi
      # tune for the host by default
       { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether C compiler accepts -mtune=native" >&5
 $as_echo_n "checking whether C compiler accepts -mtune=native... " >&6; }
-ax_save_FLAGS=$CFLAGS
-   CFLAGS="-mtune=native"
-   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+if ${ax_cv_c_flags__mtune_native+:} false; then :
+  $as_echo_n "(cached) " >&6
+else
+
+      ax_save_FLAGS=$CFLAGS
+      CFLAGS="-mtune=native"
+      cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 
 int
@@ -13467,13 +13477,15 @@ main ()
 }
 _ACEOF
 if ac_fn_c_try_compile "$LINENO"; then :
-  eval `$as_echo "ax_cv_c_flags_-mtune=native" | $as_tr_sh`=yes
+  ax_cv_c_flags__mtune_native=yes
 else
-  eval `$as_echo "ax_cv_c_flags_-mtune=native" | $as_tr_sh`=no
+  ax_cv_c_flags__mtune_native=no
 fi
 rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
-   CFLAGS=$ax_save_FLAGS
-eval ax_check_compiler_flags=$`$as_echo "ax_cv_c_flags_-mtune=native" | $as_tr_sh`
+      CFLAGS=$ax_save_FLAGS
+fi
+
+eval ax_check_compiler_flags=$ax_cv_c_flags__mtune_native
 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ax_check_compiler_flags" >&5
 $as_echo "$ax_check_compiler_flags" >&6; }
 if test "x$ax_check_compiler_flags" = xyes; then
@@ -13949,9 +13961,13 @@ if test "x$ax_check_compiler_flags" = xyes; then
 else
 	 { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether C compiler accepts -maltivec -mabi=altivec" >&5
 $as_echo_n "checking whether C compiler accepts -maltivec -mabi=altivec... " >&6; }
-ax_save_FLAGS=$CFLAGS
-   CFLAGS="-maltivec -mabi=altivec"
-   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+if ${ax_cv_c_flags__maltivec__mabi_altivec+:} false; then :
+  $as_echo_n "(cached) " >&6
+else
+
+      ax_save_FLAGS=$CFLAGS
+      CFLAGS="-maltivec -mabi=altivec"
+      cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 
 int
@@ -13963,13 +13979,15 @@ main ()
 }
 _ACEOF
 if ac_fn_c_try_compile "$LINENO"; then :
-  eval `$as_echo "ax_cv_c_flags_-maltivec -mabi=altivec" | $as_tr_sh`=yes
+  ax_cv_c_flags__maltivec__mabi_altivec=yes
 else
-  eval `$as_echo "ax_cv_c_flags_-maltivec -mabi=altivec" | $as_tr_sh`=no
+  ax_cv_c_flags__maltivec__mabi_altivec=no
 fi
 rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
-   CFLAGS=$ax_save_FLAGS
-eval ax_check_compiler_flags=$`$as_echo "ax_cv_c_flags_-maltivec -mabi=altivec" | $as_tr_sh`
+      CFLAGS=$ax_save_FLAGS
+fi
+
+eval ax_check_compiler_flags=$ax_cv_c_flags__maltivec__mabi_altivec
 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ax_check_compiler_flags" >&5
 $as_echo "$ax_check_compiler_flags" >&6; }
 if test "x$ax_check_compiler_flags" = xyes; then
@@ -14021,9 +14039,13 @@ fi
 	if test "$have_neon" = "yes" -a "x$NEON_CFLAGS" = x; then
 	     { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether C compiler accepts -mfpu=neon" >&5
 $as_echo_n "checking whether C compiler accepts -mfpu=neon... " >&6; }
-ax_save_FLAGS=$CFLAGS
-   CFLAGS="-mfpu=neon"
-   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+if ${ax_cv_c_flags__mfpu_neon+:} false; then :
+  $as_echo_n "(cached) " >&6
+else
+
+      ax_save_FLAGS=$CFLAGS
+      CFLAGS="-mfpu=neon"
+      cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 
 int
@@ -14035,13 +14057,15 @@ main ()
 }
 _ACEOF
 if ac_fn_c_try_compile "$LINENO"; then :
-  eval `$as_echo "ax_cv_c_flags_-mfpu=neon" | $as_tr_sh`=yes
+  ax_cv_c_flags__mfpu_neon=yes
 else
-  eval `$as_echo "ax_cv_c_flags_-mfpu=neon" | $as_tr_sh`=no
+  ax_cv_c_flags__mfpu_neon=no
 fi
 rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
-   CFLAGS=$ax_save_FLAGS
-eval ax_check_compiler_flags=$`$as_echo "ax_cv_c_flags_-mfpu=neon" | $as_tr_sh`
+      CFLAGS=$ax_save_FLAGS
+fi
+
+eval ax_check_compiler_flags=$ax_cv_c_flags__mfpu_neon
 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ax_check_compiler_flags" >&5
 $as_echo "$ax_check_compiler_flags" >&6; }
 if test "x$ax_check_compiler_flags" = xyes; then
@@ -14257,11 +14281,11 @@ else
 int
 main ()
 {
-/* FIXME: Include the comments suggested by Paul. */
+
 #ifndef __cplusplus
-  /* Ultrix mips cc rejects this.  */
+  /* Ultrix mips cc rejects this sort of thing.  */
   typedef int charset[2];
-  const charset cs;
+  const charset cs = { 0, 0 };
   /* SunOS 4.1.1 cc rejects this.  */
   char const *const *pcpcc;
   char **ppc;
@@ -14278,8 +14302,9 @@ main ()
   ++pcpcc;
   ppc = (char**) pcpcc;
   pcpcc = (char const *const *) ppc;
-  { /* SCO 3.2v4 cc rejects this.  */
-    char *t;
+  { /* SCO 3.2v4 cc rejects this sort of thing.  */
+    char tx;
+    char *t = &tx;
     char const *s = 0 ? (char *) 0 : (char const *) 0;
 
     *t++ = 0;
@@ -14295,10 +14320,10 @@ main ()
     iptr p = 0;
     ++p;
   }
-  { /* AIX XL C 1.02.0.0 rejects this saying
+  { /* AIX XL C 1.02.0.0 rejects this sort of thing, saying
        "k.c", line 2.27: 1506-025 (S) Operand must be a modifiable lvalue. */
-    struct s { int j; const int *ap[3]; };
-    struct s *b; b->j = 5;
+    struct s { int j; const int *ap[3]; } bx;
+    struct s *b = &bx; b->j = 5;
   }
   { /* ULTRIX-32 V3.1 (Rev 9) vcc rejects this */
     const int foo = 10;
@@ -15014,23 +15039,20 @@ else
 /* end confdefs.h.  */
 $ac_includes_default
 int
-find_stack_direction ()
+find_stack_direction (int *addr, int depth)
 {
-  static char *addr = 0;
-  auto char dummy;
-  if (addr == 0)
-    {
-      addr = &dummy;
-      return find_stack_direction ();
-    }
-  else
-    return (&dummy > addr) ? 1 : -1;
+  int dir, dummy = 0;
+  if (! addr)
+    addr = &dummy;
+  *addr = addr < &dummy ? 1 : addr == &dummy ? 0 : -1;
+  dir = depth ? find_stack_direction (addr, depth - 1) : 0;
+  return dir + dummy;
 }
 
 int
-main ()
+main (int argc, char **argv)
 {
-  return find_stack_direction () < 0;
+  return find_stack_direction (0, argc + !argv + 20) < 0;
 }
 _ACEOF
 if ac_fn_c_try_run "$LINENO"; then :
@@ -15494,9 +15516,13 @@ ax_gcc_aligns_stack=no
 if test "$GCC" = "yes"; then
  { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether C compiler accepts -mpreferred-stack-boundary=4" >&5
 $as_echo_n "checking whether C compiler accepts -mpreferred-stack-boundary=4... " >&6; }
-ax_save_FLAGS=$CFLAGS
-   CFLAGS="-mpreferred-stack-boundary=4"
-   cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+if ${ax_cv_c_flags__mpreferred_stack_boundary_4+:} false; then :
+  $as_echo_n "(cached) " >&6
+else
+
+      ax_save_FLAGS=$CFLAGS
+      CFLAGS="-mpreferred-stack-boundary=4"
+      cat confdefs.h - <<_ACEOF >conftest.$ac_ext
 /* end confdefs.h.  */
 
 int
@@ -15508,13 +15534,15 @@ main ()
 }
 _ACEOF
 if ac_fn_c_try_compile "$LINENO"; then :
-  eval `$as_echo "ax_cv_c_flags_-mpreferred-stack-boundary=4" | $as_tr_sh`=yes
+  ax_cv_c_flags__mpreferred_stack_boundary_4=yes
 else
-  eval `$as_echo "ax_cv_c_flags_-mpreferred-stack-boundary=4" | $as_tr_sh`=no
+  ax_cv_c_flags__mpreferred_stack_boundary_4=no
 fi
 rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
-   CFLAGS=$ax_save_FLAGS
-eval ax_check_compiler_flags=$`$as_echo "ax_cv_c_flags_-mpreferred-stack-boundary=4" | $as_tr_sh`
+      CFLAGS=$ax_save_FLAGS
+fi
+
+eval ax_check_compiler_flags=$ax_cv_c_flags__mpreferred_stack_boundary_4
 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ax_check_compiler_flags" >&5
 $as_echo "$ax_check_compiler_flags" >&6; }
 if test "x$ax_check_compiler_flags" = xyes; then
@@ -15666,7 +15694,7 @@ ac_compile='$F77 -c $FFLAGS conftest.$ac_ext >&5'
 ac_link='$F77 -o conftest$ac_exeext $FFLAGS $LDFLAGS conftest.$ac_ext $LIBS >&5'
 ac_compiler_gnu=$ac_cv_f77_compiler_gnu
 if test -n "$ac_tool_prefix"; then
-  for ac_prog in g77 xlf f77 frt pgf77 cf77 fort77 fl32 af77 xlf90 f90 pgf90 pghpf epcf90 gfortran g95 xlf95 f95 fort ifort ifc efc pgfortran pgf95 lf95 ftn
+  for ac_prog in g77 xlf f77 frt pgf77 cf77 fort77 fl32 af77 xlf90 f90 pgf90 pghpf epcf90 gfortran g95 xlf95 f95 fort ifort ifc efc pgfortran pgf95 lf95 ftn nagfor
   do
     # Extract the first word of "$ac_tool_prefix$ac_prog", so it can be a program name with args.
 set dummy $ac_tool_prefix$ac_prog; ac_word=$2
@@ -15684,7 +15712,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_F77="$ac_tool_prefix$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -15710,7 +15738,7 @@ fi
 fi
 if test -z "$F77"; then
   ac_ct_F77=$F77
-  for ac_prog in g77 xlf f77 frt pgf77 cf77 fort77 fl32 af77 xlf90 f90 pgf90 pghpf epcf90 gfortran g95 xlf95 f95 fort ifort ifc efc pgfortran pgf95 lf95 ftn
+  for ac_prog in g77 xlf f77 frt pgf77 cf77 fort77 fl32 af77 xlf90 f90 pgf90 pghpf epcf90 gfortran g95 xlf95 f95 fort ifort ifc efc pgfortran pgf95 lf95 ftn nagfor
 do
   # Extract the first word of "$ac_prog", so it can be a program name with args.
 set dummy $ac_prog; ac_word=$2
@@ -15728,7 +15756,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_ac_ct_F77="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -16170,7 +16198,7 @@ lt_prog_compiler_static_F77=
       lt_prog_compiler_static_F77='-non_shared'
       ;;
 
-    linux* | k*bsd*-gnu | kopensolaris*-gnu)
+    linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
       case $cc_basename in
       # old Intel for x86_64 which still supported -KPIC.
       ecc*)
@@ -18110,17 +18138,6 @@ freebsd* | dragonfly*)
   esac
   ;;
 
-gnu*)
-  version_type=linux # correct to gnu/linux during the next big refactor
-  need_lib_prefix=no
-  need_version=no
-  library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}${major} ${libname}${shared_ext}'
-  soname_spec='${libname}${release}${shared_ext}$major'
-  shlibpath_var=LD_LIBRARY_PATH
-  shlibpath_overrides_runpath=no
-  hardcode_into_libs=yes
-  ;;
-
 haiku*)
   version_type=linux # correct to gnu/linux during the next big refactor
   need_lib_prefix=no
@@ -18237,7 +18254,7 @@ linux*oldld* | linux*aout* | linux*coff*)
   ;;
 
 # This must be glibc/ELF.
-linux* | k*bsd*-gnu | kopensolaris*-gnu)
+linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
   version_type=linux # correct to gnu/linux during the next big refactor
   need_lib_prefix=no
   need_version=no
@@ -18665,9 +18682,16 @@ ac_f77_v_output="`echo $ac_f77_v_output |
 #        that detects unbalanced quotes in FLIBS should be implemented
 #        and (ugh) tested at some point.
 case $ac_f77_v_output in
-  # If we are using xlf then replace all the commas with spaces.
+  # With xlf replace commas with spaces,
+  # and remove "-link" and closing parenthesis.
   *xlfentry*)
-    ac_f77_v_output=`echo $ac_f77_v_output | sed 's/,/ /g'` ;;
+    ac_f77_v_output=`echo $ac_f77_v_output |
+      sed '
+        s/,/ /g
+        s/ -link / /g
+        s/) *$//
+      '
+    ` ;;
 
   # With Intel ifc, ignore the quoted -mGLOB_options_string stuff (quoted
   # $LIBS confuse us, and the libraries appear later in the output anyway).
@@ -18683,6 +18707,16 @@ case $ac_f77_v_output in
 	s/-ignore  *'[^']*'/ /g; s/-ignore  *\"[^\"]*\"/ /g
 	s/-def  *'[^']*'/ /g; s/-def  *\"[^\"]*\"/ /g"` ;;
 
+  # If we are using fort77 (the f2c wrapper) then filter output and delete quotes.
+  *fort77*f2c*gcc*)
+    ac_f77_v_output=`echo "$ac_f77_v_output" | sed -n '
+        /:[	 ]\+Running[	 ]\{1,\}"gcc"/{
+          /"-c"/d
+          /[.]c"*/d
+          s/^.*"gcc"/"gcc"/
+          s/"//gp
+        }'` ;;
+
   # If we are using Cray Fortran then delete quotes.
   *cft90*)
     ac_f77_v_output=`echo $ac_f77_v_output | sed 's/"//g'` ;;
@@ -18756,9 +18790,16 @@ ac_f77_v_output="`echo $ac_f77_v_output |
 #        that detects unbalanced quotes in FLIBS should be implemented
 #        and (ugh) tested at some point.
 case $ac_f77_v_output in
-  # If we are using xlf then replace all the commas with spaces.
+  # With xlf replace commas with spaces,
+  # and remove "-link" and closing parenthesis.
   *xlfentry*)
-    ac_f77_v_output=`echo $ac_f77_v_output | sed 's/,/ /g'` ;;
+    ac_f77_v_output=`echo $ac_f77_v_output |
+      sed '
+        s/,/ /g
+        s/ -link / /g
+        s/) *$//
+      '
+    ` ;;
 
   # With Intel ifc, ignore the quoted -mGLOB_options_string stuff (quoted
   # $LIBS confuse us, and the libraries appear later in the output anyway).
@@ -18774,6 +18815,16 @@ case $ac_f77_v_output in
 	s/-ignore  *'[^']*'/ /g; s/-ignore  *\"[^\"]*\"/ /g
 	s/-def  *'[^']*'/ /g; s/-def  *\"[^\"]*\"/ /g"` ;;
 
+  # If we are using fort77 (the f2c wrapper) then filter output and delete quotes.
+  *fort77*f2c*gcc*)
+    ac_f77_v_output=`echo "$ac_f77_v_output" | sed -n '
+        /:[	 ]\+Running[	 ]\{1,\}"gcc"/{
+          /"-c"/d
+          /[.]c"*/d
+          s/^.*"gcc"/"gcc"/
+          s/"//gp
+        }'` ;;
+
   # If we are using Cray Fortran then delete quotes.
   *cft90*)
     ac_f77_v_output=`echo $ac_f77_v_output | sed 's/"//g'` ;;
@@ -18832,7 +18883,11 @@ fi
 	  |-LANG:=* | -LIST:* | -LNO:* | -link)
 	  ;;
 	-lkernel32)
-	  test x"$CYGWIN" != xyes && ac_cv_f77_libs="$ac_cv_f77_libs $ac_arg"
+	  case $host_os in
+	  *cygwin*) ;;
+	  *) ac_cv_f77_libs="$ac_cv_f77_libs $ac_arg"
+	    ;;
+	  esac
 	  ;;
 	-[LRuYz])
 	  # These flags, when seen by themselves, take an argument.
@@ -19595,7 +19650,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_acx_pthread_config="yes"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -19770,7 +19825,7 @@ do
   IFS=$as_save_IFS
   test -z "$as_dir" && as_dir=.
     for ac_exec_ext in '' $ac_executable_extensions; do
-  if { test -f "$as_dir/$ac_word$ac_exec_ext" && $as_test_x "$as_dir/$ac_word$ac_exec_ext"; }; then
+  if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then
     ac_cv_prog_PTHREAD_CC="$ac_prog"
     $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5
     break 2
@@ -20414,16 +20469,16 @@ if (echo >conf$$.file) 2>/dev/null; then
     # ... but there are two gotchas:
     # 1) On MSYS, both `ln -s file dir' and `ln file dir' fail.
     # 2) DJGPP < 2.04 has no symlinks; `ln -s' creates a wrapper executable.
-    # In both cases, we have to default to `cp -p'.
+    # In both cases, we have to default to `cp -pR'.
     ln -s conf$$.file conf$$.dir 2>/dev/null && test ! -f conf$$.exe ||
-      as_ln_s='cp -p'
+      as_ln_s='cp -pR'
   elif ln conf$$.file conf$$ 2>/dev/null; then
     as_ln_s=ln
   else
-    as_ln_s='cp -p'
+    as_ln_s='cp -pR'
   fi
 else
-  as_ln_s='cp -p'
+  as_ln_s='cp -pR'
 fi
 rm -f conf$$ conf$$.exe conf$$.dir/conf$$.file conf$$.file
 rmdir conf$$.dir 2>/dev/null
@@ -20483,28 +20538,16 @@ else
   as_mkdir_p=false
 fi
 
-if test -x / >/dev/null 2>&1; then
-  as_test_x='test -x'
-else
-  if ls -dL / >/dev/null 2>&1; then
-    as_ls_L_option=L
-  else
-    as_ls_L_option=
-  fi
-  as_test_x='
-    eval sh -c '\''
-      if test -d "$1"; then
-	test -d "$1/.";
-      else
-	case $1 in #(
-	-*)set "./$1";;
-	esac;
-	case `ls -ld'$as_ls_L_option' "$1" 2>/dev/null` in #((
-	???[sx]*):;;*)false;;esac;fi
-    '\'' sh
-  '
-fi
-as_executable_p=$as_test_x
+
+# as_fn_executable_p FILE
+# -----------------------
+# Test if FILE is an executable regular file.
+as_fn_executable_p ()
+{
+  test -f "$1" && test -x "$1"
+} # as_fn_executable_p
+as_test_x='test -x'
+as_executable_p=as_fn_executable_p
 
 # Sed expression to map a string onto a valid CPP name.
 as_tr_cpp="eval sed 'y%*$as_cr_letters%P$as_cr_LETTERS%;s%[^_$as_cr_alnum]%_%g'"
@@ -20525,8 +20568,8 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by fftw $as_me 3.3.2, which was
-generated by GNU Autoconf 2.68.  Invocation command line was
+This file was extended by fftw $as_me 3.3.3, which was
+generated by GNU Autoconf 2.69.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
   CONFIG_HEADERS  = $CONFIG_HEADERS
@@ -20591,11 +20634,11 @@ _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
 ac_cs_version="\\
-fftw config.status 3.3.2
-configured by $0, generated by GNU Autoconf 2.68,
+fftw config.status 3.3.3
+configured by $0, generated by GNU Autoconf 2.69,
   with options \\"\$ac_cs_config\\"
 
-Copyright (C) 2010 Free Software Foundation, Inc.
+Copyright (C) 2012 Free Software Foundation, Inc.
 This config.status script is free software; the Free Software Foundation
 gives unlimited permission to copy, distribute and modify it."
 
@@ -20686,7 +20729,7 @@ fi
 _ACEOF
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 if \$ac_cs_recheck; then
-  set X '$SHELL' '$0' $ac_configure_args \$ac_configure_extra_args --no-create --no-recursion
+  set X $SHELL '$0' $ac_configure_args \$ac_configure_extra_args --no-create --no-recursion
   shift
   \$as_echo "running CONFIG_SHELL=$SHELL \$*" >&6
   CONFIG_SHELL='$SHELL'
diff --git a/configure.ac b/configure.ac
index 5966e24..3978d9d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,5 +1,5 @@
 dnl Process this file with autoconf to produce a configure script.
-AC_INIT(fftw, 3.3.2, fftw at fftw.org)
+AC_INIT(fftw, 3.3.3, fftw at fftw.org)
 AC_CONFIG_SRCDIR(kernel/ifftw.h)
 # fftw-3.1.x was 4:X:1
 # fftw-3.2.x was 5:X:2
diff --git a/depcomp b/depcomp
index bd0ac08..25a39e6 100755
--- a/depcomp
+++ b/depcomp
@@ -1,10 +1,10 @@
 #! /bin/sh
 # depcomp - compile a program generating dependencies as side-effects
 
-scriptversion=2011-12-04.11; # UTC
+scriptversion=2012-03-27.16; # UTC
 
 # Copyright (C) 1999, 2000, 2003, 2004, 2005, 2006, 2007, 2009, 2010,
-# 2011 Free Software Foundation, Inc.
+# 2011, 2012 Free Software Foundation, Inc.
 
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -28,7 +28,7 @@ scriptversion=2011-12-04.11; # UTC
 
 case $1 in
   '')
-     echo "$0: No command.  Try \`$0 --help' for more information." 1>&2
+     echo "$0: No command.  Try '$0 --help' for more information." 1>&2
      exit 1;
      ;;
   -h | --h*)
@@ -40,8 +40,8 @@ as side-effects.
 
 Environment variables:
   depmode     Dependency tracking mode.
-  source      Source file read by `PROGRAMS ARGS'.
-  object      Object file output by `PROGRAMS ARGS'.
+  source      Source file read by 'PROGRAMS ARGS'.
+  object      Object file output by 'PROGRAMS ARGS'.
   DEPDIR      directory where to store dependencies.
   depfile     Dependency file to output.
   tmpdepfile  Temporary file to use when outputting dependencies.
@@ -57,6 +57,12 @@ EOF
     ;;
 esac
 
+# A tabulation character.
+tab='	'
+# A newline character.
+nl='
+'
+
 if test -z "$depmode" || test -z "$source" || test -z "$object"; then
   echo "depcomp: Variables source, object and depmode must be set" 1>&2
   exit 1
@@ -102,6 +108,12 @@ if test "$depmode" = msvc7msys; then
    depmode=msvc7
 fi
 
+if test "$depmode" = xlc; then
+   # IBM C/C++ Compilers xlc/xlC can output gcc-like dependency informations.
+   gccflag=-qmakedep=gcc,-MF
+   depmode=gcc
+fi
+
 case "$depmode" in
 gcc3)
 ## gcc 3 implements dependency tracking that does exactly what
@@ -156,15 +168,14 @@ gcc)
 ## The second -e expression handles DOS-style file names with drive letters.
   sed -e 's/^[^:]*: / /' \
       -e 's/^['$alpha']:\/[^:]*: / /' < "$tmpdepfile" >> "$depfile"
-## This next piece of magic avoids the `deleted header file' problem.
+## This next piece of magic avoids the "deleted header file" problem.
 ## The problem is that when a header file which appears in a .P file
 ## is deleted, the dependency causes make to die (because there is
 ## typically no way to rebuild the header).  We avoid this by adding
 ## dummy dependencies for each header file.  Too bad gcc doesn't do
 ## this for us directly.
-  tr ' ' '
-' < "$tmpdepfile" |
-## Some versions of gcc put a space before the `:'.  On the theory
+  tr ' ' "$nl" < "$tmpdepfile" |
+## Some versions of gcc put a space before the ':'.  On the theory
 ## that the space means something, we add a space to the output as
 ## well.  hp depmode also adds that space, but also prefixes the VPATH
 ## to the object.  Take care to not repeat it in the output.
@@ -203,18 +214,15 @@ sgi)
     # clever and replace this with sed code, as IRIX sed won't handle
     # lines with more than a fixed number of characters (4096 in
     # IRIX 6.2 sed, 8192 in IRIX 6.5).  We also remove comment lines;
-    # the IRIX cc adds comments like `#:fec' to the end of the
+    # the IRIX cc adds comments like '#:fec' to the end of the
     # dependency line.
-    tr ' ' '
-' < "$tmpdepfile" \
+    tr ' ' "$nl" < "$tmpdepfile" \
     | sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' | \
-    tr '
-' ' ' >> "$depfile"
+    tr "$nl" ' ' >> "$depfile"
     echo >> "$depfile"
 
     # The second pass generates a dummy entry for each header file.
-    tr ' ' '
-' < "$tmpdepfile" \
+    tr ' ' "$nl" < "$tmpdepfile" \
    | sed -e 's/^.*\.o://' -e 's/#.*$//' -e '/^$/ d' -e 's/$/:/' \
    >> "$depfile"
   else
@@ -226,10 +234,17 @@ sgi)
   rm -f "$tmpdepfile"
   ;;
 
+xlc)
+  # This case exists only to let depend.m4 do its work.  It works by
+  # looking at the text of this script.  This case will never be run,
+  # since it is checked for above.
+  exit 1
+  ;;
+
 aix)
   # The C for AIX Compiler uses -M and outputs the dependencies
   # in a .u file.  In older versions, this file always lives in the
-  # current directory.  Also, the AIX compiler puts `$object:' at the
+  # current directory.  Also, the AIX compiler puts '$object:' at the
   # start of each line; $object doesn't have directory information.
   # Version 6 uses the directory in both cases.
   dir=`echo "$object" | sed -e 's|/[^/]*$|/|'`
@@ -259,12 +274,11 @@ aix)
     test -f "$tmpdepfile" && break
   done
   if test -f "$tmpdepfile"; then
-    # Each line is of the form `foo.o: dependent.h'.
+    # Each line is of the form 'foo.o: dependent.h'.
     # Do two passes, one to just change these to
-    # `$object: dependent.h' and one to simply `dependent.h:'.
+    # '$object: dependent.h' and one to simply 'dependent.h:'.
     sed -e "s,^.*\.[a-z]*:,$object:," < "$tmpdepfile" > "$depfile"
-    # That's a tab and a space in the [].
-    sed -e 's,^.*\.[a-z]*:[	 ]*,,' -e 's,$,:,' < "$tmpdepfile" >> "$depfile"
+    sed -e 's,^.*\.[a-z]*:['"$tab"' ]*,,' -e 's,$,:,' < "$tmpdepfile" >> "$depfile"
   else
     # The sourcefile does not contain any dependencies, so just
     # store a dummy comment line, to avoid errors with the Makefile
@@ -275,23 +289,26 @@ aix)
   ;;
 
 icc)
-  # Intel's C compiler understands `-MD -MF file'.  However on
-  #    icc -MD -MF foo.d -c -o sub/foo.o sub/foo.c
+  # Intel's C compiler anf tcc (Tiny C Compiler) understand '-MD -MF file'.
+  # However on
+  #    $CC -MD -MF foo.d -c -o sub/foo.o sub/foo.c
   # ICC 7.0 will fill foo.d with something like
   #    foo.o: sub/foo.c
   #    foo.o: sub/foo.h
-  # which is wrong.  We want:
+  # which is wrong.  We want
   #    sub/foo.o: sub/foo.c
   #    sub/foo.o: sub/foo.h
   #    sub/foo.c:
   #    sub/foo.h:
   # ICC 7.1 will output
   #    foo.o: sub/foo.c sub/foo.h
-  # and will wrap long lines using \ :
+  # and will wrap long lines using '\':
   #    foo.o: sub/foo.c ... \
   #     sub/foo.h ... \
   #     ...
-
+  # tcc 0.9.26 (FIXME still under development at the moment of writing)
+  # will emit a similar output, but also prepend the continuation lines
+  # with horizontal tabulation characters.
   "$@" -MD -MF "$tmpdepfile"
   stat=$?
   if test $stat -eq 0; then :
@@ -300,15 +317,21 @@ icc)
     exit $stat
   fi
   rm -f "$depfile"
-  # Each line is of the form `foo.o: dependent.h',
-  # or `foo.o: dep1.h dep2.h \', or ` dep3.h dep4.h \'.
+  # Each line is of the form 'foo.o: dependent.h',
+  # or 'foo.o: dep1.h dep2.h \', or ' dep3.h dep4.h \'.
   # Do two passes, one to just change these to
-  # `$object: dependent.h' and one to simply `dependent.h:'.
-  sed "s,^[^:]*:,$object :," < "$tmpdepfile" > "$depfile"
-  # Some versions of the HPUX 10.20 sed can't process this invocation
-  # correctly.  Breaking it into two sed invocations is a workaround.
-  sed 's,^[^:]*: \(.*\)$,\1,;s/^\\$//;/^$/d;/:$/d' < "$tmpdepfile" |
-    sed -e 's/$/ :/' >> "$depfile"
+  # '$object: dependent.h' and one to simply 'dependent.h:'.
+  sed -e "s/^[ $tab][ $tab]*/  /" -e "s,^[^:]*:,$object :," \
+    < "$tmpdepfile" > "$depfile"
+  sed '
+    s/[ '"$tab"'][ '"$tab"']*/ /g
+    s/^ *//
+    s/ *\\*$//
+    s/^[^:]*: *//
+    /^$/d
+    /:$/d
+    s/$/ :/
+  ' < "$tmpdepfile" >> "$depfile"
   rm -f "$tmpdepfile"
   ;;
 
@@ -344,7 +367,7 @@ hp2)
   done
   if test -f "$tmpdepfile"; then
     sed -e "s,^.*\.[a-z]*:,$object:," "$tmpdepfile" > "$depfile"
-    # Add `dependent.h:' lines.
+    # Add 'dependent.h:' lines.
     sed -ne '2,${
 	       s/^ *//
 	       s/ \\*$//
@@ -359,9 +382,9 @@ hp2)
 
 tru64)
    # The Tru64 compiler uses -MD to generate dependencies as a side
-   # effect.  `cc -MD -o foo.o ...' puts the dependencies into `foo.o.d'.
+   # effect.  'cc -MD -o foo.o ...' puts the dependencies into 'foo.o.d'.
    # At least on Alpha/Redhat 6.1, Compaq CCC V6.2-504 seems to put
-   # dependencies in `foo.d' instead, so we check for that too.
+   # dependencies in 'foo.d' instead, so we check for that too.
    # Subdirectories are respected.
    dir=`echo "$object" | sed -e 's|/[^/]*$|/|'`
    test "x$dir" = "x$object" && dir=
@@ -407,8 +430,7 @@ tru64)
    done
    if test -f "$tmpdepfile"; then
       sed -e "s,^.*\.[a-z]*:,$object:," < "$tmpdepfile" > "$depfile"
-      # That's a tab and a space in the [].
-      sed -e 's,^.*\.[a-z]*:[	 ]*,,' -e 's,$,:,' < "$tmpdepfile" >> "$depfile"
+      sed -e 's,^.*\.[a-z]*:['"$tab"' ]*,,' -e 's,$,:,' < "$tmpdepfile" >> "$depfile"
    else
       echo "#dummy" > "$depfile"
    fi
@@ -443,11 +465,11 @@ msvc7)
   p
 }' | $cygpath_u | sort -u | sed -n '
 s/ /\\ /g
-s/\(.*\)/	\1 \\/p
+s/\(.*\)/'"$tab"'\1 \\/p
 s/.\(.*\) \\/\1:/
 H
 $ {
-  s/.*/	/
+  s/.*/'"$tab"'/
   G
   p
 }' >> "$depfile"
@@ -478,7 +500,7 @@ dashmstdout)
     shift
   fi
 
-  # Remove `-o $object'.
+  # Remove '-o $object'.
   IFS=" "
   for arg
   do
@@ -498,15 +520,14 @@ dashmstdout)
   done
 
   test -z "$dashmflag" && dashmflag=-M
-  # Require at least two characters before searching for `:'
+  # Require at least two characters before searching for ':'
   # in the target name.  This is to cope with DOS-style filenames:
-  # a dependency such as `c:/foo/bar' could be seen as target `c' otherwise.
+  # a dependency such as 'c:/foo/bar' could be seen as target 'c' otherwise.
   "$@" $dashmflag |
-    sed 's:^[  ]*[^: ][^:][^:]*\:[    ]*:'"$object"'\: :' > "$tmpdepfile"
+    sed 's:^['"$tab"' ]*[^:'"$tab"' ][^:][^:]*\:['"$tab"' ]*:'"$object"'\: :' > "$tmpdepfile"
   rm -f "$depfile"
   cat < "$tmpdepfile" > "$depfile"
-  tr ' ' '
-' < "$tmpdepfile" | \
+  tr ' ' "$nl" < "$tmpdepfile" | \
 ## Some versions of the HPUX 10.20 sed can't process this invocation
 ## correctly.  Breaking it into two sed invocations is a workaround.
     sed -e 's/^\\$//' -e '/^$/d' -e '/:$/d' | sed -e 's/$/ :/' >> "$depfile"
@@ -562,8 +583,7 @@ makedepend)
   # makedepend may prepend the VPATH from the source file name to the object.
   # No need to regex-escape $object, excess matching of '.' is harmless.
   sed "s|^.*\($object *:\)|\1|" "$tmpdepfile" > "$depfile"
-  sed '1,2d' "$tmpdepfile" | tr ' ' '
-' | \
+  sed '1,2d' "$tmpdepfile" | tr ' ' "$nl" | \
 ## Some versions of the HPUX 10.20 sed can't process this invocation
 ## correctly.  Breaking it into two sed invocations is a workaround.
     sed -e 's/^\\$//' -e '/^$/d' -e '/:$/d' | sed -e 's/$/ :/' >> "$depfile"
@@ -583,7 +603,7 @@ cpp)
     shift
   fi
 
-  # Remove `-o $object'.
+  # Remove '-o $object'.
   IFS=" "
   for arg
   do
@@ -652,8 +672,8 @@ msvisualcpp)
   sed -n '/^#line [0-9][0-9]* "\([^"]*\)"/ s::\1:p' | $cygpath_u | sort -u > "$tmpdepfile"
   rm -f "$depfile"
   echo "$object : \\" > "$depfile"
-  sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::	\1 \\:p' >> "$depfile"
-  echo "	" >> "$depfile"
+  sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::'"$tab"'\1 \\:p' >> "$depfile"
+  echo "$tab" >> "$depfile"
   sed < "$tmpdepfile" -n -e 's% %\\ %g' -e '/^\(.*\)$/ s::\1\::p' >> "$depfile"
   rm -f "$tmpdepfile"
   ;;
diff --git a/dft/Makefile.in b/dft/Makefile.in
index f8718c3..6ea4ec4 100644
--- a/dft/Makefile.in
+++ b/dft/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -85,6 +102,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -545,13 +567,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/dft/bluestein.c b/dft/bluestein.c
index dcb4f90..41210f2 100644
--- a/dft/bluestein.c
+++ b/dft/bluestein.c
@@ -145,20 +145,6 @@ static void awake(plan *ego_, enum wakefulness wakefulness)
      }
 }
 
-static int applicable0(const problem *p_)
-{
-     const problem_dft *p = (const problem_dft *) p_;
-     return (1
-	     && p->sz->rnk == 1
-	     && p->vecsz->rnk == 0
-	     /* FIXME: allow other sizes */
-	     && X(is_prime)(p->sz->dims[0].n)
-
-	     /* FIXME: infinite recursion of bluestein with itself */
-	     && p->sz->dims[0].n > 16
-	  );
-}
-
 static int applicable(const solver *ego, const problem *p_, 
 		      const planner *plnr)
 {
diff --git a/dft/scalar/Makefile.in b/dft/scalar/Makefile.in
index 29f00fd..79ad7d4 100644
--- a/dft/scalar/Makefile.in
+++ b/dft/scalar/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -80,6 +97,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -510,13 +532,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/dft/scalar/codelets/Makefile.in b/dft/scalar/codelets/Makefile.in
index 3aa68b9..44d0600 100644
--- a/dft/scalar/codelets/Makefile.in
+++ b/dft/scalar/codelets/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -32,6 +32,23 @@
 # Most of this file requires GNU make .
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -105,6 +122,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libdft_scalar_codelets_la_SOURCES)
 DIST_SOURCES = $(libdft_scalar_codelets_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/dft/scalar/codelets/n1_10.c b/dft/scalar/codelets/n1_10.c
index 4016771..4ec0660 100644
--- a/dft/scalar/codelets/n1_10.c
+++ b/dft/scalar/codelets/n1_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:43 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       E T1g, T1a, T18, T1m, T1k, T1f, T19, T11, T1h, T1l;
 	       {
 		    E Tj, T3, T1b, TN, T1j, TU, T1i, TV, Tq, T10, Ti, Ts, Tw, T15, Tx;
@@ -211,7 +211,7 @@ static void n1_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       E T3, Tj, TQ, T1e, TU, TV, T1c, T1b, Tm, Tp, Tq, Ta, Th, Ti, TA;
 	       E TH, T17, T14, T1f, T1g, T1h, TL, TM, TR;
 	       {
diff --git a/dft/scalar/codelets/n1_11.c b/dft/scalar/codelets/n1_11.c
index 03fd51c..14f12e0 100644
--- a/dft/scalar/codelets/n1_11.c
+++ b/dft/scalar/codelets/n1_11.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:43 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n1_11(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP521108558, +0.521108558113202722944698153526659300680427422);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(44, is), MAKE_VOLATILE_STRIDE(44, os)) {
 	       E T1, TA, T1p, T1y, T19, T1d, T1a, T1e;
 	       {
 		    E T1f, T1u, T4, T1q, Tg, T1t, T7, T1s, Ta, Td, T1r, TP, T1X, T26, Ti;
@@ -292,7 +292,7 @@ static void n1_11(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP755749574, +0.755749574354258283774035843972344420179717445);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(44, is), MAKE_VOLATILE_STRIDE(44, os)) {
 	       E T1, TM, T4, TG, Tk, TR, Tw, TN, T7, TK, Ta, TH, Tn, TQ, Td;
 	       E TJ, Tq, TO, Tt, TP, Tg, TI;
 	       {
diff --git a/dft/scalar/codelets/n1_12.c b/dft/scalar/codelets/n1_12.c
index 70ebbd8..74931fe 100644
--- a/dft/scalar/codelets/n1_12.c
+++ b/dft/scalar/codelets/n1_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:43 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void n1_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(48, is), MAKE_VOLATILE_STRIDE(48, os)) {
 	       E TT, TW, TF, T1q, TY, TQ, TX, T1n;
 	       {
 		    E TA, TS, TR, T5, Ts, Tz, TD, TV, TU, Ta, Tx, TC, T1d, Th, TJ;
@@ -217,7 +217,7 @@ static void n1_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(48, is), MAKE_VOLATILE_STRIDE(48, os)) {
 	       E T5, TR, TA, Ts, TS, Tz, Ta, TU, TD, Tx, TV, TC, Tg, T1a, TG;
 	       E TJ, T1u, T1d, Tl, T1f, TL, TO, T1v, T1i;
 	       {
diff --git a/dft/scalar/codelets/n1_13.c b/dft/scalar/codelets/n1_13.c
index ab5c280..1ce073e 100644
--- a/dft/scalar/codelets/n1_13.c
+++ b/dft/scalar/codelets/n1_13.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:43 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -63,7 +63,7 @@ static void n1_13(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(52, is), MAKE_VOLATILE_STRIDE(52, os)) {
 	       E T2B, T2H, T2I, T2G;
 	       {
 		    E T1, T1P, T2n, T2o, To, TH, T2h, T2k, TE, TB, TF, Tw, T2j, T2c, T1m;
@@ -385,7 +385,7 @@ static void n1_13(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(52, is), MAKE_VOLATILE_STRIDE(52, os)) {
 	       E T1, T1q, Tt, Tu, To, T22, T20, T24, TF, TH, TA, TI, T1X, T25, T2a;
 	       E T2d, T18, T1n, T2k, T2n, T1l, T1r, T1f, T1o, T2h, T2m;
 	       T1 = ri[0];
diff --git a/dft/scalar/codelets/n1_14.c b/dft/scalar/codelets/n1_14.c
index 8c0a87b..c6c5cf1 100644
--- a/dft/scalar/codelets/n1_14.c
+++ b/dft/scalar/codelets/n1_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:43 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP356895867, +0.356895867892209443894399510021300583399127187);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(56, is), MAKE_VOLATILE_STRIDE(56, os)) {
 	       E Tp, T1L, T24, T1W, T1X, T28, T2a, T1Y, T29, T2b;
 	       {
 		    E T3, T1x, T1b, To, T1i, T1M, Ts, Ta, T1k, Tv, Th, T1j, T1K, Ty, TZ;
@@ -304,7 +304,7 @@ static void n1_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP974927912, +0.974927912181823607018131682993931217232785801);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(56, is), MAKE_VOLATILE_STRIDE(56, os)) {
 	       E T3, Tp, T16, T1f, Ta, T1q, Ts, T10, TG, T1z, T19, T1i, Th, T1s, Tv;
 	       E T12, TU, T1B, T17, T1o, To, T1r, Ty, T11, TN, T1A, T18, T1l;
 	       {
diff --git a/dft/scalar/codelets/n1_15.c b/dft/scalar/codelets/n1_15.c
index 61eec12..d30beab 100644
--- a/dft/scalar/codelets/n1_15.c
+++ b/dft/scalar/codelets/n1_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:43 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(60, is), MAKE_VOLATILE_STRIDE(60, os)) {
 	       E T1r, T1g, T14, T13;
 	       {
 		    E T5, T2l, Tx, TV, T1z, T1X, T2s, Tr, T24, TT, T2e, T2n, T1Z, T1Q, T1B;
@@ -333,7 +333,7 @@ static void n1_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(60, is), MAKE_VOLATILE_STRIDE(60, os)) {
 	       E T5, T2l, Tx, TV, T1C, T20, Tl, Tq, Tr, TN, TS, TT, T2c, T2d, T2n;
 	       E T1O, T1P, T22, T1l, T1q, T1w, TZ, T10, T11, Ta, Tf, Tg, TC, TH, TI;
 	       E T2f, T2g, T2m, T1R, T1S, T21, T1a, T1f, T1v, TW, TX, TY;
diff --git a/dft/scalar/codelets/n1_16.c b/dft/scalar/codelets/n1_16.c
index 474bf45..6fea631 100644
--- a/dft/scalar/codelets/n1_16.c
+++ b/dft/scalar/codelets/n1_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:44 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void n1_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       E T1z, T1L, T1M, T1N, T1P, T1J, T1K, T1G, T1O, T1Q;
 	       {
 		    E T1l, T1H, T1R, T7, T1x, TN, TC, T25, T1E, T1b, T1Z, Tt, T2h, T22, T1D;
@@ -303,7 +303,7 @@ static void n1_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       E T7, T1R, T25, TC, TN, T1x, T1H, T1l, Tt, T22, T2h, T1b, T1g, T1E, T1Z;
 	       E T1D, Te, T1S, T26, TJ, TQ, T1m, T1n, TT, Tm, T1X, T2g, T10, T15, T1B;
 	       E T1U, T1A;
diff --git a/dft/scalar/codelets/n1_2.c b/dft/scalar/codelets/n1_2.c
index 21ea89c..9285c48 100644
--- a/dft/scalar/codelets/n1_2.c
+++ b/dft/scalar/codelets/n1_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void n1_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       E T1, T2, T3, T4;
 	       T1 = ri[0];
 	       T2 = ri[WS(is, 1)];
@@ -73,7 +73,7 @@ static void n1_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       E T1, T2, T3, T4;
 	       T1 = ri[0];
 	       T2 = ri[WS(is, 1)];
diff --git a/dft/scalar/codelets/n1_20.c b/dft/scalar/codelets/n1_20.c
index e05b6bd..a46037b 100644
--- a/dft/scalar/codelets/n1_20.c
+++ b/dft/scalar/codelets/n1_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:46 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(80, is), MAKE_VOLATILE_STRIDE(80, os)) {
 	       E T1Y, T1Z, T1W, T1V;
 	       {
 		    E T1d, TP, TD, T7, T3b, T2N, T2f, T1R, T2U, TB, T2P, T2A, T3d, T37, T3j;
@@ -418,7 +418,7 @@ static void n1_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(80, is), MAKE_VOLATILE_STRIDE(80, os)) {
 	       E T7, T2Q, T3h, TD, TP, T1U, T2l, T1d, Tt, TA, TB, T2w, T2z, T2S, T35;
 	       E T36, T3f, TH, TI, TJ, T15, T1a, T1b, T1s, T1x, T1W, T29, T2a, T2j, T1h;
 	       E T1i, T1j, Te, Tl, Tm, T2D, T2G, T2R, T32, T33, T3e, TE, TF, TG, TU;
diff --git a/dft/scalar/codelets/n1_25.c b/dft/scalar/codelets/n1_25.c
index ad6aeec..671c70e 100644
--- a/dft/scalar/codelets/n1_25.c
+++ b/dft/scalar/codelets/n1_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:46 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -85,7 +85,7 @@ static void n1_25(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(100, is), MAKE_VOLATILE_STRIDE(100, os)) {
 	       E T3Y, T3U, T3W, T42, T44, T3X, T3R, T3V, T3Z, T43;
 	       {
 		    E T4Q, T1U, T9, T3b, T45, T3e, T46, T1D, T4P, T1R, Ts, T1K, T18, T1E, T4z;
@@ -653,7 +653,7 @@ static void n1_25(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(100, is), MAKE_VOLATILE_STRIDE(100, os)) {
 	       E T9, T4u, T2T, TP, T3H, TW, T5y, T3I, T2Q, T4v, Ti, Tr, Ts, T5m, T5n;
 	       E T5v, T18, T4G, T34, T3M, T1G, T4J, T38, T3T, T1v, T4K, T37, T3W, T1j, T4H;
 	       E T35, T3P, TB, TK, TL, T5p, T5q, T5w, T1T, T4N, T3c, T41, T2r, T4Q, T3e;
diff --git a/dft/scalar/codelets/n1_3.c b/dft/scalar/codelets/n1_3.c
index 0f1aa4d..94c4b02 100644
--- a/dft/scalar/codelets/n1_3.c
+++ b/dft/scalar/codelets/n1_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void n1_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       E T1, T9, T2, T3, T6, T7;
 	       T1 = ri[0];
 	       T9 = ii[0];
@@ -90,7 +90,7 @@ static void n1_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       E T1, Ta, T4, T9, T8, Tb, T5, Tc;
 	       T1 = ri[0];
 	       Ta = ii[0];
diff --git a/dft/scalar/codelets/n1_32.c b/dft/scalar/codelets/n1_32.c
index fe77909..d16fcb0 100644
--- a/dft/scalar/codelets/n1_32.c
+++ b/dft/scalar/codelets/n1_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:45 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void n1_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       E T3g, T3f, T3n, T3b, T3r, T3l, T3o, T3e, T3h, T3p;
 	       {
 		    E T2T, T3T, T4r, T7, T3t, T1z, T18, T4Z, Te, T50, T4s, T1f, T2W, T3u, T3U;
@@ -663,7 +663,7 @@ static void n1_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       E T7, T4r, T4Z, T18, T1z, T3t, T3T, T2T, Te, T1f, T50, T4s, T2W, T3u, T1G;
 	       E T3U, Tm, T1n, T1O, T2Z, T3y, T3X, T4w, T53, Tt, T1u, T1V, T2Y, T3B, T3W;
 	       E T4z, T52, T2t, T3L, T3O, T2K, TR, TY, T5F, T5G, T5H, T5I, T4R, T5j, T2E;
diff --git a/dft/scalar/codelets/n1_4.c b/dft/scalar/codelets/n1_4.c
index 2f4e4d5..fc54a23 100644
--- a/dft/scalar/codelets/n1_4.c
+++ b/dft/scalar/codelets/n1_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void n1_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       E T4, Tb, T3, Tf, T9, T5, Tc, Td;
 	       {
 		    E T1, T2, T7, T8;
@@ -95,7 +95,7 @@ static void n1_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       E T3, Tb, T9, Tf, T6, Ta, Te, Tg;
 	       {
 		    E T1, T2, T7, T8;
diff --git a/dft/scalar/codelets/n1_5.c b/dft/scalar/codelets/n1_5.c
index 3b0936f..9389b67 100644
--- a/dft/scalar/codelets/n1_5.c
+++ b/dft/scalar/codelets/n1_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       E Tq, Ti, Tk, Tu, Tw, Tp, Tb, Tj, Tr, Tv;
 	       {
 		    E T1, Tl, Ts, Tt, T8, Ta, Te, Tm, Tn, Th, To, T9;
@@ -124,7 +124,7 @@ static void n1_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       E T1, To, T8, Tt, T9, Ts, Te, Tp, Th, Tn;
 	       T1 = ri[0];
 	       To = ii[0];
diff --git a/dft/scalar/codelets/n1_6.c b/dft/scalar/codelets/n1_6.c
index dbdc1ca..747eda1 100644
--- a/dft/scalar/codelets/n1_6.c
+++ b/dft/scalar/codelets/n1_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void n1_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       E TA, Tz;
 	       {
 		    E Tb, T3, Tx, Tp, Tj, Te, Ts, Ta, Tu, Ti, Tk;
@@ -135,7 +135,7 @@ static void n1_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       E T3, Tb, Tq, Tx, T6, Tc, T9, Td, Ta, Te, Ti, Tu, Tl, Tv, Tr;
 	       E Ty;
 	       {
diff --git a/dft/scalar/codelets/n1_64.c b/dft/scalar/codelets/n1_64.c
index 74761a6..c273982 100644
--- a/dft/scalar/codelets/n1_64.c
+++ b/dft/scalar/codelets/n1_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:46 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -53,7 +53,7 @@ static void n1_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       E T9b, T9e;
 	       {
 		    E T7B, T37, T5Z, T8F, Td9, Tf, TcB, TbB, T7C, T62, TdH, T2i, Tcb, Tah, T8G;
@@ -1477,7 +1477,7 @@ static void n1_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       E T37, T7B, T8F, T5Z, Tf, Td9, TbB, TcB, T62, T7C, T2i, TdH, Tah, Tcb, T3e;
 	       E T8G, Tu, TdI, Tak, TbD, Tan, TbC, T2x, Tda, T3m, T65, T7G, T8J, T7J, T8I;
 	       E T3t, T64, TK, Tdd, Tas, Tce, Tav, Tcf, T2N, Tdc, T3G, T6G, T7O, T9k, T7R;
diff --git a/dft/scalar/codelets/n1_7.c b/dft/scalar/codelets/n1_7.c
index 8a106c0..a4e6a97 100644
--- a/dft/scalar/codelets/n1_7.c
+++ b/dft/scalar/codelets/n1_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP554958132, +0.554958132087371191422194871006410481067288862);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       E Tz, TP, Ty, TK, TN, TE, Tw, TF;
 	       {
 		    E T1, TI, T4, TG, Ta, TT, Tp, TH, T7, Tk, TJ, TO, Tu, Tb, TB;
@@ -166,7 +166,7 @@ static void n1_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP974927912, +0.974927912181823607018131682993931217232785801);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       E T1, Tu, T4, Tq, Te, Tx, T7, Ts, Tk, Tv, Ta, Tr, Th, Tw;
 	       T1 = ri[0];
 	       Tu = ii[0];
diff --git a/dft/scalar/codelets/n1_8.c b/dft/scalar/codelets/n1_8.c
index c18532c..708f60c 100644
--- a/dft/scalar/codelets/n1_8.c
+++ b/dft/scalar/codelets/n1_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void n1_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       E TF, TE, TD, TI;
 	       {
 		    E Tn, T3, TC, Ti, TB, T6, To, Tl, Td, TN, Tz, TH, Ta, Tq, Tt;
@@ -160,7 +160,7 @@ static void n1_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       E T3, Tn, Ti, TC, T6, TB, Tl, To, Td, TN, Tz, TH, Ta, TM, Tu;
 	       E TG;
 	       {
diff --git a/dft/scalar/codelets/n1_9.c b/dft/scalar/codelets/n1_9.c
index 8c85c14..a4b3cff 100644
--- a/dft/scalar/codelets/n1_9.c
+++ b/dft/scalar/codelets/n1_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n1_9(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(36, is), MAKE_VOLATILE_STRIDE(36, os)) {
 	       E T17, TV, T14, TY, T11, T15;
 	       {
 		    E Tm, TM, TL, T5, Tl, T1f, Tb, Tt, Ta, T1c, TI, TX, TF, TW, Tc;
@@ -215,7 +215,7 @@ static void n1_9(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, I
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, ri = ri + ivs, ii = ii + ivs, ro = ro + ovs, io = io + ovs, MAKE_VOLATILE_STRIDE(36, is), MAKE_VOLATILE_STRIDE(36, os)) {
 	       E T5, TO, Th, Tk, T1g, TR, Ta, T1c, Tq, TW, Tv, TX, Tf, T1d, TB;
 	       E T10, TG, TZ;
 	       {
diff --git a/dft/scalar/codelets/q1_2.c b/dft/scalar/codelets/q1_2.c
index fa61f80..f772706 100644
--- a/dft/scalar/codelets/q1_2.c
+++ b/dft/scalar/codelets/q1_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:17 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void q1_2(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T9, T6, T5;
 	       {
 		    E T1, T2, T7, T8, Tb, T4, Tc, Th, Ti, Te, Tj, Td, Tg;
@@ -103,7 +103,7 @@ static void q1_2(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T1, T2, T4, T6, T7, T8, T9, Ta, Tc, Te, Tf, Tg;
 	       T1 = rio[0];
 	       T2 = rio[WS(rs, 1)];
diff --git a/dft/scalar/codelets/q1_3.c b/dft/scalar/codelets/q1_3.c
index 72f61a1..371c517 100644
--- a/dft/scalar/codelets/q1_3.c
+++ b/dft/scalar/codelets/q1_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:41 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:23 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void q1_3(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E Tk, Tn, Tm, To, Tl;
 	       {
 		    E T1, Td, T4, Tg, Tp, T9, Te, T6, Tf, TB, TE, Ts, TZ, Tu, Tx;
@@ -191,7 +191,7 @@ static void q1_3(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T1, T4, T6, Tc, Td, Te, T9, Tf, Tl, To, Tq, Tw, Tx, Ty, Tt;
 	       E Tz, TR, TS, TN, TT, TF, TI, TK, TQ;
 	       {
diff --git a/dft/scalar/codelets/q1_4.c b/dft/scalar/codelets/q1_4.c
index eca2cfd..a57821d 100644
--- a/dft/scalar/codelets/q1_4.c
+++ b/dft/scalar/codelets/q1_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:17 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void q1_4(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T1X, T1S, T1L, T1Y, T1R;
 	       {
 		    E T3, Tf, Tv, Ti, Tw, Tx, T6, Tm, Tc, Ts, T1T, T1H, T29, T1W, T2a;
@@ -293,7 +293,7 @@ static void q1_4(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T3, Te, Tb, Tq, T6, T8, Th, Tr, Tv, TG, TD, TS, Ty, TA, TJ;
 	       E TT, TX, T18, T15, T1k, T10, T12, T1b, T1l, T1p, T1A, T1x, T1M, T1s, T1u;
 	       E T1D, T1N;
diff --git a/dft/scalar/codelets/q1_5.c b/dft/scalar/codelets/q1_5.c
index 67d09a9..2f4e51d 100644
--- a/dft/scalar/codelets/q1_5.c
+++ b/dft/scalar/codelets/q1_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:41 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:23 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void q1_5(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T1x, T1w, T1v;
 	       {
 		    E T1, Tn, TM, Tw, Tb, T8, Ta, TV, Tq, Ts, TH, Tj, Tr, T1h, T1q;
@@ -551,7 +551,7 @@ static void q1_5(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T1, Ta, TG, Tv, T8, Tb, Tp, Tj, TD, To, Tq, Tr, TN, TW, T1s;
 	       E T1h, TU, TX, T1b, T15, T1p, T1a, T1c, T1d, T1z, T1I, T2e, T23, T1G, T1J;
 	       E T1X, T1R, T2b, T1W, T1Y, T1Z, T3v, T3p, T3J, T3u, T3w, T3x, T37, T3g, T3M;
diff --git a/dft/scalar/codelets/q1_6.c b/dft/scalar/codelets/q1_6.c
index 184cdd6..3288706 100644
--- a/dft/scalar/codelets/q1_6.c
+++ b/dft/scalar/codelets/q1_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:24 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void q1_6(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T4c, T4f, T4e, T4g, T4d;
 	       {
 		    E T3, Tw, Ta, TW, Tg, TG, TM, TT, TU, TP, Tn, T17, TV, TJ, Tv;
@@ -702,7 +702,7 @@ static void q1_6(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T3, Tc, Tt, TM, TX, T16, T1n, T1G, T2h, T2A, T1R, T20, T2L, T2U, T3b;
 	       E T3u, T3F, T3O, T45, T4o, T4Z, T5i, T4z, T4I, Ta, TP, Tf, Tq, Tn, TN;
 	       E Tu, TJ, T14, T1J, T19, T1k, T1h, T1H, T1o, T1D, T2b, T2B, T2i, T2x, T1Y;
diff --git a/dft/scalar/codelets/q1_8.c b/dft/scalar/codelets/q1_8.c
index 653d74c..bb695e5 100644
--- a/dft/scalar/codelets/q1_8.c
+++ b/dft/scalar/codelets/q1_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:17 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void q1_8(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T9C, T9N, T9l, T9E, T9D, T9O;
 	       {
 		    E TV, Tk, T1d, T7, T18, T1t, TQ, TD, T5t, T4S, T5L, T4F, T5G, T61, T5o;
@@ -1289,7 +1289,7 @@ static void q1_8(R *rio, R *iio, const R *W, stride rs, stride vs, INT mb, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, rio = rio + ms, iio = iio + ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(0, vs)) {
 	       E T7, T14, T1g, Tk, TC, TQ, T10, TM, T1w, T2p, T2z, T1H, T1M, T1W, T2j;
 	       E T1V, T7R, T8O, T90, T84, T8m, T8A, T8K, T8w, T9g, Ta9, Taj, T9r, T9w, T9G;
 	       E Ta3, T9F, Te, T17, T1h, Tp, Tu, TE, T11, TD, T1p, T2m, T2y, T1C, T1U;
diff --git a/dft/scalar/codelets/t1_10.c b/dft/scalar/codelets/t1_10.c
index d173ea3..597fdf4 100644
--- a/dft/scalar/codelets/t1_10.c
+++ b/dft/scalar/codelets/t1_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 18); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 18); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 18, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T1X, T21, T20, T22;
 	       {
 		    E T23, T1U, T8, T12, T1y, T25, T1P, T1H, T1Y, T18, T10, T2b, T1K, T1O, T15;
@@ -293,7 +293,7 @@ static void t1_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 18); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 18); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 18, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T7, T1O, TT, T1C, TF, TQ, TR, T1o, T1p, T1y, TX, TY, TZ, T1d, T1g;
 	       E T1M, Ti, Tt, Tu, T1r, T1s, T1x, TU, TV, TW, T16, T19, T1L;
 	       {
diff --git a/dft/scalar/codelets/t1_12.c b/dft/scalar/codelets/t1_12.c
index bc3d60e..f285787 100644
--- a/dft/scalar/codelets/t1_12.c
+++ b/dft/scalar/codelets/t1_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:50 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 22); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 22); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 22, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T2B, T2C;
 	       {
 		    E T1, T2i, T2e, Tl, T1Y, T10, T1S, TG, T2f, T1s, T2r, Ty, T1Z, T1H, T21;
@@ -309,7 +309,7 @@ static void t1_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 22); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 22); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 22, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T1, T1W, T18, T21, Tc, T15, T1V, T22, TR, T1E, T1o, T1D, T12, T1l, T1F;
 	       E T1G, Ti, T1S, T1d, T24, Tt, T1a, T1T, T25, TA, T1z, T1j, T1y, TL, T1g;
 	       E T1A, T1B;
diff --git a/dft/scalar/codelets/t1_15.c b/dft/scalar/codelets/t1_15.c
index 5013068..fd68333 100644
--- a/dft/scalar/codelets/t1_15.c
+++ b/dft/scalar/codelets/t1_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:50 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1_15(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 28); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 28, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 28); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 28, MAKE_VOLATILE_STRIDE(30, rs)) {
 	       E T2d, T2O, T2Q, T2m, T2k, T2l, T2P, T2n;
 	       {
 		    E T1G, T3u, T3k, T3t, T1B, Tf, T37, T1y, T2V, T2M, T2a, T2i, T39, Tz, T2X;
@@ -443,7 +443,7 @@ static void t1_15(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 28); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 28, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 28); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 28, MAKE_VOLATILE_STRIDE(30, rs)) {
 	       E T1q, T34, Td, T1n, T2S, T35, T13, T1k, T1l, T2E, T2F, T2O, T1H, T1T, T2k;
 	       E T2t, T2f, T2s, T1M, T1U, Tu, TL, TM, T2H, T2I, T2N, T1w, T1Q, T29, T2w;
 	       E T24, T2v, T1B, T1R;
diff --git a/dft/scalar/codelets/t1_16.c b/dft/scalar/codelets/t1_16.c
index a83dd03..2b3d3ad 100644
--- a/dft/scalar/codelets/t1_16.c
+++ b/dft/scalar/codelets/t1_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:13 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:51 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t1_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 30); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 30); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 30, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T3G, T3F;
 	       {
 		    E T3z, T3o, T8, T1I, T2o, T35, T2r, T1s, T2w, T36, T2p, T1F, T3k, T1N, T3A;
@@ -423,7 +423,7 @@ static void t1_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 30); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 30); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 30, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T37, T1t, T2U, Ti, T38, T1w, T2R, Tu, T2s, T1C, T2c, TF, T2t, T1H;
 	       E T2d, T1f, T1q, T2B, T2C, T2D, T2E, T1Z, T2j, T24, T2k, TS, T13, T2w, T2x;
 	       E T2y, T2z, T1O, T2g, T1T, T2h;
diff --git a/dft/scalar/codelets/t1_2.c b/dft/scalar/codelets/t1_2.c
index a588f10..8f8ceef 100644
--- a/dft/scalar/codelets/t1_2.c
+++ b/dft/scalar/codelets/t1_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:47 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void t1_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
 	       E T1, Ta, T3, T6, T2, T5;
 	       T1 = ri[0];
 	       Ta = ii[0];
@@ -86,7 +86,7 @@ static void t1_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 2); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
 	       E T1, T8, T6, T7;
 	       T1 = ri[0];
 	       T8 = ii[0];
diff --git a/dft/scalar/codelets/t1_20.c b/dft/scalar/codelets/t1_20.c
index a353382..bda5470 100644
--- a/dft/scalar/codelets/t1_20.c
+++ b/dft/scalar/codelets/t1_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:14 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:53 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 38); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 38); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 38, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T4P, T4Y, T50, T4U, T4S, T4T, T4Z, T4V;
 	       {
 		    E T4N, T4r, T8, T2i, T4n, T2n, T4O, Tl, T2v, T3v, T40, T4b, TN, T2b, T3F;
@@ -561,7 +561,7 @@ static void t1_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 38); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 38); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 38, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E Tj, T1R, T4g, T4p, T2q, T37, T3Q, T42, T1r, T1O, T1P, T3i, T3l, T44, T3D;
 	       E T3E, T3K, T1V, T1W, T1X, T23, T28, T4r, T2W, T2X, T4c, T33, T34, T35, T2G;
 	       E T2L, T2M, TG, T13, T14, T3p, T3s, T43, T3A, T3B, T3J, T1S, T1T, T1U, T2e;
diff --git a/dft/scalar/codelets/t1_25.c b/dft/scalar/codelets/t1_25.c
index db3379e..6b4f1a6 100644
--- a/dft/scalar/codelets/t1_25.c
+++ b/dft/scalar/codelets/t1_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:54 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -85,7 +85,7 @@ static void t1_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 48); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 48, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 48); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 48, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T7I, T6Q, T6O, T7O, T7M, T7H, T6P, T6H, T7J, T7N;
 	       {
 		    E T78, T5G, T3Y, T3M, T7C, T7c, T77, T6Y, Tt, T3L, T5T, T4P, T5Q, T4W, T3G;
@@ -847,7 +847,7 @@ static void t1_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 48); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 48, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 48); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 48, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T1, T6b, T2l, T6o, To, T2m, T6a, T6p, T6t, T6S, T2u, T4I, T2i, T60, T3O;
 	       E T5D, T4r, T58, T3Z, T5C, T4q, T5b, TS, T5W, T2G, T5s, T4g, T4M, T2R, T5t;
 	       E T4h, T4P, T1l, T5X, T33, T5w, T4j, T4W, T3e, T5v, T4k, T4T, T1P, T5Z, T3r;
diff --git a/dft/scalar/codelets/t1_3.c b/dft/scalar/codelets/t1_3.c
index 0fbe19c..db65388 100644
--- a/dft/scalar/codelets/t1_3.c
+++ b/dft/scalar/codelets/t1_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:47 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs)) {
 	       E T1, Tm, T9, Tc, Tb, Th, T7, Ti, Ta, Tj, Td;
 	       T1 = ri[0];
 	       Tm = ii[0];
@@ -109,7 +109,7 @@ static void t1_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs)) {
 	       E T1, Ti, T6, Te, Tb, Tf, Tc, Th;
 	       T1 = ri[0];
 	       Ti = ii[0];
diff --git a/dft/scalar/codelets/t1_32.c b/dft/scalar/codelets/t1_32.c
index c9dd964..0f7465e 100644
--- a/dft/scalar/codelets/t1_32.c
+++ b/dft/scalar/codelets/t1_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:13 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:51 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void t1_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 62); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 62); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 62, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T90, T8Z;
 	       {
 		    E T8x, T87, T8, T3w, T83, T3B, T8y, Tl, T6F, Tz, T3J, T5T, T6G, TM, T3Q;
@@ -927,7 +927,7 @@ static void t1_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 62); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 62); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 62, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E Tj, T5F, T7C, T7Q, T35, T4T, T78, T7m, T1Q, T61, T5Y, T6J, T3K, T59, T41;
 	       E T56, T2B, T67, T6e, T6O, T4b, T5d, T4s, T5g, TG, T7l, T5I, T73, T3a, T4U;
 	       E T3f, T4V, T14, T5N, T5M, T6E, T3m, T4Y, T3r, T4Z, T1r, T5P, T5S, T6F, T3x;
diff --git a/dft/scalar/codelets/t1_4.c b/dft/scalar/codelets/t1_4.c
index 378d85c..932dbe9 100644
--- a/dft/scalar/codelets/t1_4.c
+++ b/dft/scalar/codelets/t1_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:47 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void t1_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E To, Te, Tm, T8, Tw, Tx, Tq, Tk;
 	       {
 		    E T1, Tv, Tu, T7, Tg, Tj, Tf, Ti, Tp, Th;
@@ -123,7 +123,7 @@ static void t1_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T1, Tp, T6, To, Tc, Tk, Th, Tl;
 	       T1 = ri[0];
 	       Tp = ii[0];
diff --git a/dft/scalar/codelets/t1_5.c b/dft/scalar/codelets/t1_5.c
index 17b894b..10451a7 100644
--- a/dft/scalar/codelets/t1_5.c
+++ b/dft/scalar/codelets/t1_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:47 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T1, TM, TJ, TA, TQ, Te, TC, Tk, TE, Tq;
 	       {
 		    E Tg, Tj, Tm, TB, Th, Tp, Tl, Ti, To, TD, Tn;
@@ -161,7 +161,7 @@ static void t1_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T1, TE, Tu, Tx, TJ, TI, TB, TC, TD, Tc, Tn, To;
 	       T1 = ri[0];
 	       TE = ii[0];
diff --git a/dft/scalar/codelets/t1_6.c b/dft/scalar/codelets/t1_6.c
index f7c78e6..dcb48ed 100644
--- a/dft/scalar/codelets/t1_6.c
+++ b/dft/scalar/codelets/t1_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:48 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs)) {
 	       E TY, TU, T10, TZ;
 	       {
 		    E T1, TX, TW, T7, Tn, Tq, TJ, TR, TB, Tl, To, TK, Tt, Tw, Ts;
@@ -173,7 +173,7 @@ static void t1_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs)) {
 	       E T7, TS, Tv, TO, Tt, TJ, Tx, TF, Ti, TI, Tw, TC;
 	       {
 		    E T1, TN, T6, TM;
diff --git a/dft/scalar/codelets/t1_64.c b/dft/scalar/codelets/t1_64.c
index 89aa609..cec1250 100644
--- a/dft/scalar/codelets/t1_64.c
+++ b/dft/scalar/codelets/t1_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:14 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -53,7 +53,7 @@ static void t1_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 126); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 126, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 126); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 126, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E TeI, Tkk, Tkj, TeL;
 	       {
 		    E TiV, Tjm, T7e, TcA, TjR, Tkl, Tm, TeM, TeZ, Ths, T7Q, TcJ, T1G, TeW, TcI;
@@ -2030,7 +2030,7 @@ static void t1_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 126); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 126, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 126); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 126, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tj, TcL, ThT, Tin, T6b, Taz, TgT, Thn, TG, Thm, TcO, TgO, T6m, ThQ, TaC;
 	       E Tim, T14, Tfq, T6y, T9O, TaG, Tc0, TcU, TeE, T1r, Tfr, T6J, T9P, TaJ, Tc1;
 	       E TcZ, TeF, T1Q, T2d, Tfx, Tfu, Tfv, Tfw, T6Q, TaM, Tdb, TeJ, T71, TaQ, T7a;
diff --git a/dft/scalar/codelets/t1_7.c b/dft/scalar/codelets/t1_7.c
index 3848292..4049786 100644
--- a/dft/scalar/codelets/t1_7.c
+++ b/dft/scalar/codelets/t1_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:10 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:48 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP356895867, +0.356895867892209443894399510021300583399127187);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 12); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 12, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 12); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 12, MAKE_VOLATILE_STRIDE(14, rs)) {
 	       E T1c, T19, T1i, T18, T16, T1q, T1t, T1r, T1u, T1s;
 	       {
 		    E T1, TR, T1h, Te, Tt, Tw, T1a, TM, T1g, Tr, Tu, TS, Tz, TC, Ty;
@@ -224,7 +224,7 @@ static void t1_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP974927912, +0.974927912181823607018131682993931217232785801);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 12); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 12, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 12); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 12, MAKE_VOLATILE_STRIDE(14, rs)) {
 	       E T1, TR, Tc, TS, TC, TO, Tn, TT, TI, TP, Ty, TU, TF, TQ;
 	       T1 = ri[0];
 	       TR = ii[0];
diff --git a/dft/scalar/codelets/t1_8.c b/dft/scalar/codelets/t1_8.c
index d4fd20e..5be7e1e 100644
--- a/dft/scalar/codelets/t1_8.c
+++ b/dft/scalar/codelets/t1_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:10 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void t1_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T1g, T1f, T1e, Tm, T1q, T1o, T1p, TN, T1h, T1i;
 	       {
 		    E T1, T1m, T1l, T7, TS, Tk, TQ, Te, To, Tr, T17, TM, T12, Tu, TW;
@@ -209,7 +209,7 @@ static void t1_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 14); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T7, T1e, TH, T19, TF, T13, TR, TU, Ti, T1f, TK, T16, Tu, T12, TM;
 	       E TP;
 	       {
diff --git a/dft/scalar/codelets/t1_9.c b/dft/scalar/codelets/t1_9.c
index 43c0ee9..a42d283 100644
--- a/dft/scalar/codelets/t1_9.c
+++ b/dft/scalar/codelets/t1_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void t1_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 16); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 16, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 16); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 16, MAKE_VOLATILE_STRIDE(18, rs)) {
 	       E T1K, T24, T1H, T23;
 	       {
 		    E T1, T1R, T1Q, T10, T1W, Te, TB, T1l, T1r, T1q, T1M, TE, T1g, Tz, T12;
@@ -274,7 +274,7 @@ static void t1_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 16); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 16, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 16); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 16, MAKE_VOLATILE_STRIDE(18, rs)) {
 	       E T1, T1B, TQ, T1G, Tc, TN, T1A, T1H, TL, T1x, T17, T1o, T1c, T1n, Tu;
 	       E T1w, TW, T1k, T11, T1l;
 	       {
diff --git a/dft/scalar/codelets/t2_10.c b/dft/scalar/codelets/t2_10.c
index b5b621d..3092580 100644
--- a/dft/scalar/codelets/t2_10.c
+++ b/dft/scalar/codelets/t2_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t2_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T27, T2b, T2a, T2c;
 	       {
 		    E T2, T3, T8, Tc, T5, T4, TX, T11, TE, T6, TB, TA;
@@ -303,7 +303,7 @@ static void t2_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T2, T5, T3, T6, T8, Tm, Tc, Tk, T9, Td, Te, TM, TO, Tg, Tp;
 	       E Tv, Tx, Tr;
 	       {
diff --git a/dft/scalar/codelets/t2_16.c b/dft/scalar/codelets/t2_16.c
index 3b0ec5a..4a95b29 100644
--- a/dft/scalar/codelets/t2_16.c
+++ b/dft/scalar/codelets/t2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:00 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t2_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T3S, T3R;
 	       {
 		    E T2, Tf, TM, TO, T3, Tg, TN, TS, T4, Tp, T6, T5, Th;
@@ -450,7 +450,7 @@ static void t2_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T2, T5, Tg, Ti, Tk, To, TE, TC, T6, T3, T8, TW, TJ, Tt, TU;
 	       E Tc, Tx, TH, TN, TO, TP, TR, T1f, T1k, T1b, T1i, T1y, T1H, T1u, T1F;
 	       {
diff --git a/dft/scalar/codelets/t2_20.c b/dft/scalar/codelets/t2_20.c
index 8b1b6be..272f538 100644
--- a/dft/scalar/codelets/t2_20.c
+++ b/dft/scalar/codelets/t2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:29 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t2_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T59, T5i, T5k, T5e, T5c, T5d, T5j, T5f;
 	       {
 		    E T2, Th, Tf, T6, T5, Tl, T1p, T1n, Ti, T3, Tt, Tv, T24, T1f, T1D;
@@ -586,7 +586,7 @@ static void t2_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T2, T5, Tg, Ti, Tk, To, T1h, T1f, T6, T3, T8, T14, T1Q, Tc, T1O;
 	       E T1v, T18, T1t, T1n, T24, T1j, T22, Tq, Tu, T1E, T1G, Tx, Ty, Tz, TJ;
 	       E T1Z, TB, T1X, T1A, TZ, TL, T1y, TX;
diff --git a/dft/scalar/codelets/t2_25.c b/dft/scalar/codelets/t2_25.c
index 3253802..47dc7a5 100644
--- a/dft/scalar/codelets/t2_25.c
+++ b/dft/scalar/codelets/t2_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:31 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:11 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -85,7 +85,7 @@ static void t2_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T8c, T7k, T7i, T8i, T8g, T8b, T7j, T7b, T8d, T8h;
 	       {
 		    E T2, T8, T3, T6, Tk, Tv, TS, T4, Ta, TD, T2L, T10, Tm, T5, Tc;
@@ -909,7 +909,7 @@ static void t2_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T2, T5, T3, T6, T8, Td, T16, T14, Te, T9, T21, T23, Tx, TR, T1g;
 	       E TB, T1f, TV, T1Q, Tg, T1S, Tk, T18, T2s, T1c, T2q, Tn, To, Tp, Tr;
 	       E T28, T2x, TY, T2k, T2m, T2v, TG, TE, T10, T1h, T1E, T26, T1B, T1G, T1V;
diff --git a/dft/scalar/codelets/t2_32.c b/dft/scalar/codelets/t2_32.c
index b7d5167..2551a19 100644
--- a/dft/scalar/codelets/t2_32.c
+++ b/dft/scalar/codelets/t2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:00 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void t2_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T9A, T9z;
 	       {
 		    E T2, T8, T3, T6, Te, Tr, T18, T4, Ta, Tz, T1n, T10, Ti, T5, Tc;
@@ -983,7 +983,7 @@ static void t2_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T2, T5, T3, T6, T8, TM, TO, Td, T9, Te, Th, Tl, TD, TH, T1y;
 	       E T1H, T15, T1A, T11, T1F, T1n, T1p, T2q, T2I, T2u, T2K, T2V, T3b, T2Z, T3d;
 	       E Tu, Ty, T3l, T3n, T1t, T1v, T2f, T2h, T1a, T1e, T32, T34, T1W, T1Y, T2C;
diff --git a/dft/scalar/codelets/t2_4.c b/dft/scalar/codelets/t2_4.c
index 433a491..73c03e9 100644
--- a/dft/scalar/codelets/t2_4.c
+++ b/dft/scalar/codelets/t2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void t2_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E Ti, Tq, To, Te, Ty, Tz, Tm, Ts;
 	       {
 		    E T2, T6, T3, T5;
@@ -129,7 +129,7 @@ static void t2_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T2, T4, T3, T5, T6, T8;
 	       T2 = W[0];
 	       T4 = W[1];
diff --git a/dft/scalar/codelets/t2_5.c b/dft/scalar/codelets/t2_5.c
index eee662c..7e1e83a 100644
--- a/dft/scalar/codelets/t2_5.c
+++ b/dft/scalar/codelets/t2_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t2_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E Ta, T1, TO, Tp, TS, Ti, TL, TC, To, TE, Ts, TF, T2, T8, T5;
 	       E TT, Tt, TG;
 	       T2 = W[0];
@@ -168,7 +168,7 @@ static void t2_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 4, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T2, T4, T7, T9, Tb, Tl, Tf, Tj;
 	       {
 		    E T8, Te, Ta, Td;
diff --git a/dft/scalar/codelets/t2_64.c b/dft/scalar/codelets/t2_64.c
index ffa611e..0fbd9cd 100644
--- a/dft/scalar/codelets/t2_64.c
+++ b/dft/scalar/codelets/t2_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -53,7 +53,7 @@ static void t2_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tg0, TlC, TlB, Tg3;
 	       {
 		    E T2, T3, Tc, T8, Te, T5, T6, T14, T3d, T3i, TJ, T7, Tr, T3g, TG;
@@ -2142,7 +2142,7 @@ static void t2_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 10); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 10, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T2, T5, T3, T6, Te, T9, TP, T3e, T1e, T39, T3c, TT, T1a, T37, T8;
 	       E Tw, Td, Ty, Tm, Th, T1C, T3K, T1V, T3x, T3I, T1G, T1R, T3v, T2m, T2q;
 	       E T5Y, T6u, T53, T5B, T62, T6w, T57, T5D, T2V, T2X, Tg, TE, T3Y, T3V, T3j;
diff --git a/dft/scalar/codelets/t2_8.c b/dft/scalar/codelets/t2_8.c
index 0958b61..cb1dcc2 100644
--- a/dft/scalar/codelets/t2_8.c
+++ b/dft/scalar/codelets/t2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 10:59:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:35:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void t2_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E TS, T1m, TJ, T1l, T1k, Tw, T1w, T1u;
 	       {
 		    E T2, T3, Tl, Tn, T5, T4, Tm, Tr, T6;
@@ -230,7 +230,7 @@ static void t2_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + 1, ri = ri + ms, ii = ii + ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T2, T5, T3, T6, T8, Tc, Tg, Ti, Tl, Tm, Tn, Tz, Tp, Tx;
 	       {
 		    E T4, Tb, T7, Ta;
diff --git a/dft/simd/Makefile.in b/dft/simd/Makefile.in
index be393bb..2cb081a 100644
--- a/dft/simd/Makefile.in
+++ b/dft/simd/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -15,6 +15,23 @@
 
