Merge remote-tracking branch 'upstream/devel' into asm-elf-pie

45 files changed, 1218 insertions, 425 deletions

diff --git a/.travis.yml b/.travis.yml
index 33b1a80..9d7d9c0 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -13,13 +13,16 @@ addons:
   apt:
     packages:
       - nasm
+      - gcc-multilib
+      - g++-multilib
 
 env:
-  - CONF_FLAGS=""
+  - CONF_FLAGS="" DISTCHECK=1
   - CONF_FLAGS="--without-simd"
+  - CONF_FLAGS="--host=i686-linux" CFLAGS=-m32
 
 script:
   - ./bootstrap
   - ./configure $CONF_FLAGS
-  - make CFLAGS="-O2 -Wall -Wwrite-strings -Werror"
-  - make distcheck
+  - make CFLAGS="$CFLAGS -O2 -Wall -Wwrite-strings -Werror"
+  - test -z "$DISTCHECK" || make distcheck
diff --git a/configure.ac b/configure.ac
index 8ff62c6..4d85b01 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,7 +1,7 @@
 # Process this file with autoconf to produce a configure script
 
 AC_PREREQ(2.59)
-AC_INIT([rfxcodec], [0.1.0], [jay.sorg@gmail.com])
+AC_INIT([rfxcodec], [0.1.0], [xrdp-devel@googlegroups.com])
 AC_CONFIG_HEADERS(config_ac.h:config_ac-h.in)
 AM_INIT_AUTOMAKE([1.6 foreign])
 AC_CONFIG_MACRO_DIR([m4])
@@ -11,6 +11,9 @@ AC_C_CONST
 AC_PROG_LIBTOOL
 PKG_INSTALLDIR
 
