diff --git a/build/make/configure.sh b/build/make/configure.sh
index 88698fd..7234b79 100755
--- a/build/make/configure.sh
+++ b/build/make/configure.sh
@@ -711,7 +711,7 @@
             TOOLCHAIN_PATH=${SDK_PATH}/usr/bin
             CC=${TOOLCHAIN_PATH}/gcc
             AR=${TOOLCHAIN_PATH}/ar
-            LD=${TOOLCHAIN_PATH}/arm-apple-darwin9-gcc-4.2.1
+            LD=${TOOLCHAIN_PATH}/arm-apple-darwin10-gcc-4.2.1
             AS=${TOOLCHAIN_PATH}/as
             STRIP=${TOOLCHAIN_PATH}/strip
             NM=${TOOLCHAIN_PATH}/nm
@@ -725,14 +725,14 @@
             add_cflags -arch ${tgt_isa}
             add_ldflags -arch_only ${tgt_isa}
 
-            add_cflags  "-isysroot /Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS3.1.sdk"
+            add_cflags  "-isysroot /Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS4.2.sdk"
 
             # This should be overridable
-            alt_libc=${SDK_PATH}/SDKs/iPhoneOS3.1.sdk
+            alt_libc=${SDK_PATH}/SDKs/iPhoneOS4.2.sdk
 
             # Add the paths for the alternate libc
 #            for d in usr/include usr/include/gcc/darwin/4.0/; do
-            for d in usr/include usr/include/gcc/darwin/4.0/ usr/lib/gcc/arm-apple-darwin9/4.0.1/include/; do
+            for d in usr/include usr/include/gcc/darwin/4.0/ usr/lib/gcc/arm-apple-darwin10/4.2.1/include/; do
                 try_dir="${alt_libc}/${d}"
                 [ -d "${try_dir}" ] && add_cflags -I"${try_dir}"
             done
diff --git a/build/make/obj_int_extract.c b/build/make/obj_int_extract.c
index e01870f..3c54b24 100644
--- a/build/make/obj_int_extract.c
+++ b/build/make/obj_int_extract.c
@@ -590,7 +590,7 @@
     //log_msg("COFF: Symbol table at offset %u\n", symtab_ptr);
     //log_msg("COFF: raw data pointer ofset for section .data is %u\n", sectionrawdata_ptr);
 
-    fp = fopen("vpx_asm_offsets.asm", "w");
+    fp = fopen("assembly_offsets.asm", "w");
 
     if (fp == NULL)
     {
diff --git a/libs.mk b/libs.mk
index 9ded394..338822c 100644
--- a/libs.mk
+++ b/libs.mk
@@ -230,10 +230,39 @@
 #
 # Add assembler dependencies for configuration and offsets
 #
-#$(filter %$(ASM).o,$(OBJS-yes)): $(BUILD_PFX)vpx_config.asm $(BUILD_PFX)vpx_asm_offsets.asm
 $(filter %.s.o,$(OBJS-yes)):   $(BUILD_PFX)vpx_config.asm
 $(filter %.asm.o,$(OBJS-yes)): $(BUILD_PFX)vpx_config.asm
 
+#
+# Calculate platform- and compiler-specific offsets for hand coded assembly
+#
+ifeq ($(ARCH_ARM), yes)
+  asm_com_offsets.asm: obj_int_extract
+  asm_com_offsets.asm: $(VP8_PREFIX)common/asm_com_offsets.c.o
+	./obj_int_extract rvds $< $(ADS2GAS) > $@
+  OBJS-yes += $(VP8_PREFIX)common/asm_com_offsets.c.o
+  CLEAN-OBJS += asm_com_offsets.asm
+  $(filter %$(ASM).o,$(OBJS-yes)): $(BUILD_PFX)asm_com_offsets.asm
+
+  ifeq ($(CONFIG_VP8_ENCODER), yes)
+    asm_enc_offsets.asm: obj_int_extract
+    asm_enc_offsets.asm: $(VP8_PREFIX)encoder/asm_enc_offsets.c.o
+	./obj_int_extract rvds $< $(ADS2GAS) > $@
+    OBJS-yes += $(VP8_PREFIX)encoder/asm_enc_offsets.c.o
+    CLEAN-OBJS += asm_enc_offsets.asm
+    $(filter %$(ASM).o,$(OBJS-yes)): $(BUILD_PFX)asm_enc_offsets.asm
+  endif
+
+  ifeq ($(CONFIG_VP8_DECODER), yes)
+    asm_dec_offsets.asm: obj_int_extract
+    asm_dec_offsets.asm: $(VP8_PREFIX)decoder/asm_dec_offsets.c.o
+	./obj_int_extract rvds $< $(ADS2GAS) > $@
+    OBJS-yes += $(VP8_PREFIX)decoder/asm_dec_offsets.c.o
+    CLEAN-OBJS += asm_dec_offsets.asm
+    $(filter %$(ASM).o,$(OBJS-yes)): $(BUILD_PFX)asm_dec_offsets.asm
+  endif
+endif
+
 $(shell $(SRC_PATH_BARE)/build/make/version.sh "$(SRC_PATH_BARE)" $(BUILD_PFX)vpx_version.h)
 CLEAN-OBJS += $(BUILD_PFX)vpx_version.h
 
diff --git a/vp8/common/arm/armv6/bilinearfilter_v6.asm b/vp8/common/arm/armv6/bilinearfilter_v6.asm
index 09d7338..a86ed5d 100644
--- a/vp8/common/arm/armv6/bilinearfilter_v6.asm
+++ b/vp8/common/arm/armv6/bilinearfilter_v6.asm
@@ -15,19 +15,19 @@
     AREA    |.text|, CODE, READONLY  ; name this block of code
 
 ;-------------------------------------
-; r0    unsigned char *src_ptr,
-; r1    unsigned short *output_ptr,
-; r2    unsigned int src_pixels_per_line,
-; r3    unsigned int output_height,
-; stack    unsigned int output_width,
-; stack    const short *vp8_filter
+; r0    unsigned char  *src_ptr,
+; r1    unsigned short *dst_ptr,
+; r2    unsigned int    src_pitch,
+; r3    unsigned int    height,
+; stack unsigned int    width,
+; stack const short    *vp8_filter
 ;-------------------------------------
 ; The output is transposed stroed in output array to make it easy for second pass filtering.
 |vp8_filter_block2d_bil_first_pass_armv6| PROC
     stmdb   sp!, {r4 - r11, lr}
 
     ldr     r11, [sp, #40]                  ; vp8_filter address
-    ldr     r4, [sp, #36]                   ; output width
+    ldr     r4, [sp, #36]                   ; width
 
     mov     r12, r3                         ; outer-loop counter
     sub     r2, r2, r4                      ; src increment for height loop
@@ -38,10 +38,10 @@
 
     ldr     r5, [r11]                       ; load up filter coefficients
 
-    mov     r3, r3, lsl #1                  ; output_height*2
+    mov     r3, r3, lsl #1                  ; height*2
     add     r3, r3, #2                      ; plus 2 to make output buffer 4-bit aligned since height is actually (height+1)
 
-    mov     r11, r1                         ; save output_ptr for each row
+    mov     r11, r1                         ; save dst_ptr for each row
 
     cmp     r5, #128                        ; if filter coef = 128, then skip the filter
     beq     bil_null_1st_filter
@@ -140,17 +140,17 @@
 
 ;---------------------------------
 ; r0    unsigned short *src_ptr,
-; r1    unsigned char *output_ptr,
-; r2    int output_pitch,
-; r3    unsigned int  output_height,
-; stack unsigned int  output_width,
-; stack const short *vp8_filter
+; r1    unsigned char  *dst_ptr,
+; r2    int             dst_pitch,
+; r3    unsigned int    height,
+; stack unsigned int    width,
+; stack const short    *vp8_filter
 ;---------------------------------
 |vp8_filter_block2d_bil_second_pass_armv6| PROC
     stmdb   sp!, {r4 - r11, lr}
 
     ldr     r11, [sp, #40]                  ; vp8_filter address
-    ldr     r4, [sp, #36]                   ; output width
+    ldr     r4, [sp, #36]                   ; width
 
     ldr     r5, [r11]                       ; load up filter coefficients
     mov     r12, r4                         ; outer-loop counter = width, since we work on transposed data matrix
diff --git a/vp8/common/arm/bilinearfilter_arm.c b/vp8/common/arm/bilinearfilter_arm.c
index 65afb41..6a46ef6 100644
--- a/vp8/common/arm/bilinearfilter_arm.c
+++ b/vp8/common/arm/bilinearfilter_arm.c
@@ -10,128 +10,29 @@
 
 
 #include <math.h>
+#include "filter.h"
 #include "subpixel.h"
-
-#define BLOCK_HEIGHT_WIDTH 4
-#define VP8_FILTER_WEIGHT 128
-#define VP8_FILTER_SHIFT  7
-
-static const short bilinear_filters[8][2] =
-{
-    { 128,   0 },
-    { 112,  16 },
-    {  96,  32 },
-    {  80,  48 },
-    {  64,  64 },
-    {  48,  80 },
-    {  32,  96 },
-    {  16, 112 }
-};
-
-
-extern void vp8_filter_block2d_bil_first_pass_armv6
-(
-    unsigned char *src_ptr,
-    unsigned short *output_ptr,
-    unsigned int src_pixels_per_line,
-    unsigned int output_height,
-    unsigned int output_width,
-    const short *vp8_filter
-);
-
-extern void vp8_filter_block2d_bil_second_pass_armv6
-(
-    unsigned short *src_ptr,
-    unsigned char  *output_ptr,
-    int output_pitch,
-    unsigned int  output_height,
-    unsigned int  output_width,
-    const short *vp8_filter
-);
-
-#if 0
-void vp8_filter_block2d_bil_first_pass_6
-(
-    unsigned char *src_ptr,
-    unsigned short *output_ptr,
-    unsigned int src_pixels_per_line,
-    unsigned int output_height,
-    unsigned int output_width,
-    const short *vp8_filter
-)
-{
-    unsigned int i, j;
-
-    for ( i=0; i<output_height; i++ )
-    {
-        for ( j=0; j<output_width; j++ )
-        {
-            /* Apply bilinear filter */
-            output_ptr[j] = ( ( (int)src_ptr[0]          * vp8_filter[0]) +
-                               ((int)src_ptr[1] * vp8_filter[1]) +
-                                (VP8_FILTER_WEIGHT/2) ) >> VP8_FILTER_SHIFT;
-            src_ptr++;
-        }
-
-        /* Next row... */
-        src_ptr    += src_pixels_per_line - output_width;
-        output_ptr += output_width;
-    }
-}
-
-void vp8_filter_block2d_bil_second_pass_6
-(
-    unsigned short *src_ptr,
-    unsigned char  *output_ptr,
-    int output_pitch,
-    unsigned int  output_height,
-    unsigned int  output_width,
-    const short *vp8_filter
-)
-{
-    unsigned int  i,j;
-    int  Temp;
-
-    for ( i=0; i<output_height; i++ )
-    {
-        for ( j=0; j<output_width; j++ )
-        {
-            /* Apply filter */
-            Temp =  ((int)src_ptr[0]         * vp8_filter[0]) +
-                    ((int)src_ptr[output_width] * vp8_filter[1]) +
-                    (VP8_FILTER_WEIGHT/2);
-            output_ptr[j] = (unsigned int)(Temp >> VP8_FILTER_SHIFT);
-            src_ptr++;
-        }
-
-        /* Next row... */
-        /*src_ptr    += src_pixels_per_line - output_width;*/
-        output_ptr += output_pitch;
-    }
-}
-#endif
+#include "arm/bilinearfilter_arm.h"
 
 void vp8_filter_block2d_bil_armv6
 (
     unsigned char *src_ptr,
-    unsigned char *output_ptr,
-    unsigned int   src_pixels_per_line,
+    unsigned char *dst_ptr,
+    unsigned int   src_pitch,
     unsigned int   dst_pitch,
-    const short      *HFilter,
-    const short      *VFilter,
+    const short   *HFilter,
+    const short   *VFilter,
     int            Width,
     int            Height
 )
 {
-
-    unsigned short FData[36*16]; /* Temp data bufffer used in filtering */
+    unsigned short FData[36*16]; /* Temp data buffer used in filtering */
 