 @SET_MAKE@
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -62,6 +79,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -450,13 +472,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/dft/simd/altivec/Makefile.in b/dft/simd/altivec/Makefile.in
index b916b07..9f26f3f 100644
--- a/dft/simd/altivec/Makefile.in
+++ b/dft/simd/altivec/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -164,6 +181,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libdft_altivec_codelets_la_SOURCES)
 DIST_SOURCES = $(am__libdft_altivec_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/dft/simd/avx/Makefile.in b/dft/simd/avx/Makefile.in
index a32b6b6..d784a35 100644
--- a/dft/simd/avx/Makefile.in
+++ b/dft/simd/avx/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -162,6 +179,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libdft_avx_codelets_la_SOURCES)
 DIST_SOURCES = $(am__libdft_avx_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/dft/simd/common/Makefile.in b/dft/simd/common/Makefile.in
index 9c593c5..833d281 100644
--- a/dft/simd/common/Makefile.in
+++ b/dft/simd/common/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -33,6 +33,23 @@
 # This file contains special make rules to generate codelets.
 # Most of this file requires GNU make .
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -75,6 +92,11 @@ CONFIG_CLEAN_FILES =
 CONFIG_CLEAN_VPATH_FILES =
 SOURCES =
 DIST_SOURCES =
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 ACLOCAL = @ACLOCAL@
 ALLOCA = @ALLOCA@
diff --git a/dft/simd/common/n1bv_10.c b/dft/simd/common/n1bv_10.c
index e149d38..280f82c 100644
--- a/dft/simd/common/n1bv_10.c
+++ b/dft/simd/common/n1bv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n1bv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Tb, Tr, T3, Ts, T6, Tw, Tg, Tt, T9, Tc, T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
@@ -149,7 +149,7 @@ static void n1bv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Tl, Ty, T7, Te, Tw, Tt, Tz, TA, TB, Tg, Th, Tm, Tj, Tk;
 	       Tj = LD(&(xi[0]), ivs, &(xi[0]));
 	       Tk = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1bv_11.c b/dft/simd/common/n1bv_11.c
index 6230da0..a3bf25a 100644
--- a/dft/simd/common/n1bv_11.c
+++ b/dft/simd/common/n1bv_11.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:00 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -53,7 +53,7 @@ static void n1bv_11(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(22, is), MAKE_VOLATILE_STRIDE(22, os)) {
 	       V T1, Tb, T4, Tq, Tg, Tm, T7, Tp, Ta, To, Tc, T11;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
@@ -199,7 +199,7 @@ static void n1bv_11(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(22, is), MAKE_VOLATILE_STRIDE(22, os)) {
 	       V Th, T3, Tm, Tf, Ti, Tc, Tj, T9, Tk, T6, Tl, Ta, Tb, Ts, Tt;
 	       Th = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
diff --git a/dft/simd/common/n1bv_12.c b/dft/simd/common/n1bv_12.c
index 4036351..4abc70c 100644
--- a/dft/simd/common/n1bv_12.c
+++ b/dft/simd/common/n1bv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:10 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T1, T6, Tc, Th, Td, Te, Ti, Tz, T4, TA, T9, Tj, Tf, Tw;
 	       {
 		    V T2, T3, T7, T8;
@@ -153,7 +153,7 @@ static void n1bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T5, Ta, TG, TF, Ty, Tm, Ti, Tp, TJ, TI, Tx, Ts;
 	       {
 		    V T1, T6, T4, Tk, T9, Tl;
diff --git a/dft/simd/common/n1bv_128.c b/dft/simd/common/n1bv_128.c
index 19004eb..35abd3d 100644
--- a/dft/simd/common/n1bv_128.c
+++ b/dft/simd/common/n1bv_128.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:18 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:08 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -73,7 +73,7 @@ static void n1bv_128(const R *ri, const R *ii, R *ro, R *io, stride is, stride o
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       V T6a, T5J, T6b, T5K, T6B, T6C, T6J, T6A, T6o, T6j, T6r, T68, T6d, T5O, T5R;
 	       V T6e, T6D, T6K;
 	       {
@@ -1760,7 +1760,7 @@ static void n1bv_128(const R *ri, const R *ii, R *ro, R *io, stride is, stride o
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       V T49, T6e, Tev, TgK, TfA, TgL, T4U, T5J, T7R, T9o, Tah, TdG, Tcw, TdB, T84;
 	       V T8T, Tfk, Tfo, T1G, T64, Tgs, Th6, T2p, T62, T7t, T9c, Tce, Tdm, T7i, T9e;
 	       V Tc8, Tdp, TgF, TgG, T4q, T4V, TeC, Tfx, T4H, T4W, T7X, T86, Tcr, TdH, T7U;
diff --git a/dft/simd/common/n1bv_13.c b/dft/simd/common/n1bv_13.c
index 011d420..8d2b260 100644
--- a/dft/simd/common/n1bv_13.c
+++ b/dft/simd/common/n1bv_13.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -65,7 +65,7 @@ static void n1bv_13(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(26, is), MAKE_VOLATILE_STRIDE(26, os)) {
 	       V T1, T7, T2, Tg, Tf, TN, Th, Tq, Ta, Tj, T5, Tr, Tk;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
@@ -251,7 +251,7 @@ static void n1bv_13(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(26, is), MAKE_VOLATILE_STRIDE(26, os)) {
 	       V TW, Tb, Tm, Ts, TB, TR, TX, TK, TU, Tz, TC, TN, TT;
 	       TW = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
diff --git a/dft/simd/common/n1bv_14.c b/dft/simd/common/n1bv_14.c
index c68a52f..7dcf9cc 100644
--- a/dft/simd/common/n1bv_14.c
+++ b/dft/simd/common/n1bv_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n1bv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V TH, T3, TP, Tn, Ta, Tu, TU, TK, TO, Tk, TM, Tg, TL, Td, T1;
 	       V T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
@@ -198,7 +198,7 @@ static void n1bv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V Tp, Ty, Tl, TL, Tq, TE, T7, TJ, Ts, TB, Te, TK, Tr, TH, Tn;
 	       V To;
 	       Tn = LD(&(xi[0]), ivs, &(xi[0]));
diff --git a/dft/simd/common/n1bv_15.c b/dft/simd/common/n1bv_15.c
index 223b1f4..4e6feec 100644
--- a/dft/simd/common/n1bv_15.c
+++ b/dft/simd/common/n1bv_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:13 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -50,7 +50,7 @@ static void n1bv_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(30, is), MAKE_VOLATILE_STRIDE(30, os)) {
 	       V Tb, TH, Tw, TA, Th, T11, T5, Ti, T12, Ta, Tx, Te, Tq, T16, Tj;
 	       V T1, T2, T3;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
@@ -210,7 +210,7 @@ static void n1bv_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(30, is), MAKE_VOLATILE_STRIDE(30, os)) {
 	       V Ti, T11, TH, Ts, TL, TM, Tz, TC, TD, TI, T12, T13, T14, T15, T16;
 	       V T17, Tf, Tj, TZ, T10;
 	       {
diff --git a/dft/simd/common/n1bv_16.c b/dft/simd/common/n1bv_16.c
index ac22c55..b495a9e 100644
--- a/dft/simd/common/n1bv_16.c
+++ b/dft/simd/common/n1bv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:13 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void n1bv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V T7, Tu, TF, TB, T13, TL, TO, TX, TC, Te, TP, Th, TQ, Tk, TW;
 	       V T16;
 	       {
@@ -196,7 +196,7 @@ static void n1bv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V Tp, T13, Tu, TY, Tm, T14, Tv, TU, T7, T16, Tx, TN, Te, T17, Ty;
 	       V TQ;
 	       {
diff --git a/dft/simd/common/n1bv_2.c b/dft/simd/common/n1bv_2.c
index c304ee6..526c449 100644
--- a/dft/simd/common/n1bv_2.c
+++ b/dft/simd/common/n1bv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:58 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1bv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -78,7 +78,7 @@ static void n1bv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1bv_20.c b/dft/simd/common/n1bv_20.c
index e44c9ba..93ff25d 100644
--- a/dft/simd/common/n1bv_20.c
+++ b/dft/simd/common/n1bv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:14 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n1bv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V TS, TA, TN, TV, TK, TU, TR, Tl;
 	       {
 		    V T3, TE, T1r, T13, Ta, TL, Tz, TG, Ts, TF, Th, TM, T1u, T1C, T1n;
@@ -238,7 +238,7 @@ static void n1bv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V T3, T1y, TH, T1i, Ts, TL, TM, Tz, T13, T16, T1j, T1u, T1v, T1w, T1r;
 	       V T1s, T1t, T1a, T1d, T1k, Ti, Tk, TE, TI, TZ, T10;
 	       {
diff --git a/dft/simd/common/n1bv_25.c b/dft/simd/common/n1bv_25.c
index 0c9bcdf..4f6239c 100644
--- a/dft/simd/common/n1bv_25.c
+++ b/dft/simd/common/n1bv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:14 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -109,7 +109,7 @@ static void n1bv_25(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(50, is), MAKE_VOLATILE_STRIDE(50, os)) {
 	       V T1g, T1k, T1I, T24, T2a, T1G, T1A, T1l, T1B, T1H, T1d;
 	       {
 		    V T2z, T1q, Ta, T9, T3n, Ty, Tl, T2O, T2W, T2l, T2s, TV, T1i, T1K, T1S;
@@ -512,7 +512,7 @@ static void n1bv_25(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(50, is), MAKE_VOLATILE_STRIDE(50, os)) {
 	       V T1b, T2o, T1v, T1e, T2W, T2P, T2Q, T2U, T11, T27, TY, T26, T12, T2f, T1j;
 	       V T28, TM, T24, TJ, T23, TN, T2e, T1i, T25, T2M, T2N, T2T, Tm, T1W, Tt;
 	       V T1X, Tu, T20, Tw, T1Y, T7, T1U, Te, T1T, Tf, T21, Tx, T1V;
diff --git a/dft/simd/common/n1bv_3.c b/dft/simd/common/n1bv_3.c
index 5f9901e..94cc2d7 100644
--- a/dft/simd/common/n1bv_3.c
+++ b/dft/simd/common/n1bv_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:58 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1bv_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(6, is), MAKE_VOLATILE_STRIDE(6, os)) {
 	       V T1, T2, T3, T6, T4, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -87,7 +87,7 @@ static void n1bv_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(6, is), MAKE_VOLATILE_STRIDE(6, os)) {
 	       V T4, T3, T5, T1, T2, T6;
 	       T4 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T1 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1bv_32.c b/dft/simd/common/n1bv_32.c
index a818f72..d394933 100644
--- a/dft/simd/common/n1bv_32.c
+++ b/dft/simd/common/n1bv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:13 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -49,7 +49,7 @@ static void n1bv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T1h, Tr, T1a, T1k, TI, T1b, T1L, T1P, T1I, T1G, T1O, T1Q, T1H, T1z, T1c;
 	       V TZ;
 	       {
@@ -367,7 +367,7 @@ static void n1bv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T2f, T2k, T2N, T2M, T19, T1B, Tb, T1p, TT, T1v, TY, T1w, T2E, T2F, T2G;
 	       V T24, T2o, TC, T1s, TH, T1t, T2B, T2C, T2D, T1X, T2n, T2I, T2J, Tq, T1A;
 	       V T14, T1q, T2c, T2l;
diff --git a/dft/simd/common/n1bv_4.c b/dft/simd/common/n1bv_4.c
index b24f224..d167a6c 100644
--- a/dft/simd/common/n1bv_4.c
+++ b/dft/simd/common/n1bv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:58 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1bv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T1, T2, T4, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
@@ -89,7 +89,7 @@ static void n1bv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T3, T7, T6, T8;
 	       {
 		    V T1, T2, T4, T5;
diff --git a/dft/simd/common/n1bv_5.c b/dft/simd/common/n1bv_5.c
index 12f6417..a132543 100644
--- a/dft/simd/common/n1bv_5.c
+++ b/dft/simd/common/n1bv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:58 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n1bv_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(10, is), MAKE_VOLATILE_STRIDE(10, os)) {
 	       V T1, T2, T3, T5, T6;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -109,7 +109,7 @@ static void n1bv_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(10, is), MAKE_VOLATILE_STRIDE(10, os)) {
 	       V Tb, T3, Tc, T6, Ta;
 	       Tb = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
diff --git a/dft/simd/common/n1bv_6.c b/dft/simd/common/n1bv_6.c
index 20f551d..8686a75 100644
--- a/dft/simd/common/n1bv_6.c
+++ b/dft/simd/common/n1bv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:58 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1bv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V T1, T2, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
@@ -108,7 +108,7 @@ static void n1bv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V Ta, Td, T3, Te, T6, Tf, Tb, Tg, T8, T9;
 	       T8 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T9 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1bv_64.c b/dft/simd/common/n1bv_64.c
index 8063124..bee58f5 100644
--- a/dft/simd/common/n1bv_64.c
+++ b/dft/simd/common/n1bv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:15 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:05 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -57,7 +57,7 @@ static void n1bv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T5T, T5S, T5X, T65, T5Z, T5R, T67, T63, T5U, T64;
 	       {
 		    V T7, T26, T5k, T6A, T47, T69, T2V, T3z, T6B, T4e, T6a, T5n, T3M, T2Y, T27;
@@ -795,7 +795,7 @@ static void n1bv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T4p, T5u, Tb, T3A, T2q, T3v, T6G, T78, Tq, T3w, T6B, T79, T2l, T3B, T4w;
 	       V T5r, TI, T2g, T6u, T74, T3q, T3D, T4E, T5o, TZ, T2h, T6x, T75, T3t, T3E;
 	       V T4L, T5p, T23, T2N, T6m, T70, T6p, T71, T2c, T2O, T3i, T3Y, T5f, T5R, T5k;
diff --git a/dft/simd/common/n1bv_7.c b/dft/simd/common/n1bv_7.c
index 5d110bb..74e58be 100644
--- a/dft/simd/common/n1bv_7.c
+++ b/dft/simd/common/n1bv_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:58 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n1bv_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(14, is), MAKE_VOLATILE_STRIDE(14, os)) {
 	       V T1, T2, T3, T8, T9, T5, T6;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -133,7 +133,7 @@ static void n1bv_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(14, is), MAKE_VOLATILE_STRIDE(14, os)) {
 	       V Tb, T9, Tc, T3, Te, T6, Td, T7, T8, Ti, Tj;
 	       Tb = LD(&(xi[0]), ivs, &(xi[0]));
 	       T7 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
diff --git a/dft/simd/common/n1bv_8.c b/dft/simd/common/n1bv_8.c
index 7485537..e7115d1 100644
--- a/dft/simd/common/n1bv_8.c
+++ b/dft/simd/common/n1bv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void n1bv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T1, T2, Tc, Td, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
@@ -119,7 +119,7 @@ static void n1bv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V Ta, Tk, Te, Tj, T7, Tn, Tf, Tm;
 	       {
 		    V T8, T9, Tc, Td;
diff --git a/dft/simd/common/n1bv_9.c b/dft/simd/common/n1bv_9.c
index d50ea58..4f2462b 100644
--- a/dft/simd/common/n1bv_9.c
+++ b/dft/simd/common/n1bv_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -61,7 +61,7 @@ static void n1bv_9(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(18, is), MAKE_VOLATILE_STRIDE(18, os)) {
 	       V T1, T2, T3, T6, Tf, T7, T8, Tb, Tc, Tp, T4;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
@@ -177,7 +177,7 @@ static void n1bv_9(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(18, is), MAKE_VOLATILE_STRIDE(18, os)) {
 	       V T5, Ty, Tm, Ti, Tw, Th, Tj, To, Tb, Tv, Ta, Tc, Tn;
 	       {
 		    V T1, T2, T3, T4;
diff --git a/dft/simd/common/n1fv_10.c b/dft/simd/common/n1fv_10.c
index d064ac4..e743cda 100644
--- a/dft/simd/common/n1fv_10.c
+++ b/dft/simd/common/n1fv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n1fv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Tb, Tr, T3, Ts, T6, Tw, Tg, Tt, T9, Tc, T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
@@ -149,7 +149,7 @@ static void n1fv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Ti, Ty, Tm, Tn, Tw, Tt, Tz, TA, TB, T7, Te, Tj, Tg, Th;
 	       Tg = LD(&(xi[0]), ivs, &(xi[0]));
 	       Th = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1fv_11.c b/dft/simd/common/n1fv_11.c
index 3814fd2..0e43d82 100644
--- a/dft/simd/common/n1fv_11.c
+++ b/dft/simd/common/n1fv_11.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -53,7 +53,7 @@ static void n1fv_11(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(22, is), MAKE_VOLATILE_STRIDE(22, os)) {
 	       V T1, Tb, T4, Tp, Tg, Tq, T7, Tn, Ta, Tm, Tc, Tr;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
@@ -199,7 +199,7 @@ static void n1fv_11(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(22, is), MAKE_VOLATILE_STRIDE(22, os)) {
 	       V T1, T4, Ti, Tg, Tl, Td, Tk, Ta, Tj, T7, Tm, Tb, Tc, Tt, Ts;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
diff --git a/dft/simd/common/n1fv_12.c b/dft/simd/common/n1fv_12.c
index 8db5a31..9ebdbfe 100644
--- a/dft/simd/common/n1fv_12.c
+++ b/dft/simd/common/n1fv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1fv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T1, T6, Tk, Tn, Tc, Td, Tf, Tr, T4, Ts, T9, Tg, Te, Tl;
 	       {
 		    V T2, T3, T7, T8;
@@ -156,7 +156,7 @@ static void n1fv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T5, Ta, TJ, Ty, Tq, Tp, Tg, Tl, TI, TA, Tz, Tu;
 	       {
 		    V T1, T6, T4, Tw, T9, Tx;
diff --git a/dft/simd/common/n1fv_128.c b/dft/simd/common/n1fv_128.c
index d7e10d4..737735a 100644
--- a/dft/simd/common/n1fv_128.c
+++ b/dft/simd/common/n1fv_128.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:54 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -73,7 +73,7 @@ static void n1fv_128(const R *ri, const R *ii, R *ro, R *io, stride is, stride o
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       V T6a, T5J, T6b, T5K, T6B, T6C, T6J, T6A, T6o, T6j, T6r, T68, T6e, T5O, T5R;
 	       V T6d, T6D, T6K;
 	       {
@@ -1760,7 +1760,7 @@ static void n1fv_128(const R *ri, const R *ii, R *ro, R *io, stride is, stride o
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       V Tr, T5J, Ted, Tgf, Tfq, TgH, T4U, T6b, T6Z, T8T, Tad, TcZ, Tcc, Td0, T84;
 	       V T9l, Tb6, Tbt, T2G, T5X, TeV, Tgr, T3p, T5V, T7B, T95, TeK, Tgt, T7q, T97;
 	       V Td8, TdK, TbD, Tc0, T3V, T61, Tfg, TgB, T4E, T65, T7U, T9f, Tf5, Tgx, T7J;
diff --git a/dft/simd/common/n1fv_13.c b/dft/simd/common/n1fv_13.c
index 223a27a..a5b61a4 100644
--- a/dft/simd/common/n1fv_13.c
+++ b/dft/simd/common/n1fv_13.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -65,7 +65,7 @@ static void n1fv_13(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(26, is), MAKE_VOLATILE_STRIDE(26, os)) {
 	       V T1, T7, T2, Tg, Tf, TN, Th, Tq, Ta, Tj, T5, Tr, Tk;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
@@ -251,7 +251,7 @@ static void n1fv_13(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(26, is), MAKE_VOLATILE_STRIDE(26, os)) {
 	       V TW, Tb, Tm, Tu, TC, TR, TX, TK, TU, Tz, TB, TN, TT;
 	       TW = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
diff --git a/dft/simd/common/n1fv_14.c b/dft/simd/common/n1fv_14.c
index 30630ac..147e509 100644
--- a/dft/simd/common/n1fv_14.c
+++ b/dft/simd/common/n1fv_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n1fv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V TH, T3, TP, Tn, Ta, Ts, TW, TK, TO, Tk, TM, Tg, TL, Td, T1;
 	       V T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
@@ -198,7 +198,7 @@ static void n1fv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V T3, Ty, To, TK, Tr, TE, Ta, TJ, Tq, TB, Th, TL, Ts, TH, T1;
 	       V T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
diff --git a/dft/simd/common/n1fv_15.c b/dft/simd/common/n1fv_15.c
index 9d3c940..38404f3 100644
--- a/dft/simd/common/n1fv_15.c
+++ b/dft/simd/common/n1fv_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -50,7 +50,7 @@ static void n1fv_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(30, is), MAKE_VOLATILE_STRIDE(30, os)) {
 	       V Tb, TX, TM, TQ, Th, TB, T5, Ti, Ta, TC, TN, Te, TG, Tq, Tj;
 	       V T1, T2, T3;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
@@ -210,7 +210,7 @@ static void n1fv_15(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(30, is), MAKE_VOLATILE_STRIDE(30, os)) {
 	       V T5, T10, TB, TO, TU, TV, TR, Ta, Tf, Tg, Tl, Tq, Tr, TE, TH;
 	       V TI, TZ, T11, T1f, T1g;
 	       {
diff --git a/dft/simd/common/n1fv_16.c b/dft/simd/common/n1fv_16.c
index 4efe64e..56c8451 100644
--- a/dft/simd/common/n1fv_16.c
+++ b/dft/simd/common/n1fv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:53 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void n1fv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V T7, Tu, TF, TB, T13, TL, TO, TX, TC, Te, TP, Th, TQ, Tk, TW;
 	       V T16;
 	       {
@@ -196,7 +196,7 @@ static void n1fv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V Tp, T13, Tu, TN, Tm, T14, Tv, TY, T7, T17, Ty, TT, Te, T16, Tx;
 	       V TQ;
 	       {
diff --git a/dft/simd/common/n1fv_2.c b/dft/simd/common/n1fv_2.c
index d5d149f..d5bc9ae 100644
--- a/dft/simd/common/n1fv_2.c
+++ b/dft/simd/common/n1fv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1fv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -78,7 +78,7 @@ static void n1fv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1fv_20.c b/dft/simd/common/n1fv_20.c
index 9f2487e..e819fc9 100644
--- a/dft/simd/common/n1fv_20.c
+++ b/dft/simd/common/n1fv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:54 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n1fv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V TU, TI, TP, TX, TM, TW, TT, TF;
 	       {
 		    V T3, Tm, T1r, T13, Ta, TN, TH, TA, TG, Tt, Th, TO, T1u, T1C, T1n;
@@ -238,7 +238,7 @@ static void n1fv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V T3, T1B, Tm, T1i, TG, TN, TO, TH, T13, T16, T1k, T1u, T1v, T1z, T1r;
 	       V T1s, T1y, T1a, T1d, T1j, Ti, TD, TB, TL, Tj, TC;
 	       {
diff --git a/dft/simd/common/n1fv_25.c b/dft/simd/common/n1fv_25.c
index a236d17..f1611e6 100644
--- a/dft/simd/common/n1fv_25.c
+++ b/dft/simd/common/n1fv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:55 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -109,7 +109,7 @@ static void n1fv_25(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(50, is), MAKE_VOLATILE_STRIDE(50, os)) {
 	       V T1g, T1k, T1I, T24, T2a, T1G, T1A, T1l, T1B, T1H, T1d;
 	       {
 		    V T2z, T1q, Ta, T9, T3n, Ty, Tl, T2O, T2W, T2l, T2s, TV, T1i, T1K, T1S;
@@ -512,7 +512,7 @@ static void n1fv_25(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(50, is), MAKE_VOLATILE_STRIDE(50, os)) {
 	       V T7, T1g, T26, Ta, T2R, T2N, T2O, T2P, T19, T1Y, T16, T1Z, T1a, T2v, T1l;
 	       V T2m, TU, T21, TR, T22, TV, T2u, T1k, T2l, T2K, T2L, T2M, TE, T1R, TB;
 	       V T1S, TF, T2r, T1i, T2j, Tp, T1U, Tm, T1V, Tq, T2s, T1h, T2i;
diff --git a/dft/simd/common/n1fv_3.c b/dft/simd/common/n1fv_3.c
index ba02ea4..604480a 100644
--- a/dft/simd/common/n1fv_3.c
+++ b/dft/simd/common/n1fv_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1fv_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(6, is), MAKE_VOLATILE_STRIDE(6, os)) {
 	       V T1, T2, T3, T6, T4, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -87,7 +87,7 @@ static void n1fv_3(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(6, is), MAKE_VOLATILE_STRIDE(6, os)) {
 	       V T1, T4, T6, T2, T3, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1fv_32.c b/dft/simd/common/n1fv_32.c
index 8cbeaf7..d3e4a02 100644
--- a/dft/simd/common/n1fv_32.c
+++ b/dft/simd/common/n1fv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:54 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -49,7 +49,7 @@ static void n1fv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T1h, Tr, T1a, T1k, TI, T1b, T1L, T1P, T1I, T1G, T1O, T1Q, T1H, T1z, T1c;
 	       V TZ;
 	       {
@@ -367,7 +367,7 @@ static void n1fv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T1T, T1W, T2K, T2x, T16, T1A, Tb, T1p, TT, T1v, TY, T1w, T27, T2a, T2b;
 	       V T2H, T2O, TC, T1s, TH, T1t, T20, T23, T24, T2E, T2N, T2g, T2j, Tq, T1B;
 	       V T19, T1q, T2A, T2L;
diff --git a/dft/simd/common/n1fv_4.c b/dft/simd/common/n1fv_4.c
index c2e274e..837e16f 100644
--- a/dft/simd/common/n1fv_4.c
+++ b/dft/simd/common/n1fv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n1fv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T1, T2, T4, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
@@ -89,7 +89,7 @@ static void n1fv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T3, T7, T6, T8;
 	       {
 		    V T1, T2, T4, T5;
diff --git a/dft/simd/common/n1fv_5.c b/dft/simd/common/n1fv_5.c
index dc1b6fc..f96e498 100644
--- a/dft/simd/common/n1fv_5.c
+++ b/dft/simd/common/n1fv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n1fv_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(10, is), MAKE_VOLATILE_STRIDE(10, os)) {
 	       V T1, T2, T3, T5, T6;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -109,7 +109,7 @@ static void n1fv_5(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(10, is), MAKE_VOLATILE_STRIDE(10, os)) {
 	       V T8, T7, Td, T9, Tc;
 	       T8 = LD(&(xi[0]), ivs, &(xi[0]));
 	       {
diff --git a/dft/simd/common/n1fv_6.c b/dft/simd/common/n1fv_6.c
index 903c8e0..0a27954 100644
--- a/dft/simd/common/n1fv_6.c
+++ b/dft/simd/common/n1fv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n1fv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V T1, T2, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
@@ -108,7 +108,7 @@ static void n1fv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V T3, Td, T6, Te, T9, Tf, Ta, Tg, T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1fv_64.c b/dft/simd/common/n1fv_64.c
index 2520674..cd77e5c 100644
--- a/dft/simd/common/n1fv_64.c
+++ b/dft/simd/common/n1fv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:54 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -57,7 +57,7 @@ static void n1fv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T5T, T5S, T5X, T65, T5Z, T5R, T67, T63, T5U, T64;
 	       {
 		    V T7, T26, T5k, T6A, T47, T69, T2V, T3z, T6B, T4e, T6a, T5n, T3M, T2Y, T27;
@@ -795,7 +795,7 @@ static void n1fv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T4p, T5q, Tb, T39, T2n, T3A, T6f, T6T, Tq, T3B, T6i, T76, T2i, T3a, T4w;
 	       V T5r, TI, T2p, T6C, T6V, T3h, T3E, T4L, T5u, TZ, T2q, T6F, T6U, T3e, T3D;
 	       V T4E, T5t, T23, T2N, T6t, T71, T6w, T72, T2c, T2O, T3t, T41, T5f, T5R, T5k;
diff --git a/dft/simd/common/n1fv_7.c b/dft/simd/common/n1fv_7.c
index e1f2994..821982e 100644
--- a/dft/simd/common/n1fv_7.c
+++ b/dft/simd/common/n1fv_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n1fv_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(14, is), MAKE_VOLATILE_STRIDE(14, os)) {
 	       V T1, T2, T3, T8, T9, T5, T6;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -133,7 +133,7 @@ static void n1fv_7(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(14, is), MAKE_VOLATILE_STRIDE(14, os)) {
 	       V T1, Ta, Td, T4, Tc, T7, Te, T8, T9, Tj, Ti;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T8 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n1fv_8.c b/dft/simd/common/n1fv_8.c
index 7553e21..81b3bd9 100644
--- a/dft/simd/common/n1fv_8.c
+++ b/dft/simd/common/n1fv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void n1fv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T1, T2, Tc, Td, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
@@ -119,7 +119,7 @@ static void n1fv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T3, Tj, Tf, Tk, Ta, Tn, Tc, Tm;
 	       {
 		    V T1, T2, Td, Te;
diff --git a/dft/simd/common/n1fv_9.c b/dft/simd/common/n1fv_9.c
index abee5be..4040026 100644
--- a/dft/simd/common/n1fv_9.c
+++ b/dft/simd/common/n1fv_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:36:52 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -61,7 +61,7 @@ static void n1fv_9(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(18, is), MAKE_VOLATILE_STRIDE(18, os)) {
 	       V T1, T2, T3, T6, Tb, T7, T8, Tc, Td, Tv, T4;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
@@ -177,7 +177,7 @@ static void n1fv_9(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(18, is), MAKE_VOLATILE_STRIDE(18, os)) {
 	       V T5, Ts, Tj, To, Tf, Tn, Tp, Tu, Tl, Ta, Tk, Tm, Tt;
 	       {
 		    V T1, T2, T3, T4;
diff --git a/dft/simd/common/n2bv_10.c b/dft/simd/common/n2bv_10.c
index ea6d9ea..dfcff11 100644
--- a/dft/simd/common/n2bv_10.c
+++ b/dft/simd/common/n2bv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:29 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n2bv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Tb, Tr, T3, Ts, T6, Tw, Tg, Tt, T9, Tc, T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
@@ -173,7 +173,7 @@ static void n2bv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Tl, Ty, T7, Te, Tw, Tt, Tz, TA, TB, Tg, Th, Tm, Tj, Tk;
 	       Tj = LD(&(xi[0]), ivs, &(xi[0]));
 	       Tk = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n2bv_12.c b/dft/simd/common/n2bv_12.c
index aea6e6a..ed8e869 100644
--- a/dft/simd/common/n2bv_12.c
+++ b/dft/simd/common/n2bv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:30 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n2bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T1, T6, Tc, Th, Td, Te, Ti, Tz, T4, TA, T9, Tj, Tf, Tw;
 	       {
 		    V T2, T3, T7, T8;
@@ -180,7 +180,7 @@ static void n2bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T5, Ta, TG, TF, Ty, Tm, Ti, Tp, TJ, TI, Tx, Ts;
 	       {
 		    V T1, T6, T4, Tk, T9, Tl;
diff --git a/dft/simd/common/n2bv_14.c b/dft/simd/common/n2bv_14.c
index baefde1..8814701 100644
--- a/dft/simd/common/n2bv_14.c
+++ b/dft/simd/common/n2bv_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:30 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n2bv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V TH, T3, TP, Tn, Ta, Tu, TU, TK, TO, Tk, TM, Tg, TL, Td, T1;
 	       V T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
@@ -229,7 +229,7 @@ static void n2bv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V Tp, Ty, Tl, TL, Tq, TE, T7, TJ, Ts, TB, Te, TK, Tr, TH, Tn;
 	       V To;
 	       Tn = LD(&(xi[0]), ivs, &(xi[0]));
diff --git a/dft/simd/common/n2bv_16.c b/dft/simd/common/n2bv_16.c
index 08b7f14..1a27040 100644
--- a/dft/simd/common/n2bv_16.c
+++ b/dft/simd/common/n2bv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:31 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void n2bv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V T7, Tu, TF, TB, T13, TL, TO, TX, TC, Te, TP, Th, TQ, Tk, TW;
 	       V T16;
 	       {
@@ -232,7 +232,7 @@ static void n2bv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V Tp, T13, Tu, TY, Tm, T14, Tv, TU, T7, T16, Tx, TN, Te, T17, Ty;
 	       V TQ;
 	       {
diff --git a/dft/simd/common/n2bv_2.c b/dft/simd/common/n2bv_2.c
index 8f5eca7..fe313b2 100644
--- a/dft/simd/common/n2bv_2.c
+++ b/dft/simd/common/n2bv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:29 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n2bv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2, T3, T4;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -81,7 +81,7 @@ static void n2bv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2, T3, T4;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n2bv_20.c b/dft/simd/common/n2bv_20.c
index 57fcd17..5aa39f2 100644
--- a/dft/simd/common/n2bv_20.c
+++ b/dft/simd/common/n2bv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:49 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:46 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n2bv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V T1H, T1I, TS, TA, TN, TV, T1M, T1N, T1O, T1P, T1R, T1S, TK, TU, TR;
 	       V Tl;
 	       {
@@ -275,7 +275,7 @@ static void n2bv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V T3, T1y, TH, T1i, Ts, TL, TM, Tz, T13, T16, T1j, T1u, T1v, T1w, T1r;
 	       V T1s, T1t, T1a, T1d, T1k, Ti, Tk, TE, TI;
 	       {
diff --git a/dft/simd/common/n2bv_32.c b/dft/simd/common/n2bv_32.c
index b603e89..48ab5db 100644
--- a/dft/simd/common/n2bv_32.c
+++ b/dft/simd/common/n2bv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:32 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -49,7 +49,7 @@ static void n2bv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T31, T32, T33, T34, T35, T36, T37, T38, T39, T3a, T3b, T3c, T1h, Tr, T3d;
 	       V T3e, T3f, T3g, T1a, T1k, TI, T1b, T1L, T1P, T1I, T1G, T1O, T1Q, T1H, T1z;
 	       V T1c, TZ;
@@ -428,7 +428,7 @@ static void n2bv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T2f, T2k, T2N, T2M, T19, T1B, Tb, T1p, TT, T1v, TY, T1w, T2E, T2F, T2G;
 	       V T24, T2o, TC, T1s, TH, T1t, T2B, T2C, T2D, T1X, T2n, T2I, T2J, Tq, T1A;
 	       V T14, T1q, T2c, T2l;
diff --git a/dft/simd/common/n2bv_4.c b/dft/simd/common/n2bv_4.c
index fae8160..5e6afc9 100644
--- a/dft/simd/common/n2bv_4.c
+++ b/dft/simd/common/n2bv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:29 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n2bv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T1, T2, T4, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
@@ -98,7 +98,7 @@ static void n2bv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T3, T7, T6, T8;
 	       {
 		    V T1, T2, T4, T5;
diff --git a/dft/simd/common/n2bv_6.c b/dft/simd/common/n2bv_6.c
index 23d80a5..8b715fd 100644
--- a/dft/simd/common/n2bv_6.c
+++ b/dft/simd/common/n2bv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:29 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n2bv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V T1, T2, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
@@ -123,7 +123,7 @@ static void n2bv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V Ta, Td, T3, Te, T6, Tf, Tb, Tg, T8, T9, Tj, Tk;
 	       T8 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T9 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n2bv_64.c b/dft/simd/common/n2bv_64.c
index dca1d0d..b23b8bb 100644
--- a/dft/simd/common/n2bv_64.c
+++ b/dft/simd/common/n2bv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:44 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:36 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -57,7 +57,7 @@ static void n2bv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T7z, T7A, T7B, T7C, T5T, T5S, T5X, T65, T8a, T8b, T8e, T8g, T5Z, T5R, T67;
 	       V T63, T5U, T64;
 	       {
@@ -918,7 +918,7 @@ static void n2bv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T4p, T5u, Tb, T3A, T2q, T3v, T6G, T78, Tq, T3w, T6B, T79, T2l, T3B, T4w;
 	       V T5r, TI, T2g, T6u, T74, T3q, T3D, T4E, T5o, TZ, T2h, T6x, T75, T3t, T3E;
 	       V T4L, T5p, T23, T2N, T6m, T70, T6p, T71, T2c, T2O, T3i, T3Y, T5f, T5R, T5k;
diff --git a/dft/simd/common/n2bv_8.c b/dft/simd/common/n2bv_8.c
index 0eb0518..a7300f2 100644
--- a/dft/simd/common/n2bv_8.c
+++ b/dft/simd/common/n2bv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:29 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void n2bv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T1, T2, Tc, Td, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
@@ -134,7 +134,7 @@ static void n2bv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ii;
 	  xo = io;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V Ta, Tk, Te, Tj, T7, Tn, Tf, Tm, Tr, Tu;
 	       {
 		    V T8, T9, Tc, Td;
diff --git a/dft/simd/common/n2fv_10.c b/dft/simd/common/n2fv_10.c
index 2a04cf0..01c61bd 100644
--- a/dft/simd/common/n2fv_10.c
+++ b/dft/simd/common/n2fv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:27 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:22 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n2fv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Tb, Tr, T3, Ts, T6, Tw, Tg, Tt, T9, Tc, T1, T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
@@ -173,7 +173,7 @@ static void n2fv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(20, is), MAKE_VOLATILE_STRIDE(20, os)) {
 	       V Ti, Ty, Tm, Tn, Tw, Tt, Tz, TA, TB, T7, Te, Tj, Tg, Th;
 	       Tg = LD(&(xi[0]), ivs, &(xi[0]));
 	       Th = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n2fv_12.c b/dft/simd/common/n2fv_12.c
index 6714943..5a9f03f 100644
--- a/dft/simd/common/n2fv_12.c
+++ b/dft/simd/common/n2fv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:22 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n2fv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T1, T6, Tk, Tn, Tc, Td, Tf, Tr, T4, Ts, T9, Tg, Te, Tl;
 	       {
 		    V T2, T3, T7, T8;
@@ -183,7 +183,7 @@ static void n2fv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(24, is), MAKE_VOLATILE_STRIDE(24, os)) {
 	       V T5, Ta, TJ, Ty, Tq, Tp, Tg, Tl, TI, TA, Tz, Tu;
 	       {
 		    V T1, T6, T4, Tw, T9, Tx;
diff --git a/dft/simd/common/n2fv_14.c b/dft/simd/common/n2fv_14.c
index 90eb133..4a96d90 100644
--- a/dft/simd/common/n2fv_14.c
+++ b/dft/simd/common/n2fv_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:22 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void n2fv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V TH, T3, TP, Tn, Ta, Ts, TW, TK, TO, Tk, TM, Tg, TL, Td, T1;
 	       V T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
@@ -229,7 +229,7 @@ static void n2fv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(28, is), MAKE_VOLATILE_STRIDE(28, os)) {
 	       V T3, Ty, To, TK, Tr, TE, Ta, TJ, Tq, TB, Th, TL, Ts, TH, T1;
 	       V T2;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
diff --git a/dft/simd/common/n2fv_16.c b/dft/simd/common/n2fv_16.c
index e5bda79..2c32976 100644
--- a/dft/simd/common/n2fv_16.c
+++ b/dft/simd/common/n2fv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:30 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:23 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void n2fv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V T7, Tu, TF, TB, T13, TL, TO, TX, TC, Te, TP, Th, TQ, Tk, TW;
 	       V T16;
 	       {
@@ -232,7 +232,7 @@ static void n2fv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V Tp, T13, Tu, TN, Tm, T14, Tv, TY, T7, T17, Ty, TT, Te, T16, Tx;
 	       V TQ;
 	       {
diff --git a/dft/simd/common/n2fv_2.c b/dft/simd/common/n2fv_2.c
index 0939f69..962f3eb 100644
--- a/dft/simd/common/n2fv_2.c
+++ b/dft/simd/common/n2fv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:27 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:21 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n2fv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2, T3, T4;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
@@ -81,7 +81,7 @@ static void n2fv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(4, is), MAKE_VOLATILE_STRIDE(4, os)) {
 	       V T1, T2, T3, T4;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n2fv_20.c b/dft/simd/common/n2fv_20.c
index 5c2dafc..cabc628 100644
--- a/dft/simd/common/n2fv_20.c
+++ b/dft/simd/common/n2fv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:28 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void n2fv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V T1H, T1I, TU, TI, TP, TX, T1M, T1N, T1O, T1P, T1R, T1S, TM, TW, TT;
 	       V TF;
 	       {
@@ -275,7 +275,7 @@ static void n2fv_20(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(40, is), MAKE_VOLATILE_STRIDE(40, os)) {
 	       V T3, T1B, Tm, T1i, TG, TN, TO, TH, T13, T16, T1k, T1u, T1v, T1z, T1r;
 	       V T1s, T1y, T1a, T1d, T1j, Ti, TD, TB, TL;
 	       {
diff --git a/dft/simd/common/n2fv_32.c b/dft/simd/common/n2fv_32.c
index f033e32..80e84c2 100644
--- a/dft/simd/common/n2fv_32.c
+++ b/dft/simd/common/n2fv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:30 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:24 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -49,7 +49,7 @@ static void n2fv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T31, T32, T33, T34, T35, T36, T37, T38, T39, T3a, T3b, T3c, T1h, Tr, T3d;
 	       V T3e, T3f, T3g, T1a, T1k, TI, T1b, T1L, T1P, T1I, T1G, T1O, T1Q, T1H, T1z;
 	       V T1c, TZ;
@@ -428,7 +428,7 @@ static void n2fv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T1T, T1W, T2K, T2x, T16, T1A, Tb, T1p, TT, T1v, TY, T1w, T27, T2a, T2b;
 	       V T2H, T2O, TC, T1s, TH, T1t, T20, T23, T24, T2E, T2N, T2g, T2j, Tq, T1B;
 	       V T19, T1q, T2A, T2L;
diff --git a/dft/simd/common/n2fv_4.c b/dft/simd/common/n2fv_4.c
index 03c8d53..bf33d8b 100644
--- a/dft/simd/common/n2fv_4.c
+++ b/dft/simd/common/n2fv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:27 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:21 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void n2fv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T1, T2, T4, T5;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
@@ -98,7 +98,7 @@ static void n2fv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(8, is), MAKE_VOLATILE_STRIDE(8, os)) {
 	       V T3, T7, T6, T8;
 	       {
 		    V T1, T2, T4, T5;
diff --git a/dft/simd/common/n2fv_6.c b/dft/simd/common/n2fv_6.c
index dfd1ce9..b1ad726 100644
--- a/dft/simd/common/n2fv_6.c
+++ b/dft/simd/common/n2fv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:27 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:21 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void n2fv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V T1, T2, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
@@ -123,7 +123,7 @@ static void n2fv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(12, is), MAKE_VOLATILE_STRIDE(12, os)) {
 	       V T3, Td, T6, Te, T9, Tf, Ta, Tg, T1, T2, Tj, Tk;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
diff --git a/dft/simd/common/n2fv_64.c b/dft/simd/common/n2fv_64.c
index bfc3c26..577ee17 100644
--- a/dft/simd/common/n2fv_64.c
+++ b/dft/simd/common/n2fv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:31 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:25 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -57,7 +57,7 @@ static void n2fv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T7r, T7s, T7t, T7u, T5T, T5S, T5X, T65, T8a, T8b, T8e, T8g, T5Z, T5R, T67;
 	       V T63, T5U, T64;
 	       {
@@ -918,7 +918,7 @@ static void n2fv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T4p, T5q, Tb, T39, T2n, T3A, T6f, T6T, Tq, T3B, T6i, T76, T2i, T3a, T4w;
 	       V T5r, TI, T2p, T6C, T6V, T3h, T3E, T4L, T5u, TZ, T2q, T6F, T6U, T3e, T3D;
 	       V T4E, T5t, T23, T2N, T6t, T71, T6w, T72, T2c, T2O, T3t, T41, T5f, T5R, T5k;
diff --git a/dft/simd/common/n2fv_8.c b/dft/simd/common/n2fv_8.c
index bc052df..4f5b63b 100644
--- a/dft/simd/common/n2fv_8.c
+++ b/dft/simd/common/n2fv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:27 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:22 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void n2fv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T1, T2, Tc, Td, T4, T5, T7, T8;
 	       T1 = LD(&(xi[0]), ivs, &(xi[0]));
 	       T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
@@ -134,7 +134,7 @@ static void n2fv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 	  R *xo;
 	  xi = ri;
 	  xo = ro;
-	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T3, Tj, Tf, Tk, Ta, Tn, Tc, Tm, Ts, Tu;
 	       {
 		    V T1, T2, Td, Te;
diff --git a/dft/simd/common/n2sv_16.c b/dft/simd/common/n2sv_16.c
index f18e8e2..a17d34a 100644
--- a/dft/simd/common/n2sv_16.c
+++ b/dft/simd/common/n2sv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:50 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:48 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void n2sv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
      DVK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T2p, T2q, T2r, T2s, T2x, T2y, T2z, T2A, T1M, T1N, T1L, T1P, T2F, T2G, T2H;
 	       V T2I, T1O, T1Q;
 	       {
@@ -344,7 +344,7 @@ static void n2sv_16(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(64, is), MAKE_VOLATILE_STRIDE(64, os)) {
 	       V T7, T1R, T25, TC, TN, T1x, T1H, T1l, Tt, T22, T2h, T1b, T1g, T1E, T1Z;
 	       V T1D, Te, T1S, T26, TJ, TQ, T1m, T1n, TT, Tm, T1X, T2g, T10, T15, T1B;
 	       V T1U, T1A;
diff --git a/dft/simd/common/n2sv_32.c b/dft/simd/common/n2sv_32.c
index 9fc5fb0..76e8f63 100644
--- a/dft/simd/common/n2sv_32.c
+++ b/dft/simd/common/n2sv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void n2sv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
      DVK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T61, T62, T63, T64, T65, T66, T67, T68, T69, T6a, T6b, T6c, T6d, T6e, T6f;
 	       V T6g, T6h, T6i, T6j, T6k, T6l, T6m, T6n, T6o, T6p, T6q, T6r, T6s, T6t, T6u;
 	       V T6v, T6w, T3g, T3f, T6x, T6y, T6z, T6A, T6B, T6C, T6D, T6E, T4p, T49, T4l;
@@ -727,7 +727,7 @@ static void n2sv_32(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(128, is), MAKE_VOLATILE_STRIDE(128, os)) {
 	       V T7, T4r, T4Z, T18, T1z, T3t, T3T, T2T, Te, T1f, T50, T4s, T2W, T3u, T1G;
 	       V T3U, Tm, T1n, T1O, T2Z, T3y, T3X, T4w, T53, Tt, T1u, T1V, T2Y, T3B, T3W;
 	       V T4z, T52, T2t, T3L, T3O, T2K, TR, TY, T5F, T5G, T5H, T5I, T4R, T5j, T2E;
diff --git a/dft/simd/common/n2sv_4.c b/dft/simd/common/n2sv_4.c
index 2d7a57b..6a586ae 100644
--- a/dft/simd/common/n2sv_4.c
+++ b/dft/simd/common/n2sv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:50 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:47 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void n2sv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T1, T2, T7, T8, T4, T5, Tc, Td;
 	       T1 = LD(&(ri[0]), ivs, &(ri[0]));
 	       T2 = LD(&(ri[WS(is, 2)]), ivs, &(ri[0]));
@@ -109,7 +109,7 @@ static void n2sv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       V T3, Tb, T9, Tf, T6, Ta, Te, Tg;
 	       {
 		    V T1, T2, T7, T8;
diff --git a/dft/simd/common/n2sv_64.c b/dft/simd/common/n2sv_64.c
index c790f7c..d6492ca 100644
--- a/dft/simd/common/n2sv_64.c
+++ b/dft/simd/common/n2sv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:57 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -53,7 +53,7 @@ static void n2sv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
      DVK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       V TeJ, TeK, TeP, TeQ, TfH, TfI, TfJ, TfK, Tgj, Tgk, Tgv, Tgw, T9a, T99, T9e;
 	       V T9b;
 	       {
@@ -1599,7 +1599,7 @@ static void n2sv_64(const R *ri, const R *ii, R *ro, R *io, stride is, stride os
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(256, is), MAKE_VOLATILE_STRIDE(256, os)) {
 	       V T37, T7B, T8F, T5Z, Tf, Td9, TbB, TcB, T62, T7C, T2i, TdH, Tah, Tcb, T3e;
 	       V T8G, Tu, TdI, Tak, TbD, Tan, TbC, T2x, Tda, T3m, T65, T7G, T8J, T7J, T8I;
 	       V T3t, T64, TK, Tdd, Tas, Tce, Tav, Tcf, T2N, Tdc, T3G, T6G, T7O, T9k, T7R;
diff --git a/dft/simd/common/n2sv_8.c b/dft/simd/common/n2sv_8.c
index f387ff7..1f74f76 100644
--- a/dft/simd/common/n2sv_8.c
+++ b/dft/simd/common/n2sv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:00:50 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:47 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void n2sv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V TF, TJ, TD, TR, TS, TT, TU, TV, TW, TE, TX, TY, TK, TI, TZ;
 	       V T10, T11, T12;
 	       {
@@ -173,7 +173,7 @@ static void n2sv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os,
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - (2 * VL), ri = ri + ((2 * VL) * ivs), ii = ii + ((2 * VL) * ivs), ro = ro + ((2 * VL) * ovs), io = io + ((2 * VL) * ovs), MAKE_VOLATILE_STRIDE(32, is), MAKE_VOLATILE_STRIDE(32, os)) {
 	       V T3, Tn, Ti, TC, T6, TB, Tl, To, Td, TN, Tz, TH, Ta, TM, Tu;
 	       V TG;
 	       {
diff --git a/dft/simd/common/q1bv_2.c b/dft/simd/common/q1bv_2.c
index 8e0d60b..5ae97b3 100644
--- a/dft/simd/common/q1bv_2.c
+++ b/dft/simd/common/q1bv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:27 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:33 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void q1bv_2(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(4, rs), MAKE_VOLATILE_STRIDE(4, vs)) {
 	       V T1, T2, T4, T5, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -84,7 +84,7 @@ static void q1bv_2(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(4, rs), MAKE_VOLATILE_STRIDE(4, vs)) {
 	       V T1, T2, T3, T4, T5, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/q1bv_4.c b/dft/simd/common/q1bv_4.c
index 5aa114d..b85f451 100644
--- a/dft/simd/common/q1bv_4.c
+++ b/dft/simd/common/q1bv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:33 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void q1bv_4(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, vs)) {
 	       V Tb, Tm, Tx, TI;
 	       {
 		    V Tc, T9, T3, TG, TA, TH, TD, Ta, T6, Td, Tn, To, Tq, Tr, Tf;
@@ -149,7 +149,7 @@ static void q1bv_4(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, vs)) {
 	       V T3, T9, TA, TG, TD, TH, T6, Ta, Te, Tk, Tp, Tv, Ts, Tw, Th;
 	       V Tl;
 	       {
diff --git a/dft/simd/common/q1bv_5.c b/dft/simd/common/q1bv_5.c
index 6c410b4..aea04b7 100644
--- a/dft/simd/common/q1bv_5.c
+++ b/dft/simd/common/q1bv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:33 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void q1bv_5(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(10, rs), MAKE_VOLATILE_STRIDE(10, vs)) {
 	       V Te, T1w, Ty, TS, TW, Tb, T1t, Tv, T1g, T1c, TP, TV, T1f, T19, TY;
 	       V TX;
 	       {
@@ -245,7 +245,7 @@ static void q1bv_5(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(10, rs), MAKE_VOLATILE_STRIDE(10, vs)) {
 	       V Tb, T7, Th, Ta, Tc, Td, T1t, T1p, T1z, T1s, T1u, T1v, Tv, Tr, TB;
 	       V Tu, Tw, Tx, TP, TL, TV, TO, TQ, TR, T19, T15, T1f, T18, T1a, T1b;
 	       {
diff --git a/dft/simd/common/q1bv_8.c b/dft/simd/common/q1bv_8.c
index f8fd962..282c872 100644
--- a/dft/simd/common/q1bv_8.c
+++ b/dft/simd/common/q1bv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:33 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void q1bv_8(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, vs)) {
 	       V T42, T43, T1U, T1V, T2Y, T2Z, TT, TS, T45, T44;
 	       {
 		    V T3, Te, T1E, T1P, Tv, Tp, T26, T20, T2b, T2m, T3M, T2x, T2D, T3X, TA;
@@ -504,7 +504,7 @@ static void q1bv_8(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, vs)) {
 	       V Ta, Tv, Te, Tp, T1L, T26, T1P, T20, T2i, T2D, T2m, T2x, T3T, T4e, T3X;
 	       V T48, TH, T12, TL, TW, T1e, T1z, T1i, T1t, T2P, T3a, T2T, T34, T3m, T3H;
 	       V T3q, T3B, T7, Tw, Tf, Ts, T1I, T27, T1Q, T23, T2f, T2E, T2n, T2A, T3Q;
diff --git a/dft/simd/common/q1fv_2.c b/dft/simd/common/q1fv_2.c
index 53b826c..07ef1eb 100644
--- a/dft/simd/common/q1fv_2.c
+++ b/dft/simd/common/q1fv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:25 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:30 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void q1fv_2(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(4, rs), MAKE_VOLATILE_STRIDE(4, vs)) {
 	       V T1, T2, T4, T5, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -84,7 +84,7 @@ static void q1fv_2(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(4, rs), MAKE_VOLATILE_STRIDE(4, vs)) {
 	       V T1, T2, T3, T4, T5, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/q1fv_4.c b/dft/simd/common/q1fv_4.c
index 90c6fcd..545eacf 100644
--- a/dft/simd/common/q1fv_4.c
+++ b/dft/simd/common/q1fv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:25 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:31 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void q1fv_4(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, vs)) {
 	       V Tb, Tm, Tx, TI;
 	       {
 		    V Tc, T9, T3, TG, TA, TH, TD, Ta, T6, Td, Tn, To, Tq, Tr, Tf;
@@ -149,7 +149,7 @@ static void q1fv_4(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, vs)) {
 	       V T3, T9, TA, TG, TD, TH, T6, Ta, Te, Tk, Tp, Tv, Ts, Tw, Th;
 	       V Tl;
 	       {
diff --git a/dft/simd/common/q1fv_5.c b/dft/simd/common/q1fv_5.c
index 6ff5a67..1c190eb 100644
--- a/dft/simd/common/q1fv_5.c
+++ b/dft/simd/common/q1fv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:25 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:31 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void q1fv_5(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(10, rs), MAKE_VOLATILE_STRIDE(10, vs)) {
 	       V Te, T1w, Ty, TS, TW, Tb, T1t, Tv, T1g, T1c, TP, TV, T1f, T19, TY;
 	       V TX;
 	       {
@@ -245,7 +245,7 @@ static void q1fv_5(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(10, rs), MAKE_VOLATILE_STRIDE(10, vs)) {
 	       V T8, T7, Th, Te, T9, Ta, T1q, T1p, T1z, T1w, T1r, T1s, Ts, Tr, TB;
 	       V Ty, Tt, Tu, TM, TL, TV, TS, TN, TO, T16, T15, T1f, T1c, T17, T18;
 	       {
diff --git a/dft/simd/common/q1fv_8.c b/dft/simd/common/q1fv_8.c
index c55f19f..1b211cc 100644
--- a/dft/simd/common/q1fv_8.c
+++ b/dft/simd/common/q1fv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:26 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:31 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void q1fv_8(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, vs)) {
 	       V T42, T43, T1U, T1V, T2Y, T2Z, TT, TS;
 	       {
 		    V T3, Te, T1E, T1P, Tu, Tp, T25, T20, T2b, T2m, T3M, T2x, T2C, T3X, TA;
@@ -501,7 +501,7 @@ static void q1fv_8(R *ri, R *ii, const R *W, stride rs, stride vs, INT mb, INT m
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(vs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, vs)) {
 	       V T3, Tu, Tf, Tp, T1E, T25, T1Q, T20, T2b, T2C, T2n, T2x, T3M, T4d, T3Y;
 	       V T48, TA, T11, TM, TW, T17, T1y, T1j, T1t, T2I, T39, T2U, T34, T3f, T3G;
 	       V T3r, T3B, Ta, Tv, Tc, Ts, T1L, T26, T1N, T23, T2i, T2D, T2k, T2A, T3T;
diff --git a/dft/simd/common/t1buv_10.c b/dft/simd/common/t1buv_10.c
index 02079ec..9accf45 100644
--- a/dft/simd/common/t1buv_10.c
+++ b/dft/simd/common/t1buv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1buv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
@@ -172,7 +172,7 @@ static void t1buv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Tu, TH, Tg, Tl, Tp, TD, TE, TJ, T5, Ta, To, TA, TB, TI, Tr;
 	       V Tt, Ts;
 	       Tr = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t1buv_2.c b/dft/simd/common/t1buv_2.c
index 10802ba..47ffed3 100644
--- a/dft/simd/common/t1buv_2.c
+++ b/dft/simd/common/t1buv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:56 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1buv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T2, T3;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -79,7 +79,7 @@ static void t1buv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1buv_3.c b/dft/simd/common/t1buv_3.c
index 0ad7628..bbb3880 100644
--- a/dft/simd/common/t1buv_3.c
+++ b/dft/simd/common/t1buv_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:56 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1buv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T1, T2, T4;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -93,7 +93,7 @@ static void t1buv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T6, T2, T4, T7, T1, T3, T5, T8;
 	       T6 = LD(&(x[0]), ms, &(x[0]));
 	       T1 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1buv_4.c b/dft/simd/common/t1buv_4.c
index d866954..9e5ccea 100644
--- a/dft/simd/common/t1buv_4.c
+++ b/dft/simd/common/t1buv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1buv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T7, T2, T5, T8, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -94,7 +94,7 @@ static void t1buv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T3, T6, T7, T2, T5;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1buv_5.c b/dft/simd/common/t1buv_5.c
index c31164f..7a8a92c 100644
--- a/dft/simd/common/t1buv_5.c
+++ b/dft/simd/common/t1buv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1buv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T2, T9, T4, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -119,7 +119,7 @@ static void t1buv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V Tf, T5, Ta, Tc, Td, Tg;
 	       Tf = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1buv_6.c b/dft/simd/common/t1buv_6.c
index 766da77..b4c7629 100644
--- a/dft/simd/common/t1buv_6.c
+++ b/dft/simd/common/t1buv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1buv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V T1, T2, Ta, Tc, T5, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -120,7 +120,7 @@ static void t1buv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V Tf, Ti, Ta, Tk, T5, Tj, Tc, Te, Td;
 	       Tc = LD(&(x[0]), ms, &(x[0]));
 	       Td = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1buv_7.c b/dft/simd/common/t1buv_7.c
index 5856caa..56f03a7 100644
--- a/dft/simd/common/t1buv_7.c
+++ b/dft/simd/common/t1buv_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void t1buv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V T1, T2, T4, Te, Tc, T9, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -144,7 +144,7 @@ static void t1buv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V Th, Tf, Ti, T5, Tk, Ta, Tj, To, Tp;
 	       Th = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1buv_8.c b/dft/simd/common/t1buv_8.c
index d90c0e6..290a5b1 100644
--- a/dft/simd/common/t1buv_8.c
+++ b/dft/simd/common/t1buv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t1buv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, Th, Tj, T5, T7, Ta, Tc;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
@@ -135,7 +135,7 @@ static void t1buv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V Tl, Tq, Tg, Tr, T5, Tt, Ta, Tu, Ti, Tk, Tj;
 	       Ti = LD(&(x[0]), ms, &(x[0]));
 	       Tj = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
diff --git a/dft/simd/common/t1buv_9.c b/dft/simd/common/t1buv_9.c
index 6b67981..35b736d 100644
--- a/dft/simd/common/t1buv_9.c
+++ b/dft/simd/common/t1buv_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -59,7 +59,7 @@ static void t1buv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T3, T5, T9, Tn, Tb, Td, Th, Tj, Tx, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
@@ -198,7 +198,7 @@ static void t1buv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T6, Tu, Tg, Tf, TD, Tq, Tp, TE;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1bv_10.c b/dft/simd/common/t1bv_10.c
index 65a46a1..90df617 100644
--- a/dft/simd/common/t1bv_10.c
+++ b/dft/simd/common/t1bv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:00 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
@@ -172,7 +172,7 @@ static void t1bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Tu, TH, Tg, Tl, Tp, TD, TE, TJ, T5, Ta, To, TA, TB, TI, Tr;
 	       V Tt, Ts;
 	       Tr = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t1bv_12.c b/dft/simd/common/t1bv_12.c
index 1777f99..43f7d93 100644
--- a/dft/simd/common/t1bv_12.c
+++ b/dft/simd/common/t1bv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:00 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1bv_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(12, rs)) {
 	       V TI, Ti, TA, T7, Tm, TE, Tw, Tk, Tf, TB, TU, TM;
 	       {
 		    V T9, TK, Tj, TL, Te;
@@ -183,7 +183,7 @@ static void t1bv_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(12, rs)) {
 	       V T1, Tt, T6, T7, TB, Tq, TC, TD, T9, Tu, Te, Tf, Tx, Tl, Ty;
 	       V Tz;
 	       {
diff --git a/dft/simd/common/t1bv_15.c b/dft/simd/common/t1bv_15.c
index c7d536d..4aa3c69 100644
--- a/dft/simd/common/t1bv_15.c
+++ b/dft/simd/common/t1bv_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:00 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void t1bv_15(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(15, rs)) {
 	       V Tq, Ty, Th, TV, TK, Ts, T1f, T7, Tu, TA, TC, Tj, Tk, T1g, Tf;
 	       {
 		    V T1, T4, T2, T9, Te;
@@ -245,7 +245,7 @@ static void t1bv_15(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(15, rs)) {
 	       V Ts, TV, T1f, TZ, T10, Tb, Tm, Tt, T1j, T1k, T1l, TI, TM, TR, Tz;
 	       V TD, TQ, T1g, T1h, T1i;
 	       {
diff --git a/dft/simd/common/t1bv_16.c b/dft/simd/common/t1bv_16.c
index ce9507f..a0338ef 100644
--- a/dft/simd/common/t1bv_16.c
+++ b/dft/simd/common/t1bv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:00 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void t1bv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TO, Ta, TJ, TP, T14, Tq, T1i, T10, T1b, T1l, T13, T1c, TR, Tl, T15;
 	       V Tv;
 	       {
@@ -235,7 +235,7 @@ static void t1bv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TJ, T1b, TD, T1c, T17, T18, Ty, TK, T10, T11, T12, Tb, TM, T13, T14;
 	       V T15, Tm, TN, TG, TI, TH;
 	       TG = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t1bv_2.c b/dft/simd/common/t1bv_2.c
index 0110313..0a7a4ea 100644
--- a/dft/simd/common/t1bv_2.c
+++ b/dft/simd/common/t1bv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1bv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T2, T3;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -79,7 +79,7 @@ static void t1bv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1bv_20.c b/dft/simd/common/t1bv_20.c
index a3addd3..c5989c9 100644
--- a/dft/simd/common/t1bv_20.c
+++ b/dft/simd/common/t1bv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:05 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1bv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, TX, T1m, T1K, T1y, Tk, Tf, T14, TQ, TZ, T1O, T1w, T1L, T1p, T1M;
 	       V T1s, TF, TY, T1x, Tp;
 	       {
@@ -288,7 +288,7 @@ static void t1bv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, T10, T1B, T1R, TF, T14, T15, TQ, Tf, Tq, Tr, T1N, T1O, T1P, T1t;
 	       V T1w, T1D, TT, TU, T11, T1K, T1L, T1M, T1m, T1p, T1C, T1i, T1j;
 	       {
diff --git a/dft/simd/common/t1bv_25.c b/dft/simd/common/t1bv_25.c
index 62381dc..cb22321 100644
--- a/dft/simd/common/t1bv_25.c
+++ b/dft/simd/common/t1bv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:06 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -107,7 +107,7 @@ static void t1bv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T25, T1B, T2y, T1K, T2s, T23, T1S, T26, T20, T1X;
 	       {
 		    V T1O, T2X, Te, T3L, Td, T3Q, T3j, T3b, T2R, T2M, T2f, T27, T1y, T1H, T3M;
@@ -574,7 +574,7 @@ static void t1bv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T1A, T1z, T1R, T1S, T1B, T1C, T1Q, T2L, T1l, T2v, T1i, T3e, T2u, Tb, T2i;
 	       V Tj, T3b, T2h, Tv, T2k, TD, T3a, T2l, T11, T2s, TY, T3d, T2r;
 	       {
diff --git a/dft/simd/common/t1bv_3.c b/dft/simd/common/t1bv_3.c
index 463d95f..9359333 100644
--- a/dft/simd/common/t1bv_3.c
+++ b/dft/simd/common/t1bv_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1bv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T1, T2, T4;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -93,7 +93,7 @@ static void t1bv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T6, T2, T4, T7, T1, T3, T5, T8;
 	       T6 = LD(&(x[0]), ms, &(x[0]));
 	       T1 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1bv_32.c b/dft/simd/common/t1bv_32.c
index 8e97da3..e81a8c5 100644
--- a/dft/simd/common/t1bv_32.c
+++ b/dft/simd/common/t1bv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -47,7 +47,7 @@ static void t1bv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T26, T25, T2a, T2i, T24, T2c, T2g, T2k, T2h, T27;
 	       {
 		    V T4, T1z, T2o, T32, T2r, T3f, Tf, T1A, T34, T2O, T1D, TC, T33, T2L, T1C;
@@ -450,7 +450,7 @@ static void t1bv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T4, T1D, T2P, T3h, Tf, T1y, T2K, T3i, TC, T1w, T2G, T3e, Tr, T1v, T2D;
 	       V T3d, T1k, T20, T2y, T3a, T1r, T21, T2v, T39, TV, T1X, T2r, T37, T12, T1Y;
 	       V T2o, T36;
diff --git a/dft/simd/common/t1bv_4.c b/dft/simd/common/t1bv_4.c
index b73f2f1..c909f33 100644
--- a/dft/simd/common/t1bv_4.c
+++ b/dft/simd/common/t1bv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1bv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T7, T2, T5, T8, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -94,7 +94,7 @@ static void t1bv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T3, T6, T7, T2, T5;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1bv_5.c b/dft/simd/common/t1bv_5.c
index 8a0a13f..36318a1 100644
--- a/dft/simd/common/t1bv_5.c
+++ b/dft/simd/common/t1bv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1bv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T2, T9, T4, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -119,7 +119,7 @@ static void t1bv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V Tf, T5, Ta, Tc, Td, Tg;
 	       Tf = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1bv_6.c b/dft/simd/common/t1bv_6.c
index 58d5af6..4e63e9f 100644
--- a/dft/simd/common/t1bv_6.c
+++ b/dft/simd/common/t1bv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1bv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V T1, T2, Ta, Tc, T5, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -120,7 +120,7 @@ static void t1bv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V Tf, Ti, Ta, Tk, T5, Tj, Tc, Te, Td;
 	       Tc = LD(&(x[0]), ms, &(x[0]));
 	       Td = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1bv_64.c b/dft/simd/common/t1bv_64.c
index 411e4e2..2e2a7dc 100644
--- a/dft/simd/common/t1bv_64.c
+++ b/dft/simd/common/t1bv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:05 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -55,7 +55,7 @@ static void t1bv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T6L, T6M, T6O, T6P, T75, T6V, T5A, T6A, T72, T6K, T6t, T6D, T6w, T6B, T6h;
 	       V T6E;
 	       {
@@ -945,7 +945,7 @@ static void t1bv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V Tg, T4B, T6v, T7G, T3r, T4w, T5q, T7F, T5Y, T62, T28, T4d, T2g, T4a, T7g;
 	       V T7Y, T6f, T6j, T2Z, T4k, T37, T4h, T7n, T81, T7w, T7x, T7y, T5M, T6q, T1k;
 	       V T4s, T1r, T4t, T7t, T7u, T7v, T5F, T6p, TV, T4p, T12, T4q, T7A, T7B, TD;
diff --git a/dft/simd/common/t1bv_7.c b/dft/simd/common/t1bv_7.c
index bb548ff..666f755 100644
--- a/dft/simd/common/t1bv_7.c
+++ b/dft/simd/common/t1bv_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void t1bv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V T1, T2, T4, Te, Tc, T9, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -144,7 +144,7 @@ static void t1bv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V Th, Tf, Ti, T5, Tk, Ta, Tj, To, Tp;
 	       Th = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1bv_8.c b/dft/simd/common/t1bv_8.c
index 839a7a1..467742d 100644
--- a/dft/simd/common/t1bv_8.c
+++ b/dft/simd/common/t1bv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t1bv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, Th, Tj, T5, T7, Ta, Tc;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
@@ -135,7 +135,7 @@ static void t1bv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V Tl, Tq, Tg, Tr, T5, Tt, Ta, Tu, Ti, Tk, Tj;
 	       Ti = LD(&(x[0]), ms, &(x[0]));
 	       Tj = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
diff --git a/dft/simd/common/t1bv_9.c b/dft/simd/common/t1bv_9.c
index fc7aeac..dcad5c1 100644
--- a/dft/simd/common/t1bv_9.c
+++ b/dft/simd/common/t1bv_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -59,7 +59,7 @@ static void t1bv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T3, T5, T9, Tn, Tb, Td, Th, Tj, Tx, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
@@ -198,7 +198,7 @@ static void t1bv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T6, Tu, Tg, Tf, TD, Tq, Tp, TE;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1fuv_10.c b/dft/simd/common/t1fuv_10.c
index 0392df5..193c7ad 100644
--- a/dft/simd/common/t1fuv_10.c
+++ b/dft/simd/common/t1fuv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1fuv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
@@ -172,7 +172,7 @@ static void t1fuv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Tr, TH, Tg, Tl, Tm, TA, TB, TJ, T5, Ta, Tb, TD, TE, TI, To;
 	       V Tq, Tp;
 	       To = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t1fuv_2.c b/dft/simd/common/t1fuv_2.c
index fca9e19..cb5d478 100644
--- a/dft/simd/common/t1fuv_2.c
+++ b/dft/simd/common/t1fuv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1fuv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T2, T3;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -79,7 +79,7 @@ static void t1fuv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fuv_3.c b/dft/simd/common/t1fuv_3.c
index f8d91c6..a3994bc 100644
--- a/dft/simd/common/t1fuv_3.c
+++ b/dft/simd/common/t1fuv_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:03 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1fuv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T1, T2, T4;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -93,7 +93,7 @@ static void t1fuv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T1, T3, T5, T6, T2, T4, T7, T8;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fuv_4.c b/dft/simd/common/t1fuv_4.c
index fed6261..ba6aba3 100644
--- a/dft/simd/common/t1fuv_4.c
+++ b/dft/simd/common/t1fuv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1fuv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T7, T2, T5, T8, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -94,7 +94,7 @@ static void t1fuv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T3, T6, T7, T2, T5;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fuv_5.c b/dft/simd/common/t1fuv_5.c
index e7ecb26..09a3e7b 100644
--- a/dft/simd/common/t1fuv_5.c
+++ b/dft/simd/common/t1fuv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1fuv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T2, T9, T4, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -119,7 +119,7 @@ static void t1fuv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V Tc, Tg, Th, T5, Ta, Td;
 	       Tc = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1fuv_6.c b/dft/simd/common/t1fuv_6.c
index 4942fa1..d6abc14 100644
--- a/dft/simd/common/t1fuv_6.c
+++ b/dft/simd/common/t1fuv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1fuv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V T1, T2, Ta, Tc, T5, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -120,7 +120,7 @@ static void t1fuv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V T4, Ti, Te, Tk, T9, Tj, T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fuv_7.c b/dft/simd/common/t1fuv_7.c
index b4c72b2..6bc4de7 100644
--- a/dft/simd/common/t1fuv_7.c
+++ b/dft/simd/common/t1fuv_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:37:59 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void t1fuv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V T1, T2, T4, Te, Tc, T9, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -144,7 +144,7 @@ static void t1fuv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V T1, Tg, Tj, T6, Ti, Tb, Tk, Tp, To;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1fuv_8.c b/dft/simd/common/t1fuv_8.c
index a5b7615..957d822 100644
--- a/dft/simd/common/t1fuv_8.c
+++ b/dft/simd/common/t1fuv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:00 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t1fuv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, Th, Tj, T5, T7, Ta, Tc;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
@@ -135,7 +135,7 @@ static void t1fuv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T4, Tq, Tm, Tr, T9, Tt, Te, Tu, T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
diff --git a/dft/simd/common/t1fuv_9.c b/dft/simd/common/t1fuv_9.c
index 75dc191..9d70b50 100644
--- a/dft/simd/common/t1fuv_9.c
+++ b/dft/simd/common/t1fuv_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:00 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -59,7 +59,7 @@ static void t1fuv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T3, T5, T9, Th, Tb, Td, Tj, Tl, TD, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
@@ -198,7 +198,7 @@ static void t1fuv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T6, TA, Tt, Tf, Ts, Tw, Tn, Tv;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1fv_10.c b/dft/simd/common/t1fv_10.c
index c1f98c0..6eec1f5 100644
--- a/dft/simd/common/t1fv_10.c
+++ b/dft/simd/common/t1fv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1fv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
@@ -172,7 +172,7 @@ static void t1fv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Tr, TH, Tg, Tl, Tm, TA, TB, TJ, T5, Ta, Tb, TD, TE, TI, To;
 	       V Tq, Tp;
 	       To = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t1fv_12.c b/dft/simd/common/t1fv_12.c
index cf87c22..fc9350a 100644
--- a/dft/simd/common/t1fv_12.c
+++ b/dft/simd/common/t1fv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:03 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1fv_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(12, rs)) {
 	       V Tq, Ti, T7, TQ, Tu, TA, TU, Tk, TR, Tf, TE, TM;
 	       {
 		    V T9, TC, Tj, TD, Te;
@@ -183,7 +183,7 @@ static void t1fv_12(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 22)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(12, rs)) {
 	       V T1, TH, T6, TA, Tq, TE, Tv, TL, T9, TI, Te, TB, Ti, TD, Tn;
 	       V TK;
 	       {
diff --git a/dft/simd/common/t1fv_15.c b/dft/simd/common/t1fv_15.c
index 33afc5e..131883f 100644
--- a/dft/simd/common/t1fv_15.c
+++ b/dft/simd/common/t1fv_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:10 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:04 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -48,7 +48,7 @@ static void t1fv_15(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(15, rs)) {
 	       V Tq, Ty, Th, T1b, T10, Ts, TP, T7, Tu, TA, TC, Tj, Tk, TQ, Tf;
 	       {
 		    V T1, T4, T2, T9, Te;
@@ -245,7 +245,7 @@ static void t1fv_15(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 28)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 28), MAKE_VOLATILE_STRIDE(15, rs)) {
 	       V T1e, T7, TP, T12, T15, Tf, Tn, To, T1b, T1c, T1f, TQ, TR, TS, Tw;
 	       V TE, TF, TT, TU, TV;
 	       {
diff --git a/dft/simd/common/t1fv_16.c b/dft/simd/common/t1fv_16.c
index c338487..9a4d26b 100644
--- a/dft/simd/common/t1fv_16.c
+++ b/dft/simd/common/t1fv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:05 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void t1fv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TO, Ta, TJ, TP, T14, Tq, T1i, T10, T1b, T1l, T13, T1c, TR, Tl, T15;
 	       V Tv;
 	       {
@@ -235,7 +235,7 @@ static void t1fv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TJ, T10, TD, T11, T1b, T1c, Ty, TK, T16, T17, T18, Tb, TN, T13, T14;
 	       V T15, Tm, TM, TG, TI, TH;
 	       TG = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t1fv_2.c b/dft/simd/common/t1fv_2.c
index 8c234d5..1eb2514 100644
--- a/dft/simd/common/t1fv_2.c
+++ b/dft/simd/common/t1fv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1fv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T2, T3;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -79,7 +79,7 @@ static void t1fv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fv_20.c b/dft/simd/common/t1fv_20.c
index 4eee715..9df6794 100644
--- a/dft/simd/common/t1fv_20.c
+++ b/dft/simd/common/t1fv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:30 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:32 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1fv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, Tx, T1m, T1K, T1y, Tk, Tf, T16, T10, TT, T1O, T1w, T1L, T1p, T1M;
 	       V T1s, TZ, TI, T1x, Tp;
 	       {
@@ -288,7 +288,7 @@ static void t1fv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, Tx, T1B, T1U, TZ, T16, T17, T10, Tf, Tq, Tr, T1N, T1O, T1S, T1t;
 	       V T1w, T1C, TI, TT, TU, T1K, T1L, T1R, T1m, T1p, T1D, Ts, TV;
 	       {
diff --git a/dft/simd/common/t1fv_25.c b/dft/simd/common/t1fv_25.c
index debe9f7..833a646 100644
--- a/dft/simd/common/t1fv_25.c
+++ b/dft/simd/common/t1fv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:31 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:33 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -107,7 +107,7 @@ static void t1fv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T25, T1B, T2y, T1K, T2s, T23, T1S, T26, T20, T1X;
 	       {
 		    V T1O, T2X, Te, T3L, Td, T3Q, T3j, T3b, T2R, T2M, T2f, T27, T1y, T1H, T3M;
@@ -574,7 +574,7 @@ static void t1fv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V Tc, Tb, Td, Te, T1C, T2t, T1E, T1x, T2m, T1u, T3c, T2n, Ty, T2i, Tv;
 	       V T38, T2j, TS, T2f, TP, T39, T2g, T1d, T2p, T1a, T3b, T2q;
 	       {
diff --git a/dft/simd/common/t1fv_3.c b/dft/simd/common/t1fv_3.c
index 33a74f7..9eb0e68 100644
--- a/dft/simd/common/t1fv_3.c
+++ b/dft/simd/common/t1fv_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1fv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T1, T2, T4;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -93,7 +93,7 @@ static void t1fv_3(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(3, rs)) {
 	       V T1, T3, T5, T6, T2, T4, T7, T8;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fv_32.c b/dft/simd/common/t1fv_32.c
index d0a4158..1a3fb08 100644
--- a/dft/simd/common/t1fv_32.c
+++ b/dft/simd/common/t1fv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:06 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -47,7 +47,7 @@ static void t1fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T26, T25, T1Z, T22, T1W, T2a, T2k, T2g;
 	       {
 		    V T4, T1z, T2o, T32, T2r, T3f, Tf, T1A, T34, T2L, T1D, TC, T33, T2O, T1C;
@@ -448,7 +448,7 @@ static void t1fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T4, T1A, T2o, T32, Tf, T1v, T2r, T3f, TC, T1C, T2L, T34, Tr, T1D, T2O;
 	       V T33, T1k, T20, T2F, T3b, T1r, T21, T2C, T3a, TV, T1X, T2y, T38, T12, T1Y;
 	       V T2v, T37;
diff --git a/dft/simd/common/t1fv_4.c b/dft/simd/common/t1fv_4.c
index 1e55fd6..a85469c 100644
--- a/dft/simd/common/t1fv_4.c
+++ b/dft/simd/common/t1fv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t1fv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T7, T2, T5, T8, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -94,7 +94,7 @@ static void t1fv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T3, T6, T7, T2, T5;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fv_5.c b/dft/simd/common/t1fv_5.c
index 990831f..0c6280e 100644
--- a/dft/simd/common/t1fv_5.c
+++ b/dft/simd/common/t1fv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t1fv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T2, T9, T4, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -119,7 +119,7 @@ static void t1fv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V Tc, Tg, Th, T5, Ta, Td;
 	       Tc = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1fv_6.c b/dft/simd/common/t1fv_6.c
index cc7177e..ed00355 100644
--- a/dft/simd/common/t1fv_6.c
+++ b/dft/simd/common/t1fv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -42,7 +42,7 @@ static void t1fv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V T1, T2, Ta, Tc, T5, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -120,7 +120,7 @@ static void t1fv_6(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 10)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(6, rs)) {
 	       V T4, Ti, Te, Tk, T9, Tj, T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t1fv_64.c b/dft/simd/common/t1fv_64.c
index e4270e0..3eb5182 100644
--- a/dft/simd/common/t1fv_64.c
+++ b/dft/simd/common/t1fv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:10 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -55,7 +55,7 @@ static void t1fv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T6L, T6M, T6O, T6P, T75, T6V, T5A, T6A, T72, T6K, T6t, T6D, T6w, T6B, T6h;
 	       V T6E;
 	       {
@@ -945,7 +945,7 @@ static void t1fv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V Tg, T4a, T6r, T7f, T3o, T4B, T5q, T7e, T5R, T62, T28, T4o, T2g, T4l, T7n;
 	       V T7Z, T68, T6j, T2C, T4s, T3a, T4v, T7u, T82, T7E, T7F, T7V, T5F, T6u, T1k;
 	       V T4e, T1r, T4d, T7B, T7C, T7W, T5M, T6v, TV, T4g, T12, T4h, T7h, T7i, TD;
diff --git a/dft/simd/common/t1fv_7.c b/dft/simd/common/t1fv_7.c
index 071e4f2..388e986 100644
--- a/dft/simd/common/t1fv_7.c
+++ b/dft/simd/common/t1fv_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:01 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -46,7 +46,7 @@ static void t1fv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V T1, T2, T4, Te, Tc, T9, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -144,7 +144,7 @@ static void t1fv_7(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 12)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 12), MAKE_VOLATILE_STRIDE(7, rs)) {
 	       V T1, Tg, Tj, T6, Ti, Tb, Tk, Tp, To;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1fv_8.c b/dft/simd/common/t1fv_8.c
index 1d750d4..e806b91 100644
--- a/dft/simd/common/t1fv_8.c
+++ b/dft/simd/common/t1fv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t1fv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, Th, Tj, T5, T7, Ta, Tc;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
@@ -135,7 +135,7 @@ static void t1fv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T4, Tq, Tm, Tr, T9, Tt, Te, Tu, T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
diff --git a/dft/simd/common/t1fv_9.c b/dft/simd/common/t1fv_9.c
index a4b2b9d..2f87f00 100644
--- a/dft/simd/common/t1fv_9.c
+++ b/dft/simd/common/t1fv_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:02 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -59,7 +59,7 @@ static void t1fv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T3, T5, T9, Th, Tb, Td, Tj, Tl, TD, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
@@ -198,7 +198,7 @@ static void t1fv_9(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 16)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 16), MAKE_VOLATILE_STRIDE(9, rs)) {
 	       V T1, T6, TA, Tt, Tf, Ts, Tw, Tn, Tv;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t1sv_16.c b/dft/simd/common/t1sv_16.c
index 76f36e2..59183a1 100644
--- a/dft/simd/common/t1sv_16.c
+++ b/dft/simd/common/t1sv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:24 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t1sv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 30); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 30); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 30), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T2S, T2O, T2B, T2j, T2A, T24, T3J, T3L, T2Q, T2I, T2R, T2L, T2C, T2y, T3D;
 	       V T3F;
 	       {
@@ -432,7 +432,7 @@ static void t1sv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 30); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 30); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 30), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T7, T37, T1t, T2U, Ti, T38, T1w, T2R, Tu, T2s, T1C, T2c, TF, T2t, T1H;
 	       V T2d, T1f, T1q, T2B, T2C, T2D, T2E, T1Z, T2j, T24, T2k, TS, T13, T2w, T2x;
 	       V T2y, T2z, T1O, T2g, T1T, T2h;
diff --git a/dft/simd/common/t1sv_2.c b/dft/simd/common/t1sv_2.c
index 42a26ea..51fce80 100644
--- a/dft/simd/common/t1sv_2.c
+++ b/dft/simd/common/t1sv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:23 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void t1sv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 2); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 2); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 2), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, Ta, T3, T6, T2, T5;
 	       T1 = LD(&(ri[0]), ms, &(ri[0]));
 	       Ta = LD(&(ii[0]), ms, &(ii[0]));
@@ -87,7 +87,7 @@ static void t1sv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 2); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 2); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 2), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T6, T7;
 	       T1 = LD(&(ri[0]), ms, &(ri[0]));
 	       T8 = LD(&(ii[0]), ms, &(ii[0]));
diff --git a/dft/simd/common/t1sv_32.c b/dft/simd/common/t1sv_32.c
index cab44c0..02acab9 100644
--- a/dft/simd/common/t1sv_32.c
+++ b/dft/simd/common/t1sv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:25 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void t1sv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 62); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 62); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 62), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T8Z, T90;
 	       {
 		    V T87, T8x, T3w, T8, T3B, T83, Tl, T8y, T6F, Tz, T3J, T5T, T6G, TM, T3Q;
@@ -909,7 +909,7 @@ static void t1sv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 62); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 62); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 62), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V Tj, T5F, T7C, T7Q, T35, T4T, T78, T7m, T1Q, T61, T5Y, T6J, T3K, T59, T41;
 	       V T56, T2B, T67, T6e, T6O, T4b, T5d, T4s, T5g, TG, T7l, T5I, T73, T3a, T4U;
 	       V T3f, T4V, T14, T5N, T5M, T6E, T3m, T4Y, T3r, T4Z, T1r, T5P, T5S, T6F, T3x;
diff --git a/dft/simd/common/t1sv_4.c b/dft/simd/common/t1sv_4.c
index 86d462f..fda67a5 100644
--- a/dft/simd/common/t1sv_4.c
+++ b/dft/simd/common/t1sv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:24 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void t1sv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, Tv, T3, T6, T5, Ta, Td, Tc, Tg, Tj, Tt, T4, Tf, Ti, Tn;
 	       V Tb, T2, T9;
 	       T1 = LD(&(ri[0]), ms, &(ri[0]));
@@ -124,7 +124,7 @@ static void t1sv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, Tp, T6, To, Tc, Tk, Th, Tl;
 	       T1 = LD(&(ri[0]), ms, &(ri[0]));
 	       Tp = LD(&(ii[0]), ms, &(ii[0]));
diff --git a/dft/simd/common/t1sv_8.c b/dft/simd/common/t1sv_8.c
index 2127410..e6ab438 100644
--- a/dft/simd/common/t1sv_8.c
+++ b/dft/simd/common/t1sv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:24 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void t1sv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 14); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 14); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 14), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T1, T1m, T1l, T7, TS, Tk, TQ, Te, To, Tr, Tu, T14, TF, Tx, T16;
 	       V TL, Tt, TW, Tp, Tq, Tw;
 	       {
@@ -211,7 +211,7 @@ static void t1sv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 14); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 14); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 14), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T7, T1e, TH, T19, TF, T13, TR, TU, Ti, T1f, TK, T16, Tu, T12, TM;
 	       V TP;
 	       {
diff --git a/dft/simd/common/t2bv_10.c b/dft/simd/common/t2bv_10.c
index af59931..6e2e412 100644
--- a/dft/simd/common/t2bv_10.c
+++ b/dft/simd/common/t2bv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:14 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t2bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
@@ -172,7 +172,7 @@ static void t2bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Tu, TH, Tg, Tl, Tp, TD, TE, TJ, T5, Ta, To, TA, TB, TI, Tr;
 	       V Tt, Ts;
 	       Tr = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t2bv_16.c b/dft/simd/common/t2bv_16.c
index 3e623e7..5df4397 100644
--- a/dft/simd/common/t2bv_16.c
+++ b/dft/simd/common/t2bv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void t2bv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TO, Ta, TJ, TP, T14, Tq, T1i, T10, T1b, T1l, T13, T1c, TR, Tl, T15;
 	       V Tv;
 	       {
@@ -235,7 +235,7 @@ static void t2bv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TJ, T1b, TD, T1c, T17, T18, Ty, TK, T10, T11, T12, Tb, TM, T13, T14;
 	       V T15, Tm, TN, TG, TI, TH;
 	       TG = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t2bv_2.c b/dft/simd/common/t2bv_2.c
index 93205c8..bca8c39 100644
--- a/dft/simd/common/t2bv_2.c
+++ b/dft/simd/common/t2bv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t2bv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T2, T3;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -79,7 +79,7 @@ static void t2bv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t2bv_20.c b/dft/simd/common/t2bv_20.c
index 8866420..35ca9ab 100644
--- a/dft/simd/common/t2bv_20.c
+++ b/dft/simd/common/t2bv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:14 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t2bv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, TX, T1m, T1K, T1y, Tk, Tf, T14, TQ, TZ, T1O, T1w, T1L, T1p, T1M;
 	       V T1s, TF, TY, T1x, Tp;
 	       {
@@ -288,7 +288,7 @@ static void t2bv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, T10, T1B, T1R, TF, T14, T15, TQ, Tf, Tq, Tr, T1N, T1O, T1P, T1t;
 	       V T1w, T1D, TT, TU, T11, T1K, T1L, T1M, T1m, T1p, T1C, T1i, T1j;
 	       {
diff --git a/dft/simd/common/t2bv_25.c b/dft/simd/common/t2bv_25.c
index b293256..1351b45 100644
--- a/dft/simd/common/t2bv_25.c
+++ b/dft/simd/common/t2bv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:15 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -107,7 +107,7 @@ static void t2bv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T25, T1B, T2y, T1K, T2s, T23, T1S, T26, T20, T1X;
 	       {
 		    V T1O, T2X, Te, T3L, Td, T3Q, T3j, T3b, T2R, T2M, T2f, T27, T1y, T1H, T3M;
@@ -574,7 +574,7 @@ static void t2bv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T1A, T1z, T1R, T1S, T1B, T1C, T1Q, T2L, T1l, T2v, T1i, T3e, T2u, Tb, T2i;
 	       V Tj, T3b, T2h, Tv, T2k, TD, T3a, T2l, T11, T2s, TY, T3d, T2r;
 	       {
diff --git a/dft/simd/common/t2bv_32.c b/dft/simd/common/t2bv_32.c
index aa40bd8..c16d74e 100644
--- a/dft/simd/common/t2bv_32.c
+++ b/dft/simd/common/t2bv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:10 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -47,7 +47,7 @@ static void t2bv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T26, T25, T2a, T2i, T24, T2c, T2g, T2k, T2h, T27;
 	       {
 		    V T4, T1z, T2o, T32, T2r, T3f, Tf, T1A, T34, T2O, T1D, TC, T33, T2L, T1C;
@@ -450,7 +450,7 @@ static void t2bv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T4, T1D, T2P, T3h, Tf, T1y, T2K, T3i, TC, T1w, T2G, T3e, Tr, T1v, T2D;
 	       V T3d, T1k, T20, T2y, T3a, T1r, T21, T2v, T39, TV, T1X, T2r, T37, T12, T1Y;
 	       V T2o, T36;
diff --git a/dft/simd/common/t2bv_4.c b/dft/simd/common/t2bv_4.c
index bc2f062..36e262b 100644
--- a/dft/simd/common/t2bv_4.c
+++ b/dft/simd/common/t2bv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t2bv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T7, T2, T5, T8, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -94,7 +94,7 @@ static void t2bv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T3, T6, T7, T2, T5;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t2bv_5.c b/dft/simd/common/t2bv_5.c
index 93cf42a..0698bc6 100644
--- a/dft/simd/common/t2bv_5.c
+++ b/dft/simd/common/t2bv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:14 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t2bv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T2, T9, T4, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -119,7 +119,7 @@ static void t2bv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V Tf, T5, Ta, Tc, Td, Tg;
 	       Tf = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t2bv_64.c b/dft/simd/common/t2bv_64.c
index 280966f..fbcddbf 100644
--- a/dft/simd/common/t2bv_64.c
+++ b/dft/simd/common/t2bv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:10 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:13 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -55,7 +55,7 @@ static void t2bv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T6L, T6M, T6O, T6P, T75, T6V, T5A, T6A, T72, T6K, T6t, T6D, T6w, T6B, T6h;
 	       V T6E;
 	       {
@@ -945,7 +945,7 @@ static void t2bv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V Tg, T4B, T6v, T7G, T3r, T4w, T5q, T7F, T5Y, T62, T28, T4d, T2g, T4a, T7g;
 	       V T7Y, T6f, T6j, T2Z, T4k, T37, T4h, T7n, T81, T7w, T7x, T7y, T5M, T6q, T1k;
 	       V T4s, T1r, T4t, T7t, T7u, T7v, T5F, T6p, TV, T4p, T12, T4q, T7A, T7B, TD;
diff --git a/dft/simd/common/t2bv_8.c b/dft/simd/common/t2bv_8.c
index 11a4459..7d36c02 100644
--- a/dft/simd/common/t2bv_8.c
+++ b/dft/simd/common/t2bv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:09 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t2bv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, Th, Tj, T5, T7, Ta, Tc;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
@@ -135,7 +135,7 @@ static void t2bv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V Tl, Tq, Tg, Tr, T5, Tt, Ta, Tu, Ti, Tk, Tj;
 	       Ti = LD(&(x[0]), ms, &(x[0]));
 	       Tj = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
diff --git a/dft/simd/common/t2fv_10.c b/dft/simd/common/t2fv_10.c
index 2ae3ac4..674f178 100644
--- a/dft/simd/common/t2fv_10.c
+++ b/dft/simd/common/t2fv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:41 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:41 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t2fv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Td, TA, T4, Ta, Tk, TE, Tp, TF, TB, T9, T1, T2, Tb;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 5)]), ms, &(x[WS(rs, 1)]));
@@ -172,7 +172,7 @@ static void t2fv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 18)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V Tr, TH, Tg, Tl, Tm, TA, TB, TJ, T5, Ta, Tb, TD, TE, TI, To;
 	       V Tq, Tp;
 	       To = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t2fv_16.c b/dft/simd/common/t2fv_16.c
index cdd21b6..829f4c1 100644
--- a/dft/simd/common/t2fv_16.c
+++ b/dft/simd/common/t2fv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:35 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void t2fv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TO, Ta, TJ, TP, T14, Tq, T1i, T10, T1b, T1l, T13, T1c, TR, Tl, T15;
 	       V Tv;
 	       {
@@ -235,7 +235,7 @@ static void t2fv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 30)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V TJ, T10, TD, T11, T1b, T1c, Ty, TK, T16, T17, T18, Tb, TN, T13, T14;
 	       V T15, Tm, TM, TG, TI, TH;
 	       TG = LD(&(x[0]), ms, &(x[0]));
diff --git a/dft/simd/common/t2fv_2.c b/dft/simd/common/t2fv_2.c
index 5cc98f8..091ba0e 100644
--- a/dft/simd/common/t2fv_2.c
+++ b/dft/simd/common/t2fv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:35 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t2fv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T2, T3;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -79,7 +79,7 @@ static void t2fv_2(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 2)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(2, rs)) {
 	       V T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t2fv_20.c b/dft/simd/common/t2fv_20.c
index 03e916e..2a10442 100644
--- a/dft/simd/common/t2fv_20.c
+++ b/dft/simd/common/t2fv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:41 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:41 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t2fv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, Tx, T1m, T1K, T1y, Tk, Tf, T16, T10, TT, T1O, T1w, T1L, T1p, T1M;
 	       V T1s, TZ, TI, T1x, Tp;
 	       {
@@ -288,7 +288,7 @@ static void t2fv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 38)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T4, Tx, T1B, T1U, TZ, T16, T17, T10, Tf, Tq, Tr, T1N, T1O, T1S, T1t;
 	       V T1w, T1C, TI, TT, TU, T1K, T1L, T1R, T1m, T1p, T1D, Ts, TV;
 	       {
diff --git a/dft/simd/common/t2fv_25.c b/dft/simd/common/t2fv_25.c
index dfde034..ee7100a 100644
--- a/dft/simd/common/t2fv_25.c
+++ b/dft/simd/common/t2fv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:42 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -107,7 +107,7 @@ static void t2fv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T25, T1B, T2y, T1K, T2s, T23, T1S, T26, T20, T1X;
 	       {
 		    V T1O, T2X, Te, T3L, Td, T3Q, T3j, T3b, T2R, T2M, T2f, T27, T1y, T1H, T3M;
@@ -574,7 +574,7 @@ static void t2fv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 48)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 48), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V Tc, Tb, Td, Te, T1C, T2t, T1E, T1x, T2m, T1u, T3c, T2n, Ty, T2i, Tv;
 	       V T38, T2j, TS, T2f, TP, T39, T2g, T1d, T2p, T1a, T3b, T2q;
 	       {
diff --git a/dft/simd/common/t2fv_32.c b/dft/simd/common/t2fv_32.c
index dcaeece..940b7de 100644
--- a/dft/simd/common/t2fv_32.c
+++ b/dft/simd/common/t2fv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:36 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -47,7 +47,7 @@ static void t2fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T26, T25, T1Z, T22, T1W, T2a, T2k, T2g;
 	       {
 		    V T4, T1z, T2o, T32, T2r, T3f, Tf, T1A, T34, T2L, T1D, TC, T33, T2O, T1C;
@@ -448,7 +448,7 @@ static void t2fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 62)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T4, T1A, T2o, T32, Tf, T1v, T2r, T3f, TC, T1C, T2L, T34, Tr, T1D, T2O;
 	       V T33, T1k, T20, T2F, T3b, T1r, T21, T2C, T3a, TV, T1X, T2y, T38, T12, T1Y;
 	       V T2v, T37;
diff --git a/dft/simd/common/t2fv_4.c b/dft/simd/common/t2fv_4.c
index 0dee1ca..8ec4e8b 100644
--- a/dft/simd/common/t2fv_4.c
+++ b/dft/simd/common/t2fv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:35 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t2fv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T7, T2, T5, T8, T3, T6;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
@@ -94,7 +94,7 @@ static void t2fv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T1, T8, T3, T6, T7, T2, T5;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T7 = LD(&(x[WS(rs, 3)]), ms, &(x[WS(rs, 1)]));
diff --git a/dft/simd/common/t2fv_5.c b/dft/simd/common/t2fv_5.c
index cdfcec5..97eefe0 100644
--- a/dft/simd/common/t2fv_5.c
+++ b/dft/simd/common/t2fv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:41 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t2fv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T2, T9, T4, T7;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 1)]), ms, &(x[WS(rs, 1)]));
@@ -119,7 +119,7 @@ static void t2fv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V Tc, Tg, Th, T5, Ta, Td;
 	       Tc = LD(&(x[0]), ms, &(x[0]));
 	       {
diff --git a/dft/simd/common/t2fv_64.c b/dft/simd/common/t2fv_64.c
index de157d4..63412d8 100644
--- a/dft/simd/common/t2fv_64.c
+++ b/dft/simd/common/t2fv_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:40 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -55,7 +55,7 @@ static void t2fv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T6L, T6M, T6O, T6P, T75, T6V, T5A, T6A, T72, T6K, T6t, T6D, T6w, T6B, T6h;
 	       V T6E;
 	       {
@@ -945,7 +945,7 @@ static void t2fv_64(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 126)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 126), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V Tg, T4a, T6r, T7f, T3o, T4B, T5q, T7e, T5R, T62, T28, T4o, T2g, T4l, T7n;
 	       V T7Z, T68, T6j, T2C, T4s, T3a, T4v, T7u, T82, T7E, T7F, T7V, T5F, T6u, T1k;
 	       V T4e, T1r, T4d, T7B, T7C, T7W, T5M, T6v, TV, T4g, T12, T4h, T7h, T7i, TD;
diff --git a/dft/simd/common/t2fv_8.c b/dft/simd/common/t2fv_8.c
index f183745..a873352 100644
--- a/dft/simd/common/t2fv_8.c
+++ b/dft/simd/common/t2fv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:35 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t2fv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, Th, Tj, T5, T7, Ta, Tc;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
@@ -135,7 +135,7 @@ static void t2fv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 14)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T4, Tq, Tm, Tr, T9, Tt, Te, Tu, T1, T3, T2;
 	       T1 = LD(&(x[0]), ms, &(x[0]));
 	       T2 = LD(&(x[WS(rs, 4)]), ms, &(x[0]));
diff --git a/dft/simd/common/t2sv_16.c b/dft/simd/common/t2sv_16.c
index 25616e9..cd3edb9 100644
--- a/dft/simd/common/t2sv_16.c
+++ b/dft/simd/common/t2sv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:26 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t2sv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T34, T30, T2N, T2v, T2M, T2g, T3V, T3X, T32, T2U, T33, T2X, T2O, T2K, T3P;
 	       V T3R;
 	       {
@@ -446,7 +446,7 @@ static void t2sv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T2, T5, Tg, Ti, Tk, To, TE, TC, T6, T3, T8, TW, TJ, Tt, TU;
 	       V Tc, Tx, TH, TN, TO, TP, TR, T1f, T1k, T1b, T1i, T1y, T1H, T1u, T1F;
 	       {
diff --git a/dft/simd/common/t2sv_32.c b/dft/simd/common/t2sv_32.c
index a75087b..e358810 100644
--- a/dft/simd/common/t2sv_32.c
+++ b/dft/simd/common/t2sv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:24 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:28 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -45,7 +45,7 @@ static void t2sv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T6H, T74, T6U, T6E, T9r, T9t, T78, T7c, T6W, T6S, T73, T6K, T7a, T72, T9x;
 	       V T9z;
 	       {
@@ -938,7 +938,7 @@ static void t2sv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 8); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 8), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T2, T5, T3, T6, T8, TM, TO, Td, T9, Te, Th, Tl, TD, TH, T1y;
 	       V T1H, T15, T1A, T11, T1F, T1n, T1p, T2q, T2I, T2u, T2K, T2V, T3b, T2Z, T3d;
 	       V Tu, Ty, T3l, T3n, T1t, T1v, T2f, T2h, T1a, T1e, T32, T34, T1W, T1Y, T2C;
diff --git a/dft/simd/common/t2sv_4.c b/dft/simd/common/t2sv_4.c
index 1623c6d..e92994f 100644
--- a/dft/simd/common/t2sv_4.c
+++ b/dft/simd/common/t2sv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:26 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -38,7 +38,7 @@ static void t2sv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 4), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T2, T6, T3, T5, T1, Tx, T8, Tc, Tf, Ta, T4, Th, Tj, Tl;
 	       T2 = LDW(&(W[0]));
 	       T6 = LDW(&(W[TWVL * 3]));
@@ -127,7 +127,7 @@ static void t2sv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 4); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 4); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 4), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T2, T4, T3, T5, T6, T8;
 	       T2 = LDW(&(W[0]));
 	       T4 = LDW(&(W[TWVL * 1]));
diff --git a/dft/simd/common/t2sv_8.c b/dft/simd/common/t2sv_8.c
index 9cae71b..2acc36c 100644
--- a/dft/simd/common/t2sv_8.c
+++ b/dft/simd/common/t2sv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:26 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -39,7 +39,7 @@ static void t2sv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T1m, T1l, T1k, T1u, T1n, T1o;
 	       {
 		    V T2, T3, Tl, Tn, T5, T6;
@@ -227,7 +227,7 @@ static void t2sv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * 6); m < me; m = m + (2 * VL), ri = ri + ((2 * VL) * ms), ii = ii + ((2 * VL) * ms), W = W + ((2 * VL) * 6), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T2, T5, T3, T6, T8, Tc, Tg, Ti, Tl, Tm, Tn, Tz, Tp, Tx;
 	       {
 		    V T4, Tb, T7, Ta;
diff --git a/dft/simd/common/t3bv_10.c b/dft/simd/common/t3bv_10.c
index e73b100..d7b68f6 100644
--- a/dft/simd/common/t3bv_10.c
+++ b/dft/simd/common/t3bv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:18 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:22 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t3bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V T1, T7, Th, Tx, Tr, Td, Tp, T6, Tv, Tc, Te, Ti, Tl, T2, T3;
 	       V T5;
 	       T2 = LDW(&(W[0]));
@@ -173,7 +173,7 @@ static void t3bv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V T1, T2, T3, Ti, T6, T7, TA, Tb, To;
 	       T1 = LDW(&(W[0]));
 	       T2 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3bv_16.c b/dft/simd/common/t3bv_16.c
index ea64e39..b229bb6 100644
--- a/dft/simd/common/t3bv_16.c
+++ b/dft/simd/common/t3bv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:18 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void t3bv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T13, Tg, TY, T14, T1A, T1q, T1f, T1x, T1r, T1i, Tt, T16, TB, T1j, T1k;
 	       V TH;
 	       {
@@ -245,7 +245,7 @@ static void t3bv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T1, T8, T9, Tl, Ti, TE, T4, Ta, TO, TV, Td, Tm, TA, TH, Ts;
 	       T1 = LDW(&(W[0]));
 	       T8 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3bv_20.c b/dft/simd/common/t3bv_20.c
index a39d8df..000fd1a 100644
--- a/dft/simd/common/t3bv_20.c
+++ b/dft/simd/common/t3bv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:18 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:23 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t3bv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T19, T1u, T1p, T1x, T1m, T1w, T1t, TI;
 	       {
 		    V T2, T8, T3, Td;
@@ -294,7 +294,7 @@ static void t3bv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T2, T8, T9, TA, T3, Tc, T4, TV, T14, Tl, Tq, Tx, TQ, Td, Te;
 	       V T1g, Ti, Tt, T11;
 	       T2 = LDW(&(W[0]));
diff --git a/dft/simd/common/t3bv_25.c b/dft/simd/common/t3bv_25.c
index faede63..faf2bfe 100644
--- a/dft/simd/common/t3bv_25.c
+++ b/dft/simd/common/t3bv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:23 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -107,7 +107,7 @@ static void t3bv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T2t, T1Z, T2W, T28, T2Q, T2r, T2g, T2u, T2o, T2l;
 	       {
 		    V T2, T5, T3, T9;
@@ -582,7 +582,7 @@ static void t3bv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T1, Td, T8, T9, TF, Te, Tu, TB, TC, T1s, T15, Tf, TY, T4, Ta;
 	       V Tx, T1T, Tg, T1N, T1v, T18, TG, T1o, T11;
 	       T1 = LDW(&(W[TWVL * 4]));
diff --git a/dft/simd/common/t3bv_32.c b/dft/simd/common/t3bv_32.c
index 251659c..a2c46e0 100644
--- a/dft/simd/common/t3bv_32.c
+++ b/dft/simd/common/t3bv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:18 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:19 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -47,7 +47,7 @@ static void t3bv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T2B, T2A, T2F, T2N, T2H, T2z, T2P, T2L, T2C, T2M;
 	       {
 		    V T2, T5, T3, T7;
@@ -459,7 +459,7 @@ static void t3bv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T2, T5, T3, T4, Tc, T1v, TH, Tz, Tn, T6, TS, Tf, TK, T7, T8;
 	       V Tv, T1I, T25, Tg, Tk, T1N, T1Q, TC, T16, T12, T1w, TL, TP, TT, T1m;
 	       V T1f;
diff --git a/dft/simd/common/t3bv_4.c b/dft/simd/common/t3bv_4.c
index 220dd50..a5aa561 100644
--- a/dft/simd/common/t3bv_4.c
+++ b/dft/simd/common/t3bv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:18 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t3bv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T2, T3, T1, Ta, T5, T8;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
@@ -99,7 +99,7 @@ static void t3bv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T2, T3, T4;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3bv_5.c b/dft/simd/common/t3bv_5.c
index aa1e0bb..9b3d0be 100644
--- a/dft/simd/common/t3bv_5.c
+++ b/dft/simd/common/t3bv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:18 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:22 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t3bv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T2, T5, T1, T3, Td, T7, Tb;
 	       T2 = LDW(&(W[0]));
 	       T5 = LDW(&(W[TWVL * 2]));
@@ -121,7 +121,7 @@ static void t3bv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T4, T5, T9;
 	       T1 = LDW(&(W[0]));
 	       T4 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3bv_8.c b/dft/simd/common/t3bv_8.c
index 5db1d7f..3134f60 100644
--- a/dft/simd/common/t3bv_8.c
+++ b/dft/simd/common/t3bv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:18 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t3bv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T2, T3, Tb, T1, T5, Tn, Tq, T8, Td, T4, Ta, Tp, Tg, Ti, T9;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
@@ -141,7 +141,7 @@ static void t3bv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ii;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T4, T5, Tp, T6, T7, Tj;
 	       T1 = LDW(&(W[0]));
 	       T4 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3fv_10.c b/dft/simd/common/t3fv_10.c
index e52b88f..32a685c 100644
--- a/dft/simd/common/t3fv_10.c
+++ b/dft/simd/common/t3fv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:55 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:55 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t3fv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V T1, T7, Th, Tx, Tr, Td, Tp, T6, Tv, Tc, Te, Ti, Tl, T2, T3;
 	       V T5;
 	       T2 = LDW(&(W[0]));
@@ -173,7 +173,7 @@ static void t3fv_10(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(10, rs)) {
 	       V T1, T2, T3, Ti, T6, T7, Tx, Tb, To;
 	       T1 = LDW(&(W[0]));
 	       T2 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3fv_16.c b/dft/simd/common/t3fv_16.c
index cefc55a..fb3747a 100644
--- a/dft/simd/common/t3fv_16.c
+++ b/dft/simd/common/t3fv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:49 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -43,7 +43,7 @@ static void t3fv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T13, Tg, TY, T14, T1A, T1q, T1f, T1x, T1r, T1i, Tt, T16, TB, T1j, T1k;
 	       V TH;
 	       {
@@ -245,7 +245,7 @@ static void t3fv_16(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T4, T5, T6, To, T1, Ty, T7, T8, TO, TV, Te, Tp, TB, TH, Ts;
 	       T4 = LDW(&(W[0]));
 	       T5 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3fv_20.c b/dft/simd/common/t3fv_20.c
index 2871854..d6e1a36 100644
--- a/dft/simd/common/t3fv_20.c
+++ b/dft/simd/common/t3fv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:55 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:55 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t3fv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T1k, T1w, T1r, T1z, T1o, T1y, T1v, T1h;
 	       {
 		    V T2, T8, T3, Td;
@@ -294,7 +294,7 @@ static void t3fv_20(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(20, rs)) {
 	       V T2, T8, T9, TA, T3, Tc, T4, TZ, T18, Tl, Tq, Tx, TU, Td, Te;
 	       V T15, Ti, Tt, TJ;
 	       T2 = LDW(&(W[0]));
diff --git a/dft/simd/common/t3fv_25.c b/dft/simd/common/t3fv_25.c
index 5923779..a7a0cb3 100644
--- a/dft/simd/common/t3fv_25.c
+++ b/dft/simd/common/t3fv_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:56 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:56 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -107,7 +107,7 @@ static void t3fv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T2t, T1Z, T2W, T28, T2Q, T2r, T2g, T2u, T2o, T2l;
 	       {
 		    V T2, T5, T3, T9;
@@ -582,7 +582,7 @@ static void t3fv_25(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(25, rs)) {
 	       V T1, T4, T2, T3, TA, Td, Tp, Tw, Tx, T1G, T1j, T5, T1c, T8, T9;
 	       V Ts, T1J, Tg, T1C, T1m, TX, TB, T1f, TU;
 	       T1 = LDW(&(W[0]));
diff --git a/dft/simd/common/t3fv_32.c b/dft/simd/common/t3fv_32.c
index 506aa35..884afe6 100644
--- a/dft/simd/common/t3fv_32.c
+++ b/dft/simd/common/t3fv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:50 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:50 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -47,7 +47,7 @@ static void t3fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T2B, T2A, T2u, T2x, T2r, T2F, T2L, T2P;
 	       {
 		    V T2, T5, T3, T7;
@@ -457,7 +457,7 @@ static void t3fv_32(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 8)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 8), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T2, T5, T3, T4, Tc, T1C, TP, Tz, Tn, T6, TS, Tf, TK, T7, T8;
 	       V Tv, T1w, T22, Tg, Tk, T1D, T1R, TC, T18, T12, T1t, TH, TL, TT, T1n;
 	       V T1c;
diff --git a/dft/simd/common/t3fv_4.c b/dft/simd/common/t3fv_4.c
index 8d9247b..16c8bdf 100644
--- a/dft/simd/common/t3fv_4.c
+++ b/dft/simd/common/t3fv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:49 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -40,7 +40,7 @@ static void t3fv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T2, T3, T1, Ta, T5, T8;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
@@ -99,7 +99,7 @@ static void t3fv_4(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(4, rs)) {
 	       V T2, T3, T4;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3fv_5.c b/dft/simd/common/t3fv_5.c
index 7bc130a..c380348 100644
--- a/dft/simd/common/t3fv_5.c
+++ b/dft/simd/common/t3fv_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:54 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -44,7 +44,7 @@ static void t3fv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T2, T5, T1, T3, Td, T7, Tb;
 	       T2 = LDW(&(W[0]));
 	       T5 = LDW(&(W[TWVL * 2]));
@@ -121,7 +121,7 @@ static void t3fv_5(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 4)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 4), MAKE_VOLATILE_STRIDE(5, rs)) {
 	       V T1, T4, T5, T9;
 	       T1 = LDW(&(W[0]));
 	       T4 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/common/t3fv_8.c b/dft/simd/common/t3fv_8.c
index 296542b..3f4e5ba 100644
--- a/dft/simd/common/t3fv_8.c
+++ b/dft/simd/common/t3fv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:01:49 EDT 2012 */
+/* Generated on Sun Nov 25 07:38:49 EST 2012 */
 