+AX_CFLAGS_WARN_ALL
+AX_APPEND_COMPILE_FLAGS([-Wwrite-strings])
+
 # SIMD is optional
 AC_ARG_WITH([simd],
     AC_HELP_STRING([--without-simd],[Omit SIMD extensions.]))
diff --git a/include/rfxcodec_encode.h b/include/rfxcodec_encode.h
index 04112b0..1f9082d 100644
--- a/include/rfxcodec_encode.h
+++ b/include/rfxcodec_encode.h
@@ -62,13 +62,13 @@ rfxcodec_encode_destroy(void *handle);
  * 9 - HH1 */
 int
 rfxcodec_encode(void *handle, char *cdata, int *cdata_bytes,
-                char *buf, int width, int height, int stride_bytes,
+                const char *buf, int width, int height, int stride_bytes,
                 const struct rfx_rect *region, int num_region,
                 const struct rfx_tile *tiles, int num_tiles,
                 const char *quants, int num_quants);
 int
 rfxcodec_encode_ex(void *handle, char *cdata, int *cdata_bytes,
-                   char *buf, int width, int height, int stride_bytes,
+                   const char *buf, int width, int height, int stride_bytes,
                    const struct rfx_rect *region, int num_region,
                    const struct rfx_tile *tiles, int num_tiles,
                    const char *quants, int num_quants, int flags);
diff --git a/m4/ax_append_compile_flags.m4 b/m4/ax_append_compile_flags.m4
new file mode 100644
index 0000000..5b6f1af
--- /dev/null
+++ b/m4/ax_append_compile_flags.m4
@@ -0,0 +1,67 @@
+# ============================================================================
+#  https://www.gnu.org/software/autoconf-archive/ax_append_compile_flags.html
+# ============================================================================
+#
+# SYNOPSIS
+#
+#   AX_APPEND_COMPILE_FLAGS([FLAG1 FLAG2 ...], [FLAGS-VARIABLE], [EXTRA-FLAGS], [INPUT])
+#
+# DESCRIPTION
+#
+#   For every FLAG1, FLAG2 it is checked whether the compiler works with the
+#   flag.  If it does, the flag is added FLAGS-VARIABLE
+#
+#   If FLAGS-VARIABLE is not specified, the current language's flags (e.g.
+#   CFLAGS) is used.  During the check the flag is always added to the
+#   current language's flags.
+#
+#   If EXTRA-FLAGS is defined, it is added to the current language's default
+#   flags (e.g. CFLAGS) when the check is done.  The check is thus made with
+#   the flags: "CFLAGS EXTRA-FLAGS FLAG".  This can for example be used to
+#   force the compiler to issue an error when a bad flag is given.
+#
+#   INPUT gives an alternative input source to AC_COMPILE_IFELSE.
+#
+#   NOTE: This macro depends on the AX_APPEND_FLAG and
+#   AX_CHECK_COMPILE_FLAG. Please keep this macro in sync with
+#   AX_APPEND_LINK_FLAGS.
+#
+# LICENSE
+#
+#   Copyright (c) 2011 Maarten Bosmans <mkbosmans@gmail.com>
+#
+#   This program is free software: you can redistribute it and/or modify it
+#   under the terms of the GNU General Public License as published by the
+#   Free Software Foundation, either version 3 of the License, or (at your
+#   option) any later version.
+#
+#   This program is distributed in the hope that it will be useful, but
+#   WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+#   Public License for more details.
+#
+#   You should have received a copy of the GNU General Public License along
+#   with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+#   As a special exception, the respective Autoconf Macro's copyright owner
+#   gives unlimited permission to copy, distribute and modify the configure
+#   scripts that are the output of Autoconf when processing the Macro. You
+#   need not follow the terms of the GNU General Public License when using
+#   or distributing such scripts, even though portions of the text of the
+#   Macro appear in them. The GNU General Public License (GPL) does govern
+#   all other use of the material that constitutes the Autoconf Macro.
+#
+#   This special exception to the GPL applies to versions of the Autoconf
+#   Macro released by the Autoconf Archive. When you make and distribute a
+#   modified version of the Autoconf Macro, you may extend this special
+#   exception to the GPL to apply to your modified version as well.
+
+#serial 6
+
+AC_DEFUN([AX_APPEND_COMPILE_FLAGS],
+[AX_REQUIRE_DEFINED([AX_CHECK_COMPILE_FLAG])
+AX_REQUIRE_DEFINED([AX_APPEND_FLAG])
+for flag in $1; do
+  AX_CHECK_COMPILE_FLAG([$flag], [AX_APPEND_FLAG([$flag], [$2])], [], [$3], [$4])
+done
+])dnl AX_APPEND_COMPILE_FLAGS
diff --git a/m4/ax_append_flag.m4 b/m4/ax_append_flag.m4
new file mode 100644
index 0000000..e8c5312
--- /dev/null
+++ b/m4/ax_append_flag.m4
@@ -0,0 +1,71 @@
+# ===========================================================================
+#      https://www.gnu.org/software/autoconf-archive/ax_append_flag.html
+# ===========================================================================
+#
+# SYNOPSIS
+#
+#   AX_APPEND_FLAG(FLAG, [FLAGS-VARIABLE])
+#
+# DESCRIPTION
+#
+#   FLAG is appended to the FLAGS-VARIABLE shell variable, with a space
+#   added in between.
+#
+#   If FLAGS-VARIABLE is not specified, the current language's flags (e.g.
+#   CFLAGS) is used.  FLAGS-VARIABLE is not changed if it already contains
+#   FLAG.  If FLAGS-VARIABLE is unset in the shell, it is set to exactly
+#   FLAG.
+#
+#   NOTE: Implementation based on AX_CFLAGS_GCC_OPTION.
+#
+# LICENSE
+#
+#   Copyright (c) 2008 Guido U. Draheim <guidod@gmx.de>
+#   Copyright (c) 2011 Maarten Bosmans <mkbosmans@gmail.com>
+#
+#   This program is free software: you can redistribute it and/or modify it
+#   under the terms of the GNU General Public License as published by the
+#   Free Software Foundation, either version 3 of the License, or (at your
+#   option) any later version.
+#
+#   This program is distributed in the hope that it will be useful, but
+#   WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+#   Public License for more details.
+#
+#   You should have received a copy of the GNU General Public License along
+#   with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+#   As a special exception, the respective Autoconf Macro's copyright owner
+#   gives unlimited permission to copy, distribute and modify the configure
+#   scripts that are the output of Autoconf when processing the Macro. You
+#   need not follow the terms of the GNU General Public License when using
+#   or distributing such scripts, even though portions of the text of the
+#   Macro appear in them. The GNU General Public License (GPL) does govern
+#   all other use of the material that constitutes the Autoconf Macro.
+#
+#   This special exception to the GPL applies to versions of the Autoconf
+#   Macro released by the Autoconf Archive. When you make and distribute a
+#   modified version of the Autoconf Macro, you may extend this special
+#   exception to the GPL to apply to your modified version as well.
+
+#serial 7
+
+AC_DEFUN([AX_APPEND_FLAG],
+[dnl
+AC_PREREQ(2.64)dnl for _AC_LANG_PREFIX and AS_VAR_SET_IF
+AS_VAR_PUSHDEF([FLAGS], [m4_default($2,_AC_LANG_PREFIX[FLAGS])])
+AS_VAR_SET_IF(FLAGS,[
+  AS_CASE([" AS_VAR_GET(FLAGS) "],
+    [*" $1 "*], [AC_RUN_LOG([: FLAGS already contains $1])],
+    [
+     AS_VAR_APPEND(FLAGS,[" $1"])
+     AC_RUN_LOG([: FLAGS="$FLAGS"])
+    ])
+  ],
+  [
+  AS_VAR_SET(FLAGS,[$1])
+  AC_RUN_LOG([: FLAGS="$FLAGS"])
+  ])
+AS_VAR_POPDEF([FLAGS])dnl
+])dnl AX_APPEND_FLAG
diff --git a/m4/ax_cflags_warn_all.m4 b/m4/ax_cflags_warn_all.m4
new file mode 100644
index 0000000..094577e
--- /dev/null
+++ b/m4/ax_cflags_warn_all.m4
@@ -0,0 +1,122 @@
+# ===========================================================================
+#    https://www.gnu.org/software/autoconf-archive/ax_cflags_warn_all.html
+# ===========================================================================
+#
+# SYNOPSIS
+#
+#   AX_CFLAGS_WARN_ALL   [(shellvar [,default, [A/NA]])]
+#   AX_CXXFLAGS_WARN_ALL [(shellvar [,default, [A/NA]])]
+#   AX_FCFLAGS_WARN_ALL  [(shellvar [,default, [A/NA]])]
+#
+# DESCRIPTION
+#
+#   Try to find a compiler option that enables most reasonable warnings.
+#
+#   For the GNU compiler it will be -Wall (and -ansi -pedantic) The result
+#   is added to the shellvar being CFLAGS, CXXFLAGS, or FCFLAGS by default.
+#
+#   Currently this macro knows about the GCC, Solaris, Digital Unix, AIX,
+#   HP-UX, IRIX, NEC SX-5 (Super-UX 10), Cray J90 (Unicos 10.0.0.8), and
+#   Intel compilers.  For a given compiler, the Fortran flags are much more
+#   experimental than their C equivalents.
+#
+#    - $1 shell-variable-to-add-to : CFLAGS, CXXFLAGS, or FCFLAGS
+#    - $2 add-value-if-not-found : nothing
+#    - $3 action-if-found : add value to shellvariable
+#    - $4 action-if-not-found : nothing
+#
+#   NOTE: These macros depend on AX_APPEND_FLAG.
+#
+# LICENSE
+#
+#   Copyright (c) 2008 Guido U. Draheim <guidod@gmx.de>
+#   Copyright (c) 2010 Rhys Ulerich <rhys.ulerich@gmail.com>
+#
+#   This program is free software; you can redistribute it and/or modify it
+#   under the terms of the GNU General Public License as published by the
+#   Free Software Foundation; either version 3 of the License, or (at your
+#   option) any later version.
+#
+#   This program is distributed in the hope that it will be useful, but
+#   WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+#   Public License for more details.
+#
+#   You should have received a copy of the GNU General Public License along
+#   with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+#   As a special exception, the respective Autoconf Macro's copyright owner
+#   gives unlimited permission to copy, distribute and modify the configure
+#   scripts that are the output of Autoconf when processing the Macro. You
+#   need not follow the terms of the GNU General Public License when using
+#   or distributing such scripts, even though portions of the text of the
+#   Macro appear in them. The GNU General Public License (GPL) does govern
+#   all other use of the material that constitutes the Autoconf Macro.
+#
+#   This special exception to the GPL applies to versions of the Autoconf
+#   Macro released by the Autoconf Archive. When you make and distribute a
+#   modified version of the Autoconf Macro, you may extend this special
+#   exception to the GPL to apply to your modified version as well.
+
+#serial 16
+
+AC_DEFUN([AX_FLAGS_WARN_ALL],[dnl
+AS_VAR_PUSHDEF([FLAGS],[_AC_LANG_PREFIX[]FLAGS])dnl
+AS_VAR_PUSHDEF([VAR],[ac_cv_[]_AC_LANG_ABBREV[]flags_warn_all])dnl
+AC_CACHE_CHECK([m4_ifval($1,$1,FLAGS) for maximum warnings],
+VAR,[VAR="no, unknown"
+ac_save_[]FLAGS="$[]FLAGS"
+for ac_arg dnl
+in "-warn all  % -warn all"   dnl Intel
+   "-pedantic  % -Wall"       dnl GCC
+   "-xstrconst % -v"          dnl Solaris C
+   "-std1      % -verbose -w0 -warnprotos" dnl Digital Unix
+   "-qlanglvl=ansi % -qsrcmsg -qinfo=all:noppt:noppc:noobs:nocnd" dnl AIX
+   "-ansi -ansiE % -fullwarn" dnl IRIX
+   "+ESlit     % +w1"         dnl HP-UX C
+   "-Xc        % -pvctl[,]fullmsg" dnl NEC SX-5 (Super-UX 10)
+   "-h conform % -h msglevel 2" dnl Cray C (Unicos)
+   #
+do FLAGS="$ac_save_[]FLAGS "`echo $ac_arg | sed -e 's,%%.*,,' -e 's,%,,'`
+   AC_COMPILE_IFELSE([AC_LANG_PROGRAM],
+                     [VAR=`echo $ac_arg | sed -e 's,.*% *,,'` ; break])
+done
+FLAGS="$ac_save_[]FLAGS"
+])
+AS_VAR_POPDEF([FLAGS])dnl
+AX_REQUIRE_DEFINED([AX_APPEND_FLAG])
+case ".$VAR" in
+     .ok|.ok,*) m4_ifvaln($3,$3) ;;
+   .|.no|.no,*) m4_default($4,[m4_ifval($2,[AX_APPEND_FLAG([$2], [$1])])]) ;;
+   *) m4_default($3,[AX_APPEND_FLAG([$VAR], [$1])]) ;;
+esac
+AS_VAR_POPDEF([VAR])dnl
+])dnl AX_FLAGS_WARN_ALL
+dnl  implementation tactics:
+dnl   the for-argument contains a list of options. The first part of
+dnl   these does only exist to detect the compiler - usually it is
+dnl   a global option to enable -ansi or -extrawarnings. All other
+dnl   compilers will fail about it. That was needed since a lot of
+dnl   compilers will give false positives for some option-syntax
+dnl   like -Woption or -Xoption as they think of it is a pass-through
+dnl   to later compile stages or something. The "%" is used as a
+dnl   delimiter. A non-option comment can be given after "%%" marks
+dnl   which will be shown but not added to the respective C/CXXFLAGS.
+
+AC_DEFUN([AX_CFLAGS_WARN_ALL],[dnl
+AC_LANG_PUSH([C])
+AX_FLAGS_WARN_ALL([$1], [$2], [$3], [$4])
+AC_LANG_POP([C])
+])
+
+AC_DEFUN([AX_CXXFLAGS_WARN_ALL],[dnl
+AC_LANG_PUSH([C++])
+AX_FLAGS_WARN_ALL([$1], [$2], [$3], [$4])
+AC_LANG_POP([C++])
+])
+
+AC_DEFUN([AX_FCFLAGS_WARN_ALL],[dnl
+AC_LANG_PUSH([Fortran])
+AX_FLAGS_WARN_ALL([$1], [$2], [$3], [$4])
+AC_LANG_POP([Fortran])
+])
diff --git a/m4/ax_check_compile_flag.m4 b/m4/ax_check_compile_flag.m4
new file mode 100644
index 0000000..dcabb92
--- /dev/null
+++ b/m4/ax_check_compile_flag.m4
@@ -0,0 +1,74 @@
+# ===========================================================================
+#  https://www.gnu.org/software/autoconf-archive/ax_check_compile_flag.html
+# ===========================================================================
+#
+# SYNOPSIS
+#
+#   AX_CHECK_COMPILE_FLAG(FLAG, [ACTION-SUCCESS], [ACTION-FAILURE], [EXTRA-FLAGS], [INPUT])
+#
+# DESCRIPTION
+#
+#   Check whether the given FLAG works with the current language's compiler
+#   or gives an error.  (Warnings, however, are ignored)
+#
+#   ACTION-SUCCESS/ACTION-FAILURE are shell commands to execute on
+#   success/failure.
+#
+#   If EXTRA-FLAGS is defined, it is added to the current language's default
+#   flags (e.g. CFLAGS) when the check is done.  The check is thus made with
+#   the flags: "CFLAGS EXTRA-FLAGS FLAG".  This can for example be used to
+#   force the compiler to issue an error when a bad flag is given.
+#
+#   INPUT gives an alternative input source to AC_COMPILE_IFELSE.
+#
+#   NOTE: Implementation based on AX_CFLAGS_GCC_OPTION. Please keep this
+#   macro in sync with AX_CHECK_{PREPROC,LINK}_FLAG.
+#
+# LICENSE
+#
+#   Copyright (c) 2008 Guido U. Draheim <guidod@gmx.de>
+#   Copyright (c) 2011 Maarten Bosmans <mkbosmans@gmail.com>
+#
+#   This program is free software: you can redistribute it and/or modify it
+#   under the terms of the GNU General Public License as published by the
+#   Free Software Foundation, either version 3 of the License, or (at your
+#   option) any later version.
+#
+#   This program is distributed in the hope that it will be useful, but
+#   WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+#   Public License for more details.
+#
+#   You should have received a copy of the GNU General Public License along
+#   with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+#   As a special exception, the respective Autoconf Macro's copyright owner
+#   gives unlimited permission to copy, distribute and modify the configure
+#   scripts that are the output of Autoconf when processing the Macro. You
+#   need not follow the terms of the GNU General Public License when using
+#   or distributing such scripts, even though portions of the text of the
+#   Macro appear in them. The GNU General Public License (GPL) does govern
+#   all other use of the material that constitutes the Autoconf Macro.
+#
+#   This special exception to the GPL applies to versions of the Autoconf
+#   Macro released by the Autoconf Archive. When you make and distribute a
+#   modified version of the Autoconf Macro, you may extend this special
+#   exception to the GPL to apply to your modified version as well.
+
+#serial 5
+
+AC_DEFUN([AX_CHECK_COMPILE_FLAG],
+[AC_PREREQ(2.64)dnl for _AC_LANG_PREFIX and AS_VAR_IF
+AS_VAR_PUSHDEF([CACHEVAR],[ax_cv_check_[]_AC_LANG_ABBREV[]flags_$4_$1])dnl
+AC_CACHE_CHECK([whether _AC_LANG compiler accepts $1], CACHEVAR, [
+  ax_check_save_flags=$[]_AC_LANG_PREFIX[]FLAGS
+  _AC_LANG_PREFIX[]FLAGS="$[]_AC_LANG_PREFIX[]FLAGS $4 $1"
+  AC_COMPILE_IFELSE([m4_default([$5],[AC_LANG_PROGRAM()])],
+    [AS_VAR_SET(CACHEVAR,[yes])],
+    [AS_VAR_SET(CACHEVAR,[no])])
+  _AC_LANG_PREFIX[]FLAGS=$ax_check_save_flags])
+AS_VAR_IF(CACHEVAR,yes,
+  [m4_default([$2], :)],
+  [m4_default([$3], :)])
+AS_VAR_POPDEF([CACHEVAR])dnl
+])dnl AX_CHECK_COMPILE_FLAGS
diff --git a/m4/ax_require_defined.m4 b/m4/ax_require_defined.m4
new file mode 100644
index 0000000..17c3eab
--- /dev/null
+++ b/m4/ax_require_defined.m4
@@ -0,0 +1,37 @@
+# ===========================================================================
+#    https://www.gnu.org/software/autoconf-archive/ax_require_defined.html
+# ===========================================================================
+#
+# SYNOPSIS
+#
+#   AX_REQUIRE_DEFINED(MACRO)
+#
+# DESCRIPTION
+#
+#   AX_REQUIRE_DEFINED is a simple helper for making sure other macros have
+#   been defined and thus are available for use.  This avoids random issues
+#   where a macro isn't expanded.  Instead the configure script emits a
+#   non-fatal:
+#
+#     ./configure: line 1673: AX_CFLAGS_WARN_ALL: command not found
+#
+#   It's like AC_REQUIRE except it doesn't expand the required macro.
+#
+#   Here's an example:
+#
+#     AX_REQUIRE_DEFINED([AX_CHECK_LINK_FLAG])
+#
+# LICENSE
+#
+#   Copyright (c) 2014 Mike Frysinger <vapier@gentoo.org>
+#
+#   Copying and distribution of this file, with or without modification, are
+#   permitted in any medium without royalty provided the copyright notice
+#   and this notice are preserved. This file is offered as-is, without any
+#   warranty.
+
+#serial 2
+
+AC_DEFUN([AX_REQUIRE_DEFINED], [dnl
+  m4_ifndef([$1], [m4_fatal([macro ]$1[ is not defined; is a m4 file missing?])])
+])dnl AX_REQUIRE_DEFINED
diff --git a/m4/nasm.m4 b/m4/nasm.m4
index a8816cc..9911684 100644
--- a/m4/nasm.m4
+++ b/m4/nasm.m4
@@ -106,13 +106,14 @@ AC_MSG_CHECKING([whether the assembler ($NASM $NAFLAGS) works])
 cat > conftest.asm <<EOF
 [%line __oline__ "configure"
         section .text
-        global  _main,main
-_main:
-main:   xor     eax,eax
+        global  _nasmfunc, nasmfunc
+_nasmfunc:
+nasmfunc:
+        xor     eax, eax
         ret
 ]EOF
-try_nasm='$NASM $NAFLAGS -o conftest.o conftest.asm'
-if AC_TRY_EVAL(try_nasm) && test -s conftest.o; then
+try_nasm='$NASM $NAFLAGS -o conftest-nasm.o conftest.asm'
+if AC_TRY_EVAL(try_nasm) && test -s conftest-nasm.o; then
   AC_MSG_RESULT(yes)
 else
   echo "configure: failed program was:" >&AC_FD_CC
@@ -123,13 +124,19 @@ else
 fi
 