     /* First filter 1-D horizontally... */
-    /* pixel_step = 1; */
-    vp8_filter_block2d_bil_first_pass_armv6(src_ptr, FData, src_pixels_per_line, Height + 1, Width, HFilter);
+    vp8_filter_block2d_bil_first_pass_armv6(src_ptr, FData, src_pitch, Height + 1, Width, HFilter);
 
     /* then 1-D vertically... */
-    vp8_filter_block2d_bil_second_pass_armv6(FData, output_ptr, dst_pitch, Height, Width, VFilter);
+    vp8_filter_block2d_bil_second_pass_armv6(FData, dst_ptr, dst_pitch, Height, Width, VFilter);
 }
 
 
@@ -148,8 +49,8 @@
     const short  *HFilter;
     const short  *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil_armv6(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 4, 4);
 }
@@ -167,8 +68,8 @@
     const short  *HFilter;
     const short  *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil_armv6(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 8, 8);
 }
@@ -186,8 +87,8 @@
     const short  *HFilter;
     const short  *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil_armv6(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 8, 4);
 }
@@ -205,8 +106,8 @@
     const short  *HFilter;
     const short  *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil_armv6(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 16, 16);
 }
diff --git a/vp8/common/arm/bilinearfilter_arm.h b/vp8/common/arm/bilinearfilter_arm.h
new file mode 100644
index 0000000..b7155d3
--- /dev/null
+++ b/vp8/common/arm/bilinearfilter_arm.h
@@ -0,0 +1,35 @@
+/*
+ *  Copyright (c) 2011 The WebM project authors. All Rights Reserved.
+ *
+ *  Use of this source code is governed by a BSD-style license
+ *  that can be found in the LICENSE file in the root of the source
+ *  tree. An additional intellectual property rights grant can be found
+ *  in the file PATENTS.  All contributing project authors may
+ *  be found in the AUTHORS file in the root of the source tree.
+ */
+
+
+#ifndef BILINEARFILTER_ARM_H
+#define BILINEARFILTER_ARM_H
+
+extern void vp8_filter_block2d_bil_first_pass_armv6
+(
+    const unsigned char  *src_ptr,
+    unsigned short       *dst_ptr,
+    unsigned int          src_pitch,
+    unsigned int          height,
+    unsigned int          width,
+    const short          *vp8_filter
+);
+
+extern void vp8_filter_block2d_bil_second_pass_armv6
+(
+    const unsigned short *src_ptr,
+    unsigned char        *dst_ptr,
+    int                   dst_pitch,
+    unsigned int          height,
+    unsigned int          width,
+    const short         *vp8_filter
+);
+
+#endif /* BILINEARFILTER_ARM_H */
diff --git a/vp8/common/arm/filter_arm.c b/vp8/common/arm/filter_arm.c
index b4f2fe6..2612fc1 100644
--- a/vp8/common/arm/filter_arm.c
+++ b/vp8/common/arm/filter_arm.c
@@ -11,26 +11,10 @@
 
 #include "vpx_ports/config.h"
 #include <math.h>
+#include "filter.h"
 #include "subpixel.h"
 #include "vpx_ports/mem.h"
 
-#define BLOCK_HEIGHT_WIDTH 4
-#define VP8_FILTER_WEIGHT 128
-#define VP8_FILTER_SHIFT  7
-
-DECLARE_ALIGNED(16, static const short, sub_pel_filters[8][6]) =
-{
-    { 0,  0,  128,    0,   0,  0 },         /* note that 1/8 pel positions are just as per alpha -0.5 bicubic */
-    { 0, -6,  123,   12,  -1,  0 },
-    { 2, -11, 108,   36,  -8,  1 },         /* New 1/4 pel 6 tap filter */
-    { 0, -9,   93,   50,  -6,  0 },
-    { 3, -16,  77,   77, -16,  3 },         /* New 1/2 pel 6 tap filter */
-    { 0, -6,   50,   93,  -9,  0 },
-    { 1, -8,   36,  108, -11,  2 },         /* New 1/4 pel 6 tap filter */
-    { 0, -1,   12,  123,  -6,  0 },
-};
-
-
 extern void vp8_filter_block2d_first_pass_armv6
 (
     unsigned char *src_ptr,
@@ -93,11 +77,11 @@
 {
     const short  *HFilter;
     const short  *VFilter;
-    DECLARE_ALIGNED_ARRAY(4, short, FData, 12*4); /* Temp data bufffer used in filtering */
+    DECLARE_ALIGNED_ARRAY(4, short, FData, 12*4); /* Temp data buffer used in filtering */
 
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];       /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     /* Vfilter is null. First pass only */
     if (xoffset && !yoffset)
@@ -129,47 +113,6 @@
     }
 }
 
-#if 0
-void vp8_sixtap_predict8x4_armv6
-(
-    unsigned char  *src_ptr,
-    int  src_pixels_per_line,
-    int  xoffset,
-    int  yoffset,
-    unsigned char *dst_ptr,
-    int  dst_pitch
-)
-{
-    const short  *HFilter;
-    const short  *VFilter;
-    DECLARE_ALIGNED_ARRAY(4, short, FData, 16*8); /* Temp data bufffer used in filtering */
-
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];       /* 6 tap */
-
-
-    /*if (xoffset && !yoffset)
-    {
-        vp8_filter_block2d_first_pass_only_armv6 (  src_ptr, dst_ptr, src_pixels_per_line, 8, dst_pitch, HFilter );
-    }*/
-    /* Hfilter is null. Second pass only */
-    /*else if (!xoffset && yoffset)
-    {
-        vp8_filter_block2d_second_pass_only_armv6 ( src_ptr, dst_ptr, src_pixels_per_line, 8, dst_pitch, VFilter );
-    }
-    else
-    {
-        if (yoffset & 0x1)
-            vp8_filter_block2d_first_pass_armv6 ( src_ptr-src_pixels_per_line, FData+1, src_pixels_per_line, 8, 7, HFilter );
-        else*/
-
-        vp8_filter_block2d_first_pass_armv6 ( src_ptr-(2*src_pixels_per_line), FData, src_pixels_per_line, 8, 9, HFilter );
-
-        vp8_filter_block2d_second_pass_armv6 ( FData+2, dst_ptr, dst_pitch, 4, 8, VFilter );
-    /*}*/
-}
-#endif
-
 void vp8_sixtap_predict8x8_armv6
 (
     unsigned char  *src_ptr,
@@ -182,10 +125,10 @@
 {
     const short  *HFilter;
     const short  *VFilter;
-    DECLARE_ALIGNED_ARRAY(4, short, FData, 16*8); /* Temp data bufffer used in filtering */
+    DECLARE_ALIGNED_ARRAY(4, short, FData, 16*8); /* Temp data buffer used in filtering */
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];       /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     if (xoffset && !yoffset)
     {
@@ -224,10 +167,10 @@
 {
     const short  *HFilter;
     const short  *VFilter;
-    DECLARE_ALIGNED_ARRAY(4, short, FData, 24*16);    /* Temp data bufffer used in filtering */
+    DECLARE_ALIGNED_ARRAY(4, short, FData, 24*16);    /* Temp data buffer used in filtering */
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];       /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     if (xoffset && !yoffset)
     {
diff --git a/vp8/common/asm_com_offsets.c b/vp8/common/asm_com_offsets.c
new file mode 100644
index 0000000..d299dd2
--- /dev/null
+++ b/vp8/common/asm_com_offsets.c
@@ -0,0 +1,49 @@
+/*
+ *  Copyright (c) 2011 The WebM project authors. All Rights Reserved.
+ *
+ *  Use of this source code is governed by a BSD-style license
+ *  that can be found in the LICENSE file in the root of the source
+ *  tree. An additional intellectual property rights grant can be found
+ *  in the file PATENTS.  All contributing project authors may
+ *  be found in the AUTHORS file in the root of the source tree.
+ */
+
+
+#include "vpx_ports/config.h"
+#include <stddef.h>
+
+#include "vpx_scale/yv12config.h"
+
+#define ct_assert(name,cond) \
+    static void assert_##name(void) UNUSED;\
+    static void assert_##name(void) {switch(0){case 0:case !!(cond):;}}
+
+#define DEFINE(sym, val) int sym = val;
+
+/*
+#define BLANK() asm volatile("\n->" : : )
+*/
+
+/*
+ * int main(void)
+ * {
+ */
+
+//vpx_scale
+DEFINE(yv12_buffer_config_y_width,              offsetof(YV12_BUFFER_CONFIG, y_width));
+DEFINE(yv12_buffer_config_y_height,             offsetof(YV12_BUFFER_CONFIG, y_height));
+DEFINE(yv12_buffer_config_y_stride,             offsetof(YV12_BUFFER_CONFIG, y_stride));
+DEFINE(yv12_buffer_config_uv_width,             offsetof(YV12_BUFFER_CONFIG, uv_width));
+DEFINE(yv12_buffer_config_uv_height,            offsetof(YV12_BUFFER_CONFIG, uv_height));
+DEFINE(yv12_buffer_config_uv_stride,            offsetof(YV12_BUFFER_CONFIG, uv_stride));
+DEFINE(yv12_buffer_config_y_buffer,             offsetof(YV12_BUFFER_CONFIG, y_buffer));
+DEFINE(yv12_buffer_config_u_buffer,             offsetof(YV12_BUFFER_CONFIG, u_buffer));
+DEFINE(yv12_buffer_config_v_buffer,             offsetof(YV12_BUFFER_CONFIG, v_buffer));
+DEFINE(yv12_buffer_config_border,               offsetof(YV12_BUFFER_CONFIG, border));
+
+//add asserts for any offset that is not supported by assembly code
+//add asserts for any size that is not supported by assembly code
+/*
+ * return 0;
+ * }
+ */
diff --git a/vp8/common/filter_c.c b/vp8/common/filter.c
similarity index 66%
rename from vp8/common/filter_c.c
rename to vp8/common/filter.c
index 399a847..6e364a9 100644
--- a/vp8/common/filter_c.c
+++ b/vp8/common/filter.c
@@ -10,13 +10,10 @@
 
 
 #include <stdlib.h>
+#include "filter.h"
+#include "vpx_ports/mem.h"
 
-#define BLOCK_HEIGHT_WIDTH 4
-#define VP8_FILTER_WEIGHT 128
-#define VP8_FILTER_SHIFT  7
-
-
-static const int bilinear_filters[8][2] =
+DECLARE_ALIGNED(16, const short, vp8_bilinear_filters[8][2]) =
 {
     { 128,   0 },
     { 112,  16 },
@@ -28,8 +25,7 @@
     {  16, 112 }
 };
 
-
-static const short sub_pel_filters[8][6] =
+DECLARE_ALIGNED(16, const short, vp8_sub_pel_filters[8][6]) =
 {
 
     { 0,  0,  128,    0,   0,  0 },         /* note that 1/8 pel positions are just as per alpha -0.5 bicubic */
@@ -40,9 +36,6 @@
     { 0, -6,   50,   93,  -9,  0 },
     { 1, -8,   36,  108, -11,  2 },         /* New 1/4 pel 6 tap filter */
     { 0, -1,   12,  123,  -6,  0 },
-
-
-
 };
 
 void vp8_filter_block2d_first_pass
@@ -146,7 +139,7 @@
     const short  *VFilter
 )
 {
-    int FData[9*4]; /* Temp data bufffer used in filtering */
+    int FData[9*4]; /* Temp data buffer used in filtering */
 
     /* First filter 1-D horizontally... */
     vp8_filter_block2d_first_pass(src_ptr - (2 * src_pixels_per_line), FData, src_pixels_per_line, 1, 9, 4, HFilter);
@@ -195,8 +188,8 @@
     const short  *HFilter;
     const short  *VFilter;
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];   /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     vp8_filter_block2d(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter);
 }
@@ -212,10 +205,10 @@
 {
     const short  *HFilter;
     const short  *VFilter;
-    int FData[13*16];   /* Temp data bufffer used in filtering */
+    int FData[13*16];   /* Temp data buffer used in filtering */
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];   /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     /* First filter 1-D horizontally... */
     vp8_filter_block2d_first_pass(src_ptr - (2 * src_pixels_per_line), FData, src_pixels_per_line, 1, 13, 8, HFilter);
@@ -238,10 +231,10 @@
 {
     const short  *HFilter;
     const short  *VFilter;
-    int FData[13*16];   /* Temp data bufffer used in filtering */
+    int FData[13*16];   /* Temp data buffer used in filtering */
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];   /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     /* First filter 1-D horizontally... */
     vp8_filter_block2d_first_pass(src_ptr - (2 * src_pixels_per_line), FData, src_pixels_per_line, 1, 9, 8, HFilter);
@@ -264,11 +257,11 @@
 {
     const short  *HFilter;
     const short  *VFilter;
-    int FData[21*24];   /* Temp data bufffer used in filtering */
+    int FData[21*24];   /* Temp data buffer used in filtering */
 