 #include "codelet-dft.h"
 
@@ -41,7 +41,7 @@ static void t3fv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T2, T3, Tb, T1, T5, Tn, Tq, T8, Td, T4, Ta, Tp, Tg, Ti, T9;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
@@ -141,7 +141,7 @@ static void t3fv_8(R *ri, R *ii, const R *W, stride rs, INT mb, INT me, INT ms)
 	  INT m;
 	  R *x;
 	  x = ri;
-	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + (mb * ((TWVL / VL) * 6)); m < me; m = m + VL, x = x + (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T2, T3, Ta, T4, Tb, Tc, Tq;
 	       T2 = LDW(&(W[0]));
 	       T3 = LDW(&(W[TWVL * 2]));
diff --git a/dft/simd/neon/Makefile.in b/dft/simd/neon/Makefile.in
index b61485e..03753bb 100644
--- a/dft/simd/neon/Makefile.in
+++ b/dft/simd/neon/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -164,6 +181,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libdft_neon_codelets_la_SOURCES)
 DIST_SOURCES = $(am__libdft_neon_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/dft/simd/sse2/Makefile.in b/dft/simd/sse2/Makefile.in
index f7a26d9..e6802b2 100644
--- a/dft/simd/sse2/Makefile.in
+++ b/dft/simd/sse2/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -164,6 +181,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libdft_sse2_codelets_la_SOURCES)
 DIST_SOURCES = $(am__libdft_sse2_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/doc/FAQ/Makefile.in b/doc/FAQ/Makefile.in
index b1dfe00..8365b94 100644
--- a/doc/FAQ/Makefile.in
+++ b/doc/FAQ/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -15,6 +15,23 @@
 
 @SET_MAKE@
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -55,6 +72,11 @@ CONFIG_CLEAN_FILES =
 CONFIG_CLEAN_VPATH_FILES =
 SOURCES =
 DIST_SOURCES =
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 ACLOCAL = @ACLOCAL@
 ALLOCA = @ALLOCA@
diff --git a/doc/FAQ/fftw-faq.ascii b/doc/FAQ/fftw-faq.ascii
index ab1bdff..c048c87 100644
--- a/doc/FAQ/fftw-faq.ascii
+++ b/doc/FAQ/fftw-faq.ascii
@@ -1,5 +1,5 @@
             FFTW FREQUENTLY ASKED QUESTIONS WITH ANSWERS
-                            28 Apr 2012
+                            25 Nov 2012
 			     Matteo Frigo
 			   Steven G. Johnson
  			    <fftw at fftw.org>
diff --git a/doc/FAQ/fftw-faq.html/index.html b/doc/FAQ/fftw-faq.html/index.html
index 65f7c17..0a0615e 100644
--- a/doc/FAQ/fftw-faq.html/index.html
+++ b/doc/FAQ/fftw-faq.html/index.html
@@ -103,7 +103,7 @@ Solaris.</a>
 </ul><hr>
 <address>
 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw at fftw.org">fftw at fftw.org</A>
-- 28 April 2012
+- 25 November 2012
 </address><br>
 Extracted from FFTW Frequently Asked Questions with Answers,
 Copyright © 2012 Matteo Frigo and Massachusetts Institute of Technology.
diff --git a/doc/FAQ/fftw-faq.html/section1.html b/doc/FAQ/fftw-faq.html/section1.html
index e3f5e7f..c34c865 100644
--- a/doc/FAQ/fftw-faq.html/section1.html
+++ b/doc/FAQ/fftw-faq.html/section1.html
@@ -78,7 +78,7 @@ Next: <a href="section2.html" rel=precedes>Installing FFTW</a>.<br>
 <a href="index.html" rev=subdocument>Return to contents</a>.<p>
 <address>
 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw at fftw.org">fftw at fftw.org</A>
-- 28 April 2012
+- 25 November 2012
 </address><br>
 Extracted from FFTW Frequently Asked Questions with Answers,
 Copyright © 2012 Matteo Frigo and Massachusetts Institute of Technology.
diff --git a/doc/FAQ/fftw-faq.html/section2.html b/doc/FAQ/fftw-faq.html/section2.html
index bd17e33..4b59c5d 100644
--- a/doc/FAQ/fftw-faq.html/section2.html
+++ b/doc/FAQ/fftw-faq.html/section2.html
@@ -278,7 +278,7 @@ Back: <a href="section1.html" rev=precedes>Introduction and General Information<
 <a href="index.html" rev=subdocument>Return to contents</a>.<p>
 <address>
 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw at fftw.org">fftw at fftw.org</A>
-- 28 April 2012
+- 25 November 2012
 </address><br>
 Extracted from FFTW Frequently Asked Questions with Answers,
 Copyright © 2012 Matteo Frigo and Massachusetts Institute of Technology.
diff --git a/doc/FAQ/fftw-faq.html/section3.html b/doc/FAQ/fftw-faq.html/section3.html
index 93c74e3..2a4bf77 100644
--- a/doc/FAQ/fftw-faq.html/section3.html
+++ b/doc/FAQ/fftw-faq.html/section3.html
@@ -327,7 +327,7 @@ Back: <a href="section2.html" rev=precedes>Installing FFTW</a>.<br>
 <a href="index.html" rev=subdocument>Return to contents</a>.<p>
 <address>
 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw at fftw.org">fftw at fftw.org</A>
-- 28 April 2012
+- 25 November 2012
 </address><br>
 Extracted from FFTW Frequently Asked Questions with Answers,
 Copyright © 2012 Matteo Frigo and Massachusetts Institute of Technology.
diff --git a/doc/FAQ/fftw-faq.html/section4.html b/doc/FAQ/fftw-faq.html/section4.html
index e3f9f72..a5299c2 100644
--- a/doc/FAQ/fftw-faq.html/section4.html
+++ b/doc/FAQ/fftw-faq.html/section4.html
@@ -57,7 +57,7 @@ Back: <a href="section3.html" rev=precedes>Using FFTW</a>.<br>
 <a href="index.html" rev=subdocument>Return to contents</a>.<p>
 <address>
 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw at fftw.org">fftw at fftw.org</A>
-- 28 April 2012
+- 25 November 2012
 </address><br>
 Extracted from FFTW Frequently Asked Questions with Answers,
 Copyright © 2012 Matteo Frigo and Massachusetts Institute of Technology.
diff --git a/doc/FAQ/fftw-faq.html/section5.html b/doc/FAQ/fftw-faq.html/section5.html
index 3a1f0e4..12cbcdc 100644
--- a/doc/FAQ/fftw-faq.html/section5.html
+++ b/doc/FAQ/fftw-faq.html/section5.html
@@ -141,7 +141,7 @@ Back: <a href="section4.html" rev=precedes>Internals of FFTW</a>.<br>
 <a href="index.html" rev=subdocument>Return to contents</a>.<p>
 <address>
 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw at fftw.org">fftw at fftw.org</A>
-- 28 April 2012
+- 25 November 2012
 </address><br>
 Extracted from FFTW Frequently Asked Questions with Answers,
 Copyright © 2012 Matteo Frigo and Massachusetts Institute of Technology.
diff --git a/doc/Makefile.in b/doc/Makefile.in
index 2591c72..31ec1e8 100644
--- a/doc/Makefile.in
+++ b/doc/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -15,6 +15,23 @@
 
 @SET_MAKE@
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -75,6 +92,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 am__installdirs = "$(DESTDIR)$(infodir)"
 am__vpath_adj_setup = srcdirstrip=`echo "$(srcdir)" | sed 's|.|.|g'`;
 am__vpath_adj = case $$p in \
@@ -425,9 +447,7 @@ uninstall-html-am:
 
 uninstall-info-am:
 	@$(PRE_UNINSTALL)
-	@if test -d '$(DESTDIR)$(infodir)' && \
-	    (install-info --version && \
-	     install-info --version 2>&1 | sed 1q | grep -i -v debian) >/dev/null 2>&1; then \
+	@if test -d '$(DESTDIR)$(infodir)' && $(am__can_run_installinfo); then \
 	  list='$(INFO_DEPS)'; \
 	  for file in $$list; do \
 	    relfile=`echo "$$file" | sed 's|^.*/||'`; \
@@ -666,13 +686,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
@@ -757,8 +774,11 @@ install-dvi: install-dvi-recursive
 
 install-dvi-am: $(DVIS)
 	@$(NORMAL_INSTALL)
-	test -z "$(dvidir)" || $(MKDIR_P) "$(DESTDIR)$(dvidir)"
 	@list='$(DVIS)'; test -n "$(dvidir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(dvidir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(dvidir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
@@ -773,8 +793,11 @@ install-html: install-html-recursive
 
 install-html-am: $(HTMLS)
 	@$(NORMAL_INSTALL)
-	test -z "$(htmldir)" || $(MKDIR_P) "$(DESTDIR)$(htmldir)"
 	@list='$(HTMLS)'; list2=; test -n "$(htmldir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(htmldir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(htmldir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p" || test -d "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  $(am__strip_dir) \
@@ -797,9 +820,12 @@ install-info: install-info-recursive
 
 install-info-am: $(INFO_DEPS)
 	@$(NORMAL_INSTALL)
-	test -z "$(infodir)" || $(MKDIR_P) "$(DESTDIR)$(infodir)"
 	@srcdirstrip=`echo "$(srcdir)" | sed 's|.|.|g'`; \
 	list='$(INFO_DEPS)'; test -n "$(infodir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(infodir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(infodir)" || exit 1; \
+	fi; \
 	for file in $$list; do \
 	  case $$file in \
 	    $(srcdir)/*) file=`echo "$$file" | sed "s|^$$srcdirstrip/||"`;; \
@@ -817,13 +843,7 @@ install-info-am: $(INFO_DEPS)
 	  echo " $(INSTALL_DATA) $$files '$(DESTDIR)$(infodir)'"; \
 	  $(INSTALL_DATA) $$files "$(DESTDIR)$(infodir)" || exit $$?; done
 	@$(POST_INSTALL)
-	@am__run_installinfo=yes; \
-	case $$AM_UPDATE_INFO_DIR in \
-	  n|no|NO) am__run_installinfo=no;; \
-	  *) (install-info --version) >/dev/null 2>&1 \
-	       || am__run_installinfo=no;; \
-	esac; \
-	if test $$am__run_installinfo = yes; then \
+	@if $(am__can_run_installinfo); then \
 	  list='$(INFO_DEPS)'; test -n "$(infodir)" || list=; \
 	  for file in $$list; do \
 	    relfile=`echo "$$file" | sed 's|^.*/||'`; \
@@ -837,8 +857,11 @@ install-pdf: install-pdf-recursive
 
 install-pdf-am: $(PDFS)
 	@$(NORMAL_INSTALL)
-	test -z "$(pdfdir)" || $(MKDIR_P) "$(DESTDIR)$(pdfdir)"
 	@list='$(PDFS)'; test -n "$(pdfdir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(pdfdir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(pdfdir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
@@ -850,8 +873,11 @@ install-ps: install-ps-recursive
 
 install-ps-am: $(PSS)
 	@$(NORMAL_INSTALL)
-	test -z "$(psdir)" || $(MKDIR_P) "$(DESTDIR)$(psdir)"
 	@list='$(PSS)'; test -n "$(psdir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(psdir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(psdir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
diff --git a/doc/fftw3.info b/doc/fftw3.info
index 9d8fa71..478e74e 100644
--- a/doc/fftw3.info
+++ b/doc/fftw3.info
@@ -1,6 +1,6 @@
 This is fftw3.info, produced by makeinfo version 4.13 from fftw3.texi.
 