 AC_MSG_CHECKING([whether the linker accepts assembler output])
-try_nasm='${CC-cc} -o conftest${ac_exeext} $LDFLAGS conftest.o $LIBS 1>&AC_FD_CC'
-if AC_TRY_EVAL(try_nasm) && test -s conftest${ac_exeext}; then
-  rm -rf conftest*
-  AC_MSG_RESULT(yes)
-else
-  rm -rf conftest*
-  AC_MSG_RESULT(no)
+nasm_save_LIBS="$LIBS"
+LIBS="conftest-nasm.o $LIBS"
+AC_LINK_IFELSE(
+  [AC_LANG_PROGRAM([[
+    #ifdef __cplusplus
+    extern "C"
+    #endif
+    void nasmfunc(void);
+  ]], [[nasmfunc();]])], [nasm_link_ok=yes], [nasm_link_ok=no])
+LIBS="$nasm_save_LIBS"
+AC_MSG_RESULT([$nasm_link_ok])
+
+if test "x$nasm_link_ok" = "xno"; then
   AC_MSG_ERROR([configuration problem: maybe object file format mismatch.])
 fi
 
diff --git a/src/Makefile.am b/src/Makefile.am
index 3449dd7..4e20371 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -1,4 +1,4 @@
-EXTRA_DIST = nasm_lt.sh
+EXTRA_DIST = common.asm nasm_lt.sh
 
 SUBDIRS =
 
@@ -29,11 +29,14 @@ noinst_HEADERS = \
   rfxencode_quantization.h \
   rfxencode_rlgr1.h \
   rfxencode_rlgr3.h \
-  rfxencode_tile.h
+  rfxencode_tile.h \
+  rfxencode_diff_rlgr1.h \
+  rfxencode_diff_rlgr3.h
 
 lib_LTLIBRARIES = librfxencode.la
 
 librfxencode_la_SOURCES = $(noinst_HEADERS) rfxencode.c \
   rfxcompose.c rfxencode_tile.c rfxencode_dwt.c \
   rfxencode_quantization.c rfxencode_differential.c \
-  rfxencode_rlgr1.c rfxencode_rlgr3.c rfxencode_alpha.c
+  rfxencode_rlgr1.c rfxencode_rlgr3.c rfxencode_alpha.c \
+  rfxencode_diff_rlgr1.c rfxencode_diff_rlgr3.c
diff --git a/src/amd64/Makefile.am b/src/amd64/Makefile.am
index 575212e..07aae69 100644
--- a/src/amd64/Makefile.am
+++ b/src/amd64/Makefile.am
@@ -1,7 +1,5 @@
 AMD64_ASM = \
   cpuid_amd64.asm \
-  rfxcodec_encode_diff_rlgr1_amd64_sse2.asm \
-  rfxcodec_encode_diff_rlgr3_amd64_sse2.asm \
   rfxcodec_encode_dwt_shift_amd64_sse2.asm \
   rfxcodec_encode_dwt_shift_amd64_sse41.asm
 
@@ -14,7 +12,7 @@ noinst_LTLIBRARIES = librfxencode-amd64.la
 librfxencode_amd64_la_SOURCES = \
   funcs_amd64.h \
   rfxencode_tile_amd64.c \
- $(AMD64_ASM)
+  $(AMD64_ASM)
 
 nasm_verbose = $(nasm_verbose_@AM_V@)
 nasm_verbose_ = $(nasm_verbose_@AM_DEFAULT_V@)
@@ -22,4 +20,5 @@ nasm_verbose_0 = @echo "  NASM     $@";
 
 .asm.lo:
 	$(nasm_verbose)$(LIBTOOL) $(AM_V_lt) --mode=compile \
-	  $(top_srcdir)/src/nasm_lt.sh $(NASM) $(NAFLAGS) -I$(srcdir) -I. $< -o $@
+	  $(top_srcdir)/src/nasm_lt.sh $(NASM) $(NAFLAGS) \
+	  -I$(top_srcdir)/src $< -o $@
diff --git a/src/amd64/cpuid_amd64.asm b/src/amd64/cpuid_amd64.asm
index e561b2d..acc738e 100644
--- a/src/amd64/cpuid_amd64.asm
+++ b/src/amd64/cpuid_amd64.asm
@@ -1,14 +1,6 @@
-%ifidn __OUTPUT_FORMAT__,elf64
-SECTION .note.GNU-stack noalloc noexec nowrite progbits
-%endif
+%include "common.asm"
 
-SECTION .text
-
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
+section .text
 
 ;The first six integer or pointer arguments are passed in registers
 ;RDI, RSI, RDX, RCX, R8, and R9
@@ -16,11 +8,7 @@ SECTION .text
 ;int
 ;cpuid_amd64(int eax_in, int ecx_in, int *eax, int *ebx, int *ecx, int *edx)
 
-%ifidn __OUTPUT_FORMAT__,elf64
 PROC cpuid_amd64
-%else
-PROC _cpuid_amd64
-%endif
     ; save registers
     push rbx
 
diff --git a/src/amd64/funcs_amd64.h b/src/amd64/funcs_amd64.h
index 124f838..d93e40b 100644
--- a/src/amd64/funcs_amd64.h
+++ b/src/amd64/funcs_amd64.h
@@ -33,36 +33,14 @@ cpuid_amd64(int eax_in, int ecx_in, int *eax, int *ebx, int *ecx, int *edx);
 
 int
 rfxcodec_encode_dwt_shift_amd64_sse2(const char *qtable,
-                                     unsigned char *data,
+                                     const unsigned char *data,
                                      short *dwt_buffer1,
                                      short *dwt_buffer);
 int
 rfxcodec_encode_dwt_shift_amd64_sse41(const char *qtable,
-                                      unsigned char *data,
+                                      const unsigned char *data,
                                       short *dwt_buffer1,
                                       short *dwt_buffer);
-int
-rfxcodec_encode_diff_rlgr1_amd64_sse2(short *co,
-                                      void *dst, int dst_bytes);
-int
-rfxcodec_encode_diff_rlgr3_amd64_sse2(short *co,
-                                      void *dst, int dst_bytes);
-
-int
-rfxcodec_decode_rlgr1_diff_amd64_sse2(void *data, int data_bytes,
-                                      short *out_data);
-int
-rfxcodec_decode_rlgr3_diff_amd64_sse2(void *data, int data_bytes,
-                                      short *out_data);
-int
-rfxcodec_decode_shift_idwt_amd64_sse2(char *qtable, short *src, short *dst);
-int
-rfxcodec_decode_yuv2rgb_amd64_sse2(short *ydata, short *udata, short *vdata,
-                                   unsigned int *rgbdata, int stride);
-int
-rfxcodec_decode_yuva2argb_amd64_sse2(short *ydata, short *udata,
-                                     short *vdata, char *adata,
-                                     unsigned int *rgbdata, int stride);
 
 #ifdef __cplusplus
 }
diff --git a/src/amd64/rfxcodec_encode_diff_rlgr1_amd64_sse2.asm b/src/amd64/rfxcodec_encode_diff_rlgr1_amd64_sse2.asm
deleted file mode 100644
index b2de84f..0000000
--- a/src/amd64/rfxcodec_encode_diff_rlgr1_amd64_sse2.asm
+++ /dev/null
@@ -1,36 +0,0 @@
-%ifidn __OUTPUT_FORMAT__,elf64
-section .note.GNU-stack noalloc noexec nowrite progbits
-%endif
-
-section .data
-    const1 times 8 dw 1
-
-section .text
-
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
-;The first six integer or pointer arguments are passed in registers
-;RDI, RSI, RDX, RCX, R8, and R9
-
-;int
-;rfxcodec_encode_diff_rlgr1_amd64_sse2(short *co,
-;                                      void *dst, int dst_bytes);
-
-%ifidn __OUTPUT_FORMAT__,elf64
-PROC rfxcodec_encode_diff_rlgr1_amd64_sse2
-%else
-PROC _rfxcodec_encode_diff_rlgr1_amd64_sse2
-%endif
-    ; save registers
-    push rbx
-
-    mov rax, 0
-    ; restore registers
-    pop rbx
-    ret
-    align 16
-
diff --git a/src/amd64/rfxcodec_encode_diff_rlgr3_amd64_sse2.asm b/src/amd64/rfxcodec_encode_diff_rlgr3_amd64_sse2.asm
deleted file mode 100644
index f5712be..0000000
--- a/src/amd64/rfxcodec_encode_diff_rlgr3_amd64_sse2.asm
+++ /dev/null
@@ -1,31 +0,0 @@
-%ifidn __OUTPUT_FORMAT__,elf64
-section .note.GNU-stack noalloc noexec nowrite progbits
-%endif
-
-section .data
-    const1 times 8 dw 1
-
-section .text
-
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
-;int
-;rfxcodec_encode_diff_rlgr3_amd64_sse2(short *co,
-;                                      void *dst, int dst_bytes);
-
-%ifidn __OUTPUT_FORMAT__,elf64
-PROC rfxcodec_encode_diff_rlgr3_amd64_sse2
-%else
-PROC _rfxcodec_encode_diff_rlgr3_amd64_sse2
-%endif
-    ; save registers
-    push rbx
-    mov rax, 0
-    pop rbx
-    ret
-    align 16
-
diff --git a/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse2.asm b/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse2.asm
index ee97588..69ccf07 100644
--- a/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse2.asm
+++ b/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse2.asm
@@ -19,9 +19,7 @@
 ;
 ;amd64 asm dwt
 
-%ifidn __OUTPUT_FORMAT__,elf64
-section .note.GNU-stack noalloc noexec nowrite progbits
-%endif
+%include "common.asm"
 
 section .data
     align 16
@@ -47,12 +45,6 @@ section .data
 
 section .text
 
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
 ;******************************************************************************
 ; source 16 bit signed, 16 pixel width
 rfx_dwt_2d_encode_block_horiz_16_16:
@@ -1346,11 +1338,7 @@ set_quants_lo:
 ;                                     short *work_buffer);
 
 ;******************************************************************************
-%ifidn __OUTPUT_FORMAT__,elf64
 PROC rfxcodec_encode_dwt_shift_amd64_sse2
-%else
-PROC _rfxcodec_encode_dwt_shift_amd64_sse2
-%endif
     ; save registers
     push rbx
     push rdx
diff --git a/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse41.asm b/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse41.asm
index ab52808..2b19f81 100644
--- a/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse41.asm
+++ b/src/amd64/rfxcodec_encode_dwt_shift_amd64_sse41.asm
@@ -19,9 +19,7 @@
 ;
 ;amd64 asm dwt
 