 
-    HFilter = sub_pel_filters[xoffset];   /* 6 tap */
-    VFilter = sub_pel_filters[yoffset];   /* 6 tap */
+    HFilter = vp8_sub_pel_filters[xoffset];   /* 6 tap */
+    VFilter = vp8_sub_pel_filters[yoffset];   /* 6 tap */
 
     /* First filter 1-D horizontally... */
     vp8_filter_block2d_first_pass(src_ptr - (2 * src_pixels_per_line), FData, src_pixels_per_line, 1, 21, 16, HFilter);
@@ -283,57 +276,50 @@
  *
  *  ROUTINE       : filter_block2d_bil_first_pass
  *
- *  INPUTS        : UINT8  *src_ptr          : Pointer to source block.
- *                  UINT32 src_pixels_per_line : Stride of input block.
- *                  UINT32 pixel_step        : Offset between filter input samples (see notes).
- *                  UINT32 output_height     : Input block height.
- *                  UINT32 output_width      : Input block width.
- *                  INT32  *vp8_filter          : Array of 2 bi-linear filter taps.
+ *  INPUTS        : UINT8  *src_ptr    : Pointer to source block.
+ *                  UINT32  src_stride : Stride of source block.
+ *                  UINT32  height     : Block height.
+ *                  UINT32  width      : Block width.
+ *                  INT32  *vp8_filter : Array of 2 bi-linear filter taps.
  *
- *  OUTPUTS       : INT32 *output_ptr        : Pointer to filtered block.
+ *  OUTPUTS       : INT32  *dst_ptr    : Pointer to filtered block.
  *
  *  RETURNS       : void
  *
- *  FUNCTION      : Applies a 1-D 2-tap bi-linear filter to the source block in
- *                  either horizontal or vertical direction to produce the
- *                  filtered output block. Used to implement first-pass
- *                  of 2-D separable filter.
+ *  FUNCTION      : Applies a 1-D 2-tap bi-linear filter to the source block
+ *                  in the horizontal direction to produce the filtered output
+ *                  block. Used to implement first-pass of 2-D separable filter.
  *
  *  SPECIAL NOTES : Produces INT32 output to retain precision for next pass.
  *                  Two filter taps should sum to VP8_FILTER_WEIGHT.
- *                  pixel_step defines whether the filter is applied
- *                  horizontally (pixel_step=1) or vertically (pixel_step=stride).
- *                  It defines the offset required to move from one input
- *                  to the next.
  *
  ****************************************************************************/
 void vp8_filter_block2d_bil_first_pass
 (
-    unsigned char *src_ptr,
-    unsigned short *output_ptr,
-    unsigned int src_pixels_per_line,
-    int pixel_step,
-    unsigned int output_height,
-    unsigned int output_width,
-    const int *vp8_filter
+    unsigned char  *src_ptr,
+    unsigned short *dst_ptr,
+    unsigned int    src_stride,
+    unsigned int    height,
+    unsigned int    width,
+    const short    *vp8_filter
 )
 {
     unsigned int i, j;
 
-    for (i = 0; i < output_height; i++)
+    for (i = 0; i < height; i++)
     {
-        for (j = 0; j < output_width; j++)
+        for (j = 0; j < width; j++)
         {
             /* Apply bilinear filter */
-            output_ptr[j] = (((int)src_ptr[0]          * vp8_filter[0]) +
-                             ((int)src_ptr[pixel_step] * vp8_filter[1]) +
-                             (VP8_FILTER_WEIGHT / 2)) >> VP8_FILTER_SHIFT;
+            dst_ptr[j] = (((int)src_ptr[0] * vp8_filter[0]) +
+                          ((int)src_ptr[1] * vp8_filter[1]) +
+                          (VP8_FILTER_WEIGHT / 2)) >> VP8_FILTER_SHIFT;
             src_ptr++;
         }
 
         /* Next row... */
-        src_ptr    += src_pixels_per_line - output_width;
-        output_ptr += output_width;
+        src_ptr += src_stride - width;
+        dst_ptr += width;
     }
 }
 
@@ -341,60 +327,51 @@
  *
  *  ROUTINE       : filter_block2d_bil_second_pass
  *
- *  INPUTS        : INT32  *src_ptr          : Pointer to source block.
- *                  UINT32 src_pixels_per_line : Stride of input block.
- *                  UINT32 pixel_step        : Offset between filter input samples (see notes).
- *                  UINT32 output_height     : Input block height.
- *                  UINT32 output_width      : Input block width.
- *                  INT32  *vp8_filter          : Array of 2 bi-linear filter taps.
+ *  INPUTS        : INT32  *src_ptr    : Pointer to source block.
+ *                  UINT32  dst_pitch  : Destination block pitch.
+ *                  UINT32  height     : Block height.
+ *                  UINT32  width      : Block width.
+ *                  INT32  *vp8_filter : Array of 2 bi-linear filter taps.
  *
- *  OUTPUTS       : UINT16 *output_ptr       : Pointer to filtered block.
+ *  OUTPUTS       : UINT16 *dst_ptr    : Pointer to filtered block.
  *
  *  RETURNS       : void
  *
- *  FUNCTION      : Applies a 1-D 2-tap bi-linear filter to the source block in
- *                  either horizontal or vertical direction to produce the
- *                  filtered output block. Used to implement second-pass
- *                  of 2-D separable filter.
+ *  FUNCTION      : Applies a 1-D 2-tap bi-linear filter to the source block
+ *                  in the vertical direction to produce the filtered output
+ *                  block. Used to implement second-pass of 2-D separable filter.
  *
  *  SPECIAL NOTES : Requires 32-bit input as produced by filter_block2d_bil_first_pass.
  *                  Two filter taps should sum to VP8_FILTER_WEIGHT.
- *                  pixel_step defines whether the filter is applied
- *                  horizontally (pixel_step=1) or vertically (pixel_step=stride).
- *                  It defines the offset required to move from one input
- *                  to the next.
  *
  ****************************************************************************/
 void vp8_filter_block2d_bil_second_pass
 (
     unsigned short *src_ptr,
-    unsigned char  *output_ptr,
-    int output_pitch,
-    unsigned int  src_pixels_per_line,
-    unsigned int  pixel_step,
-    unsigned int  output_height,
-    unsigned int  output_width,
-    const int *vp8_filter
+    unsigned char  *dst_ptr,
+    int             dst_pitch,
+    unsigned int    height,
+    unsigned int    width,
+    const short    *vp8_filter
 )
 {
     unsigned int  i, j;
     int  Temp;
 
-    for (i = 0; i < output_height; i++)
+    for (i = 0; i < height; i++)
     {
-        for (j = 0; j < output_width; j++)
+        for (j = 0; j < width; j++)
         {
             /* Apply filter */
-            Temp = ((int)src_ptr[0]         * vp8_filter[0]) +
-                   ((int)src_ptr[pixel_step] * vp8_filter[1]) +
+            Temp = ((int)src_ptr[0]     * vp8_filter[0]) +
+                   ((int)src_ptr[width] * vp8_filter[1]) +
                    (VP8_FILTER_WEIGHT / 2);
-            output_ptr[j] = (unsigned int)(Temp >> VP8_FILTER_SHIFT);
+            dst_ptr[j] = (unsigned int)(Temp >> VP8_FILTER_SHIFT);
             src_ptr++;
         }
 
         /* Next row... */
-        src_ptr    += src_pixels_per_line - output_width;
-        output_ptr += output_pitch;
+        dst_ptr += dst_pitch;
     }
 }
 
@@ -404,11 +381,14 @@
  *  ROUTINE       : filter_block2d_bil
  *
  *  INPUTS        : UINT8  *src_ptr          : Pointer to source block.
- *                  UINT32 src_pixels_per_line : Stride of input block.
- *                  INT32  *HFilter         : Array of 2 horizontal filter taps.
- *                  INT32  *VFilter         : Array of 2 vertical filter taps.
+ *                  UINT32  src_pitch        : Stride of source block.
+ *                  UINT32  dst_pitch        : Stride of destination block.
+ *                  INT32  *HFilter          : Array of 2 horizontal filter taps.
+ *                  INT32  *VFilter          : Array of 2 vertical filter taps.
+ *                  INT32  Width             : Block width
+ *                  INT32  Height            : Block height
  *
- *  OUTPUTS       : UINT16 *output_ptr       : Pointer to filtered block.
+ *  OUTPUTS       : UINT16 *dst_ptr       : Pointer to filtered block.
  *
  *  RETURNS       : void
  *
@@ -422,23 +402,23 @@
 void vp8_filter_block2d_bil
 (
     unsigned char *src_ptr,
-    unsigned char *output_ptr,
-    unsigned int   src_pixels_per_line,
+    unsigned char *dst_ptr,
+    unsigned int   src_pitch,
     unsigned int   dst_pitch,
-    const int      *HFilter,
-    const int      *VFilter,
+    const short   *HFilter,
+    const short   *VFilter,
     int            Width,
     int            Height
 )
 {
 
-    unsigned short FData[17*16];    /* Temp data bufffer used in filtering */
+    unsigned short FData[17*16];    /* Temp data buffer used in filtering */
 
     /* First filter 1-D horizontally... */
-    vp8_filter_block2d_bil_first_pass(src_ptr, FData, src_pixels_per_line, 1, Height + 1, Width, HFilter);
+    vp8_filter_block2d_bil_first_pass(src_ptr, FData, src_pitch, Height + 1, Width, HFilter);
 
     /* then 1-D vertically... */
-    vp8_filter_block2d_bil_second_pass(FData, output_ptr, dst_pitch, Width, Width, Height, Width, VFilter);
+    vp8_filter_block2d_bil_second_pass(FData, dst_ptr, dst_pitch, Height, Width, VFilter);
 }
 
 
@@ -452,11 +432,11 @@
     int dst_pitch
 )
 {
-    const int  *HFilter;
-    const int  *VFilter;
+    const short *HFilter;
+    const short *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 #if 0
     {
         int i;
@@ -490,11 +470,11 @@
     int  dst_pitch
 )
 {
-    const int  *HFilter;
-    const int  *VFilter;
+    const short *HFilter;
+    const short *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 8, 8);
 
@@ -510,11 +490,11 @@
     int  dst_pitch
 )
 {
-    const int  *HFilter;
-    const int  *VFilter;
+    const short *HFilter;
+    const short *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 8, 4);
 
@@ -530,11 +510,11 @@
     int  dst_pitch
 )
 {
-    const int  *HFilter;
-    const int  *VFilter;
+    const short *HFilter;
+    const short *VFilter;
 
-    HFilter = bilinear_filters[xoffset];
-    VFilter = bilinear_filters[yoffset];
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
 
     vp8_filter_block2d_bil(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 16, 16);
 }
diff --git a/vp8/common/filter.h b/vp8/common/filter.h
new file mode 100644
index 0000000..0f225c2
--- /dev/null
+++ b/vp8/common/filter.h
@@ -0,0 +1,22 @@
+/*
+ *  Copyright (c) 2011 The WebM project authors. All Rights Reserved.
+ *
+ *  Use of this source code is governed by a BSD-style license
+ *  that can be found in the LICENSE file in the root of the source
+ *  tree. An additional intellectual property rights grant can be found
+ *  in the file PATENTS.  All contributing project authors may
+ *  be found in the AUTHORS file in the root of the source tree.
+ */
+
+
+#ifndef FILTER_H
+#define FILTER_H
+
+#define BLOCK_HEIGHT_WIDTH 4
+#define VP8_FILTER_WEIGHT 128
+#define VP8_FILTER_SHIFT  7
+
+extern const short vp8_bilinear_filters[8][2];
+extern const short vp8_sub_pel_filters[8][6];
+
+#endif //FILTER_H
diff --git a/vp8/common/threading.h b/vp8/common/threading.h
index 0dd7bbc..bfd4916 100644
--- a/vp8/common/threading.h
+++ b/vp8/common/threading.h
@@ -14,7 +14,7 @@
 
 #define VPXINFINITE 10000       /* 10second. */
 
-#if CONFIG_OS_SUPPORT
+#if CONFIG_OS_SUPPORT && CONFIG_MULTITHREAD
 
 /* Thread management macros */
 #ifdef _WIN32
@@ -90,8 +90,6 @@
 #define x86_pause_hint()
 #endif
 