-This manual is for FFTW (version 3.3.2, 28 April 2012).
+This manual is for FFTW (version 3.3.3, 25 November 2012).
 
    Copyright (C) 2003 Matteo Frigo.
 
@@ -27,139 +27,139 @@ END-INFO-DIR-ENTRY
 
 
 Indirect:
-fftw3.info-1: 1073
-fftw3.info-2: 297043
+fftw3.info-1: 1076
+fftw3.info-2: 297310
 
 Tag Table:
 (Indirect)
-Node: Top1073
-Node: Introduction1746
-Node: Tutorial8176
-Ref: Tutorial-Footnote-19420
-Node: Complex One-Dimensional DFTs9514
-Node: Complex Multi-Dimensional DFTs15270
-Ref: Complex Multi-Dimensional DFTs-Footnote-118702
-Node: One-Dimensional DFTs of Real Data18837
-Node: Multi-Dimensional DFTs of Real Data23282
-Node: More DFTs of Real Data26854
-Node: The Halfcomplex-format DFT30356
-Node: Real even/odd DFTs (cosine/sine transforms)32965
-Ref: Real even/odd DFTs (cosine/sine transforms)-Footnote-138575
-Ref: Real even/odd DFTs (cosine/sine transforms)-Footnote-238764
-Node: The Discrete Hartley Transform39697
-Ref: The Discrete Hartley Transform-Footnote-141882
-Node: Other Important Topics42131
-Node: SIMD alignment and fftw_malloc42424
-Node: Multi-dimensional Array Format44684
-Node: Row-major Format45305
-Node: Column-major Format46998
-Node: Fixed-size Arrays in C48082
-Node: Dynamic Arrays in C49518
-Node: Dynamic Arrays in C-The Wrong Way51156
-Node: Words of Wisdom-Saving Plans52904
-Node: Caveats in Using Wisdom55579
-Node: FFTW Reference57667
-Node: Data Types and Files58155
-Node: Complex numbers58587
-Node: Precision60328
-Node: Memory Allocation61890
-Node: Using Plans63461
-Node: Basic Interface67341
-Ref: Basic Interface-Footnote-168085
-Node: Complex DFTs68149
-Node: Planner Flags72116
-Node: Real-data DFTs77473
-Node: Real-data DFT Array Format82469
-Node: Real-to-Real Transforms84724
-Node: Real-to-Real Transform Kinds88694
-Node: Advanced Interface91162
-Node: Advanced Complex DFTs91902
-Node: Advanced Real-data DFTs96161
-Node: Advanced Real-to-real Transforms98488
-Node: Guru Interface99594
-Node: Interleaved and split arrays100517
-Node: Guru vector and transform sizes101560
-Node: Guru Complex DFTs104125
-Node: Guru Real-data DFTs106961
-Node: Guru Real-to-real Transforms109884
-Node: 64-bit Guru Interface111203
-Node: New-array Execute Functions113526
-Node: Wisdom117524
-Node: Wisdom Export117883
-Node: Wisdom Import119857
-Node: Forgetting Wisdom121879
-Node: Wisdom Utilities122251
-Node: What FFTW Really Computes123618
-Node: The 1d Discrete Fourier Transform (DFT)124443
-Node: The 1d Real-data DFT125802
-Node: 1d Real-even DFTs (DCTs)127456
-Node: 1d Real-odd DFTs (DSTs)130665
-Node: 1d Discrete Hartley Transforms (DHTs)133607
-Node: Multi-dimensional Transforms134283
-Node: Multi-threaded FFTW136886
-Node: Installation and Supported Hardware/Software138355
-Node: Usage of Multi-threaded FFTW140180
-Node: How Many Threads to Use?143488
-Node: Thread safety144512
-Node: Distributed-memory FFTW with MPI146680
-Node: FFTW MPI Installation149259
-Node: Linking and Initializing MPI FFTW151051
-Node: 2d MPI example152281
-Node: MPI Data Distribution156517
-Node: Basic and advanced distribution interfaces159395
-Node: Load balancing163830
-Node: Transposed distributions165516
-Node: One-dimensional distributions169288
-Node: Multi-dimensional MPI DFTs of Real Data171857
-Node: Other Multi-dimensional Real-data MPI Transforms176505
-Node: FFTW MPI Transposes178678
-Node: Basic distributed-transpose interface179518
-Node: Advanced distributed-transpose interface181702
-Node: An improved replacement for MPI_Alltoall182990
-Node: FFTW MPI Wisdom184966
-Ref: FFTW MPI Wisdom-Footnote-1187709
-Node: Avoiding MPI Deadlocks188622
-Node: FFTW MPI Performance Tips189651
-Node: Combining MPI and Threads191120
-Node: FFTW MPI Reference194591
-Node: MPI Files and Data Types195170
-Node: MPI Initialization196166
-Node: Using MPI Plans197265
-Node: MPI Data Distribution Functions199091
-Node: MPI Plan Creation204547
-Node: MPI Wisdom Communication215224
-Node: FFTW MPI Fortran Interface216150
-Ref: FFTW MPI Fortran Interface-Footnote-1222179
-Node: Calling FFTW from Modern Fortran222586
-Node: Overview of Fortran interface223937
-Node: Extended and quadruple precision in Fortran227389
-Node: Reversing array dimensions228770
-Node: FFTW Fortran type reference232305
-Node: Plan execution in Fortran236792
-Node: Allocating aligned memory in Fortran239688
-Node: Accessing the wisdom API from Fortran243052
-Node: Wisdom File Export/Import from Fortran243829
-Node: Wisdom String Export/Import from Fortran245491
-Node: Wisdom Generic Export/Import from Fortran247479
-Node: Defining an FFTW module249709
-Node: Calling FFTW from Legacy Fortran250778
-Node: Fortran-interface routines252335
-Ref: Fortran-interface routines-Footnote-1255993
-Ref: Fortran-interface routines-Footnote-2256196
-Node: FFTW Constants in Fortran256329
-Node: FFTW Execution in Fortran257484
-Node: Fortran Examples260240
-Node: Wisdom of Fortran?263659
-Node: Upgrading from FFTW version 2265339
-Ref: Upgrading from FFTW version 2-Footnote-1274962
-Node: Installation and Customization275145
-Node: Installation on Unix276789
-Node: Installation on non-Unix systems285452
-Node: Cycle Counters287667
-Node: Generating your own code289419
-Node: Acknowledgments291454
-Node: License and Copyright295174
-Node: Concept Index297043
-Node: Library Index333685
+Node: Top1076
+Node: Introduction1749
+Node: Tutorial8085
+Ref: Tutorial-Footnote-19329
+Node: Complex One-Dimensional DFTs9423
+Node: Complex Multi-Dimensional DFTs15179
+Ref: Complex Multi-Dimensional DFTs-Footnote-118611
+Node: One-Dimensional DFTs of Real Data18746
+Node: Multi-Dimensional DFTs of Real Data23191
+Node: More DFTs of Real Data27121
+Node: The Halfcomplex-format DFT30623
+Node: Real even/odd DFTs (cosine/sine transforms)33232
+Ref: Real even/odd DFTs (cosine/sine transforms)-Footnote-138842
+Ref: Real even/odd DFTs (cosine/sine transforms)-Footnote-239031
+Node: The Discrete Hartley Transform39964
+Ref: The Discrete Hartley Transform-Footnote-142149
+Node: Other Important Topics42398
+Node: SIMD alignment and fftw_malloc42691
+Node: Multi-dimensional Array Format44951
+Node: Row-major Format45572
+Node: Column-major Format47265
+Node: Fixed-size Arrays in C48349
+Node: Dynamic Arrays in C49785
+Node: Dynamic Arrays in C-The Wrong Way51423
+Node: Words of Wisdom-Saving Plans53171
+Node: Caveats in Using Wisdom55846
+Node: FFTW Reference57934
+Node: Data Types and Files58422
+Node: Complex numbers58854
+Node: Precision60595
+Node: Memory Allocation62157
+Node: Using Plans63728
+Node: Basic Interface67608
+Ref: Basic Interface-Footnote-168352
+Node: Complex DFTs68416
+Node: Planner Flags72383
+Node: Real-data DFTs77740
+Node: Real-data DFT Array Format82736
+Node: Real-to-Real Transforms84991
+Node: Real-to-Real Transform Kinds88961
+Node: Advanced Interface91429
+Node: Advanced Complex DFTs92169
+Node: Advanced Real-data DFTs96428
+Node: Advanced Real-to-real Transforms98755
+Node: Guru Interface99861
+Node: Interleaved and split arrays100784
+Node: Guru vector and transform sizes101827
+Node: Guru Complex DFTs104392
+Node: Guru Real-data DFTs107228
+Node: Guru Real-to-real Transforms110151
+Node: 64-bit Guru Interface111470
+Node: New-array Execute Functions113793
+Node: Wisdom117791
+Node: Wisdom Export118150
+Node: Wisdom Import120124
+Node: Forgetting Wisdom122146
+Node: Wisdom Utilities122518
+Node: What FFTW Really Computes123885
+Node: The 1d Discrete Fourier Transform (DFT)124710
+Node: The 1d Real-data DFT126069
+Node: 1d Real-even DFTs (DCTs)127723
+Node: 1d Real-odd DFTs (DSTs)130932
+Node: 1d Discrete Hartley Transforms (DHTs)133874
+Node: Multi-dimensional Transforms134550
+Node: Multi-threaded FFTW137153
+Node: Installation and Supported Hardware/Software138622
+Node: Usage of Multi-threaded FFTW140447
+Node: How Many Threads to Use?143755
+Node: Thread safety144779
+Node: Distributed-memory FFTW with MPI146947
+Node: FFTW MPI Installation149526
+Node: Linking and Initializing MPI FFTW151318
+Node: 2d MPI example152548
+Node: MPI Data Distribution156784
+Node: Basic and advanced distribution interfaces159662
+Node: Load balancing164097
+Node: Transposed distributions165783
+Node: One-dimensional distributions169555
+Node: Multi-dimensional MPI DFTs of Real Data172124
+Node: Other Multi-dimensional Real-data MPI Transforms176772
+Node: FFTW MPI Transposes178945
+Node: Basic distributed-transpose interface179785
+Node: Advanced distributed-transpose interface181969
+Node: An improved replacement for MPI_Alltoall183257
+Node: FFTW MPI Wisdom185233
+Ref: FFTW MPI Wisdom-Footnote-1187976
+Node: Avoiding MPI Deadlocks188889
+Node: FFTW MPI Performance Tips189918
+Node: Combining MPI and Threads191387
+Node: FFTW MPI Reference194858
+Node: MPI Files and Data Types195437
+Node: MPI Initialization196433
+Node: Using MPI Plans197532
+Node: MPI Data Distribution Functions199358
+Node: MPI Plan Creation204814
+Node: MPI Wisdom Communication215491
+Node: FFTW MPI Fortran Interface216417
+Ref: FFTW MPI Fortran Interface-Footnote-1222446
+Node: Calling FFTW from Modern Fortran222853
+Node: Overview of Fortran interface224204
+Node: Extended and quadruple precision in Fortran227656
+Node: Reversing array dimensions229037
+Node: FFTW Fortran type reference232572
+Node: Plan execution in Fortran237059
+Node: Allocating aligned memory in Fortran239955
+Node: Accessing the wisdom API from Fortran243319
+Node: Wisdom File Export/Import from Fortran244096
+Node: Wisdom String Export/Import from Fortran245758
+Node: Wisdom Generic Export/Import from Fortran247746
+Node: Defining an FFTW module249976
+Node: Calling FFTW from Legacy Fortran251045
+Node: Fortran-interface routines252602
+Ref: Fortran-interface routines-Footnote-1256260
+Ref: Fortran-interface routines-Footnote-2256463
+Node: FFTW Constants in Fortran256596
+Node: FFTW Execution in Fortran257751
+Node: Fortran Examples260507
+Node: Wisdom of Fortran?263926
+Node: Upgrading from FFTW version 2265606
+Ref: Upgrading from FFTW version 2-Footnote-1275229
+Node: Installation and Customization275412
+Node: Installation on Unix277056
+Node: Installation on non-Unix systems285719
+Node: Cycle Counters287934
+Node: Generating your own code289686
+Node: Acknowledgments291721
+Node: License and Copyright295441
+Node: Concept Index297310
+Node: Library Index333952
 
 End Tag Table
diff --git a/doc/fftw3.info-1 b/doc/fftw3.info-1
index e04c3ad..2aa546c 100644
--- a/doc/fftw3.info-1
+++ b/doc/fftw3.info-1
@@ -1,6 +1,6 @@
 This is fftw3.info, produced by makeinfo version 4.13 from fftw3.texi.
 
-This manual is for FFTW (version 3.3.2, 28 April 2012).
+This manual is for FFTW (version 3.3.3, 25 November 2012).
 
    Copyright (C) 2003 Matteo Frigo.
 
@@ -33,7 +33,7 @@ FFTW User Manual
 
 Welcome to FFTW, the Fastest Fourier Transform in the West.  FFTW is a
 collection of fast C routines to compute the discrete Fourier transform.
-This manual documents FFTW version 3.3.2.
+This manual documents FFTW version 3.3.3.
 
 * Menu:
 
@@ -58,7 +58,7 @@ File: fftw3.info,  Node: Introduction,  Next: Tutorial,  Prev: Top,  Up: Top
 1 Introduction
 **************
 
-This manual documents version 3.3.2 of FFTW, the _Fastest Fourier
+This manual documents version 3.3.3 of FFTW, the _Fastest Fourier
 Transform in the West_.  FFTW is a comprehensive collection of fast C
 routines for computing the discrete Fourier transform (DFT) and various
 special cases thereof.  
@@ -75,11 +75,11 @@ special cases thereof.
    * FFTW supports the SSE, SSE2, AVX, Altivec, and MIPS PS instruction
            sets.
 
-   * FFTW 3.3.2 includes parallel (multi-threaded) transforms   for
-     shared-memory systems.    FFTW 3.3.2 does not include
-     distributed-memory parallel   transforms, but we plan to implement
-     an MPI version soon.  (Meanwhile,   you can use the MPI
-     implementation from FFTW 2.1.3.)
+   * FFTW includes parallel (multi-threaded) transforms        for
+     shared-memory systems.
+
+   * Starting with version 3.3, FFTW includes distributed-memory
+     parallel        transforms using MPI.
 
    We assume herein that you are familiar with the properties and uses
 of the DFT that are relevant to your application.  Otherwise, see e.g.
@@ -520,16 +520,22 @@ slightly over half of the output of the corresponding complex DFT.
 (The division is rounded down.)  The ordering of the data is otherwise
 exactly the same as in the complex-DFT case.
 
-   Since the complex data is slightly larger than the real data, some
-complications arise for in-place transforms.  In this case, the final
-dimension of the real data must be padded with extra values to
-accommodate the size of the complex data--two values if the last
-dimension is even and one if it is odd.  That is, the last dimension of
-the real data must physically contain 2 * (n[d-1]/2+1) `double' values
-(exactly enough to hold the complex data).  This physical array size
-does not, however, change the _logical_ array size--only n[d-1] values
-are actually stored in the last dimension, and n[d-1] is the last
-dimension passed to the plan-creation routine.
+   For out-of-place transforms, this is the end of the story: the real
+data is stored as a row-major array of size n[0] x n[1] x n[2] x ... x
+n[d-1]  and the complex data is stored as a row-major array of size
+n[0] x n[1] x n[2] x ... x (n[d-1]/2 + 1) .
+
+   For in-place transforms, however, extra padding of the real-data
+array is necessary because the complex array is larger than the real
+array, and the two arrays share the same memory locations.  Thus, for
+in-place transforms, the final dimension of the real-data array must be
+padded with extra values to accommodate the size of the complex
+data--two values if the last dimension is even and one if it is odd.  That
+is, the last dimension of the real data must physically contain 2 *
+(n[d-1]/2+1) `double' values (exactly enough to hold the complex data).
+This physical array size does not, however, change the _logical_ array
+size--only n[d-1] values are actually stored in the last dimension, and
+n[d-1] is the last dimension passed to the plan-creation routine.
 
    For example, consider the transform of a two-dimensional real array
 of size `n0' by `n1'.  The output of the r2c transform is a
@@ -3050,7 +3056,7 @@ subsequent plans.
    With OpenMP, to configure FFTW to use all of the currently running
 OpenMP threads (set by `omp_set_num_threads(nthreads)' or by the
 `OMP_NUM_THREADS' environment variable), you can do:
-`fftw_plan_with_nthreads(omp_get_num_threads())'. (The `omp_' OpenMP
+`fftw_plan_with_nthreads(omp_get_max_threads())'. (The `omp_' OpenMP
 functions are declared via `#include <omp.h>'.)
 
    Given a plan, you then execute it as usual with
diff --git a/doc/fftw3.info-2 b/doc/fftw3.info-2
index b62bdca..d261793 100644
Binary files a/doc/fftw3.info-2 and b/doc/fftw3.info-2 differ
diff --git a/doc/fftw3.pdf b/doc/fftw3.pdf
index f8bf6f6..9178add 100644
Binary files a/doc/fftw3.pdf and b/doc/fftw3.pdf differ
diff --git a/doc/html/1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html b/doc/html/1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html
index 7e88041..0c54d96 100644
--- a/doc/html/1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html
+++ b/doc/html/1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>1d Discrete Hartley Transforms (DHTs) - FFTW 3.3.2</title>
+<title>1d Discrete Hartley Transforms (DHTs) - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="What-FFTW-Really-Computes.html#What-FFTW-Really-Computes" title="What FFTW Really Computes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/1d-Real_002deven-DFTs-_0028DCTs_0029.html b/doc/html/1d-Real_002deven-DFTs-_0028DCTs_0029.html
index 8b34dc3..3ef9563 100644
--- a/doc/html/1d-Real_002deven-DFTs-_0028DCTs_0029.html
+++ b/doc/html/1d-Real_002deven-DFTs-_0028DCTs_0029.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>1d Real-even DFTs (DCTs) - FFTW 3.3.2</title>
+<title>1d Real-even DFTs (DCTs) - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="What-FFTW-Really-Computes.html#What-FFTW-Really-Computes" title="What FFTW Really Computes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/1d-Real_002dodd-DFTs-_0028DSTs_0029.html b/doc/html/1d-Real_002dodd-DFTs-_0028DSTs_0029.html
index b3f567a..717fad5 100644
--- a/doc/html/1d-Real_002dodd-DFTs-_0028DSTs_0029.html
+++ b/doc/html/1d-Real_002dodd-DFTs-_0028DSTs_0029.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>1d Real-odd DFTs (DSTs) - FFTW 3.3.2</title>
+<title>1d Real-odd DFTs (DSTs) - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="What-FFTW-Really-Computes.html#What-FFTW-Really-Computes" title="What FFTW Really Computes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/2d-MPI-example.html b/doc/html/2d-MPI-example.html
index 91a2089..1a8ae37 100644
--- a/doc/html/2d-MPI-example.html
+++ b/doc/html/2d-MPI-example.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>2d MPI example - FFTW 3.3.2</title>
+<title>2d MPI example - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/64_002dbit-Guru-Interface.html b/doc/html/64_002dbit-Guru-Interface.html
index 4b74096..249598d 100644
--- a/doc/html/64_002dbit-Guru-Interface.html
+++ b/doc/html/64_002dbit-Guru-Interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>64-bit Guru Interface - FFTW 3.3.2</title>
+<title>64-bit Guru Interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Accessing-the-wisdom-API-from-Fortran.html b/doc/html/Accessing-the-wisdom-API-from-Fortran.html
index 307242e..7bb8caa 100644
--- a/doc/html/Accessing-the-wisdom-API-from-Fortran.html
+++ b/doc/html/Accessing-the-wisdom-API-from-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Accessing the wisdom API from Fortran - FFTW 3.3.2</title>
+<title>Accessing the wisdom API from Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Acknowledgments.html b/doc/html/Acknowledgments.html
index 99a4c78..e49ba4b 100644
--- a/doc/html/Acknowledgments.html
+++ b/doc/html/Acknowledgments.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Acknowledgments - FFTW 3.3.2</title>
+<title>Acknowledgments - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Installation-and-Customization.html#Installation-and-Customization" title="Installation and Customization">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Advanced-Complex-DFTs.html b/doc/html/Advanced-Complex-DFTs.html
index b5d4604..2a0fbdb 100644
--- a/doc/html/Advanced-Complex-DFTs.html
+++ b/doc/html/Advanced-Complex-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Advanced Complex DFTs - FFTW 3.3.2</title>
+<title>Advanced Complex DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Advanced-Interface.html#Advanced-Interface" title="Advanced Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Advanced-Interface.html b/doc/html/Advanced-Interface.html
index 022501d..5732943 100644
--- a/doc/html/Advanced-Interface.html
+++ b/doc/html/Advanced-Interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Advanced Interface - FFTW 3.3.2</title>
+<title>Advanced Interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Advanced-Real_002ddata-DFTs.html b/doc/html/Advanced-Real_002ddata-DFTs.html
index 095deb8..4622ccf 100644
--- a/doc/html/Advanced-Real_002ddata-DFTs.html
+++ b/doc/html/Advanced-Real_002ddata-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Advanced Real-data DFTs - FFTW 3.3.2</title>
+<title>Advanced Real-data DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Advanced-Interface.html#Advanced-Interface" title="Advanced Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Advanced-Real_002dto_002dreal-Transforms.html b/doc/html/Advanced-Real_002dto_002dreal-Transforms.html
index 2decbf1..6dbd802 100644
--- a/doc/html/Advanced-Real_002dto_002dreal-Transforms.html
+++ b/doc/html/Advanced-Real_002dto_002dreal-Transforms.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Advanced Real-to-real Transforms - FFTW 3.3.2</title>
+<title>Advanced Real-to-real Transforms - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Advanced-Interface.html#Advanced-Interface" title="Advanced Interface">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Advanced-distributed_002dtranspose-interface.html b/doc/html/Advanced-distributed_002dtranspose-interface.html
index c7c1730..79ac420 100644
--- a/doc/html/Advanced-distributed_002dtranspose-interface.html
+++ b/doc/html/Advanced-distributed_002dtranspose-interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Advanced distributed-transpose interface - FFTW 3.3.2</title>
+<title>Advanced distributed-transpose interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes" title="FFTW MPI Transposes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Allocating-aligned-memory-in-Fortran.html b/doc/html/Allocating-aligned-memory-in-Fortran.html
index 2657767..94de5cb 100644
--- a/doc/html/Allocating-aligned-memory-in-Fortran.html
+++ b/doc/html/Allocating-aligned-memory-in-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Allocating aligned memory in Fortran - FFTW 3.3.2</title>
+<title>Allocating aligned memory in Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/An-improved-replacement-for-MPI_005fAlltoall.html b/doc/html/An-improved-replacement-for-MPI_005fAlltoall.html
index 0f42392..f6a9671 100644
--- a/doc/html/An-improved-replacement-for-MPI_005fAlltoall.html
+++ b/doc/html/An-improved-replacement-for-MPI_005fAlltoall.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>An improved replacement for MPI_Alltoall - FFTW 3.3.2</title>
+<title>An improved replacement for MPI_Alltoall - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes" title="FFTW MPI Transposes">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Avoiding-MPI-Deadlocks.html b/doc/html/Avoiding-MPI-Deadlocks.html
index 5c8e435..5e72e0e 100644
--- a/doc/html/Avoiding-MPI-Deadlocks.html
+++ b/doc/html/Avoiding-MPI-Deadlocks.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Avoiding MPI Deadlocks - FFTW 3.3.2</title>
+<title>Avoiding MPI Deadlocks - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Basic-Interface.html b/doc/html/Basic-Interface.html
index 7bc00f6..827715c 100644
--- a/doc/html/Basic-Interface.html
+++ b/doc/html/Basic-Interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Basic Interface - FFTW 3.3.2</title>
+<title>Basic Interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Basic-and-advanced-distribution-interfaces.html b/doc/html/Basic-and-advanced-distribution-interfaces.html
index d7a65b7..5a6864d 100644
--- a/doc/html/Basic-and-advanced-distribution-interfaces.html
+++ b/doc/html/Basic-and-advanced-distribution-interfaces.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Basic and advanced distribution interfaces - FFTW 3.3.2</title>
+<title>Basic and advanced distribution interfaces - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="MPI-Data-Distribution.html#MPI-Data-Distribution" title="MPI Data Distribution">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Basic-distributed_002dtranspose-interface.html b/doc/html/Basic-distributed_002dtranspose-interface.html
index 4e9d9f4..bd28410 100644
--- a/doc/html/Basic-distributed_002dtranspose-interface.html
+++ b/doc/html/Basic-distributed_002dtranspose-interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Basic distributed-transpose interface - FFTW 3.3.2</title>
+<title>Basic distributed-transpose interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes" title="FFTW MPI Transposes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Calling-FFTW-from-Legacy-Fortran.html b/doc/html/Calling-FFTW-from-Legacy-Fortran.html
index 5c6a3ba..e9facd1 100644
--- a/doc/html/Calling-FFTW-from-Legacy-Fortran.html
+++ b/doc/html/Calling-FFTW-from-Legacy-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Calling FFTW from Legacy Fortran - FFTW 3.3.2</title>
+<title>Calling FFTW from Legacy Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Calling-FFTW-from-Modern-Fortran.html b/doc/html/Calling-FFTW-from-Modern-Fortran.html
index 4f32271..9057bd6 100644
--- a/doc/html/Calling-FFTW-from-Modern-Fortran.html
+++ b/doc/html/Calling-FFTW-from-Modern-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Calling FFTW from Modern Fortran - FFTW 3.3.2</title>
+<title>Calling FFTW from Modern Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Caveats-in-Using-Wisdom.html b/doc/html/Caveats-in-Using-Wisdom.html
index 85ed4cd..1dc61c3 100644
--- a/doc/html/Caveats-in-Using-Wisdom.html
+++ b/doc/html/Caveats-in-Using-Wisdom.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Caveats in Using Wisdom - FFTW 3.3.2</title>
+<title>Caveats in Using Wisdom - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Other-Important-Topics.html#Other-Important-Topics" title="Other Important Topics">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Column_002dmajor-Format.html b/doc/html/Column_002dmajor-Format.html
index e15028a..f972b48 100644
--- a/doc/html/Column_002dmajor-Format.html
+++ b/doc/html/Column_002dmajor-Format.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Column-major Format - FFTW 3.3.2</title>
+<title>Column-major Format - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002ddimensional-Array-Format.html#Multi_002ddimensional-Array-Format" title="Multi-dimensional Array Format">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Combining-MPI-and-Threads.html b/doc/html/Combining-MPI-and-Threads.html
index 3c05bef..cd14af9 100644
--- a/doc/html/Combining-MPI-and-Threads.html
+++ b/doc/html/Combining-MPI-and-Threads.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Combining MPI and Threads - FFTW 3.3.2</title>
+<title>Combining MPI and Threads - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Complex-DFTs.html b/doc/html/Complex-DFTs.html
index 167fe95..6b537d7 100644
--- a/doc/html/Complex-DFTs.html
+++ b/doc/html/Complex-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Complex DFTs - FFTW 3.3.2</title>
+<title>Complex DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Basic-Interface.html#Basic-Interface" title="Basic Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Complex-Multi_002dDimensional-DFTs.html b/doc/html/Complex-Multi_002dDimensional-DFTs.html
index 980e8b7..aa9d9d1 100644
--- a/doc/html/Complex-Multi_002dDimensional-DFTs.html
+++ b/doc/html/Complex-Multi_002dDimensional-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Complex Multi-Dimensional DFTs - FFTW 3.3.2</title>
+<title>Complex Multi-Dimensional DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Tutorial.html#Tutorial" title="Tutorial">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Complex-One_002dDimensional-DFTs.html b/doc/html/Complex-One_002dDimensional-DFTs.html
index fc87959..bdbe662 100644
--- a/doc/html/Complex-One_002dDimensional-DFTs.html
+++ b/doc/html/Complex-One_002dDimensional-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Complex One-Dimensional DFTs - FFTW 3.3.2</title>
+<title>Complex One-Dimensional DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Tutorial.html#Tutorial" title="Tutorial">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Complex-numbers.html b/doc/html/Complex-numbers.html
index aaad739..5836d19 100644
--- a/doc/html/Complex-numbers.html
+++ b/doc/html/Complex-numbers.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Complex numbers - FFTW 3.3.2</title>
+<title>Complex numbers - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Data-Types-and-Files.html#Data-Types-and-Files" title="Data Types and Files">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Concept-Index.html b/doc/html/Concept-Index.html
index 72688ea..eb2f56a 100644
--- a/doc/html/Concept-Index.html
+++ b/doc/html/Concept-Index.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Concept Index - FFTW 3.3.2</title>
+<title>Concept Index - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="License-and-Copyright.html#License-and-Copyright" title="License and Copyright">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Cycle-Counters.html b/doc/html/Cycle-Counters.html
index 0006c2a..1105b64 100644
--- a/doc/html/Cycle-Counters.html
+++ b/doc/html/Cycle-Counters.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Cycle Counters - FFTW 3.3.2</title>
+<title>Cycle Counters - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Installation-and-Customization.html#Installation-and-Customization" title="Installation and Customization">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Data-Types-and-Files.html b/doc/html/Data-Types-and-Files.html
index 2b27952..d6fa400 100644
--- a/doc/html/Data-Types-and-Files.html
+++ b/doc/html/Data-Types-and-Files.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Data Types and Files - FFTW 3.3.2</title>
+<title>Data Types and Files - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Defining-an-FFTW-module.html b/doc/html/Defining-an-FFTW-module.html
index 46d8653..76bc96b 100644
--- a/doc/html/Defining-an-FFTW-module.html
+++ b/doc/html/Defining-an-FFTW-module.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Defining an FFTW module - FFTW 3.3.2</title>
+<title>Defining an FFTW module - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Distributed_002dmemory-FFTW-with-MPI.html b/doc/html/Distributed_002dmemory-FFTW-with-MPI.html
index eb81ece..8581f1a 100644
--- a/doc/html/Distributed_002dmemory-FFTW-with-MPI.html
+++ b/doc/html/Distributed_002dmemory-FFTW-with-MPI.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Distributed-memory FFTW with MPI - FFTW 3.3.2</title>
+<title>Distributed-memory FFTW with MPI - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW" title="Multi-threaded FFTW">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Dynamic-Arrays-in-C.html b/doc/html/Dynamic-Arrays-in-C.html
index c0c925f..0fff16e 100644
--- a/doc/html/Dynamic-Arrays-in-C.html
+++ b/doc/html/Dynamic-Arrays-in-C.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Dynamic Arrays in C - FFTW 3.3.2</title>
+<title>Dynamic Arrays in C - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002ddimensional-Array-Format.html#Multi_002ddimensional-Array-Format" title="Multi-dimensional Array Format">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Dynamic-Arrays-in-C_002dThe-Wrong-Way.html b/doc/html/Dynamic-Arrays-in-C_002dThe-Wrong-Way.html
index bb8d011..339a1ed 100644
--- a/doc/html/Dynamic-Arrays-in-C_002dThe-Wrong-Way.html
+++ b/doc/html/Dynamic-Arrays-in-C_002dThe-Wrong-Way.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Dynamic Arrays in C-The Wrong Way - FFTW 3.3.2</title>
+<title>Dynamic Arrays in C-The Wrong Way - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002ddimensional-Array-Format.html#Multi_002ddimensional-Array-Format" title="Multi-dimensional Array Format">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Extended-and-quadruple-precision-in-Fortran.html b/doc/html/Extended-and-quadruple-precision-in-Fortran.html
index af71e95..1b2974b 100644
--- a/doc/html/Extended-and-quadruple-precision-in-Fortran.html
+++ b/doc/html/Extended-and-quadruple-precision-in-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Extended and quadruple precision in Fortran - FFTW 3.3.2</title>
+<title>Extended and quadruple precision in Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Overview-of-Fortran-interface.html#Overview-of-Fortran-interface" title="Overview of Fortran interface">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-Constants-in-Fortran.html b/doc/html/FFTW-Constants-in-Fortran.html
index 5e48254..e2f1b35 100644
--- a/doc/html/FFTW-Constants-in-Fortran.html
+++ b/doc/html/FFTW-Constants-in-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW Constants in Fortran - FFTW 3.3.2</title>
+<title>FFTW Constants in Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Legacy-Fortran.html#Calling-FFTW-from-Legacy-Fortran" title="Calling FFTW from Legacy Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-Execution-in-Fortran.html b/doc/html/FFTW-Execution-in-Fortran.html
index 95bfe00..56353e5 100644
--- a/doc/html/FFTW-Execution-in-Fortran.html
+++ b/doc/html/FFTW-Execution-in-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW Execution in Fortran - FFTW 3.3.2</title>
+<title>FFTW Execution in Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Legacy-Fortran.html#Calling-FFTW-from-Legacy-Fortran" title="Calling FFTW from Legacy Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-Fortran-type-reference.html b/doc/html/FFTW-Fortran-type-reference.html
index b7582be..223bc4a 100644
--- a/doc/html/FFTW-Fortran-type-reference.html
+++ b/doc/html/FFTW-Fortran-type-reference.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW Fortran type reference - FFTW 3.3.2</title>
+<title>FFTW Fortran type reference - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-MPI-Fortran-Interface.html b/doc/html/FFTW-MPI-Fortran-Interface.html
index cac6770..105d1fa 100644
--- a/doc/html/FFTW-MPI-Fortran-Interface.html
+++ b/doc/html/FFTW-MPI-Fortran-Interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW MPI Fortran Interface - FFTW 3.3.2</title>
+<title>FFTW MPI Fortran Interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-MPI-Installation.html b/doc/html/FFTW-MPI-Installation.html
index aa6afd9..54bd05d 100644
--- a/doc/html/FFTW-MPI-Installation.html
+++ b/doc/html/FFTW-MPI-Installation.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW MPI Installation - FFTW 3.3.2</title>
+<title>FFTW MPI Installation - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-MPI-Performance-Tips.html b/doc/html/FFTW-MPI-Performance-Tips.html
index fd6d31b..e970287 100644
--- a/doc/html/FFTW-MPI-Performance-Tips.html
+++ b/doc/html/FFTW-MPI-Performance-Tips.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW MPI Performance Tips - FFTW 3.3.2</title>
+<title>FFTW MPI Performance Tips - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-MPI-Reference.html b/doc/html/FFTW-MPI-Reference.html
index 3ee8e8b..bbcfd59 100644
--- a/doc/html/FFTW-MPI-Reference.html
+++ b/doc/html/FFTW-MPI-Reference.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW MPI Reference - FFTW 3.3.2</title>
+<title>FFTW MPI Reference - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-MPI-Transposes.html b/doc/html/FFTW-MPI-Transposes.html
index 78724cb..8d5ba38 100644
--- a/doc/html/FFTW-MPI-Transposes.html
+++ b/doc/html/FFTW-MPI-Transposes.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW MPI Transposes - FFTW 3.3.2</title>
+<title>FFTW MPI Transposes - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-MPI-Wisdom.html b/doc/html/FFTW-MPI-Wisdom.html
index 9d588dd..a31c0d9 100644
--- a/doc/html/FFTW-MPI-Wisdom.html
+++ b/doc/html/FFTW-MPI-Wisdom.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW MPI Wisdom - FFTW 3.3.2</title>
+<title>FFTW MPI Wisdom - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/FFTW-Reference.html b/doc/html/FFTW-Reference.html
index e171a4a..708f26e 100644
--- a/doc/html/FFTW-Reference.html
+++ b/doc/html/FFTW-Reference.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>FFTW Reference - FFTW 3.3.2</title>
+<title>FFTW Reference - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Other-Important-Topics.html#Other-Important-Topics" title="Other Important Topics">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Fixed_002dsize-Arrays-in-C.html b/doc/html/Fixed_002dsize-Arrays-in-C.html
index c3d9f1b..3117987 100644
--- a/doc/html/Fixed_002dsize-Arrays-in-C.html
+++ b/doc/html/Fixed_002dsize-Arrays-in-C.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Fixed-size Arrays in C - FFTW 3.3.2</title>
+<title>Fixed-size Arrays in C - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002ddimensional-Array-Format.html#Multi_002ddimensional-Array-Format" title="Multi-dimensional Array Format">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Forgetting-Wisdom.html b/doc/html/Forgetting-Wisdom.html
index 60da331..ef4ce1d 100644
--- a/doc/html/Forgetting-Wisdom.html
+++ b/doc/html/Forgetting-Wisdom.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Forgetting Wisdom - FFTW 3.3.2</title>
+<title>Forgetting Wisdom - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Wisdom.html#Wisdom" title="Wisdom">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Fortran-Examples.html b/doc/html/Fortran-Examples.html
index bd12a31..c63e08b 100644
--- a/doc/html/Fortran-Examples.html
+++ b/doc/html/Fortran-Examples.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Fortran Examples - FFTW 3.3.2</title>
+<title>Fortran Examples - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Legacy-Fortran.html#Calling-FFTW-from-Legacy-Fortran" title="Calling FFTW from Legacy Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Fortran_002dinterface-routines.html b/doc/html/Fortran_002dinterface-routines.html
index 421a9d9..c7c7de8 100644
--- a/doc/html/Fortran_002dinterface-routines.html
+++ b/doc/html/Fortran_002dinterface-routines.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Fortran-interface routines - FFTW 3.3.2</title>
+<title>Fortran-interface routines - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Legacy-Fortran.html#Calling-FFTW-from-Legacy-Fortran" title="Calling FFTW from Legacy Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Generating-your-own-code.html b/doc/html/Generating-your-own-code.html
index 91d3059..eb6f541 100644
--- a/doc/html/Generating-your-own-code.html
+++ b/doc/html/Generating-your-own-code.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Generating your own code - FFTW 3.3.2</title>
+<title>Generating your own code - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Installation-and-Customization.html#Installation-and-Customization" title="Installation and Customization">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Guru-Complex-DFTs.html b/doc/html/Guru-Complex-DFTs.html
index 0384c09..1d00b9a 100644
--- a/doc/html/Guru-Complex-DFTs.html
+++ b/doc/html/Guru-Complex-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Guru Complex DFTs - FFTW 3.3.2</title>
+<title>Guru Complex DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Guru-Interface.html b/doc/html/Guru-Interface.html
index 7905300..eb52bf1 100644
--- a/doc/html/Guru-Interface.html
+++ b/doc/html/Guru-Interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Guru Interface - FFTW 3.3.2</title>
+<title>Guru Interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Guru-Real_002ddata-DFTs.html b/doc/html/Guru-Real_002ddata-DFTs.html
index 3ca35cf..6cbb9ba 100644
--- a/doc/html/Guru-Real_002ddata-DFTs.html
+++ b/doc/html/Guru-Real_002ddata-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Guru Real-data DFTs - FFTW 3.3.2</title>
+<title>Guru Real-data DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Guru-Real_002dto_002dreal-Transforms.html b/doc/html/Guru-Real_002dto_002dreal-Transforms.html
index b00295e..9f9b61c 100644
--- a/doc/html/Guru-Real_002dto_002dreal-Transforms.html
+++ b/doc/html/Guru-Real_002dto_002dreal-Transforms.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Guru Real-to-real Transforms - FFTW 3.3.2</title>
+<title>Guru Real-to-real Transforms - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Guru-vector-and-transform-sizes.html b/doc/html/Guru-vector-and-transform-sizes.html
index 6b5a0c5..617d027 100644
--- a/doc/html/Guru-vector-and-transform-sizes.html
+++ b/doc/html/Guru-vector-and-transform-sizes.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Guru vector and transform sizes - FFTW 3.3.2</title>
+<title>Guru vector and transform sizes - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/How-Many-Threads-to-Use_003f.html b/doc/html/How-Many-Threads-to-Use_003f.html
index c00b4a2..80e50eb 100644
--- a/doc/html/How-Many-Threads-to-Use_003f.html
+++ b/doc/html/How-Many-Threads-to-Use_003f.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>How Many Threads to Use? - FFTW 3.3.2</title>
+<title>How Many Threads to Use? - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW" title="Multi-threaded FFTW">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Installation-and-Customization.html b/doc/html/Installation-and-Customization.html
index 9e1b4ee..c748985 100644
--- a/doc/html/Installation-and-Customization.html
+++ b/doc/html/Installation-and-Customization.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Installation and Customization - FFTW 3.3.2</title>
+<title>Installation and Customization - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Upgrading-from-FFTW-version-2.html#Upgrading-from-FFTW-version-2" title="Upgrading from FFTW version 2">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Installation-and-Supported-Hardware_002fSoftware.html b/doc/html/Installation-and-Supported-Hardware_002fSoftware.html
index d2ddbf8..919fdf2 100644
--- a/doc/html/Installation-and-Supported-Hardware_002fSoftware.html
+++ b/doc/html/Installation-and-Supported-Hardware_002fSoftware.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Installation and Supported Hardware/Software - FFTW 3.3.2</title>
+<title>Installation and Supported Hardware/Software - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW" title="Multi-threaded FFTW">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Installation-on-Unix.html b/doc/html/Installation-on-Unix.html
index a215791..79447b8 100644
--- a/doc/html/Installation-on-Unix.html
+++ b/doc/html/Installation-on-Unix.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Installation on Unix - FFTW 3.3.2</title>
+<title>Installation on Unix - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Installation-and-Customization.html#Installation-and-Customization" title="Installation and Customization">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Installation-on-non_002dUnix-systems.html b/doc/html/Installation-on-non_002dUnix-systems.html
index e28f2fb..26fa4f0 100644
--- a/doc/html/Installation-on-non_002dUnix-systems.html
+++ b/doc/html/Installation-on-non_002dUnix-systems.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Installation on non-Unix systems - FFTW 3.3.2</title>
+<title>Installation on non-Unix systems - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Installation-and-Customization.html#Installation-and-Customization" title="Installation and Customization">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Interleaved-and-split-arrays.html b/doc/html/Interleaved-and-split-arrays.html
index ab94552..67eb0ae 100644
--- a/doc/html/Interleaved-and-split-arrays.html
+++ b/doc/html/Interleaved-and-split-arrays.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Interleaved and split arrays - FFTW 3.3.2</title>
+<title>Interleaved and split arrays - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Introduction.html b/doc/html/Introduction.html
index e932d4d..a310fa0 100644
--- a/doc/html/Introduction.html
+++ b/doc/html/Introduction.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Introduction - FFTW 3.3.2</title>
+<title>Introduction - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="index.html#Top" title="Top">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
@@ -55,7 +55,7 @@ Up: <a rel="up" accesskey="u" href="index.html#Top">Top</a>
 
 <h2 class="chapter">1 Introduction</h2>
 
-<p>This manual documents version 3.3.2 of FFTW, the
+<p>This manual documents version 3.3.3 of FFTW, the
 <em>Fastest Fourier Transform in the West</em>.  FFTW is a comprehensive
 collection of fast C routines for computing the discrete Fourier
 transform (DFT) and various special cases thereof. 
@@ -75,11 +75,10 @@ transform (DFT) and various special cases thereof.
      <li>FFTW supports the SSE, SSE2, AVX, Altivec, and MIPS PS instruction
        sets.
 
-     <li>FFTW 3.3.2 includes parallel (multi-threaded) transforms
-  for shared-memory systems. 
-  FFTW 3.3.2 does not include distributed-memory parallel
-  transforms, but we plan to implement an MPI version soon.  (Meanwhile,
-  you can use the MPI implementation from FFTW 2.1.3.) 
+     <li>FFTW includes parallel (multi-threaded) transforms
+       for shared-memory systems. 
+<li>Starting with version 3.3, FFTW includes distributed-memory parallel
+       transforms using MPI. 
 </ul>
 
    <p>We assume herein that you are familiar with the properties and uses of
diff --git a/doc/html/Library-Index.html b/doc/html/Library-Index.html
index a6a966b..5eec48c 100644
--- a/doc/html/Library-Index.html
+++ b/doc/html/Library-Index.html
@@ -1,15 +1,15 @@
 <html lang="en">
 <head>
-<title>Library Index - FFTW 3.3.2</title>
+<title>Library Index - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Concept-Index.html#Concept-Index" title="Concept Index">
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/License-and-Copyright.html b/doc/html/License-and-Copyright.html
index fa52418..eba4536 100644
--- a/doc/html/License-and-Copyright.html
+++ b/doc/html/License-and-Copyright.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>License and Copyright - FFTW 3.3.2</title>
+<title>License and Copyright - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Acknowledgments.html#Acknowledgments" title="Acknowledgments">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Linking-and-Initializing-MPI-FFTW.html b/doc/html/Linking-and-Initializing-MPI-FFTW.html
index ed28b2a..f1e1582 100644
--- a/doc/html/Linking-and-Initializing-MPI-FFTW.html
+++ b/doc/html/Linking-and-Initializing-MPI-FFTW.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Linking and Initializing MPI FFTW - FFTW 3.3.2</title>
+<title>Linking and Initializing MPI FFTW - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Load-balancing.html b/doc/html/Load-balancing.html
index b7afc43..1f28c32 100644
--- a/doc/html/Load-balancing.html
+++ b/doc/html/Load-balancing.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Load balancing - FFTW 3.3.2</title>
+<title>Load balancing - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="MPI-Data-Distribution.html#MPI-Data-Distribution" title="MPI Data Distribution">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/MPI-Data-Distribution-Functions.html b/doc/html/MPI-Data-Distribution-Functions.html
index 18c1253..c38dc40 100644
--- a/doc/html/MPI-Data-Distribution-Functions.html
+++ b/doc/html/MPI-Data-Distribution-Functions.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>MPI Data Distribution Functions - FFTW 3.3.2</title>
+<title>MPI Data Distribution Functions - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/MPI-Data-Distribution.html b/doc/html/MPI-Data-Distribution.html
index 8b8bcfd..6f34660 100644
--- a/doc/html/MPI-Data-Distribution.html
+++ b/doc/html/MPI-Data-Distribution.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>MPI Data Distribution - FFTW 3.3.2</title>
+<title>MPI Data Distribution - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/MPI-Files-and-Data-Types.html b/doc/html/MPI-Files-and-Data-Types.html
index f3acf6c..0e8e2bb 100644
--- a/doc/html/MPI-Files-and-Data-Types.html
+++ b/doc/html/MPI-Files-and-Data-Types.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>MPI Files and Data Types - FFTW 3.3.2</title>
+<title>MPI Files and Data Types - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/MPI-Initialization.html b/doc/html/MPI-Initialization.html
index 7f42ffc..a6ea3b4 100644
--- a/doc/html/MPI-Initialization.html
+++ b/doc/html/MPI-Initialization.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>MPI Initialization - FFTW 3.3.2</title>
+<title>MPI Initialization - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/MPI-Plan-Creation.html b/doc/html/MPI-Plan-Creation.html
index a695996..7ea7813 100644
--- a/doc/html/MPI-Plan-Creation.html
+++ b/doc/html/MPI-Plan-Creation.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>MPI Plan Creation - FFTW 3.3.2</title>
+<title>MPI Plan Creation - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/MPI-Wisdom-Communication.html b/doc/html/MPI-Wisdom-Communication.html
index e43d527..367ee5f 100644
--- a/doc/html/MPI-Wisdom-Communication.html
+++ b/doc/html/MPI-Wisdom-Communication.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>MPI Wisdom Communication - FFTW 3.3.2</title>
+<title>MPI Wisdom Communication - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Memory-Allocation.html b/doc/html/Memory-Allocation.html
index 412f594..d4efddd 100644
--- a/doc/html/Memory-Allocation.html
+++ b/doc/html/Memory-Allocation.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Memory Allocation - FFTW 3.3.2</title>
+<title>Memory Allocation - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Data-Types-and-Files.html#Data-Types-and-Files" title="Data Types and Files">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/More-DFTs-of-Real-Data.html b/doc/html/More-DFTs-of-Real-Data.html
index b706e04..9f76c3e 100644
--- a/doc/html/More-DFTs-of-Real-Data.html
+++ b/doc/html/More-DFTs-of-Real-Data.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>More DFTs of Real Data - FFTW 3.3.2</title>
+<title>More DFTs of Real Data - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Tutorial.html#Tutorial" title="Tutorial">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Multi_002dDimensional-DFTs-of-Real-Data.html b/doc/html/Multi_002dDimensional-DFTs-of-Real-Data.html
index 9bdcf44..0511400 100644
--- a/doc/html/Multi_002dDimensional-DFTs-of-Real-Data.html
+++ b/doc/html/Multi_002dDimensional-DFTs-of-Real-Data.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Multi-Dimensional DFTs of Real Data - FFTW 3.3.2</title>
+<title>Multi-Dimensional DFTs of Real Data - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Tutorial.html#Tutorial" title="Tutorial">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
@@ -86,11 +86,16 @@ over half of the output of the corresponding complex DFT.  (The division
 is rounded down.)  The ordering of the data is otherwise exactly the
 same as in the complex-DFT case.
 
-   <p>Since the complex data is slightly larger than the real data, some
-complications arise for in-place transforms.  In this case, the final
-dimension of the real data must be padded with extra values to
-accommodate the size of the complex data—two values if the last
-dimension is even and one if it is odd. 
+   <p>For out-of-place transforms, this is the end of the story: the real
+data is stored as a row-major array of size n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × n<sub>d-1</sub> and the complex
+data is stored as a row-major array of size n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × (n<sub>d-1</sub>/2 + 1).
+
+   <p>For in-place transforms, however, extra padding of the real-data array
+is necessary because the complex array is larger than the real array,
+and the two arrays share the same memory locations.  Thus, for
+in-place transforms, the final dimension of the real-data array must
+be padded with extra values to accommodate the size of the complex
+data—two values if the last dimension is even and one if it is odd. 
 <a name="index-padding-63"></a>That is, the last dimension of the real data must physically contain
 2 * (n<sub>d-1</sub>/2+1)<code>double</code> values (exactly enough to hold the complex data). 
 This physical array size does not, however, change the <em>logical</em>
diff --git a/doc/html/Multi_002ddimensional-Array-Format.html b/doc/html/Multi_002ddimensional-Array-Format.html
index d5ebf89..ab28ebe 100644
--- a/doc/html/Multi_002ddimensional-Array-Format.html
+++ b/doc/html/Multi_002ddimensional-Array-Format.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Multi-dimensional Array Format - FFTW 3.3.2</title>
+<title>Multi-dimensional Array Format - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Other-Important-Topics.html#Other-Important-Topics" title="Other Important Topics">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Multi_002ddimensional-MPI-DFTs-of-Real-Data.html b/doc/html/Multi_002ddimensional-MPI-DFTs-of-Real-Data.html
index dae42cb..33b7f73 100644
--- a/doc/html/Multi_002ddimensional-MPI-DFTs-of-Real-Data.html
+++ b/doc/html/Multi_002ddimensional-MPI-DFTs-of-Real-Data.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Multi-dimensional MPI DFTs of Real Data - FFTW 3.3.2</title>
+<title>Multi-dimensional MPI DFTs of Real Data - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Multi_002ddimensional-Transforms.html b/doc/html/Multi_002ddimensional-Transforms.html
index 07c08c8..939aacf 100644
--- a/doc/html/Multi_002ddimensional-Transforms.html
+++ b/doc/html/Multi_002ddimensional-Transforms.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Multi-dimensional Transforms - FFTW 3.3.2</title>
+<title>Multi-dimensional Transforms - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="What-FFTW-Really-Computes.html#What-FFTW-Really-Computes" title="What FFTW Really Computes">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Multi_002dthreaded-FFTW.html b/doc/html/Multi_002dthreaded-FFTW.html
index f5ce2f6..d736c17 100644
--- a/doc/html/Multi_002dthreaded-FFTW.html
+++ b/doc/html/Multi_002dthreaded-FFTW.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Multi-threaded FFTW - FFTW 3.3.2</title>
+<title>Multi-threaded FFTW - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/New_002darray-Execute-Functions.html b/doc/html/New_002darray-Execute-Functions.html
index 14318ab..61abf32 100644
--- a/doc/html/New_002darray-Execute-Functions.html
+++ b/doc/html/New_002darray-Execute-Functions.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>New-array Execute Functions - FFTW 3.3.2</title>
+<title>New-array Execute Functions - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/One_002dDimensional-DFTs-of-Real-Data.html b/doc/html/One_002dDimensional-DFTs-of-Real-Data.html
index c5d02ce..29792da 100644
--- a/doc/html/One_002dDimensional-DFTs-of-Real-Data.html
+++ b/doc/html/One_002dDimensional-DFTs-of-Real-Data.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>One-Dimensional DFTs of Real Data - FFTW 3.3.2</title>
+<title>One-Dimensional DFTs of Real Data - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Tutorial.html#Tutorial" title="Tutorial">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/One_002ddimensional-distributions.html b/doc/html/One_002ddimensional-distributions.html
index 50d273b..46a284d 100644
--- a/doc/html/One_002ddimensional-distributions.html
+++ b/doc/html/One_002ddimensional-distributions.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>One-dimensional distributions - FFTW 3.3.2</title>
+<title>One-dimensional distributions - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="MPI-Data-Distribution.html#MPI-Data-Distribution" title="MPI Data Distribution">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Other-Important-Topics.html b/doc/html/Other-Important-Topics.html
index 83d703b..b35e56c 100644
--- a/doc/html/Other-Important-Topics.html
+++ b/doc/html/Other-Important-Topics.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Other Important Topics - FFTW 3.3.2</title>
+<title>Other Important Topics - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Tutorial.html#Tutorial" title="Tutorial">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html b/doc/html/Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html
index 37d428f..d75298c 100644
--- a/doc/html/Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html
+++ b/doc/html/Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Other Multi-dimensional Real-data MPI Transforms - FFTW 3.3.2</title>
+<title>Other Multi-dimensional Real-data MPI Transforms - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Overview-of-Fortran-interface.html b/doc/html/Overview-of-Fortran-interface.html
index 310688c..63b3d12 100644
--- a/doc/html/Overview-of-Fortran-interface.html
+++ b/doc/html/Overview-of-Fortran-interface.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Overview of Fortran interface - FFTW 3.3.2</title>
+<title>Overview of Fortran interface - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Plan-execution-in-Fortran.html b/doc/html/Plan-execution-in-Fortran.html
index 64e8a27..c258549 100644
--- a/doc/html/Plan-execution-in-Fortran.html
+++ b/doc/html/Plan-execution-in-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Plan execution in Fortran - FFTW 3.3.2</title>
+<title>Plan execution in Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Planner-Flags.html b/doc/html/Planner-Flags.html
index 9f679f7..d248467 100644
--- a/doc/html/Planner-Flags.html
+++ b/doc/html/Planner-Flags.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Planner Flags - FFTW 3.3.2</title>
+<title>Planner Flags - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Basic-Interface.html#Basic-Interface" title="Basic Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Precision.html b/doc/html/Precision.html
index bd60f31..093018d 100644
--- a/doc/html/Precision.html
+++ b/doc/html/Precision.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Precision - FFTW 3.3.2</title>
+<title>Precision - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Data-Types-and-Files.html#Data-Types-and-Files" title="Data Types and Files">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html b/doc/html/Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html
index a4ba592..3018893 100644
--- a/doc/html/Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html
+++ b/doc/html/Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Real even/odd DFTs (cosine/sine transforms) - FFTW 3.3.2</title>
+<title>Real even/odd DFTs (cosine/sine transforms) - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="More-DFTs-of-Real-Data.html#More-DFTs-of-Real-Data" title="More DFTs of Real Data">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Real_002ddata-DFT-Array-Format.html b/doc/html/Real_002ddata-DFT-Array-Format.html
index 151bf49..bd401f5 100644
--- a/doc/html/Real_002ddata-DFT-Array-Format.html
+++ b/doc/html/Real_002ddata-DFT-Array-Format.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Real-data DFT Array Format - FFTW 3.3.2</title>
+<title>Real-data DFT Array Format - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Basic-Interface.html#Basic-Interface" title="Basic Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Real_002ddata-DFTs.html b/doc/html/Real_002ddata-DFTs.html
index 69d5b34..88c7eba 100644
--- a/doc/html/Real_002ddata-DFTs.html
+++ b/doc/html/Real_002ddata-DFTs.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Real-data DFTs - FFTW 3.3.2</title>
+<title>Real-data DFTs - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Basic-Interface.html#Basic-Interface" title="Basic Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Real_002dto_002dReal-Transform-Kinds.html b/doc/html/Real_002dto_002dReal-Transform-Kinds.html
index 42b98f1..5cbb9fb 100644
--- a/doc/html/Real_002dto_002dReal-Transform-Kinds.html
+++ b/doc/html/Real_002dto_002dReal-Transform-Kinds.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Real-to-Real Transform Kinds - FFTW 3.3.2</title>
+<title>Real-to-Real Transform Kinds - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Basic-Interface.html#Basic-Interface" title="Basic Interface">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Real_002dto_002dReal-Transforms.html b/doc/html/Real_002dto_002dReal-Transforms.html
index 2b0014a..a04723b 100644
--- a/doc/html/Real_002dto_002dReal-Transforms.html
+++ b/doc/html/Real_002dto_002dReal-Transforms.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Real-to-Real Transforms - FFTW 3.3.2</title>
+<title>Real-to-Real Transforms - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Basic-Interface.html#Basic-Interface" title="Basic Interface">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Reversing-array-dimensions.html b/doc/html/Reversing-array-dimensions.html
index 85125ed..5f73a58 100644
--- a/doc/html/Reversing-array-dimensions.html
+++ b/doc/html/Reversing-array-dimensions.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Reversing array dimensions - FFTW 3.3.2</title>
+<title>Reversing array dimensions - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran" title="Calling FFTW from Modern Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Row_002dmajor-Format.html b/doc/html/Row_002dmajor-Format.html
index 287743e..4cd1d93 100644
--- a/doc/html/Row_002dmajor-Format.html
+++ b/doc/html/Row_002dmajor-Format.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Row-major Format - FFTW 3.3.2</title>
+<title>Row-major Format - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002ddimensional-Array-Format.html#Multi_002ddimensional-Array-Format" title="Multi-dimensional Array Format">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/SIMD-alignment-and-fftw_005fmalloc.html b/doc/html/SIMD-alignment-and-fftw_005fmalloc.html
index c7fefb5..0325c84 100644
--- a/doc/html/SIMD-alignment-and-fftw_005fmalloc.html
+++ b/doc/html/SIMD-alignment-and-fftw_005fmalloc.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>SIMD alignment and fftw_malloc - FFTW 3.3.2</title>
+<title>SIMD alignment and fftw_malloc - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Other-Important-Topics.html#Other-Important-Topics" title="Other Important Topics">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html b/doc/html/The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html
index 332f52a..6747161 100644
--- a/doc/html/The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html
+++ b/doc/html/The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>The 1d Discrete Fourier Transform (DFT) - FFTW 3.3.2</title>
+<title>The 1d Discrete Fourier Transform (DFT) - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="What-FFTW-Really-Computes.html#What-FFTW-Really-Computes" title="What FFTW Really Computes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/The-1d-Real_002ddata-DFT.html b/doc/html/The-1d-Real_002ddata-DFT.html
index 3b06c1a..84f96bb 100644
--- a/doc/html/The-1d-Real_002ddata-DFT.html
+++ b/doc/html/The-1d-Real_002ddata-DFT.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>The 1d Real-data DFT - FFTW 3.3.2</title>
+<title>The 1d Real-data DFT - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="What-FFTW-Really-Computes.html#What-FFTW-Really-Computes" title="What FFTW Really Computes">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/The-Discrete-Hartley-Transform.html b/doc/html/The-Discrete-Hartley-Transform.html
index de4a47c..5ce0eb6 100644
--- a/doc/html/The-Discrete-Hartley-Transform.html
+++ b/doc/html/The-Discrete-Hartley-Transform.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>The Discrete Hartley Transform - FFTW 3.3.2</title>
+<title>The Discrete Hartley Transform - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="More-DFTs-of-Real-Data.html#More-DFTs-of-Real-Data" title="More DFTs of Real Data">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/The-Halfcomplex_002dformat-DFT.html b/doc/html/The-Halfcomplex_002dformat-DFT.html
index 0844c50..2c1b97b 100644
--- a/doc/html/The-Halfcomplex_002dformat-DFT.html
+++ b/doc/html/The-Halfcomplex_002dformat-DFT.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>The Halfcomplex-format DFT - FFTW 3.3.2</title>
+<title>The Halfcomplex-format DFT - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="More-DFTs-of-Real-Data.html#More-DFTs-of-Real-Data" title="More DFTs of Real Data">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Thread-safety.html b/doc/html/Thread-safety.html
index c1c37f8..2351f3c 100644
--- a/doc/html/Thread-safety.html
+++ b/doc/html/Thread-safety.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Thread safety - FFTW 3.3.2</title>
+<title>Thread safety - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW" title="Multi-threaded FFTW">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Transposed-distributions.html b/doc/html/Transposed-distributions.html
index e0c78b5..7c61c1f 100644
--- a/doc/html/Transposed-distributions.html
+++ b/doc/html/Transposed-distributions.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Transposed distributions - FFTW 3.3.2</title>
+<title>Transposed distributions - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="MPI-Data-Distribution.html#MPI-Data-Distribution" title="MPI Data Distribution">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Tutorial.html b/doc/html/Tutorial.html
index 3983030..a84d49d 100644
--- a/doc/html/Tutorial.html
+++ b/doc/html/Tutorial.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Tutorial - FFTW 3.3.2</title>
+<title>Tutorial - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Introduction.html#Introduction" title="Introduction">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Upgrading-from-FFTW-version-2.html b/doc/html/Upgrading-from-FFTW-version-2.html
index 4934cf4..964e08c 100644
--- a/doc/html/Upgrading-from-FFTW-version-2.html
+++ b/doc/html/Upgrading-from-FFTW-version-2.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Upgrading from FFTW version 2 - FFTW 3.3.2</title>
+<title>Upgrading from FFTW version 2 - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="prev" href="Calling-FFTW-from-Legacy-Fortran.html#Calling-FFTW-from-Legacy-Fortran" title="Calling FFTW from Legacy Fortran">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Usage-of-Multi_002dthreaded-FFTW.html b/doc/html/Usage-of-Multi_002dthreaded-FFTW.html
index fb05152..1226e76 100644
--- a/doc/html/Usage-of-Multi_002dthreaded-FFTW.html
+++ b/doc/html/Usage-of-Multi_002dthreaded-FFTW.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Usage of Multi-threaded FFTW - FFTW 3.3.2</title>
+<title>Usage of Multi-threaded FFTW - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW" title="Multi-threaded FFTW">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
@@ -100,7 +100,7 @@ disabled for subsequent plans.
    <p><a name="index-OpenMP-336"></a>With OpenMP, to configure FFTW to use all of the currently running
 OpenMP threads (set by <code>omp_set_num_threads(nthreads)</code> or by the
 <code>OMP_NUM_THREADS</code> environment variable), you can do:
-<code>fftw_plan_with_nthreads(omp_get_num_threads())</code>. (The ‘<samp><span class="samp">omp_</span></samp>’
+<code>fftw_plan_with_nthreads(omp_get_max_threads())</code>. (The ‘<samp><span class="samp">omp_</span></samp>’
 OpenMP functions are declared via <code>#include <omp.h></code>.)
 