-%ifidn __OUTPUT_FORMAT__,elf64
-section .note.GNU-stack noalloc noexec nowrite progbits
-%endif
+%include "common.asm"
 
 section .data
     align 16
@@ -47,12 +45,6 @@ section .data
 
 section .text
 
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
 ;******************************************************************************
 ; source 16 bit signed, 16 pixel width
 rfx_dwt_2d_encode_block_horiz_16_16:
@@ -1214,11 +1206,7 @@ set_quants_lo:
 ;                                      short *work_buffer);
 
 ;******************************************************************************
-%ifidn __OUTPUT_FORMAT__,elf64
 PROC rfxcodec_encode_dwt_shift_amd64_sse41
-%else
-PROC _rfxcodec_encode_dwt_shift_amd64_sse41
-%endif
     ; save registers
     push rbx
     push rdx
diff --git a/src/amd64/rfxencode_tile_amd64.c b/src/amd64/rfxencode_tile_amd64.c
index 431b3d2..1619752 100644
--- a/src/amd64/rfxencode_tile_amd64.c
+++ b/src/amd64/rfxencode_tile_amd64.c
@@ -18,6 +18,10 @@
  * limitations under the License.
  */
 
+#if defined(HAVE_CONFIG_H)
+#include <config_ac.h>
+#endif
+
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -28,6 +32,8 @@
 #include "rfxencode_rlgr1.h"
 #include "rfxencode_rlgr3.h"
 #include "rfxencode_alpha.h"
+#include "rfxencode_diff_rlgr1.h"
+#include "rfxencode_diff_rlgr3.h"
 
 #include "amd64/funcs_amd64.h"
 
@@ -38,7 +44,7 @@
 /******************************************************************************/
 int
 rfx_encode_component_rlgr1_amd64_sse2(struct rfxencode *enc, const char *qtable,
-                                      uint8 *data,
+                                      const uint8 *data,
                                       uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr1_amd64_sse2:"));
@@ -47,20 +53,14 @@ rfx_encode_component_rlgr1_amd64_sse2(struct rfxencode *enc, const char *qtable,
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr1_amd64_sse2(enc->dwt_buffer1,
-    //                                              buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr1_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr1(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
 
 /******************************************************************************/
 int
 rfx_encode_component_rlgr3_amd64_sse2(struct rfxencode *enc, const char *qtable,
-                                      uint8 *data,
+                                      const uint8 *data,
                                       uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr3_amd64_sse2:"));
@@ -69,56 +69,38 @@ rfx_encode_component_rlgr3_amd64_sse2(struct rfxencode *enc, const char *qtable,
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr3_amd64_sse2(enc->dwt_buffer1,
-    //                                              buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr3_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr3(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
 
 /******************************************************************************/
 int
 rfx_encode_component_rlgr1_amd64_sse41(struct rfxencode *enc, const char *qtable,
-                                       uint8 *data,
+                                       const uint8 *data,
                                        uint8 *buffer, int buffer_size, int *size)
 {
-    LLOGLN(10, ("rfx_encode_component_rlgr1_amd64_sse2:"));
+    LLOGLN(10, ("rfx_encode_component_rlgr1_amd64_sse41:"));
     if (rfxcodec_encode_dwt_shift_amd64_sse41(qtable, data, enc->dwt_buffer1,
                                               enc->dwt_buffer) != 0)
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr1_amd64_sse2(enc->dwt_buffer1,
-    //                                              buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr1_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr1(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
 
 /******************************************************************************/
 int
 rfx_encode_component_rlgr3_amd64_sse41(struct rfxencode *enc, const char *qtable,
-                                       uint8 *data,
+                                       const uint8 *data,
                                        uint8 *buffer, int buffer_size, int *size)
 {
-    LLOGLN(10, ("rfx_encode_component_rlgr3_amd64_sse2:"));
+    LLOGLN(10, ("rfx_encode_component_rlgr3_amd64_sse41:"));
     if (rfxcodec_encode_dwt_shift_amd64_sse41(qtable, data, enc->dwt_buffer1,
                                               enc->dwt_buffer) != 0)
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr3_amd64_sse2(enc->dwt_buffer1,
-    //                                              buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr3_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr3(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
diff --git a/src/common.asm b/src/common.asm
new file mode 100644
index 0000000..7e2b84b
--- /dev/null
+++ b/src/common.asm
@@ -0,0 +1,51 @@
+;
+;Copyright 2017 Pavel Roskin
+;
+;Permission to use, copy, modify, distribute, and sell this software and its
+;documentation for any purpose is hereby granted without fee, provided that
+;the above copyright notice appear in all copies and that both that
+;copyright notice and this permission notice appear in supporting
+;documentation.
+;
+;The above copyright notice and this permission notice shall be included in
+;all copies or substantial portions of the Software.
+;
+;THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+;IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+;FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
+;OPEN GROUP BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+;AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+;CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+;
+;Common nasm code
+;
+
+; Detect ELF formats
+%ifidn __OUTPUT_FORMAT__,elf
+%define is_elf 1
+%endif
+
+%ifidn __OUTPUT_FORMAT__,elf32
+%define is_elf 1
+%endif
+
+%ifidn __OUTPUT_FORMAT__,elf64
+%define is_elf 1
+%endif
+
+; Mark stack non-executable
+%ifdef is_elf
+section .note.GNU-stack noalloc noexec nowrite progbits
+%endif
+
+; Global function header
+%macro PROC 1
+    align 16
+%ifdef is_elf
+    global %1:function
+    %1:
+%else
+    global _%1
+    _%1:
+%endif
+%endmacro
diff --git a/src/rfxcommon.h b/src/rfxcommon.h
index c7970ad..dad376c 100644
--- a/src/rfxcommon.h
+++ b/src/rfxcommon.h
@@ -1,7 +1,7 @@
 /**
  * RFX codec
  *
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -96,4 +96,35 @@ typedef struct _STREAM STREAM;
 
 #define xnew(_type) (_type *) calloc(1, sizeof(_type))
 
+/*
+  GCC has __builtin_clz that translates to BSR on x86/x64, CLZ on ARM, etc.
+  and emulates the instruction if the hardware does not implement it.
+  Visual C++ 2005 and up has _BitScanReverse
+
+  LZCNT = BSR ^ 31
+
+*/
+#if defined(__GNUC__)
+#define GBSR(_in, _r) do { \
+    _r = __builtin_clz(_in) ^ 31; \
+} while (0)
+#elif defined(_MSC_VER) && (_MSC_VER > 1000)
+#define GBSR(_in, _r) do { \
+    unsigned long rv = 0; \
+    _BitScanReverse(&rv, _in); \
+    _r = rv; \
+} while (0)
+#else
+#define GBSR(_in, _r) do { \
+    int rv = -1; \
+    int x = _in; \
+    while (x != 0) \
+    { \
+        rv++; \
+        x = x >> 1; \
+    } \
+    _r = rv; \
+} while (0)
+#endif
+
 #endif
diff --git a/src/rfxcompose.c b/src/rfxcompose.c
index 81b36a2..3c0f87a 100644
--- a/src/rfxcompose.c
+++ b/src/rfxcompose.c
@@ -3,7 +3,7 @@
  * RemoteFX Codec Library
  *
  * Copyright 2011 Vic Lee
- * Copyright 2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2015-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -208,7 +208,8 @@ rfx_compose_message_region(struct rfxencode *enc, STREAM *s,
 /******************************************************************************/
 static int
 rfx_compose_message_tile_yuv(struct rfxencode *enc, STREAM *s,
-                             char *tile_data, int tile_width, int tile_height,
+                             const char *tile_data,
+                             int tile_width, int tile_height,
                              int stride_bytes, const char *quantVals,
                              int quantIdxY, int quantIdxCb, int quantIdxCr,
                              int xIdx, int yIdx)
@@ -251,7 +252,8 @@ rfx_compose_message_tile_yuv(struct rfxencode *enc, STREAM *s,
 /******************************************************************************/
 static int
 rfx_compose_message_tile_yuva(struct rfxencode *enc, STREAM *s,
-                              char *tile_data, int tile_width, int tile_height,
+                              const char *tile_data,
+                              int tile_width, int tile_height,
                               int stride_bytes, const char *quantVals,
                               int quantIdxY, int quantIdxCb, int quantIdxCr,
                               int xIdx, int yIdx)
@@ -296,7 +298,8 @@ rfx_compose_message_tile_yuva(struct rfxencode *enc, STREAM *s,
 /******************************************************************************/
 static int
 rfx_compose_message_tile_rgb(struct rfxencode *enc, STREAM *s,
-                             char *tile_data, int tile_width, int tile_height,
+                             const char *tile_data,
+                             int tile_width, int tile_height,
                              int stride_bytes, const char *quantVals,
                              int quantIdxY, int quantIdxCb, int quantIdxCr,
                              int xIdx, int yIdx)
@@ -339,7 +342,8 @@ rfx_compose_message_tile_rgb(struct rfxencode *enc, STREAM *s,
 /******************************************************************************/
 static int
 rfx_compose_message_tile_argb(struct rfxencode *enc, STREAM *s,
-                              char *tile_data, int tile_width, int tile_height,
+                              const char *tile_data,
+                              int tile_width, int tile_height,
                               int stride_bytes, const char *quantVals,
                               int quantIdxY, int quantIdxCb, int quantIdxCr,
                               int xIdx, int yIdx)
@@ -386,7 +390,7 @@ rfx_compose_message_tile_argb(struct rfxencode *enc, STREAM *s,
 /******************************************************************************/
 static int
 rfx_compose_message_tileset(struct rfxencode *enc, STREAM *s,
-                            char *buf, int width, int height,
+                            const char *buf, int width, int height,
                             int stride_bytes,
                             const struct rfx_tile *tiles, int num_tiles,
                             const char *quants, int num_quants,
@@ -407,7 +411,7 @@ rfx_compose_message_tileset(struct rfxencode *enc, STREAM *s,
     int y;
     int cx;
     int cy;
-    char *tile_data;
+    const char *tile_data;
 