-#else /* CONFIG_OS_SUPPORT = 0 */
-#define THREAD_FUNCTION void *
-#endif /* CONFIG_OS_SUPPORT */
+#endif /* CONFIG_OS_SUPPORT && CONFIG_MULTITHREAD */
 
 #endif
diff --git a/vp8/decoder/arm/detokenize.asm b/vp8/decoder/arm/detokenize.asm
index 45e068a..0c164f1 100644
--- a/vp8/decoder/arm/detokenize.asm
+++ b/vp8/decoder/arm/detokenize.asm
@@ -13,7 +13,7 @@
 
     AREA    |.text|, CODE, READONLY  ; name this block of code
 
-    INCLUDE vpx_asm_offsets.asm
+    INCLUDE asm_dec_offsets.asm
 
 l_qcoeff    EQU     0
 l_i         EQU     4
diff --git a/vp8/common/arm/vpx_asm_offsets.c b/vp8/decoder/asm_dec_offsets.c
similarity index 65%
rename from vp8/common/arm/vpx_asm_offsets.c
rename to vp8/decoder/asm_dec_offsets.c
index 074d350..e485cb4 100644
--- a/vp8/common/arm/vpx_asm_offsets.c
+++ b/vp8/decoder/asm_dec_offsets.c
@@ -1,5 +1,5 @@
 /*
- *  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
+ *  Copyright (c) 2011 The WebM project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
@@ -12,13 +12,7 @@
 #include "vpx_ports/config.h"
 #include <stddef.h>
 
-#if CONFIG_VP8_ENCODER
-#include "vpx_scale/yv12config.h"
-#endif
-
-#if CONFIG_VP8_DECODER
 #include "onyxd_int.h"
-#endif
 
 #define DEFINE(sym, val) int sym = val;
 
@@ -31,20 +25,6 @@
  * {
  */
 
-#if CONFIG_VP8_DECODER || CONFIG_VP8_ENCODER
-DEFINE(yv12_buffer_config_y_width,              offsetof(YV12_BUFFER_CONFIG, y_width));
-DEFINE(yv12_buffer_config_y_height,             offsetof(YV12_BUFFER_CONFIG, y_height));
-DEFINE(yv12_buffer_config_y_stride,             offsetof(YV12_BUFFER_CONFIG, y_stride));
-DEFINE(yv12_buffer_config_uv_width,             offsetof(YV12_BUFFER_CONFIG, uv_width));
-DEFINE(yv12_buffer_config_uv_height,            offsetof(YV12_BUFFER_CONFIG, uv_height));
-DEFINE(yv12_buffer_config_uv_stride,            offsetof(YV12_BUFFER_CONFIG, uv_stride));
-DEFINE(yv12_buffer_config_y_buffer,             offsetof(YV12_BUFFER_CONFIG, y_buffer));
-DEFINE(yv12_buffer_config_u_buffer,             offsetof(YV12_BUFFER_CONFIG, u_buffer));
-DEFINE(yv12_buffer_config_v_buffer,             offsetof(YV12_BUFFER_CONFIG, v_buffer));
-DEFINE(yv12_buffer_config_border,               offsetof(YV12_BUFFER_CONFIG, border));
-#endif
-
-#if CONFIG_VP8_DECODER
 DEFINE(detok_scan,                              offsetof(DETOK, scan));
 DEFINE(detok_ptr_block2leftabove,               offsetof(DETOK, ptr_block2leftabove));
 DEFINE(detok_coef_tree_ptr,                     offsetof(DETOK, vp8_coef_tree_ptr));
@@ -68,7 +48,6 @@
 
 DEFINE(tokenextrabits_min_val,                  offsetof(TOKENEXTRABITS, min_val));
 DEFINE(tokenextrabits_length,                   offsetof(TOKENEXTRABITS, Length));
-#endif
 
 //add asserts for any offset that is not supported by assembly code
 //add asserts for any size that is not supported by assembly code
diff --git a/vp8/decoder/decodframe.c b/vp8/decoder/decodframe.c
index bf79b31..f4bb664 100644
--- a/vp8/decoder/decodframe.c
+++ b/vp8/decoder/decodframe.c
@@ -484,9 +484,11 @@
         bool_decoder++;
     }
 
+#if CONFIG_MULTITHREAD
     /* Clamp number of decoder threads */
     if (pbi->decoding_thread_count > num_part - 1)
         pbi->decoding_thread_count = num_part - 1;
+#endif
 }
 
 
@@ -844,7 +846,9 @@
     vpx_memcpy(&xd->dst, &pc->yv12_fb[pc->new_fb_idx], sizeof(YV12_BUFFER_CONFIG));
 
     /* set up frame new frame for intra coded blocks */
+#if CONFIG_MULTITHREAD
     if (!(pbi->b_multithreaded_rd) || pc->multi_token_partition == ONE_PARTITION || !(pc->filter_level))
+#endif
         vp8_setup_intra_recon(&pc->yv12_fb[pc->new_fb_idx]);
 
     vp8_setup_block_dptrs(xd);
@@ -864,6 +868,7 @@
 
     vpx_memcpy(&xd->block[0].bmi, &xd->mode_info_context->bmi[0], sizeof(B_MODE_INFO));
 
+#if CONFIG_MULTITHREAD
     if (pbi->b_multithreaded_rd && pc->multi_token_partition != ONE_PARTITION)
     {
         vp8mt_decode_mb_rows(pbi, xd);
@@ -878,6 +883,7 @@
         vp8_yv12_extend_frame_borders_ptr(&pc->yv12_fb[pc->new_fb_idx]);    /*cm->frame_to_show);*/
     }
     else
+#endif
     {
         int ibc = 0;
         int num_part = 1 << pc->multi_token_partition;
diff --git a/vp8/decoder/onyxd_if.c b/vp8/decoder/onyxd_if.c
index adbb1f3..41966b9 100644
--- a/vp8/decoder/onyxd_if.c
+++ b/vp8/decoder/onyxd_if.c
@@ -114,8 +114,10 @@
     pbi->ready_for_new_data = 1;
 
     pbi->CPUFreq = 0; /*vp8_get_processor_freq();*/
+#if CONFIG_MULTITHREAD
     pbi->max_threads = oxcf->max_threads;
     vp8_decoder_create_threads(pbi);
+#endif
 
     /* vp8cx_init_de_quantizer() is first called here. Add check in frame_init_dequantizer() to avoid
      *  unnecessary calling of vp8cx_init_de_quantizer() for every frame.
@@ -149,8 +151,8 @@
 #if CONFIG_MULTITHREAD
     if (pbi->b_multithreaded_rd)
         vp8mt_de_alloc_temp_buffers(pbi, pbi->common.mb_rows);
-#endif
     vp8_decoder_remove_threads(pbi);
+#endif
     vp8_remove_common(&pbi->common);
     vpx_free(pbi);
 }
@@ -407,6 +409,7 @@
         return retcode;
     }
 
+#if CONFIG_MULTITHREAD
     if (pbi->b_multithreaded_rd && cm->multi_token_partition != ONE_PARTITION)
     {
         if (swap_frame_buffers (cm))
@@ -424,6 +427,7 @@
             return -1;
         }
     } else
+#endif
     {
         if (swap_frame_buffers (cm))
         {
diff --git a/vp8/decoder/onyxd_int.h b/vp8/decoder/onyxd_int.h
index 2f0fe70..09a8976 100644
--- a/vp8/decoder/onyxd_int.h
+++ b/vp8/decoder/onyxd_int.h
@@ -87,14 +87,15 @@
     unsigned int time_decoding;
     unsigned int time_loop_filtering;
 
+#if CONFIG_MULTITHREAD
+    /* variable for threading */
+
     volatile int b_multithreaded_rd;
     int max_threads;
     int current_mb_col_main;
     int decoding_thread_count;
     int allocated_decoding_thread_count;
 
-    /* variable for threading */
-#if CONFIG_MULTITHREAD
     int mt_baseline_filter_level[MAX_MB_SEGMENTS];
     int sync_range;
     int *mt_current_mb_col;                  /* Each row remembers its already decoded column. */
diff --git a/vp8/decoder/reconintra_mt.c b/vp8/decoder/reconintra_mt.c
index ad4324b..854aba3 100644
--- a/vp8/decoder/reconintra_mt.c
+++ b/vp8/decoder/reconintra_mt.c
@@ -21,7 +21,6 @@
 
 void vp8mt_build_intra_predictors_mby(VP8D_COMP *pbi, MACROBLOCKD *x, int mb_row, int mb_col)
 {
-#if CONFIG_MULTITHREAD
     unsigned char *yabove_row;    /* = x->dst.y_buffer - x->dst.y_stride; */
     unsigned char *yleft_col;
     unsigned char yleft_buf[16];
@@ -146,17 +145,10 @@
     case MB_MODE_COUNT:
         break;
     }
-#else
-    (void) pbi;
-    (void) x;
-    (void) mb_row;
-    (void) mb_col;
-#endif
 }
 
 void vp8mt_build_intra_predictors_mby_s(VP8D_COMP *pbi, MACROBLOCKD *x, int mb_row, int mb_col)
 {
-#if CONFIG_MULTITHREAD
     unsigned char *yabove_row;    /* = x->dst.y_buffer - x->dst.y_stride; */
     unsigned char *yleft_col;
     unsigned char yleft_buf[16];
@@ -289,17 +281,10 @@
     case MB_MODE_COUNT:
         break;
     }
-#else
-    (void) pbi;
-    (void) x;
-    (void) mb_row;
-    (void) mb_col;
-#endif
 }
 
 void vp8mt_build_intra_predictors_mbuv(VP8D_COMP *pbi, MACROBLOCKD *x, int mb_row, int mb_col)
 {
-#if CONFIG_MULTITHREAD
     unsigned char *uabove_row;   /* = x->dst.u_buffer - x->dst.uv_stride; */
     unsigned char *uleft_col;    /*[16];*/
     unsigned char uleft_buf[8];
@@ -452,17 +437,10 @@
     case MB_MODE_COUNT:
         break;
     }
-#else
-    (void) pbi;
-    (void) x;
-    (void) mb_row;
-    (void) mb_col;
-#endif
 }
 
 void vp8mt_build_intra_predictors_mbuv_s(VP8D_COMP *pbi, MACROBLOCKD *x, int mb_row, int mb_col)
 {
-#if CONFIG_MULTITHREAD
     unsigned char *uabove_row;  /* = x->dst.u_buffer - x->dst.uv_stride; */
     unsigned char *uleft_col;   /*[16];*/
     unsigned char uleft_buf[8];
@@ -621,12 +599,6 @@
     case MB_MODE_COUNT:
         break;
     }
-#else
-    (void) pbi;
-    (void) x;
-    (void) mb_row;
-    (void) mb_col;
-#endif
 }
 
 
@@ -638,7 +610,6 @@
                           int mb_col,
                           int num)
 {
-#if CONFIG_MULTITHREAD
     int i, r, c;
 
     unsigned char *Above;   /* = *(x->base_dst) + x->dst - x->dst_stride; */
@@ -935,15 +906,6 @@
 
 
     }
-#else
-    (void) pbi;
-    (void) xd;
-    (void) b_mode;
-    (void) predictor;
-    (void) mb_row;
-    (void) mb_col;
-    (void) num;
-#endif
 }
 
 /* copy 4 bytes from the above right down so that the 4x4 prediction modes using pixels above and
@@ -951,7 +913,6 @@
  */
 void vp8mt_intra_prediction_down_copy(VP8D_COMP *pbi, MACROBLOCKD *x, int mb_row, int mb_col)
 {
-#if CONFIG_MULTITHREAD
     unsigned char *above_right;   /* = *(x->block[0].base_dst) + x->block[0].dst - x->block[0].dst_stride + 16; */
     unsigned int *src_ptr;
     unsigned int *dst_ptr0;
@@ -973,10 +934,4 @@
     *dst_ptr0 = *src_ptr;
     *dst_ptr1 = *src_ptr;
     *dst_ptr2 = *src_ptr;
-#else
-    (void) pbi;
-    (void) x;
-    (void) mb_row;
-    (void) mb_col;
-#endif
 }
diff --git a/vp8/decoder/threading.c b/vp8/decoder/threading.c
index ec2cb2b..271249a 100644
--- a/vp8/decoder/threading.c
+++ b/vp8/decoder/threading.c
@@ -38,7 +38,6 @@
 
 void vp8_setup_decoding_thread_data(VP8D_COMP *pbi, MACROBLOCKD *xd, MB_ROW_DEC *mbrd, int count)
 {
-#if CONFIG_MULTITHREAD
     VP8_COMMON *const pc = & pbi->common;
     int i, j;
 