    <p><a name="index-thread-safety-337"></a>Given a plan, you then execute it as usual with
diff --git a/doc/html/Using-MPI-Plans.html b/doc/html/Using-MPI-Plans.html
index c58f7c5..11dcc89 100644
--- a/doc/html/Using-MPI-Plans.html
+++ b/doc/html/Using-MPI-Plans.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Using MPI Plans - FFTW 3.3.2</title>
+<title>Using MPI Plans - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Using-Plans.html b/doc/html/Using-Plans.html
index 759c2a8..54dada2 100644
--- a/doc/html/Using-Plans.html
+++ b/doc/html/Using-Plans.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Using Plans - FFTW 3.3.2</title>
+<title>Using Plans - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/What-FFTW-Really-Computes.html b/doc/html/What-FFTW-Really-Computes.html
index 9298e5a..0ed93a2 100644
--- a/doc/html/What-FFTW-Really-Computes.html
+++ b/doc/html/What-FFTW-Really-Computes.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>What FFTW Really Computes - FFTW 3.3.2</title>
+<title>What FFTW Really Computes - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-Export.html b/doc/html/Wisdom-Export.html
index 6840a8e..3daa384 100644
--- a/doc/html/Wisdom-Export.html
+++ b/doc/html/Wisdom-Export.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom Export - FFTW 3.3.2</title>
+<title>Wisdom Export - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Wisdom.html#Wisdom" title="Wisdom">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-File-Export_002fImport-from-Fortran.html b/doc/html/Wisdom-File-Export_002fImport-from-Fortran.html
index c26c329..8ae20c1 100644
--- a/doc/html/Wisdom-File-Export_002fImport-from-Fortran.html
+++ b/doc/html/Wisdom-File-Export_002fImport-from-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom File Export/Import from Fortran - FFTW 3.3.2</title>
+<title>Wisdom File Export/Import from Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Accessing-the-wisdom-API-from-Fortran.html#Accessing-the-wisdom-API-from-Fortran" title="Accessing the wisdom API from Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-Generic-Export_002fImport-from-Fortran.html b/doc/html/Wisdom-Generic-Export_002fImport-from-Fortran.html
index e817ae8..b6d344d 100644
--- a/doc/html/Wisdom-Generic-Export_002fImport-from-Fortran.html
+++ b/doc/html/Wisdom-Generic-Export_002fImport-from-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom Generic Export/Import from Fortran - FFTW 3.3.2</title>
+<title>Wisdom Generic Export/Import from Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Accessing-the-wisdom-API-from-Fortran.html#Accessing-the-wisdom-API-from-Fortran" title="Accessing the wisdom API from Fortran">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-Import.html b/doc/html/Wisdom-Import.html
index 8cd0181..4f83632 100644
--- a/doc/html/Wisdom-Import.html
+++ b/doc/html/Wisdom-Import.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom Import - FFTW 3.3.2</title>
+<title>Wisdom Import - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Wisdom.html#Wisdom" title="Wisdom">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-String-Export_002fImport-from-Fortran.html b/doc/html/Wisdom-String-Export_002fImport-from-Fortran.html
index 9b97712..aab80ca 100644
--- a/doc/html/Wisdom-String-Export_002fImport-from-Fortran.html
+++ b/doc/html/Wisdom-String-Export_002fImport-from-Fortran.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom String Export/Import from Fortran - FFTW 3.3.2</title>
+<title>Wisdom String Export/Import from Fortran - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Accessing-the-wisdom-API-from-Fortran.html#Accessing-the-wisdom-API-from-Fortran" title="Accessing the wisdom API from Fortran">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-Utilities.html b/doc/html/Wisdom-Utilities.html
index 1138b55..0165aa2 100644
--- a/doc/html/Wisdom-Utilities.html
+++ b/doc/html/Wisdom-Utilities.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom Utilities - FFTW 3.3.2</title>
+<title>Wisdom Utilities - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Wisdom.html#Wisdom" title="Wisdom">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom-of-Fortran_003f.html b/doc/html/Wisdom-of-Fortran_003f.html
index 271e446..cb483c1 100644
--- a/doc/html/Wisdom-of-Fortran_003f.html
+++ b/doc/html/Wisdom-of-Fortran_003f.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom of Fortran? - FFTW 3.3.2</title>
+<title>Wisdom of Fortran? - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Calling-FFTW-from-Legacy-Fortran.html#Calling-FFTW-from-Legacy-Fortran" title="Calling FFTW from Legacy Fortran">
@@ -10,7 +10,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Wisdom.html b/doc/html/Wisdom.html
index ddc6209..0bfe3e0 100644
--- a/doc/html/Wisdom.html
+++ b/doc/html/Wisdom.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Wisdom - FFTW 3.3.2</title>
+<title>Wisdom - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/Words-of-Wisdom_002dSaving-Plans.html b/doc/html/Words-of-Wisdom_002dSaving-Plans.html
index d47cd07..c9b7cb8 100644
--- a/doc/html/Words-of-Wisdom_002dSaving-Plans.html
+++ b/doc/html/Words-of-Wisdom_002dSaving-Plans.html
@@ -1,8 +1,8 @@
 <html lang="en">
 <head>
-<title>Words of Wisdom-Saving Plans - FFTW 3.3.2</title>
+<title>Words of Wisdom-Saving Plans - FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="index.html#Top">
 <link rel="up" href="Other-Important-Topics.html#Other-Important-Topics" title="Other Important Topics">
@@ -11,7 +11,7 @@
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
diff --git a/doc/html/index.html b/doc/html/index.html
index 1aed0bf..f2b258c 100644
--- a/doc/html/index.html
+++ b/doc/html/index.html
@@ -1,14 +1,14 @@
 <html lang="en">
 <head>
-<title>FFTW 3.3.2</title>
+<title>FFTW 3.3.3</title>
 <meta http-equiv="Content-Type" content="text/html">
-<meta name="description" content="FFTW 3.3.2">
+<meta name="description" content="FFTW 3.3.3">
 <meta name="generator" content="makeinfo 4.13">
 <link title="Top" rel="start" href="#Top">
 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
 <!--
 This manual is for FFTW
-(version 3.3.2, 28 April 2012).
+(version 3.3.3, 25 November 2012).
 
 Copyright (C) 2003 Matteo Frigo.
 
@@ -42,7 +42,7 @@ Copyright (C) 2003 Massachusetts Institute of Technology.
 --></style>
 </head>
 <body>
-<h1 class="settitle">FFTW 3.3.2</h1>
+<h1 class="settitle">FFTW 3.3.3</h1>
 <div class="contents">
 <h2>Table of Contents</h2>
 <ul>
@@ -224,7 +224,7 @@ Up: <a rel="up" accesskey="u" href="../index.html#dir">(dir)</a>
 
 <p>Welcome to FFTW, the Fastest Fourier Transform in the West.  FFTW is a
 collection of fast C routines to compute the discrete Fourier transform. 
-This manual documents FFTW version 3.3.2.
+This manual documents FFTW version 3.3.3.
 
 <ul class="menu">
 <li><a accesskey="1" href="Introduction.html#Introduction">Introduction</a>
diff --git a/doc/intro.texi b/doc/intro.texi
index cc0e7a6..48dfe6f 100644
--- a/doc/intro.texi
+++ b/doc/intro.texi
@@ -21,11 +21,10 @@ transform (DFT) and various special cases thereof.
 @item  FFTW supports the SSE, SSE2, AVX, Altivec, and MIPS PS instruction
        sets.
 
- at item  FFTW @value{VERSION} includes parallel (multi-threaded) transforms
-  for shared-memory systems.
-  FFTW @value{VERSION} does not include distributed-memory parallel
-  transforms, but we plan to implement an MPI version soon.  (Meanwhile,
-  you can use the MPI implementation from FFTW 2.1.3.)
+ at item  FFTW includes parallel (multi-threaded) transforms
+       for shared-memory systems.
+ at item  Starting with version 3.3, FFTW includes distributed-memory parallel
+       transforms using MPI.
 @end itemize
 
 We assume herein that you are familiar with the properties and uses of
diff --git a/doc/rfftwnd.eps b/doc/rfftwnd.eps
index 0d457f8..cf80e04 100644
--- a/doc/rfftwnd.eps
+++ b/doc/rfftwnd.eps
@@ -1,7 +1,7 @@
 %!PS-Adobe-3.0 EPSF-3.0
 %%Title: ./rfftwnd.fig
 %%Creator: fig2dev Version 3.2 Patchlevel 5d
-%%CreationDate: Sat Apr 28 11:05:17 2012
+%%CreationDate: Sun Nov 25 07:42:44 2012
 %%BoundingBox: 0 0 270 405
 %Magnification: 0.7000
 %%EndComments
diff --git a/doc/rfftwnd.pdf b/doc/rfftwnd.pdf
index 1229aba..c8ce682 100644
Binary files a/doc/rfftwnd.pdf and b/doc/rfftwnd.pdf differ
diff --git a/doc/stamp-vti b/doc/stamp-vti
index b8568b0..0b04663 100644
--- a/doc/stamp-vti
+++ b/doc/stamp-vti
@@ -1,4 +1,4 @@
- at set UPDATED 28 April 2012
- at set UPDATED-MONTH April 2012
- at set EDITION 3.3.2
- at set VERSION 3.3.2
+ at set UPDATED 25 November 2012
+ at set UPDATED-MONTH November 2012
+ at set EDITION 3.3.3
+ at set VERSION 3.3.3
diff --git a/doc/texinfo.tex b/doc/texinfo.tex
index 2abda0f..85b68e7 100644
--- a/doc/texinfo.tex
+++ b/doc/texinfo.tex
@@ -3,11 +3,11 @@
 % Load plain if necessary, i.e., if running under initex.
 \expandafter\ifx\csname fmtname\endcsname\relax\input plain\fi
 %
-\def\texinfoversion{2012-01-03.09}
+\def\texinfoversion{2012-03-11.15}
 %
 % Copyright 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
 % 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
-% 2007, 2008, 2009, 2010, 2011 Free Software Foundation, Inc.
+% 2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation, Inc.
 %
 % This texinfo.tex file is free software: you can redistribute it and/or
 % modify it under the terms of the GNU General Public License as
@@ -887,7 +887,7 @@ where each line of input produces a line of output.}
 \def\popthisfilestack{\errthisfilestackempty}
 \def\errthisfilestackempty{\errmessage{Internal error:
   the stack of filenames is empty.}}
-
+%
 \def\thisfile{}
 
 % @center line
@@ -895,36 +895,46 @@ where each line of input produces a line of output.}
 %
 \parseargdef\center{%
   \ifhmode
-    \let\next\centerH
+    \let\centersub\centerH
   \else
-    \let\next\centerV
+    \let\centersub\centerV
   \fi
-  \next{\hfil \ignorespaces#1\unskip \hfil}%
+  \centersub{\hfil \ignorespaces#1\unskip \hfil}%
+  \let\centersub\relax % don't let the definition persist, just in case
 }
-\def\centerH#1{%
-  {%
-    \hfil\break
-    \advance\hsize by -\leftskip
-    \advance\hsize by -\rightskip
-    \line{#1}%
-    \break
-  }%
+\def\centerH#1{{%
+  \hfil\break
+  \advance\hsize by -\leftskip
+  \advance\hsize by -\rightskip
+  \line{#1}%
+  \break
+}}
+%
+\newcount\centerpenalty
+\def\centerV#1{%
+  % The idea here is the same as in \startdefun, \cartouche, etc.: if
+  % @center is the first thing after a section heading, we need to wipe
+  % out the negative parskip inserted by \sectionheading, but still
+  % prevent a page break here.
+  \centerpenalty = \lastpenalty
+  \ifnum\centerpenalty>10000 \vskip\parskip \fi
+  \ifnum\centerpenalty>9999 \penalty\centerpenalty \fi
+  \line{\kern\leftskip #1\kern\rightskip}%
 }
-\def\centerV#1{\line{\kern\leftskip #1\kern\rightskip}}
 
 % @sp n   outputs n lines of vertical space
-
+%
 \parseargdef\sp{\vskip #1\baselineskip}
 
 % @comment ...line which is ignored...
 % @c is the same as @comment
 % @ignore ... @end ignore  is another way to write a comment
-
+%
 \def\comment{\begingroup \catcode`\^^M=\other%
 \catcode`\@=\other \catcode`\{=\other \catcode`\}=\other%
 \commentxxx}
 {\catcode`\^^M=\other \gdef\commentxxx#1^^M{\endgroup}}
-
+%
 \let\c=\comment
 
 % @paragraphindent NCHARS
@@ -1173,8 +1183,8 @@ output) for that.)}
   %
   % #1 is image name, #2 width (might be empty/whitespace), #3 height (ditto).
   \def\dopdfimage#1#2#3{%
-    \def\imagewidth{#2}\setbox0 = \hbox{\ignorespaces #2}%
-    \def\imageheight{#3}\setbox2 = \hbox{\ignorespaces #3}%
+    \def\pdfimagewidth{#2}\setbox0 = \hbox{\ignorespaces #2}%
+    \def\pdfimageheight{#3}\setbox2 = \hbox{\ignorespaces #3}%
     %
     % pdftex (and the PDF format) support .pdf, .png, .jpg (among
     % others).  Let's try in that order, PDF first since if
@@ -1212,8 +1222,8 @@ output) for that.)}
     \else
       \immediate\pdfximage
     \fi
-      \ifdim \wd0 >0pt width \imagewidth \fi
-      \ifdim \wd2 >0pt height \imageheight \fi
+      \ifdim \wd0 >0pt width \pdfimagewidth \fi
+      \ifdim \wd2 >0pt height \pdfimageheight \fi
       \ifnum\pdftexversion<13
          #1.\pdfimgext
        \else
@@ -1362,7 +1372,13 @@ output) for that.)}
       \fi
     \fi
     \nextsp}
-  \def\getfilename#1{\filenamelength=0\expandafter\skipspaces#1|\relax}
+  \def\getfilename#1{%
+    \filenamelength=0
+    % If we don't expand the argument now, \skipspaces will get
+    % snagged on things like "@value{foo}".
+    \edef\temp{#1}%
+    \expandafter\skipspaces\temp|\relax
+  }
   \ifnum\pdftexversion < 14
     \let \startlink \pdfannotlink
   \else
@@ -2842,8 +2858,8 @@ end
 % 
 \def\outfmtnametex{tex}
 %
-\def\inlinefmt#1{\doinlinefmt #1,\finish}
-\def\doinlinefmt#1,#2,\finish{%
+\long\def\inlinefmt#1{\doinlinefmt #1,\finish}
+\long\def\doinlinefmt#1,#2,\finish{%
   \def\inlinefmtname{#1}%
   \ifx\inlinefmtname\outfmtnametex \ignorespaces #2\fi
 }
@@ -2855,8 +2871,8 @@ end
 % well use a command to get a left brace too.  We could re-use the
 % delimiter character idea from \verb, but it seems like overkill.
 % 
-\def\inlineraw{\tex \doinlineraw}
-\def\doinlineraw#1{\doinlinerawtwo #1,\finish}
+\long\def\inlineraw{\tex \doinlineraw}
+\long\def\doinlineraw#1{\doinlinerawtwo #1,\finish}
 \def\doinlinerawtwo#1,#2,\finish{%
   \def\inlinerawname{#1}%
   \ifx\inlinerawname\outfmtnametex \ignorespaces #2\fi
@@ -4732,10 +4748,9 @@ end
 %
 % ..., ready, GO:
 %
-\def\safewhatsit#1{%
-\ifhmode
+\def\safewhatsit#1{\ifhmode
   #1%
-\else
+ \else
   % \lastskip and \lastpenalty cannot both be nonzero simultaneously.
   \whatsitskip = \lastskip
   \edef\lastskipmacro{\the\lastskip}%
@@ -4759,7 +4774,6 @@ end
     % to re-insert the same penalty (values >10000 are used for various
     % signals); since we just inserted a non-discardable item, any
     % following glue (such as a \parskip) would be a breakpoint.  For example:
-    %
     %   @deffn deffn-whatever
     %   @vindex index-whatever
     %   Description.
@@ -4772,8 +4786,7 @@ end
     % (the whatsit from the \write), so we must insert a \nobreak.
     \nobreak\vskip\whatsitskip
   \fi
-\fi
-}
+\fi}
 
 % The index entry written in the file actually looks like
 %  \entry {sortstring}{page}{topic}
@@ -5876,14 +5889,15 @@ end
   %
   % We'll almost certainly start a paragraph next, so don't let that
   % glue accumulate.  (Not a breakpoint because it's preceded by a
-  % discardable item.)
+  % discardable item.)  However, when a paragraph is not started next
+  % (\startdefun, \cartouche, \center, etc.), this needs to be wiped out
+  % or the negative glue will cause weirdly wrong output, typically
+  % obscuring the section heading with something else.
   \vskip-\parskip
   %
-  % This is purely so the last item on the list is a known \penalty >
-  % 10000.  This is so \startdefun can avoid allowing breakpoints after
-  % section headings.  Otherwise, it would insert a valid breakpoint between:
-  %   @section sec-whatever
-  %   @deffn def-whatever
+  % This is so the last item on the main vertical list is a known
+  % \penalty > 10000, so \startdefun, etc., can recognize the situation
+  % and do the needful.
   \penalty 10001
 }
 
@@ -6303,7 +6317,7 @@ end
   % If this cartouche directly follows a sectioning command, we need the
   % \parskip glue (backspaced over by default) or the cartouche can
   % collide with the section heading.
-  \ifnum\lastpenalty>10000 \vskip\parskip \fi
+  \ifnum\lastpenalty>10000 \vskip\parskip \penalty\lastpenalty \fi
   %
   \vbox\bgroup
       \baselineskip=0pt\parskip=0pt\lineskip=0pt
@@ -7802,26 +7816,36 @@ end
 \def\pxref#1{\putwordsee{} \xrefX[#1,,,,,,,]}
 \def\xref#1{\putwordSee{} \xrefX[#1,,,,,,,]}
 \def\ref#1{\xrefX[#1,,,,,,,]}
+%
+\newbox\topbox
+\newbox\printedrefnamebox
+\newbox\printedmanualbox
+%
 \def\xrefX[#1,#2,#3,#4,#5,#6]{\begingroup
   \unsepspaces
-  \def\printedmanual{\ignorespaces #5}%
+  %
   \def\printedrefname{\ignorespaces #3}%
-  \setbox1=\hbox{\printedmanual\unskip}%
-  \setbox0=\hbox{\printedrefname\unskip}%
-  \ifdim \wd0 = 0pt
+  \setbox\printedrefnamebox = \hbox{\printedrefname\unskip}%
+  %
+  \def\printedmanual{\ignorespaces #5}%
+  \setbox\printedmanualbox  = \hbox{\printedmanual\unskip}%
+  %
+  % If the printed reference name (arg #3) was not explicitly given in
+  % the @xref, figure out what we want to use.
+  \ifdim \wd\printedrefnamebox = 0pt
     % No printed node name was explicitly given.
     \expandafter\ifx\csname SETxref-automatic-section-title\endcsname \relax
-      % Use the node name inside the square brackets.
+      % Not auto section-title: use node name inside the square brackets.
       \def\printedrefname{\ignorespaces #1}%
     \else
-      % Use the actual chapter/section title appear inside
-      % the square brackets.  Use the real section title if we have it.
-      \ifdim \wd1 > 0pt
-        % It is in another manual, so we don't have it.
+      % Auto section-title: use chapter/section title inside
+      % the square brackets if we have it.
+      \ifdim \wd\printedmanualbox > 0pt
+        % It is in another manual, so we don't have it; use node name.
         \def\printedrefname{\ignorespaces #1}%
       \else
         \ifhavexrefs
-          % We know the real title if we have the xref values.
+          % We (should) know the real title if we have the xref values.
           \def\printedrefname{\refx{#1-title}{}}%
         \else
           % Otherwise just copy the Info node name.
@@ -7835,6 +7859,7 @@ end
   \ifpdf
     {\indexnofonts
      \turnoffactive
+     \makevalueexpandable
      % This expands tokens, so do it after making catcode changes, so _
      % etc. don't get their TeX definitions.
      \getfilename{#4}%
@@ -7867,7 +7892,7 @@ end
   \iffloat\Xthisreftitle
     % If the user specified the print name (third arg) to the ref,
     % print it instead of our usual "Figure 1.2".
-    \ifdim\wd0 = 0pt
+    \ifdim\wd\printedrefnamebox = 0pt
       \refx{#1-snt}{}%
     \else
       \printedrefname
@@ -7875,21 +7900,46 @@ end
     %
     % if the user also gave the printed manual name (fifth arg), append
     % "in MANUALNAME".
-    \ifdim \wd1 > 0pt
+    \ifdim \wd\printedmanualbox > 0pt
       \space \putwordin{} \cite{\printedmanual}%
     \fi
   \else
     % node/anchor (non-float) references.
-    %
-    % If we use \unhbox0 and \unhbox1 to print the node names, TeX does not
-    % insert empty discretionaries after hyphens, which means that it will
-    % not find a line break at a hyphen in a node names.  Since some manuals
-    % are best written with fairly long node names, containing hyphens, this
-    % is a loss.  Therefore, we give the text of the node name again, so it
-    % is as if TeX is seeing it for the first time.
-    \ifdim \wd1 > 0pt
-      \putwordSection{} ``\printedrefname'' \putwordin{} \cite{\printedmanual}%
+    % 
+    % If we use \unhbox to print the node names, TeX does not insert
+    % empty discretionaries after hyphens, which means that it will not
+    % find a line break at a hyphen in a node names.  Since some manuals
+    % are best written with fairly long node names, containing hyphens,
+    % this is a loss.  Therefore, we give the text of the node name
+    % again, so it is as if TeX is seeing it for the first time.
+    % 
+    % Cross-manual reference.  Only include the "Section ``foo'' in" if
+    % the foo is neither missing or Top.  Thus, @xref{,,,foo,The Foo Manual}
+    % outputs simply "see The Foo Manual".
+    \ifdim \wd\printedmanualbox > 0pt
+      % What is the 7sp about?  The idea is that we also want to omit
+      % the Section part if we would be printing "Top", since they are
+      % clearly trying to refer to the whole manual.  But, this being
+      % TeX, we can't easily compare strings while ignoring the possible
+      % spaces before and after in the input.  By adding the arbitrary
+      % 7sp, we make it much less likely that a real node name would
+      % happen to have the same width as "Top" (e.g., in a monospaced font).
+      % I hope it will never happen in practice.
+      % 
+      % For the same basic reason, we retypeset the "Top" at every
+      % reference, since the current font is indeterminate.
+      % 
+      \setbox\topbox = \hbox{Top\kern7sp}%
+      \setbox2 = \hbox{\ignorespaces \printedrefname \unskip \kern7sp}%
+      \ifdim \wd2 > 7sp
+        \ifdim \wd2 = \wd\topbox \else
+          \putwordSection{} ``\printedrefname'' \putwordin{}\space
+        \fi
+      \fi
+      \cite{\printedmanual}%
     \else
+      % Reference in this manual.
+      %
       % _ (for example) has to be the character _ for the purposes of the
       % control sequence corresponding to the node, but it has to expand
       % into the usual \leavevmode...\vrule stuff for purposes of
@@ -7901,7 +7951,7 @@ end
        \setbox2 = \hbox{\ignorespaces \refx{#1-snt}{}}%
        \ifdim \wd2 > 0pt \refx{#1-snt}\space\fi
       }%
-      % output the `[mynode]' via a macro so it can be overridden.
+      % output the `[mynode]' via the macro below so it can be overridden.
       \xrefprintnodename\printedrefname
       %
       % But we always want a comma and a space:
@@ -8281,7 +8331,7 @@ end
   it from ftp://tug.org/tex/epsf.tex.}
 %
 \def\image#1{%
-  \ifx\epsfbox\thisiundefined
+  \ifx\epsfbox\thisisundefined
     \ifwarnednoepsf \else
       \errhelp = \noepsfhelp
       \errmessage{epsf.tex not found, images will be ignored}%
@@ -8305,6 +8355,13 @@ end
   % If the image is by itself, center it.
   \ifvmode
     \imagevmodetrue
+  \else \ifx\centersub\centerV
+    % for @center @image, we need a vbox so we can have our vertical space
+    \imagevmodetrue
+    \vbox\bgroup % vbox has better behavior than vtop herev
+  \fi\fi
+  %
+  \ifimagevmode
     \nobreak\medskip
     % Usually we'll have text after the image which will insert
     % \parskip glue, so insert it here too to equalize the space
@@ -8314,9 +8371,13 @@ end
   \fi
   %
   % Leave vertical mode so that indentation from an enclosing
-  % environment such as @quotation is respected.  On the other hand, if
-  % it's at the top level, we don't want the normal paragraph indentation.
-  \noindent
+  %  environment such as @quotation is respected.
+  % However, if we're at the top level, we don't want the
+  %  normal paragraph indentation.
+  % On the other hand, if we are in the case of @center @image, we don't
+  %  want to start a paragraph, which will create a hsize-width box and
+  %  eradicate the centering.
+  \ifx\centersub\centerV\else \noindent \fi
   %
   % Output the image.
   \ifpdf
@@ -8328,7 +8389,10 @@ end
     \epsfbox{#1.eps}%
   \fi
   %
-  \ifimagevmode \medskip \fi  % space after the standalone image
+  \ifimagevmode
+    \medskip  % space after a standalone image
+  \fi  
+  \ifx\centersub\centerV \egroup \fi
 \endgroup}
 
 
diff --git a/doc/threads.texi b/doc/threads.texi
index 68d9e7f..e5047c3 100644
--- a/doc/threads.texi
+++ b/doc/threads.texi
@@ -123,7 +123,7 @@ disabled for subsequent plans.
 With OpenMP, to configure FFTW to use all of the currently running
 OpenMP threads (set by @code{omp_set_num_threads(nthreads)} or by the
 @code{OMP_NUM_THREADS} environment variable), you can do:
- at code{fftw_plan_with_nthreads(omp_get_num_threads())}. (The @samp{omp_}
+ at code{fftw_plan_with_nthreads(omp_get_max_threads())}. (The @samp{omp_}
 OpenMP functions are declared via @code{#include <omp.h>}.)
 
 @cindex thread safety
diff --git a/doc/tutorial.texi b/doc/tutorial.texi
index 24a6d57..5763275 100644
--- a/doc/tutorial.texi
+++ b/doc/tutorial.texi
@@ -421,11 +421,16 @@ over half of the output of the corresponding complex DFT.  (The division
 is rounded down.)  The ordering of the data is otherwise exactly the
 same as in the complex-DFT case.
 
-Since the complex data is slightly larger than the real data, some
-complications arise for in-place transforms.  In this case, the final
-dimension of the real data must be padded with extra values to
-accommodate the size of the complex data---two values if the last
-dimension is even and one if it is odd.
+For out-of-place transforms, this is the end of the story: the real
+data is stored as a row-major array of size @ndims and the complex
+data is stored as a row-major array of size @ndimshalf{}.
+
+For in-place transforms, however, extra padding of the real-data array
+is necessary because the complex array is larger than the real array,
+and the two arrays share the same memory locations.  Thus, for
+in-place transforms, the final dimension of the real-data array must
+be padded with extra values to accommodate the size of the complex
+data---two values if the last dimension is even and one if it is odd.
 @cindex padding
 That is, the last dimension of the real data must physically contain
 @tex
diff --git a/doc/version.texi b/doc/version.texi
index b8568b0..0b04663 100644
--- a/doc/version.texi
+++ b/doc/version.texi
@@ -1,4 +1,4 @@
- at set UPDATED 28 April 2012
- at set UPDATED-MONTH April 2012
- at set EDITION 3.3.2
- at set VERSION 3.3.2
+ at set UPDATED 25 November 2012
+ at set UPDATED-MONTH November 2012
+ at set EDITION 3.3.3
+ at set VERSION 3.3.3
diff --git a/genfft/Makefile.in b/genfft/Makefile.in
index 92f38c6..05d57b2 100644
--- a/genfft/Makefile.in
+++ b/genfft/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -17,6 +17,23 @@
 
 # this makefile requires GNU make.
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -57,6 +74,11 @@ CONFIG_CLEAN_FILES =
 CONFIG_CLEAN_VPATH_FILES =
 SOURCES =
 DIST_SOURCES =
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 ACLOCAL = @ACLOCAL@
 ALLOCA = @ALLOCA@
diff --git a/genfft/gen_hc2c.ml b/genfft/gen_hc2c.ml
index 788ffba..14bf408 100644
--- a/genfft/gen_hc2c.ml
+++ b/genfft/gen_hc2c.ml
@@ -138,7 +138,7 @@ let generate n =
 	     Expr_assign (vaim, CPlus [vaim; CUminus (byvl vms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       byvl (Integer nt)]);
-	     make_volatile_stride (CVar rs)
+	     make_volatile_stride (4*n) (CVar rs)
 	   ],
 	  Asch asch)])
   in
diff --git a/genfft/gen_hc2cdft.ml b/genfft/gen_hc2cdft.ml
index 0fcce6e..268b85b 100644
--- a/genfft/gen_hc2cdft.ml
+++ b/genfft/gen_hc2cdft.ml
@@ -159,7 +159,7 @@ let generate n =
 	     Expr_assign (vaim, CPlus [vaim; CUminus (byvl vms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       byvl (Integer nt)]);
-	     make_volatile_stride (CVar rs)
+	     make_volatile_stride (4*n) (CVar rs)
 	   ],
 	  Asch asch)]
     )
diff --git a/genfft/gen_hc2cdft_c.ml b/genfft/gen_hc2cdft_c.ml
index c83f716..1a4e9ae 100644
--- a/genfft/gen_hc2cdft_c.ml
+++ b/genfft/gen_hc2cdft_c.ml
@@ -171,7 +171,7 @@ let generate n =
 	     Expr_assign (vaim, CPlus [vaim; CUminus (byvl vms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       bytwvl (Integer nt)]);
-	     make_volatile_stride (CVar rs)
+	     make_volatile_stride (4*n) (CVar rs)
 	   ],
 	  Asch asch)]
     )
diff --git a/genfft/gen_hc2hc.ml b/genfft/gen_hc2hc.ml
index 47288be..192caa5 100644
--- a/genfft/gen_hc2hc.ml
+++ b/genfft/gen_hc2hc.ml
@@ -124,7 +124,7 @@ let generate n =
 			  CPlus [viioarray; CUminus (byvl vms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       byvl (Integer nt)]);
-	     make_volatile_stride (CVar rs)
+	     make_volatile_stride (2*n) (CVar rs)
 	   ],
 	  Asch asch)])
   in
diff --git a/genfft/gen_notw.ml b/genfft/gen_notw.ml
index 461eb93..69bc940 100644
--- a/genfft/gen_notw.ml
+++ b/genfft/gen_notw.ml
@@ -122,8 +122,8 @@ let generate n =
 					       byvl (CVar sovs)]);
 	     Expr_assign (CVar ioarray, CPlus [CVar ioarray; 
 					       byvl (CVar sovs)]);
-	     make_volatile_stride (CVar istride);
-	     make_volatile_stride (CVar ostride)
+	     make_volatile_stride (4*n) (CVar istride);
+	     make_volatile_stride (4*n) (CVar ostride)
 	   ],
 	  Asch annot)
    ])
diff --git a/genfft/gen_notw_c.ml b/genfft/gen_notw_c.ml
index 3b5853f..86609da 100644
--- a/genfft/gen_notw_c.ml
+++ b/genfft/gen_notw_c.ml
@@ -117,8 +117,8 @@ let generate n =
 					       byvl (CVar sivs)]);
 	     Expr_assign (CVar roarray, CPlus [CVar roarray; 
 					       byvl (CVar sovs)]);
-	     make_volatile_stride (CVar istride);
-	     make_volatile_stride (CVar ostride)
+	     make_volatile_stride (2*n) (CVar istride);
+	     make_volatile_stride (2*n) (CVar ostride)
 	   ],
 	  Asch annot);
    ])
diff --git a/genfft/gen_r2cb.ml b/genfft/gen_r2cb.ml
index f1335fc..074b3ab 100644
--- a/genfft/gen_r2cb.ml
+++ b/genfft/gen_r2cb.ml
@@ -121,9 +121,9 @@ let generate n =
 	     Expr_assign (CVar ar1, CPlus [CVar ar1; CVar sovs]);
 	     Expr_assign (CVar acr, CPlus [CVar acr; CVar sivs]);
 	     Expr_assign (CVar aci, CPlus [CVar aci; CVar sivs]);
-	     make_volatile_stride (CVar rs);
-	     make_volatile_stride (CVar csr);
-	     make_volatile_stride (CVar csi)
+	     make_volatile_stride (4*n) (CVar rs);
+	     make_volatile_stride (4*n) (CVar csr);
+	     make_volatile_stride (4*n) (CVar csi)
 	   ],
 	  Asch annot)
    ])
diff --git a/genfft/gen_r2cf.ml b/genfft/gen_r2cf.ml
index 47d1052..a23c377 100644
--- a/genfft/gen_r2cf.ml
+++ b/genfft/gen_r2cf.ml
@@ -118,9 +118,9 @@ let generate n =
 	     Expr_assign (CVar ar1, CPlus [CVar ar1; CVar sivs]);
 	     Expr_assign (CVar acr, CPlus [CVar acr; CVar sovs]);
 	     Expr_assign (CVar aci, CPlus [CVar aci; CVar sovs]);
-	     make_volatile_stride (CVar rs);
-	     make_volatile_stride (CVar csr);
-	     make_volatile_stride (CVar csi)
+	     make_volatile_stride (4*n) (CVar rs);
+	     make_volatile_stride (4*n) (CVar csr);
+	     make_volatile_stride (4*n) (CVar csi)
 	   ],
 	  Asch annot)
    ])
diff --git a/genfft/gen_r2r.ml b/genfft/gen_r2r.ml
index b926359..a63c910 100644
--- a/genfft/gen_r2r.ml
+++ b/genfft/gen_r2r.ml
@@ -197,8 +197,8 @@ let generate n mode =
 	    [Expr_assign (CVar i, CPlus [CVar i; CUminus (Integer 1)]);
 	     Expr_assign (CVar iarray, CPlus [CVar iarray; CVar sivs]);
 	     Expr_assign (CVar oarray, CPlus [CVar oarray; CVar sovs]);
-	     make_volatile_stride (CVar istride);
-	     make_volatile_stride (CVar ostride)
+	     make_volatile_stride (2*n) (CVar istride);
+	     make_volatile_stride (2*n) (CVar ostride)
 	   ],
 	  Asch annot)
    ])
diff --git a/genfft/gen_twiddle.ml b/genfft/gen_twiddle.ml
index 35bdf7b..6a8dc16 100644
--- a/genfft/gen_twiddle.ml
+++ b/genfft/gen_twiddle.ml
@@ -108,7 +108,7 @@ let generate n =
 						byvl (CVar sms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       byvl (Integer nt)]);
-	     make_volatile_stride (CVar rs)
+	     make_volatile_stride (2*n) (CVar rs)
 	    ],
 	  Asch annot)])
   in
diff --git a/genfft/gen_twiddle_c.ml b/genfft/gen_twiddle_c.ml
index 1896d35..6fea68f 100644
--- a/genfft/gen_twiddle_c.ml
+++ b/genfft/gen_twiddle_c.ml
@@ -112,7 +112,7 @@ let generate n =
 						byvl (CVar sms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       bytwvl (Integer nt)]);
-	     make_volatile_stride (CVar rs)
+	     make_volatile_stride n (CVar rs)
 	    ],
 	  Asch annot)])
   in
diff --git a/genfft/gen_twidsq.ml b/genfft/gen_twidsq.ml
index 7d58022..aee6b01 100644
--- a/genfft/gen_twidsq.ml
+++ b/genfft/gen_twidsq.ml
@@ -123,8 +123,8 @@ let generate n =
 	     Expr_assign (CVar rioarray, CPlus [CVar rioarray; CVar ms]);
 	     Expr_assign (CVar iioarray, CPlus [CVar iioarray; CVar ms]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; Integer nt]);
-	     make_volatile_stride (CVar rs);
-	     make_volatile_stride (CVar vs)
+	     make_volatile_stride (2*n) (CVar rs);
+	     make_volatile_stride (2*0) (CVar vs)
 	   ],
 	  Asch annot)]) in
 
diff --git a/genfft/gen_twidsq_c.ml b/genfft/gen_twidsq_c.ml
index 2a443ec..ecbe3f7 100644
--- a/genfft/gen_twidsq_c.ml
+++ b/genfft/gen_twidsq_c.ml
@@ -133,8 +133,8 @@ let generate n =
 						byvl (CVar sms)]);
 	     Expr_assign (CVar twarray, CPlus [CVar twarray; 
 					       bytwvl (Integer nt)]);
-	     make_volatile_stride (CVar rs);
-	     make_volatile_stride (CVar vs)
+	     make_volatile_stride (2*n) (CVar rs);
+	     make_volatile_stride (2*n) (CVar vs)
 	   ],
 	  Asch annot)]) in
 
diff --git a/genfft/genutil.ml b/genfft/genutil.ml
index 96c6880..b8f70a5 100644
--- a/genfft/genutil.ml
+++ b/genfft/genutil.ml
@@ -324,4 +324,5 @@ let twinstr_to_string vl x =
   else
     Twiddle.twinstr_to_c_string x
 
-let make_volatile_stride x = C.CCall ("MAKE_VOLATILE_STRIDE", x)
+let make_volatile_stride n x = 
+  C.CCall ("MAKE_VOLATILE_STRIDE", C.Comma((C.Integer n), x))
diff --git a/kernel/Makefile.in b/kernel/Makefile.in
index 4b24357..f8d4940 100644
--- a/kernel/Makefile.in
+++ b/kernel/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -80,6 +97,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libkernel_la_SOURCES)
 DIST_SOURCES = $(libkernel_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/kernel/buffered.c b/kernel/buffered.c
index 5c26779..186ea78 100644
--- a/kernel/buffered.c
+++ b/kernel/buffered.c
@@ -74,6 +74,7 @@ int X(nbuf_redundant)(INT n, INT vl, int which,
 		      const INT *maxnbuf, int nmaxnbuf)
 {
      int i;
+     (void)nmaxnbuf; /* UNUSED */
      for (i = 0; i < which; ++i)
 	  if (X(nbuf)(n, vl, maxnbuf[i]) == X(nbuf)(n, vl, maxnbuf[which]))
 	       return 1;
diff --git a/kernel/ifftw.h b/kernel/ifftw.h
index 932767c..8245858 100644
--- a/kernel/ifftw.h
+++ b/kernel/ifftw.h
@@ -741,6 +741,7 @@ struct planner_s {
      double (*cost_hook)(const problem *p, double t, cost_kind k);
      int (*wisdom_ok_hook)(const problem *p, flags_t flags);
      void (*nowisdom_hook)(const problem *p);
+     wisdom_state_t (*bogosity_hook)(wisdom_state_t state, const problem *p);
 
      /* solver descriptors */
      slvdesc *slvdescs;
@@ -827,7 +828,7 @@ extern stride X(mkstride)(INT n, INT s);
 void X(stride_destroy)(stride p);
 /* hackery to prevent the compiler from copying the strides array
    onto the stack */
-#define MAKE_VOLATILE_STRIDE(x) (x) = (x) + X(an_INT_guaranteed_to_be_zero)
+#define MAKE_VOLATILE_STRIDE(nptr, x) (x) = (x) + X(an_INT_guaranteed_to_be_zero)
 #else
 
 typedef INT stride;
@@ -840,9 +841,26 @@ typedef INT stride;
 #define fftwl_stride_destroy(p) ((void) p)
 
 /* hackery to prevent the compiler from ``optimizing'' induction
-   variables in codelet loops. */
-#define MAKE_VOLATILE_STRIDE(x) (x) = (x) ^ X(an_INT_guaranteed_to_be_zero)
-
+   variables in codelet loops.  The problem is that for each K and for
+   each expression of the form P[I + STRIDE * K] in a loop, most
+   compilers will try to lift an induction variable PK := &P[I + STRIDE * K].
+   For large values of K this behavior overflows the
+   register set, which is likely worse than doing the index computation
+   in the first place.
+
+   If we guess that there are more than
+   ESTIMATED_AVAILABLE_INDEX_REGISTERS such pointers, we deliberately confuse
+   the compiler by setting STRIDE ^= ZERO, where ZERO is a value guaranteed to
+   be 0, but the compiler does not know this. 
+
+   16 registers ought to be enough for anybody, or so the amd64 and ARM ISA's
+   seem to imply.
+*/
+#define ESTIMATED_AVAILABLE_INDEX_REGISTERS 16
+#define MAKE_VOLATILE_STRIDE(nptr, x)                   \
+     (nptr <= ESTIMATED_AVAILABLE_INDEX_REGISTERS ?     \
+        0 :                                             \
+      ((x) = (x) ^ X(an_INT_guaranteed_to_be_zero)))
 #endif /* PRECOMPUTE_ARRAY_INDICES */
 
 /*-----------------------------------------------------------------------*/
diff --git a/kernel/planner.c b/kernel/planner.c
index 2958981..172c398 100644
--- a/kernel/planner.c
+++ b/kernel/planner.c
@@ -614,9 +614,11 @@ static plan *search(planner *ego, const problem *p, unsigned *slvndx,
      return pln;
 }
 
-#define CHECK_FOR_BOGOSITY			\
-     if (ego->wisdom_state == WISDOM_IS_BOGUS)	\
-	  goto wisdom_is_bogus
+#define CHECK_FOR_BOGOSITY						\
+     if ((ego->bogosity_hook ?						\
+	  (ego->wisdom_state = ego->bogosity_hook(ego->wisdom_state, p)) \
+	  : ego->wisdom_state) == WISDOM_IS_BOGUS)			\
+	  goto wisdom_is_bogus;
 
 static plan *mkplan(planner *ego, const problem *p)
 {
@@ -923,6 +925,7 @@ planner *X(mkplanner)(void)
      p->cost_hook = 0;
      p->wisdom_ok_hook = 0;
      p->nowisdom_hook = 0;
+     p->bogosity_hook = 0;
      p->cur_reg_nam = 0;
      p->wisdom_state = WISDOM_NORMAL;
 
diff --git a/kernel/timer.c b/kernel/timer.c
index a1e5527..1d713b4 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -77,7 +77,7 @@ static double elapsed_since(crude_time t0)
      return elapsed_sec(t1, t0);
 }
 
-#  define TIME_MIN_SEC 1.0e-2 /* from fftw2 */
+#  define TIME_MIN_SEC 1.0e-3
 
 #else /* !HAVE_GETTIMEOFDAY */
 
diff --git a/libbench2/Makefile.in b/libbench2/Makefile.in
index 00d50ca..e9095aa 100644
--- a/libbench2/Makefile.in
+++ b/libbench2/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -87,6 +104,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(libbench2_a_SOURCES)
 DIST_SOURCES = $(libbench2_a_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/ltmain.sh b/ltmain.sh
index c7d06c3..33f642a 100644
--- a/ltmain.sh
+++ b/ltmain.sh
@@ -70,7 +70,7 @@
 #         compiler:		$LTCC
 #         compiler flags:		$LTCFLAGS
 #         linker:		$LD (gnu? $with_gnu_ld)
-#         $progname:	(GNU libtool) 2.4.2 Debian-2.4.2-1
+#         $progname:	(GNU libtool) 2.4.2 Debian-2.4.2-1.1
 #         automake:	$automake_version
 #         autoconf:	$autoconf_version
 #
@@ -80,7 +80,7 @@
 
 PROGRAM=libtool
 PACKAGE=libtool
-VERSION="2.4.2 Debian-2.4.2-1"
+VERSION="2.4.2 Debian-2.4.2-1.1"
 TIMESTAMP=""
 package_revision=1.3337
 
diff --git a/m4/Makefile.in b/m4/Makefile.in
index 678dca1..28b3eb0 100644
--- a/m4/Makefile.in
+++ b/m4/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -15,6 +15,23 @@
 
 @SET_MAKE@
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -55,6 +72,11 @@ CONFIG_CLEAN_FILES =
 CONFIG_CLEAN_VPATH_FILES =
 SOURCES =
 DIST_SOURCES =
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 ACLOCAL = @ACLOCAL@
 ALLOCA = @ALLOCA@
diff --git a/m4/acx_mpi.m4 b/m4/acx_mpi.m4
index 54cdc0a..36fcabd 100644
--- a/m4/acx_mpi.m4
+++ b/m4/acx_mpi.m4
@@ -63,10 +63,10 @@ AC_LANG_CASE([C], [
 ])
 
 if test x = x"$MPILIBS"; then
-	AC_LANG_CASE([C], [AC_CHECK_FUNC(MPI_Init, [MPILIBS=""])],
-		[C++], [AC_CHECK_FUNC(MPI_Init, [MPILIBS=""])],
+	AC_LANG_CASE([C], [AC_CHECK_FUNC(MPI_Init, [MPILIBS=" "])],
+		[C++], [AC_CHECK_FUNC(MPI_Init, [MPILIBS=" "])],
 		[Fortran 77], [AC_MSG_CHECKING([for MPI_Init])
-			AC_TRY_LINK([],[      call MPI_Init], [MPILIBS=""
+			AC_TRY_LINK([],[      call MPI_Init], [MPILIBS=" "
 				AC_MSG_RESULT(yes)], [AC_MSG_RESULT(no)])])
 fi
 if test x = x"$MPILIBS"; then
diff --git a/m4/libtool.m4 b/m4/libtool.m4
index 828104c..534d1cc 100644
--- a/m4/libtool.m4
+++ b/m4/libtool.m4
@@ -2512,17 +2512,6 @@ freebsd* | dragonfly*)
   esac
   ;;
 
-gnu*)
-  version_type=linux # correct to gnu/linux during the next big refactor
-  need_lib_prefix=no
-  need_version=no
-  library_names_spec='${libname}${release}${shared_ext}$versuffix ${libname}${release}${shared_ext}${major} ${libname}${shared_ext}'
-  soname_spec='${libname}${release}${shared_ext}$major'
-  shlibpath_var=LD_LIBRARY_PATH
-  shlibpath_overrides_runpath=no
-  hardcode_into_libs=yes
-  ;;
-
 haiku*)
   version_type=linux # correct to gnu/linux during the next big refactor
   need_lib_prefix=no
@@ -2639,7 +2628,7 @@ linux*oldld* | linux*aout* | linux*coff*)
   ;;
 
 # This must be glibc/ELF.
-linux* | k*bsd*-gnu | kopensolaris*-gnu)
+linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
   version_type=linux # correct to gnu/linux during the next big refactor
   need_lib_prefix=no
   need_version=no
@@ -3255,10 +3244,6 @@ freebsd* | dragonfly*)
   fi
   ;;
 
-gnu*)
-  lt_cv_deplibs_check_method=pass_all
-  ;;
-
 haiku*)
   lt_cv_deplibs_check_method=pass_all
   ;;
@@ -3297,7 +3282,7 @@ irix5* | irix6* | nonstopux*)
   ;;
 
 # This must be glibc/ELF.
-linux* | k*bsd*-gnu | kopensolaris*-gnu)
+linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
   lt_cv_deplibs_check_method=pass_all
   ;;
 
@@ -4049,7 +4034,7 @@ m4_if([$1], [CXX], [
 	    ;;
 	esac
 	;;
-      linux* | k*bsd*-gnu | kopensolaris*-gnu)
+      linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
 	case $cc_basename in
 	  KCC*)
 	    # KAI C++ Compiler
@@ -4348,7 +4333,7 @@ m4_if([$1], [CXX], [
       _LT_TAGVAR(lt_prog_compiler_static, $1)='-non_shared'
       ;;
 
-    linux* | k*bsd*-gnu | kopensolaris*-gnu)
+    linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
       case $cc_basename in
       # old Intel for x86_64 which still supported -KPIC.
       ecc*)
@@ -6241,9 +6226,6 @@ if test "$_lt_caught_CXX_error" != yes; then
         _LT_TAGVAR(ld_shlibs, $1)=yes
         ;;
 
-      gnu*)
-        ;;
-
       haiku*)
         _LT_TAGVAR(archive_cmds, $1)='$CC -shared $libobjs $deplibs $compiler_flags ${wl}-soname $wl$soname -o $lib'
         _LT_TAGVAR(link_all_deplibs, $1)=yes
@@ -6405,7 +6387,7 @@ if test "$_lt_caught_CXX_error" != yes; then
         _LT_TAGVAR(inherit_rpath, $1)=yes
         ;;
 
-      linux* | k*bsd*-gnu | kopensolaris*-gnu)
+      linux* | k*bsd*-gnu | kopensolaris*-gnu | gnu*)
         case $cc_basename in
           KCC*)
 	    # Kuck and Associates, Inc. (KAI) C++ Compiler
diff --git a/mpi/Makefile.in b/mpi/Makefile.in
index 94f02b1..9a0fafd 100644
--- a/mpi/Makefile.in
+++ b/mpi/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -18,6 +18,23 @@
 
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -140,6 +157,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 SOURCES = $(libfftw3 at PREC_SUFFIX@_mpi_la_SOURCES) $(mpi_bench_SOURCES)
 DIST_SOURCES = $(libfftw3 at PREC_SUFFIX@_mpi_la_SOURCES) \
 	$(mpi_bench_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 am__include_HEADERS_DIST = fftw3-mpi.h
 HEADERS = $(include_HEADERS) $(nodist_include_HEADERS)
 ETAGS = etags
@@ -353,7 +375,6 @@ $(ACLOCAL_M4): @MAINTAINER_MODE_TRUE@ $(am__aclocal_m4_deps)
 $(am__aclocal_m4_deps):
 install-libLTLIBRARIES: $(lib_LTLIBRARIES)
 	@$(NORMAL_INSTALL)
-	test -z "$(libdir)" || $(MKDIR_P) "$(DESTDIR)$(libdir)"
 	@list='$(lib_LTLIBRARIES)'; test -n "$(libdir)" || list=; \
 	list2=; for p in $$list; do \
 	  if test -f $$p; then \
@@ -361,6 +382,8 @@ install-libLTLIBRARIES: $(lib_LTLIBRARIES)
 	  else :; fi; \
 	done; \
 	test -z "$$list2" || { \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(libdir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(libdir)" || exit 1; \
 	  echo " $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL) $(INSTALL_STRIP_FLAG) $$list2 '$(DESTDIR)$(libdir)'"; \
 	  $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL) $(INSTALL_STRIP_FLAG) $$list2 "$(DESTDIR)$(libdir)"; \
 	}
@@ -509,8 +532,11 @@ clean-libtool:
 	-rm -rf .libs _libs
 install-includeHEADERS: $(include_HEADERS)
 	@$(NORMAL_INSTALL)
-	test -z "$(includedir)" || $(MKDIR_P) "$(DESTDIR)$(includedir)"
 	@list='$(include_HEADERS)'; test -n "$(includedir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(includedir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(includedir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
@@ -527,8 +553,11 @@ uninstall-includeHEADERS:
 	dir='$(DESTDIR)$(includedir)'; $(am__uninstall_files_from_dir)
 install-nodist_includeHEADERS: $(nodist_include_HEADERS)
 	@$(NORMAL_INSTALL)
-	test -z "$(includedir)" || $(MKDIR_P) "$(DESTDIR)$(includedir)"
 	@list='$(nodist_include_HEADERS)'; test -n "$(includedir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(includedir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(includedir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; \
diff --git a/mpi/api.c b/mpi/api.c
index ee1d51a..11bd8d4 100644
--- a/mpi/api.c
+++ b/mpi/api.c
@@ -118,6 +118,17 @@ static void nowisdom_hook(const problem *p)
      XM(any_true)(1, comm); /* signal nowisdom to any wisdom_ok_hook */
 }
 
+/* needed to synchronize planner bogosity flag, in case non-MPI problems
+   on a subset of processes encountered bogus wisdom */
+static wisdom_state_t bogosity_hook(wisdom_state_t state, const problem *p)
+{
+     MPI_Comm comm = problem_comm(p);
+     if (comm != MPI_COMM_NULL /* an MPI problem */
+	 && XM(any_true)(state == WISDOM_IS_BOGUS, comm)) /* bogus somewhere */
+	  return WISDOM_IS_BOGUS;
+     return state;
+}
+
 void XM(init)(void)
 {
      if (!mpi_inited) {
@@ -125,6 +136,7 @@ void XM(init)(void)
 	  plnr->cost_hook = cost_hook;
 	  plnr->wisdom_ok_hook = wisdom_ok_hook;
 	  plnr->nowisdom_hook = nowisdom_hook;
+	  plnr->bogosity_hook = bogosity_hook;
           XM(conf_standard)(plnr);
 	  mpi_inited = 1;	  
      }
diff --git a/mpi/transpose-alltoall.c b/mpi/transpose-alltoall.c
index 8e0f3d0..71e7f77 100644
--- a/mpi/transpose-alltoall.c
+++ b/mpi/transpose-alltoall.c
@@ -216,6 +216,8 @@ static plan *mkplan(const solver *ego_, const problem *p_, planner *plnr)
 	  INT db, dbt; /* destination block sizes */
 	  db = XM(block)(p->nx, p->block, pe);
 	  dbt = XM(block)(p->ny, p->tblock, pe);
+	  if (db != p->block || dbt != p->tblock)
+	       equal_blocks = 0;
 