     LLOGLN(10, ("rfx_compose_message_tileset:"));
     if (quants == 0)
@@ -569,7 +573,8 @@ rfx_compose_message_frame_end(struct rfxencode *enc, STREAM *s)
 int
 rfx_compose_message_data(struct rfxencode *enc, STREAM *s,
                          const struct rfx_rect *regions, int num_regions,
-                         char *buf, int width, int height, int stride_bytes,
+                         const char *buf, int width, int height,
+                         int stride_bytes,
                          const struct rfx_tile *tiles, int num_tiles,
                          const char *quants, int num_quants, int flags)
 {
diff --git a/src/rfxcompose.h b/src/rfxcompose.h
index 7a3ff3a..6fde0f3 100644
--- a/src/rfxcompose.h
+++ b/src/rfxcompose.h
@@ -1,7 +1,7 @@
 /**
  * RFX codec encoder
  *
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -26,7 +26,8 @@ rfx_compose_message_header(struct rfxencode *enc, STREAM *s);
 int
 rfx_compose_message_data(struct rfxencode *enc, STREAM *s,
                          const struct rfx_rect *regions, int num_regions,
-                         char *buf, int width, int height, int stride_bytes,
+                         const char *buf, int width, int height,
+                         int stride_bytes,
                          const struct rfx_tile *tiles, int num_tiles,
                          const char *quants, int num_quants, int flags);
 
diff --git a/src/rfxencode.c b/src/rfxencode.c
index e4b67aa..315c6e9 100644
--- a/src/rfxencode.c
+++ b/src/rfxencode.c
@@ -1,7 +1,7 @@
 /**
  * RFX codec encoder
  *
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -298,7 +298,7 @@ rfxcodec_encode_destroy(void *handle)
 /******************************************************************************/
 int
 rfxcodec_encode_ex(void *handle, char *cdata, int *cdata_bytes,
-                   char *buf, int width, int height, int stride_bytes,
+                   const char *buf, int width, int height, int stride_bytes,
                    const struct rfx_rect *regions, int num_regions,
                    const struct rfx_tile *tiles, int num_tiles,
                    const char *quants, int num_quants, int flags)
@@ -334,7 +334,7 @@ rfxcodec_encode_ex(void *handle, char *cdata, int *cdata_bytes,
 /******************************************************************************/
 int
 rfxcodec_encode(void *handle, char *cdata, int *cdata_bytes,
-                char *buf, int width, int height, int stride_bytes,
+                const char *buf, int width, int height, int stride_bytes,
                 const struct rfx_rect *regions, int num_regions,
                 const struct rfx_tile *tiles, int num_tiles,
                 const char *quants, int num_quants)
diff --git a/src/rfxencode.h b/src/rfxencode.h
index 827acec..55147de 100644
--- a/src/rfxencode.h
+++ b/src/rfxencode.h
@@ -22,8 +22,8 @@
 struct rfxencode;
 
 typedef int (*rfx_encode_proc)(struct rfxencode *enc, const char *qtable,
-                               uint8 *data, uint8 *buffer,
-                               int buffer_size, int *size);
+                               const uint8 *data,
+                               uint8 *buffer, int buffer_size, int *size);
 
 struct rfxencode
 {
diff --git a/src/rfxencode_alpha.c b/src/rfxencode_alpha.c
index 637788b..2b04923 100644
--- a/src/rfxencode_alpha.c
+++ b/src/rfxencode_alpha.c
@@ -2,7 +2,7 @@
  * librfxcodec: A Remote Desktop Protocol client.
  * RemoteFX Codec Library
  *
- * Copyright 2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2015-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -39,10 +39,10 @@
 #if 1
 /*****************************************************************************/
 static int
-fdelta(char *in_plane, char *out_plane, int cx, int cy)
+fdelta(const char *in_plane, char *out_plane, int cx, int cy)
 {
     char delta;
-    char *src8;
+    const char *src8;
     char *dst8;
     int index;
     int jndex;
@@ -251,15 +251,15 @@ fpack(char *plane, int cx, int cy, STREAM *s)
 
 /*****************************************************************************/
 int
-rfx_encode_plane(struct rfxencode *enc, uint8 *plane, int cx, int cy,
+rfx_encode_plane(struct rfxencode *enc, const uint8 *plane, int cx, int cy,
                  STREAM *s)
 {
-    char *org_plane;
+    const char *org_plane;
     char *delta_plane;
     int bytes;
     uint8 *holdp;
 
-    org_plane = (char *) plane;
+    org_plane = (const char *) plane;
     delta_plane = (char *) (enc->dwt_buffer1);
     fdelta(org_plane, delta_plane, cx, cy);
     holdp = s->p;
diff --git a/src/rfxencode_alpha.h b/src/rfxencode_alpha.h
index 3f01218..c5d1d1d 100644
--- a/src/rfxencode_alpha.h
+++ b/src/rfxencode_alpha.h
@@ -21,7 +21,7 @@
 #define __RFXCODEC_ENCODE_ALPHA_H
 
 int
-rfx_encode_plane(struct rfxencode *enc, uint8 *plane, int cx, int cy,
+rfx_encode_plane(struct rfxencode *enc, const uint8 *plane, int cx, int cy,
                  STREAM *s);
 