@@ -88,18 +87,11 @@
 
     for (i=0; i< pc->mb_rows; i++)
         pbi->mt_current_mb_col[i]=-1;
-#else
-    (void) pbi;
-    (void) xd;
-    (void) mbrd;
-    (void) count;
-#endif
 }
 
 
 void vp8mt_decode_macroblock(VP8D_COMP *pbi, MACROBLOCKD *xd, int mb_row, int mb_col)
 {
-#if CONFIG_MULTITHREAD
     int eobtotal = 0;
     int i, do_clamp = xd->mode_info_context->mbmi.need_to_clamp_mvs;
     VP8_COMMON *pc = &pbi->common;
@@ -222,18 +214,11 @@
                     (xd->qcoeff+16*16, xd->block[16].dequant,
                      xd->predictor+16*16, xd->dst.u_buffer, xd->dst.v_buffer,
                      xd->dst.uv_stride, xd->eobs+16);
-#else
-    (void) pbi;
-    (void) xd;
-    (void) mb_row;
-    (void) mb_col;
-#endif
 }
 
 
 THREAD_FUNCTION vp8_thread_decoding_proc(void *p_data)
 {
-#if CONFIG_MULTITHREAD
     int ithread = ((DECODETHREAD_DATA *)p_data)->ithread;
     VP8D_COMP *pbi = (VP8D_COMP *)(((DECODETHREAD_DATA *)p_data)->ptr1);
     MB_ROW_DEC *mbrd = (MB_ROW_DEC *)(((DECODETHREAD_DATA *)p_data)->ptr2);
@@ -438,9 +423,6 @@
             sem_post(&pbi->h_event_end_decoding);
         }
     }
-#else
-    (void) p_data;
-#endif
 
     return 0 ;
 }
@@ -448,7 +430,6 @@
 
 void vp8_decoder_create_threads(VP8D_COMP *pbi)
 {
-#if CONFIG_MULTITHREAD
     int core_count = 0;
     int ithread;
 
@@ -482,16 +463,11 @@
 
         pbi->allocated_decoding_thread_count = pbi->decoding_thread_count;
     }
-
-#else
-    (void) pbi;
-#endif
 }
 
 
 void vp8mt_de_alloc_temp_buffers(VP8D_COMP *pbi, int mb_rows)
 {
-#if CONFIG_MULTITHREAD
     VP8_COMMON *const pc = & pbi->common;
     int i;
 
@@ -589,15 +565,11 @@
             pbi->mt_vleft_col = NULL ;
         }
     }
-#else
-    (void) pbi;
-#endif
 }
 
 
 void vp8mt_alloc_temp_buffers(VP8D_COMP *pbi, int width, int prev_mb_rows)
 {
-#if CONFIG_MULTITHREAD
     VP8_COMMON *const pc = & pbi->common;
     int i;
     int uv_width;
@@ -646,17 +618,11 @@
         for (i=0; i< pc->mb_rows; i++)
             CHECK_MEM_ERROR(pbi->mt_vleft_col[i], vpx_calloc(sizeof(unsigned char) * 8, 1));
     }
-#else
-    (void) pbi;
-    (void) width;
-#endif
 }
 
 
 void vp8_decoder_remove_threads(VP8D_COMP *pbi)
 {
-#if CONFIG_MULTITHREAD
-
     /* shutdown MB Decoding thread; */
     if (pbi->b_multithreaded_rd)
     {
@@ -702,15 +668,11 @@
             pbi->de_thread_data = NULL;
         }
     }
-#else
-    (void) pbi;
-#endif
 }
 
 
 void vp8mt_lpf_init( VP8D_COMP *pbi, int default_filt_lvl)
 {
-#if CONFIG_MULTITHREAD
     VP8_COMMON *cm  = &pbi->common;
     MACROBLOCKD *mbd = &pbi->mb;
     /*YV12_BUFFER_CONFIG *post = &cm->new_frame;*/  /*frame_to_show;*/
@@ -752,16 +714,11 @@
         vp8_init_loop_filter(cm);
     else if (frame_type != cm->last_frame_type)
         vp8_frame_init_loop_filter(lfi, frame_type);
-#else
-    (void) pbi;
-    (void) default_filt_lvl;
-#endif
 }
 
 
 void vp8mt_decode_mb_rows( VP8D_COMP *pbi, MACROBLOCKD *xd)
 {
-#if CONFIG_MULTITHREAD
     int mb_row;
     VP8_COMMON *pc = &pbi->common;
 
@@ -981,8 +938,4 @@
     }
 
     sem_wait(&pbi->h_event_end_decoding);   /* add back for each frame */
-#else
-    (void) pbi;
-    (void) xd;
-#endif
 }
diff --git a/vp8/encoder/arm/arm_csystemdependent.c b/vp8/encoder/arm/arm_csystemdependent.c
index a1f1102..6c17a79 100644
--- a/vp8/encoder/arm/arm_csystemdependent.c
+++ b/vp8/encoder/arm/arm_csystemdependent.c
@@ -38,14 +38,14 @@
         /*cpi->rtcd.variance.var4x4                = vp8_variance4x4_c;
         cpi->rtcd.variance.var8x8                = vp8_variance8x8_c;
         cpi->rtcd.variance.var8x16               = vp8_variance8x16_c;
-        cpi->rtcd.variance.var16x8               = vp8_variance16x8_c;
-        cpi->rtcd.variance.var16x16              = vp8_variance16x16_c;*/
+        cpi->rtcd.variance.var16x8               = vp8_variance16x8_c;*/
+        cpi->rtcd.variance.var16x16              = vp8_variance16x16_armv6;
 
         /*cpi->rtcd.variance.subpixvar4x4          = vp8_sub_pixel_variance4x4_c;
         cpi->rtcd.variance.subpixvar8x8          = vp8_sub_pixel_variance8x8_c;
         cpi->rtcd.variance.subpixvar8x16         = vp8_sub_pixel_variance8x16_c;
-        cpi->rtcd.variance.subpixvar16x8         = vp8_sub_pixel_variance16x8_c;
-        cpi->rtcd.variance.subpixvar16x16        = vp8_sub_pixel_variance16x16_c;*/
+        cpi->rtcd.variance.subpixvar16x8         = vp8_sub_pixel_variance16x8_c;*/
+        cpi->rtcd.variance.subpixvar16x16        = vp8_sub_pixel_variance16x16_armv6;
 
         /*cpi->rtcd.variance.mse16x16              = vp8_mse16x16_c;
         cpi->rtcd.variance.getmbss               = vp8_get_mb_ss_c;*/
diff --git a/vp8/encoder/arm/armv5te/boolhuff_armv5te.asm b/vp8/encoder/arm/armv5te/boolhuff_armv5te.asm
index e78dc33..3c05f57 100644
--- a/vp8/encoder/arm/armv5te/boolhuff_armv5te.asm
+++ b/vp8/encoder/arm/armv5te/boolhuff_armv5te.asm
@@ -14,7 +14,7 @@
     EXPORT |vp8_stop_encode|
     EXPORT |vp8_encode_value|
 
-    INCLUDE vpx_vp8_enc_asm_offsets.asm
+    INCLUDE asm_enc_offsets.asm
 
     ARM
     REQUIRE8
diff --git a/vp8/encoder/arm/armv5te/vp8_packtokens_armv5.asm b/vp8/encoder/arm/armv5te/vp8_packtokens_armv5.asm
index b2abadf..d939287 100644
--- a/vp8/encoder/arm/armv5te/vp8_packtokens_armv5.asm
+++ b/vp8/encoder/arm/armv5te/vp8_packtokens_armv5.asm
@@ -11,7 +11,7 @@
 
     EXPORT |vp8cx_pack_tokens_armv5|
 
-    INCLUDE vpx_vp8_enc_asm_offsets.asm
+    INCLUDE asm_enc_offsets.asm
 
     ARM
     REQUIRE8
diff --git a/vp8/encoder/arm/armv5te/vp8_packtokens_mbrow_armv5.asm b/vp8/encoder/arm/armv5te/vp8_packtokens_mbrow_armv5.asm
index f9c3852..ac2bba6 100644
--- a/vp8/encoder/arm/armv5te/vp8_packtokens_mbrow_armv5.asm
+++ b/vp8/encoder/arm/armv5te/vp8_packtokens_mbrow_armv5.asm
@@ -11,7 +11,7 @@
 
     EXPORT |vp8cx_pack_mb_row_tokens_armv5|
 
-    INCLUDE vpx_vp8_enc_asm_offsets.asm
+    INCLUDE asm_enc_offsets.asm
 
     ARM
     REQUIRE8
diff --git a/vp8/encoder/arm/armv5te/vp8_packtokens_partitions_armv5.asm b/vp8/encoder/arm/armv5te/vp8_packtokens_partitions_armv5.asm
index 42dae13..c2eccdb 100644
--- a/vp8/encoder/arm/armv5te/vp8_packtokens_partitions_armv5.asm
+++ b/vp8/encoder/arm/armv5te/vp8_packtokens_partitions_armv5.asm
@@ -11,7 +11,7 @@
 