 	  /* MPI requires type "int" here; apparently it
 	     has no 64-bit API?  Grrr. */
@@ -223,9 +225,6 @@ static plan *mkplan(const solver *ego_, const problem *p_, planner *plnr)
 	  sbo[pe] = (int) (pe * (b * p->tblock) * vn);
 	  rbs[pe] = (int) (db * bt * vn);
 	  rbo[pe] = (int) (pe * (p->block * bt) * vn);
-	  if (sbs[pe] != (b * p->tblock) * vn
-	      || rbs[pe] != (p->block * bt) * vn)
-	       equal_blocks = 0;
      }
      pln->send_block_sizes = sbs;
      pln->send_block_offsets = sbo;
diff --git a/rdft/Makefile.in b/rdft/Makefile.in
index 6f77047..3220ace 100644
--- a/rdft/Makefile.in
+++ b/rdft/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -90,6 +107,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -571,13 +593,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/rdft/rank0.c b/rdft/rank0.c
index 685ecc7..ce2d40e 100644
--- a/rdft/rank0.c
+++ b/rdft/rank0.c
@@ -331,6 +331,7 @@ static plan *mkplan(const solver *ego_, const problem *p_, planner *plnr)
      pln = MKPLAN_RDFT(P, &padt, ego->apply);
 