 #endif
diff --git a/src/rfxencode_diff_rlgr1.c b/src/rfxencode_diff_rlgr1.c
new file mode 100644
index 0000000..cd09f18
--- /dev/null
+++ b/src/rfxencode_diff_rlgr1.c
@@ -0,0 +1,252 @@
+/**
+ * FreeRDP: A Remote Desktop Protocol client.
+ * RemoteFX Codec Library - RLGR
+ *
+ * Copyright 2011 Vic Lee
+ * Copyright 2016-2017 Jay Sorg <jay.sorg@gmail.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+ * This implementation of RLGR refers to
+ * [MS-RDPRFX] 3.1.8.1.7.3 RLGR1/RLGR3 Pseudocode
+ */
+
+#if defined(HAVE_CONFIG_H)
+#include <config_ac.h>
+#endif
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "rfxcommon.h"
+
+#define PIXELS_IN_TILE 4096
+
+/* Constants used within the RLGR1/RLGR3 algorithm */
+#define KPMAX   (80)  /* max value for kp or krp */
+#define LSGR    (3)   /* shift count to convert kp to k */
+#define UP_GR   (4)   /* increase in kp after a zero run in RL mode */
+#define DN_GR   (6)   /* decrease in kp after a nonzero symbol in RL mode */
+#define UQ_GR   (3)   /* increase in kp after nonzero symbol in GR mode */
+#define DQ_GR   (3)   /* decrease in kp after zero symbol in GR mode */
+
+#define GetNextInput do { \
+    input = *coef; \
+    coef++; \
+    coef_size--; \
+} while (0)
+
+#define CheckWrite do { \
+    while (bit_count >= 8) \
+    { \
+        bit_count -= 8; \
+        *cdata = bits >> bit_count; \
+        cdata++; \
+    } \
+} while (0)
+
+/* output GR code for (mag - 1) */
+#define CodeGR(_krp, _lmag) do { \
+    int lkr = _krp >> LSGR; \
+    /* unary part of GR code */ \
+    int lvk = _lmag >> lkr; \
+    int llvk = lvk; \
+    while (llvk >= 8) \
+    { \
+        bits <<= 8; \
+        bits |= 0xFF; \
+        llvk -= 8; \
+        *cdata = bits >> bit_count; \
+        cdata++; \
+    } \
+    bits <<= llvk; \
+    bits |= (1 << llvk) - 1; \
+    bit_count += llvk; \
+    bits <<= 1; \
+    bit_count++; \
+    CheckWrite; \
+    /* remainder part of GR code, if needed */ \
+    if (lkr) \
+    { \
+        bits <<= lkr; \
+        bits |= _lmag & ((1 << lkr) - 1); \
+        bit_count += lkr; \
+    } \
+    /* update _krp, only if it is not equal to 1 */ \
+    if (lvk == 0) \
+    { \
+        _krp = MAX(0, _krp - 2); \
+    } \
+    else if (lvk > 1) \
+    { \
+        _krp = MIN(KPMAX, _krp + lvk); \
+    } \
+} while (0)
+
+int
+rfx_encode_diff_rlgr1(sint16 *coef, uint8 *cdata, int cdata_size)
+{
+    int k;
+    int kp;
+    int krp;
+
+    int input;
+    int numZeros;
+    int runmax;
+    int mag;
+    int sign;
+    int processed_size;
+    int lmag;
+    int coef_size;
+    int y;
+
+    int bit_count;
+    unsigned int bits;
+    uint8 *cdata_org;
+
+    uint32 twoMs;
+
+    /* the last 64 bytes are diff */
+    for (k = PIXELS_IN_TILE - 1; k > PIXELS_IN_TILE - 64; k--)
+    {
+        coef[k] -= coef[k - 1];
+    }
+
+    /* initialize the parameters */
+    k = 1;
+    kp = 1 << LSGR;
+    krp = 1 << LSGR;
+
+    bit_count = 0;
+    bits = 0;
+    cdata_org = cdata;
+
+    /* process all the input coefficients */
+    coef_size = PIXELS_IN_TILE;
+    while (coef_size > 0)
+    {
+        if (k)
+        {
+
+            /* RUN-LENGTH MODE */
+
+            /* collect the run of zeros in the input stream */
+            numZeros = 0;
+
+            GetNextInput;
+            while (input == 0 && coef_size > 0)
+            {
+                numZeros++;
+                GetNextInput;
+            }
+
+            /* emit output zeros */
+            runmax = 1 << k;
+            while (numZeros >= runmax)
+            {
+
+                bits <<= 1;
+                bit_count++;
+
+                CheckWrite;
+
+                numZeros -= runmax;
+
+                kp = MIN(KPMAX, kp + UP_GR);
+                k = kp >> LSGR;
+
+                runmax = 1 << k;
+            }
+
+            /* output a 1 to terminate runs */
+            bits <<= 1;
+            bits |= 1;
+            bit_count++;
+
+            /* output the remaining run length using k bits */
+            bits <<= k;
+            bits |= numZeros;
+            bit_count += k;
+
+            CheckWrite;
+
+            /* encode the nonzero value using GR coding */
+            if (input < 0)
+            {
+                mag = -input;
+                sign = 1;
+            }
+            else
+            {
+                mag = input;
+                sign = 0;
+            }
+
+            bits <<= 1;
+            bits |= sign;
+            bit_count++;
+
+            lmag = mag ? mag - 1 : 0;
+
+            CodeGR(krp, lmag); /* output GR code for (mag - 1) */
+            CheckWrite;
+
+            kp = MAX(0, kp - DN_GR);
+            k = kp >> LSGR;
+
+        }
+        else
+        {
+
+            /* GOLOMB-RICE MODE */
+
+            /* RLGR1 variant */
+
+            /* convert input to (2*magnitude - sign), encode using GR code */
+            GetNextInput;
+            y = input >> 15;
+            twoMs = (((input ^ y) - y) << 1) + y;
+            CodeGR(krp, twoMs);
+            CheckWrite;
+
+            /* update k, kp */
+            if (twoMs)
+            {
+                kp = MAX(0, kp - DQ_GR);
+                k = kp >> LSGR;
+            }
+            else
+            {
+                kp = MIN(KPMAX, kp + UQ_GR);
+                k = kp >> LSGR;
+            }
+
+        }
+    }
+
+    if (bit_count > 0)
+    {
+        bits <<= 8 - bit_count;
+        *cdata = bits;
+        cdata++;
+        bit_count = 0;
+    }
+
+    processed_size = cdata - cdata_org;
+
+    return processed_size;
+}
+
diff --git a/src/rfxencode_diff_rlgr1.h b/src/rfxencode_diff_rlgr1.h
new file mode 100644
index 0000000..796f8f2
--- /dev/null
+++ b/src/rfxencode_diff_rlgr1.h
@@ -0,0 +1,29 @@
+/**
+ * FreeRDP: A Remote Desktop Protocol client.
+ * RemoteFX Codec Library - RLGR
+ *
+ * Copyright 2011 Vic Lee
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef __RFX_DIFF_RLGR1_H
+#define __RFX_DIFF_RLGR1_H
+
+#include "rfxcommon.h"
+
+int
+rfx_encode_diff_rlgr1(sint16 *coef, uint8 *cdata, int cdata_size);
+
+#endif /* __RFX_DIFF_RLGR1_H */
+
diff --git a/src/rfxencode_diff_rlgr3.c b/src/rfxencode_diff_rlgr3.c
new file mode 100644
index 0000000..0b68db3
--- /dev/null
+++ b/src/rfxencode_diff_rlgr3.c
@@ -0,0 +1,288 @@
+/**
+ * FreeRDP: A Remote Desktop Protocol client.
+ * RemoteFX Codec Library - RLGR
+ *
+ * Copyright 2011 Vic Lee
+ * Copyright 2016-2017 Jay Sorg <jay.sorg@gmail.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+ * This implementation of RLGR refers to
+ * [MS-RDPRFX] 3.1.8.1.7.3 RLGR1/RLGR3 Pseudocode
+ */
+
+#if defined(HAVE_CONFIG_H)
+#include <config_ac.h>
+#endif
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "rfxcommon.h"
+
+#define PIXELS_IN_TILE 4096
+
+/* Constants used within the RLGR1/RLGR3 algorithm */
+#define KPMAX   (80)  /* max value for kp or krp */
+#define LSGR    (3)   /* shift count to convert kp to k */
+#define UP_GR   (4)   /* increase in kp after a zero run in RL mode */
+#define DN_GR   (6)   /* decrease in kp after a nonzero symbol in RL mode */
+#define UQ_GR   (3)   /* increase in kp after nonzero symbol in GR mode */
+#define DQ_GR   (3)   /* decrease in kp after zero symbol in GR mode */
+
+#define GetNextInput do { \
+    input = *coef; \
+    coef++; \
+    coef_size--; \
+} while (0)
+
+#define CheckWrite do { \
+    while (bit_count >= 8) \
+    { \
+        bit_count -= 8; \
+        *cdata = bits >> bit_count; \
+        cdata++; \
+    } \
+} while (0)
+
+/* output GR code for (mag - 1) */
+#define CodeGR(_krp, _lmag) do { \
+    int lkr = _krp >> LSGR; \
+    /* unary part of GR code */ \
+    int lvk = _lmag >> lkr; \
+    int llvk = lvk; \
+    while (llvk >= 8) \
+    { \
+        bits <<= 8; \
+        bits |= 0xFF; \
+        llvk -= 8; \
+        *cdata = bits >> bit_count; \
+        cdata++; \
+    } \
+    bits <<= llvk; \
+    bits |= (1 << llvk) - 1; \
+    bit_count += llvk; \
+    bits <<= 1; \
+    bit_count++; \
+    CheckWrite; \
+    /* remainder part of GR code, if needed */ \
+    if (lkr) \
+    { \
+        bits <<= lkr; \
+        bits |= _lmag & ((1 << lkr) - 1); \
+        bit_count += lkr; \
+    } \
+    /* update _krp, only if it is not equal to 1 */ \
+    if (lvk == 0) \
+    { \
+        _krp = MAX(0, _krp - 2); \
+    } \
+    else if (lvk > 1) \
+    { \
+        _krp = MIN(KPMAX, _krp + lvk); \
+    } \
+} while (0)
+
+int
+rfx_encode_diff_rlgr3(sint16 *coef, uint8 *cdata, int cdata_size)
+{
+    int k;
+    int kp;
+    int krp;
+
+    int input;
+    int numZeros;
+    int runmax;
+    int mag;
+    int sign;
+    int processed_size;
+    int lmag;
+    int coef_size;
+    int y;
+
+    int bit_count;
+    unsigned int bits;
+    uint8 *cdata_org;
+
+    uint32 twoMs1;
+    uint32 twoMs2;
+    uint32 sum2Ms;
+    uint32 nIdx;
+
+    /* the last 64 bytes are diff */
+    for (k = PIXELS_IN_TILE - 1; k > PIXELS_IN_TILE - 64; k--)
+    {
+        coef[k] -= coef[k - 1];
+    }
+
+    /* initialize the parameters */
+    k = 1;
+    kp = 1 << LSGR;
+    krp = 1 << LSGR;
+
+    bit_count = 0;
+    bits = 0;
+    cdata_org = cdata;
+
+    /* process all the input coefficients */
+    coef_size = PIXELS_IN_TILE;
+    while (coef_size > 0)
+    {
+        if (k)
+        {
+
+            /* RUN-LENGTH MODE */
+
+            /* collect the run of zeros in the input stream */
+            numZeros = 0;
+
+            GetNextInput;
+            while (input == 0 && coef_size > 0)
+            {
+                numZeros++;
+                GetNextInput;
+            }
+
+            /* emit output zeros */
+            runmax = 1 << k;
+            while (numZeros >= runmax)
+            {
+
+                bits <<= 1;
+                bit_count++;
+
+                CheckWrite;
+
+                numZeros -= runmax;
+
+                kp = MIN(KPMAX, kp + UP_GR);
+                k = kp >> LSGR;
+
+                runmax = 1 << k;
+            }
+
+            /* output a 1 to terminate runs */
+            bits <<= 1;
+            bits |= 1;
+            bit_count++;
+
+            /* output the remaining run length using k bits */
+            bits <<= k;
+            bits |= numZeros;
+            bit_count += k;
+
+            CheckWrite;
+
+            /* encode the nonzero value using GR coding */
+            if (input < 0)
+            {
+                mag = -input;
+                sign = 1;
+            }
+            else
+            {
+                mag = input;
+                sign = 0;
+            }
+
+            bits <<= 1;
+            bits |= sign;
+            bit_count++;
+
+            lmag = mag ? mag - 1 : 0;
+
+            CodeGR(krp, lmag); /* output GR code for (mag - 1) */
+            CheckWrite;
+
+            kp = MAX(0, kp - DN_GR);
+            k = kp >> LSGR;
+
+        }
+        else
+        {
+
+            /* GOLOMB-RICE MODE */
+
+            /* RLGR3 variant */
+
+            /* convert the next two input values to (2*magnitude - sign) and */
+            /* encode their sum using GR code */
+
+            GetNextInput;
+
+            y = input >> 15;
+            twoMs1 = (((input ^ y) - y) << 1) + y;
+
+            GetNextInput;
+
+            y = input >> 15;
+            twoMs2 = (((input ^ y) - y) << 1) + y;
+
+            sum2Ms = twoMs1 + twoMs2;
+
+            CodeGR(krp, sum2Ms);
+
+            CheckWrite;
+
+            /* encode binary representation of the first input (twoMs1). */
+            if (sum2Ms != 0)
+            {
+                GBSR(sum2Ms, nIdx);
+                nIdx++;
+            }
+            else
+            {
+                nIdx = 0;
+            }
+
+            bits <<= nIdx;
+            bits |= twoMs1;
+            bit_count += nIdx;
+
+            CheckWrite;
+
+            /* update k,kp for the two input values */
+            if (twoMs1 != 0)
+            {
+                if (twoMs2 != 0)
+                {
+                    kp = MAX(0, kp - (2 * DQ_GR));
+                    k = kp >> LSGR;
+                }
+            }
+            else
+            {
+                if (twoMs2 == 0)
+                {
+                    kp = MIN(KPMAX, kp + (2 * UQ_GR));
+                    k = kp >> LSGR;
+                }
+            }
+        }
+    }
+
+    if (bit_count > 0)
+    {
+        bits <<= 8 - bit_count;
+        *cdata = bits;
+        cdata++;
+        bit_count = 0;
+    }
+
+    processed_size = cdata - cdata_org;
+
+    return processed_size;
+}
diff --git a/src/rfxencode_diff_rlgr3.h b/src/rfxencode_diff_rlgr3.h
new file mode 100644
index 0000000..dd61309
--- /dev/null
+++ b/src/rfxencode_diff_rlgr3.h
@@ -0,0 +1,29 @@
+/**
+ * FreeRDP: A Remote Desktop Protocol client.
+ * RemoteFX Codec Library - RLGR
+ *
+ * Copyright 2011 Vic Lee
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef __RFX_DIFF_RLGR3_H
+#define __RFX_DIFF_RLGR3_H
+
+#include "rfxcommon.h"
+
+int
+rfx_encode_diff_rlgr3(sint16 *coef, uint8 *cdata, int cdata_size);
+
+#endif /* __RFX_DIFF_RLGR3_H */
+
diff --git a/src/rfxencode_dwt.c b/src/rfxencode_dwt.c
index 3c649fa..5d82044 100644
--- a/src/rfxencode_dwt.c
+++ b/src/rfxencode_dwt.c
@@ -3,7 +3,7 @@
  * RemoteFX Codec Library - DWT
  *
  * Copyright 2011 Vic Lee
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -150,10 +150,10 @@ rfx_dwt_2d_encode_block(sint16 *buffer, sint16 *dwt, int subband_width)
 
 /******************************************************************************/
 static int
-rfx_dwt_2d_encode_block8(uint8 *in_buffer,
+rfx_dwt_2d_encode_block8(const uint8 *in_buffer,
                          sint16 *buffer, sint16 *dwt, int subband_width)
 {
-    uint8 *src;
+    const uint8 *src;
     sint16 *l, *h;
     sint16 s1, s2, s3;
     int total_width;
@@ -213,7 +213,7 @@ rfx_dwt_2d_encode_block8(uint8 *in_buffer,
 
 /******************************************************************************/
 int
-rfx_dwt_2d_encode(uint8 *in_buffer, sint16 *buffer, sint16 *dwt_buffer)
+rfx_dwt_2d_encode(const uint8 *in_buffer, sint16 *buffer, sint16 *dwt_buffer)
 {
     rfx_dwt_2d_encode_block8(in_buffer, buffer, dwt_buffer, 32);
     rfx_dwt_2d_encode_block(buffer + 3072, dwt_buffer, 16);
diff --git a/src/rfxencode_dwt.h b/src/rfxencode_dwt.h
index 36a62ed..2d91176 100644
--- a/src/rfxencode_dwt.h
+++ b/src/rfxencode_dwt.h
@@ -1,7 +1,7 @@
 /**
  * RFX codec encoder
  *
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -20,6 +20,6 @@
 #define __RFXENCODE_RFX_H
 
 int
-rfx_dwt_2d_encode(uint8 *in_buffer, sint16 *buffer, sint16 *dwt_buffer);
+rfx_dwt_2d_encode(const uint8 *in_buffer, sint16 *buffer, sint16 *dwt_buffer);
 
 #endif
diff --git a/src/rfxencode_quantization.c b/src/rfxencode_quantization.c
index e36c6a8..b10a70e 100644
--- a/src/rfxencode_quantization.c
+++ b/src/rfxencode_quantization.c
@@ -3,7 +3,7 @@
  * RemoteFX Codec Library - Quantization
  *
  * Copyright 2011 Vic Lee
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
diff --git a/src/rfxencode_tile.c b/src/rfxencode_tile.c
index 5279169..3f87241 100644
--- a/src/rfxencode_tile.c
+++ b/src/rfxencode_tile.c
@@ -3,7 +3,7 @@
  * RemoteFX Codec Library - Encode
  *
  * Copyright 2011 Vic Lee
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -53,7 +53,7 @@
 