     EXPORT |vp8cx_pack_tokens_into_partitions_armv5|
 
-    INCLUDE vpx_vp8_enc_asm_offsets.asm
+    INCLUDE asm_enc_offsets.asm
 
     ARM
     REQUIRE8
diff --git a/vp8/encoder/arm/armv6/vp8_variance16x16_armv6.asm b/vp8/encoder/arm/armv6/vp8_variance16x16_armv6.asm
new file mode 100644
index 0000000..8d7258a
--- /dev/null
+++ b/vp8/encoder/arm/armv6/vp8_variance16x16_armv6.asm
@@ -0,0 +1,147 @@
+;
+;  Copyright (c) 2011 The WebM project authors. All Rights Reserved.
+;
+;  Use of this source code is governed by a BSD-style license
+;  that can be found in the LICENSE file in the root of the source
+;  tree. An additional intellectual property rights grant can be found
+;  in the file PATENTS.  All contributing project authors may
+;  be found in the AUTHORS file in the root of the source tree.
+;
+
+
+    EXPORT  |vp8_variance16x16_armv6|
+
+    ARM
+    REQUIRE8
+    PRESERVE8
+
+    AREA ||.text||, CODE, READONLY, ALIGN=2
+
+; r0    unsigned char *src_ptr
+; r1    int source_stride
+; r2    unsigned char *ref_ptr
+; r3    int  recon_stride
+; stack unsigned int *sse
+|vp8_variance16x16_armv6| PROC
+
+    stmfd   sp!, {r4-r12, lr}
+    mov     r12, #16            ; set loop counter to 16 (=block height)
+    mov     r8, #0              ; initialize sum = 0
+    mov     r11, #0             ; initialize sse = 0
+
+loop
+    ; 1st 4 pixels
+    ldr     r4, [r0, #0x0]      ; load 4 src pixels
+    ldr     r5, [r2, #0x0]      ; load 4 ref pixels
+
+    mov     lr, #0              ; constant zero
+
+    usub8   r6, r4, r5          ; calculate difference
+    sel     r7, r6, lr          ; select bytes with positive difference
+    usub8   r9, r5, r4          ; calculate difference with reversed operands
+    sel     r6, r9, lr          ; select bytes with negative difference
+
+    ; calculate partial sums
+    usad8   r4, r7, lr          ; calculate sum of positive differences
+    usad8   r5, r6, lr          ; calculate sum of negative differences
+    orr     r6, r6, r7          ; differences of all 4 pixels
+    ; calculate total sum
+    adds    r8, r8, r4          ; add positive differences to sum
+    subs    r8, r8, r5          ; substract negative differences from sum
+
+    ; calculate sse
+    uxtb16  r5, r6              ; byte (two pixels) to halfwords
+    uxtb16  r10, r6, ror #8     ; another two pixels to halfwords
+    smlad   r11, r5, r5, r11    ; dual signed multiply, add and accumulate (1)
+
+    ; 2nd 4 pixels
+    ldr     r4, [r0, #0x4]      ; load 4 src pixels
+    ldr     r5, [r2, #0x4]      ; load 4 ref pixels
+    smlad   r11, r10, r10, r11  ; dual signed multiply, add and accumulate (2)
+
+    usub8   r6, r4, r5          ; calculate difference
+    sel     r7, r6, lr          ; select bytes with positive difference
+    usub8   r9, r5, r4          ; calculate difference with reversed operands
+    sel     r6, r9, lr          ; select bytes with negative difference
+
+    ; calculate partial sums
+    usad8   r4, r7, lr          ; calculate sum of positive differences
+    usad8   r5, r6, lr          ; calculate sum of negative differences
+    orr     r6, r6, r7          ; differences of all 4 pixels
+
+    ; calculate total sum
+    add     r8, r8, r4          ; add positive differences to sum
+    sub     r8, r8, r5          ; substract negative differences from sum
+
+    ; calculate sse
+    uxtb16  r5, r6              ; byte (two pixels) to halfwords
+    uxtb16  r10, r6, ror #8     ; another two pixels to halfwords
+    smlad   r11, r5, r5, r11    ; dual signed multiply, add and accumulate (1)
+
+    ; 3rd 4 pixels
+    ldr     r4, [r0, #0x8]      ; load 4 src pixels
+    ldr     r5, [r2, #0x8]      ; load 4 ref pixels
+    smlad   r11, r10, r10, r11  ; dual signed multiply, add and accumulate (2)
+
+    usub8   r6, r4, r5          ; calculate difference
+    sel     r7, r6, lr          ; select bytes with positive difference
+    usub8   r9, r5, r4          ; calculate difference with reversed operands
+    sel     r6, r9, lr          ; select bytes with negative difference
+
+    ; calculate partial sums
+    usad8   r4, r7, lr          ; calculate sum of positive differences
+    usad8   r5, r6, lr          ; calculate sum of negative differences
+    orr     r6, r6, r7          ; differences of all 4 pixels
+
+    ; calculate total sum
+    add     r8, r8, r4          ; add positive differences to sum
+    sub     r8, r8, r5          ; substract negative differences from sum
+
+    ; calculate sse
+    uxtb16  r5, r6              ; byte (two pixels) to halfwords
+    uxtb16  r10, r6, ror #8     ; another two pixels to halfwords
+    smlad   r11, r5, r5, r11    ; dual signed multiply, add and accumulate (1)
+
+    ; 4th 4 pixels
+    ldr     r4, [r0, #0xc]      ; load 4 src pixels
+    ldr     r5, [r2, #0xc]      ; load 4 ref pixels
+    smlad   r11, r10, r10, r11  ; dual signed multiply, add and accumulate (2)
+
+    usub8   r6, r4, r5          ; calculate difference
+    add     r0, r0, r1          ; set src_ptr to next row
+    sel     r7, r6, lr          ; select bytes with positive difference
+    usub8   r9, r5, r4          ; calculate difference with reversed operands
+    add     r2, r2, r3          ; set dst_ptr to next row
+    sel     r6, r9, lr          ; select bytes with negative difference
+
+    ; calculate partial sums
+    usad8   r4, r7, lr          ; calculate sum of positive differences
+    usad8   r5, r6, lr          ; calculate sum of negative differences
+    orr     r6, r6, r7          ; differences of all 4 pixels
+
+    ; calculate total sum
+    add     r8, r8, r4          ; add positive differences to sum
+    sub     r8, r8, r5          ; substract negative differences from sum
+
+    ; calculate sse
+    uxtb16  r5, r6              ; byte (two pixels) to halfwords
+    uxtb16  r10, r6, ror #8     ; another two pixels to halfwords
+    smlad   r11, r5, r5, r11    ; dual signed multiply, add and accumulate (1)
+    smlad   r11, r10, r10, r11  ; dual signed multiply, add and accumulate (2)
+
+
+    subs    r12, r12, #1
+
+    bne     loop
+
+    ; return stuff
+    ldr     r6, [sp, #0x28]     ; get address of sse
+    mul     r0, r8, r8          ; sum * sum
+    str     r11, [r6]           ; store sse
+    sub     r0, r11, r0, ASR #8 ; return (sse - ((sum * sum) >> 8))
+
+    ldmfd   sp!, {r4-r12, pc}
+
+    ENDP
+
+    END
diff --git a/vp8/encoder/arm/variance_arm.c b/vp8/encoder/arm/variance_arm.c
index b40c048..9737bef 100644
--- a/vp8/encoder/arm/variance_arm.c
+++ b/vp8/encoder/arm/variance_arm.c
@@ -10,6 +10,40 @@
 
 #include "vpx_config.h"
 #include "variance.h"
+#include "filter.h"
+#include "arm/bilinearfilter_arm.h"
+
+#if HAVE_ARMV6
+
+unsigned int vp8_sub_pixel_variance16x16_armv6
+(
+    const unsigned char  *src_ptr,
+    int  src_pixels_per_line,
+    int  xoffset,
+    int  yoffset,
+    const unsigned char *dst_ptr,
+    int dst_pixels_per_line,
+    unsigned int *sse
+)
+{
+    unsigned short first_pass[36*16];
+    unsigned char  second_pass[20*16];
+    const short *HFilter, *VFilter;
+
+    HFilter = vp8_bilinear_filters[xoffset];
+    VFilter = vp8_bilinear_filters[yoffset];
+
+    vp8_filter_block2d_bil_first_pass_armv6(src_ptr, first_pass,
+                                            src_pixels_per_line,
+                                            17, 16, HFilter);
+    vp8_filter_block2d_bil_second_pass_armv6(first_pass, second_pass,
+                                             16, 16, 16, VFilter);
+
+    return vp8_variance16x16_armv6(second_pass, 16, dst_ptr,
+                                   dst_pixels_per_line, sse);
+}
+
+#endif
 
 #if HAVE_ARMV7
 
diff --git a/vp8/encoder/arm/variance_arm.h b/vp8/encoder/arm/variance_arm.h
index 3cbacfa..06d7287 100644
--- a/vp8/encoder/arm/variance_arm.h
+++ b/vp8/encoder/arm/variance_arm.h
@@ -12,6 +12,23 @@
 #ifndef VARIANCE_ARM_H
 #define VARIANCE_ARM_H
 
+#if HAVE_ARMV6
+
+extern prototype_variance(vp8_variance16x16_armv6);
+extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_armv6);
+
+#if !CONFIG_RUNTIME_CPU_DETECT
+
+#undef  vp8_variance_subpixvar16x16
+#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_armv6
+
+#undef  vp8_variance_var16x16
+#define vp8_variance_var16x16 vp8_variance16x16_armv6
+
+#endif /* !CONFIG_RUNTIME_CPU_DETECT */
+
+#endif /* HAVE_ARMV6 */
+
 #if HAVE_ARMV7
 extern prototype_sad(vp8_sad4x4_neon);
 extern prototype_sad(vp8_sad8x8_neon);
diff --git a/vp8/encoder/arm/vpx_vp8_enc_asm_offsets.c b/vp8/encoder/asm_enc_offsets.c
similarity index 72%
rename from vp8/encoder/arm/vpx_vp8_enc_asm_offsets.c
rename to vp8/encoder/asm_enc_offsets.c
index 4703a84..cd49532 100644
--- a/vp8/encoder/arm/vpx_vp8_enc_asm_offsets.c
+++ b/vp8/encoder/asm_enc_offsets.c
@@ -1,5 +1,5 @@
 /*
- *  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
+ *  Copyright (c) 2011 The WebM project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
@@ -12,9 +12,9 @@
 #include "vpx_ports/config.h"
 #include <stddef.h>
 
-#include "../treewriter.h"
-#include "../tokenize.h"
-#include "../onyx_int.h"
+#include "treewriter.h"
+#include "tokenize.h"
+#include "onyx_int.h"
 
 #define ct_assert(name,cond) \
     static void assert_##name(void) UNUSED;\
@@ -31,6 +31,7 @@
  * {
  */
 
+//pack tokens
 DEFINE(vp8_writer_lowvalue,                     offsetof(vp8_writer, lowvalue));
 DEFINE(vp8_writer_range,                        offsetof(vp8_writer, range));
 DEFINE(vp8_writer_value,                        offsetof(vp8_writer, value));
@@ -40,19 +41,19 @@
 
 DEFINE(tokenextra_token,                        offsetof(TOKENEXTRA, Token));
 DEFINE(tokenextra_extra,                        offsetof(TOKENEXTRA, Extra));
-DEFINE(tokenextra_context_tree,                  offsetof(TOKENEXTRA, context_tree));
+DEFINE(tokenextra_context_tree,                 offsetof(TOKENEXTRA, context_tree));
 DEFINE(tokenextra_skip_eob_node,                offsetof(TOKENEXTRA, skip_eob_node));
 DEFINE(TOKENEXTRA_SZ,                           sizeof(TOKENEXTRA));
 
-DEFINE(vp8_extra_bit_struct_sz,                   sizeof(vp8_extra_bit_struct));
+DEFINE(vp8_extra_bit_struct_sz,                 sizeof(vp8_extra_bit_struct));
 
 DEFINE(vp8_token_value,                         offsetof(vp8_token, value));
 DEFINE(vp8_token_len,                           offsetof(vp8_token, Len));
 
-DEFINE(vp8_extra_bit_struct_tree,                 offsetof(vp8_extra_bit_struct, tree));
-DEFINE(vp8_extra_bit_struct_prob,                 offsetof(vp8_extra_bit_struct, prob));
-DEFINE(vp8_extra_bit_struct_len,                  offsetof(vp8_extra_bit_struct, Len));
-DEFINE(vp8_extra_bit_struct_base_val,              offsetof(vp8_extra_bit_struct, base_val));
+DEFINE(vp8_extra_bit_struct_tree,               offsetof(vp8_extra_bit_struct, tree));
+DEFINE(vp8_extra_bit_struct_prob,               offsetof(vp8_extra_bit_struct, prob));
+DEFINE(vp8_extra_bit_struct_len,                offsetof(vp8_extra_bit_struct, Len));
+DEFINE(vp8_extra_bit_struct_base_val,           offsetof(vp8_extra_bit_struct, base_val));
 
 DEFINE(vp8_comp_tplist,                         offsetof(VP8_COMP, tplist));
 DEFINE(vp8_comp_common,                         offsetof(VP8_COMP, common));
@@ -62,12 +63,14 @@
 DEFINE(tokenlist_stop,                          offsetof(TOKENLIST, stop));
 DEFINE(TOKENLIST_SZ,                            sizeof(TOKENLIST));
 
-DEFINE(vp8_common_mb_rows,                       offsetof(VP8_COMMON, mb_rows));
+DEFINE(vp8_common_mb_rows,                      offsetof(VP8_COMMON, mb_rows));
 
-// These two sizes are used in vp7cx_pack_tokens.  They are hard coded
-//  so if the size changes this will have to be adjusted.
+// These two sizes are used in vp8cx_pack_tokens.  They are hard coded
+// so if the size changes this will have to be adjusted.
+#if HAVE_ARMV5TE
 ct_assert(TOKENEXTRA_SZ, sizeof(TOKENEXTRA) == 8)
 ct_assert(vp8_extra_bit_struct_sz, sizeof(vp8_extra_bit_struct) == 16)
+#endif
 
 //add asserts for any offset that is not supported by assembly code
 //add asserts for any size that is not supported by assembly code
diff --git a/vp8/encoder/bitstream.c b/vp8/encoder/bitstream.c
index 412542d..52ad0d6 100644
--- a/vp8/encoder/bitstream.c
+++ b/vp8/encoder/bitstream.c
@@ -1654,10 +1654,12 @@
     {
         vp8_start_encode(&cpi->bc2, cx_data + bc->pos);
 
-        if (!cpi->b_multi_threaded)
-            pack_tokens(&cpi->bc2, cpi->tok, cpi->tok_count);
-        else
+#if CONFIG_MULTITHREAD
+        if (cpi->b_multi_threaded)
             pack_mb_row_tokens(cpi, &cpi->bc2);
+        else
+#endif
+            pack_tokens(&cpi->bc2, cpi->tok, cpi->tok_count);
 
         vp8_stop_encode(&cpi->bc2);
         oh.first_partition_length_in_bytes = cpi->bc.pos ;
diff --git a/vp8/encoder/encodeframe.c b/vp8/encoder/encodeframe.c
index f06a4ca..c871e60 100644
--- a/vp8/encoder/encodeframe.c
+++ b/vp8/encoder/encodeframe.c
@@ -800,28 +800,9 @@
         struct vpx_usec_timer  emr_timer;
         vpx_usec_timer_start(&emr_timer);
 