      retval = fill_iodim(pln, p);
+     (void)retval; /* UNUSED unless DEBUG */
      A(retval);
      A(pln->vl > 0); /* because FINITE_RNK(p->vecsz->rnk) holds */
      pln->nam = ego->nam;
diff --git a/rdft/scalar/Makefile.in b/rdft/scalar/Makefile.in
index f4b0558..af72052 100644
--- a/rdft/scalar/Makefile.in
+++ b/rdft/scalar/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -80,6 +97,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -514,13 +536,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/rdft/scalar/r2cb/Makefile.in b/rdft/scalar/r2cb/Makefile.in
index 489108d..29c396e 100644
--- a/rdft/scalar/r2cb/Makefile.in
+++ b/rdft/scalar/r2cb/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -32,6 +32,23 @@
 # Most of this file requires GNU make .
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -117,6 +134,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_scalar_r2cb_la_SOURCES)
 DIST_SOURCES = $(librdft_scalar_r2cb_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/rdft/scalar/r2cb/hb2_16.c b/rdft/scalar/r2cb/hb2_16.c
index 24b0f8c..32ee150 100644
--- a/rdft/scalar/r2cb/hb2_16.c
+++ b/rdft/scalar/r2cb/hb2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:24 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hb2_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E Tv, TB, TF, Ty, T1J, T1O, T1N, T1K;
 	       {
 		    E Tw, T2z, T2C, Tx, T3f, T3l, T2F, T3r, Tz;
@@ -445,7 +445,7 @@ static void hb2_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E Tv, Ty, T1l, T1n, T1p, T1t, T27, T25, Tz, Tw, TB, T21, T1P, T1H, T1X;
 	       E T17, T1L, T1N, T1v, T1w, T1x, T1B, T2F, T2T, T2b, T2R, T3j, T3x, T35, T3t;
 	       {
diff --git a/rdft/scalar/r2cb/hb2_20.c b/rdft/scalar/r2cb/hb2_20.c
index e56bd27..eaf4a13 100644
--- a/rdft/scalar/r2cb/hb2_20.c
+++ b/rdft/scalar/r2cb/hb2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:12 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hb2_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T1S, T1O, T1s, TI, T24, T1Y, T2g, T2k, TS, TR, T1I, T26, T1o, T20, T1F;
 	       E T25, TT, T1Z;
 	       {
@@ -600,7 +600,7 @@ static void hb2_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E TD, TG, TE, TH, TJ, T1t, T27, T25, T1T, T1R, T1V, T2j, T2Z, T21, T2X;
 	       E T2T, T2n, T2P, T3V, T41, T3R, T3X, T29, T2c, T4H, T4L, T1L, T1M, T1N, T2d;
 	       E T4R, T1P, T4P, T49, T2N, T2f, T47, T2L;
diff --git a/rdft/scalar/r2cb/hb2_25.c b/rdft/scalar/r2cb/hb2_25.c
index b6094a8..e920e10 100644
--- a/rdft/scalar/r2cb/hb2_25.c
+++ b/rdft/scalar/r2cb/hb2_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:13 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -85,7 +85,7 @@ static void hb2_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E TN, TQ, T4e, T2y, T4i, T3U, T4u, T4o, T4G, T4C, T2F, T41, T3Q, T4q, T3a;
 	       E T3F, T4a, T4w, T46, T44;
 	       {
@@ -957,7 +957,7 @@ static void hb2_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E TN, TQ, TO, TR, TT, TY, T2t, T2r, TZ, TU, T4f, T4l, T2d, T4v, T5m;
 	       E T2j, T5l, T4X, T2v, T11, T3R, T1L, T5d, T6x, T5h, T6t, T25, T26, T27, T29;
 	       E T6D, T7v, T49, T7l, T7p, T7t, T2p, T2n, T4b, T4p, T5n, T6B, T5b, T5p, T6p;
diff --git a/rdft/scalar/r2cb/hb2_32.c b/rdft/scalar/r2cb/hb2_32.c
index 0d2d38c..8136dc1 100644
--- a/rdft/scalar/r2cb/hb2_32.c
+++ b/rdft/scalar/r2cb/hb2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:10 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:25 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hb2_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T5u, T6b, T6e, T5I, T66, T60, T5U, T5R, T67, T5L, T61, T5x, T5A, T5D, T5O;
 	       E T62, T5V, T5P;
 	       {
@@ -983,7 +983,7 @@ static void hb2_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T11, T14, T12, T15, T17, T2z, T2B, T1c, T18, T1d, T1g, T1k, T2F, T2L, T3t;
 	       E T4H, T3h, T3V, T3b, T4v, T4T, T4X, T6t, T71, T6z, T75, T81, T8x, T8f, T8z;
 	       E T2R, T2V, T8p, T8t, T4r, T4t, T53, T69, T3n, T3r, T7P, T7T, T4P, T4R, T6F;
diff --git a/rdft/scalar/r2cb/hb2_4.c b/rdft/scalar/r2cb/hb2_4.c
index 0125fff..3319230 100644
--- a/rdft/scalar/r2cb/hb2_4.c
+++ b/rdft/scalar/r2cb/hb2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:23 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hb2_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E Tg, Tc, Te, To, Tn;
 	       {
 		    E T7, Tb, T8, Ta;
@@ -132,7 +132,7 @@ static void hb2_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T7, T9, T8, Ta, Tb, Td;
 	       T7 = W[0];
 	       T9 = W[1];
diff --git a/rdft/scalar/r2cb/hb2_5.c b/rdft/scalar/r2cb/hb2_5.c
index 6e31ce1..59e8fb9 100644
--- a/rdft/scalar/r2cb/hb2_5.c
+++ b/rdft/scalar/r2cb/hb2_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:11 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hb2_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T9, TB, Tz, Tm, T1, TG, TO, TJ, TC, Tn, Tg, To, Tf, Tw, TQ;
 	       E T8, Tb, Th, Ta, Ti, Tp;
 	       T9 = W[0];
@@ -174,7 +174,7 @@ static void hb2_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E Th, Tk, Ti, Tl, Tn, TP, Tx, TN;
 	       {
 		    E Tj, Tw, Tm, Tv;
diff --git a/rdft/scalar/r2cb/hb2_8.c b/rdft/scalar/r2cb/hb2_8.c
index eb4ef29..88860ea 100644
--- a/rdft/scalar/r2cb/hb2_8.c
+++ b/rdft/scalar/r2cb/hb2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:24 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hb2_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Tf, Tg, Tl, Tp, Ti, Tj, T1o, T1u, Tk, T1b, To, T1e, TK, Tq, T13;
 	       E TP, T1p, T7, T1h, T1v, TZ, Tv, Tw, Ta, Tx, T1j, TE, TB, Td, Ty;
 	       E Th, T1n, T1t;
@@ -236,7 +236,7 @@ static void hb2_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Tf, Ti, Tg, Tj, Tl, Tp, TP, TR, TF, TG, TH, T15, TL, TT;
 	       {
 		    E Th, To, Tk, Tn;
diff --git a/rdft/scalar/r2cb/hb_10.c b/rdft/scalar/r2cb/hb_10.c
index 3f00430..9208477 100644
--- a/rdft/scalar/r2cb/hb_10.c
+++ b/rdft/scalar/r2cb/hb_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hb_10(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T21, T1Y, T1X;
 	       {
 		    E T1B, TH, T1g, T3, T1V, T1x, T1G, T1E, TM, TK, T11, TB, T7, T1m, T1J;
@@ -287,7 +287,7 @@ static void hb_10(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T3, T18, TE, TF, T1B, T1A, T1f, T1t, Ti, Tl, TJ, T1i, Tt, TA, T1w;
 	       E T1v, T1p, T1E, TM, TO;
 	       {
diff --git a/rdft/scalar/r2cb/hb_12.c b/rdft/scalar/r2cb/hb_12.c
index e74bbe4..00575a8 100644
--- a/rdft/scalar/r2cb/hb_12.c
+++ b/rdft/scalar/r2cb/hb_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:14 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hb_12(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T1U, T1X, T1W, T1Y, T1V;
 	       {
 		    E T18, T20, T2a, T1s, T21, T1b, T29, T1p, TO, T11, To, Tb, Tg, T23, T1f;
@@ -326,7 +326,7 @@ static void hb_12(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T5, TH, T12, T1M, T1i, T1U, Tg, Tt, T19, T1X, T1p, T1P, Ta, TM, T15;
 	       E T1N, T1l, T1V, Tl, Ty, T1c, T1Y, T1s, T1Q;
 	       {
diff --git a/rdft/scalar/r2cb/hb_15.c b/rdft/scalar/r2cb/hb_15.c
index d18a6f2..c66a795 100644
--- a/rdft/scalar/r2cb/hb_15.c
+++ b/rdft/scalar/r2cb/hb_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:14 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void hb_15(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(30, rs)) {
 	       E T3v, T3u, T3r, T3w, T3t;
 	       {
 		    E T5, T11, T1C, T2U, T2f, T3f, T19, T18, TS, TH, T14, T16, T3g, T3a, Ts;
@@ -443,7 +443,7 @@ static void hb_15(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(30, rs)) {
 	       E T5, T10, T1J, T2C, T2c, T2M, TH, T18, T17, TS, T2Q, T2R, T2S, Tg, Tr;
 	       E Ts, T11, T12, T13, T2N, T2O, T2P, T1u, T1x, T1y, T1W, T1Z, T28, T1P, T1S;
 	       E T27, T1B, T1E, T1F, T2G, T2H, T2I, T2D, T2E, T2F;
diff --git a/rdft/scalar/r2cb/hb_16.c b/rdft/scalar/r2cb/hb_16.c
index a83442e..aa770ee 100644
--- a/rdft/scalar/r2cb/hb_16.c
+++ b/rdft/scalar/r2cb/hb_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:14 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hb_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1I, T1L, T1K, T1M, T1J;
 	       {
 		    E T1O, TA, T1h, T21, T3b, T2T, T3D, T3r, T1k, T1P, T3y, Tf, T36, T2A, T22;
@@ -441,7 +441,7 @@ static void hb_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T2K, T2W, Tw, T17, T1S, T2k, T1w, Te, TD, T1x, T10, T2n, T2L, T1Z;
 	       E T2X, Tm, T1z, TN, T19, T2e, T2p, T2P, T2Z, Tt, T1A, TW, T1a, T27, T2q;
 	       E T2S, T30;
diff --git a/rdft/scalar/r2cb/hb_2.c b/rdft/scalar/r2cb/hb_2.c
index c70cfd8..2dc242f 100644
--- a/rdft/scalar/r2cb/hb_2.c
+++ b/rdft/scalar/r2cb/hb_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:12 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hb_2(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
 	       E T5, T6, T9, T8, T7, Ta;
 	       {
 		    E T1, T2, T3, T4;
@@ -86,7 +86,7 @@ static void hb_2(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
 	       E T1, T2, T6, T3, T4, T8, T5, T7;
 	       T1 = cr[0];
 	       T2 = ci[0];
diff --git a/rdft/scalar/r2cb/hb_20.c b/rdft/scalar/r2cb/hb_20.c
index 528c834..11d073f 100644
--- a/rdft/scalar/r2cb/hb_20.c
+++ b/rdft/scalar/r2cb/hb_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:22 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hb_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T1T, T1Q, T1P;
 	       {
 		    E T2W, T4e, T7, TE, T3z, T4z, T1t, T2l, T3a, T3G, T13, T33, T3H, T1i, T2g;
@@ -572,7 +572,7 @@ static void hb_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T7, T3T, T49, TE, T1v, T2T, T3g, T2d, T13, T3n, T3o, T1i, T26, T4e, T4d;
 	       E T23, T1n, T42, T3Z, T1m, T2h, T2I, T2i, T2P, T30, T37, T38, Tm, TB, TC;
 	       E T46, T47, T4a, T2a, T2b, T2e, T1w, T1x, T1y, T3O, T3R, T3U, T3h, T3i, T3j;
diff --git a/rdft/scalar/r2cb/hb_25.c b/rdft/scalar/r2cb/hb_25.c
index ed1a8c0..2a796af 100644
--- a/rdft/scalar/r2cb/hb_25.c
+++ b/rdft/scalar/r2cb/hb_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:23 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -85,7 +85,7 @@ static void hb_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T3w, T3P, T2d, T3y, T3x, T3Q;
 	       {
 		    E T9, T3E, T1F, T3B, T6f, T7d, T5u, T6U, T4k, T2k, T5G, T1G, T19, T1H, T1s;
@@ -897,7 +897,7 @@ static void hb_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T9, T5Q, T3y, T39, T5v, Ti, Tr, Ts, TZ, T18, T1z, T2k, T4l, T3h, T44;
 	       E T5d, T6C, T5C, T6o, T56, T6B, T5B, T6l, T2z, T4m, T3i, T47, T1K, T5w, T3c;
 	       E T3B, T5R, TB, TK, TL, T1i, T1r, T1A, T2P, T4o, T3k, T4b, T5s, T6F, T5F;
diff --git a/rdft/scalar/r2cb/hb_3.c b/rdft/scalar/r2cb/hb_3.c
index d41dd6d..960dd48 100644
--- a/rdft/scalar/r2cb/hb_3.c
+++ b/rdft/scalar/r2cb/hb_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:12 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hb_3(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs)) {
 	       E Tk, Tj, Tn, Tl, Tm, To;
 	       {
 		    E T1, Td, T7, T8, T4, Tg, T2, T3;
@@ -109,7 +109,7 @@ static void hb_3(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs)) {
 	       E T1, T4, Ta, Te, T5, T8, Tb, Tf;
 	       {
 		    E T2, T3, T6, T7;
diff --git a/rdft/scalar/r2cb/hb_32.c b/rdft/scalar/r2cb/hb_32.c
index 7fde105..53063e0 100644
--- a/rdft/scalar/r2cb/hb_32.c
+++ b/rdft/scalar/r2cb/hb_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:00 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:15 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hb_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T5o, T5r, T5q, T5n, T5s, T5p;
 	       {
 		    E T5K, Tf, T8k, T7k, T8x, T7N, T3i, T1i, T3v, T2L, T5f, T4v, T6T, T6m, T52;
@@ -928,7 +928,7 @@ static void hb_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T4o, T6y, T70, T5u, Tf, T12, T5x, T6z, T3m, T3Y, T29, T2y, T4v, T71, T2U;
 	       E T3M, Tu, T1U, T6D, T73, T6G, T74, T1h, T2z, T2X, T3o, T4D, T5A, T4K, T5z;
 	       E T30, T3n, TK, T1j, T6S, T7w, T6V, T7v, T1y, T2B, T3c, T3S, T4X, T61, T54;
diff --git a/rdft/scalar/r2cb/hb_4.c b/rdft/scalar/r2cb/hb_4.c
index 7ecea54..747be7f 100644
--- a/rdft/scalar/r2cb/hb_4.c
+++ b/rdft/scalar/r2cb/hb_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:12 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hb_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T8, Th, Ta, T7, Ti, T9;
 	       {
 		    E Td, Tg, T3, T6, Tu, Tm, Tx, Tr;
@@ -126,7 +126,7 @@ static void hb_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T3, Ti, T6, Tm, Tc, Tn, Tf, Tj;
 	       {
 		    E T1, T2, T4, T5;
diff --git a/rdft/scalar/r2cb/hb_5.c b/rdft/scalar/r2cb/hb_5.c
index 637ad30..f0537a9 100644
--- a/rdft/scalar/r2cb/hb_5.c
+++ b/rdft/scalar/r2cb/hb_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:12 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hb_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E TQ, TP, TT, TR, TS, TU;
 	       {
 		    E T1, Tn, TM, Tw, Tb, T8, To, Tf, Ta, Tg, Th;
@@ -164,7 +164,7 @@ static void hb_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T1, Tj, TG, Ts, T8, Ti, T9, Tn, TD, Tu, Tg, Tt;
 	       {
 		    E T4, Tq, T7, Tr;
diff --git a/rdft/scalar/r2cb/hb_6.c b/rdft/scalar/r2cb/hb_6.c
index 8c1b816..8d1dfc1 100644
--- a/rdft/scalar/r2cb/hb_6.c
+++ b/rdft/scalar/r2cb/hb_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:12 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hb_6(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs)) {
 	       E TK, TR, TB, TM, TL, TS;
 	       {
 		    E Td, TN, TO, TJ, Tn, Tk, TC, T3, Tr, T4, T5, T7, T8;
@@ -175,7 +175,7 @@ static void hb_6(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs)) {
 	       E T3, Ty, Ta, TO, Tr, TB, Td, TE, Tk, TL, Tn, TH;
 	       {
 		    E T1, T2, Tb, Tc;
diff --git a/rdft/scalar/r2cb/hb_64.c b/rdft/scalar/r2cb/hb_64.c
index 4f18d36..582d3b2 100644
--- a/rdft/scalar/r2cb/hb_64.c
+++ b/rdft/scalar/r2cb/hb_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:01 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:15 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -53,7 +53,7 @@ static void hb_64(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tcx, Tcw, Tcv;
 	       {
 		    E Thy, Tv, T7n, T5B, TfP, Tey, Tkl, TjB, T6U, T2k, T7o, T2H, TiH, Tia, Tk8;
@@ -2043,7 +2043,7 @@ static void hb_64(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tf, T8C, Tfa, Thk, Tgg, ThM, T2c, T5O, T4K, T6g, Tag, TdE, TcA, Te6, T7P;
 	       E T94, TK, T7o, T38, T4P, Tfv, Thn, T5W, T6j, Tb0, TdK, Tfs, Tho, T8K, T97;
 	       E Tb7, TdL, TZ, T7l, T2P, T4Q, Tfo, Thq, T5T, T6k, TaH, TdH, Tfl, Thr, T8H;
diff --git a/rdft/scalar/r2cb/hb_7.c b/rdft/scalar/r2cb/hb_7.c
index 8b5e5aa..e9ca4c2 100644
--- a/rdft/scalar/r2cb/hb_7.c
+++ b/rdft/scalar/r2cb/hb_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:12 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void hb_7(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP554958132, +0.554958132087371191422194871006410481067288862);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(14, rs)) {
 	       E T1q, T1p, T1t, T1r, T1s, T1u;
 	       {
 		    E T1, T4, TC, T7, TB, Tt, TD, Ta, TA, T1l, TZ, T1b, Th, Tw, Td;
@@ -227,7 +227,7 @@ static void hb_7(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP433883739, +0.433883739117558120475768332848358754609990728);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(14, rs)) {
 	       E T1, T4, T7, Ta, Tx, TI, TV, TQ, TE, Tm, Tb, Te, Th, Tk, Tq;
 	       E TF, TR, TU, TJ, Tt;
 	       {
diff --git a/rdft/scalar/r2cb/hb_8.c b/rdft/scalar/r2cb/hb_8.c
index 3f04a68..968eb26 100644
--- a/rdft/scalar/r2cb/hb_8.c
+++ b/rdft/scalar/r2cb/hb_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hb_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Tw, TH, Tf, Ty, Tx, TI;
 	       {
 		    E TV, TD, T1i, T7, T1b, T1n, TQ, Tk, Tb, Tl, Ta, T1d, Tt, Tc, Tm;
@@ -221,7 +221,7 @@ static void hb_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T7, T18, T1c, To, Ty, TM, TY, TC, Te, TZ, T10, Tv, Tz, TP, TS;
 	       E TD;
 	       {
diff --git a/rdft/scalar/r2cb/hb_9.c b/rdft/scalar/r2cb/hb_9.c
index f387ce4..54a1812 100644
--- a/rdft/scalar/r2cb/hb_9.c
+++ b/rdft/scalar/r2cb/hb_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:59 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -48,7 +48,7 @@ static void hb_9(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(18, rs)) {
 	       E T1X, T1S, T1U, T1P, T1Y, T1T;
 	       {
 		    E T5, Tl, TQ, T1y, T1b, T1J, Tg, TE, TW, T13, T10, Tz, Tw, TT, T1K;
@@ -291,7 +291,7 @@ static void hb_9(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(18, rs)) {
 	       E T5, Tl, TM, T1o, T16, T1y, Ta, Tf, Tg, Tq, Tv, Tw, TT, T17, T1u;
 	       E T1A, T1r, T1z, T10, T18;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cb2_16.c b/rdft/scalar/r2cb/hc2cb2_16.c
index 9fe9fb6..ef85f4a 100644
--- a/rdft/scalar/r2cb/hc2cb2_16.c
+++ b/rdft/scalar/r2cb/hc2cb2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:57 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cb2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E Tv, TB, TF, Ty, T1J, T1O, T1N, T1K;
 	       {
 		    E Tw, T2z, T2C, Tx, T3f, T3l, T2F, T3r, Tz;
@@ -454,7 +454,7 @@ static void hc2cb2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E Tv, Ty, T1l, T1n, T1p, T1t, T27, T25, Tz, Tw, TB, T21, T1P, T1H, T1X;
 	       E T17, T1L, T1N, T1v, T1w, T1x, T1B, T2F, T2T, T2b, T2R, T3j, T3x, T35, T3t;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cb2_20.c b/rdft/scalar/r2cb/hc2cb2_20.c
index e3bc337..3ec63c6 100644
--- a/rdft/scalar/r2cb/hc2cb2_20.c
+++ b/rdft/scalar/r2cb/hc2cb2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:43 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:02 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cb2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T1S, T1O, T1s, TI, T24, T1Y, T2g, T2k, TS, TR, T1I, T26, T1o, T20, T1F;
 	       E T25, TT, T1Z;
 	       {
@@ -600,7 +600,7 @@ static void hc2cb2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E TD, TG, TE, TH, TJ, T1t, T27, T25, T1T, T1R, T1V, T2j, T2Z, T21, T2X;
 	       E T2T, T2n, T2P, T3V, T41, T3R, T3X, T29, T2c, T4H, T4L, T1L, T1M, T1N, T2d;
 	       E T4R, T1P, T4P, T49, T2N, T2f, T47, T2L;
diff --git a/rdft/scalar/r2cb/hc2cb2_32.c b/rdft/scalar/r2cb/hc2cb2_32.c
index c39f99a..d821304 100644
--- a/rdft/scalar/r2cb/hc2cb2_32.c
+++ b/rdft/scalar/r2cb/hc2cb2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:58 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cb2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T5u, T6b, T6e, T5I, T66, T60, T5U, T5R, T67, T5L, T61, T5x, T5A, T5D, T5O;
 	       E T62, T5V, T5P;
 	       {
@@ -993,7 +993,7 @@ static void hc2cb2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T11, T14, T12, T15, T17, T2z, T2B, T1c, T18, T1d, T1g, T1k, T2F, T2L, T3t;
 	       E T4H, T3h, T3V, T3b, T4v, T4T, T4X, T6t, T71, T6z, T75, T81, T8x, T8f, T8z;
 	       E T2R, T2V, T8p, T8t, T4r, T4t, T53, T69, T3n, T3r, T7P, T7T, T4P, T4R, T6F;
diff --git a/rdft/scalar/r2cb/hc2cb2_4.c b/rdft/scalar/r2cb/hc2cb2_4.c
index 852f9df..de33449 100644
--- a/rdft/scalar/r2cb/hc2cb2_4.c
+++ b/rdft/scalar/r2cb/hc2cb2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:57 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cb2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Tg, Tc, Te, To, Tn;
 	       {
 		    E T7, Tb, T8, Ta;
@@ -132,7 +132,7 @@ static void hc2cb2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T7, T9, T8, Ta, Tb, Td;
 	       T7 = W[0];
 	       T9 = W[1];
diff --git a/rdft/scalar/r2cb/hc2cb2_8.c b/rdft/scalar/r2cb/hc2cb2_8.c
index 3d77afa..5fa33e4 100644
--- a/rdft/scalar/r2cb/hc2cb2_8.c
+++ b/rdft/scalar/r2cb/hc2cb2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:57 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cb2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E Tf, Ti, TK, Tq, TH, TT, TX, TW, TY, TU, TI;
 	       {
 		    E Tg, Tl, Tp, Th, T1n, T1t, Tj;
@@ -229,7 +229,7 @@ static void hc2cb2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E Tf, Ti, Tg, Tj, Tl, Tp, TP, TR, TF, TG, TH, T15, TL, TT;
 	       {
 		    E Th, To, Tk, Tn;
diff --git a/rdft/scalar/r2cb/hc2cb_10.c b/rdft/scalar/r2cb/hc2cb_10.c
index 51f04be..1e6807b 100644
--- a/rdft/scalar/r2cb/hc2cb_10.c
+++ b/rdft/scalar/r2cb/hc2cb_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:53 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cb_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T21, T1Y, T1X;
 	       {
 		    E T1B, TH, T1g, T3, T1V, T1x, T1G, T1E, TM, TK, T11, TB, T7, T1m, T1J;
@@ -287,7 +287,7 @@ static void hc2cb_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T3, T18, TJ, T1i, TE, TF, T1B, T1A, T1f, T1t, Ti, Tl, Tt, TA, T1w;
 	       E T1v, T1p, T1E, TM, TO;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cb_12.c b/rdft/scalar/r2cb/hc2cb_12.c
index 169d848..cd49e98 100644
--- a/rdft/scalar/r2cb/hc2cb_12.c
+++ b/rdft/scalar/r2cb/hc2cb_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:53 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cb_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E T1U, T1X, T1W, T1Y, T1V;
 	       {
 		    E T18, T20, T21, T1b, T2a, T1s, T29, T1p, TO, T11, To, Tb, Tg, T23, T1f;
@@ -326,7 +326,7 @@ static void hc2cb_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E T5, TH, T12, T1M, T1i, T1U, Tl, Ty, T1c, T1Y, T1s, T1Q, Ta, TM, T15;
 	       E T1N, T1l, T1V, Tg, Tt, T19, T1X, T1p, T1P;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cb_16.c b/rdft/scalar/r2cb/hc2cb_16.c
index 50e4330..02b9b98 100644
--- a/rdft/scalar/r2cb/hc2cb_16.c
+++ b/rdft/scalar/r2cb/hc2cb_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:53 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cb_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T1I, T1L, T1K, T1M, T1J;
 	       {
 		    E T1O, TA, T1h, T21, T3b, T2T, T3D, T3r, T1k, T1P, T3y, Tf, T36, T2A, T22;
@@ -441,7 +441,7 @@ static void hc2cb_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T7, T2K, T2W, Tw, T17, T1S, T2k, T1w, Te, TD, T1x, T10, T2n, T2L, T1Z;
 	       E T2X, Tm, T1z, TN, T19, T2e, T2p, T2P, T2Z, Tt, T1A, TW, T1a, T27, T2q;
 	       E T2S, T30;
diff --git a/rdft/scalar/r2cb/hc2cb_2.c b/rdft/scalar/r2cb/hc2cb_2.c
index d689e2e..4e64f3f 100644
--- a/rdft/scalar/r2cb/hc2cb_2.c
+++ b/rdft/scalar/r2cb/hc2cb_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cb_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T5, T6, T9, T8, T7, Ta;
 	       {
 		    E T1, T2, T3, T4;
@@ -86,7 +86,7 @@ static void hc2cb_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T1, T2, T6, T3, T4, T8, T5, T7;
 	       T1 = Rp[0];
 	       T2 = Rm[0];
diff --git a/rdft/scalar/r2cb/hc2cb_20.c b/rdft/scalar/r2cb/hc2cb_20.c
index b825d58..601c84f 100644
--- a/rdft/scalar/r2cb/hc2cb_20.c
+++ b/rdft/scalar/r2cb/hc2cb_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:55 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cb_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T1T, T1Q, T1P;
 	       {
 		    E T3z, T4z, TE, T7, T2W, T4e, T2l, T1t, T33, T3H, T3G, T3a, T1i, T2g, T13;
@@ -572,7 +572,7 @@ static void hc2cb_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T7, T3T, T49, TE, T1v, T2T, T3g, T2d, T13, T3n, T3o, T1i, T26, T4e, T4d;
 	       E T23, T1n, T42, T3Z, T1m, T2h, T2I, T2i, T2P, T30, T37, T38, Tm, TB, TC;
 	       E T46, T47, T4a, T2a, T2b, T2e, T1w, T1x, T1y, T3O, T3R, T3U, T3h, T3i, T3j;
diff --git a/rdft/scalar/r2cb/hc2cb_32.c b/rdft/scalar/r2cb/hc2cb_32.c
index 7c1992f..ac7b4d2 100644
--- a/rdft/scalar/r2cb/hc2cb_32.c
+++ b/rdft/scalar/r2cb/hc2cb_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:55 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cb_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T5o, T5r, T5q, T5n, T5s, T5p;
 	       {
 		    E T5K, Tf, T8k, T7k, T8x, T7N, T3i, T1i, T3v, T2L, T5f, T4v, T6T, T6m, T52;
@@ -928,7 +928,7 @@ static void hc2cb_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T4o, T6y, T70, T5u, Tf, T12, T5x, T6z, T3m, T3Y, T29, T2y, T4v, T71, T2U;
 	       E T3M, Tu, T1U, T6D, T73, T6G, T74, T1h, T2z, T2X, T3o, T4D, T5A, T4K, T5z;
 	       E T30, T3n, TK, T1j, T6S, T7w, T6V, T7v, T1y, T2B, T3c, T3S, T4X, T61, T54;
diff --git a/rdft/scalar/r2cb/hc2cb_4.c b/rdft/scalar/r2cb/hc2cb_4.c
index 1f5a49c..05f20b1 100644
--- a/rdft/scalar/r2cb/hc2cb_4.c
+++ b/rdft/scalar/r2cb/hc2cb_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cb_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Th, Ta, T7, Ti, T9;
 	       {
 		    E Tq, Td, T3, Tg, Tu, Tm, T6, Tp;
@@ -126,7 +126,7 @@ static void hc2cb_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T3, Ti, Tc, Tn, T6, Tm, Tf, Tj;
 	       {
 		    E T1, T2, Ta, Tb;
diff --git a/rdft/scalar/r2cb/hc2cb_6.c b/rdft/scalar/r2cb/hc2cb_6.c
index 1b64bfc..cd2e263 100644
--- a/rdft/scalar/r2cb/hc2cb_6.c
+++ b/rdft/scalar/r2cb/hc2cb_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cb_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E TK, TR, TB, TM, TL, TS;
 	       {
 		    E Td, TN, TO, TJ, Tn, Tk, TC, T3, Tr, T7, T8, T4, T5;
@@ -175,7 +175,7 @@ static void hc2cb_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T3, Ty, Td, TE, Ta, TO, Tr, TB, Tk, TL, Tn, TH;
 	       {
 		    E T1, T2, Tb, Tc;
diff --git a/rdft/scalar/r2cb/hc2cb_8.c b/rdft/scalar/r2cb/hc2cb_8.c
index fb2975c..f3cefdd 100644
--- a/rdft/scalar/r2cb/hc2cb_8.c
+++ b/rdft/scalar/r2cb/hc2cb_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:36 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cb_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E Tw, TH, Tf, Ty, Tx, TI;
 	       {
 		    E TV, TD, T1i, T7, T1b, T1n, TQ, Tk, Tp, TE, Te, T1o, T1e, T1j, Tu;
@@ -218,7 +218,7 @@ static void hc2cb_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T18, T1c, To, Ty, TM, TY, TC, Te, TZ, T10, Tv, Tz, TP, TS;
 	       E TD;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cbdft2_16.c b/rdft/scalar/r2cb/hc2cbdft2_16.c
index 174296e..b746c23 100644
--- a/rdft/scalar/r2cb/hc2cbdft2_16.c
+++ b/rdft/scalar/r2cb/hc2cbdft2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:49 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cbdft2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T3w, T3z, T2Y, T3D, T3x, T3m, T3u, T3C, T3y, T3o, T3k, T3E, T3A;
 	       {
 		    E T20, Tf, T3Q, T32, T3V, T3f, T2a, TN, T2f, T1m, T3G, T2G, T3L, T2T, T26;
@@ -469,7 +469,7 @@ static void hc2cbdft2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E TB, T2L, T30, T1n, Tf, T1U, T2H, T3p, T1E, T1Z, TM, T31, T2s, T3k, T1i;
 	       E T2M, Tu, T1Y, T2Q, T2X, T2T, T2Y, TY, T1d, T19, T1e, T2v, T2C, T2y, T2D;
 	       E T1x, T1V;
diff --git a/rdft/scalar/r2cb/hc2cbdft2_20.c b/rdft/scalar/r2cb/hc2cbdft2_20.c
index d1982ad..2d5926f 100644
--- a/rdft/scalar/r2cb/hc2cbdft2_20.c
+++ b/rdft/scalar/r2cb/hc2cbdft2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:50 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:09 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cbdft2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T5s, T5v, T5t, T5z, T5q, T5y, T5u, T5A, T5w;
 	       {
 		    E T3T, T27, T2o, T41, T2p, T40, TU, T15, T2Q, T1N, T2L, T1w, T59, T4n, T5e;
@@ -599,7 +599,7 @@ static void hc2cbdft2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T7, T3N, T4a, T16, T1G, T3g, T3D, T26, T1k, T3A, T3B, T1v, T2e, T48, T47;
 	       E T2d, T1L, T43, T40, T1K, T2l, T3t, T2m, T3w, T3n, T3p, TC, T2b, T4d, T4f;
 	       E T23, T2j, T1B, T1H, T3U, T3W, T3G, T3I, T11, T17;
diff --git a/rdft/scalar/r2cb/hc2cbdft2_32.c b/rdft/scalar/r2cb/hc2cbdft2_32.c
index 8c14a57..f804340 100644
--- a/rdft/scalar/r2cb/hc2cbdft2_32.c
+++ b/rdft/scalar/r2cb/hc2cbdft2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:49 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:08 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cbdft2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T8e, T8h, T7S, T8l, T8f, T84, T8c, T8k, T8g, T86, T82, T8m, T8i;
 	       {
 		    E T4B, T3h, T3K, Tv, T8Y, T6T, T8L, T7i, T8X, T7f, T4Y, T1G, T4K, T1j, T4X;
@@ -966,7 +966,7 @@ static void hc2cbdft2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tf, T4a, T6h, T7Z, T6P, T8e, T1j, T4v, T2R, T4L, T5C, T7E, T6a, T7U, T3n;
 	       E T4q, TZ, T38, T2p, T4B, T7M, T7R, T2y, T4C, T5Y, T63, T6C, T86, T4i, T4n;
 	       E T6z, T85, TK, T31, T1Y, T4y, T7J, T7Q, T27, T4z, T5R, T62, T6v, T83, T4f;
diff --git a/rdft/scalar/r2cb/hc2cbdft2_4.c b/rdft/scalar/r2cb/hc2cbdft2_4.c
index 8856ff6..7aafbaa 100644
--- a/rdft/scalar/r2cb/hc2cbdft2_4.c
+++ b/rdft/scalar/r2cb/hc2cbdft2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:48 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cbdft2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Ty, TB, Tw, TE, TA, TF, Tz, TG, TC;
 	       {
 		    E T4, Tg, T3, Tm, Tc, T5, Th, Ti;
@@ -134,7 +134,7 @@ static void hc2cbdft2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T3, Tl, T6, Tm, Td, Tj, Tx, Tv, Ts, Tq;
 	       {
 		    E Tf, Tc, T9, Ti;
diff --git a/rdft/scalar/r2cb/hc2cbdft2_8.c b/rdft/scalar/r2cb/hc2cbdft2_8.c
index 6181bf1..bee8efc 100644
--- a/rdft/scalar/r2cb/hc2cbdft2_8.c
+++ b/rdft/scalar/r2cb/hc2cbdft2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:48 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cbdft2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1m, T1r, T1i, T1u, T1o, T1v, T1n, T1w, T1s;
 	       {
 		    E T1k, Tl, T1p, TE, TP, T1g, TM, T1b, T1f, T1a, TU, Tf, T1l, TH, Tw;
@@ -241,7 +241,7 @@ static void hc2cbdft2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T1d, T1h, Tl, TG, T14, T19, TO, Te, TL, T18, T15, TB, T1e, Tw;
 	       E T1i;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cbdft_10.c b/rdft/scalar/r2cb/hc2cbdft_10.c
index d1396ac..522b072 100644
--- a/rdft/scalar/r2cb/hc2cbdft_10.c
+++ b/rdft/scalar/r2cb/hc2cbdft_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:46 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:04 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cbdft_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T2d, T2f;
 	       {
 		    E T1g, TQ, T1z, TZ, Tu, T23, T1p, T14, Tt, T27, T13, Tj, Tz, T1i, T18;
@@ -313,7 +313,7 @@ static void hc2cbdft_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T3, TS, TR, T13, Ti, T12, TT, TU, T1g, T1T, Tr, T1s, TJ, T1h, TG;
 	       E T1m, TK, TL, T1k, T1l, T1b, T1P, TY, T1w;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cbdft_12.c b/rdft/scalar/r2cb/hc2cbdft_12.c
index 54cadac..c7249c0 100644
--- a/rdft/scalar/r2cb/hc2cbdft_12.c
+++ b/rdft/scalar/r2cb/hc2cbdft_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:47 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:04 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cbdft_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E T2S, T2V, T2w, T2Z, T2T, T2I, T2Q, T2Y, T2U, T2K, T2G, T30, T2W;
 	       {
 		    E Tb, T1Z, T2D, T1E, T1N, T2y, TD, T2t, T1U, T1e, T2o, TY, T1f, TI, T1g;
@@ -343,7 +343,7 @@ static void hc2cbdft_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E Tv, T1E, TC, T1F, TW, T1x, TT, T1w, T1d, T1N, Tb, T1R, TI, T1z, TN;
 	       E T1A, T17, T1I, T12, T1H, T1g, T1S, Tm, T1O;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cbdft_16.c b/rdft/scalar/r2cb/hc2cbdft_16.c
index 13830d7..515c0d4 100644
--- a/rdft/scalar/r2cb/hc2cbdft_16.c
+++ b/rdft/scalar/r2cb/hc2cbdft_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:47 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:05 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cbdft_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T3w, T3z, T2Y, T3D, T3x, T3m, T3u, T3C, T3y, T3o, T3k, T3E, T3A;
 	       {
 		    E T20, Tf, T3Q, T32, T3V, T3f, T2a, TN, T2f, T1m, T3G, T2G, T3L, T2T, T26;
@@ -469,7 +469,7 @@ static void hc2cbdft_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E TB, T2L, T30, T1n, Tf, T1U, T2H, T3p, T1E, T1Z, TM, T31, T2s, T3k, T1i;
 	       E T2M, Tu, T1Y, T2Q, T2X, T2T, T2Y, TY, T1d, T19, T1e, T2v, T2C, T2y, T2D;
 	       E T1x, T1V;
diff --git a/rdft/scalar/r2cb/hc2cbdft_2.c b/rdft/scalar/r2cb/hc2cbdft_2.c
index 15a9c93..2bac682 100644
--- a/rdft/scalar/r2cb/hc2cbdft_2.c
+++ b/rdft/scalar/r2cb/hc2cbdft_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:03 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cbdft_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T9, Ta, T3, Tc, T7, T4;
 	       {
 		    E T1, T2, T5, T6;
@@ -93,7 +93,7 @@ static void hc2cbdft_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T3, T9, T7, Tb;
 	       {
 		    E T1, T2, T5, T6;
diff --git a/rdft/scalar/r2cb/hc2cbdft_20.c b/rdft/scalar/r2cb/hc2cbdft_20.c
index 3ec05ab..84fd4de 100644
--- a/rdft/scalar/r2cb/hc2cbdft_20.c
+++ b/rdft/scalar/r2cb/hc2cbdft_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:48 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cbdft_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T5s, T5v, T5t, T5z, T5q, T5y, T5u, T5A, T5w;
 	       {
 		    E T3T, T27, T2o, T41, T2p, T40, TU, T15, T2Q, T1N, T2L, T1w, T59, T4n, T5e;
@@ -599,7 +599,7 @@ static void hc2cbdft_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T7, T3N, T4a, T16, T1G, T3g, T3D, T26, T1k, T3A, T3B, T1v, T2e, T48, T47;
 	       E T2d, T1L, T43, T40, T1K, T2l, T3t, T2m, T3w, T3n, T3p, TC, T2b, T4d, T4f;
 	       E T23, T2j, T1B, T1H, T3U, T3W, T3G, T3I, T11, T17;
diff --git a/rdft/scalar/r2cb/hc2cbdft_32.c b/rdft/scalar/r2cb/hc2cbdft_32.c
index 871c6f5..4cce754 100644
--- a/rdft/scalar/r2cb/hc2cbdft_32.c
+++ b/rdft/scalar/r2cb/hc2cbdft_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:47 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:06 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cbdft_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T8e, T8h, T7S, T8l, T8f, T84, T8c, T8k, T8g, T86, T82, T8m, T8i;
 	       {
 		    E T4B, T3h, T3K, Tv, T8Y, T6T, T8L, T7i, T8X, T7f, T4Y, T1G, T4K, T1j, T4X;
@@ -966,7 +966,7 @@ static void hc2cbdft_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tf, T4a, T6h, T7Z, T6P, T8e, T1j, T4v, T2R, T4L, T5C, T7E, T6a, T7U, T3n;
 	       E T4q, TZ, T38, T2p, T4B, T7M, T7R, T2y, T4C, T5Y, T63, T6C, T86, T4i, T4n;
 	       E T6z, T85, TK, T31, T1Y, T4y, T7J, T7Q, T27, T4z, T5R, T62, T6v, T83, T4f;
diff --git a/rdft/scalar/r2cb/hc2cbdft_4.c b/rdft/scalar/r2cb/hc2cbdft_4.c
index 40d89ca..554a170 100644
--- a/rdft/scalar/r2cb/hc2cbdft_4.c
+++ b/rdft/scalar/r2cb/hc2cbdft_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:03 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cbdft_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Ty, TB, Tw, TE, TA, TF, Tz, TG, TC;
 	       {
 		    E T4, Tg, T3, Tm, Tc, T5, Th, Ti;
@@ -134,7 +134,7 @@ static void hc2cbdft_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T3, Tl, T6, Tm, Td, Tj, Tx, Tv, Ts, Tq;
 	       {
 		    E Tf, Tc, T9, Ti;
diff --git a/rdft/scalar/r2cb/hc2cbdft_6.c b/rdft/scalar/r2cb/hc2cbdft_6.c
index bcce28d..60e48bd 100644
--- a/rdft/scalar/r2cb/hc2cbdft_6.c
+++ b/rdft/scalar/r2cb/hc2cbdft_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:04 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cbdft_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T18, T1b, T16, T1e, T1a, T1f, T19, T1g, T1c;
 	       {
 		    E Tw, T4, TV, Tj, TP, TH, Tr, TY, T5, T6, Ta, Ty;
@@ -193,7 +193,7 @@ static void hc2cbdft_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T4, Tv, Tr, TL, Tb, Tc, Ty, TP, To, TB, Tj, TQ, Tp, Tq, TE;
 	       E TM;
 	       {
diff --git a/rdft/scalar/r2cb/hc2cbdft_8.c b/rdft/scalar/r2cb/hc2cbdft_8.c
index 2a0726d..42a4d7f 100644
--- a/rdft/scalar/r2cb/hc2cbdft_8.c
+++ b/rdft/scalar/r2cb/hc2cbdft_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:04 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cbdft_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1m, T1r, T1i, T1u, T1o, T1v, T1n, T1w, T1s;
 	       {
 		    E T1k, Tl, T1p, TE, TP, T1g, TM, T1b, T1f, T1a, TU, Tf, T1l, TH, Tw;
@@ -241,7 +241,7 @@ static void hc2cbdft_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T1d, T1h, Tl, TG, T14, T19, TO, Te, TL, T18, T15, TB, T1e, Tw;
 	       E T1i;
 	       {
diff --git a/rdft/scalar/r2cb/r2cbIII_10.c b/rdft/scalar/r2cb/r2cbIII_10.c
index dfa8034..559b235 100644
--- a/rdft/scalar/r2cb/r2cbIII_10.c
+++ b/rdft/scalar/r2cb/r2cbIII_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:35 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void r2cbIII_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E Tq, Ti, Tk, Tu, Tw, Tp, Tb, Tj, Tr, Tv;
 	       {
 		    E T1, To, Ts, Tt, T8, Ta, Te, Tl, Tm, Th, Tn, T9;
@@ -126,7 +126,7 @@ static void r2cbIII_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_118033988, +1.118033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E T1, To, T8, Tq, Ta, Tp, Te, Ts, Th, Tn;
 	       T1 = Cr[WS(csr, 2)];
 	       To = Ci[WS(csi, 2)];
diff --git a/rdft/scalar/r2cb/r2cbIII_12.c b/rdft/scalar/r2cb/r2cbIII_12.c
index f4bbdc8..61cc97c 100644
--- a/rdft/scalar/r2cb/r2cbIII_12.c
+++ b/rdft/scalar/r2cb/r2cbIII_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:36 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cbIII_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E TE, TD, TF, TG;
 	       {
 		    E Tx, T6, Te, Tb, T5, Tw, Ts, To, Th, Ti, T9, TA;
@@ -145,7 +145,7 @@ static void r2cbIII_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E T5, Tw, Tb, Te, Tx, Ts, Ta, TA, Tg, Tj, Tz, Tp, Tt, Tu;
 	       {
 		    E T1, T2, T3, T4;
diff --git a/rdft/scalar/r2cb/r2cbIII_15.c b/rdft/scalar/r2cb/r2cbIII_15.c
index f5cd424..fec22f6 100644
--- a/rdft/scalar/r2cb/r2cbIII_15.c
+++ b/rdft/scalar/r2cb/r2cbIII_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:36 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -47,7 +47,7 @@ static void r2cbIII_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E TX, Tv, To, TW, Tl, Tx, Ty, Tw;
 	       {
 		    E TA, Tk, T6, T5, Tz, Th, TI, Tp, Tu, TK, TR, Tn, Td, Tq;
@@ -197,7 +197,7 @@ static void r2cbIII_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E Tv, TD, T5, Ts, TC, T6, Tf, TW, TK, Td, Tg, TP, To, TN, TA;
 	       E TO, TQ, Tt, Tu, T12, Te, T11;
 	       Tt = Ci[WS(csi, 4)];
diff --git a/rdft/scalar/r2cb/r2cbIII_16.c b/rdft/scalar/r2cb/r2cbIII_16.c
index d31678b..7c1f2e2 100644
--- a/rdft/scalar/r2cb/r2cbIII_16.c
+++ b/rdft/scalar/r2cb/r2cbIII_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:20 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:36 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -47,7 +47,7 @@ static void r2cbIII_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E TA, TD, Tv, TG, TE, TF;
 	       {
 		    E TK, TP, T7, T13, TW, TH, Tj, TC, To, Te, TX, TS, T12, Tt, TB;
@@ -195,7 +195,7 @@ static void r2cbIII_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E T7, TW, T13, Tj, TD, TK, TP, TH, Te, TX, T12, To, Tt, Tx, TS;
 	       E Tw, TT, TY;
 	       {
diff --git a/rdft/scalar/r2cb/r2cbIII_2.c b/rdft/scalar/r2cb/r2cbIII_2.c
index 7bb7024..a08f523 100644
--- a/rdft/scalar/r2cb/r2cbIII_2.c
+++ b/rdft/scalar/r2cb/r2cbIII_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:32 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void r2cbIII_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = Cr[0];
 	       T2 = Ci[0];
@@ -71,7 +71,7 @@ static void r2cbIII_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = Cr[0];
 	       T2 = Ci[0];
diff --git a/rdft/scalar/r2cb/r2cbIII_20.c b/rdft/scalar/r2cb/r2cbIII_20.c
index 98f5ad7..4857711 100644
--- a/rdft/scalar/r2cb/r2cbIII_20.c
+++ b/rdft/scalar/r2cb/r2cbIII_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:28 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:44 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void r2cbIII_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E TZ, TD, TW, Tw, Tt, TF, T1f, T1b;
 	       {
 		    E T1l, Tk, T9, Tj, Ta, TV, TI, Ts, TU, T1t, T11, Tx, T13, TC, T1a;
@@ -234,7 +234,7 @@ static void r2cbIII_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E T1, Tj, T1k, T13, T8, Tk, T17, Ts, T16, TI, T18, T19, Ta, Tu, T1i;
 	       E TS, Th, Tv, TX, TD, TV, TL, TW, TY;
 	       {
diff --git a/rdft/scalar/r2cb/r2cbIII_25.c b/rdft/scalar/r2cb/r2cbIII_25.c
index 3a94185..7fd93fd 100644
--- a/rdft/scalar/r2cb/r2cbIII_25.c
+++ b/rdft/scalar/r2cb/r2cbIII_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:29 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -82,7 +82,7 @@ static void r2cbIII_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E T1P, T2c, T2a, T24, T26, T25, T27, T2b;
 	       {
 		    E T1O, TS, T5, T1N, TP, Te, TA, T2i, T1V, T17, T1B, T2h, T1S, T10, T1C;
@@ -365,7 +365,7 @@ static void r2cbIII_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E TS, T1O, T5, TP, T1N, TI, TH, Te, T17, T2h, T1y, T1V, T10, T2g, T1x;
 	       E T1S, Tz, Ty, Tn, T1m, T2e, T1B, T22, T1f, T2d, T1A, T1Z, TQ, TR;
 	       TQ = Ci[WS(csi, 2)];
diff --git a/rdft/scalar/r2cb/r2cbIII_3.c b/rdft/scalar/r2cb/r2cbIII_3.c
index c1391d6..f5cff8e 100644
--- a/rdft/scalar/r2cb/r2cbIII_3.c
+++ b/rdft/scalar/r2cb/r2cbIII_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:32 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cbIII_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T4, T1, T2, T3;
 	       T4 = Ci[0];
 	       T1 = Cr[WS(csr, 1)];
@@ -76,7 +76,7 @@ static void r2cbIII_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T5, T1, T2, T3, T4;
 	       T4 = Ci[0];
 	       T5 = KP1_732050807 * T4;
diff --git a/rdft/scalar/r2cb/r2cbIII_32.c b/rdft/scalar/r2cb/r2cbIII_32.c
index 9d083b1..b030bad 100644
--- a/rdft/scalar/r2cb/r2cbIII_32.c
+++ b/rdft/scalar/r2cb/r2cbIII_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:37 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -56,7 +56,7 @@ static void r2cbIII_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T1N, T1K, T1Q, T1H, T1O, T1P;
 	       {
 		    E T1I, T1e, T1Z, T7, T2E, T2i, T1x, Tz, Te, T2j, T22, T2F, T1h, T1y, TK;
@@ -382,7 +382,7 @@ static void r2cbIII_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T7, T2i, T2F, Tz, T1k, T1I, T1Z, T1x, Te, T22, T2E, T2j, T1f, T1y, TK;
 	       E T1J, Tm, T2B, TW, T1a, T1C, T1L, T28, T2l, Tt, T2A, T17, T1b, T1F, T1M;
 	       E T2d, T2m;
diff --git a/rdft/scalar/r2cb/r2cbIII_4.c b/rdft/scalar/r2cb/r2cbIII_4.c
index c2ab4c5..d81ae77 100644
--- a/rdft/scalar/r2cb/r2cbIII_4.c
+++ b/rdft/scalar/r2cb/r2cbIII_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:17 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:33 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cbIII_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T1, T2, T4, T5, T3, T6;
 	       T1 = Cr[0];
 	       T2 = Cr[WS(csr, 1)];
@@ -79,7 +79,7 @@ static void r2cbIII_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T1, T2, T3, T4, T5, T6;
 	       T1 = Cr[0];
 	       T2 = Cr[WS(csr, 1)];
diff --git a/rdft/scalar/r2cb/r2cbIII_5.c b/rdft/scalar/r2cb/r2cbIII_5.c
index 211ec0c..b919ee4 100644
--- a/rdft/scalar/r2cb/r2cbIII_5.c
+++ b/rdft/scalar/r2cb/r2cbIII_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:17 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:33 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void r2cbIII_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E T1, T2, T3, Tc, Ta, T8, T9;
 	       T8 = Ci[WS(csi, 1)];
 	       T9 = Ci[0];
@@ -95,7 +95,7 @@ static void r2cbIII_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E Ta, Tc, T1, T4, T5, T6, Tb, T7;
 	       {
 		    E T8, T9, T2, T3;
diff --git a/rdft/scalar/r2cb/r2cbIII_6.c b/rdft/scalar/r2cb/r2cbIII_6.c
index d0b361e..d339629 100644
--- a/rdft/scalar/r2cb/r2cbIII_6.c
+++ b/rdft/scalar/r2cb/r2cbIII_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:17 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:33 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cbIII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T1, T8, T2, T3, T5, T6;
 	       T1 = Cr[WS(csr, 1)];
 	       T8 = Ci[WS(csi, 1)];
@@ -90,7 +90,7 @@ static void r2cbIII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T1, T6, T4, T5, T9, Tb, Ta, Tc;
 	       T1 = Cr[WS(csr, 1)];
 	       T6 = Ci[WS(csi, 1)];
diff --git a/rdft/scalar/r2cb/r2cbIII_64.c b/rdft/scalar/r2cb/r2cbIII_64.c
index 6705917..98062ea 100644
--- a/rdft/scalar/r2cb/r2cbIII_64.c
+++ b/rdft/scalar/r2cb/r2cbIII_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:38 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -74,7 +74,7 @@ static void r2cbIII_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E T43, T4b, T49, T4e, T3T, T46, T40, T4a;
 	       {
 		    E T3t, T15, T2E, T3U, T6b, Tf, T6Q, T6u, T5J, T4L, T3V, T1g, T5U, T5q, T3u;
@@ -814,7 +814,7 @@ static void r2cbIII_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E T15, T3t, T3U, T2N, Tf, T6b, T6u, T6R, T4L, T5J, T1g, T3V, T5q, T5U, T2I;
 	       E T3u, Tu, T6v, T4V, T5s, T6e, T6Q, T1s, T2D, T1D, T2E, T3B, T3Y, T4Q, T5r;
 	       E T3y, T3X, TK, T6g, T57, T5N, T6j, T6N, T1W, T34, T25, T35, T3J, T4j, T52;
diff --git a/rdft/scalar/r2cb/r2cbIII_7.c b/rdft/scalar/r2cb/r2cbIII_7.c
index dd44b41..8d63413 100644
--- a/rdft/scalar/r2cb/r2cbIII_7.c
+++ b/rdft/scalar/r2cb/r2cbIII_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:17 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:33 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void r2cbIII_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP554958132, +0.554958132087371191422194871006410481067288862);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E Tn, Td, Tg, Ti, Tl, T8;
 	       {
 		    E T1, T9, Tb, Ta, T2, T4, Th, Tm, Tc, T3, Te;
@@ -115,7 +115,7 @@ static void r2cbIII_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_563662964, +1.563662964936059617416889053348115500464669037);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E T9, Td, Tb, T1, T4, T2, T3, T5, Tc, Ta, T6, T8, T7;
 	       T6 = Ci[WS(csi, 2)];
 	       T8 = Ci[0];
diff --git a/rdft/scalar/r2cb/r2cbIII_8.c b/rdft/scalar/r2cb/r2cbIII_8.c
index 7c1a5b4..9e011d5 100644
--- a/rdft/scalar/r2cb/r2cbIII_8.c
+++ b/rdft/scalar/r2cb/r2cbIII_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:34 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cbIII_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E T4, T7, T3, Tl, Tf, T5, T8, T9, T6, Tc;
 	       {
 		    E T1, T2, Td, Te;
@@ -112,7 +112,7 @@ static void r2cbIII_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E T3, T7, Tf, Tl, T6, Tc, Ta, Tk, Tb, Tg;
 	       {
 		    E T1, T2, Td, Te;
diff --git a/rdft/scalar/r2cb/r2cbIII_9.c b/rdft/scalar/r2cb/r2cbIII_9.c
index 5fe4bcc..50b3c2b 100644
--- a/rdft/scalar/r2cb/r2cbIII_9.c
+++ b/rdft/scalar/r2cb/r2cbIII_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:04:19 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:35 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -50,7 +50,7 @@ static void r2cbIII_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E T4, Td, T3, Th, Tr, Tm, T7, Tc, Tj, Tg, T1, T2;
 	       Tg = Ci[WS(csi, 1)];
 	       T1 = Cr[WS(csr, 4)];
@@ -142,7 +142,7 @@ static void r2cbIII_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E T3, Ts, Ti, Td, Tc, T8, To, Tu, Tl, Tt, T9, Te;
 	       {
 		    E Th, T1, T2, Tf, Tg;
diff --git a/rdft/scalar/r2cb/r2cb_10.c b/rdft/scalar/r2cb/r2cb_10.c
index 1faacc0..f441529 100644
--- a/rdft/scalar/r2cb/r2cb_10.c
+++ b/rdft/scalar/r2cb/r2cb_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void r2cb_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E Tb, T3, Tc, T6, Tq, To, Ty, Tw, Td, T9;
 	       {
 		    E Tu, Tn, T7, Tv, Tk, T8;
@@ -134,7 +134,7 @@ static void r2cb_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP1_118033988, +1.118033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E T3, Tb, Tn, Tv, Tk, Tu, Ta, Ts, Te, Tg, Ti, Tj;
 	       {
 		    E T1, T2, Tl, Tm;
diff --git a/rdft/scalar/r2cb/r2cb_11.c b/rdft/scalar/r2cb/r2cb_11.c
index 1f34bd7..86d8702 100644
--- a/rdft/scalar/r2cb/r2cb_11.c
+++ b/rdft/scalar/r2cb/r2cb_11.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -49,7 +49,7 @@ static void r2cb_11(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP521108558, +0.521108558113202722944698153526659300680427422);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(44, rs), MAKE_VOLATILE_STRIDE(44, csr), MAKE_VOLATILE_STRIDE(44, csi)) {
 	       E Tf, Tq, Tt, Tu;
 	       {
 		    E T1, Td, Th, Te, Tg, T2, Ts, TK, TB, TT, Tj, T6, T3, T4, T5;
@@ -183,7 +183,7 @@ static void r2cb_11(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP1_081281634, +1.081281634911195164215271908637383390863541216);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(44, rs), MAKE_VOLATILE_STRIDE(44, csr), MAKE_VOLATILE_STRIDE(44, csi)) {
 	       E Td, Tl, Tf, Th, Tj, T1, T2, T6, T5, T4, T3, T7, Tk, Te, Tg;
 	       E Ti;
 	       {
diff --git a/rdft/scalar/r2cb/r2cb_12.c b/rdft/scalar/r2cb/r2cb_12.c
index 4de6b28..37cfcba 100644
--- a/rdft/scalar/r2cb/r2cb_12.c
+++ b/rdft/scalar/r2cb/r2cb_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:53 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cb_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E Ts, Tr;
 	       {
 		    E Tz, Te, Tn, Tk, Tc, Tw, Ty, Th, T4, T3, Td, T5;
@@ -131,7 +131,7 @@ static void r2cb_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E T8, Tb, Tm, TA, Tw, Tx, Tp, TB, T3, Tr, Tg, T6, Ts, Tk;
 	       {
 		    E T9, Ta, Tn, To;
diff --git a/rdft/scalar/r2cb/r2cb_128.c b/rdft/scalar/r2cb/r2cb_128.c
index 2a74024..690cdc8 100644
--- a/rdft/scalar/r2cb/r2cb_128.c
+++ b/rdft/scalar/r2cb/r2cb_128.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:09 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -74,7 +74,7 @@ static void r2cb_128(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(512, rs), MAKE_VOLATILE_STRIDE(512, csr), MAKE_VOLATILE_STRIDE(512, csi)) {
 	       E T9H, T9I, T9X, T9Y;
 	       {
 		    E Tdr, T9, Tcl, Ta9, T6b, T2d, T91, T7j, Tg, Tds, Tcm, Tae, T92, T7m, T6c;
@@ -1617,7 +1617,7 @@ static void r2cb_128(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(512, rs), MAKE_VOLATILE_STRIDE(512, csr), MAKE_VOLATILE_STRIDE(512, csi)) {
 	       E Ta, T6q, T2a, T5k, T8x, Tbx, TcF, Ten, Th, T6r, T2j, T5l, T8E, Tby, TcI;
 	       E Teo, Tx, T6t, TcM, Teq, TcP, Ter, T2t, T5n, T2C, T5o, T8Q, TbA, T8X, TbB;
 	       E T6w, T7L, T1j, T6L, Tde, TeC, TdL, TeR, T3v, T5z, T4I, T5O, T9O, TbM, TaV;
diff --git a/rdft/scalar/r2cb/r2cb_13.c b/rdft/scalar/r2cb/r2cb_13.c
index 2e6ab27..2be6af7 100644
--- a/rdft/scalar/r2cb/r2cb_13.c
+++ b/rdft/scalar/r2cb/r2cb_13.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:53 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:08 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -64,7 +64,7 @@ static void r2cb_13(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(52, rs), MAKE_VOLATILE_STRIDE(52, csr), MAKE_VOLATILE_STRIDE(52, csi)) {
 	       E TW, T14, TS, TO, T18, T1e, TY, TX, TQ, Tq, TP, Tl, T1d, Tr;
 	       {
 		    E T1, TN, T16, TJ, TV, TG, TU, Tf, T2, T3, Tb, Ti, T4;
@@ -232,7 +232,7 @@ static void r2cb_13(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(52, rs), MAKE_VOLATILE_STRIDE(52, csr), MAKE_VOLATILE_STRIDE(52, csi)) {
 	       E TG, TS, TR, T15, TJ, TT, T1, Tm, Tc, Td, Tg, Tj, Tk, Tn, To;
 	       E Tp;
 	       {
diff --git a/rdft/scalar/r2cb/r2cb_14.c b/rdft/scalar/r2cb/r2cb_14.c
index 4679040..c0e933a 100644
--- a/rdft/scalar/r2cb/r2cb_14.c
+++ b/rdft/scalar/r2cb/r2cb_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:53 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:08 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void r2cb_14(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP554958132, +0.554958132087371191422194871006410481067288862);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(56, rs), MAKE_VOLATILE_STRIDE(56, csr), MAKE_VOLATILE_STRIDE(56, csi)) {
 	       E Te, TO, TT, TG, TJ, TD, TR, TE;
 	       {
 		    E T3, TK, To, TM, Tu, TL, Tr, TS, TA, TN, TX, TF, Tv, T7, Tf;
@@ -172,7 +172,7 @@ static void r2cb_14(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(56, rs), MAKE_VOLATILE_STRIDE(56, csr), MAKE_VOLATILE_STRIDE(56, csi)) {
 	       E T3, Td, T6, Te, Tq, Tz, Tn, Ty, Tc, Tg, Tk, Tx, T9, Tf, T1;
 	       E T2;
 	       T1 = Cr[0];
diff --git a/rdft/scalar/r2cb/r2cb_15.c b/rdft/scalar/r2cb/r2cb_15.c
index d36e50d..ad5d3bf 100644
--- a/rdft/scalar/r2cb/r2cb_15.c
+++ b/rdft/scalar/r2cb/r2cb_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:53 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:08 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -47,7 +47,7 @@ static void r2cb_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E TL, Tz, TM, TK;
 	       {
 		    E T3, Th, Tt, TD, TI, TH, TY, TC, TZ, Tu, Tm, Tv, Tr, Te, TW;
@@ -179,7 +179,7 @@ static void r2cb_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E T3, Tu, Ti, TB, TZ, T10, TE, TG, TJ, Tn, Tv, Ts, Tw, T8, Td;
 	       E Te;
 	       {
diff --git a/rdft/scalar/r2cb/r2cb_16.c b/rdft/scalar/r2cb/r2cb_16.c
index da72218..76ace1f 100644
--- a/rdft/scalar/r2cb/r2cb_16.c
+++ b/rdft/scalar/r2cb/r2cb_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:08 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cb_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E TN, TS, TF, TI;
 	       {
 		    E T8, TD, Tj, TL, T5, TM, TE, To, Td, Tq, Tc, TP, Ty, Te, Tr;
@@ -172,7 +172,7 @@ static void r2cb_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E T9, TS, Tl, TG, T6, TR, Ti, TD, Td, Tq, Tg, Tt, Tn, Tu, TV;
 	       E TU, TN, TK;
 	       {
diff --git a/rdft/scalar/r2cb/r2cb_2.c b/rdft/scalar/r2cb/r2cb_2.c
index 7dee848..c28fc34 100644
--- a/rdft/scalar/r2cb/r2cb_2.c
+++ b/rdft/scalar/r2cb/r2cb_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void r2cb_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = Cr[0];
 	       T2 = Cr[WS(csr, 1)];
@@ -69,7 +69,7 @@ static void r2cb_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = Cr[0];
 	       T2 = Cr[WS(csr, 1)];
diff --git a/rdft/scalar/r2cb/r2cb_20.c b/rdft/scalar/r2cb/r2cb_20.c
index 8f60f02..89bd78e 100644
--- a/rdft/scalar/r2cb/r2cb_20.c
+++ b/rdft/scalar/r2cb/r2cb_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:55 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:09 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void r2cb_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E TY, T1o, T1m, T14, T12, TX, T1n, T1j, TZ, T13;
 	       {
 		    E Tr, TD, Tl, T5, T1a, T1l, T1d, T1k, TT, T10, TO, T11, TE, TF, Tk;
@@ -207,7 +207,7 @@ static void r2cb_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E T6, TF, Tm, Tt, TQ, T1n, T1f, T12, T1m, TV, T13, T1c, Td, Tk, Tl;
 	       E Ty, TD, TE, Tn, To, Tp, TG, TH, TI;
 	       {
diff --git a/rdft/scalar/r2cb/r2cb_25.c b/rdft/scalar/r2cb/r2cb_25.c
index 9b04396..773ab72 100644
--- a/rdft/scalar/r2cb/r2cb_25.c
+++ b/rdft/scalar/r2cb/r2cb_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:55 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:10 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -82,7 +82,7 @@ static void r2cb_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E T1H, T24, T22, T1W, T1Y, T1X, T1Z, T23;
 	       {
 		    E T1G, Tu, T5, T1F, Tr, Te, T2o, T1N, T2a, T1t, TR, T1K, T29, T1u, TG;
@@ -362,7 +362,7 @@ static void r2cb_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP1_902113032, +1.902113032590307144232878666758764286811397268);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E Tu, T1G, T5, Tr, T1F, TN, TO, Te, TR, T27, T1r, T1N, TG, T26, T1q;
 	       E T1K, T1a, T1b, Tn, T1e, T2a, T1u, T1U, T13, T29, T1t, T1R, Ts, Tt;
 	       Ts = Ci[WS(csi, 5)];
diff --git a/rdft/scalar/r2cb/r2cb_3.c b/rdft/scalar/r2cb/r2cb_3.c
index 8df340b..3338e3e 100644
--- a/rdft/scalar/r2cb/r2cb_3.c
+++ b/rdft/scalar/r2cb/r2cb_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cb_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T4, T1, T2, T3;
 	       T4 = Ci[WS(csi, 1)];
 	       T1 = Cr[0];
@@ -76,7 +76,7 @@ static void r2cb_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T5, T1, T2, T3, T4;
 	       T4 = Ci[WS(csi, 1)];
 	       T5 = KP1_732050807 * T4;
diff --git a/rdft/scalar/r2cb/r2cb_32.c b/rdft/scalar/r2cb/r2cb_32.c
index b2b05a8..9875019 100644
--- a/rdft/scalar/r2cb/r2cb_32.c
+++ b/rdft/scalar/r2cb/r2cb_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:09 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -47,7 +47,7 @@ static void r2cb_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T1F, T1C, T1H, T1z, T1G, T1I;
 	       {
 		    E T8, T1t, Tz, T1R, T5, T1S, T1u, TE, T1w, TP, T1U, Tg, T2m, T1X, T1x;
@@ -336,7 +336,7 @@ static void r2cb_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T9, T2c, TB, T1y, T6, T2b, Ty, T1v, Th, T2e, T2f, TD, TK, T1C, T1F;
 	       E T1h, Tp, T2i, T2m, TN, T13, T1K, T1Y, T1k, Tw, TU, T1l, TW, T1V, T2j;
 	       E T1R, T2l;
diff --git a/rdft/scalar/r2cb/r2cb_4.c b/rdft/scalar/r2cb/r2cb_4.c
index 6d336bf..2645033 100644
--- a/rdft/scalar/r2cb/r2cb_4.c
+++ b/rdft/scalar/r2cb/r2cb_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void r2cb_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T4, T6, T1, T2, T3, T5;
 	       T4 = Cr[WS(csr, 1)];
 	       T6 = Ci[WS(csi, 1)];
@@ -77,7 +77,7 @@ static void r2cb_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T5, T8, T3, T6;
 	       {
 		    E T4, T7, T1, T2;
diff --git a/rdft/scalar/r2cb/r2cb_5.c b/rdft/scalar/r2cb/r2cb_5.c
index bb7313c..8552553 100644
--- a/rdft/scalar/r2cb/r2cb_5.c
+++ b/rdft/scalar/r2cb/r2cb_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void r2cb_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E T1, T2, T3, Tc, Ta, T8, T9;
 	       T8 = Ci[WS(csi, 1)];
 	       T9 = Ci[WS(csi, 2)];
@@ -95,7 +95,7 @@ static void r2cb_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP1_175570504, +1.175570504584946258337411909278145537195304875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E Ta, Tc, T1, T4, T5, T6, Tb, T7;
 	       {
 		    E T8, T9, T2, T3;
diff --git a/rdft/scalar/r2cb/r2cb_6.c b/rdft/scalar/r2cb/r2cb_6.c
index 8df74b8..e5f3243 100644
--- a/rdft/scalar/r2cb/r2cb_6.c
+++ b/rdft/scalar/r2cb/r2cb_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cb_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T4, T7, T3, Te, Tc, T5;
 	       {
 		    E T1, T2, Ta, Tb;
@@ -95,7 +95,7 @@ static void r2cb_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T3, T7, Tc, Te, T6, T8, T1, T2, T9, Td;
 	       T1 = Cr[0];
 	       T2 = Cr[WS(csr, 3)];
diff --git a/rdft/scalar/r2cb/r2cb_64.c b/rdft/scalar/r2cb/r2cb_64.c
index aefcabe..08c6237 100644
--- a/rdft/scalar/r2cb/r2cb_64.c
+++ b/rdft/scalar/r2cb/r2cb_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:09 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -56,7 +56,7 @@ static void r2cb_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E T3d, T32, T37, T2Z, T3f, T3b, T3c, T35;
 	       {
 		    E T5H, T9, T5j, T4p, T2T, T1b, T3Z, T3j, Tg, T5I, T5k, T4u, T40, T3m, T2U;
@@ -727,7 +727,7 @@ static void r2cb_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E Ta, T2S, T18, T2u, T3F, T4V, T5l, T61, Th, T2T, T1h, T2v, T3M, T4W, T5o;
 	       E T62, T3Q, T5q, T5u, T44, Tp, Tw, T2V, T2W, T2X, T2Y, T3X, T5t, T1r, T2x;
 	       E T41, T5r, T1A, T2y, T4a, T5y, T5N, T4H, TN, T31, T4E, T5z, T39, T3q, T1L;
diff --git a/rdft/scalar/r2cb/r2cb_7.c b/rdft/scalar/r2cb/r2cb_7.c
index 56d1ca7..798b39a 100644
--- a/rdft/scalar/r2cb/r2cb_7.c
+++ b/rdft/scalar/r2cb/r2cb_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void r2cb_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP554958132, +0.554958132087371191422194871006410481067288862);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E Tn, Td, Tg, Ti, Tl, T8;
 	       {
 		    E T1, T9, Tb, Ta, T2, T4, Th, Tm, Tc, T3, Te;
@@ -115,7 +115,7 @@ static void r2cb_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP1_563662964, +1.563662964936059617416889053348115500464669037);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E T9, Td, Tb, T1, T4, T2, T3, T5, Tc, Ta, T6, T8, T7;
 	       T6 = Ci[WS(csi, 2)];
 	       T8 = Ci[WS(csi, 1)];
diff --git a/rdft/scalar/r2cb/r2cb_8.c b/rdft/scalar/r2cb/r2cb_8.c
index a0bb017..74d47af 100644
--- a/rdft/scalar/r2cb/r2cb_8.c
+++ b/rdft/scalar/r2cb/r2cb_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cb_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E Th, Tb, Tg, Ti;
 	       {
 		    E T4, Ta, Td, T9, T3, Tc, T8, Te;
@@ -106,7 +106,7 @@ static void r2cb_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E T5, Tg, T3, Te, T9, Ti, Td, Tj, T6, Ta;
 	       {
 		    E T4, Tf, T1, T2;
diff --git a/rdft/scalar/r2cb/r2cb_9.c b/rdft/scalar/r2cb/r2cb_9.c
index 017346f..8cabcf0 100644
--- a/rdft/scalar/r2cb/r2cb_9.c
+++ b/rdft/scalar/r2cb/r2cb_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:41:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -50,7 +50,7 @@ static void r2cb_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP2_000000000, +2.000000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E T4, Th, T3, Tb, Tp, Tk, T7, Tf, Ti, Ta, T1, T2;
 	       Ta = Ci[WS(csi, 3)];
 	       T1 = Cr[0];
@@ -142,7 +142,7 @@ static void r2cb_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP1_732050807, +1.732050807568877293527446341505872366942805254);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ovs, R1 = R1 + ovs, Cr = Cr + ivs, Ci = Ci + ivs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E T3, Tq, Tc, Tk, Tj, T8, Tm, Ts, Th, Tr, Tw, Tx;
 	       {
 		    E Tb, T1, T2, T9, Ta;
diff --git a/rdft/scalar/r2cf/Makefile.in b/rdft/scalar/r2cf/Makefile.in
index b4600b4..e86f4a8 100644
--- a/rdft/scalar/r2cf/Makefile.in
+++ b/rdft/scalar/r2cf/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -32,6 +32,23 @@
 # Most of this file requires GNU make .
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -117,6 +134,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_scalar_r2cf_la_SOURCES)
 DIST_SOURCES = $(librdft_scalar_r2cf_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/rdft/scalar/r2cf/hc2cf2_16.c b/rdft/scalar/r2cf/hc2cf2_16.c
index 1bdfd30..29718d1 100644
--- a/rdft/scalar/r2cf/hc2cf2_16.c
+++ b/rdft/scalar/r2cf/hc2cf2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:31 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:42 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cf2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T3S, T3R;
 	       {
 		    E T2, Tf, TM, TO, T3, Tg, TN, TS, T4, Tp, T6, T5, Th;
@@ -450,7 +450,7 @@ static void hc2cf2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T2, T5, Tg, Ti, Tk, To, TE, TC, T6, T3, T8, TW, TJ, Tt, TU;
 	       E Tc, Tx, TH, TN, TO, TP, TR, T1f, T1k, T1b, T1i, T1y, T1H, T1u, T1F;
 	       {
diff --git a/rdft/scalar/r2cf/hc2cf2_20.c b/rdft/scalar/r2cf/hc2cf2_20.c
index a1133ef..0ae08d7 100644
--- a/rdft/scalar/r2cf/hc2cf2_20.c
+++ b/rdft/scalar/r2cf/hc2cf2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:32 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:44 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cf2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T59, T5i, T5k, T5e, T5c, T5d, T5j, T5f;
 	       {
 		    E T2, Th, Tf, T6, T5, Tl, T1p, T1n, Ti, T3, Tt, Tv, T24, T1f, T1D;
@@ -586,7 +586,7 @@ static void hc2cf2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T2, T5, Tg, Ti, Tk, To, T1h, T1f, T6, T3, T8, T14, T1Q, Tc, T1O;
 	       E T1v, T18, T1t, T1n, T24, T1j, T22, Tq, Tu, T1E, T1G, Tx, Ty, Tz, TJ;
 	       E T1Z, TB, T1X, T1A, TZ, TL, T1y, TX;
diff --git a/rdft/scalar/r2cf/hc2cf2_32.c b/rdft/scalar/r2cf/hc2cf2_32.c
index ee5b127..b7c6146 100644
--- a/rdft/scalar/r2cf/hc2cf2_32.c
+++ b/rdft/scalar/r2cf/hc2cf2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:32 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:43 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cf2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T9A, T9z;
 	       {
 		    E T2, T8, T3, T6, Te, Tr, T18, T4, Ta, Tz, T1n, T10, Ti, T5, Tc;
@@ -980,7 +980,7 @@ static void hc2cf2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T2, T5, T3, T6, T8, TM, TO, Td, T9, Te, Th, Tl, TD, TH, T1y;
 	       E T1H, T15, T1A, T11, T1F, T1n, T1p, T2q, T2I, T2u, T2K, T2V, T3b, T2Z, T3d;
 	       E Tu, Ty, T3l, T3n, T1t, T1v, T2f, T2h, T1a, T1e, T32, T34, T1W, T1Y, T2C;
diff --git a/rdft/scalar/r2cf/hc2cf2_4.c b/rdft/scalar/r2cf/hc2cf2_4.c
index e0759f6..bc3f89e 100644
--- a/rdft/scalar/r2cf/hc2cf2_4.c
+++ b/rdft/scalar/r2cf/hc2cf2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:31 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:42 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cf2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Ti, Tq, To, Te, Ty, TA, Tm, Ts;
 	       {
 		    E T2, T6, T3, T5;
@@ -129,7 +129,7 @@ static void hc2cf2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T2, T4, T3, T5, T6, T8;
 	       T2 = W[0];
 	       T4 = W[1];
diff --git a/rdft/scalar/r2cf/hc2cf2_8.c b/rdft/scalar/r2cf/hc2cf2_8.c
index 5d4a180..83cf72a 100644
--- a/rdft/scalar/r2cf/hc2cf2_8.c
+++ b/rdft/scalar/r2cf/hc2cf2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:31 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:42 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cf2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E TS, T1m, TJ, T1l, T1k, Tw, T1w, T1u;
 	       {
 		    E T2, T3, Tl, Tn, T5, T4, Tm, Tr, T6;
@@ -230,7 +230,7 @@ static void hc2cf2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T2, T5, T3, T6, T8, Tc, Tg, Ti, Tl, Tm, Tn, Tz, Tp, Tx;
 	       {
 		    E T4, Tb, T7, Ta;
diff --git a/rdft/scalar/r2cf/hc2cf_10.c b/rdft/scalar/r2cf/hc2cf_10.c
index fc23516..9a07a63 100644
--- a/rdft/scalar/r2cf/hc2cf_10.c
+++ b/rdft/scalar/r2cf/hc2cf_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cf_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T1X, T21, T20, T22;
 	       {
 		    E T26, T1U, T8, T12, T1n, T1P, T24, T1K, T1Y, T18, T10, T2b, T1H, T23, T15;
@@ -293,7 +293,7 @@ static void hc2cf_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T7, T1O, TT, T1C, TF, TQ, TR, T1r, T1s, T1L, TX, TY, TZ, T16, T19;
 	       E T1y, Ti, Tt, Tu, T1o, T1p, T1M, TU, TV, TW, T1d, T1g, T1x;
 	       {
diff --git a/rdft/scalar/r2cf/hc2cf_12.c b/rdft/scalar/r2cf/hc2cf_12.c
index 425c3f9..bc92f7a 100644
--- a/rdft/scalar/r2cf/hc2cf_12.c
+++ b/rdft/scalar/r2cf/hc2cf_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:31 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cf_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E T2n, T2u;
 	       {
 		    E T1, T2i, T2e, Tl, T1Y, T10, T1S, TG, T2f, T1s, T2s, Ty, T1Z, T1H, T21;
@@ -309,7 +309,7 @@ static void hc2cf_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E T1, T1W, T18, T22, Tc, T15, T1V, T23, TR, T1E, T1o, T1D, T12, T1l, T1F;
 	       E T1G, Ti, T1S, T1d, T25, Tt, T1a, T1T, T26, TA, T1y, T1j, T1B, TL, T1g;
 	       E T1z, T1A;
diff --git a/rdft/scalar/r2cf/hc2cf_16.c b/rdft/scalar/r2cf/hc2cf_16.c
index 48ff89c..562a92b 100644
--- a/rdft/scalar/r2cf/hc2cf_16.c
+++ b/rdft/scalar/r2cf/hc2cf_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:22 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:31 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cf_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T3G, T3F;
 	       {
 		    E T3z, T3o, T8, T1I, T2p, T35, T2r, T1s, T2w, T36, T2k, T1F, T3k, T1N, T3A;
@@ -423,7 +423,7 @@ static void hc2cf_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T7, T37, T1t, T2U, Ti, T38, T1w, T2R, Tu, T2s, T1C, T2c, TF, T2t, T1H;
 	       E T2d, T1f, T1q, T2B, T2C, T2D, T2E, T1Z, T2j, T24, T2k, TS, T13, T2w, T2x;
 	       E T2y, T2z, T1O, T2g, T1T, T2h;
diff --git a/rdft/scalar/r2cf/hc2cf_2.c b/rdft/scalar/r2cf/hc2cf_2.c
index 2ed0042..2f15a9f 100644
--- a/rdft/scalar/r2cf/hc2cf_2.c
+++ b/rdft/scalar/r2cf/hc2cf_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cf_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T1, Ta, T3, T6, T2, T5;
 	       T1 = Rp[0];
 	       Ta = Rm[0];
@@ -86,7 +86,7 @@ static void hc2cf_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T1, T8, T6, T7;
 	       T1 = Rp[0];
 	       T8 = Rm[0];
diff --git a/rdft/scalar/r2cf/hc2cf_20.c b/rdft/scalar/r2cf/hc2cf_20.c
index dac7c6f..c574c92 100644
--- a/rdft/scalar/r2cf/hc2cf_20.c
+++ b/rdft/scalar/r2cf/hc2cf_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:29 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:41 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cf_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T4P, T4Y, T50, T4U, T4S, T4T, T4Z, T4V;
 	       {
 		    E T4N, T4r, T8, T2i, T4n, T2n, T4O, Tl, T2v, T3v, T3T, T4f, TN, T2b, T3F;
@@ -561,7 +561,7 @@ static void hc2cf_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E Tj, T1R, T4j, T4s, T2q, T37, T3Q, T42, T1r, T1O, T1P, T3p, T3s, T3K, T3A;
 	       E T3B, T3Z, T1V, T1W, T1X, T23, T28, T4q, T2W, T2X, T4f, T33, T34, T35, T2G;
 	       E T2L, T2M, TG, T13, T14, T3i, T3l, T3J, T3D, T3E, T40, T1S, T1T, T1U, T2e;
diff --git a/rdft/scalar/r2cf/hc2cf_32.c b/rdft/scalar/r2cf/hc2cf_32.c
index 6eae895..f5e9f6e 100644
--- a/rdft/scalar/r2cf/hc2cf_32.c
+++ b/rdft/scalar/r2cf/hc2cf_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:24 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:33 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cf_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T90, T8Z;
 	       {
 		    E T8x, T87, T8, T3w, T83, T3B, T8y, Tl, T6F, Tz, T3J, T5T, T6G, TM, T3Q;
@@ -927,7 +927,7 @@ static void hc2cf_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb,
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tj, T5F, T7C, T7Q, T35, T4T, T78, T7m, T1Q, T61, T5Y, T6J, T3K, T59, T41;
 	       E T56, T2B, T67, T6e, T6O, T4b, T5d, T4s, T5g, TG, T7l, T5I, T73, T3a, T4U;
 	       E T3f, T4V, T14, T5N, T5M, T6E, T3m, T4Y, T3r, T4Z, T1r, T5P, T5S, T6F, T3x;
diff --git a/rdft/scalar/r2cf/hc2cf_4.c b/rdft/scalar/r2cf/hc2cf_4.c
index 4def811..4276c0b 100644
--- a/rdft/scalar/r2cf/hc2cf_4.c
+++ b/rdft/scalar/r2cf/hc2cf_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cf_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E To, Te, Tm, T8, Tw, Ty, Tq, Tk;
 	       {
 		    E T1, Tv, Tu, T7, Tg, Tj, Tf, Ti, Tp, Th;
@@ -123,7 +123,7 @@ static void hc2cf_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T1, Tp, T6, To, Tc, Tk, Th, Tl;
 	       T1 = Rp[0];
 	       Tp = Rm[0];
diff --git a/rdft/scalar/r2cf/hc2cf_6.c b/rdft/scalar/r2cf/hc2cf_6.c
index e0e4d5f..ea08e6f 100644
--- a/rdft/scalar/r2cf/hc2cf_6.c
+++ b/rdft/scalar/r2cf/hc2cf_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cf_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E TY, TU, T10, TZ;
 	       {
 		    E T1, TX, TW, T7, Tn, Tq, TJ, TS, TB, Tl, To, TK, Tt, Tw, Ts;
@@ -173,7 +173,7 @@ static void hc2cf_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T7, TS, Tv, TO, Tt, TJ, Tx, TF, Ti, TI, Tw, TC;
 	       {
 		    E T1, TN, T6, TM;
diff --git a/rdft/scalar/r2cf/hc2cf_8.c b/rdft/scalar/r2cf/hc2cf_8.c
index 714e568..cfeb379 100644
--- a/rdft/scalar/r2cf/hc2cf_8.c
+++ b/rdft/scalar/r2cf/hc2cf_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:21 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cf_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1g, T1f, T1e, Tm, T1q, T1o, T1p, TN, T1h, T1i;
 	       {
 		    E T1, T1m, T1l, T7, TS, Tk, TQ, Te, To, Tr, T17, TM, T12, Tu, TW;
@@ -209,7 +209,7 @@ static void hc2cf_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb, I
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T1e, TH, T19, TF, T13, TR, TU, Ti, T1f, TK, T16, Tu, T12, TM;
 	       E TP;
 	       {
diff --git a/rdft/scalar/r2cf/hc2cfdft2_16.c b/rdft/scalar/r2cf/hc2cfdft2_16.c
index c890f5e..835ccd2 100644
--- a/rdft/scalar/r2cf/hc2cfdft2_16.c
+++ b/rdft/scalar/r2cf/hc2cfdft2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cfdft2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T4p, T4o, T4n, T4s;
 	       {
 		    E T1, T2, Tw, Ty, Th, T3, Tx, TE, Ti, TK, Tj, T4, T5;
@@ -485,7 +485,7 @@ static void hc2cfdft2_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T1, T4, T2, T5, T7, Td, T12, TY, Tk, Ti, Tm, T1l, T1b, TL, T1h;
 	       E Ts, TR, T17, Ty, Tz, TA, TE, T1L, T1Q, T1H, T1O, T24, T2d, T20, T2b;
 	       {
diff --git a/rdft/scalar/r2cf/hc2cfdft2_20.c b/rdft/scalar/r2cf/hc2cfdft2_20.c
index 5c1bad3..4b4b5d0 100644
--- a/rdft/scalar/r2cf/hc2cfdft2_20.c
+++ b/rdft/scalar/r2cf/hc2cfdft2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:53 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void hc2cfdft2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T5h, T5C, T5E, T5y, T5w, T5x, T5D, T5z;
 	       {
 		    E Tm, Tq, Tn, T1, T6, Tg, Tp, Tb, T1i, TU, Tr, TW, Tx, T2B, T1A;
@@ -638,7 +638,7 @@ static void hc2cfdft2_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP475528258, +0.475528258147576786058219666689691071702849317);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T4, T7, Tm, To, Tq, Tu, T1I, T1G, T8, T5, Ta, T1u, T2u, Tg, T2s;
 	       E T21, T1A, T1Z, T1O, T2I, T1K, T2G, Tw, TC, T2a, T2e, TH, TI, TJ, TX;
 	       E T2D, TN, T2B, T26, T1n, TZ, T24, T1j;
diff --git a/rdft/scalar/r2cf/hc2cfdft2_32.c b/rdft/scalar/r2cf/hc2cfdft2_32.c
index dfadbc4..ce07b29 100644
--- a/rdft/scalar/r2cf/hc2cfdft2_32.c
+++ b/rdft/scalar/r2cf/hc2cfdft2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -46,7 +46,7 @@ static void hc2cfdft2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tax, TaA;
 	       {
 		    E T1, Th, T2, T5, Ti, Ty, T1t, T3, Tb, Tj, TY, TK, Tl, T4, Tk;
@@ -1055,7 +1055,7 @@ static void hc2cfdft2_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 8, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T1, T4, T2, T5, T7, T1b, T1d, Td, Ti, Tk, Tj, Tl, TL, TR, T2h;
 	       E T2O, T16, T2l, T10, T2K, Tm, Tq, T3s, T3K, T3w, T3M, T4e, T4u, T4i, T4w;
 	       E Ty, TE, T3h, T3j, T2q, T2u, T4l, T4n, T1v, T1B, T3E, T3G, T2B, T2F, T3Y;
diff --git a/rdft/scalar/r2cf/hc2cfdft2_4.c b/rdft/scalar/r2cf/hc2cfdft2_4.c
index d940320..69c87e9 100644
--- a/rdft/scalar/r2cf/hc2cfdft2_4.c
+++ b/rdft/scalar/r2cf/hc2cfdft2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cfdft2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T1, T5, T2, T4;
 	       T1 = W[0];
 	       T5 = W[3];
@@ -142,7 +142,7 @@ static void hc2cfdft2_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 4, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T1, T3, T2, T4, T5, T9;
 	       T1 = W[0];
 	       T3 = W[1];
diff --git a/rdft/scalar/r2cf/hc2cfdft2_8.c b/rdft/scalar/r2cf/hc2cfdft2_8.c
index 7ef13d2..d63abc4 100644
--- a/rdft/scalar/r2cf/hc2cfdft2_8.c
+++ b/rdft/scalar/r2cf/hc2cfdft2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdft2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1G, T1F, T1C, T1D, T1N, T1B, T1R, T1L;
 	       {
 		    E T1, T2, Th, Tj, T4, T3, Ti, Tp, T5;
@@ -256,7 +256,7 @@ static void hc2cfdft2_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1, T4, T2, T5, Tu, Ty, T7, Td, Ti, Tj, Tk, TP, To, TN;
 	       {
 		    E T3, Tc, T6, Tb;
diff --git a/rdft/scalar/r2cf/hc2cfdft_10.c b/rdft/scalar/r2cf/hc2cfdft_10.c
index 8e3aed3..70a78b8 100644
--- a/rdft/scalar/r2cf/hc2cfdft_10.c
+++ b/rdft/scalar/r2cf/hc2cfdft_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:34 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void hc2cfdft_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T1x, T1I, T1T, T22, T20;
 	       {
 		    E T3, T1u, T1S, T2f, Td, T1w, T14, T1p, T1j, T1q, T1N, T2e, T1z, To, T2i;
@@ -314,7 +314,7 @@ static void hc2cfdft_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP279508497, +0.279508497187473712051146708591409529430077295);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 18, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E Tw, TL, TM, T1W, T1X, T27, T1Z, T20, T26, TX, T1a, T1b, T1d, T1e, T1f;
 	       E T1q, T1t, T1u, T1x, T1A, T1B, T1g, T1h, T1i, Td, T25, T1k, T1F;
 	       {
diff --git a/rdft/scalar/r2cf/hc2cfdft_12.c b/rdft/scalar/r2cf/hc2cfdft_12.c
index 72bbbb4..3c2b21c 100644
--- a/rdft/scalar/r2cf/hc2cfdft_12.c
+++ b/rdft/scalar/r2cf/hc2cfdft_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:34 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdft_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E T2z, T2M;
 	       {
 		    E To, T1E, T2H, T1m, T1W, Tl, T1J, T2i, T2K, T1B, T2I, T2e, T19, T2E, T2C;
@@ -356,7 +356,7 @@ static void hc2cfdft_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP433012701, +0.433012701892219323381861585376468091735701313);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 22, MAKE_VOLATILE_STRIDE(48, rs)) {
 	       E Tm, T1t, T1d, T2j, Tj, T1Y, T1w, T1G, T1q, T2q, T1U, T2k, Tw, T1y, T17;
 	       E T2g, TP, T21, T1B, T1J, T12, T2u, T1P, T2h;
 	       {
diff --git a/rdft/scalar/r2cf/hc2cfdft_16.c b/rdft/scalar/r2cf/hc2cfdft_16.c
index 3c0a250..02dae42 100644
--- a/rdft/scalar/r2cf/hc2cfdft_16.c
+++ b/rdft/scalar/r2cf/hc2cfdft_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:46 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cfdft_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T4d, T4g;
 	       {
 		    E T1f, T2e, T3D, T1K, T2g, T1c, T3H, T2W, T2j, TR, T3E, T2R, T2l, T11, T3G;
@@ -480,7 +480,7 @@ static void hc2cfdft_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 30, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T19, T3h, T21, T2Y, T1o, T3d, T2s, T39, TW, T3i, T24, T2Z, T1z, T3c, T2p;
 	       E T3a, Tj, T2S, T28, T2R, T1L, T36, T2i, T32, TC, T2V, T2b, T2U, T1W, T35;
 	       E T2l, T33;
diff --git a/rdft/scalar/r2cf/hc2cfdft_2.c b/rdft/scalar/r2cf/hc2cfdft_2.c
index ba4b272..6dfac89 100644
--- a/rdft/scalar/r2cf/hc2cfdft_2.c
+++ b/rdft/scalar/r2cf/hc2cfdft_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:33 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:44 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cfdft_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T9, Ta, T3, Tc, T7, T4;
 	       {
 		    E T1, T2, T5, T6;
@@ -95,7 +95,7 @@ static void hc2cfdft_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 2, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T3, T9, T7, Tb;
 	       {
 		    E T1, T2, T5, T6;
diff --git a/rdft/scalar/r2cf/hc2cfdft_20.c b/rdft/scalar/r2cf/hc2cfdft_20.c
index 7bc43c6..b734e63 100644
--- a/rdft/scalar/r2cf/hc2cfdft_20.c
+++ b/rdft/scalar/r2cf/hc2cfdft_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:37 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:48 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void hc2cfdft_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T4X, T5i, T5k, T5e, T5c, T5d, T5j, T5f;
 	       {
 		    E T2E, T4W, T3v, T4k, T2M, T3w, T4V, T4j, T2p, T2T, T5a, T5A, T3D, T3o, T4b;
@@ -606,7 +606,7 @@ static void hc2cfdft_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP475528258, +0.475528258147576786058219666689691071702849317);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 38, MAKE_VOLATILE_STRIDE(80, rs)) {
 	       E T12, T2w, T4o, T4V, T2H, T3a, T4y, T4Y, T1z, T2v, T25, T2y, T2s, T2z, T4v;
 	       E T4X, T4r, T4U, T3A, T3Z, T2X, T37, T3k, T41, T2M, T39, T3v, T3Y, T2S, T36;
 	       E T3p, T42, Td, T4G, T33, T3N, Tw, T4H, T32, T3O;
diff --git a/rdft/scalar/r2cf/hc2cfdft_32.c b/rdft/scalar/r2cf/hc2cfdft_32.c
index 5527258..f52d9c6 100644
--- a/rdft/scalar/r2cf/hc2cfdft_32.c
+++ b/rdft/scalar/r2cf/hc2cfdft_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:35 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:46 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -46,7 +46,7 @@ static void hc2cfdft_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T9X, Ta0;
 	       {
 		    E T3B, T89, T61, T8l, T2F, T7p, T8t, T4B, T7I, T5e, T7L, T1n, T7R, T5E, T82;
@@ -1009,7 +1009,7 @@ static void hc2cfdft_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 62, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E T2S, T5K, T52, T5N, T7p, T8r, T7i, T8o, T2q, T7t, T45, T6L, T2d, T7u, T48;
 	       E T6M, T1A, T4c, T4f, T1T, T3f, T5M, T7e, T7l, T6J, T7x, T4V, T5J, T7b, T7k;
 	       E T6G, T7w, Tj, TC, T5r, T4k, T4n, T5s, T3D, T5C, T6V, T72, T4G, T5F, T6u;
diff --git a/rdft/scalar/r2cf/hc2cfdft_4.c b/rdft/scalar/r2cf/hc2cfdft_4.c
index 9208a37..4ea1e2a 100644
--- a/rdft/scalar/r2cf/hc2cfdft_4.c
+++ b/rdft/scalar/r2cf/hc2cfdft_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:33 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:44 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cfdft_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Td, Tu, Tr, T4, Tm, To, T9, T5, TA, Tp, Tv, TD, T6, Tq;
 	       {
 		    E Tk, Tl, Tf, TC, Tj, T7, T8, T1, Tn, Tb, Tc;
@@ -136,7 +136,7 @@ static void hc2cfdft_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E Tc, Tr, Tk, Tx, T9, Ts, Tp, Tw;
 	       {
 		    E Ta, Tb, Tj, Tf, Tg, Th, Te, Ti;
diff --git a/rdft/scalar/r2cf/hc2cfdft_6.c b/rdft/scalar/r2cf/hc2cfdft_6.c
index 5d9a2a1..d413113 100644
--- a/rdft/scalar/r2cf/hc2cfdft_6.c
+++ b/rdft/scalar/r2cf/hc2cfdft_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:33 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:44 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdft_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E TP, TT, TN, TM, TY, T13;
 	       {
 		    E T3, TQ, TJ, T12, Tu, TB, TX, T10, Tj, Tf, Ti, Td, Th, TU, TS;
@@ -197,7 +197,7 @@ static void hc2cfdft_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP433012701, +0.433012701892219323381861585376468091735701313);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 10, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T3, TM, Tc, TN, Ts, T10, TI, TR, TF, T11, TH, TU;
 	       {
 		    E T1, T2, TD, Tz, TA, TB, T7, Tf, Tb, Th, Tq, Tw, Tm, Tu, T4;
diff --git a/rdft/scalar/r2cf/hc2cfdft_8.c b/rdft/scalar/r2cf/hc2cfdft_8.c
index 6c8d7da..f256e96 100644
--- a/rdft/scalar/r2cf/hc2cfdft_8.c
+++ b/rdft/scalar/r2cf/hc2cfdft_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:34 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:44 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdft_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T1A, T1w, T1z, T1x, T1H, T1v, T1L, T1F;
 	       {
 		    E Ty, T14, TO, T1o, Tv, TG, T16, T1m, Ta, T19, T1h, TV, T10, TX, TZ;
@@ -248,7 +248,7 @@ static void hc2cfdft_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT mb
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, Rp = Rp + ms, Ip = Ip + ms, Rm = Rm - ms, Im = Im - ms, W = W + 14, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E Tv, TX, Ts, TY, TE, T1a, TJ, T19, T1l, T1m, T9, T10, Ti, T11, TP;
 	       E T16, TU, T17, T1i, T1j;
 	       {
diff --git a/rdft/scalar/r2cf/hf2_16.c b/rdft/scalar/r2cf/hf2_16.c
index 7d781fd..58e01ea 100644
--- a/rdft/scalar/r2cf/hf2_16.c
+++ b/rdft/scalar/r2cf/hf2_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:55 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:02 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hf2_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T35, T32;
 	       {
 		    E T2, Tf, TM, TO, T3, Tg, TN, TS, T4, Tp, T6, T5, Th;
@@ -447,7 +447,7 @@ static void hf2_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T2, T5, Tg, Ti, Tk, To, TE, TC, T6, T3, T8, TW, TJ, Tt, TU;
 	       E Tc, Tx, TH, TN, TO, TP, TR, T1f, T1k, T1b, T1i, T1y, T1H, T1u, T1F;
 	       {
diff --git a/rdft/scalar/r2cf/hf2_20.c b/rdft/scalar/r2cf/hf2_20.c
index b4a6703..3850cbe 100644
--- a/rdft/scalar/r2cf/hf2_20.c
+++ b/rdft/scalar/r2cf/hf2_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hf2_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T5o, T5u, T5w, T5q, T5n, T5p, T5v, T5r;
 	       {
 		    E T2, Th, Tf, T6, T5, Tl, T1p, T1n, Ti, T3, Tt, Tv, T24, T1f, T1D;
@@ -584,7 +584,7 @@ static void hf2_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T2, T5, Tg, Ti, Tk, To, T1h, T1f, T6, T3, T8, T14, T1Q, Tc, T1O;
 	       E T1v, T18, T1t, T1n, T24, T1j, T22, Tq, Tu, T1E, T1G, Tx, Ty, Tz, TJ;
 	       E T1Z, TB, T1X, T1A, TZ, TL, T1y, TX;
diff --git a/rdft/scalar/r2cf/hf2_25.c b/rdft/scalar/r2cf/hf2_25.c
index 15a491f..069152b 100644
--- a/rdft/scalar/r2cf/hf2_25.c
+++ b/rdft/scalar/r2cf/hf2_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:00 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:09 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -85,7 +85,7 @@ static void hf2_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T7M, T6S, T6Q, T7S, T7Q, T7L, T6R, T6J, T7N, T7R;
 	       {
 		    E T2, T8, T3, T6, Tk, Tv, TS, T4, Ta, TD, T2L, T10, Tm, T5, Tc;
@@ -915,7 +915,7 @@ static void hf2_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T2, T5, T3, T6, T8, Td, T16, T14, Te, T9, T21, T23, Tx, TR, T1g;
 	       E TB, T1f, TV, T1Q, Tg, T1S, Tk, T18, T2s, T1c, T2q, Tn, To, Tp, Tr;
 	       E T28, T2x, TY, T2k, T2m, T2v, TG, TE, T10, T1h, T1E, T26, T1B, T1G, T1V;
diff --git a/rdft/scalar/r2cf/hf2_32.c b/rdft/scalar/r2cf/hf2_32.c
index 6b1209b..21ca6cc 100644
--- a/rdft/scalar/r2cf/hf2_32.c
+++ b/rdft/scalar/r2cf/hf2_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:57 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:04 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hf2_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T7d, T7a;
 	       {
 		    E T2, T8, T3, T6, Te, Tr, T18, T4, Ta, Tz, T1n, T10, Ti, T5, Tc;
@@ -981,7 +981,7 @@ static void hf2_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T2, T5, T3, T6, T8, TM, TO, Td, T9, Te, Th, Tl, TD, TH, T1y;
 	       E T1H, T15, T1A, T11, T1F, T1n, T1p, T2q, T2I, T2u, T2K, T2V, T3b, T2Z, T3d;
 	       E Tu, Ty, T3l, T3n, T1t, T1v, T2f, T2h, T1a, T1e, T32, T34, T1W, T1Y, T2C;
diff --git a/rdft/scalar/r2cf/hf2_4.c b/rdft/scalar/r2cf/hf2_4.c
index 78ff302..2e066e2 100644
--- a/rdft/scalar/r2cf/hf2_4.c
+++ b/rdft/scalar/r2cf/hf2_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:02 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hf2_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E Ti, Tq, To, Te, TA, Ty, Tm, Ts;
 	       {
 		    E T2, T6, T3, T5;
@@ -129,7 +129,7 @@ static void hf2_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T2, T4, T3, T5, T6, T8;
 	       T2 = W[0];
 	       T4 = W[1];
diff --git a/rdft/scalar/r2cf/hf2_5.c b/rdft/scalar/r2cf/hf2_5.c
index 24c41a9..84ed277 100644
--- a/rdft/scalar/r2cf/hf2_5.c
+++ b/rdft/scalar/r2cf/hf2_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:58 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:07 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hf2_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E Ta, T1, TL, Tp, TT, Ti, TM, TC, To, TE, Ts, TF, T2, T8, T5;
 	       E TS, Tt, TG;
 	       T2 = W[0];
@@ -168,7 +168,7 @@ static void hf2_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T2, T4, T7, T9, Tb, Tl, Tf, Tj;
 	       {
 		    E T8, Te, Ta, Td;
diff --git a/rdft/scalar/r2cf/hf2_8.c b/rdft/scalar/r2cf/hf2_8.c
index 6397143..16e1c59 100644
--- a/rdft/scalar/r2cf/hf2_8.c
+++ b/rdft/scalar/r2cf/hf2_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:54 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:02 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hf2_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E TS, T1l, TJ, T1m, T1k, Tw, T1w, T1u;
 	       {
 		    E T2, T3, Tl, Tn, T5, T4, Tm, Tr, T6;
@@ -230,7 +230,7 @@ static void hf2_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T2, T5, T3, T6, T8, Tc, Tg, Ti, Tl, Tm, Tn, Tz, Tp, Tx;
 	       {
 		    E T4, Tb, T7, Ta;
diff --git a/rdft/scalar/r2cf/hf_10.c b/rdft/scalar/r2cf/hf_10.c
index ac25903..07b945d 100644
--- a/rdft/scalar/r2cf/hf_10.c
+++ b/rdft/scalar/r2cf/hf_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hf_10(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T29, T2d, T2c, T2e;
 	       {
 		    E T23, T1U, T8, T12, T1y, T1P, T25, T1H, T2b, T18, T10, T1Y, T1I, Tl, T13;
@@ -293,7 +293,7 @@ static void hf_10(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 18); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 18, MAKE_VOLATILE_STRIDE(20, rs)) {
 	       E T7, T1R, TT, T1C, TF, TQ, TR, T1o, T1p, T1P, TX, TY, TZ, T1d, T1g;
 	       E T1x, Ti, Tt, Tu, T1r, T1s, T1O, TU, TV, TW, T16, T19, T1y;
 	       {
diff --git a/rdft/scalar/r2cf/hf_12.c b/rdft/scalar/r2cf/hf_12.c
index 34c5329..92351c8 100644
--- a/rdft/scalar/r2cf/hf_12.c
+++ b/rdft/scalar/r2cf/hf_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hf_12(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T2u, T2n;
 	       {
 		    E T1, T2i, T2e, Tl, T1Y, T10, T1S, TG, T2f, T1s, T2s, Ty, T1Z, T1H, T21;
@@ -309,7 +309,7 @@ static void hf_12(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 22); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 22, MAKE_VOLATILE_STRIDE(24, rs)) {
 	       E T1, T1W, T18, T23, Tc, T15, T1V, T22, TR, T1E, T1o, T1D, T12, T1l, T1F;
 	       E T1G, Ti, T1S, T1d, T26, Tt, T1a, T1T, T25, TA, T1y, T1j, T1B, TL, T1g;
 	       E T1z, T1A;
diff --git a/rdft/scalar/r2cf/hf_15.c b/rdft/scalar/r2cf/hf_15.c
index b7c38fb..7308565 100644
--- a/rdft/scalar/r2cf/hf_15.c
+++ b/rdft/scalar/r2cf/hf_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:45 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:52 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void hf_15(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(30, rs)) {
 	       E T3v, T3E, T3G, T3A, T3y, T3z, T3F, T3B;
 	       {
 		    E T1G, T3l, T3H, T3k, T1B, Tf, T37, T1y, T2Y, T2M, T2a, T2i, T39, Tz, T2U;
@@ -444,7 +444,7 @@ static void hf_15(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 28); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 28, MAKE_VOLATILE_STRIDE(30, rs)) {
 	       E T1q, T2Q, Td, T1n, T2T, T3l, T13, T1k, T1l, T2E, T2F, T3j, T1H, T1T, T2k;
 	       E T2w, T2f, T2v, T1M, T1U, Tu, TL, TM, T2H, T2I, T3i, T1w, T1Q, T29, T2t;
 	       E T24, T2s, T1B, T1R;
diff --git a/rdft/scalar/r2cf/hf_16.c b/rdft/scalar/r2cf/hf_16.c
index 5b283fc..21eaa14 100644
--- a/rdft/scalar/r2cf/hf_16.c
+++ b/rdft/scalar/r2cf/hf_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:46 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:54 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hf_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T2T, T2Q;
 	       {
 		    E T3A, T3o, T8, T1I, T2w, T35, T2k, T1s, T2p, T36, T2r, T1F, T3k, T1N, T3z;
@@ -425,7 +425,7 @@ static void hf_16(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 30); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 30, MAKE_VOLATILE_STRIDE(32, rs)) {
 	       E T7, T38, T1t, T2U, Ti, T37, T1w, T2R, Tu, T2t, T1C, T2c, TF, T2s, T1H;
 	       E T2d, T1f, T1q, T2B, T2C, T2D, T2E, T1Z, T2k, T24, T2j, TS, T13, T2w, T2x;
 	       E T2y, T2z, T1O, T2h, T1T, T2g;
diff --git a/rdft/scalar/r2cf/hf_2.c b/rdft/scalar/r2cf/hf_2.c
index 3fbe7bc..87f81f9 100644
--- a/rdft/scalar/r2cf/hf_2.c
+++ b/rdft/scalar/r2cf/hf_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:49 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hf_2(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
 	       E T1, Ta, T3, T6, T2, T5;
 	       T1 = cr[0];
 	       Ta = ci[0];
@@ -86,7 +86,7 @@ static void hf_2(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 2); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 2, MAKE_VOLATILE_STRIDE(4, rs)) {
 	       E T1, T8, T6, T7;
 	       T1 = cr[0];
 	       T8 = ci[0];
diff --git a/rdft/scalar/r2cf/hf_20.c b/rdft/scalar/r2cf/hf_20.c
index b9dd152..33aa315 100644
--- a/rdft/scalar/r2cf/hf_20.c
+++ b/rdft/scalar/r2cf/hf_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:50 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:00 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hf_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E T54, T5a, T5c, T56, T53, T55, T5b, T57;
 	       {
 		    E T4N, T4q, T8, T2i, T4r, T2n, T4O, Tl, T2v, T3v, T43, T4b, TN, T2b, T3F;
@@ -559,7 +559,7 @@ static void hf_20(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 38); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 38, MAKE_VOLATILE_STRIDE(40, rs)) {
 	       E Tj, T1R, T4j, T4s, T2q, T37, T3Q, T42, T1r, T1O, T1P, T3i, T3l, T3J, T3D;
 	       E T3E, T44, T1V, T1W, T1X, T2e, T2j, T2k, T2W, T2X, T4f, T33, T34, T35, T2J;
 	       E T2O, T4q, TG, T13, T14, T3p, T3s, T3K, T3A, T3B, T43, T1S, T1T, T1U, T23;
diff --git a/rdft/scalar/r2cf/hf_25.c b/rdft/scalar/r2cf/hf_25.c
index 59b04cc..1907212 100644
--- a/rdft/scalar/r2cf/hf_25.c
+++ b/rdft/scalar/r2cf/hf_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:52 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:01 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -85,7 +85,7 @@ static void hf_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T7i, T6o, T6m, T7o, T7m, T7h, T6n, T6f, T7j, T7n;
 	       {
 		    E T6W, T5G, T3Y, T3M, T7q, T70, T6V, T7P, Tt, T3L, T5T, T45, T5Q, T4c, T3G;
@@ -859,7 +859,7 @@ static void hf_25(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 48); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 48, MAKE_VOLATILE_STRIDE(50, rs)) {
 	       E T1, T6b, T2l, T6g, To, T2m, T6e, T6f, T6a, T6H, T2u, T4I, T2i, T60, T3S;
 	       E T5D, T4r, T58, T3Z, T5C, T4q, T5b, TS, T5W, T2G, T5s, T4g, T4M, T2R, T5t;
 	       E T4h, T4P, T1l, T5X, T37, T5v, T4k, T4T, T3e, T5w, T4j, T4W, T1P, T5Z, T3v;
diff --git a/rdft/scalar/r2cf/hf_3.c b/rdft/scalar/r2cf/hf_3.c
index e8c2a05..b0370e8 100644
--- a/rdft/scalar/r2cf/hf_3.c
+++ b/rdft/scalar/r2cf/hf_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:49 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hf_3(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs)) {
 	       E T1, Tl, T9, Tc, Tb, Th, T7, Ti, Ta, Tj, Td;
 	       T1 = cr[0];
 	       Tl = ci[0];
@@ -109,7 +109,7 @@ static void hf_3(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 4); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 4, MAKE_VOLATILE_STRIDE(6, rs)) {
 	       E T1, Ti, T6, Te, Tb, Tf, Tc, Tj;
 	       T1 = cr[0];
 	       Ti = ci[0];
diff --git a/rdft/scalar/r2cf/hf_32.c b/rdft/scalar/r2cf/hf_32.c
index f58ebc4..c2d530b 100644
--- a/rdft/scalar/r2cf/hf_32.c
+++ b/rdft/scalar/r2cf/hf_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:47 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:54 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hf_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E T6D, T6A;
 	       {
 		    E T8y, T87, T8, T3w, T83, T3B, T8x, Tl, T6G, Tz, T3J, T5T, T6F, TM, T3Q;
@@ -925,7 +925,7 @@ static void hf_32(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 62); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 62, MAKE_VOLATILE_STRIDE(64, rs)) {
 	       E Tj, T5F, T7C, T7Q, T35, T4T, T78, T7m, T1Q, T61, T5Y, T6J, T3K, T56, T41;
 	       E T59, T2B, T67, T6e, T6O, T4b, T5g, T4s, T5d, TG, T7l, T5I, T73, T3a, T4U;
 	       E T3f, T4V, T14, T5K, T5N, T6F, T3m, T4Z, T3r, T4Y, T1r, T5P, T5S, T6E, T3x;
diff --git a/rdft/scalar/r2cf/hf_4.c b/rdft/scalar/r2cf/hf_4.c
index 7a31c1e..c55fcf6 100644
--- a/rdft/scalar/r2cf/hf_4.c
+++ b/rdft/scalar/r2cf/hf_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hf_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E To, Te, Tm, T8, Ty, Tw, Tq, Tk;
 	       {
 		    E T1, Tv, Tu, T7, Tg, Tj, Tf, Ti, Tp, Th;
@@ -123,7 +123,7 @@ static void hf_4(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 6); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 6, MAKE_VOLATILE_STRIDE(8, rs)) {
 	       E T1, Tp, T6, To, Tc, Tk, Th, Tl;
 	       T1 = cr[0];
 	       Tp = ci[0];
diff --git a/rdft/scalar/r2cf/hf_5.c b/rdft/scalar/r2cf/hf_5.c
index c3805ab..03547b0 100644
--- a/rdft/scalar/r2cf/hf_5.c
+++ b/rdft/scalar/r2cf/hf_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:42 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hf_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T1, TJ, TK, TA, TR, Te, TC, Tk, TE, Tq;
 	       {
 		    E Tg, Tj, Tm, TB, Th, Tp, Tl, Ti, To, TD, Tn;
@@ -161,7 +161,7 @@ static void hf_5(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 8); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 8, MAKE_VOLATILE_STRIDE(10, rs)) {
 	       E T1, TE, Tu, Tx, TC, TB, TF, TG, TH, Tc, Tn, To;
 	       T1 = cr[0];
 	       TE = ci[0];
diff --git a/rdft/scalar/r2cf/hf_6.c b/rdft/scalar/r2cf/hf_6.c
index 57fa5f1..056da65 100644
--- a/rdft/scalar/r2cf/hf_6.c
+++ b/rdft/scalar/r2cf/hf_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:43 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hf_6(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs)) {
 	       E T11, T12, T14, T13;
 	       {
 		    E T1, TV, TX, T7, Tn, Tq, TO, TR, TB, Tl, To, TH, Tt, Tw, Ts;
@@ -173,7 +173,7 @@ static void hf_6(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 10); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 10, MAKE_VOLATILE_STRIDE(12, rs)) {
 	       E T7, TS, Tv, TO, Tt, TJ, Tx, TF, Ti, TI, Tw, TC;
 	       {
 		    E T1, TM, T6, TN;
diff --git a/rdft/scalar/r2cf/hf_64.c b/rdft/scalar/r2cf/hf_64.c
index b2afef1..8a02eb5 100644
--- a/rdft/scalar/r2cf/hf_64.c
+++ b/rdft/scalar/r2cf/hf_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:47 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:55 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -53,7 +53,7 @@ static void hf_64(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tku, Tky, Tkt, Tkx;
 	       {
 		    E TiV, Tjm, T7e, TcA, TjR, Tkl, Tm, TeM, T7Q, TcI, TeZ, Thr, T1G, TeW, TcJ;
@@ -2003,7 +2003,7 @@ static void hf_64(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 126); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 126, MAKE_VOLATILE_STRIDE(128, rs)) {
 	       E Tj, TcL, ThT, Tin, T6b, Taz, TgT, Thn, TG, Thm, TcO, TgO, T6m, Tim, TaC;
 	       E ThQ, T14, Tfr, T6y, T9O, TaG, Tc0, TcU, TeE, T1r, Tfq, T6J, T9P, TaJ, Tc1;
 	       E TcZ, TeF, T1Q, T2d, Tfu, Tfv, Tfw, Tfx, T6Q, TaM, Tdb, TeI, T71, TaQ, T7a;
diff --git a/rdft/scalar/r2cf/hf_7.c b/rdft/scalar/r2cf/hf_7.c
index 7d060e5..bfd96f6 100644
--- a/rdft/scalar/r2cf/hf_7.c
+++ b/rdft/scalar/r2cf/hf_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:43 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:50 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void hf_7(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP356895867, +0.356895867892209443894399510021300583399127187);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(14, rs)) {
 	       E T1, TR, T18, T10, T12, T16, T11, T13;
 	       {
 		    E T19, T1a, T1i, Te, Tt, Tw, T1b, TM, T1h, Tr, Tu, TS, Tz, TC, Ty;
@@ -221,7 +221,7 @@ static void hf_7(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP781831482, +0.781831482468029808708444526674057750232334519);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 12); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 12, MAKE_VOLATILE_STRIDE(14, rs)) {
 	       E T1, TT, Tc, TV, TC, TO, Tn, TS, TI, TP, Ty, TU, TF, TQ;
 	       T1 = cr[0];
 	       TT = ci[0];
diff --git a/rdft/scalar/r2cf/hf_8.c b/rdft/scalar/r2cf/hf_8.c
index 976c330..9410cfd 100644
--- a/rdft/scalar/r2cf/hf_8.c
+++ b/rdft/scalar/r2cf/hf_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:44 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:51 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hf_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T1f, T1g, T1e, Tm, T1q, T1o, T1p, TN, T1h, T1i;
 	       {
 		    E T1, T1m, T1l, T7, TS, Tk, TQ, Te, To, Tr, T17, TM, T12, Tu, TW;
@@ -209,7 +209,7 @@ static void hf_8(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 14); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 14, MAKE_VOLATILE_STRIDE(16, rs)) {
 	       E T7, T1f, TH, T19, TF, T12, TR, TU, Ti, T1e, TK, T16, Tu, T13, TM;
 	       E TP;
 	       {
diff --git a/rdft/scalar/r2cf/hf_9.c b/rdft/scalar/r2cf/hf_9.c
index 43bac02..ff894de 100644
--- a/rdft/scalar/r2cf/hf_9.c
+++ b/rdft/scalar/r2cf/hf_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:44 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:51 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -48,7 +48,7 @@ static void hf_9(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(18, rs)) {
 	       E T20, T1Z;
 	       {
 		    E T1, T1P, T1Q, T10, T1S, Te, TB, T1d, T1a, T19, T1M, TE, T1c, Tz, T1n;
@@ -277,7 +277,7 @@ static void hf_9(R *cr, R *ci, const R *W, stride rs, INT mb, INT me, INT ms)
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * 16); m < me; m = m + 1, cr = cr + ms, ci = ci - ms, W = W + 16, MAKE_VOLATILE_STRIDE(18, rs)) {
 	       E T1, T1B, TQ, T1A, Tc, TN, T1C, T1D, TL, T1x, T19, T1o, T1c, T1n, Tu;
 	       E T1w, TW, T1k, T11, T1l;
 	       {
diff --git a/rdft/scalar/r2cf/r2cfII_10.c b/rdft/scalar/r2cf/r2cfII_10.c
index 7737325..345ac7a 100644
--- a/rdft/scalar/r2cf/r2cfII_10.c
+++ b/rdft/scalar/r2cf/r2cfII_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:15 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cfII_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E Tq, Ti, Tk, Tu, Tw, Tp, Tb, Tj, Tr, Tv;
 	       {
 		    E T1, To, Ts, Tt, T8, Ta, Te, Tm, Tl, Th, Tn, T9;
@@ -124,7 +124,7 @@ static void r2cfII_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E T1, To, T8, Tq, T9, Tp, Te, Ts, Th, Tn;
 	       T1 = R0[0];
 	       To = R1[WS(rs, 2)];
diff --git a/rdft/scalar/r2cf/r2cfII_12.c b/rdft/scalar/r2cf/r2cfII_12.c
index 1aab93b..a86498e 100644
--- a/rdft/scalar/r2cf/r2cfII_12.c
+++ b/rdft/scalar/r2cf/r2cfII_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:15 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void r2cfII_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E TD, TB, Tp, T9, Tq, Tr, TE, To, Ts, TC;
 	       {
 		    E T8, T1, Tv, Tm, TF, Tz, Tl, Ta, Tb, Tt, TA, T4, Tc;
@@ -139,7 +139,7 @@ static void r2cfII_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E Tx, Tg, T4, Tz, Ty, Tj, TA, T9, Tm, Tl, Te, Tp, To, Tf, TE;
 	       E TF;
 	       {
diff --git a/rdft/scalar/r2cf/r2cfII_15.c b/rdft/scalar/r2cf/r2cfII_15.c
index f6ed4f8..452d8a6 100644
--- a/rdft/scalar/r2cf/r2cfII_15.c
+++ b/rdft/scalar/r2cf/r2cfII_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:15 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -50,7 +50,7 @@ static void r2cfII_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP250000000, +0.250000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E T9, TQ, TV, TW, Tw, TJ;
 	       {
 		    E Ta, Tl, Tg, T8, T7, TF, TX, TT, Tm, Th, TM, TZ, Tr, Tn, Tj;
@@ -191,7 +191,7 @@ static void r2cfII_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E T1, T2, Tx, TR, TE, T7, TD, Th, Tm, Tr, TQ, TA, TB, Tf, Te;
 	       E Tu, TS, Td, TH, TO;
 	       T1 = R0[WS(rs, 5)];
diff --git a/rdft/scalar/r2cf/r2cfII_16.c b/rdft/scalar/r2cf/r2cfII_16.c
index cc47832..2caec84 100644
--- a/rdft/scalar/r2cf/r2cfII_16.c
+++ b/rdft/scalar/r2cf/r2cfII_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:15 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void r2cfII_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E TN, TF, TX, TV, TO, TP, TY, TM, TQ, TW;
 	       {
 		    E TT, TZ, TB, T5, Tu, TK, TJ, Tr, T9, TC, T8, Tl, TH, TG, Ti;
@@ -183,7 +183,7 @@ static void r2cfII_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E T5, T11, TB, TV, Tr, TK, Tu, TJ, Ti, TH, Tl, TG, Tc, T10, TE;
 	       E TS;
 	       {
diff --git a/rdft/scalar/r2cf/r2cfII_2.c b/rdft/scalar/r2cf/r2cfII_2.c
index fddfbb1..4d9d77e 100644
--- a/rdft/scalar/r2cf/r2cfII_2.c
+++ b/rdft/scalar/r2cf/r2cfII_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void r2cfII_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = R0[0];
 	       T2 = R1[0];
@@ -69,7 +69,7 @@ static void r2cfII_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = R0[0];
 	       T2 = R1[0];
diff --git a/rdft/scalar/r2cf/r2cfII_20.c b/rdft/scalar/r2cf/r2cfII_20.c
index f0d2f5c..c3c49cf 100644
--- a/rdft/scalar/r2cf/r2cfII_20.c
+++ b/rdft/scalar/r2cf/r2cfII_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:14 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:23 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -48,7 +48,7 @@ static void r2cfII_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP381966011, +0.381966011250105151795413165634361882279690820);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E Tv, TK, TN, Th, T1l, T1n, Ts, TH;
 	       {
 		    E Ti, T1d, T1f, T1e, T1g, T1p, TS, Tg, To, T8, T7, T19, T1r, T1k, Tx;
@@ -243,7 +243,7 @@ static void r2cfII_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E T8, TD, Tm, TN, T9, TC, TY, TE, Te, TF, Tl, TK, T12, TL, Tk;
 	       E TM, T1, T6, Tq, T1l, T1c, Tp, T1f, T1e, T1d, Ty, TW, T1g, T1m, Tx;
 	       E Tu;
diff --git a/rdft/scalar/r2cf/r2cfII_25.c b/rdft/scalar/r2cf/r2cfII_25.c
index 721528e..44086c0 100644
--- a/rdft/scalar/r2cf/r2cfII_25.c
+++ b/rdft/scalar/r2cf/r2cfII_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:16 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:24 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -105,7 +105,7 @@ static void r2cfII_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E T2R, T2T, T2D, T2C, T2H, T2G, T2B, T2P, T2S;
 	       {
 		    E T2A, TJ, T1K, T3l, T2z, TB, T2d, T2l, T1N, T21, T15, T1g, T1s, T1D, T9;
@@ -489,7 +489,7 @@ static void r2cfII_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E TE, TR, T2i, T1z, TL, TS, TB, T2d, T1l, T1i, T2c, T9, T23, TZ, TW;
 	       E T22, Ti, T26, T16, T13, T25, Ts, T2a, T1e, T1b, T29, TP, TQ;
 	       {
diff --git a/rdft/scalar/r2cf/r2cfII_3.c b/rdft/scalar/r2cf/r2cfII_3.c
index dc351e7..503c1b0 100644
--- a/rdft/scalar/r2cf/r2cfII_3.c
+++ b/rdft/scalar/r2cf/r2cfII_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cfII_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T3, T1, T2, T4;
 	       T3 = R0[0];
 	       T1 = R1[0];
@@ -76,7 +76,7 @@ static void r2cfII_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T1, T2, T3, T4;
 	       T1 = R0[0];
 	       T2 = R1[0];
diff --git a/rdft/scalar/r2cf/r2cfII_32.c b/rdft/scalar/r2cf/r2cfII_32.c
index 3b94237..a44ed17 100644
--- a/rdft/scalar/r2cf/r2cfII_32.c
+++ b/rdft/scalar/r2cf/r2cfII_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:08 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:16 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -53,7 +53,7 @@ static void r2cfII_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T23, T1S, T21, T1L, T2z, T2x, T1Z, T22;
 	       {
 		    E T2n, T2B, T1z, T5, T1C, T2C, T2o, Tc, T27, T1J, T1l, Tm, T26, T1G, T1k;
@@ -364,7 +364,7 @@ static void r2cfII_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T5, T2D, T1z, T2q, Tc, T2C, T1C, T2n, Tm, T1k, T1J, T26, Tv, T1l, T1G;
 	       E T27, T15, T1r, T1Y, T2e, T1c, T1s, T1V, T2d, TK, T1o, T1R, T2b, TR, T1p;
 	       E T1O, T2a;
diff --git a/rdft/scalar/r2cf/r2cfII_4.c b/rdft/scalar/r2cf/r2cfII_4.c
index 0c54b5a..27080f0 100644
--- a/rdft/scalar/r2cf/r2cfII_4.c
+++ b/rdft/scalar/r2cf/r2cfII_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void r2cfII_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T1, T5, T2, T3, T4, T6;
 	       T1 = R0[0];
 	       T5 = R0[WS(rs, 1)];
@@ -77,7 +77,7 @@ static void r2cfII_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T1, T6, T4, T5, T2, T3;
 	       T1 = R0[0];
 	       T6 = R0[WS(rs, 1)];
diff --git a/rdft/scalar/r2cf/r2cfII_5.c b/rdft/scalar/r2cf/r2cfII_5.c
index f1ec8fc..ed8b659 100644
--- a/rdft/scalar/r2cf/r2cfII_5.c
+++ b/rdft/scalar/r2cf/r2cfII_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cfII_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E T1, T2, T3, T5, T6;
 	       T1 = R0[0];
 	       T2 = R0[WS(rs, 1)];
@@ -93,7 +93,7 @@ static void r2cfII_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP587785252, +0.587785252292473129168705954639072768597652438);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E T8, T3, T6, T9, Tc, Tb, T7, Ta;
 	       T8 = R0[0];
 	       {
diff --git a/rdft/scalar/r2cf/r2cfII_6.c b/rdft/scalar/r2cf/r2cfII_6.c
index 6bae9f5..895d605 100644
--- a/rdft/scalar/r2cf/r2cfII_6.c
+++ b/rdft/scalar/r2cf/r2cfII_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cfII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T1, T9, T2, T3, T6, T7;
 	       T1 = R0[0];
 	       T9 = R1[WS(rs, 1)];
@@ -90,7 +90,7 @@ static void r2cfII_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E Ta, T7, T9, T1, T3, T2, T8, T4, T5, T6, Tb;
 	       Ta = R1[WS(rs, 1)];
 	       T5 = R1[WS(rs, 2)];
diff --git a/rdft/scalar/r2cf/r2cfII_64.c b/rdft/scalar/r2cf/r2cfII_64.c
index 8711884..eff6cc2 100644
--- a/rdft/scalar/r2cf/r2cfII_64.c
+++ b/rdft/scalar/r2cf/r2cfII_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:09 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:17 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -69,7 +69,7 @@ static void r2cfII_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E T5b, T6q, T6p, T5e;
 	       {
 		    E T5h, T3Z, T35, Tm, T5g, T3W, T34, Tv, T5f, T3T, T6N, T6z, T6j, T65, T33;
@@ -816,7 +816,7 @@ static void r2cfII_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E Tm, T34, T3Z, T5g, Tv, T35, T3W, T5h, Td, T33, T6B, T6Q, T3T, T5f, T68;
 	       E T6m, T2b, T3n, T4O, T5D, T2F, T3r, T4K, T5z, TK, T3c, T47, T5n, TR, T3b;
 	       E T44, T5o, T15, T38, T4e, T5l, T1c, T39, T4b, T5k, T1s, T3g, T4v, T5w, T1W;
diff --git a/rdft/scalar/r2cf/r2cfII_7.c b/rdft/scalar/r2cf/r2cfII_7.c
index 3b680c9..9f2b60f 100644
--- a/rdft/scalar/r2cf/r2cfII_7.c
+++ b/rdft/scalar/r2cf/r2cfII_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:05 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:13 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void r2cfII_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP356895867, +0.356895867892209443894399510021300583399127187);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E Td, Tk;
 	       {
 		    E T4, T3, Te, T5, T9, Tf, T6, Tg, Tj;
@@ -115,7 +115,7 @@ static void r2cfII_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP781831482, +0.781831482468029808708444526674057750232334519);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E T1, Ta, Td, T4, Tb, T7, Tc, T8, T9;
 	       T1 = R0[0];
 	       T8 = R1[0];
diff --git a/rdft/scalar/r2cf/r2cfII_8.c b/rdft/scalar/r2cf/r2cfII_8.c
index 1a2d0e6..a40d93c 100644
--- a/rdft/scalar/r2cf/r2cfII_8.c
+++ b/rdft/scalar/r2cf/r2cfII_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:14 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void r2cfII_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E Te, T8, Td, T5, Tj, Tl, Tf, Tb;
 	       {
 		    E T1, Th, T9, Ti, T4, Ta;
@@ -110,7 +110,7 @@ static void r2cfII_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E T1, Tj, T4, Ti, T8, Te, Tb, Tf, T2, T3;
 	       T1 = R0[0];
 	       Tj = R0[WS(rs, 2)];
diff --git a/rdft/scalar/r2cf/r2cfII_9.c b/rdft/scalar/r2cf/r2cfII_9.c
index cc1b056..87f86f3 100644
--- a/rdft/scalar/r2cf/r2cfII_9.c
+++ b/rdft/scalar/r2cf/r2cfII_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:03:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:40:14 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -55,7 +55,7 @@ static void r2cfII_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E To, T5, Tp, Ta, Ti, Tm, TB, Tq, Tt, Tf, Th;
 	       {
 		    E T1, T6, T4, Tb, Tk, T9, Tc, Td, Tl, Te;
@@ -165,7 +165,7 @@ static void r2cfII_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E T1, T4, To, Ta, Tl, Tk, Tf, Ti, Th, T2, T3, T5, Tg;
 	       T1 = R0[0];
 	       T2 = R1[WS(rs, 1)];
diff --git a/rdft/scalar/r2cf/r2cf_10.c b/rdft/scalar/r2cf/r2cf_10.c
index d9b2d5a..a998454 100644
--- a/rdft/scalar/r2cf/r2cf_10.c
+++ b/rdft/scalar/r2cf/r2cf_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cf_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E Tt, T3, T7, Tq, T6, Tv, Tp, Tm, Th, T8, T1, T2, T9, Tr;
 	       T1 = R0[0];
 	       T2 = R1[WS(rs, 2)];
@@ -129,7 +129,7 @@ static void r2cf_10(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP587785252, +0.587785252292473129168705954639072768597652438);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(40, rs), MAKE_VOLATILE_STRIDE(40, csr), MAKE_VOLATILE_STRIDE(40, csi)) {
 	       E Ti, Tt, Ta, Tn, Td, To, Te, Tv, T3, Tq, T6, Tr, T7, Tu, Tg;
 	       E Th;
 	       Tg = R0[0];
diff --git a/rdft/scalar/r2cf/r2cf_11.c b/rdft/scalar/r2cf/r2cf_11.c
index 64a2e39..eb3cd2d 100644
--- a/rdft/scalar/r2cf/r2cf_11.c
+++ b/rdft/scalar/r2cf/r2cf_11.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -48,7 +48,7 @@ static void r2cf_11(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP521108558, +0.521108558113202722944698153526659300680427422);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(44, rs), MAKE_VOLATILE_STRIDE(44, csr), MAKE_VOLATILE_STRIDE(44, csi)) {
 	       E T1, Tg, TF, TB, TI, TL, Tz, TA;
 	       {
 		    E T4, TC, TE, T7, TD, Ta, TS, TG, TJ, Td, TP, TM, Ty, Tq, Th;
@@ -177,7 +177,7 @@ static void r2cf_11(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP755749574, +0.755749574354258283774035843972344420179717445);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(44, rs), MAKE_VOLATILE_STRIDE(44, csr), MAKE_VOLATILE_STRIDE(44, csi)) {
 	       E T1, T4, Tl, Tg, Th, Td, Ti, Ta, Tk, T7, Tj, Tb, Tc;
 	       T1 = R0[0];
 	       {
diff --git a/rdft/scalar/r2cf/r2cf_12.c b/rdft/scalar/r2cf/r2cf_12.c
index 7aa7073..0f38c1f 100644
--- a/rdft/scalar/r2cf/r2cf_12.c
+++ b/rdft/scalar/r2cf/r2cf_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:46 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cf_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E Tm, T6, Ty, Tp, T5, Tk, Tt, Tb, Tc, Td, T9, Tn;
 	       {
 		    E T1, Tg, Th, Ti, T4, T2, T3, T7, T8, Tj;
@@ -134,7 +134,7 @@ static void r2cf_12(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(48, rs), MAKE_VOLATILE_STRIDE(48, csr), MAKE_VOLATILE_STRIDE(48, csi)) {
 	       E T5, Tp, Tb, Tn, Ty, Tt, Ta, Tq, Tc, Ti, Tz, Tu, Td, To;
 	       {
 		    E T1, T2, T3, T4;
diff --git a/rdft/scalar/r2cf/r2cf_128.c b/rdft/scalar/r2cf/r2cf_128.c
index 4d7359d..f603ba3 100644
--- a/rdft/scalar/r2cf/r2cf_128.c
+++ b/rdft/scalar/r2cf/r2cf_128.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:47 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -69,7 +69,7 @@ static void r2cf_128(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(512, rs), MAKE_VOLATILE_STRIDE(512, csr), MAKE_VOLATILE_STRIDE(512, csi)) {
 	       E T95, T96;
 	       {
 		    E TcD, TdR, T5P, T8v, T27, T7r, Tf, Ta5, T7s, T5S, T8w, T2e, TdS, TcG, Tbn;
@@ -1613,7 +1613,7 @@ static void r2cf_128(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride c
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(512, rs), MAKE_VOLATILE_STRIDE(512, csr), MAKE_VOLATILE_STRIDE(512, csi)) {
 	       E TcD, TdU, T27, T7r, T5S, T8y, Tf, Ta5, Tu, Tbq, TcG, TdV, T2e, T8z, T5V;
 	       E T7s, TK, Ta6, TcK, TdX, T2o, T5X, T7w, T8B, TZ, Ta7, TcN, TdY, T2x, T5Y;
 	       E T7z, T8C, T1g, Taa, TcU, TeA, TcX, Tez, T1v, Tab, T2M, T6z, T7E, T9e, T7H;
diff --git a/rdft/scalar/r2cf/r2cf_13.c b/rdft/scalar/r2cf/r2cf_13.c
index fe8d9f0..4ce98ab 100644
--- a/rdft/scalar/r2cf/r2cf_13.c
+++ b/rdft/scalar/r2cf/r2cf_13.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:46 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -61,7 +61,7 @@ static void r2cf_13(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(52, rs), MAKE_VOLATILE_STRIDE(52, csr), MAKE_VOLATILE_STRIDE(52, csi)) {
 	       E T15, T1a, T11, T17, T14, T1b;
 	       {
 		    E TN, TD, TV, TA, Tb, TZ, T12, TS, Tx, Tu, Ti, TU;
@@ -230,7 +230,7 @@ static void r2cf_13(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(52, rs), MAKE_VOLATILE_STRIDE(52, csr), MAKE_VOLATILE_STRIDE(52, csi)) {
 	       E T13, Tb, Tm, TW, TX, T14, TU, T10, Tz, TB, Tu, TC, TR, T11;
 	       T13 = R0[0];
 	       {
diff --git a/rdft/scalar/r2cf/r2cf_14.c b/rdft/scalar/r2cf/r2cf_14.c
index c7a3f65..dd1b431 100644
--- a/rdft/scalar/r2cf/r2cf_14.c
+++ b/rdft/scalar/r2cf/r2cf_14.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:46 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void r2cf_14(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP554958132, +0.554958132087371191422194871006410481067288862);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(56, rs), MAKE_VOLATILE_STRIDE(56, csr), MAKE_VOLATILE_STRIDE(56, csi)) {
 	       E TN, T3, TG, TQ, Tx, To, TH, Td, TD, TO, Tw, Ta, TL, Ty, TT;
 	       E TI, Tg, Tr, Te, Tf, TP, TJ;
 	       {
@@ -177,7 +177,7 @@ static void r2cf_14(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP781831482, +0.781831482468029808708444526674057750232334519);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(56, rs), MAKE_VOLATILE_STRIDE(56, csr), MAKE_VOLATILE_STRIDE(56, csi)) {
 	       E T3, TB, T6, Tv, Tn, Ts, Tk, Tt, Td, Ty, T9, Tw, Tg, Tz, T1;
 	       E T2;
 	       T1 = R0[0];
diff --git a/rdft/scalar/r2cf/r2cf_15.c b/rdft/scalar/r2cf/r2cf_15.c
index edee269..ffb708b 100644
--- a/rdft/scalar/r2cf/r2cf_15.c
+++ b/rdft/scalar/r2cf/r2cf_15.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:39 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:46 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -46,7 +46,7 @@ static void r2cf_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E Tw, Tz, Tp, Ty;
 	       {
 		    E Ti, TF, TR, TN, TX, T11, TM, TS, Tl, TH, Tf, To, TT, TD, Tg;
@@ -184,7 +184,7 @@ static void r2cf_15(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(60, rs), MAKE_VOLATILE_STRIDE(60, csr), MAKE_VOLATILE_STRIDE(60, csi)) {
 	       E Ti, TR, TL, TD, TE, T7, Te, Tf, TV, TW, TX, Tv, Ty, TH, To;
 	       E Tr, TG, TS, TT, TU;
 	       {
diff --git a/rdft/scalar/r2cf/r2cf_16.c b/rdft/scalar/r2cf/r2cf_16.c
index c47dda5..d77371c 100644
--- a/rdft/scalar/r2cf/r2cf_16.c
+++ b/rdft/scalar/r2cf/r2cf_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:47 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void r2cf_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E TQ, TP;
 	       {
 		    E TB, TN, Tf, T7, Te, Tv, TO, TE, Tq, TJ, Tp, TI, TT, Ty, Tm;
@@ -170,7 +170,7 @@ static void r2cf_16(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(64, rs), MAKE_VOLATILE_STRIDE(64, csr), MAKE_VOLATILE_STRIDE(64, csi)) {
 	       E T3, T6, T7, Tz, Ti, Ta, Td, Te, TA, Th, Tq, TV, TF, TP, Tx;
 	       E TU, TE, TM, Tg, Tf, TJ, TQ;
 	       {
diff --git a/rdft/scalar/r2cf/r2cf_2.c b/rdft/scalar/r2cf/r2cf_2.c
index 4392140..08e28be 100644
--- a/rdft/scalar/r2cf/r2cf_2.c
+++ b/rdft/scalar/r2cf/r2cf_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void r2cf_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = R0[0];
 	       T2 = R1[0];
@@ -69,7 +69,7 @@ static void r2cf_2(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(8, rs), MAKE_VOLATILE_STRIDE(8, csr), MAKE_VOLATILE_STRIDE(8, csi)) {
 	       E T1, T2;
 	       T1 = R0[0];
 	       T2 = R1[0];
diff --git a/rdft/scalar/r2cf/r2cf_20.c b/rdft/scalar/r2cf/r2cf_20.c
index aa9aa55..c7979b8 100644
--- a/rdft/scalar/r2cf/r2cf_20.c
+++ b/rdft/scalar/r2cf/r2cf_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:48 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cf_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E T1i, T1c, T1a, T1o, T1m, T1h, T1b, T13, T1j, T1n;
 	       {
 		    E T3, T1d, TJ, TV, T1k, T16, T19, T1l, Ty, Ti, T12, TD, T1g, TR, TX;
@@ -208,7 +208,7 @@ static void r2cf_20(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(80, rs), MAKE_VOLATILE_STRIDE(80, csr), MAKE_VOLATILE_STRIDE(80, csi)) {
 	       E T3, T1m, TF, T17, Ts, TM, TN, Tz, Ta, Th, Ti, T1g, T1h, T1k, T10;
 	       E T13, T19, TG, TH, TI, T1d, T1e, T1j, TT, TW, T18;
 	       {
diff --git a/rdft/scalar/r2cf/r2cf_25.c b/rdft/scalar/r2cf/r2cf_25.c
index b218d17..1b1c46f 100644
--- a/rdft/scalar/r2cf/r2cf_25.c
+++ b/rdft/scalar/r2cf/r2cf_25.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:41 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:48 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -104,7 +104,7 @@ static void r2cf_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E T2H, T2w, T2x, T2A, T2C, T2v, T2M, T2y, T2B, T2N;
 	       {
 		    E T2u, TJ, T1O, T39, T2t, TB, T21, T1M, T2e, T26, T1B, T1r, T1k, T1c, T9;
@@ -471,7 +471,7 @@ static void r2cf_25(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(100, rs), MAKE_VOLATILE_STRIDE(100, csr), MAKE_VOLATILE_STRIDE(100, csi)) {
 	       E T8, T1j, T1V, T1l, T7, T9, Ta, T12, T2u, T1O, T19, T1P, Ti, T2r, T1K;
 	       E Tp, T1L, Tx, T2q, T1H, TE, T1I, TN, T2t, T1R, TU, T1S, T6, T1k, T3;
 	       E T2s, T2v;
diff --git a/rdft/scalar/r2cf/r2cf_3.c b/rdft/scalar/r2cf/r2cf_3.c
index 418fd2f..5e879fe 100644
--- a/rdft/scalar/r2cf/r2cf_3.c
+++ b/rdft/scalar/r2cf/r2cf_3.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cf_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T1, T2, T3, T4;
 	       T1 = R0[0];
 	       T2 = R1[0];
@@ -76,7 +76,7 @@ static void r2cf_3(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(12, rs), MAKE_VOLATILE_STRIDE(12, csr), MAKE_VOLATILE_STRIDE(12, csi)) {
 	       E T1, T2, T3, T4;
 	       T1 = R0[0];
 	       T2 = R1[0];
diff --git a/rdft/scalar/r2cf/r2cf_32.c b/rdft/scalar/r2cf/r2cf_32.c
index b4a55e4..fefff09 100644
--- a/rdft/scalar/r2cf/r2cf_32.c
+++ b/rdft/scalar/r2cf/r2cf_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:47 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void r2cf_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T1x, T1M, T1I, T1E, T1J, T1H;
 	       {
 		    E Tv, T1h, T7, T2b, Te, T2n, Ty, T1i, T1l, TF, T2d, Tt, T1k, TC, T2c;
@@ -335,7 +335,7 @@ static void r2cf_32(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(128, rs), MAKE_VOLATILE_STRIDE(128, csr), MAKE_VOLATILE_STRIDE(128, csi)) {
 	       E T7, T2b, Tv, T1l, Te, T2o, Ty, T1k, Tt, T2d, TF, T1h, Tm, T2c, TC;
 	       E T1i, T1Z, T22, T2k, T2j, T1e, T1C, T19, T1B, T1S, T1V, T2h, T2g, TX, T1z;
 	       E TS, T1y;
diff --git a/rdft/scalar/r2cf/r2cf_4.c b/rdft/scalar/r2cf/r2cf_4.c
index a7bed6f..f22a1f6 100644
--- a/rdft/scalar/r2cf/r2cf_4.c
+++ b/rdft/scalar/r2cf/r2cf_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void r2cf_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T1, T2, T4, T5, T3, T6;
 	       T1 = R0[0];
 	       T2 = R0[WS(rs, 1)];
@@ -75,7 +75,7 @@ static void r2cf_4(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
 {
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(16, rs), MAKE_VOLATILE_STRIDE(16, csr), MAKE_VOLATILE_STRIDE(16, csi)) {
 	       E T1, T2, T3, T4, T5, T6;
 	       T1 = R0[0];
 	       T2 = R0[WS(rs, 1)];
diff --git a/rdft/scalar/r2cf/r2cf_5.c b/rdft/scalar/r2cf/r2cf_5.c
index 7e02dde..211ed17 100644
--- a/rdft/scalar/r2cf/r2cf_5.c
+++ b/rdft/scalar/r2cf/r2cf_5.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void r2cf_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E T7, T1, T2, T4, T5;
 	       T7 = R0[0];
 	       T1 = R0[WS(rs, 2)];
@@ -93,7 +93,7 @@ static void r2cf_5(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(20, rs), MAKE_VOLATILE_STRIDE(20, csr), MAKE_VOLATILE_STRIDE(20, csi)) {
 	       E Ta, T7, T8, T3, Tb, T6, T9, Tc;
 	       Ta = R0[0];
 	       {
diff --git a/rdft/scalar/r2cf/r2cf_6.c b/rdft/scalar/r2cf/r2cf_6.c
index 9134a49..9f31742 100644
--- a/rdft/scalar/r2cf/r2cf_6.c
+++ b/rdft/scalar/r2cf/r2cf_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void r2cf_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T4, Td, T3, Tc, T9, T5;
 	       {
 		    E T1, T2, T7, T8;
@@ -95,7 +95,7 @@ static void r2cf_6(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(24, rs), MAKE_VOLATILE_STRIDE(24, csr), MAKE_VOLATILE_STRIDE(24, csi)) {
 	       E T3, Td, T9, Tc, T6, Tb, T1, T2, Ta, Te;
 	       T1 = R0[0];
 	       T2 = R1[WS(rs, 1)];
diff --git a/rdft/scalar/r2cf/r2cf_64.c b/rdft/scalar/r2cf/r2cf_64.c
index 736cf48..a006910 100644
--- a/rdft/scalar/r2cf/r2cf_64.c
+++ b/rdft/scalar/r2cf/r2cf_64.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:40 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:47 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -53,7 +53,7 @@ static void r2cf_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E T5n, T5o;
 	       {
 		    E T11, T2j, T4P, T5P, T3D, T5p, T3d, Tf, T1k, T1H, T5D, T4l, T5A, T4a, T3i;
@@ -710,7 +710,7 @@ static void r2cf_64(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride cs
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(256, rs), MAKE_VOLATILE_STRIDE(256, csr), MAKE_VOLATILE_STRIDE(256, csi)) {
 	       E T4l, T5a, T15, T3n, T2T, T3Q, T7, Te, Tf, T4A, T4L, T1X, T3B, T23, T3y;
 	       E T5I, T66, T4R, T52, T2j, T3F, T2H, T3I, T5P, T69, T1i, T3t, T1l, T3u, TZ;
 	       E T63, T4v, T58, T1r, T3r, T1u, T3q, TK, T62, T4s, T57, Tm, Tt, Tu, T4o;
diff --git a/rdft/scalar/r2cf/r2cf_7.c b/rdft/scalar/r2cf/r2cf_7.c
index 6dcd1fc..0502f15 100644
--- a/rdft/scalar/r2cf/r2cf_7.c
+++ b/rdft/scalar/r2cf/r2cf_7.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -44,7 +44,7 @@ static void r2cf_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP356895867, +0.356895867892209443894399510021300583399127187);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E T1, Tg, Tc;
 	       {
 		    E Th, T4, Ti, Ta, Tj, T7, Td, T5, T6, Tl, Tk;
@@ -115,7 +115,7 @@ static void r2cf_7(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP974927912, +0.974927912181823607018131682993931217232785801);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(28, rs), MAKE_VOLATILE_STRIDE(28, csr), MAKE_VOLATILE_STRIDE(28, csi)) {
 	       E T1, Ta, Tb, T4, Td, T7, Tc, T8, T9;
 	       T1 = R0[0];
 	       T8 = R1[0];
diff --git a/rdft/scalar/r2cf/r2cf_8.c b/rdft/scalar/r2cf/r2cf_8.c
index b20a0bf..0fae650 100644
--- a/rdft/scalar/r2cf/r2cf_8.c
+++ b/rdft/scalar/r2cf/r2cf_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void r2cf_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E T4, T7, T3, Tj, Td, T5, T8, T9;
 	       {
 		    E T1, T2, Tb, Tc;
@@ -104,7 +104,7 @@ static void r2cf_8(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(32, rs), MAKE_VOLATILE_STRIDE(32, csr), MAKE_VOLATILE_STRIDE(32, csi)) {
 	       E T3, T7, Td, Tj, T6, Tg, Ta, Ti;
 	       {
 		    E T1, T2, Tb, Tc;
diff --git a/rdft/scalar/r2cf/r2cf_9.c b/rdft/scalar/r2cf/r2cf_9.c
index 4b0aef7..2eb0c80 100644
--- a/rdft/scalar/r2cf/r2cf_9.c
+++ b/rdft/scalar/r2cf/r2cf_9.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:02:38 EDT 2012 */
+/* Generated on Sun Nov 25 07:39:45 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -56,7 +56,7 @@ static void r2cf_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E Tp, Tz, Tw, Ts, TA;
 	       {
 		    E T1, T6, Tb, T7, T4, To, T8, Tc, Td, T2, T3;
@@ -162,7 +162,7 @@ static void r2cf_9(R *R0, R *R1, R *Cr, R *Ci, stride rs, stride csr, stride csi
      DK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(rs), MAKE_VOLATILE_STRIDE(csr), MAKE_VOLATILE_STRIDE(csi)) {
+	  for (i = v; i > 0; i = i - 1, R0 = R0 + ivs, R1 = R1 + ivs, Cr = Cr + ovs, Ci = Ci + ovs, MAKE_VOLATILE_STRIDE(36, rs), MAKE_VOLATILE_STRIDE(36, csr), MAKE_VOLATILE_STRIDE(36, csi)) {
 	       E T1, T4, Tr, Ta, Tl, Ti, Tf, Tk, Tj, T2, T3, T5, Tg;
 	       T1 = R0[0];
 	       T2 = R1[WS(rs, 1)];
diff --git a/rdft/scalar/r2r/Makefile.in b/rdft/scalar/r2r/Makefile.in
index c6aca3f..a5fb612 100644
--- a/rdft/scalar/r2r/Makefile.in
+++ b/rdft/scalar/r2r/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -32,6 +32,23 @@
 # Most of this file requires GNU make .
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -98,6 +115,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_scalar_r2r_la_SOURCES)
 DIST_SOURCES = $(librdft_scalar_r2r_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/rdft/scalar/r2r/e01_8.c b/rdft/scalar/r2r/e01_8.c
index 30c12d6..62f7a63 100644
--- a/rdft/scalar/r2r/e01_8.c
+++ b/rdft/scalar/r2r/e01_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:27 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -46,7 +46,7 @@ static void e01_8(const R *I, R *O, stride is, stride os, INT v, INT ivs, INT ov
      DK(KP1_414213562, +1.414213562373095048801688724209698078569671875);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       E T8, Td, Th, T7, Tp, Tl, Te, Tb;
 	       {
 		    E Tj, T3, Tk, T6, T9, Ta;
@@ -127,7 +127,7 @@ static void e01_8(const R *I, R *O, stride is, stride os, INT v, INT ivs, INT ov
      DK(KP1_847759065, +1.847759065022573512256366378793576573644833252);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       E T7, Tl, T4, Tk, Td, To, Tg, Tn;
 	       {
 		    E T5, T6, T1, T3, T2;
diff --git a/rdft/scalar/r2r/e10_8.c b/rdft/scalar/r2r/e10_8.c
index 32cecac..a768e4b 100644
--- a/rdft/scalar/r2r/e10_8.c
+++ b/rdft/scalar/r2r/e10_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:04 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:27 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -47,7 +47,7 @@ static void e10_8(const R *I, R *O, stride is, stride os, INT v, INT ivs, INT ov
      DK(KP1_847759065, +1.847759065022573512256366378793576573644833252);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       E T3, Te, Tl, Tp, Tm, T6, Tn, T9;
 	       {
 		    E T4, Tj, Tk, T5, T7, T8;
@@ -129,7 +129,7 @@ static void e10_8(const R *I, R *O, stride is, stride os, INT v, INT ivs, INT ov
      DK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT i;
-	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
+	  for (i = v; i > 0; i = i - 1, I = I + ivs, O = O + ovs, MAKE_VOLATILE_STRIDE(16, is), MAKE_VOLATILE_STRIDE(16, os)) {
 	       E T3, Tj, Tf, Tk, Ta, Tn, Tc, Tm;
 	       {
 		    E T1, T2, Td, Te;
diff --git a/rdft/simd/Makefile.in b/rdft/simd/Makefile.in
index 232eaa2..79dfafa 100644
--- a/rdft/simd/Makefile.in
+++ b/rdft/simd/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -15,6 +15,23 @@
 