 /******************************************************************************/
 static int
-rfx_encode_format_rgb(char *rgb_data, int width, int height,
+rfx_encode_format_rgb(const char *rgb_data, int width, int height,
                       int stride_bytes, int pixel_format,
                       uint8 *r_buf, uint8 *g_buf, uint8 *b_buf)
 {
@@ -221,7 +221,7 @@ rfx_encode_format_rgb(char *rgb_data, int width, int height,
 
 /******************************************************************************/
 static int
-rfx_encode_format_argb(char *argb_data, int width, int height,
+rfx_encode_format_argb(const char *argb_data, int width, int height,
                        int stride_bytes, int pixel_format,
                        uint8 *a_buf, uint8 *r_buf, uint8 *g_buf, uint8 *b_buf)
 {
@@ -439,7 +439,8 @@ rfx_encode_rgb_to_yuv(uint8 *y_r_buf, uint8 *u_g_buf, uint8 *v_b_buf)
 /******************************************************************************/
 int
 rfx_encode_component_rlgr1(struct rfxencode *enc, const char *qtable,
-                           uint8 *data, uint8 *buffer, int buffer_size, int *size)
+                           const uint8 *data,
+                           uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr1:"));
     if (rfx_dwt_2d_encode(data, enc->dwt_buffer1, enc->dwt_buffer) != 0)
@@ -461,7 +462,8 @@ rfx_encode_component_rlgr1(struct rfxencode *enc, const char *qtable,
 /******************************************************************************/
 int
 rfx_encode_component_rlgr3(struct rfxencode *enc, const char *qtable,
-                           uint8 *data, uint8 *buffer, int buffer_size, int *size)
+                           const uint8 *data,
+                           uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr3:"));
     if (rfx_dwt_2d_encode(data, enc->dwt_buffer1, enc->dwt_buffer) != 0)
@@ -482,7 +484,7 @@ rfx_encode_component_rlgr3(struct rfxencode *enc, const char *qtable,
 
 /******************************************************************************/
 int
-rfx_encode_rgb(struct rfxencode *enc, char *rgb_data,
+rfx_encode_rgb(struct rfxencode *enc, const char *rgb_data,
                int width, int height, int stride_bytes,
                const char *y_quants, const char *u_quants,
                const char *v_quants,
@@ -537,7 +539,7 @@ rfx_encode_rgb(struct rfxencode *enc, char *rgb_data,
 
 /******************************************************************************/
 int
-rfx_encode_argb(struct rfxencode *enc, char *rgb_data,
+rfx_encode_argb(struct rfxencode *enc, const char *rgb_data,
                 int width, int height, int stride_bytes,
                 const char *y_quants, const char *u_quants,
                 const char *v_quants,
@@ -598,19 +600,19 @@ rfx_encode_argb(struct rfxencode *enc, char *rgb_data,
 
 /******************************************************************************/
 int
-rfx_encode_yuv(struct rfxencode *enc, char *yuv_data,
+rfx_encode_yuv(struct rfxencode *enc, const char *yuv_data,
                int width, int height, int stride_bytes,
                const char *y_quants, const char *u_quants,
                const char *v_quants,
                STREAM *data_out, int *y_size, int *u_size, int *v_size)
 {
-    uint8 *y_buffer;
-    uint8 *u_buffer;
-    uint8 *v_buffer;
+    const uint8 *y_buffer;
+    const uint8 *u_buffer;
+    const uint8 *v_buffer;
 
-    y_buffer = (uint8 *) yuv_data;
-    u_buffer = (uint8 *) (yuv_data + RFX_YUV_BTES);
-    v_buffer = (uint8 *) (yuv_data + RFX_YUV_BTES * 2);
+    y_buffer = (const uint8 *) yuv_data;
+    u_buffer = (const uint8 *) (yuv_data + RFX_YUV_BTES);
+    v_buffer = (const uint8 *) (yuv_data + RFX_YUV_BTES * 2);
     if (enc->rfx_encode(enc, y_quants, y_buffer,
                         stream_get_tail(data_out),
                         stream_get_left(data_out),
@@ -640,22 +642,22 @@ rfx_encode_yuv(struct rfxencode *enc, char *yuv_data,
 
 /******************************************************************************/
 int
-rfx_encode_yuva(struct rfxencode *enc, char *yuva_data,
+rfx_encode_yuva(struct rfxencode *enc, const char *yuva_data,
                 int width, int height, int stride_bytes,
                 const char *y_quants, const char *u_quants,
                 const char *v_quants,
                 STREAM *data_out, int *y_size, int *u_size,
                 int *v_size, int *a_size)
 {
-    uint8 *y_buffer;
-    uint8 *u_buffer;
-    uint8 *v_buffer;
-    uint8 *a_buffer;
+    const uint8 *y_buffer;
+    const uint8 *u_buffer;
+    const uint8 *v_buffer;
+    const uint8 *a_buffer;
 
-    y_buffer = (uint8 *) yuva_data;
-    u_buffer = (uint8 *) (yuva_data + RFX_YUV_BTES);
-    v_buffer = (uint8 *) (yuva_data + RFX_YUV_BTES * 2);
-    a_buffer = (uint8 *) (yuva_data + RFX_YUV_BTES * 3);
+    y_buffer = (const uint8 *) yuva_data;
+    u_buffer = (const uint8 *) (yuva_data + RFX_YUV_BTES);
+    v_buffer = (const uint8 *) (yuva_data + RFX_YUV_BTES * 2);
+    a_buffer = (const uint8 *) (yuva_data + RFX_YUV_BTES * 3);
     if (enc->rfx_encode(enc, y_quants, y_buffer,
                         stream_get_tail(data_out),
                         stream_get_left(data_out),
diff --git a/src/rfxencode_tile.h b/src/rfxencode_tile.h
index 6195d8d..74667a4 100644
--- a/src/rfxencode_tile.h
+++ b/src/rfxencode_tile.h
@@ -3,7 +3,7 @@
  * RemoteFX Codec Library - Encode
  *
  * Copyright 2011 Vic Lee
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -28,33 +28,33 @@
 
 int
 rfx_encode_component_rlgr1(struct rfxencode *enc, const char *qtable,
-                           uint8 *data,
+                           const uint8 *data,
                            uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr3(struct rfxencode *enc, const char *qtable,
-                           uint8 *data,
+                           const uint8 *data,
                            uint8 *buffer, int buffer_size, int *size);
 int
-rfx_encode_rgb(struct rfxencode *enc, char *rgb_data,
+rfx_encode_rgb(struct rfxencode *enc, const char *rgb_data,
                int width, int height, int stride_bytes,
                const char *y_quants, const char *u_quants,
                const char *v_quants,
                STREAM *data_out, int *y_size, int *cb_size, int *cr_size);
 int
-rfx_encode_argb(struct rfxencode *enc, char *argb_data,
+rfx_encode_argb(struct rfxencode *enc, const char *argb_data,
                 int width, int height, int stride_bytes,
                 const char *y_quants, const char *cb_quants,
                 const char *cr_quants,
                 STREAM *data_out, int *y_size, int *u_size,
                 int *v_size, int *a_size);
 int
-rfx_encode_yuv(struct rfxencode *enc, char *yuv_data,
+rfx_encode_yuv(struct rfxencode *enc, const char *yuv_data,
                int width, int height, int stride_bytes,
                const char *y_quants, const char *u_quants,
                const char *v_quants,
                STREAM *data_out, int *y_size, int *u_size, int *v_size);
 int
-rfx_encode_yuva(struct rfxencode *enc, char *yuv_data,
+rfx_encode_yuva(struct rfxencode *enc, const char *yuv_data,
                 int width, int height, int stride_bytes,
                 const char *y_quants, const char *u_quants,
                 const char *v_quants,
@@ -63,35 +63,35 @@ rfx_encode_yuva(struct rfxencode *enc, char *yuv_data,
 
 int
 rfx_encode_component_rlgr1_x86_sse2(struct rfxencode *enc, const char *qtable,
-                                    uint8 *data,
+                                    const uint8 *data,
                                     uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr3_x86_sse2(struct rfxencode *enc, const char *qtable,
-                                    uint8 *data,
+                                    const uint8 *data,
                                     uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr1_x86_sse41(struct rfxencode *enc, const char *qtable,
-                                     uint8 *data,
+                                     const uint8 *data,
                                      uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr3_x86_sse41(struct rfxencode *enc, const char *qtable,
-                                     uint8 *data,
+                                     const uint8 *data,
                                      uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr1_amd64_sse2(struct rfxencode *enc, const char *qtable,
-                                      uint8 *data,
+                                      const uint8 *data,
                                       uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr3_amd64_sse2(struct rfxencode *enc, const char *qtable,
-                                      uint8 *data,
+                                      const uint8 *data,
                                       uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr1_amd64_sse41(struct rfxencode *enc, const char *qtable,
-                                       uint8 *data,
+                                       const uint8 *data,
                                        uint8 *buffer, int buffer_size, int *size);
 int
 rfx_encode_component_rlgr3_amd64_sse41(struct rfxencode *enc, const char *qtable,
-                                       uint8 *data,
+                                       const uint8 *data,
                                        uint8 *buffer, int buffer_size, int *size);
 
 #endif
diff --git a/src/x86/Makefile.am b/src/x86/Makefile.am
index 969fc0e..2d099e5 100644
--- a/src/x86/Makefile.am
+++ b/src/x86/Makefile.am
@@ -1,7 +1,5 @@
 X86_ASM = \
   cpuid_x86.asm \
-  rfxcodec_encode_diff_rlgr1_x86_sse2.asm \
-  rfxcodec_encode_diff_rlgr3_x86_sse2.asm \
   rfxcodec_encode_dwt_shift_x86_sse2.asm \
   rfxcodec_encode_dwt_shift_x86_sse41.asm
 
@@ -22,4 +20,5 @@ nasm_verbose_0 = @echo "  NASM     $@";
 
 .asm.lo:
 	$(nasm_verbose)$(LIBTOOL) $(AM_V_lt) --mode=compile \
-	  $(top_srcdir)/src/nasm_lt.sh $(NASM) $(NAFLAGS) -I$(srcdir) -I. $< -o $@
+	  $(top_srcdir)/src/nasm_lt.sh $(NASM) $(NAFLAGS) \
+	  -I$(top_srcdir)/src $< -o $@
diff --git a/src/x86/cpuid_x86.asm b/src/x86/cpuid_x86.asm
index fe19a90..4ddb8a2 100644
--- a/src/x86/cpuid_x86.asm
+++ b/src/x86/cpuid_x86.asm
@@ -1,23 +1,11 @@
-%ifidn __OUTPUT_FORMAT__,elf
-SECTION .note.GNU-stack noalloc noexec nowrite progbits
-%endif
+%include "common.asm"
 
-SECTION .text
-
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
+section .text
 
 ;int
 ;cpuid_x86(int eax_in, int ecx_in, int *eax, int *ebx, int *ecx, int *edx)
 