-        if (!cpi->b_multi_threaded)
-        {
-            // for each macroblock row in image
-            for (mb_row = 0; mb_row < cm->mb_rows; mb_row++)
-            {
-
-                vp8_zero(cm->left_context)
-
-                encode_mb_row(cpi, cm, mb_row, x, xd, &tp, segment_counts, &totalrate);
-
-                // adjust to the next row of mbs
-                x->src.y_buffer += 16 * x->src.y_stride - 16 * cm->mb_cols;
-                x->src.u_buffer += 8 * x->src.uv_stride - 8 * cm->mb_cols;
-                x->src.v_buffer += 8 * x->src.uv_stride - 8 * cm->mb_cols;
-            }
-
-            cpi->tok_count = tp - cpi->tok;
-
-        }
-        else
-        {
 #if CONFIG_MULTITHREAD
+        if (cpi->b_multi_threaded)
+        {
             int i;
 
             vp8cx_init_mbrthread_data(cpi, x, cpi->mb_row_ei, 1,  cpi->encoding_thread_count);
@@ -886,7 +867,25 @@
                 x->activity_sum += cpi->mb_row_ei[i].mb.activity_sum;
             }
 
+        }
+        else
 #endif
+        {
+            // for each macroblock row in image
+            for (mb_row = 0; mb_row < cm->mb_rows; mb_row++)
+            {
+
+                vp8_zero(cm->left_context)
+
+                encode_mb_row(cpi, cm, mb_row, x, xd, &tp, segment_counts, &totalrate);
+
+                // adjust to the next row of mbs
+                x->src.y_buffer += 16 * x->src.y_stride - 16 * cm->mb_cols;
+                x->src.u_buffer += 8 * x->src.uv_stride - 8 * cm->mb_cols;
+                x->src.v_buffer += 8 * x->src.uv_stride - 8 * cm->mb_cols;
+            }
+
+            cpi->tok_count = tp - cpi->tok;
 
         }
 
@@ -1166,7 +1165,7 @@
 
         Error16x16 = vp8_rd_pick_intra16x16mby_mode(cpi, x, &rate16x16, &rate16x16_tokenonly, &dist16x16);
 
-        Error4x4 = vp8_rd_pick_intra4x4mby_modes(cpi, x, &rate4x4, &rate4x4_tokenonly, &dist4x4);
+        Error4x4 = vp8_rd_pick_intra4x4mby_modes(cpi, x, &rate4x4, &rate4x4_tokenonly, &dist4x4, Error16x16);
 
         rate += (Error4x4 < Error16x16) ? rate4x4 : rate16x16;
     }
diff --git a/vp8/encoder/encodemv.c b/vp8/encoder/encodemv.c
index cce7530..4cb4c6e 100644
--- a/vp8/encoder/encodemv.c
+++ b/vp8/encoder/encodemv.c
@@ -128,7 +128,7 @@
 
         while (--i > 3);
 
-        if (x & 240)
+        if (x & 0xFFF0)
             cost += vp8_cost_bit(p [MVPbits + 3], (x >> 3) & 1);
     }
 
diff --git a/vp8/encoder/ethreading.c b/vp8/encoder/ethreading.c
index 8da93a6..4ce306f 100644
--- a/vp8/encoder/ethreading.c
+++ b/vp8/encoder/ethreading.c
@@ -13,6 +13,8 @@
 #include "common.h"
 #include "extend.h"
 
+#if CONFIG_MULTITHREAD
+
 extern int vp8cx_encode_inter_macroblock(VP8_COMP *cpi, MACROBLOCK *x,
                                          TOKENEXTRA **t, int recon_yoffset,
                                          int recon_uvoffset);
@@ -25,7 +27,6 @@
 static
 THREAD_FUNCTION thread_encoding_proc(void *p_data)
 {
-#if CONFIG_MULTITHREAD
     int ithread = ((ENCODETHREAD_DATA *)p_data)->ithread;
     VP8_COMP *cpi = (VP8_COMP *)(((ENCODETHREAD_DATA *)p_data)->ptr1);
     MB_ROW_COMP *mbri = (MB_ROW_COMP *)(((ENCODETHREAD_DATA *)p_data)->ptr2);
@@ -247,10 +248,6 @@
         }
     }
 
-#else
-    (void) p_data;
-#endif
-
     //printf("exit thread %d\n", ithread);
     return 0;
 }
@@ -436,10 +433,6 @@
 
     cpi->processor_core_count = 32; //vp8_get_proc_core_count();
 
-    CHECK_MEM_ERROR(cpi->tplist, vpx_malloc(sizeof(TOKENLIST) * cpi->common.mb_rows));
-
-#if CONFIG_MULTITHREAD
-
     if (cpi->processor_core_count > 1 && cpi->oxcf.multi_threaded > 1)
     {
         int ithread;
@@ -488,13 +481,10 @@
 
     }
 
-#endif
 }
 
 void vp8cx_remove_encoder_threads(VP8_COMP *cpi)
 {
-#if CONFIG_MULTITHREAD
-
     if (cpi->b_multi_threaded)
     {
         //shutdown other threads
@@ -521,7 +511,5 @@
         vpx_free(cpi->en_thread_data);
         vpx_free(cpi->mt_current_mb_col);
     }
-
-#endif
-    vpx_free(cpi->tplist);
 }
+#endif
diff --git a/vp8/encoder/onyx_if.c b/vp8/encoder/onyx_if.c
index 595c31c..511631a 100644
--- a/vp8/encoder/onyx_if.c
+++ b/vp8/encoder/onyx_if.c
@@ -262,6 +262,9 @@
 
 void vp8_dealloc_compressor_data(VP8_COMP *cpi)
 {
+    vpx_free(cpi->tplist);
+    cpi->tplist = NULL;
+
     // Delete last frame MV storage buffers
     if (cpi->lfmv != 0)
         vpx_free(cpi->lfmv);
@@ -1476,6 +1479,8 @@
     else
         cpi->mt_sync_range = 16;
 #endif
+
+    CHECK_MEM_ERROR(cpi->tplist, vpx_malloc(sizeof(TOKENLIST) * cpi->common.mb_rows));
 }
 
 
@@ -2427,7 +2432,9 @@
     init_mv_ref_counts();
 #endif
 
+#if CONFIG_MULTITHREAD
     vp8cx_create_encoder_threads(cpi);
+#endif
 
     cpi->fn_ptr[BLOCK_16X16].sdf            = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16);
     cpi->fn_ptr[BLOCK_16X16].vf             = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16);
@@ -2702,7 +2709,9 @@
 
     }
 
+#if CONFIG_MULTITHREAD
     vp8cx_remove_encoder_threads(cpi);
+#endif
 
     vp8_dealloc_compressor_data(cpi);
     vpx_free(cpi->mb.ss);
diff --git a/vp8/encoder/onyx_int.h b/vp8/encoder/onyx_int.h
index e8a452d..0289b4a 100644
--- a/vp8/encoder/onyx_int.h
+++ b/vp8/encoder/onyx_int.h
@@ -589,6 +589,7 @@
     int cyclic_refresh_q;
     signed char *cyclic_refresh_map;
 
+#if CONFIG_MULTITHREAD
     // multithread data
     int * mt_current_mb_col;
     int mt_sync_range;
@@ -596,13 +597,10 @@
     int b_multi_threaded;
     int encoding_thread_count;
 
-#if CONFIG_MULTITHREAD
     pthread_t *h_encoding_thread;
-#endif
     MB_ROW_COMP *mb_row_ei;
     ENCODETHREAD_DATA *en_thread_data;
 
-#if CONFIG_MULTITHREAD
     //events
     sem_t *h_event_start_encoding;
     sem_t h_event_end_encoding;
diff --git a/vp8/encoder/rdopt.c b/vp8/encoder/rdopt.c
index d8b4f5c..7080425 100644
--- a/vp8/encoder/rdopt.c
+++ b/vp8/encoder/rdopt.c
@@ -584,7 +584,7 @@
     *Rate = vp8_rdcost_mby(mb);
 }
 
-static void rd_pick_intra4x4block(
+static int rd_pick_intra4x4block(
     VP8_COMP *cpi,
     MACROBLOCK *x,
     BLOCK *be,
@@ -650,16 +650,20 @@
     b->bmi.mode = (B_PREDICTION_MODE)(*best_mode);
     vp8_encode_intra4x4block_rd(IF_RTCD(&cpi->rtcd), x, be, b, b->bmi.mode);
 
+    return best_rd;
+
 }
 
 
-int vp8_rd_pick_intra4x4mby_modes(VP8_COMP *cpi, MACROBLOCK *mb, int *Rate, int *rate_y, int *Distortion)
+int vp8_rd_pick_intra4x4mby_modes(VP8_COMP *cpi, MACROBLOCK *mb, int *Rate,
+                                  int *rate_y, int *Distortion, int best_rd)
 {
     MACROBLOCKD *const xd = &mb->e_mbd;
     int i;
     int cost = mb->mbmode_cost [xd->frame_type] [B_PRED];
     int distortion = 0;
     int tot_rate_y = 0;
+    int total_rd = 0;
     ENTROPY_CONTEXT_PLANES t_above, t_left;
     ENTROPY_CONTEXT *ta;
     ENTROPY_CONTEXT *tl;
@@ -681,7 +685,7 @@
         B_PREDICTION_MODE UNINITIALIZED_IS_SAFE(best_mode);
         int UNINITIALIZED_IS_SAFE(r), UNINITIALIZED_IS_SAFE(ry), UNINITIALIZED_IS_SAFE(d);
 
-        rd_pick_intra4x4block(
+        total_rd += rd_pick_intra4x4block(
             cpi, mb, mb->block + i, xd->block + i, &best_mode, A, L,
             ta + vp8_block2above[i],
             tl + vp8_block2left[i], &r, &ry, &d);
@@ -690,8 +694,14 @@
         distortion += d;
         tot_rate_y += ry;
         mic->bmi[i].mode = xd->block[i].bmi.mode = best_mode;
+
+        if(total_rd >= best_rd)
+          break;
     }
 
+    if(total_rd >= best_rd)
+      return INT_MAX;
+
     *Rate = cost;
     *rate_y += tot_rate_y;
     *Distortion = distortion;
@@ -1960,15 +1970,28 @@
         switch (this_mode)
         {
         case B_PRED:
+        {
+            int tmp_rd;
+
             // Note the rate value returned here includes the cost of coding the BPRED mode : x->mbmode_cost[x->e_mbd.frame_type][BPRED];
-            vp8_rd_pick_intra4x4mby_modes(cpi, x, &rate, &rate_y, &distortion);
+            tmp_rd = vp8_rd_pick_intra4x4mby_modes(cpi, x, &rate, &rate_y, &distortion, best_yrd);
             rate2 += rate;
             distortion2 += distortion;
-            rate2 += uv_intra_rate;
-            rate_uv = uv_intra_rate_tokenonly;
-            distortion2 += uv_intra_distortion;
-            distortion_uv = uv_intra_distortion;
-            break;
+
+            if(tmp_rd < best_yrd)
+            {
+                rate2 += uv_intra_rate;
+                rate_uv = uv_intra_rate_tokenonly;
+                distortion2 += uv_intra_distortion;
+                distortion_uv = uv_intra_distortion;
+            }
+            else
+            {
+                this_rd = INT_MAX;
+                disable_skip = 1;
+            }
+        }
+        break;
 
         case SPLITMV:
         {
diff --git a/vp8/encoder/rdopt.h b/vp8/encoder/rdopt.h
index d874409..72ba9a0 100644
--- a/vp8/encoder/rdopt.h
+++ b/vp8/encoder/rdopt.h
@@ -12,7 +12,7 @@
 #ifndef __INC_RDOPT_H
 #define __INC_RDOPT_H
 void vp8_initialize_rd_consts(VP8_COMP *cpi, int Qvalue);
-int vp8_rd_pick_intra4x4mby_modes(VP8_COMP *cpi, MACROBLOCK *mb, int *rate, int *rate_to, int *distortion);
+int vp8_rd_pick_intra4x4mby_modes(VP8_COMP *cpi, MACROBLOCK *mb, int *rate, int *rate_to, int *distortion, int best_rd);
 int vp8_rd_pick_intra16x16mby_mode(VP8_COMP *cpi, MACROBLOCK *x, int *returnrate, int *rate_to, int *returndistortion);
 int vp8_rd_pick_intra_mbuv_mode(VP8_COMP *cpi, MACROBLOCK *x, int *rate, int *rate_to, int *distortion);
 extern int vp8_rd_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, int recon_uvoffset, int *returnrate, int *returndistortion, int *returnintra);
diff --git a/vp8/encoder/temporal_filter.c b/vp8/encoder/temporal_filter.c
index dbe1d02..e4c3db1 100644
--- a/vp8/encoder/temporal_filter.c
+++ b/vp8/encoder/temporal_filter.c
@@ -70,7 +70,7 @@
     // U & V
     mv_row >>= 1;
     mv_col >>= 1;
-    stride >>= 1;
+    stride = (stride + 1) >> 1;
     offset = (mv_row >> 3) * stride + (mv_col >> 3);
     uptr = u_mb_ptr + offset;
     vptr = v_mb_ptr + offset;
diff --git a/vp8/vp8_common.mk b/vp8/vp8_common.mk
index cf47626..af07618 100644
--- a/vp8/vp8_common.mk
+++ b/vp8/vp8_common.mk
@@ -35,7 +35,7 @@
 VP8_COMMON_SRCS-yes += common/entropymode.c
 VP8_COMMON_SRCS-yes += common/entropymv.c
 VP8_COMMON_SRCS-yes += common/extend.c
-VP8_COMMON_SRCS-yes += common/filter_c.c
+VP8_COMMON_SRCS-yes += common/filter.c
 VP8_COMMON_SRCS-yes += common/findnearmv.c
 VP8_COMMON_SRCS-yes += common/generic/systemdependent.c
 VP8_COMMON_SRCS-yes += common/idctllm.c
@@ -111,14 +111,15 @@
 VP8_COMMON_SRCS-$(HAVE_SSE2) += common/x86/postproc_sse2.asm
 endif
 