 @SET_MAKE@
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -62,6 +79,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -448,13 +470,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/rdft/simd/altivec/Makefile.in b/rdft/simd/altivec/Makefile.in
index a6fdd2a..38143fd 100644
--- a/rdft/simd/altivec/Makefile.in
+++ b/rdft/simd/altivec/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -105,6 +122,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_altivec_codelets_la_SOURCES)
 DIST_SOURCES = $(am__librdft_altivec_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/rdft/simd/avx/Makefile.in b/rdft/simd/avx/Makefile.in
index 7a2b5aa..7cd8c97 100644
--- a/rdft/simd/avx/Makefile.in
+++ b/rdft/simd/avx/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -104,6 +121,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_avx_codelets_la_SOURCES)
 DIST_SOURCES = $(am__librdft_avx_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/rdft/simd/common/Makefile.in b/rdft/simd/common/Makefile.in
index df574c3..2c0d2a0 100644
--- a/rdft/simd/common/Makefile.in
+++ b/rdft/simd/common/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -33,6 +33,23 @@
 # This file contains special make rules to generate codelets.
 # Most of this file requires GNU make .
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -75,6 +92,11 @@ CONFIG_CLEAN_FILES =
 CONFIG_CLEAN_VPATH_FILES =
 SOURCES =
 DIST_SOURCES =
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 ACLOCAL = @ACLOCAL@
 ALLOCA = @ALLOCA@
diff --git a/rdft/simd/common/hc2cbdftv_10.c b/rdft/simd/common/hc2cbdftv_10.c
index e2c7c88..73b78ef 100644
--- a/rdft/simd/common/hc2cbdftv_10.c
+++ b/rdft/simd/common/hc2cbdftv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cbdftv_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(40, rs)) {
 	       V Ts, T4, TR, T1, TZ, TD, Ty, Tn, Ti, TT, T11, TJ, T15, Tr, TN;
 	       V TE, Tv, To, Tb, T8, Tw, Te, Tx, Th, Tt, T7, T9, T2, T3, Tc;
 	       V Td, Tf, Tg, T5, T6, Tu, Ta;
@@ -174,7 +174,7 @@ static void hc2cbdftv_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(40, rs)) {
 	       V T5, TE, Ts, Tt, TC, Tz, TH, TJ, To, Tq, T2, T4, T3, T9, Tx;
 	       V Tm, TB, Td, Ty, Ti, TA, T6, T8, T7, Tl, Tk, Tj, Tc, Tb, Ta;
 	       V Tf, Th, Tg, TF, TG, Te, Tn;
diff --git a/rdft/simd/common/hc2cbdftv_12.c b/rdft/simd/common/hc2cbdftv_12.c
index 3b96428..7b2a32a 100644
--- a/rdft/simd/common/hc2cbdftv_12.c
+++ b/rdft/simd/common/hc2cbdftv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cbdftv_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(48, rs)) {
 	       V Tz, TT, T1, T1j, TN, TF, TP, TL, Tx, T15, TJ, T1b, T1g, T1l, T18;
 	       V T12, TO, TC, TK, Tl, T16, TQ, TU, TG, T1c, TM, T1k, Ty, T19, T1a;
 	       V T13, T14, T1h, T1i, TS, TR, T1m, T1n, TI, TH;
@@ -188,7 +188,7 @@ static void hc2cbdftv_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(48, rs)) {
 	       V TY, TZ, Tf, TC, Tq, TG, Tm, TF, Ty, TD, T13, T1h, T2, T9, T3;
 	       V T5, T6, Tc, Tb, Td, T8, T4, Ta, T7, Te, To, Tp, Tr, Tv, Ti;
 	       V Ts, Tl, Tw, Tu, Tg, Th, Tj, Tk, Tt, Tx, T11, T12;
diff --git a/rdft/simd/common/hc2cbdftv_16.c b/rdft/simd/common/hc2cbdftv_16.c
index 24415ea..02da837 100644
--- a/rdft/simd/common/hc2cbdftv_16.c
+++ b/rdft/simd/common/hc2cbdftv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -41,7 +41,7 @@ static void hc2cbdftv_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T1D, T1F, TV, TW, T17, T18, T1B, T1A, T1H, T1G;
 	       {
 		    V T8, Tv, Tb, TF, Tl, TJ, TP, T1w, TE, T1t, T10, T1p, TG, Te, Tg;
@@ -237,7 +237,7 @@ static void hc2cbdftv_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V Tf, T16, TZ, T1C, TI, T1a, TV, T1D, T1F, T1G, Ty, T19, TC, T17, TS;
 	       V T10;
 	       {
diff --git a/rdft/simd/common/hc2cbdftv_2.c b/rdft/simd/common/hc2cbdftv_2.c
index 7201e1b..331d703 100644
--- a/rdft/simd/common/hc2cbdftv_2.c
+++ b/rdft/simd/common/hc2cbdftv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cbdftv_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T2, T3, T1, T5, T4, T7, T6;
 	       T2 = LD(&(Rp[0]), ms, &(Rp[0]));
 	       T3 = LD(&(Rm[0]), -ms, &(Rm[0]));
@@ -79,7 +79,7 @@ static void hc2cbdftv_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T6, T5, T2, T4, T3, T1, T7, T8;
 	       T2 = LD(&(Rp[0]), ms, &(Rp[0]));
 	       T3 = LD(&(Rm[0]), -ms, &(Rm[0]));
diff --git a/rdft/simd/common/hc2cbdftv_20.c b/rdft/simd/common/hc2cbdftv_20.c
index 9f640e2..8cd8bb9 100644
--- a/rdft/simd/common/hc2cbdftv_20.c
+++ b/rdft/simd/common/hc2cbdftv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cbdftv_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(80, rs)) {
 	       V T1M, T1T, T4, TF, T12, Te, T16, Ts, Tb, TN, TA, TG, TU, T1Y, T11;
 	       V T1e, T29, T21, T15, Th, T13, Tp;
 	       {
@@ -298,7 +298,7 @@ static void hc2cbdftv_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(80, rs)) {
 	       V TK, T1v, TY, T1x, T1j, T2f, TS, TT, TO, TU, T5, To, Tp, Tq, T2a;
 	       V T2d, T2g, T2k, T2j, T1k, T1l, T18, T1m, T1f;
 	       {
diff --git a/rdft/simd/common/hc2cbdftv_32.c b/rdft/simd/common/hc2cbdftv_32.c
index 9f61f4c..1d51838 100644
--- a/rdft/simd/common/hc2cbdftv_32.c
+++ b/rdft/simd/common/hc2cbdftv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:07 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:30 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -45,7 +45,7 @@ static void hc2cbdftv_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(128, rs)) {
 	       V T3a, T3N;
 	       {
 		    V T2G, T1o, T2o, T2Y, T1b, T1V, Ts, T1S, T3A, T48, T3p, T45, T31, T2z, T2H;
@@ -464,7 +464,7 @@ static void hc2cbdftv_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(128, rs)) {
 	       V T1W, T21, Tf, T2c, T1t, T2r, T3T, T4m, Ty, T2q, T3P, T4n, T1n, T2d, T1T;
 	       V T22, T1E, T24, T3I, T4p, TU, T2n, T1i, T2h, T1L, T25, T3L, T4q, T1f, T2o;
 	       V T1j, T2k;
diff --git a/rdft/simd/common/hc2cbdftv_4.c b/rdft/simd/common/hc2cbdftv_4.c
index c64eefd..683ba9d 100644
--- a/rdft/simd/common/hc2cbdftv_4.c
+++ b/rdft/simd/common/hc2cbdftv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -38,7 +38,7 @@ static void hc2cbdftv_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T2, T3, T5, T6, Tf, T1, T9, Ta, T4, Tb, T7, Tc, Th, T8, Tg;
 	       V Te, Td, Ti, Tj;
 	       T2 = LD(&(Rp[0]), ms, &(Rp[0]));
@@ -96,7 +96,7 @@ static void hc2cbdftv_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
 {
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T5, Tc, T9, Td, T2, T4, T3, T6, T8, T7, Tj, Ti, Th, Tk, Tl;
 	       V Ta, Te, T1, Tb, Tf, Tg;
 	       T2 = LD(&(Rp[0]), ms, &(Rp[0]));
diff --git a/rdft/simd/common/hc2cbdftv_6.c b/rdft/simd/common/hc2cbdftv_6.c
index 2ccbb99..80a5d4b 100644
--- a/rdft/simd/common/hc2cbdftv_6.c
+++ b/rdft/simd/common/hc2cbdftv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cbdftv_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(24, rs)) {
 	       V Tv, Tn, Tr, Te, T4, Tg, Ta, Tf, T7, T1, Td, T2, T3, T8, T9;
 	       V T5, T6, Th, Tj, Tb, Tp, Tx, Ti, Tc, To, Tk, Ts, Tq, Tw, Tm;
 	       V Tl, Tu, Tt, Tz, Ty;
@@ -121,7 +121,7 @@ static void hc2cbdftv_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(24, rs)) {
 	       V T5, Th, Te, Ts, Tk, Tm, T2, T4, T3, T6, Tc, T8, Tb, T7, Ta;
 	       V T9, Td, Ti, Tj, TA, Tf, Tn, Tv, Tt, Tz, T1, Tl, Tg, Tu, Tr;
 	       V Tq, Ty, To, Tp, TC, TB, Tx, Tw;
diff --git a/rdft/simd/common/hc2cbdftv_8.c b/rdft/simd/common/hc2cbdftv_8.c
index 72a3dd8..962a8e7 100644
--- a/rdft/simd/common/hc2cbdftv_8.c
+++ b/rdft/simd/common/hc2cbdftv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cbdftv_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V TJ, T4, Tf, TB, TD, TE, Tm, T1, Tj, TF, Tp, Tb, Tg, Tt, Tx;
 	       V T2, T3, Td, Te, T5, T6, T8, T9, Tn, T7, To, Ta, Tk, Tl, TG;
 	       V TL, Tq, Tc, Tu, Th, Tv, Ty, Tw, TC, Ti, TK, TA, Tz, TI, TH;
@@ -138,7 +138,7 @@ static void hc2cbdftv_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T5, Tj, Tq, TI, Te, Tk, Tt, TJ, T2, Tg, T4, Ti, T3, Th, To;
 	       V Tp, T6, Tc, T8, Tb, T7, Ta, T9, Td, Tr, Ts, TP, Tu, Tm, TO;
 	       V Tn, Tf, Tl, T1, TN, Tv, TR, Tw, TQ, TC, TK, TA, TG, TB, TH;
diff --git a/rdft/simd/common/hc2cfdftv_10.c b/rdft/simd/common/hc2cfdftv_10.c
index 0dacf3a..34f5154 100644
--- a/rdft/simd/common/hc2cfdftv_10.c
+++ b/rdft/simd/common/hc2cfdftv_10.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void hc2cfdftv_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(40, rs)) {
 	       V T5, T6, Tw, Tr, Tc, Tj, Tl, Tm, Tk, Ts, Tg, Ty, T3, T4, T1;
 	       V T2, Tv, Tq, Ta, Tb, T9, Ti, Te, Tf, Td, Tx, Tn, Tt, Th, TQ;
 	       V TT, Tz, T7, TR, To, Tu, TU;
@@ -176,7 +176,7 @@ static void hc2cfdftv_10(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 18)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 18), MAKE_VOLATILE_STRIDE(40, rs)) {
 	       V Tl, Tt, Tu, TY, TZ, T10, Tz, TE, TF, TV, TW, TX, Ta, TU, TN;
 	       V TR, TH, TQ, TK, TL, TM, TI, TG, TJ, TT, TO, TP, TS, T18, T1c;
 	       V T12, T1b, T15, T16, T17, T14, T11, T13, T1e, T19, T1a, T1d;
diff --git a/rdft/simd/common/hc2cfdftv_12.c b/rdft/simd/common/hc2cfdftv_12.c
index 442665e..9bcec69 100644
--- a/rdft/simd/common/hc2cfdftv_12.c
+++ b/rdft/simd/common/hc2cfdftv_12.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdftv_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(48, rs)) {
 	       V T3, T7, TH, TE, Th, TC, Tq, T11, TU, Tx, Tb, Tz, Tu, Tw, Tp;
 	       V Tl, T9, Ta, T8, Ty, Tn, To, Tm, TG, T1, T2, Tt, T5, T6, T4;
 	       V Tv, Tj, Tk, Ti, TD, Tf, Tg, Te, TB, TT, TF, TR, Tr;
@@ -188,7 +188,7 @@ static void hc2cfdftv_12(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 22)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 22), MAKE_VOLATILE_STRIDE(48, rs)) {
 	       V TX, T13, T4, Tf, TZ, TD, TF, T17, TW, T14, Tw, Tl, T10, TL, TN;
 	       V T16;
 	       {
diff --git a/rdft/simd/common/hc2cfdftv_16.c b/rdft/simd/common/hc2cfdftv_16.c
index ed7d616..593d0e0 100644
--- a/rdft/simd/common/hc2cfdftv_16.c
+++ b/rdft/simd/common/hc2cfdftv_16.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -42,7 +42,7 @@ static void hc2cfdftv_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP414213562, +0.414213562373095048801688724209698078569671875);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T8, Tc, TQ, TZ, T1J, T1x, T12, TH, T1I, T1q, Tp, TJ, Te, Tf, Td;
 	       V TN, Tj, Tk, Ti, TK, Tg, TO, Tl, TL, T1r, Th, TR, T1y, T1s, Tq;
 	       V TM, T1z, T1N, T1t, T10, Tr, T13, TS, T1K, T1A, T1E, T1u, T1f, T11, T1c;
@@ -241,7 +241,7 @@ static void hc2cfdftv_16(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP923879532, +0.923879532511286756128183189396788286822416626);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 30)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 30), MAKE_VOLATILE_STRIDE(64, rs)) {
 	       V T1D, T1E, T1R, TP, T1b, Ta, T1w, T18, T1x, T1z, T1A, T1G, T1H, T1S, Tx;
 	       V T13, T10, T1a, T1, T3, TA, TM, TL, TN, T6, T8, TC, TH, TG, TI;
 	       V T2, Tz, TK, TJ, T7, TB, TF, TE, TD, TO, T4, T9, T5, T15, T17;
diff --git a/rdft/simd/common/hc2cfdftv_2.c b/rdft/simd/common/hc2cfdftv_2.c
index c25fe57..a6cb1e4 100644
--- a/rdft/simd/common/hc2cfdftv_2.c
+++ b/rdft/simd/common/hc2cfdftv_2.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cfdftv_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T1, T2, T4, T3, T5, T7, T6;
 	       T1 = LD(&(Rp[0]), ms, &(Rp[0]));
 	       T2 = LD(&(Rm[0]), -ms, &(Rm[0]));
@@ -81,7 +81,7 @@ static void hc2cfdftv_2(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 2)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 2), MAKE_VOLATILE_STRIDE(8, rs)) {
 	       V T4, T6, T1, T3, T2, T5, T7, T8;
 	       T1 = LD(&(Rp[0]), ms, &(Rp[0]));
 	       T2 = LD(&(Rm[0]), -ms, &(Rm[0]));
diff --git a/rdft/simd/common/hc2cfdftv_20.c b/rdft/simd/common/hc2cfdftv_20.c
index b04d75d..ef35c15 100644
--- a/rdft/simd/common/hc2cfdftv_20.c
+++ b/rdft/simd/common/hc2cfdftv_20.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -43,7 +43,7 @@ static void hc2cfdftv_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP618033988, +0.618033988749894848204586834365638117720309180);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(80, rs)) {
 	       V T2g, T2f, T2w, T2k, T2A, T2u, T2e, T2o, T1O, T2b, T2i, T1R, T1X, T1k, TN;
 	       V T1w, T1G, T1t, Ti, T2c, T12, T1x, T2j, T1U, T1y, T1d, T24, T2v, T2h, T2x;
 	       V T2B, T2p, T2l, T2z, T2y, T2D, T2C, T2r, T2q, T2n, T2m;
@@ -303,7 +303,7 @@ static void hc2cfdftv_20(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 38)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 38), MAKE_VOLATILE_STRIDE(80, rs)) {
 	       V TW, T1x, T2i, T2A, T1r, T1s, T1a, T1y, T1l, Tn, TK, TL, T1p, T1o, T27;
 	       V T2t, T2a, T2u, T2e, T2C, T20, T2w, T23, T2x, T2d, T2B, T1W, T1X, T1U, T1V;
 	       V T2z, T2K, T2G, T2N, T2J, T2v, T2y, T2F, T2D, T2E, T2M, T2H, T2I, T2L;
diff --git a/rdft/simd/common/hc2cfdftv_32.c b/rdft/simd/common/hc2cfdftv_32.c
index 86975c9..c43c34e 100644
--- a/rdft/simd/common/hc2cfdftv_32.c
+++ b/rdft/simd/common/hc2cfdftv_32.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -46,7 +46,7 @@ static void hc2cfdftv_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(128, rs)) {
 	       V T2m, T2b, T2c, T2d, T2v, T2r, T20, T2i, T2n, T2e, T2o, T2u, T2j, T2f, T2t;
 	       V T2s, T2x, T2w, T2l, T2k, T2h, T2g;
 	       {
@@ -464,7 +464,7 @@ static void hc2cfdftv_32(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 62)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 62), MAKE_VOLATILE_STRIDE(128, rs)) {
 	       V Ta, T2m, Tx, T2h, T3R, T4h, T3q, T4g, T3B, T4n, T3E, T4o, T1B, T2S, T1O;
 	       V T2R, TV, T2p, T1i, T2o, T3L, T4q, T3I, T4r, T3w, T4k, T3t, T4j, T26, T2V;
 	       V T2d, T2U;
diff --git a/rdft/simd/common/hc2cfdftv_4.c b/rdft/simd/common/hc2cfdftv_4.c
index c1b6fa1..26b0563 100644
--- a/rdft/simd/common/hc2cfdftv_4.c
+++ b/rdft/simd/common/hc2cfdftv_4.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -39,7 +39,7 @@ static void hc2cfdftv_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T1, T2, Tb, T5, T6, T4, T9, T3, Tc, T7, Ta, Tg, T8, Td, Th;
 	       V Tf, Te, Ti, Tj;
 	       T1 = LD(&(Rp[0]), ms, &(Rp[0]));
@@ -98,7 +98,7 @@ static void hc2cfdftv_4(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 6)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 6), MAKE_VOLATILE_STRIDE(16, rs)) {
 	       V T4, Tc, T9, Te, T1, T3, T2, Tb, T6, T8, T7, T5, Td, Tg, Th;
 	       V Ta, Tf, Tk, Tl, Ti, Tj;
 	       T1 = LD(&(Rp[0]), ms, &(Rp[0]));
diff --git a/rdft/simd/common/hc2cfdftv_6.c b/rdft/simd/common/hc2cfdftv_6.c
index 0ebf2d8..cf6dd82 100644
--- a/rdft/simd/common/hc2cfdftv_6.c
+++ b/rdft/simd/common/hc2cfdftv_6.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdftv_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(24, rs)) {
 	       V T5, T6, T3, Tj, T4, T9, Te, Th, T1, T2, Ti, Tc, Td, Tb, Tg;
 	       V T7, Ta, Tt, Tk, Tr, T8, Ts, Tf, Tx, Tu, To, Tl, Tw, Tv, Tn;
 	       V Tm, Tz, Ty, Tp, Tq;
@@ -122,7 +122,7 @@ static void hc2cfdftv_6(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 10)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 10), MAKE_VOLATILE_STRIDE(24, rs)) {
 	       V Ta, Tu, Tn, Tw, Ti, Tv, T1, T8, Tg, Tf, T7, T3, Te, T6, T2;
 	       V T4, T9, T5, Tk, Tm, Tj, Tl, Tc, Th, Tb, Td, Tr, Tp, Tq, To;
 	       V Tt, Ts, TA, Ty, Tz, Tx, TC, TB;
diff --git a/rdft/simd/common/hc2cfdftv_8.c b/rdft/simd/common/hc2cfdftv_8.c
index 8b6b879..7d6580c 100644
--- a/rdft/simd/common/hc2cfdftv_8.c
+++ b/rdft/simd/common/hc2cfdftv_8.c
@@ -19,7 +19,7 @@
  */
 
 /* This file was automatically generated --- DO NOT EDIT */
-/* Generated on Sat Apr 28 11:05:06 EDT 2012 */
+/* Generated on Sun Nov 25 07:42:29 EST 2012 */
 
 #include "codelet-rdft.h"
 
@@ -40,7 +40,7 @@ static void hc2cfdftv_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V T3, Tc, Tl, Ts, Tf, Tg, Te, Tp, T7, Ta, T1, T2, Tb, Tj, Tk;
 	       V Ti, Tr, T5, T6, T4, T9, Th, Tq, TC, T8, Td, TF, Tm, TG, TD;
 	       V Tt, Tu, Tn, TH, TL, TE, TK, Tz, Tv, Ty, To, TJ, TI, TN, TM;
@@ -141,7 +141,7 @@ static void hc2cfdftv_8(R *Rp, R *Ip, R *Rm, R *Im, const R *W, stride rs, INT m
      DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
      {
 	  INT m;
-	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(rs)) {
+	  for (m = mb, W = W + ((mb - 1) * ((TWVL / VL) * 14)); m < me; m = m + VL, Rp = Rp + (VL * ms), Ip = Ip + (VL * ms), Rm = Rm - (VL * ms), Im = Im - (VL * ms), W = W + (TWVL * 14), MAKE_VOLATILE_STRIDE(32, rs)) {
 	       V Ta, TE, Tr, TF, Tl, TK, Tw, TG, T1, T6, T3, T8, T2, T7, T4;
 	       V T9, T5, To, Tq, Tn, Tp, Tc, Th, Te, Tj, Td, Ti, Tf, Tk, Tb;
 	       V Tg, Tt, Tv, Ts, Tu, Ty, Tz, Tm, Tx, TC, TD, TA, TB, TI, TO;
diff --git a/rdft/simd/neon/Makefile.in b/rdft/simd/neon/Makefile.in
index 41fb30b..e8611f7 100644
--- a/rdft/simd/neon/Makefile.in
+++ b/rdft/simd/neon/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -105,6 +122,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_neon_codelets_la_SOURCES)
 DIST_SOURCES = $(am__librdft_neon_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/rdft/simd/sse2/Makefile.in b/rdft/simd/sse2/Makefile.in
index c835a79..f1986cf 100644
--- a/rdft/simd/sse2/Makefile.in
+++ b/rdft/simd/sse2/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -28,6 +28,23 @@
 # manual).
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -105,6 +122,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(librdft_sse2_codelets_la_SOURCES)
 DIST_SOURCES = $(am__librdft_sse2_codelets_la_SOURCES_DIST)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/reodft/Makefile.in b/reodft/Makefile.in
index d8d1d7d..9a1ecdd 100644
--- a/reodft/Makefile.in
+++ b/reodft/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -82,6 +99,11 @@ RECURSIVE_TARGETS = all-recursive check-recursive dvi-recursive \
 	install-pdf-recursive install-ps-recursive install-recursive \
 	installcheck-recursive installdirs-recursive pdf-recursive \
 	ps-recursive uninstall-recursive
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 RECURSIVE_CLEAN_TARGETS = mostlyclean-recursive clean-recursive	\
   distclean-recursive maintainer-clean-recursive
 AM_RECURSIVE_TARGETS = $(RECURSIVE_TARGETS:-recursive=) \
@@ -526,13 +548,10 @@ distdir: $(DISTFILES)
 	done
 	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
 	  if test "$$subdir" = .; then :; else \
-	    test -d "$(distdir)/$$subdir" \
-	    || $(MKDIR_P) "$(distdir)/$$subdir" \
-	    || exit 1; \
-	  fi; \
-	done
-	@list='$(DIST_SUBDIRS)'; for subdir in $$list; do \
-	  if test "$$subdir" = .; then :; else \
+	    $(am__make_dryrun) \
+	      || test -d "$(distdir)/$$subdir" \
+	      || $(MKDIR_P) "$(distdir)/$$subdir" \
+	      || exit 1; \
 	    dir1=$$subdir; dir2="$(distdir)/$$subdir"; \
 	    $(am__relativize); \
 	    new_distdir=$$reldir; \
diff --git a/simd-support/Makefile.in b/simd-support/Makefile.in
index b3bb866..01c6a23 100644
--- a/simd-support/Makefile.in
+++ b/simd-support/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -85,6 +102,11 @@ SOURCES = $(libsimd_sse2_nonportable_la_SOURCES) \
 	$(libsimd_support_la_SOURCES)
 DIST_SOURCES = $(libsimd_sse2_nonportable_la_SOURCES) \
 	$(libsimd_support_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/simd-support/simd-avx.h b/simd-support/simd-avx.h
index 349d4dd..741e345 100644
--- a/simd-support/simd-avx.h
+++ b/simd-support/simd-avx.h
@@ -120,8 +120,20 @@ static inline void ST(R *x, V v, INT ovs, const R *aligned_like)
      STOREL(x, l);
 }
 
-#define STM2 ST
-#define STN2(x, v0, v1, ovs) /* nop */
+#define STM2(x, v, ovs, aligned_like) /* no-op */
+static inline void STN2(R *x, V v0, V v1, INT ovs)
+{
+    V x0 = VSHUF(v0, v1, SHUFVALS(0, 1, 0, 1));
+    V x1 = VSHUF(v0, v1, SHUFVALS(2, 3, 2, 3));
+    __m128 h0 = _mm256_extractf128_ps(x0, 1);
+    __m128 l0 = _mm256_castps256_ps128(x0);
+    __m128 h1 = _mm256_extractf128_ps(x1, 1);
+    __m128 l1 = _mm256_castps256_ps128(x1);
+    *(__m128 *)(x + 3*ovs) = h1;
+    *(__m128 *)(x + 2*ovs) = h0;
+    *(__m128 *)(x + 1*ovs) = l1;
+    *(__m128 *)(x + 0*ovs) = l0;
+}
 
 #define STM4(x, v, ovs, aligned_like) /* no-op */
 #define STN4(x, v0, v1, v2, v3, ovs)				\
diff --git a/simd-support/simd-neon.h b/simd-support/simd-neon.h
index 9df7280..96a5a2d 100644
--- a/simd-support/simd-neon.h
+++ b/simd-support/simd-neon.h
@@ -25,9 +25,9 @@
 /* define these unconditionally, because they are used by
    taint.c which is compiled without neon */
 #define SIMD_SUFFIX _neon	/* for renaming */
-#define VL 1			/* SIMD complex vector length */
-#define SIMD_VSTRIDE_OKA(x) 1
-#define SIMD_STRIDE_OKPAIR SIMD_STRIDE_OKA
+#define VL 2            /* SIMD complex vector length */
+#define SIMD_VSTRIDE_OKA(x) ((x) == 2)
+#define SIMD_STRIDE_OKPAIR SIMD_STRIDE_OK
 
 #if defined(__GNUC__) && !defined(__ARM_NEON__)
 #error "compiling simd-neon.h requires -mfpu=neon or equivalent"
@@ -37,10 +37,11 @@
 
 /* FIXME: I am not sure whether this code assumes little-endian
    ordering.  VLIT may or may not be wrong for big-endian systems. */
-typedef float32x2_t V;
-#define VLIT(x0, x1) {x0, x1}
+typedef float32x4_t V;
+
+#define VLIT(x0, x1, x2, x3) {x0, x1, x2, x3}
 #define LDK(x) x
-#define DVK(var, val) const V var = VLIT(val, val)
+#define DVK(var, val) const V var = VLIT(val, val, val, val)
 
 /* NEON has FMA, but a three-operand FMA is not too useful
    for FFT purposes.  We normally compute
@@ -62,42 +63,68 @@ typedef float32x2_t V;
 #warning "--enable-fma on NEON is probably a bad idea (see source code)"
 #endif
 
-#define VADD(a, b) vadd_f32(a, b)
-#define VSUB(a, b) vsub_f32(a, b)
-#define VMUL(a, b) vmul_f32(a, b)
-#define VFMA(a, b, c) vmla_f32(c, a, b)	        /* a*b+c */
-#define VFNMS(a, b, c) vmls_f32(c, a, b)	/* FNMS=-(a*b-c) in powerpc terminology; MLS=c-a*b
+#define VADD(a, b) vaddq_f32(a, b)
+#define VSUB(a, b) vsubq_f32(a, b)
+#define VMUL(a, b) vmulq_f32(a, b)
+#define VFMA(a, b, c) vmlaq_f32(c, a, b)	        /* a*b+c */
+#define VFNMS(a, b, c) vmlsq_f32(c, a, b)	/* FNMS=-(a*b-c) in powerpc terminology; MLS=c-a*b
 						   in ARM terminology */
 #define VFMS(a, b, c) VSUB(VMUL(a, b), c)	/* FMS=a*b-c in powerpc terminology; no equivalent
-						   arm instruction */
+						   arm instruction (?) */
+
+static inline V LDA(const R *x, INT ivs, const R *aligned_like)
+{
+     (void) aligned_like;	/* UNUSED */
+     return vld1q_f32((const float32_t *)x);
+}
 
-#define LDA(x, ivs, aligned_like) vld1_f32((const float32_t *)x)
-#define LD LDA
+static inline V LD(const R *x, INT ivs, const R *aligned_like)
+{
+     (void) aligned_like;	/* UNUSED */
+     return vcombine_f32(vld1_f32((float32_t *)x), vld1_f32((float32_t *)(x + ivs)));
+}
+
+static inline void STA(R *x, V v, INT ovs, const R *aligned_like)
+{
+     (void) aligned_like;	/* UNUSED */
+     vst1q_f32((float32_t *)x, v);
+}
 
-#define STA(x, v, ovs, aligned_like) vst1_f32((float32_t *)x, v)
-#define ST STA
+static inline void ST(R *x, V v, INT ovs, const R *aligned_like)
+{
+     (void) aligned_like;	/* UNUSED */
+     /* WARNING: the extra_iter hack depends upon store-low occurring
+	after store-high */
+     vst1_f32((float32_t *)(x + ovs), vget_high_f32(v));
+     vst1_f32((float32_t *)x, vget_low_f32(v));
+}
 
-/* store and 2x2 complex transpose */
-#define STM2 STA
-#define STN2(x, v0, v1, ovs)	/* using the STM2 form */
+/* 2x2 complex transpose and store */
+#define STM2 ST
+#define STN2(x, v0, v1, ovs) /* nop */
 
-/* store and 2x2 real transpose */
+/* store and 4x4 real transpose */
 static inline void STM4(R *x, V v, INT ovs, const R *aligned_like)
 {
      (void) aligned_like;	/* UNUSED */
-     vst1_lane_f32((float32_t *)(x)      , v, 0);
-     vst1_lane_f32((float32_t *)(x + ovs), v, 1);
+     vst1_lane_f32((float32_t *)(x)      , vget_low_f32(v), 0);
+     vst1_lane_f32((float32_t *)(x + ovs), vget_low_f32(v), 1);
+     vst1_lane_f32((float32_t *)(x + 2 * ovs), vget_high_f32(v), 0);
+     vst1_lane_f32((float32_t *)(x + 3 * ovs), vget_high_f32(v), 1);
 }
+#define STN4(x, v0, v1, v2, v3, ovs)	/* use STM4 */
 
-#define STN4(x, v0, v1, v2, v3, ovs)	/* using the STM4 form */
-
-#define FLIP_RI(x) vrev64_f32(x)
+#define FLIP_RI(x) vrev64q_f32(x)
 
 static inline V VCONJ(V x)
 {
-     /* FIXME: there ought to be a way to XOR floating-point values */
-     const V pm = VLIT(1.0, -1.0);
+#if 1
+     static const uint32x4_t pm = {0, 0x80000000u, 0, 0x80000000u};
+     return vreinterpretq_f32_u32(veorq_u32(vreinterpretq_u32_f32(x), pm));
+#else
+     const V pm = VLIT(1.0, -1.0, 1.0, -1.0);
      return VMUL(x, pm);
+#endif
 }
 
 static inline V VBYI(V x)
@@ -107,25 +134,25 @@ static inline V VBYI(V x)
 
 static inline V VFMAI(V b, V c)
 {
-     const V mp = VLIT(-1.0, 1.0);
+     const V mp = VLIT(-1.0, 1.0, -1.0, 1.0);
      return VFMA(FLIP_RI(b), mp, c);
 }
 
 static inline V VFNMSI(V b, V c)
 {
-     const V mp = VLIT(-1.0, 1.0);
+     const V mp = VLIT(-1.0, 1.0, -1.0, 1.0);
      return VFNMS(FLIP_RI(b), mp, c);
 }
 
 static inline V VFMACONJ(V b, V c)
 {
-     const V pm = VLIT(1.0, -1.0);
+     const V pm = VLIT(1.0, -1.0, 1.0, -1.0);
      return VFMA(b, pm, c);
 }
 
 static inline V VFNMSCONJ(V b, V c)
 {
-     const V pm = VLIT(1.0, -1.0);
+     const V pm = VLIT(1.0, -1.0, 1.0, -1.0);
      return VFNMS(b, pm, c);
 }
 
@@ -134,13 +161,29 @@ static inline V VFMSCONJ(V b, V c)
      return VSUB(VCONJ(b), c);
 }
 
-#define VDUPL(x) vdup_lane_f32(x, 0)
-#define VDUPH(x) vdup_lane_f32(x, 1)
+#if 1
+#define VEXTRACT_REIM(tr, ti, tx)                               \
+{                                                               \
+     tr = vcombine_f32(vdup_lane_f32(vget_low_f32(tx), 0),      \
+                       vdup_lane_f32(vget_high_f32(tx), 0));    \
+     ti = vcombine_f32(vdup_lane_f32(vget_low_f32(tx), 1),      \
+                       vdup_lane_f32(vget_high_f32(tx), 1));    \
+}
+#else
+/* this alternative might be faster in an ideal world, but gcc likes
+   to spill VVV onto the stack */
+#define VEXTRACT_REIM(tr, ti, tx)               \
+{                                               \
+     float32x4x2_t vvv = vtrnq_f32(tx, tx);     \
+     tr = vvv.val[0];                           \
+     ti = vvv.val[1];                           \
+}
+#endif
 
 static inline V VZMUL(V tx, V sr)
 {
-     V tr = VDUPL(tx);
-     V ti = VDUPH(tx);
+     V tr, ti;
+     VEXTRACT_REIM(tr, ti, tx);
      tr = VMUL(sr, tr);
      sr = VBYI(sr);
      return VFMA(ti, sr, tr);
@@ -148,8 +191,8 @@ static inline V VZMUL(V tx, V sr)
 
 static inline V VZMULJ(V tx, V sr)
 {
-     V tr = VDUPL(tx);
-     V ti = VDUPH(tx);
+     V tr, ti;
+     VEXTRACT_REIM(tr, ti, tx);
      tr = VMUL(sr, tr);
      sr = VBYI(sr);
      return VFNMS(ti, sr, tr);
@@ -157,8 +200,8 @@ static inline V VZMULJ(V tx, V sr)
 
 static inline V VZMULI(V tx, V sr)
 {
-     V tr = VDUPL(tx);
-     V ti = VDUPH(tx);
+     V tr, ti;
+     VEXTRACT_REIM(tr, ti, tx);
      ti = VMUL(ti, sr);
      sr = VBYI(sr);
      return VFMS(tr, sr, ti);
@@ -166,55 +209,57 @@ static inline V VZMULI(V tx, V sr)
 
 static inline V VZMULIJ(V tx, V sr)
 {
-     V tr = VDUPL(tx);
-     V ti = VDUPH(tx);
+     V tr, ti;
+     VEXTRACT_REIM(tr, ti, tx);
      ti = VMUL(ti, sr);
      sr = VBYI(sr);
      return VFMA(tr, sr, ti);
 }
 
 /* twiddle storage #1: compact, slower */
-#define VTW1(v,x) {TW_CEXP, v, x}
-#define TWVL1 1
+#define VTW1(v,x) {TW_CEXP, v, x}, {TW_CEXP, v+1, x}
+#define TWVL1 VL
 static inline V BYTW1(const R *t, V sr)
 {
-     V tx = LD(t, 1, t);
+     V tx = LDA(t, 2, 0);
      return VZMUL(tx, sr);
 }
 
 static inline V BYTWJ1(const R *t, V sr)
 {
-     V tx = LD(t, 1, t);
+     V tx = LDA(t, 2, 0);
      return VZMULJ(tx, sr);
 }
 
 /* twiddle storage #2: twice the space, faster (when in cache) */
-#define VTW2(v,x) {TW_COS, v, x}, {TW_COS, v, x}, {TW_SIN, v, -x}, {TW_SIN, v, x}
-#define TWVL2 2
+#  define VTW2(v,x)							\
+  {TW_COS, v, x}, {TW_COS, v, x}, {TW_COS, v+1, x}, {TW_COS, v+1, x},	\
+  {TW_SIN, v, -x}, {TW_SIN, v, x}, {TW_SIN, v+1, -x}, {TW_SIN, v+1, x}
+#define TWVL2 (2 * VL)
 
 static inline V BYTW2(const R *t, V sr)
 {
-     const V *twp = (const V *) t;
      V si = FLIP_RI(sr);
-     V tr = twp[0], ti = twp[1];
+     V tr = LDA(t, 2, 0), ti = LDA(t+2*VL, 2, 0);
      return VFMA(ti, si, VMUL(tr, sr));
 }
 
 static inline V BYTWJ2(const R *t, V sr)
 {
-     const V *twp = (const V *) t;
      V si = FLIP_RI(sr);
-     V tr = twp[0], ti = twp[1];
+     V tr = LDA(t, 2, 0), ti = LDA(t+2*VL, 2, 0);
      return VFNMS(ti, si, VMUL(tr, sr));
 }
 
 /* twiddle storage #3 */
-#define VTW3(v,x) {TW_CEXP, v, x}
-#define TWVL3 1
+#  define VTW3(v,x) {TW_CEXP, v, x}, {TW_CEXP, v+1, x}
+#  define TWVL3 (VL)
 
 /* twiddle storage for split arrays */
-#define VTWS(v,x) {TW_COS, v, x}, {TW_COS, v+1, x}, {TW_SIN, v, x}, {TW_SIN, v+1, x}
-#define TWVLS 2
+#  define VTWS(v,x)							  \
+    {TW_COS, v, x}, {TW_COS, v+1, x}, {TW_COS, v+2, x}, {TW_COS, v+3, x}, \
+    {TW_SIN, v, x}, {TW_SIN, v+1, x}, {TW_SIN, v+2, x}, {TW_SIN, v+3, x}
+#define TWVLS (2 * VL)
 
 #define VLEAVE()		/* nothing */
 
diff --git a/support/Makefile.in b/support/Makefile.in
index 967b42a..9ec1d4b 100644
--- a/support/Makefile.in
+++ b/support/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -15,6 +15,23 @@
 
 @SET_MAKE@
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -55,6 +72,11 @@ CONFIG_CLEAN_FILES =
 CONFIG_CLEAN_VPATH_FILES =
 SOURCES =
 DIST_SOURCES =
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
 ACLOCAL = @ACLOCAL@
 ALLOCA = @ALLOCA@
diff --git a/tests/Makefile.in b/tests/Makefile.in
index 0db56f1..e200da0 100644
--- a/tests/Makefile.in
+++ b/tests/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -81,6 +98,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(bench_SOURCES)
 DIST_SOURCES = $(bench_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
diff --git a/threads/Makefile.in b/threads/Makefile.in
index 84102cf..51dcee3 100644
--- a/threads/Makefile.in
+++ b/threads/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -16,6 +16,23 @@
 @SET_MAKE@
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -144,6 +161,11 @@ SOURCES = $(libfftw3 at PREC_SUFFIX@_omp_la_SOURCES) \
 	$(libfftw3 at PREC_SUFFIX@_threads_la_SOURCES)
 DIST_SOURCES = $(libfftw3 at PREC_SUFFIX@_omp_la_SOURCES) \
 	$(libfftw3 at PREC_SUFFIX@_threads_la_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 ETAGS = etags
 CTAGS = ctags
 DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
@@ -352,7 +374,6 @@ $(ACLOCAL_M4): @MAINTAINER_MODE_TRUE@ $(am__aclocal_m4_deps)
 $(am__aclocal_m4_deps):
 install-libLTLIBRARIES: $(lib_LTLIBRARIES)
 	@$(NORMAL_INSTALL)
-	test -z "$(libdir)" || $(MKDIR_P) "$(DESTDIR)$(libdir)"
 	@list='$(lib_LTLIBRARIES)'; test -n "$(libdir)" || list=; \
 	list2=; for p in $$list; do \
 	  if test -f $$p; then \
@@ -360,6 +381,8 @@ install-libLTLIBRARIES: $(lib_LTLIBRARIES)
 	  else :; fi; \
 	done; \
 	test -z "$$list2" || { \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(libdir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(libdir)" || exit 1; \
 	  echo " $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL) $(INSTALL_STRIP_FLAG) $$list2 '$(DESTDIR)$(libdir)'"; \
 	  $(LIBTOOL) $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=install $(INSTALL) $(INSTALL_STRIP_FLAG) $$list2 "$(DESTDIR)$(libdir)"; \
 	}
diff --git a/tools/Makefile.in b/tools/Makefile.in
index 72d7beb..eda1e85 100644
--- a/tools/Makefile.in
+++ b/tools/Makefile.in
@@ -1,4 +1,4 @@
-# Makefile.in generated by automake 1.11.3 from Makefile.am.
+# Makefile.in generated by automake 1.11.6 from Makefile.am.
 # @configure_input@
 
 # Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
@@ -17,6 +17,23 @@
 
 
 VPATH = @srcdir@
+am__make_dryrun = \
+  { \
+    am__dry=no; \
+    case $$MAKEFLAGS in \
+      *\\[\ \	]*) \
+        echo 'am--echo: ; @echo "AM"  OK' | $(MAKE) -f - 2>/dev/null \
+          | grep '^AM OK$$' >/dev/null || am__dry=yes;; \
+      *) \
+        for am__flg in $$MAKEFLAGS; do \
+          case $$am__flg in \
+            *=*|--*) ;; \
+            *n*) am__dry=yes; break;; \
+          esac; \
+        done;; \
+    esac; \
+    test $$am__dry = yes; \
+  }
 pkgdatadir = $(datadir)/@PACKAGE@
 pkgincludedir = $(includedir)/@PACKAGE@
 pkglibdir = $(libdir)/@PACKAGE@
@@ -118,6 +135,11 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(fftw at PREC_SUFFIX@_wisdom_SOURCES)
 DIST_SOURCES = $(fftw at PREC_SUFFIX@_wisdom_SOURCES)
+am__can_run_installinfo = \
+  case $$AM_UPDATE_INFO_DIR in \
+    n|no|NO) false;; \
+    *) (install-info --version) >/dev/null 2>&1;; \
+  esac
 man1dir = $(mandir)/man1
 NROFF = nroff
 MANS = $(dist_man_MANS)
@@ -324,8 +346,11 @@ fftw-wisdom-to-conf: $(top_builddir)/config.status $(srcdir)/fftw-wisdom-to-conf
 	cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@
 install-binPROGRAMS: $(bin_PROGRAMS)
 	@$(NORMAL_INSTALL)
-	test -z "$(bindir)" || $(MKDIR_P) "$(DESTDIR)$(bindir)"
 	@list='$(bin_PROGRAMS)'; test -n "$(bindir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(bindir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(bindir)" || exit 1; \
+	fi; \
 	for p in $$list; do echo "$$p $$p"; done | \
 	sed 's/$(EXEEXT)$$//' | \
 	while read p p1; do if test -f $$p || test -f $$p1; \
@@ -370,8 +395,11 @@ fftw at PREC_SUFFIX@-wisdom$(EXEEXT): $(fftw at PREC_SUFFIX@_wisdom_OBJECTS) $(fftw at PR
 	$(fftw at PREC_SUFFIX@_wisdom_LINK) $(fftw at PREC_SUFFIX@_wisdom_OBJECTS) $(fftw at PREC_SUFFIX@_wisdom_LDADD) $(LIBS)
 install-binSCRIPTS: $(bin_SCRIPTS)
 	@$(NORMAL_INSTALL)
-	test -z "$(bindir)" || $(MKDIR_P) "$(DESTDIR)$(bindir)"
 	@list='$(bin_SCRIPTS)'; test -n "$(bindir)" || list=; \
+	if test -n "$$list"; then \
+	  echo " $(MKDIR_P) '$(DESTDIR)$(bindir)'"; \
+	  $(MKDIR_P) "$(DESTDIR)$(bindir)" || exit 1; \
+	fi; \
 	for p in $$list; do \
 	  if test -f "$$p"; then d=; else d="$(srcdir)/"; fi; \
 	  if test -f "$$d$$p"; then echo "$$d$$p"; echo "$$p"; else :; fi; \
@@ -451,11 +479,18 @@ clean-libtool:
 	-rm -rf .libs _libs
 install-man1: $(dist_man_MANS)
 	@$(NORMAL_INSTALL)
-	test -z "$(man1dir)" || $(MKDIR_P) "$(DESTDIR)$(man1dir)"
-	@list=''; test -n "$(man1dir)" || exit 0; \
-	{ for i in $$list; do echo "$$i"; done; \
-	l2='$(dist_man_MANS)'; for i in $$l2; do echo "$$i"; done | \
-	  sed -n '/\.1[a-z]*$$/p'; \
+	@list1=''; \
+	list2='$(dist_man_MANS)'; \
+	test -n "$(man1dir)" \
+	  && test -n "`echo $$list1$$list2`" \
+	  || exit 0; \
+	echo " $(MKDIR_P) '$(DESTDIR)$(man1dir)'"; \
+	$(MKDIR_P) "$(DESTDIR)$(man1dir)" || exit 1; \
+	{ for i in $$list1; do echo "$$i"; done;  \
+	if test -n "$$list2"; then \
+	  for i in $$list2; do echo "$$i"; done \
+	    | sed -n '/\.1[a-z]*$$/p'; \
+	fi; \
 	} | while read p; do \
 	  if test -f $$p; then d=; else d="$(srcdir)/"; fi; \
 	  echo "$$d$$p"; echo "$$p"; \

-- 
Library for computing Fast Fourier Transforms



More information about the debian-science-commits mailing list