-%ifidn __OUTPUT_FORMAT__,elf
 PROC cpuid_x86
-%else
-PROC _cpuid_x86
-%endif
     ; save registers
     push ebx
     push ecx
diff --git a/src/x86/funcs_x86.h b/src/x86/funcs_x86.h
index 858bc5c..4f56f27 100644
--- a/src/x86/funcs_x86.h
+++ b/src/x86/funcs_x86.h
@@ -33,38 +33,15 @@ cpuid_x86(int eax_in, int ecx_in, int *eax, int *ebx, int *ecx, int *edx);
 
 int
 rfxcodec_encode_dwt_shift_x86_sse2(const char *qtable,
-                                   unsigned char *data,
+                                   const unsigned char *data,
                                    short *dwt_buffer1,
                                    short *dwt_buffer);
 int
 rfxcodec_encode_dwt_shift_x86_sse41(const char *qtable,
-                                    unsigned char *data,
+                                    const unsigned char *data,
                                     short *dwt_buffer1,
                                     short *dwt_buffer);
 
-int
-rfxcodec_encode_diff_rlgr1_x86_sse2(short *co,
-                                    void *dst, int dst_bytes);
-int
-rfxcodec_encode_diff_rlgr3_x86_sse2(short *co,
-                                    void *dst, int dst_bytes);
-
-int
-rfxcodec_decode_rlgr1_diff_x86_sse2(void *data, int data_bytes,
-                                    short *out_data);
-int
-rfxcodec_decode_rlgr3_diff_x86_sse2(void *data, int data_bytes,
-                                    short *out_data);
-int
-rfxcodec_decode_shift_idwt_x86_sse2(const char *qtable, short *src, short *dst);
-int
-rfxcodec_decode_yuv2rgb_x86_sse2(short *ydata, short *udata, short *vdata,
-                                 unsigned int *rgbdata, int stride);
-int
-rfxcodec_decode_yuva2argb_x86_sse2(short *ydata, short *udata,
-                                   short *vdata, char *adata,
-                                   unsigned int *rgbdata, int stride);
-
 #ifdef __cplusplus
 }
 #endif
diff --git a/src/x86/rfxcodec_encode_diff_rlgr1_x86_sse2.asm b/src/x86/rfxcodec_encode_diff_rlgr1_x86_sse2.asm
deleted file mode 100644
index 13d10e9..0000000
--- a/src/x86/rfxcodec_encode_diff_rlgr1_x86_sse2.asm
+++ /dev/null
@@ -1,35 +0,0 @@
-%ifidn __OUTPUT_FORMAT__,elf
-section .note.GNU-stack noalloc noexec nowrite progbits
-%endif
-
-section .data
-    const1 times 8 dw 1
-
-section .text
-
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
-;int
-;rfxcodec_encode_diff_rlgr1_x86_sse2(short *co,
-;                                    void *dst, int dst_bytes);
-
-%ifidn __OUTPUT_FORMAT__,elf
-PROC rfxcodec_encode_diff_rlgr1_x86_sse2
-%else
-PROC _rfxcodec_encode_diff_rlgr1_x86_sse2
-%endif
-    push ebx
-    push esi
-    push edi
-
-    mov eax, 0
-    pop edi
-    pop esi
-    pop ebx
-    ret
-    align 16
-
diff --git a/src/x86/rfxcodec_encode_diff_rlgr3_x86_sse2.asm b/src/x86/rfxcodec_encode_diff_rlgr3_x86_sse2.asm
deleted file mode 100644
index a8588f2..0000000
--- a/src/x86/rfxcodec_encode_diff_rlgr3_x86_sse2.asm
+++ /dev/null
@@ -1,35 +0,0 @@
-%ifidn __OUTPUT_FORMAT__,elf
-section .note.GNU-stack noalloc noexec nowrite progbits
-%endif
-
-section .data
-    const1 times 8 dw 1
-
-section .text
-
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
-;int
-;rfxcodec_encode_diff_rlgr3_x86_sse2(short *co,
-;                                    void *dst, int dst_bytes);
-
-%ifidn __OUTPUT_FORMAT__,elf
-PROC rfxcodec_encode_diff_rlgr3_x86_sse2
-%else
-PROC _rfxcodec_encode_diff_rlgr3_x86_sse2
-%endif
-    push ebx
-    push esi
-    push edi
-
-    mov eax, 0
-    pop edi
-    pop esi
-    pop ebx
-    ret
-    align 16
-
diff --git a/src/x86/rfxcodec_encode_dwt_shift_x86_sse2.asm b/src/x86/rfxcodec_encode_dwt_shift_x86_sse2.asm
index 0b7e122..fb8dd56 100644
--- a/src/x86/rfxcodec_encode_dwt_shift_x86_sse2.asm
+++ b/src/x86/rfxcodec_encode_dwt_shift_x86_sse2.asm
@@ -20,8 +20,9 @@
 ;
 ;x86 asm dwt
 
+%include "common.asm"
+
 %ifidn __OUTPUT_FORMAT__,elf
-section .note.GNU-stack noalloc noexec nowrite progbits
 %ifdef PIC
 section .text
 extern _GLOBAL_OFFSET_TABLE_
@@ -71,12 +72,6 @@ section .data
 
 section .text
 
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
 %define LHI_ADD  [esp + 1 * 16 + 4]
 %define LHI_SFT  [esp + 2 * 16 + 4]
 %define LLO_ADD  [esp + 3 * 16 + 4]
@@ -1381,11 +1376,7 @@ set_quants_lo:
 ;                                   short *work_buffer);
 
 ;******************************************************************************
-%ifidn __OUTPUT_FORMAT__,elf
 PROC rfxcodec_encode_dwt_shift_x86_sse2
-%else
-PROC _rfxcodec_encode_dwt_shift_x86_sse2
-%endif
     ; align stack
     mov eax, esp
     sub eax, 0x10
diff --git a/src/x86/rfxcodec_encode_dwt_shift_x86_sse41.asm b/src/x86/rfxcodec_encode_dwt_shift_x86_sse41.asm
index 4bb5d68..915e6ee 100644
--- a/src/x86/rfxcodec_encode_dwt_shift_x86_sse41.asm
+++ b/src/x86/rfxcodec_encode_dwt_shift_x86_sse41.asm
@@ -20,8 +20,9 @@
 ;
 ;x86 asm dwt
 
+%include "common.asm"
+
 %ifidn __OUTPUT_FORMAT__,elf
-section .note.GNU-stack noalloc noexec nowrite progbits
 %ifdef PIC
 section .text
 extern _GLOBAL_OFFSET_TABLE_
@@ -71,12 +72,6 @@ section .data
 
 section .text
 
-%macro PROC 1
-    align 16
-    global %1
-    %1:
-%endmacro
-
 %define LHI_ADD  [esp + 1 * 16 + 4]
 %define LHI_SFT  [esp + 2 * 16 + 4]
 %define LLO_ADD  [esp + 3 * 16 + 4]
@@ -1249,11 +1244,7 @@ set_quants_lo:
 ;                                    short *work_buffer);
 
 ;******************************************************************************
-%ifidn __OUTPUT_FORMAT__,elf
 PROC rfxcodec_encode_dwt_shift_x86_sse41
-%else
-PROC _rfxcodec_encode_dwt_shift_x86_sse41
-%endif
     ; align stack
     mov eax, esp
     sub eax, 0x10
diff --git a/src/x86/rfxencode_tile_x86.c b/src/x86/rfxencode_tile_x86.c
index 3418b69..a84305e 100644
--- a/src/x86/rfxencode_tile_x86.c
+++ b/src/x86/rfxencode_tile_x86.c
@@ -18,6 +18,10 @@
  * limitations under the License.
  */
 
+#if defined(HAVE_CONFIG_H)
+#include <config_ac.h>
+#endif
+
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -28,6 +32,8 @@
 #include "rfxencode_rlgr1.h"
 #include "rfxencode_rlgr3.h"
 #include "rfxencode_alpha.h"
+#include "rfxencode_diff_rlgr1.h"
+#include "rfxencode_diff_rlgr3.h"
 
 #include "x86/funcs_x86.h"
 
@@ -38,7 +44,7 @@
 /******************************************************************************/
 int
 rfx_encode_component_rlgr1_x86_sse2(struct rfxencode *enc, const char *qtable,
-                                    uint8 *data,
+                                    const uint8 *data,
                                     uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr1_x86_sse2:"));
@@ -47,20 +53,14 @@ rfx_encode_component_rlgr1_x86_sse2(struct rfxencode *enc, const char *qtable,
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr1_x86_sse2(enc->dwt_buffer1,
-    //                                            buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr1_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr1(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
 
 /******************************************************************************/
 int
 rfx_encode_component_rlgr3_x86_sse2(struct rfxencode *enc, const char *qtable,
-                                    uint8 *data,
+                                    const uint8 *data,
                                     uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr3_x86_sse2:"));
@@ -69,20 +69,14 @@ rfx_encode_component_rlgr3_x86_sse2(struct rfxencode *enc, const char *qtable,
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr3_x86_sse2(enc->dwt_buffer1,
-    //                                            buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr3_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr3(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
 
 /******************************************************************************/
 int
 rfx_encode_component_rlgr1_x86_sse41(struct rfxencode *enc, const char *qtable,
-                                     uint8 *data,
+                                     const uint8 *data,
                                      uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr1_x86_sse41:"));
@@ -91,20 +85,14 @@ rfx_encode_component_rlgr1_x86_sse41(struct rfxencode *enc, const char *qtable,
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr1_x86_sse2(enc->dwt_buffer1,
-    //                                            buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr1_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr1(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
 
 /******************************************************************************/
 int
 rfx_encode_component_rlgr3_x86_sse41(struct rfxencode *enc, const char *qtable,
-                                     uint8 *data,
+                                     const uint8 *data,
                                      uint8 *buffer, int buffer_size, int *size)
 {
     LLOGLN(10, ("rfx_encode_component_rlgr3_x86_sse41:"));
@@ -113,12 +101,6 @@ rfx_encode_component_rlgr3_x86_sse41(struct rfxencode *enc, const char *qtable,
     {
         return 1;
     }
-    //*size = rfxcodec_encode_diff_rlgr3_x86_sse(enc->dwt_buffer1,
-    //                                            buffer, buffer_size);
-    if (rfx_differential_encode(enc->dwt_buffer1 + 4032, 64) != 0)
-    {
-        return 1;
-    }
-    *size = rfx_rlgr3_encode(enc->dwt_buffer1, buffer, buffer_size);
+    *size = rfx_encode_diff_rlgr3(enc->dwt_buffer1, buffer, buffer_size);
     return 0;
 }
diff --git a/tests/rfxcodectest.c b/tests/rfxcodectest.c
index 4d20be2..3b22dc7 100644
--- a/tests/rfxcodectest.c
+++ b/tests/rfxcodectest.c
@@ -1,7 +1,7 @@
 /**
  * RFX codec encoder test
  *
- * Copyright 2014-2015 Jay Sorg <jay.sorg@gmail.com>
+ * Copyright 2014-2017 Jay Sorg <jay.sorg@gmail.com>
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
diff --git a/tests/rfxencode.c b/tests/rfxencode.c
index 66fb173..d8d1dfc 100644
--- a/tests/rfxencode.c
+++ b/tests/rfxencode.c
@@ -16,6 +16,10 @@
  * limitations under the License.
  */
 
+#if defined(HAVE_CONFIG_H)
+#include <config_ac.h>
+#endif
+
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>