+VP8_COMMON_SRCS-$(ARCH_ARM)  += common/asm_com_offsets.c
 VP8_COMMON_SRCS-$(ARCH_ARM)  += common/arm/arm_systemdependent.c
 
 # common (c)
 VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/bilinearfilter_arm.c
+VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/bilinearfilter_arm.h
 VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/filter_arm.c
 VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/loopfilter_arm.c
 VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/reconintra_arm.c
-VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/vpx_asm_offsets.c
 
 # common (armv6)
 VP8_COMMON_SRCS-$(HAVE_ARMV6)  += common/arm/armv6/bilinearfilter_v6$(ASM)
@@ -161,16 +162,3 @@
 VP8_COMMON_SRCS-$(HAVE_ARMV7)  += common/arm/neon/buildintrapredictorsmby_neon$(ASM)
 VP8_COMMON_SRCS-$(HAVE_ARMV7)  += common/arm/neon/save_neon_reg$(ASM)
 VP8_COMMON_SRCS-$(HAVE_ARMV7)  += common/arm/neon/recon_neon.c
-
-
-#
-# Rule to extract assembly constants from C sources
-#
-ifeq ($(ARCH_ARM),yes)
-vpx_asm_offsets.asm: obj_int_extract
-vpx_asm_offsets.asm: $(VP8_PREFIX)common/arm/vpx_asm_offsets.c.o
-	./obj_int_extract rvds $< $(ADS2GAS) > $@
-OBJS-yes += $(VP8_PREFIX)common/arm/vpx_asm_offsets.c.o
-CLEAN-OBJS += vpx_asm_offsets.asm
-$(filter %$(ASM).o,$(OBJS-yes)): vpx_asm_offsets.asm
-endif
diff --git a/vp8/vp8_cx_iface.c b/vp8/vp8_cx_iface.c
index b23bd95..ea99c61 100644
--- a/vp8/vp8_cx_iface.c
+++ b/vp8/vp8_cx_iface.c
@@ -912,8 +912,8 @@
         ctx->preview_img.x_chroma_shift = 1;
         ctx->preview_img.y_chroma_shift = 1;
 
-        ctx->preview_img.d_w = ctx->cfg.g_w;
-        ctx->preview_img.d_h = ctx->cfg.g_h;
+        ctx->preview_img.d_w = sd.y_width;
+        ctx->preview_img.d_h = sd.y_height;
         ctx->preview_img.stride[VPX_PLANE_Y] = sd.y_stride;
         ctx->preview_img.stride[VPX_PLANE_U] = sd.uv_stride;
         ctx->preview_img.stride[VPX_PLANE_V] = sd.uv_stride;
diff --git a/vp8/vp8cx.mk b/vp8/vp8cx.mk
index 932f145..1ec26e6 100644
--- a/vp8/vp8cx.mk
+++ b/vp8/vp8cx.mk
@@ -42,7 +42,7 @@
 VP8_CX_SRCS-yes += encoder/encodeintra.c
 VP8_CX_SRCS-yes += encoder/encodemb.c
 VP8_CX_SRCS-yes += encoder/encodemv.c
-VP8_CX_SRCS-yes += encoder/ethreading.c
+VP8_CX_SRCS-$(CONFIG_MULTITHREAD) += encoder/ethreading.c
 VP8_CX_SRCS-yes += encoder/firstpass.c
 VP8_CX_SRCS-yes += encoder/generic/csystemdependent.c
 VP8_CX_SRCS-yes += encoder/block.h
diff --git a/vp8/vp8cx_arm.mk b/vp8/vp8cx_arm.mk
index 6b624a7..abc5dc8 100644
--- a/vp8/vp8cx_arm.mk
+++ b/vp8/vp8cx_arm.mk
@@ -14,11 +14,13 @@
 #File list for arm
 # encoder
 VP8_CX_SRCS-$(ARCH_ARM)  += encoder/arm/arm_csystemdependent.c
+VP8_CX_SRCS-$(ARCH_ARM)  += encoder/asm_enc_offsets.c
 
 VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/encodemb_arm.c
-VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/variance_arm.c
 VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/quantize_arm.c
 VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/picklpf_arm.c
+VP8_CX_SRCS-$(HAVE_ARMV6)  += encoder/arm/variance_arm.c
+VP8_CX_SRCS-$(HAVE_ARMV6)  += encoder/arm/variance_arm.h
 VP8_CX_SRCS-$(HAVE_ARMV5TE) += encoder/arm/boolhuff_arm.c
 
 VP8_CX_SRCS_REMOVE-$(HAVE_ARMV5TE)  += encoder/boolhuff.c
@@ -32,6 +34,7 @@
 
 #File list for armv6
 # encoder
+VP8_CX_SRCS-$(HAVE_ARMV6)  += encoder/arm/armv6/vp8_variance16x16_armv6$(ASM)
 VP8_CX_SRCS-$(HAVE_ARMV6)  += encoder/arm/armv6/walsh_v6$(ASM)
 
 #File list for neon
@@ -50,17 +53,3 @@
 VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/neon/vp8_subpixelvariance16x16s_neon$(ASM)
 VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/neon/vp8_memcpy_neon$(ASM)
 VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/neon/vp8_shortwalsh4x4_neon$(ASM)
-
-VP8_CX_SRCS-$(HAVE_ARMV7)  += encoder/arm/vpx_vp8_enc_asm_offsets.c
-
-#
-# Rule to extract assembly constants from C sources
-#
-ifeq ($(ARCH_ARM),yes)
-vpx_vp8_enc_asm_offsets.asm: obj_int_extract
-vpx_vp8_enc_asm_offsets.asm: $(VP8_PREFIX)encoder/arm/vpx_vp8_enc_asm_offsets.c.o
-	./obj_int_extract rvds $< $(ADS2GAS) > $@
-OBJS-yes += $(VP8_PREFIX)encoder/arm/vpx_vp8_enc_asm_offsets.c.o
-CLEAN-OBJS += vpx_vp8_enc_asm_offsets.asm
-$(filter %$(ASM).o,$(OBJS-yes)): vpx_vp8_enc_asm_offsets.asm
-endif
diff --git a/vp8/vp8dx.mk b/vp8/vp8dx.mk
index 1acd674..62f6211 100644
--- a/vp8/vp8dx.mk
+++ b/vp8/vp8dx.mk
@@ -65,7 +65,7 @@
 VP8_DX_SRCS-yes += decoder/onyxd_int.h
 VP8_DX_SRCS-yes += decoder/treereader.h
 VP8_DX_SRCS-yes += decoder/onyxd_if.c
-VP8_DX_SRCS-yes += decoder/threading.c
+VP8_DX_SRCS-$(CONFIG_MULTITHREAD) += decoder/threading.c
 VP8_DX_SRCS-yes += decoder/idct_blk.c
 VP8_DX_SRCS-$(CONFIG_MULTITHREAD) += decoder/reconintra_mt.h
 VP8_DX_SRCS-$(CONFIG_MULTITHREAD) += decoder/reconintra_mt.c
diff --git a/vp8/vp8dx_arm.mk b/vp8/vp8dx_arm.mk
index 0803a9c..080c9af 100644
--- a/vp8/vp8dx_arm.mk
+++ b/vp8/vp8dx_arm.mk
@@ -12,6 +12,7 @@
 #VP8_DX_SRCS list is modified according to different platforms.
 
 VP8_DX_SRCS-$(ARCH_ARM)  += decoder/arm/arm_dsystemdependent.c
+VP8_CX_SRCS-$(ARCH_ARM)  += decoder/asm_dec_offsets.c
 
 VP8_DX_SRCS-$(HAVE_ARMV6)  += decoder/arm/dequantize_arm.c
 VP8_DX_SRCS-$(CONFIG_ARM_ASM_DETOK)  += decoder/arm/detokenize$(ASM)
diff --git a/vpx_scale/arm/neon/vp8_vpxyv12_copyframe_func_neon.asm b/vpx_scale/arm/neon/vp8_vpxyv12_copyframe_func_neon.asm
index 24d46cb..e6bb486 100644
--- a/vpx_scale/arm/neon/vp8_vpxyv12_copyframe_func_neon.asm
+++ b/vpx_scale/arm/neon/vp8_vpxyv12_copyframe_func_neon.asm
@@ -14,7 +14,7 @@
     REQUIRE8
     PRESERVE8
 
-    INCLUDE vpx_asm_offsets.asm
+    INCLUDE asm_com_offsets.asm
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
 
diff --git a/vpx_scale/arm/neon/vp8_vpxyv12_copyframeyonly_neon.asm b/vpx_scale/arm/neon/vp8_vpxyv12_copyframeyonly_neon.asm
index 6534827..febccc2 100644
--- a/vpx_scale/arm/neon/vp8_vpxyv12_copyframeyonly_neon.asm
+++ b/vpx_scale/arm/neon/vp8_vpxyv12_copyframeyonly_neon.asm
@@ -16,7 +16,7 @@
     REQUIRE8
     PRESERVE8
 
-    INCLUDE vpx_asm_offsets.asm
+    INCLUDE asm_com_offsets.asm
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
 ;void vpxyv12_copy_frame_yonly(YV12_BUFFER_CONFIG *src_ybc, YV12_BUFFER_CONFIG *dst_ybc);
diff --git a/vpx_scale/arm/neon/vp8_vpxyv12_copysrcframe_func_neon.asm b/vpx_scale/arm/neon/vp8_vpxyv12_copysrcframe_func_neon.asm
index dfc8db5..ec64dbc 100644
--- a/vpx_scale/arm/neon/vp8_vpxyv12_copysrcframe_func_neon.asm
+++ b/vpx_scale/arm/neon/vp8_vpxyv12_copysrcframe_func_neon.asm
@@ -14,7 +14,7 @@
     REQUIRE8
     PRESERVE8
 
-    INCLUDE vpx_asm_offsets.asm
+    INCLUDE asm_com_offsets.asm
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
 ;Note: This function is used to copy source data in src_buffer[i] at beginning of
diff --git a/vpx_scale/arm/neon/vp8_vpxyv12_extendframeborders_neon.asm b/vpx_scale/arm/neon/vp8_vpxyv12_extendframeborders_neon.asm
index e475b92..b0a3b93 100644
--- a/vpx_scale/arm/neon/vp8_vpxyv12_extendframeborders_neon.asm
+++ b/vpx_scale/arm/neon/vp8_vpxyv12_extendframeborders_neon.asm
@@ -14,7 +14,7 @@
     REQUIRE8
     PRESERVE8
 
-    INCLUDE vpx_asm_offsets.asm
+    INCLUDE asm_com_offsets.asm
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
 ;void vp8_yv12_extend_frame_borders_neon (YV12_BUFFER_CONFIG *ybf);